27th INTERNATIONALTOWING TANK CONFERENCE

Copenhagen, DENMARK p g ,Aug. 31 – Sep. 5, 2014

PROCEEDINGS ‐ VOLUME I

Table of Contents

Proceedings of the 27th ITTC

Volume I Preface Organising committee Supporting organisations Report of the Executive Committee

1

Report of the Advisory Council

10

Report of the Resistance Committee

14

Report of the Propulsion Committee

60

Report of the Manoeuvring Committee

128

Report of the Seakeeping Committee

195

Report of the Ocean Engineering Committee

263

Report of the Stability in Waves Committee

332

Report of the Quality Systems Group

414

Appendix 1 Committees of the 27th ITTC

439

Appendix 2 Tasks and Structure of the 27th ITTC Technical Committees and Groups

444

Appendix 3 Tasks and Structure of the 28th ITTC Technical Committees and Groups

458

Appendix 4 Description and Rules of the ITTC – Proposed for Adoption by the 27th Full Conference

475

Appendix 5 Member Organisations

496

Appendix 6 Designated Representatives (R), Committee Members (M) and Observers (O) invited to attend the 27th ITTC Conference

513

Volume II

Report of the Specialist Committee on CFD in Marine Hydrodynamics

522

Report of the Specialist Committee on Detailed Flow Measurement Techniques

568

Report of the Specialist Committee on Performance of Ships in Service

585

Report of the Specialist Committee on Hydrodynamic Noise

639

Report of the Specialist Committee on Hydrodynamic Modelling of Marine Renewable Energy Devices Report of the Specialist Committee on Ice

680

726

iii

Preface

ITTC 2014 is held in Copenhagen, Denmark – according to a recent UN survey the happiest nation in the world as many of you saw on a large commercial on the wall in the baggage claim area when you arrived to the airport. I am personally very happy to see so many international experts joining and actively participating in this important conference in Copenhagen. Just like for the last conference, I will also this time like to applause the presence of so many senior participants. Their presence is the evidence of the long history of ITTC. As part of the opening of the conference we are very happy to be able to welcome Bill Morgan who takes us through the long and interesting history of ITTC - presented by a man that has joined the journey a long part of the way! In Denmark, we are not so happy to experience that the number of shipyards has diminished over the last many years leaving the impression that Denmark is no longer a shipbuilding nation. It is true that we do not have so many shipyards anymore, but the amount of maritime designers, suppliers and service providers has increased. From the managing director of a Danish shipping company building ships in Asia I recently heard that they are using 23 different Danish

sub-suppliers for one of their new buildings. This is indicating that we are still a shipbuilding nation – the ships are just not built here in Denmark. The story above is an indication of the globalization of shipping and the inherent international nature of shipbuilding. The international element has over the past three years for ITTC resulted in a much stronger and visible representation in IMO regarding e.g. EEDI issues. ITTC is seen as a trusted advisor and we are addressed in many important questions within naval architecture – especially regarding test of energy efficient solutions. It cannot be stressed too much that if we wish to maintain our image as an independent, trustworthy group of experts, we shall work hard to be seen as such – otherwise we will lose credibility and influence. The future of ITTC is depending on maintaining a good image. The future work of ITTC is also important. To that end, the Executive Committee this period took the initiative to form a group within ITTC to formulate the future work of ITTC, under the leadership of Jürgen Friesch, HSVA. This work will start up at this conference and it shall be very interesting to follow it.

iv

I wish to use this opportunity to thank all the committees that have worked hard and passionately in the last three years. Their work is reflected in these proceedings. It is worth mentioning that the work in the committees is often done on top of all other daily duties and sometimes in people’s spare time. Such an effort should receive our deepest respect and thankfulness. I will also like to extend a special thank to our secretary, Aage Damsgaard, who has been working hard for ITTC during the last period as well as being instrumental in organizing of

the ITTC Conference here in Copenhagen. At FORCE Technology we recently and on the same day celebrated his 25 years jubilee, 70 years birthday and retirement. Aage has been proposed as secretary for ITTC for the next three years. Being reelected as secretary for ECMAR as well means that he will not retire completely…. I hope that you will all be very happy and enjoy your stay in Copenhagen as a result of both the technical and social programme of this conference!

Copenhagen, 18th July, 2014

Peter Kr. Sørensen

Chairman, 27th ITTC Executive Committee

v

Organising Committee

Peter Kr. Sørensen, FORCE Technology (Chairman)

Steen Sabinsky, European Maritime Development Centre

Valdemar Ehlers, Maritime Group of Society of Danish Engineers

Aage Damsgaard, FORCE Technology (Secretary)

Editor

Aage Damsgaard, FORCE Technology

vi

Financial support rendered by

IDA Maritimt Selskab City of Copenhagen Lauritzen Fonden

Den Danske Maritime Fond Torm Fonden

Orients Fond (Norden) Skibsteknisk Selskabs Fond

Den A. P. Møllerske Støttefond ONR Global

ITTC FORCE Technology

Executive Committee

Final report and recommendations to the 27th ITTC

1. INTRODUCTION

The roles and responsibilities of the Execu-

tive Committee are defined in the ITTC Rules and include

• Implementing the decisions of the Full Conference

• Representing ITTC between Con-ferences

• Replacing members of technical committees and groups as necessary between Conferences

• Accepting new member organisa-tions to the ITTC

• Managing the finances • Approving the arrangements and as-

sociated costs and registration fees for the Conference

• Reporting on its activities to the Full Conference

In addition, the Executive Committee ap-

points members of the Advisory Council and reviews the members on a regular basis.

During the 27th ITTC, the Executive Com-

mittee has performed its duties in accordance with the above. In addition, circumstances have required the Executive Committee to make policy decisions which could not await the next Full Conference. These decisions concern acts in relation to IMO’s introduction of the Energy Efficiency Design Index (EEDI) and are de-scribed later in this report. This has also led to

the recommendation to modify the ITTC Rules as also described later.

Finally, the Executive Committee has

worked together with the Advisory Council on the presentation of recommendations to be adopted by the 27th Full Conference.

2. OBITUARIES

2.1 Jong H. Hwang

Professor Jong H. Hwang died of

pneumonia March 12, 2012 at the age of 84. He was born on January 25, 1928 in Seocheon, Hamkyungnam-do, Korea.

Prof. Hwang played a leading role in founding Korea Towing Tank Conference (KTTC) on April 1980 in order to incorporate effectively with ITTC and became the first

11

Chairman of KTTC until 1984. He remained a member of the Steering Committee of KTTC until 1997. He served as a representative for the towing tank of Seoul National University and actively participated in ITTC. He made contributions by presenting written discussions at the 11th , 15th , 16th, and 17th ITTC.

He attended Seoul National University and received his B.S. degree in naval architecture in 1950 and Ph.D. degree in naval architecture in 1969. His teaching career started early in 1952 as TA until 1954. He stayed at the Massachu-setts Institute of Technology as a visiting fellow at the Department of Naval Architecture and Marine Engineering for one year beginning 1956. His academic career started as a lecturer at the Department of Naval Architecture, Seoul National University from 1954 to 1959. He became an assistant professor in 1959 and then an associate professor in 1961, and professor in 1969. He retired in 1993. Then he became a professor emeritus. He was a visiting professor at the Department of Naval Architecture, The University of Tokyo for one year from 1976 to 1977.

Professor Hwang was a pioneer in ship hyd-

rodynamics research and education in Korea. Early in his life, he played a leading role in setting up the ship hydrodynamics research program in the Department of Naval Architec-ture and Ocean Engineering, Seoul National University. He was the most influential profes-sor on many Korean ship hydrodynamicists. He also made a herculean effort to establish aca-demic ties between Korea and Japan in the field of naval architecture, specifically with the University of Tokyo and Hiroshima University. He organized the Korea-Japan Ship Hydrody-namics Seminar in 1970 as the first internatio-nal cooperative activity. He served as the 12th president of the Society of Naval Architecture of Korea (SNAK) from 1973 to 1975. During

this period, he put much emphasis on interacti-ons with ITTC, ISSC, PRADS and IUTAM. He published over one hundred research re-ports and journal papers and four books in en-gineering mathematics and one in naval archi-tecture. He was elected as a member of the Korea Academy of Arts and Science in 1999. Professor Hwang received many professional honors such as the Outstanding Scientific Achievement Award in 1986 from the Society of Naval Architects of Korea. He received the ‘Mokryun Medal’ and the ‘Moran Medal’ from the Government. He also received the ‘Sunggok Award’ from Sunggok Foundation.

Prof. Hwang is survived by his wife, Sook-

hee Kim, four sons, Woonsuk (Prof. in Inha Univ.), Woonkwang(Executive V.P. in LG Eletronics), Woonbong (Prof. in POSTECH), Woonjae( Prof. of Korea Univ.), and one daughter, Meeran.

2.2 Takao Inui

The Emeritus Professor Takao Inui of the

University of Tokyo, Japan went away on 14th September 2012 at the age of 92. He was a chairman of the 18th ITTC executive committee in 1987 (Kobe, Japan ). He has achieved a great number of excellent research works in the field of wave resistance. Especially he contributed to the dissemination of bulbous bow for merchant

22

ships. He was the fourth Weinblum lecturer and received a lot of awards including Japan Academy Prize and Japan Culture Prize (Bunkakourousha).

3. COMMITTEE MEMBERSHIP

Membership of the 27th Executive Commit-

tee has been the following: Chairman: Stig Sand, FORCE Technology.

Stig retired from FORCE on 31st December 2013 and was replaced by Peter K. Sørensen, also from FORCE.

Northern Europe representative: Susanne

Abrahamsson, SSPA. Central Europe representative: Jürgen Fri-

esch, HSVA. Southern Europe representative: Daniele

Ranocchia, INSEAN (CNR) Pacific Islands representative: Masashi Ka-

shiwagi, Osaka University East Asia representative: Suak-Ho Van,

KRISO Americas representative. F. Mary Williams,

NRC. Mary resigned in January, 2013, and was replaced by Antonio Fernandes, LabOceano.

Non-voting ex-officio members of the Ex-

ecutive Committee were: Gerhard Strasser, Vienna Model Basin, as

Advisory Council Chairman, Antonio Fernandes, LabOceano, as past

Chairman, until appointed Americas Area Rep-resentative,

Aage Damsgaard, FORCE Technology, as ITTC and EC Secretary.

4. COMMITTEE MEETINGS The committee has held four meetings be-

tween 2011 and 2014, and the final meeting will be held in Copenhagen during the 27th Full Conference.

The first meeting was held in Rio de Janeiro

during the 26th Full Conference. The committee reviewed the comments made by the Confer-ence to the Terms of Reference for the new technical committees and endorsed the revised version, which was subsequently issued to the technical committee chairpersons together with their appointment. The Terms of Reference are included as Appendix 2 of Volume I of the Proceedings of the 27th Full Conference. The technical committee members had already been selected and were approved by the Full Confer-ence, see Appendix 1 of Volume I of the Pro-ceedings.

The committee further discussed the issue

of AC membership qualifications. The discus-sion was instigated by the increasing number of universities seeking membership of the AC. The valuable work of the universities in the technical committees is highly appreciated, but care should be taken to avoid the procedures becoming too “academic” and impractical for use in the commercial tanks. It was decided to propose a revision of the Rules on this point, to emphasize the importance of commercial work as a qualification for being member of the AC.

The committee confirmed the financial

support to representation of ITTC at IMO meetings, and appointed Gerhard Strasser as the representative for EEDI/EEOI matters.

The second committee meeting was held in

Daejon, Korea, hosted by KRISO. The pro-posed wording of the modification of the ITTC Rules regarding AC membership was reviewed and agreed to be submitted to the Full Confer-

33

ence for adoption, and a new AC membership review form should be prepared, reflecting the revised requirements. It was decided that all AC members should report on the new form during this period.

The committee established a new working

group to deal with the future role of ITTC, in-cluding planning of future technical commit-tees. The group is headed by Jürgen Friesch and members are Gerhard Strasser, Masashi Kashiwagi and Suak-Ho Van.

The decision on venue for the 28th ITTC

Full Conference was opened for discussion. Both Korea (KRISO) and China (CSSRC) an-nounced that they would be candidates. They were asked to give a presentation of their can-didature at the third Executive Committee meeting.

The committee decided to re-admit Brodar-

ski Institute, Croatia, to the Advisory Council after it had settled the outstanding fees.

During the AC meeting preceding the

committee meeting, a serious dispute had arisen with regard to the process of revising the speed/power sea trial procedures and submit-ting them to IMO. As a consequence, the committee decided to produce a procedure for making decisions between Conferences and recommend this for adoption by the next Full Conference.

The third meeting was held in Copenhagen,

Denmark, hosted by FORCE Technology. The two candidates for hosting the 28th Full Confer-ence, CSSRC and KRISO, both gave very con-vincing presentations of their respective pro-posals, and the committee had to vote to make a decision which venue to recommend to the Full Conference. The decision was to recom-mend CSSRC to host the 28th ITTC Full Con-ference.

Two new members of ITTC were approved,

Technical University Berlin and Cranfield Uni-versity, and a number of changes to technical committees were implemented.

Further items covered in this meeting in-

cluded a presentation by the working group, approval of the venue and fee for the 27th Full Conference, ITTC Secretary for the 28th ITTC, and cost of attending IMO meetings.

The fourth meeting was held in Wuxi,

China, hosted by CSSRC. The committee reviewed the wording of

the proposed changes to the Rules of ITTC and agreed on a few further revisions.

The committee further reviewed the finan-

cial status. As a consequence of the increased financial support to the involvement in IMO and ISO activities in relation to EEDI and sea trial procedures, and of the introduction of the new website, it is expected that the total ex-penses for this period may exceed the fee in-come by up to 10,000 USD. As this amount may be covered by the reserves and as the cor-responding costs will be less during the next period, the committee decided that the mem-bership fee should be kept at its present level. The committee also agreed to support the print-ing of the Proceedings of the 27th ITTC by an amount of 6,000 USD.

The AC membership review was performed

using the revised form, which was found ac-ceptable with a minor adjustment. The revised form is annexed to this report.

The committee briefly reviewed the evalua-

tion of the members of the technical commit-tees submitted by the chairmen using the form annexed to this report. The evaluations will be considered when appointing candidates for the

44

next committees. For this purpose, the revised form of CV for committee member candidates, also annexed, shall be used.

Two new AC member organisations were

approved, Shanghai Ship and Shipping Re-search Institute, China, and Samsung Heavy Industries, Korea, as was Universiti Teknologi Petronas, Malaysia, as new member of ITTC.

Several new Area Representatives will be

or have been appointed for the next ITTC, in-cluding Kourosh Koushan, Marintek, for Northern Europe and Fabio di Felice, CNR-INSEAN, for Southern Europe. A new repre-sentative for Central Europe will be appointed later, as may a new representative for East Asia. The representatives for Pacific Islands and Americas shall continue for one more term.

5. COMMITTEE DECISIONS

5.1 Revision of ITTC Rules The committee decided to recommend to

the 27th Full Conference to adopt two modifica-tions to the ITTC Rules, one concerning deci-sion making between Full Conferences, the other concerning the qualifications for Advi-sory Council membership. The recommended modifications appear in clause 5.2 and 7.2(a) of RP 1.0-01, Description and Rules of the ITTC, respectively, which is found in the 2014-version of the ITTC Recommended Procedures and Guidelines. RP 1.0-04 describes the proce-dure for decision making between Full Confer-ences. Both documents are also found in Ap-pendix 4 of Volume I of the Proceedings.

The involvement of ITTC in external activi-

ties, e.g. in IMO and ISO, has necessitated the ability of the Executive Committee to make and execute decisions without having them

approved by the Full Conference in advance. During the past period, the Executive Commit-tee decided that ITTC should play a major role in the implementation of EEDI requirements in order to influence this important aspect of ship performance requirements. This necessitated a revision of the speed/power sea trial procedures which was implemented in 2012 and submitted to IMO beginning of 2013. At the same time, the two revised procedures 7.5-04-01-01.1 and 01.2 were made available on the ITTC website. The recommended modification of the ITTC Rules is made in order to formalise this way of operating.

The Advisory Council was formed at the

13th ITTC in 1972 in order to emphasize the role of the commercial tanks in relation to the procedures and guidelines issued by ITTC, to ensure that they are applicable in practice. Cri-teria for becoming a member of the Advisory Council were established, to show that their primary business was model testing for clients. The recommended modification strengthens the criteria slightly, also requiring work to be per-formed by professional full-time staff. The new requirements are reflected in the form used to assess present Advisory Council members every second period and new candidates for membership of the Advisory Council, partly asking the member to specify the division of their turn-over between commercial work, re-search and education, partly to specify the number of full-time staff performing work for clients. The new form has been tested on all Advisory Council members this period and on two new candidates.

In addition to these two major changes to

the Rules, some minor revisions were made for reasons of consistency. Among these is the definition of the role of the Vice Chairman of the Executive Committee and the Advisory Council, clauses 6.2 and 7.2, respectively.

55

5.2 Committees and Tasks of the 28th ITTC

Based on input from the technical commit-

tees, AC members, ITTC members and the EC Working Group on the Future of ITTC, the Advisory Council prepared the document de-fining the structure and Terms of Reference of technical committees and groups of the 28th ITTC enclosed as Appendix 3 of Volume I of the Proceedings of the 27th Full Conference.

5.3 Working Group on Future of ITTC The Executive Committee decided to con-

tinue the working group considering the future role of ITTC. However, as the specific tasks for this group could not be finalised during this term, a small group of committee members continued subsequently to elaborate the Terms of Reference and mode of operation of the group.

The working group has proposed the fol-

lowing draft Terms of Reference: Discuss open questions of long term devel-

opments in hydrodynamics by observing also different activities of networks worldwide (EU, aerospace, automobile)

Identify new topics of importance / rele-

vance to the ITTC Develop a strategy with regard to ITTC’s

role in national and international institutions like IMO, ISO, ISSC. It should additionally outline how ITTC should be represented in those bodies

Discuss if the way how ITTC works is still

adequate

Discuss if ITTC is ready to react flexible on short term questions mainly in between two Conferences (Does the role of the committee chairmen need to be defined in a new way)

Handle all external needs which are related

to non-technical issues Define how the group should interact with

AC and EC Represent ITTC in all political, administra-

tive bodies and organizations (technical input should be given by the technical committees), and suggest how this could look like

Suggest modifications to the ITTC Rules, if

necessary Adjust the relevant procedures, if neces-

sary.

5.4 Changes in ITTC Membership The following new members of ITTC have

been approved during this period: • Technical University Berlin, Ger-

many

• Cranfield University, UK

• Universiti Teknologi PETRONAS, Malaysia

The following members terminated their

membership: • Webb Institute, USA

• Icepronav, Romania The two Japanese members, IHI Corpora-

tion and Universal Shipbuilding Corporation, merged their test facilities, which now operate under the name Japan Marine United Corpora-

66

tion. JMUC continues IHI’s membership of the Advisory Council.

Several members have changed the names

of their Designated Representatives. The mem-ber information as known by the secretariat on 1st May 2014 is included in Appendix 5 of Volume I of the Proceedings of the 27th Full Conference.

5.5 Changes in Technical Committee Membership

The Executive Committee has endorsed a

number of changes of membership of the tech-nical committees. Appendix 1 of Volume I of the Proceedings includes a list of all past and present members of the committees of the 27th ITTC.

5.6 Advisory Council Membership Review During this period, all Advisory Council

members were reviewed using the revised evaluation form. The revised review form, re-flecting the increased emphasis on commercial work required for being a member of the Advi-sory Council, was generally accepted by the AC members. This form will therefore be im-plemented for future reviews, and the Execu-tive Committee will decide on the continuation of AC membership on the basis of these forms.

5.7 ITTC Website A new website has been developed and may

be found on www.ittc.info . The website is continuously updated with member information and provides links to ITTC Recommended Procedures and Guidelines, Symbols and Ter-minology, Dictionary, Sample QA Manual, and the ITTC wiki. It furthermore holds the Pro-

ceedings of all ITTC Conferences and the Cata-logue of Facilities. Under the News menu, you will find the ITTC News and Short News, which are posted when relevant. Finally, when relevant, it will provide a link to the conference website.

5.8 IMO and ISO Activities The Executive Committee has decided that

ITTC shall continue the cooperation with IMO in relation to the implementation of EEDI and EEOI. IMO has adopted the ITTC model test procedures and recommends the ITTC speed/power trial procedure to be used until the ISO15016 standard has been revised. The committee has consequently decided to con-tinue the financial support to ITTC’s involve-ment in the revision of the ISO standard.

5.9 ITTC Accounts The final accounts of the 26th ITTC and the

projected accounts for the 27th ITTC are shown below. All amounts are in USD.

Account item 27th pro-

jected 26th Final

ITTC fee 56,000 55,200 AC fee 66,500 62,750 Financial income 0 129 Total income 122,500 118,079 Secretariat hours 95,000 86,525 Secretariat expenses 12,000 16,374 IMO/ISO activities 15,000 2,754 Support to Conference 6,000 0 Financial costs 750 535 Misc. costs 1,000 565 Total costs 129,750 106,753 Net Result -7,250 11,325 Total equity capital 44,157 51,407

77

5.10 Budget for 28th ITTC The proposed budget (in USD) for the 28th

ITTC is shown below. The budget has been based on the assumption that the ITTC and AC fees and the remuneration of the secretariat remain unchanged.

Account item 28th

Budget ITTC fee 57,000 AC fee 66,500 Financial income 0 Total income 123,500 Secretariat hours 90,000 Secretariat expenses 12,000 IMO/ISO activities 10,000 Support to Conference 6,000 Financial costs 800 Misc. costs 700 Total costs 119,500 Net Result 4,000

6. RECOMMENDATIONS The Executive Committee recommends the

following to the Full Conference: • Adopt the following new and re-

vised procedures and guidelines recommended by the technical committees and group

o 4.2.3-01-03 o 7.5-01-03-01 o 7.5-01-03-03 o 7.5-01-03-04 o 7.5-02-01-01 o 7.5-02-01-04 o 7.5-02-01-05 o 7.5-02-02-02 o 7.5-02-02-02.1 o 7.5-02-02-02.2 o 7.5-02-03-01.4 o 7.5-02-03-01.6 o 7.5-02-03-02.1

o 7.5-02-03-02.2 o 7.5-02-03-02.3 o 7.5-02-03-03.2 o 7.5-02-03-03.3 o 7.5-02-03-03.4 o 7.5-02-04-02 o 7.5-02-05-04 o 7.5-02-05-05 o 7.5-02-06-01 o 7.5-02-06-02 o 7.5-02-06-03 o 7.5-02-06-04 o 7.5-02-06-05 o 7.5-02-07-02.1 o 7.5-02-07-02.2 o 7.5-02-07-02.3 o 7.5-02-07-03.7 o 7.5-02-07-03.8 o 7.5-02-07-03.9 o 7.5-02-07-03.10 o 7.5-02-07-04.2 o 7.5-02-07-04.4 o 7.5-03-02-03 o 7.5-03-02-04 o 7.5-03-03-01 o 7.5-03-03-02 o 7.5-03-04-02 o 7.5-04-01-01.1 o 7.5-04-01-01.2 o 7.5-04-04-01

• Delete and remove the following procedures and guidelines from the ITTC Recommended Procedures and Guidelines, as being obsolete

o 7.5-02-02-03 o 7.5-02-02-04 o 7.5-02-02-05 o 7.5-02-02-06 o 7.5-02-07-02.4 o 7.5-02-07-03.3

• Adopt the revised Register of ITTC Recommended Procedures and Guidelines as prepared by the Qual-ity Systems Group

88

• Adopt the revised ITTC Rules as re-flected in procedure 1.0-01 and 1.0-04

• Adopt the committee structure and Terms of Reference for the 28th ITTC as presented in Appendix 3 of Volume I of the Proceedings of the 27th ITTC Full Conference

• Accept the proposed chairs and members of the technical commit-tees and groups

• Accept that the Executive Commit-tee establishes an informal working group to consider the future role of ITTC

• Accept prof. Zhenping WENG, CSSRC as the next Chairman of the Executive Committee and CSSRC as host of the 28th ITTC Full Con-ference

• Accept Mr. Aage Damsgaard, FORCE Technology, as Secretary of the 28th ITTC

• Accept the continued involvement of ITTC in IMO and ISO with re-gard to EEDI, EEOI and safety mat-ters and to support this activity fi-nancially

• Accept the proposed budget for the 28th ITTC and general membership fee of 600 USD for the entire period

99

Advisory Council

Report to the 27th ITTC

1. MEMBERSHIP AND MEETINGS

The membership of the 27th Advisory

Council consisted of 35 organisations. Two organisations, Ecole Centrale Nantes, France, and LabOceano, Brazil, were admitted to the Advisory Council. The two Japanese ITTC members, IHI Corporation and Universal Ship-building Corporation, merged their test facili-ties, which now operate under the name Japan Marine United Corporation. JMUC continues IHI’s membership of the Advisory Council.

Prof. Gerhard Strasser was elected Chair-

man and Dr. Takuya Omori Vice Chairman. Mr. Aage Damsgaard was appointed Secretary.

The Advisory Council held three meetings

since the last Conference, In Daejon, Korea, October 2012, in Copenhagen, Denmark, Sep-tember 2013, and Wuxi, China, in April, 2014. The fourth meeting will be held in Copenhagen during the Conference.

The roles and responsibilities of the Advi-

sory Council as defined in the ITTC Rules are primarily to support the Executive Committee on all technical matters.

2. ACTIVITIES AND RECOMMEN-DATIONS OF THE ADVISORY COUNCIL

2.1 Kick-off Meeting with Technical Committees

At the 26th Conference in Rio de Janeiro,

the Advisory Council arranged a kick-off meet-ing with all the new chairmen of the technical committees who were present at the Confer-ence. The intention of the kick-off meeting was to give the new chairmen a good understanding of their work procedures and the liaison with the Advisory Council, and to emphasize the importance of a well coordinated commence-ment of the committee work.

The kick-off meeting appeared to have little

effect on the performance of the committees, so it will be repeated at the 27th Conference with even more focus on the first year’s activities of the committees and the results expected during that year.

2.2 Review of the Work of the Technical Committees and Groups

The technical committees and group pro-

vided progress reports for the Advisory Com-mittee meetings in October 2012 and Septem-ber 2013. The first progress report demon-

1010

strated that most of the committees had not properly understood the tasks concerning new and revised procedures. In order to improve the performance of the next committees in this respect, it was decided that a new standard template for the first progress report shall be implemented, spelling out in more detail what is required. The present report template may be used for the second progress report.

The technical committees and group sub-

mitted a total of 51 revised or new procedures for review by the Advisory Council and pro-posed a number of procedures to be deleted. With such a large number of documents, it is evident that keeping deadlines is crucial, as is the standard of work expressed in the proce-dures. During this period, about 80% of all “final” procedures had to be returned to the committees for further corrections following the last AC review. Had the AC review of pro-cedures been taken literally, all these proce-dures should have been postponed to the next period.

Consequently, it is very important for the

coming periods that the agreed working proce-dures are followed.

2.3 Advisory Council Working Groups As in the latest periods, the work of the Ad-

visory Council was organised in four working groups, each dealing specifically with the work of selected committees. The division of work this time was as shown in the table below.

WG 1 WG 2 WG 3 WG 4

Resis-tance

Propulsion Manoeu-vring

Ocean Engineer-

ing

SC CFD SC Ships in Service

Seakeeping SC Ice

SC De-tailed Flow

SC Hydrody-namic Noise

Stability in Waves

SC Energy Devices

The work of the Quality Systems Group

was monitored by the AC Chairman. The main responsibilities of the working

groups are to review committee progress re-ports, review procedures and guidelines and define the Terms of Reference for the next committees. In order to use the time at the AC meetings efficiently, the working groups as far as possible perform their review before the meetings. This is of course possible, only, if the documents are submitted timely by the com-mittees. The meetings are then used to consoli-date the comments resulting from the review in discussions with the entire AC and preparing responses to the committees.

2.4 ITTC Recommended Procedures A total of 51 procedures and guidelines

were prepared by the committees and group, 18 new and 33 revised documents. In addition, the committees proposed six old procedures to be deleted as they were considered obsolete. As described above, the large number of new and revised procedures and guidelines appeared to cause problems for the committees as well as the Advisory Council, which resulted in a sig-nificant delay of the publication and a consid-erable additional work load for the Quality Systems Group.

1111

Out of the 51 new or revised documents, just four were considered too incomplete to be published this time.

Some inconsistencies in the titles of a large

part of the new and revised procedures were discovered and have been revised by the QSG. As a consequence thereof, the procedure for formatting of procedures and guidelines has been revised. During the next period, the entire register will be reviewed by the QSG and any remaining inconsistencies will be corrected.

The proposed new Register of ITTC Rec-

ommended Procedures and Guidelines has been published on the conference website and will, if adopted by the Full Conference, be available on the ITTC website.

2.5 Technical Committees for the 28th ITTC

Several proposals for new specialist com-

mittees were tabled for the Advisory Council, partly by the present committees and partly by AC members. After voting in the AC, it was recommended to establish the following com-mittees for the 28th ITTC:

General committees Resistance Propulsion Manoeuvring Seakeeping Ocean Engineering Stability in Waves

Specialist committees Performance of Ships in Service Hydrodynamic Noise Ice Hydrodynamic Modelling of Marine Re-newable Energy Devices Modelling of Environmental Conditions Energy Saving Methods Groups Quality Systems Group The specialist committees on CFD and De-

tailed Flow Measurements were discontinued after having run for two periods. CFD and Flow aspects have to be handled by the respec-tive general committees.

2.6 ITTC Rules The Advisory Council was heavily involved

in the discussion of the changes of the ITTC Rules. The proposed procedure for decision making between Conferences was instigated by discussions between AC members, and the proposed strengthening of the requirements for AC membership were directly suggested by the AC. The EC report presents the proposed changes.

2.7 Cooperation with IMO and ISO The Advisory Council, represented by its

Chairman, has played a very active role in the cooperation with IMO on EEDI matters and with ISO on the revision of the sea trial stan-dard ISO 15016.

In relation to the EEDI regulations, ITTC

was firstly able to have the ITTC model test procedures accepted as the approved test pro-cedures to be considered in validation of EEDI. Secondly, the ITTC sea trial procedures were

1212

identified by IMO as the preferred standard, at least until a revised ISO 15016 was ready. IMO also advised ISO to cooperate with ITTC in the revision of ISO 15016 and base the revision on the ITTC procedures. The cooperation between ITTC and ISO has been very successful, and the revised ISO 15016 is now nearly complete. 3. OFFICERS FOR THE 28TH ITTC

ADVISORY COUNCIL Prof. Gerhard Strasser was reappointed as

Chairman for the 28th ITTC Advisory Council and Dr. Takuya Omori as Vice Chairman.

1313

Resistance Committee

Final Report and Recommendations to the 27th ITTC

1. INTRODUCTION

1.1 Membership and Meetings The members of the Resistance Committee

of the 27th ITTC are: Prof. Stephen Turnock (Chair) Univer-

sity of Southampton Southampton, United Kingdom

Dr. Hisao Tanaka Japan Marine United Corporation Tsu, Japan

Dr. Jin Kim Maritime and Ocean Engineering Research Institute Daejeon, Korea

Prof. Baoshan Wu China Ship Scientific Research Centre Wuxi. Jiangsu, China

Dr. Thomas C. Fu (Secretary) Naval Surface Warfare Center, Carderock Division W. Bethesda, Maryland, U.S.A.

Prof. Ali Can Takinaci Istanbul Technical University Istanbul, Turkey

Dr. Tommi Mikkola

Aalto University Helsinki, Finland

Four committee meetings have been held during the work period:

Istanbul, Turkey, 27-28 February 2012

at the Istanbul Technical University.

Bethesda, Maryland, U.S.A., 13-14 September 2012 at the Naval Surface Warfare Center Carderock Division.

Espoo, Finland, 10-11 June 2013 at Aalto University, Otaniemi Campus.

Southampton, United Kingdom, 14-15 January 2014 at the University of Southampton.

1.2 Tasks

The recommendations for the work of the Resistance Committee as given by the 26th ITTC were as follows:

1. Update the state-of-the-art for predicting

the resistance of different ship concepts emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

1414

a. The potential impact of new techno-

logical developments on the ITTC.

b. New experimental techniques and ex-trapolation methods.

c. New benchmark data.

d. The practical applications of computa-tional methods to resistance predictions and scaling.

e. The need for R&D for improving meth-ods of model experiments, numerical modeling and full-scale measurements.

2. Review ITTC Recommended Procedures

relevant to resistance and: a. Identify any requirements for changes

in the light of current practice, and, if approved by the Advisory Council, up-date them.

b. Identify the need for new procedures and outline the purpose and content of these.

c. Implement updated uncertainty analy-sis spreadsheet for resistance test.

3. Continue the analysis of the ITTC world-

wide series for identifying facility biases.

4. Review definitions of surface roughness and develop a guideline for its measure-ment.

5. Review results from tests that correlate skin friction with surface roughness.

6. Review trends and new developments in experimental techniques on unsteady flows and dynamic free surface phenomena.

7. Review new developments on model manufacturing devices and methods.

8. Review the development and evaluate im-provements in design methods and the capabilities of numerical optimization ap-plications, such as Simulation Based De-sign environments, with special emphasis on design of new ship concepts, geometry manipulation and parameterization, surro-gate models and variable fidelity applica-tions. (The fundamental assumption that an optimal hull shape is one that minimizes the calm water resistance may no longer be appropriate given the developments in CFD that give the designer the ability to make assessment of both wave and viscous ef-fects for added resistance in waves as well as the interaction between hull-propulsor and appendages.)

2. STATE OF THE ART

The concern of the shipping industry to

both reduce fuel use and hence expense, as well emissions, has placed greater emphasis on the ability to accurately resolve at design small changes in hull and appendage resistance. This desire has driven many of the state-of-the-art improvements seen since 2011 as the results of funded research programmes focussed on the energy efficiency design index (EEDI) start to reach maturity.

A review by Molland et al (2014) compares

alternative techniques for improving overall ship propulsive efficiency for both drag reduc-tion and improved propulsor efficiency. Table 1 compares the relative contributions of differ-ent resistance components for a variety of ship types. The domination of skin friction, espe-cially at slow speeds, confirms the research drive to improve coatings longevity and per-formance as well the search for alternative

1515

methods of reducing friction such as air lubri-cation. One area which has received little atten-tion to date is in methods to reduce air resis-tance of ship superstructures which although they constitute 2-4% of resistance are treated effectively as a bluff body dominated by pres-sure form drag (Molland et al, 2011). As such they are well suited to a relatively simple series of design modification. Investigations using a combination of CFD and wind tunnel tests are

expected to result in a new generation of streamlined ships.

Table 1 Approximate distribution of resistance components. Air drag is shown as a percentage of total resistance, i.e. total hull plus appendages plus air.(Molland et al, 2014) Type

Lbp (m)

CB

Dw (tonnes)

Service speed (Knots)

Service power ( kW)

Fr

Hull resistance component Air Drag % total

Friction %

Form %

Wave %

Tanker 330 0.84 250000 15 24000 0.136 66 26 8 2.0 Tanker 174 0.80 41000 14.5 7300 0.181 65 25 10 3.0 Bulk carrier 290 0.83 170000 15 15800 0.145 66 24 10 2.5 Bulk carrier 180 0.80 45000 14 7200 0.171 65 25 10 3.0 Container 334 0.64 100000

10000 TEU 26 62000 0.234 63 12 25 4.5

Container 232 0.65 37000 3500 TEU

23.5 29000 0.250 60 10 30 4.0

Catamaran ferry

80 0.47 650 pass 150 cars

36 23500 0.700 30 10 60 4.0

One method of reducing resistance is that of

adjusting the in service trim of vessels and this has prompted a significant number of towing tank studies. Larsen et al (2012) examined the physics of how adjusting trim can modify both the form and wave resistance components. They used a combination of model tests, RANS CFD, and potential flow theory to investigate the behaviour of a large cargo vessel at par-tially loaded draught and reduced speed. The RANS CFD was used to calculate the resis-tance and as shown in Figure 1 captured well the changes found in the model scale self pro-pulsion tests, whereas the potential flow under predicted the power change. Overall a 10% drop in power could be achieved with the cor-rect trim, with 80% originating from reduction in residuary resistance around the bulbous bow.

Figure 1. Comparison of different trim guid-ance methods at Fn=0.128 (Larsen et al, 2012).

The ability of CFD to resolve in detail the

flow features around the bulbous wave which initiate the drag changes with trim is captured well in Figure 2. An example of a study for a shipping fleet (Takinaci and Onen,2013) on trim optimisation found 4-14% power reduc-tion for a range of size of ships. Ships in the range of 40,000-80,000 tonnes had important

1616

potential benefits, whereas for larger ships the gains were found to be less.

Figure 2. Bow wave at 2.0m trim and Fn=0.128.Model test and RANS CFD (Larsen et al, 2012).

Another area of growing importance is the

understanding of the influence of detailed hull design on added resistance effects. In the past designing a ship for a single design speed, matching the propulsor in calm water, and then adding an appropriate powering margin was acceptable. The influence of the installed power term in the EEDI formula now chal-lenges designers to at least consider how they can better quantify the performance of the ship across its whole operational profile. A prob-abilistic approach can be applied for the ex-pected voyage sea states that allows a better assessment to be made of alternative hull de-signs. For example Winden et al (2013) stud-ied, using CFD, the influence of added and

calm water resistance in steady waves of a va-riety of bow forms, shown in Figure 3.

Figure 3. Six flares used to assess influence on added resistance (Winden et al, 2013).

Figure 4. Added resistance in waves for the tested bow sections.

Significant changes can be found as given

in Figure 4 for the change in added resistance compared to the reference hull. There is lim-ited available experimental data for validation of computational approaches but as described later in section 8 the ever growing capability of simulation based techniques will require such data to ensure that valid designs are imple-mented at full scale.

1717

2.1 New Technologies Air lubrication. As one of the energy-

saving technologies, frictional drag reduction technology by air lubrication has been devel-oped and its practical use is being attempted, see for example Kawakita (2013).

Kawashima et al. (2007) gives a progress

report of a research project moving towards practical use of air bubble injection as a drag reduction device for ships. The project aims to achieve a 10% net energy-saving by air bubble injection, taking into account the work needed for injecting air bubbles.

It is difficult to estimate the actual drag re-

duction effect on a full scale ship based on model scale experiments, as the relative scale

ratio of air bubbles to boundary layer length is very different between model and actual ship. Therefore they carried out experiments using a flat plate (L = 50 m, B = 1 m) in the 400 m towing tank of NMRI. The plate was towed at 6.2 m/s (12 kt), which equivalent to the cruis-ing speed of the ship for a full scale experi-ment. Air bubbles were injected at 3 m from the bow. Both the total drag of the flat plate and local skin friction were measured. The pro-cedure of the power estimation of the full scale ship in the state of the bubble injection is shown in Figure 5. The skin friction drag re-duction values by bubbly flow in full scale ship are estimated based on tank test result of flat plate. The drag reduction value in full scale ship with bubble flow drag reduction system is estimated based on a linear approximation. An example of estimation is shown in Figure 6.

Figure 5. Schematic diagram of full scale ship power estimation with bubble flow drag reduc-tions system.(Kawashima et al, 2007)

Figure 6. Estimation of drag reduction value in full scale ship with bubble flow drag reduction system based on curve approximation. (Kawa-shima et al, 2007) 2.2 Experimental Techniques and Ex-

trapolation

The advanced model measurement technol-ogy conference series organised via an EU sponsored research programme, the hydro test-ing alliance http://www.hta-forum.eu/, provides a valuable resource of up-to-date developments in experimental testing technology. The 3rd of the series was held in Gdansk in September

0

5000

10000

15000

20000

25000

0 2 4 6 8 10 12

tbc(m m )

dR(N)

C enter laneSide laneEnd laneTotal

1818

2013, Atlar and Wilczynski, (2013). The ses-sions concentrated on noise measurements, PIV applications, optical measurements, coating assessment and drag reduction, uncertainty, control technologies, free running models and smart tank testing.

Of relevance to the later discussion of CFD validation for resistance prediction is the method of waterline registration using fluores-cence, Geerts et al (2011). Waterline registra-tion is of use in assessing squat, freeboard and bow wave dynamics. The use of a fluorescent light source applied as a coating to the hull and illuminated by UV prevents unwanted reflec-tions and allows much more accurate capture of the dynamic surface waterline as shown in Fig-ure 7.

Figure 7. Comparison of an image from the same camera position but with different light-ing at a model speed of 0.65 m/s; above: image with regular tank lights, below: image with ultra violet lights.

The rapid reduction in the cost of inertial

measurement units and their ease of use, either through use of commercial ‘smartphone’ sys-tems or as more conventional instrument pack-

ages provides an alternative method of measur-ing model sinkage and trim. Bennett et al (2014) used a combination of three 9 degree of freedom wireless sensors, strain gauges, con-ventional heave and trim potentiometers and video analysis to investigate model response. Figure 8 shows the experimental setup. An experimental uncertainty analysis demonstrated that with suitable calibration comparable levels of uncertainty were obtained between the con-ventional heave and pitch measurements and those obtained derived using calibrated wire-less sensors. Such systems are ideal for use on free running models were conventional tech-niques are not applicable and often video mo-tion capture systems are difficult to use due to lighting or location of suitable fixture locations.

Figure 8. Schematic of experimental set-up of a segmented hydroelastic model with three Shimmer sensors.

2.3 New Benchmark Data,

The results of a major new experimental study for bench-marking data are not reported in this term of the Committee. But the plan of new measurements is confirmed with the Steer-ing Committee for CFD Workshop 2015 (Lars-son et al, 2014) and it will be used as the new benchmark case. The Workshop will be held at Tokyo in December 2-4, 2015. The detail in-formation can be found at http://www.t2015.nmri.go.jp, the workshop

1919

website. The model ship is named as Japan Bulk Carrier (JBC). The lines of JBC hull form are shown in Figure 9 and 10. The JBC hull form has a duct type ESD (energy saving de-vice) and so the experimental data will be ob-tained both with and without a duct. The model size and measurement items are shown in Table 2.

Figure 9. Body plan of JBC

Figure 10. Profile of JBC

Table 2. Items of measurement for JBC Condition Hull Measurement Towing

Tank Towing 7m BH X, M NMRI

7m BH w/o ESD

V, T NMRI

7m BH w/ ESD

V, T NMRI

3m BH w/o ESD

V, T OU

3m BH w/ ESD

V, T OU

Self-prop 7m BH SP, M NMRI 7m BH w/o ESD

V, T NMRI

7m BH w/ ESD

V, T NMRI

3m BH w/o ESD

V, T OU

3m BH w/ ESD

V, T OU

*X: Resistance, M: Trim and Sinkage, SP: Self-propulsion data, V: Velocities, T: Turbulence, NMRI: National Maritime Research Institute, OU: Osaka Uni-versity

2.4 Practical Applications of CFD A good overview of the current capabilities

of the CFD methods in ship hydrodynamics is provided by the CFD Workshop series. An interested reader is also referred to an extensive review of current capabilities and future trends of CFD in ship hydrodynamics by Stern et al (2013), which includes collected results and references on various resistance, sinkage and trim verification and validation studies. The latest Gothenburg 2010 workshop was quite extensively discussed in the report of the 26th ITTC Specialist committee on CFD. However, a book (Larsson et al, 2014) about the results, findings and conclusions of the workshop has been published recently with some additional experimental and computational data. For com-pleteness it is appropriate to collect some of the conclusions of the workshop which are most relevant for resistance and the associated flow predictions.

2020

Considering all the computed resistance predictions the mean difference between measurements and simulations is practically zero (-0.1%) and the mean standard devia-tion has improved considerably since the 2005 workshop (from 4.7% to 2.1%). The average comparison errors in sinkage and trim for Fr>0.2 are around 4%, whereas larger (relative) errors are observed for lower speeds most probably due to difficul-ties in measuring the quantities accurately and due to the small absolute values. Wave profiles on the hull and at the closest cut are generally well predicted, but large differ-ences between the methods are observed further from the hulls.

Grid sizes above 3 million cells do not pro-vide discernible improvement in resistance predictions (with URANS). Above and be-low this the resistance predictions are within 4% and 8% of the measured value. Finer grids with up to tens of millions of cells are required for local flow predictions. For DTMB 5415 accurate free-surface pre-diction can be obtained with just 2 million cells whereas finer grid are required for KVLCC2 due to shorter wave length.

The results suggest that turbulence models more advanced than the two-equation mod-els do not improve the resistance predic-tions. The anisotropic explicit algebraic Reynolds stress model seems to be the best option for predicting aft body flow of U shaped hulls with strong bilge vortex. The hybrid RANS/LES models seem promising, but they show limitations for flows with limited separation or triggering turbulence for slender bodies. Furthermore, the grid resolution requirements are significantly higher than for URANS based predictions.

The results suggest that it is easier to reach

convergent behaviour with grid variation

using structured rather than unstructured grids. The established uncertainty estima-tion methods give consistent results in the vicinity of the asymptotic range, but quite different estimates far from the range. Most resistance solutions are validated. For the non-validated solutions the source of error is suggested to be the turbulence model. The favourable characteristics of an anisot-

ropic turbulence model have been demon-strated by Guo et al (2013) as well. They have studied the distribution of resistance by meas-uring and simulating the calm water resistance, sinkage and trim of a three-segment KVLCC2 model. A comprehensive verification and vali-dation study shows that both isotropic and ani-sotropic models can give good prediction in terms of the measured quantities, but the supe-riority of the anisotropic explicit algebraic stress model is revealed by the resistance pre-diction of the aft segment. The study provides particularly interesting reference data for CFD model validation.

As the methods have matured and the mod-

elling knowledge has increased, Navier-Stokes equations based methods are used for an in-creasingly wide range of cases related to resis-tance and wave making. Castiglione et al (2014) have studied the validity of the RANS based resistance prediction for a catamaran model in shallow water and the influence of water depth of the interference effects. Maki et al (2013) have compared linear potential flow and RANS based methods for the prediction of the calm-water resistance components of a sur-face effect ship. Takai et al (2011) have studied the predictive capability of RANS based ap-proach for the performance analysis of a very large high-speed ship with a transit speed of at least 36 knots. Bhushan et al (2012) have stud-ied the vortical structures and the associated transom flow and sinkage and trim instabilities of the appended Athena hull form using hybrid

2121

RANS/LES approach including validation against full-scale experimental data.

Examples of current and future capabilities

of CFD with massively parallel simulations have been provided by Nishikawa et al (2012, 2013) and Fu et al (2013). Nishikawa et al (2012) have demonstrated fully resolved LES of KVLCC2 with Reynolds numbers of 5x105 and 1x106 with up to 1x109 cells and over 1500 cores and later (Nishikawa et al, 2013) with model test Reynolds number 4.6x106 using up to 32x109 cells.

These papers provide concrete examples of

the rapid development of high performance computing and of the computational require-ments of fully resolved LES simulations with practical Reynolds numbers. Fu et al (2013) on the other hand have studied the capabilities of a Cartesian grid immersed body, volume-of-fluid method for the simulation of planing hulls. They have compared measurements and mas-sively parallel simulations with 1-8x108 cells for three validation cases. The results demon-strate excellent reproduction of the flow details such as impact pressure, wetted length and spray sheet formation and good agreement in terms of hull attitude and resistance.

Despite the rapidly growing interest in bare

hull flow and resistance predictions based on the Navier-Stokes equations with (U)RANS, LES or DES modelling, there is still an interest to apply and develop potential flow based methods also for resistance predictions (see also the section on simulation based design). The methods have been improved both in terms of predictive capability and computational effi-ciency. Huang et al (2013) discuss the numeri-cal implementation of the Neumann-Michell theory of ship generated waves. They highlight the importance of specific implementation de-tails which are fundamental for the quality of the predictions. The developed approach pro-

vides a resistance prediction with an accuracy of around 10 percent for a wide range of dis-placement hull forms and Froude numbers.

Belibassakis et al (2013) have applied iso-

geometric analysis for the Neuman-Kelvin problem of the ship wave making and resis-tance. Here the same NURBS basis is used to define the geometry and the singularity distri-butions with the intention of providing the same accuracy with a lower number of panels and a natural connection with modern ship de-sign systems. Taravella and Vorus (2012) have developed an expanded, general solution of Ogilvie's formulation for moderate and high-speed ships (0.4<Fr<1.0) accounting for the wake trench generated by a fully ventilated transom. For the three cases shown with closed stern or fully ventilated transom the accuracy of the resistance prediction is roughly 10 per-cent from Fr=0.4 up. Yan and Liu (2011) have applied the Pre-corrected Fast Fourier Trans-form (PFFT) to improve the computational efficiency of the high-order boundary element method (BEM) for nonlinear wave-wave/body interaction. The approach is based on a process, where only the near-field contributions of the influence matrix are evaluated exactly. The approach reduces the O(N2~3) expense of the conventional quadratic BEM to O(N ln N). Se-ries 60 has been used as a practical case to demonstrate the applicability of the developed approach.

In terms of verification and validation of

computational predictions Eca and Hoekstra (2014) have proposed a new procedure for the estimation of numerical uncertainty. They have combined the traditional variable order expan-sion (ahp) and three alternative fixed order ex-pansions (ah, ah2, a1h + a2h

2). The procedure includes non-weighted and weighted fits and the best fit is selected based on the standard deviation of the fit. The procedure is tested with four different test cases including the re-

2222

sistance of KVLCC2. It is demonstrated that the alternative expansions are more frequently used as the complexity of the flow case in-creases. The results further demonstrate that it is hard to avoid scatter of data in complex flow cases and, thus uncertainty estimation proce-dures which are able to handle this are re-quired.

An extensive analysis by Zou and Larsson

(2014) of all the Gothenburg 2010 compares alternative approaches for verification and validation on the large data set of mesh refine-ment triplets. These indicate that for the various approaches tested that verification and valida-tion gives a relatively reliable error and uncer-tainty estimation when used within the asymp-totic region. That the level of iterative conver-gence needs to be assessed alongside grid con-vergence and that typically modelling error is small compared to numerical and experimental uncertainty.

2.5 Need for Research and Development

As the need for increased energy efficiency has grown, the number of unconventional hull forms, drag reduction technologies, and interest in multihulls has also grown. In order to effec-tively assess the performance of these tech-nologies and designs, including their perform-ance at sea in a range of sea states much re-search is needed in improving instrumentation and testing methods both at model and full scales, in the numerical modelling and under-standing the physics related to ship resistance. Specifically work needs to be done as it relates to high-speed planing hulls, multi-hulls, drag reduction technologies, and added resistance in waves. Advances in numerical modelling con-tinue, but increased emphasis on improved tur-bulence modelling and focus on simulation of high Reynolds number boundary layers, and high Froude number flows are needed to both

support advanced concept design and accurate prediction on ship flows and hydrodynamics forces. The ability to predict accurately viscous drag, wave drag, form drag (pressure) and spray drag continues to be of importance and continued work is needed. Schemes for han-dling surface roughness numerically still re-main an area of research as is the accurate modelling of full-scale boundary layers. Wave interactions between multihull hulls continue to be a challenge, as is the accurate prediction of spray drag for high-speed craft.

With the current emphasis on energy effi-

ciency, systems that provide weather routing to save fuel have been proposed. These systems rely on accurate prediction and knowledge of a ship’s added resistance in waves, which has led to a need for model testing procedures, as well as for full-scale ship trials. This work has dem-onstrated the difficulties in accurately charac-tering ship performance in a range of environ-mental conditions.

As the development of drag reduction tech-

nologies continues it will require improved instrumentation and testing techniques to accu-rately assess these technologies where the dif-ferences in the measured drag may only be 1-2%, but would still translate to a significant cost savings over the lifetime of the ship.

3. PROCEDURES

3.1 Resistance Tests

The evolution of the procedures for uncer-tainty analysis in measurement related to resis-tance tests has been overviewed.

The well-known ISO GUM (1995) is based

on the Guide: Recommendation 1 (CI-1981) by the Comité International des Poids et Me-

2323

sures (CIPM) and Recommendation INC-1 (1980) by the Working Group on the Statement of Uncertainties of the Bureau International des Poids et Mesures (BIPM). Meanwhile, the 18th ITTC Advisory Council (1984-1987) estab-lished an ad-hoc "Working Group on Valida-tion Techniques" with the task to discuss the subject concerned numerical methods, as well as pay the attention of ITTC to the uncertainty of physical model testing. The 19th ITTC Vali-dation Panel provided "Guideline for Uncer-tainty Analysis of Measurement" in Section II.3.2 of the Panel report (ITTC, 1990). They also presented excellent examples of uncer-tainty analysis, e.g., for resistance measure-ment, detailed in Section II.4.1 of the report, although, where the terminology of precision errors (random or repeatability) and bias errors (systematic or fixed) was used.

Table 3. Guides for uncertainty analysis in ITTC community before 2008

Procedure Number

Title

7.5-02-01-01

Testing and Extrapolation methods, Gen-eral, Uncertainty Analysis in EFD, Un-certainty Analysis Methodology. (1999/Rev00)

7.5-02-01-02

Testing and Extrapolation Methods, General, Uncertainty Analysis in EFD, Guidelines for Resistance Towing Tank Tests. (1999/Rev00)

7.5-02-01-03

Testing and Extrapolation, General, Den-sity and Viscosity of Water. (1999/Rev00)

7.5-02-02-01

Testing and Extrapolation Methods, Resistance, Resistance test. (2002/Rev01)

7.5-02-02-02

Resistance, Uncertainty Analysis, Ex-ample for Resistance Test. (2002/Rev01)

7.5-02-02-03

Resistance, Uncertainty Analysis Spread-sheet for Resistance Measurements. (2002/rev00)

7.5-02-02-04

Resistance, Uncertainty Analysis Spread-sheet for Speed Measurements. (2002/rev00)

7.5-02-02-05

Resistance, Uncertainty Analysis Spread-sheet for Sinkage and Trim Measure-ments. (2002/rev00)

7.5-02-02-06

Resistance, Uncertainty Analysis Spread-sheet for Wave Profile Measurements. (2002/rev00)

ISO GUM Guide to the Expression of Uncertainty in Measurement. (1995) (Drafted in 1993)

Since 1999, ITTC has recommended a se-

ries of procedures/guidelines for uncertainty analysis according to the AIAA methodology, as shown in Table 3.

The 25th ITTC Specialist Committee on

Uncertainty Analysis (2005-2008) revised the procedure 7.5-02-01-01 and in 2008, the 25th ITTC agreed to shift the methodology for analysis of uncertainty in measurement from the AIAA standard (symbolically by bias and precision uncertainties) to the ISO GUM meth-odology (symbolically by type A and type B uncertainties).

2424

Additionally, considering there is no sub-

stantial information given by the procedure 7.5-02-01-02(1999), “General guideline for uncer-tainty analysis of resistance tests”, the special-ist committee decided to revise it as an illustra-tive example for application of ISO GUM into a specific kind of hydrodynamic experiments in towing tanks. This revised procedure 7.5-02-01-02 was accepted and however, finally re-numbered 7.5-02-02-02 in 2008. Logically, this re-allocation of numeration is more proper, as this procedure should be in the procedure group of 7.5-02-02 related to resistance tests. How-ever, the procedure originally numbered 7.5-02-02-02 (2002), “Example for uncertainty analysis of resistance tests”, was dropped in 2008, although it would be better to be re-numbered 7.5-02-02-02.1, as a supplement of 7.5-02-02-02 (2008). The 27th ITTC Resistance Committee decided to revise this disappearing original procedure and then suggest to recover it as newly numerated 7.5-02-02-02.1 (2014), see Table 4.

Table 4. Changes in ITTC procedures for un-certainty analysis related to resistance in 2008 Procedure Number

Title

7.5-02-01-01

Guide to the Expression of Uncertainty in Experimental Hydrodynamics. (2008/Rev01)

7.5-02-01-02

(Revised and re-numbered 7.5-02-02-02)

7.5-02-02-02

Testing and Extrapolation Methods, General Guidelines for Uncertainty Analysis in Resistance Towing tank Tests. (2008/Rev01)

In revised version of the dropped procedure

7.5-02-02-02.1 (2008), a methodology is pro-vided that shows how ISO GUM process can applied in experimental hydrodynamics and illustrates some specific consideration that should be taken into the uncertainty analysis of resistance measurements, such as about the

uncertainty of wetted surface area in resistance tests. Especially, the procedure is revised to focus on resistance measurement, and does not include the process of extrapolation, so as to avoid the existing disputes on the analysis of uncertainties or more correctly modelling as-sumptions related to frictional line, form factor and residuary resistance coefficient, which should be dealt with in a new procedure for uncertainty analysis of extrapolation in future.

For underwater vehicles, e.g., torpedo and

submersible, the wetted surface area can usu-ally be estimated mathematically by the toler-ance of manufacture or practically measured with a systems such as a 3D Terrestrial Laser Scanner. The displacement volume of a sub-merged model represents the size of model and can be expressed as

HBL (1)

and the wetted surface area can be expressed as,

23/1S (2) where, L is the characteristic length, B the width and H the height of model. The relative tolerance of displacement volume can be esti-mated as

2222

H

H

B

B

L

L (3)

and then the relative uncertainty of displace-ment volume can be evaluated by combination of the uncertainties of length, width and height,

222HBL uuu

uu

(4)

2525

where, u denotes the standard uncertainty. Thereafter, the uncertainty of wetted surface area can be estimated as

3/2 u

S

uu S

S (5)

For surface vessels, the size of its underwa-ter part is determined by its weight (displace-ment mass, Δ),

water

(6)

Then, its uncertainty can be estimated by

2_2

wateruuu

u

(7)

And the uncertainty of wetted surface area can be calculated by Equation 5.

On the other hand, all the procedures as listed in Table 2 for uncertainty analysis of resistance tests seem to focus mainly on inter-preting the test results to the users of CFD simulation and are too complicated to be prac-tical or even useful to routine tests in towing tank. They are seen to be much too mathemati-cal rather than of practical engineering use in regular towing tank tests.

The 27th ITTC Resistance Committee per-

formed uncertainty analysis for a real example of a new series of resistance tests of DTMB 5415 model and found that the dominant com-ponents of uncertainty are of dynamometer accuracy (evaluated by calibration) and repeat-ability (estimated by repeat tests), as shown in Table 5 and Eq.8,

2ityrepeatabil2

rdynamomete

25

22

25

24

23

22

21

uu

uu

uuuuuuC

(8)

This is in agreement with experiences in

well-controlled commercial towing tanks. Therefore, the spreadsheet for resistance meas-urement (the procedure 7.5-02-02-03) is not necessary or even not useful for routine prac-tice of commercial tests, because the total/ combined uncertainty in resistance can be esti-mated simply by RSS (Root-Sum-Square) of that of dynamometer calibration and repeat tests as in Equation 8, although such a spread-sheet may be used in investigation of UA method or for detailed comparison of intra- and inter-laboratory tests.

Table 5. Example of uncertainty analysis in resistance measurement of DTMB 5415 model Component of Uncer-tainty in RT

Type Uncertainty Compon-ent in RT (Fr =0.28)

Wetted Surface Area B 1u = 0.035 %

Dynamometer (ν=32) A 2u = 0.19 %

Towing Speed B 3u = 0.067 %

Water Temperature B 4u = 0.024 %

Repeatability (N=9) A 5u = 0.45 %

Combined uncertainty for single measurement Cu = 0.49 %

Expanded uncertainty for single measurement (kP=2) PU = 0.98 %

Furthermore, the uncertainty propagated from towing speed into resistance can usually be considered negligible when the speed can be controlled with the accuracy recommended by the ITTC procedure 7.5-02-02-01. Therefore, the spreadsheet by the procedure 7.5-02-02-04 (2002) is not needed for routine tests.

The value of resistance is closely correlated

to the running sinkage and trim, but there is no

2626

analytical relation between resistance and its corresponding sinkage and trim. If special at-tention is given to the measurement of sinkage and trim, the detailed analysis as with the spreadsheet in the procedure 7.5-02-02-05 (2002) may be needed.

Finally, the measurement of wave profile is

quite different from that of resistance. Before developing a procedure for uncertainty analy-sis, a detailed procedure should be recom-mended for testing of wave profile measure-ment itself.

It is suggested that all the spreadsheets in

the procedures 7.5-02-02-03~06 (2002) can be dropped or if needed, revised in future and ad-ditionally, when repeat tests are performed to obtain the mean as measured and evaluate the uncertainty of repeatability, the outlier detec-tion will be included in the spreadsheets.

4. WORLD WIDE CAMPAIGN The world wide campaign has occupied the

resistance committee since the 24th ITTC. No new submissions have been made to the com-mittee since the 26th ITTC. The analysis pre-sented uses the available data to draw conclu-sions about inter-tank bias. A new spreadsheet based analysis tool was developed to draw to-gether all the data for comparative purposes. Although it is disappointing not to be able to fully exploit all the tests conducted by the many tanks who participated testing the small and large geosim models unless the data is submitted there is little that can be done. Simi-larly where there are ambiguities in the data submitted due to the double blind nature of the testing these are impossible to resolve.

4.1 Inter-laboratory comparison The comparison of test data from a total of

11 towing tanks for the large model of DTMB 5415 has been performed, as in what appears to be a mistake the data from No.10 tank is identi-cal to that from No. 4.

The large model, denoted as Geosim A,

used in the ITTC worldwide comparative tests is the CEHIPAR model 2716, a wooden geosim of the model DTMB 5415, with Lpp of 5.72 m, draft of 0.248 m in calm water without trim, displacement volume of 549 m3 (ITTC, 2005), corresponding to a scale of 24.824. The nomi-nal wetted surface area of 4.786 m2 (Olivieri et al, 2001) is adopted in expressing the total re-sistance coefficient (CT).

As prescribed by the ITTC comparative

tests, there would be used 9 repeat tests at each speed in each of towing tanks to perform statis-tical analysis. All the total resistance measure-ments in a specific tank are corrected to the nominal speed (Fr=0.1, 0.28 and 0.41) and converted to the nominal temperature of fresh water 15 degrees Celsius before any statistical analysis is made.

The means of total resistance coefficients

from those repeat tests in each tank are given in Table 6. Such means can be regarded as the best measurement in each towing tank. The experimental standard deviation (StDev) of tests in each tank is also presented. Such stan-dard deviations can be used to estimate the uncertainties of repeatability of measurement in each towing tank.

2727

Table 6. Statistical analysis of resistance meas-urement in comparative tests of the large DTMB 5415 model in 11 towing tanks

Tank No.

CT(10-3)_15deg_Fresh Water of Large Model (5.72m)_DTMB 5415_S=4.786m2

Fr =0.1 Fr =0.28 Fr =0.41 Mean StDev Mean StDev Mean StDev

# 1 3.956 1.2% 4.156 0.2% 6.429 0.2% # 2 3.917 1.6% 4.160 0.5% 6.497 0.5% # 3 4.007 0.9% 4.216 0.2% 6.536 0.2% # 4 4.306 3.6% 4.270 1.8% 6.587 1.9% # 5 4.008 1.2% 4.248 0.4% 6.617 0.3%

# 6 3.918 1.1% 4.234 0.6% 6.639 0.3% # 7 N/A 4.263 0.4% 6.480 0.5% # 8 3.959 0.5% 4.166 0.5% 6.336 0.8% # 9 4.001 1.9% 4.216 0.7% 6.590 1.9% # 10 (#4) # 11 3.989 1.1% 4.190 0.4% 6.412 0.2% # 12 4.019 2.3% 4.203 0.7% 6.368 0.7% Averaged after outliers (in RED) ticked out

Baseline 3.975 0.98% 4.211 0.96% 6.499 1.6%

Before any statistical analysis, a practical

approach to detect outlier is suggested for intra-laboratory comparison as the following steps:

Step 1: Calculate the mean (R0) and standard deviation (S0) of 9 repeat tests,

),1(1

i0 NiRN

Ri

(9)

),1()(1

1 20i0 NiRR

NS

i

(10)

Step 2: Judge if there is any test result outside the scattering band of double deviation,

),1(

?2 00i

Ni

SRR

(11)

Step 3: If no test is outside the band, no outlier exists. If the kth test is outside the “double” band, it will be doubted as an outlier. Tick it out and calculate the mean (R

*) and standard

deviation (S*) of the repeat tests again, exclud-

ing the kth test.

Step 4: Judge if the kth test is outside the scat-tering band of triple deviation,

?3 **k SRR (12)

Step 5: If the kth test is outside the “triple” band, its measurement can be considered as an outlier and then the mean R

* and standard de-

viation S* are adopted as statistic parameters of

repeat tests. Otherwise, no outlier is detected and the mean R0 and standard deviation S0 of repeat tests are used.

For inter-laboratory comparison, the aver-age of measurement means of 11 towing tanks can be considered as a kind of baseline.

When averaging the means of tests in 11

tanks, the detection of outlier can be performed following the above steps. The statistical analy-sis and corresponding results are shown in Fig-ures 11-13 and given in Table 7. These devia-tions are kind of measure for the facility bias. It is interesting to note that the scattering of data between towing tanks is much larger at speed of Fr=0.41 than that of Fr=0.1 and Fr=0.28.

Figure 11a. Statistical analysis for means of total resistance coefficients of 10 tanks

(Fr=0.1/including an outlier)

2828

Figure 11b. Statistical analysis for means of total resistance coefficients of 10 tanks

(Fr=0.1/excluding outlier)

Figure 12. Statistical analysis for means of total resistance coefficients of 11 tanks

(Fr=0.28, no outlier)

Figure 13. Statistical analysis for means of total resistance coefficients of 11 tanks

(Fr=0.41, no outlier)

The normalized deviations of means of re-sistance in each tank from the overall average of all tanks are summarized in Figure 14 and it shown that almost 95% of the means are within the scattering band of 2% of the overall aver-age, when the outlier is excluded.

2929

Figure 14. Scattering of means of resistance by 11 towing tanks in comparative tests of the

large DTMB 5415 model

The measurements of running sinkage and

trim would present more information to intra- and inter-laboratory comparison of resistance tests. For intra-laboratory comparison, the sta-tistical analysis for sinkage and trim from re-peat tests in each towing tank is given in Table 7-9. Obviously, the scattering of resistance is not closely correlated to that of sinkage.

Table 7. Statistical analysis of running sinkage and trim measurement in comparative tests of

the large DTMB 5415 model (Fr=0.1)

Fr =0.1 CT(10-3)_15deg_Fresh Water of Large Model (5.72m)_DTMB 5415_S=4.786m2

Tank No.

Resistance CT Sinkage (mm) Trim (Deg)

Mean StDev Mean StDev Mean StDev # 1 3.956 1.2% -1.64 0.31 -0.015 0.002 # 2 3.917 1.6% -1.05 0.40 -0.008 0.006 # 3 4.007 0.9% -1.19 0.08 -0.012 0.001 # 4 4.306 3.6% -0.85 0.24 -0.018 0.011 # 5 4.008 1.2% N/A # 6 3.918 1.1% N/A # 7 N/A N/A # 8 3.959 0.5% -1.30 0.03 -0.012 0.000 # 9 4.001 1.9% N/A # 10 (#4) # 11 3.989 1.1% -0.89 0.31 -0.014 0.001 # 12 4.019 2.3% N/A

Averaged after outliers (in RED) ticked out Averag

e (Baseline)

3.975 0.98%

-1.05 0.19

-0.013 0.003

Table 8. Statistical analysis of running sinkage and trim measurement in comparative tests of

the large DTMB 5415 model (Fr=0.28)

Fr =0.28

CT(10-3)_15deg_Fresh Water of Large Model (5.72m)_DTMB 5415_S=4.786m2

Tank No.

Resistance CT Sinkage (mm) Trim (Deg)

Mean StDev Mean StDe

v Mean StDe

v

# 1 4.156 0.2% -

10.95 0.29 -

0.113 0.002

# 2 4.160 0.5% -

10.75 0.43 -

0.103 0.005

# 3 4.216 0.2% -

10.49 0.11 -

0.102 0.002

# 4 4.270 1.8% -

10.39 0.30 -

0.111 0.009

# 5 4.248 0.4% -9.21 0.14 -

0.098 0.003

# 6 4.234 0.6% -

12.59 0.19 -

0.118 0.003

# 7 4.263 0.4% -

10.23 0.16 -

0.104 0.002

# 8 4.166 0.5% -

10.34 0.10 -

0.101 0.001

# 9 4.216 0.7% -

10.32 0.35 -

0.097 0.004 # 10 (#4)

# 11 4.190 0.4% -

10.05 0.30 -

0.015 0.004

# 12 4.203 0.7% -9.35 0.15 -

0.016 0.002 Averaged after outliers (in RED) ticked out

Average (Baseline)

4.211 0.96%

-10.21 0.55

-0.104 0.005

3030

Table 9. Statistical analysis of running sinkage and trim measurement in comparative tests of

the large DTMB 5415 model (Fr=0.41)

Fr =0.41 CT(10-3)_15deg_Fresh Water of Large Model

(5.72m)_DTMB 5415_S=4.786m2 Tank No.

Resistance CT Sinkage (mm) Trim (Deg) Mean StDev Mean StDev Mean StDev

# 1 6.429 0.2% -27.35 0.25 0.335 0.012 # 2 6.497 0.5% -26.30 0.33 0.373 0.004 # 3 6.536 0.2% -26.67 0.16 0.430 0.004 # 4 6.587 1.9% -25.96 0.51 0.415 0.019 # 5 6.617 0.3% -22.52 0.12 0.361 0.005

# 6 6.639 0.3% -29.45 0.28 0.535 0.009 # 7 6.480 0.5% -24.40 0.16 0.403 0.009 # 8 6.336 0.8% -25.21 0.07 0.367 0.005 # 9 6.590 1.9% N/A # 10 (#4) # 11 6.412 0.2% -25.24 0.08 0.378 0.006 # 12 6.368 0.7% -24.39 0.20 0.352 0.004

Averaged after outliers (in RED) ticked out Average (Baseline) 6.499 1.6% -25.34 1.45 0.379 0.031

For intra-laboratory comparison, the statis-

tical analysis for means of sinkage and trim from repeat tests in each towing tank is shown in Figures 15-17 and also presented in Table 7-9. The scattering of resistance is not closely correlated to that of sinkage, either, as shown in Figure 18.

Figure 15. Statistical analysis of running sink-age and trim measurement in comparative tests

of the large DTMB 5415 model (Fr=0.1)

Figure 16. Statistical analysis of running sink-age and trim measurement in comparative tests

of the large DTMB 5415 model (Fr=0.28)

3131

Figure 17. Statistical analysis of running sink-age and trim measurement in comparative tests

of the large DTMB 5415 model (Fr=0.41)

Figure 18. Correlation analysis of resistance to sinkage and trim measurement in comparative tests of the large DTMB 5415 model (Fr=0.41) 4.2 Wave Resistance Evaluation from

Worldwide Campaign Tests were done during the 24th, 25th and

26th ITTC periods. In the 24th ITTC, 20 institu-tions from 15 countries have been carried on the tests while in the 25th ITTC, 35 institutions from 19 countries have been participants. In the last 26th ITTC period, 41 institutions from 20 countries have been carried on the tests.

During the tests two geosims of the DTMB

5415 Combatant with 5.720 and 3.048 meters length have been used, see Table 10. Test Froude numbers are selected as 0.1, 0.28 and 0.41 and carried on 4 different days and 10 runs each set.

3232

Table 10. Hull geometric parameters LPP (m) 5.720 BWL(m) 0.724 T(m) 0.402 (m3) 0.842 S (m2) 4.8273

The purpose of the resistance test is to pro-

duce data for the temperature-corrected resis-tance coefficient. The total measured resistance values have been given with the file system. Therefore,

212

TMTM

M M M

RC

S V (13)

The residuary resistance of the model is

calculated from the model resistance tests tak-ing the form factor equals to k=0.15 (Stern et al, 2010) which is to be independent of scale and speed. The residuary resistance can there-fore be calculated as:

1R TM FMC C C k (14)

where CFM is derived from the ITTC – 1957 correlation line.

An Excel macro based spreadsheet is de-

veloped for the evaluation of wave resistance. The extreme values of maximum and minimum of residuary resistance of two models are given in Table 11.

Table 11. Maximum and minimum values of

residuary resistance

Figure 19. Distribution of maximum and minimum values of residuary resistance for the

Large Model.

Figure 20. Distribution of maximum and minimum values of residuary resistance for the

Small Model. Figures 19 and 20 show the distribution of

maximum and minimum wave resistance val-ues for each institute. The diversity in residuary resistance is quite high for the Fr= 0.1 when comparing with the others due to the measure-ment sensitivity is poor in the low speed range. Additionally, the spread of in resistance values is higher for the small model when comparied to the large one for the Froude numbers 0.28 and 0.41.

Negative wave resistance values exist in

both models for the Froude Number 0.28 due to the definition of form factor. Negative residu-ary resistance values still exist in the small

Fr LARGE MODEL (5.72m) SMALL MODEL (3.048m)

Min CR*1000 Max.CR*1000 Min CR*1000 Max.CR*1000 0.10 -19.0788 20.84345 -18.79000 23.66018 0.28 -1.8363 3.5725 -2.43352 4.82622 0.41 1.2535 5.6700 -0.33806 6.75376

3333

model for the Froude Number 0.41. Such val-ues suggest that selecting a fixed value of form factor valid for all Fn is incorrect. Indeed for dynamic hull with significant amounts of sink-age and trim a Fn dependency is to be expected and could be resolved through use of longitudi-nal wave cuts for measuring wave resistance, (Molland et al, 2011).

The usefulness of the spreadsheet based ap-

proach and other studies can be carried out using suitable macro functions in the future as it is able to access the whole database.

4.3 Comparison with variation from Goth-enburg 2010 study

As previously reported (ITTC, 2011b) the

same hull DTMB 5415 tested at two scales in the ITTC world wide campaign was one of the test cases (3.1a,3.1b and 3.2) for the Gothen-burg 2010 CFD Workshop (Larrson et al, 2014). Figure 21 shows the % variability for the total resistance coefficient for all the CFD values as a function of computational mesh size. The benchmark value for DTMB5415 is taken as the value from a single experimental source. In looking at the variability in the CFD data especially noting the differences between fixed and free to trim calculations, it can be seen that for the vast majority of the CFD re-sults even for the smallest mesh cases lie within 5% of the mean.

The accompanying statistical analysis for

the DTMB hull extracted here as Table 12 quantifies values with mean differences from the single experimental test case varying be-tween 0.1% (free at Fn=0.28) and 4.3% (free at 0.41), In comparing these values with those presented for the large model which had a stan-dard deviation of 1% from the mean, although the CFD still has a larger variability, with lar-ger mesh calculations the uncertainty is ap-

proaching that of the general capability of tow-ing tanks to measure resistance. Another point worth emphasising is that the Gothenburg workshop used test data from INSEAN (Case 3.1 fixed , large model Fn=0.28), IIHR (Case 3.1b fixed, small model Fn=0.28) and INSEAN (Case 3.2, free to sink and trim, large model Fn=0.28,0.41). The majority of the data in Larrson et al (2014) is presented as a percent-age difference from the experimental value. For case 3.1a for the large model as shown in Table 8 the World Wide Campaign (WWC) mean value of CT is 4.21x10-3.although comparable with the value of 4.23 x10-3 shown in Fig. 5.24 of Zou and Larrson(2014). This change in-creases the error from 2.6% to 3.1% although the WWC was a different physical model. The comparable change for the more realistic free model (case 3.2) is from 0.1%D to 0.6%D.

It is worth noting that the uncertainty with

computing free sinkage and trim appears al-ready to be comparable with the capabilities of tanks to measure these quantities to a common datum. One useful facet not originally in-cluded in the values of sinkage and trim was the influence on uncertainty on the level of the IIHR rails, (Larrson et al, 2014, 53-64). After the original presentation of the data it was found that there were significant variation in both rails in the IIHR towing tank. These were subsequently re-levelled increasing the overall experimental uncertainty in sinkage for in-stance from a maximum measured variation of 1.29mm on the east rail to 0.462mm.

3434

Figure 21. Variation in Resistance Coefficient with mesh for all Gothenburg 2010 calm water resistance test cases, including worldwide cam-paign hull DTMB5415,

Table 12 Gothenburg 2010 Calm water resis-tance CFD results for DTMB5416 test cases

Case Fn %E σ % No. of Submis-sions

3.1a

Fixed S &T

0.28 2.5 5.3 11

3.1b Fixed S & T

0.28 -2.6 4.4 5

3.2

Free

0.138 -2.8 4.4 5

0.28 0.1 2.1 6

0.41 4.3 1.4 5

4.4 Recommendations for the World Wide Campaign

The worldwide campaign data should be

made available via new ITTC website. The previous committee has provided an easily used database for additional studies. Further analysis was conducted by the committee and has shown some greater understanding. No new data was received. We suggest an approach for inter tank bias comparison, established a base line by removing 'outliers' and make accessible the whole database via new ITTC website and will provide a searchable spreadsheet for use when looking at all data. A comparison is made with the corresponding data from the CFD analysis from Gothenburg 2010.

For future such campaigns, the double blind

although a good idea in reality was too much of challenge. The inability to resolve ambiguities in submission, despite the prescriptive spread-sheet based uncertainty procedures (7.5-02-02-03 to 7.5-02-2.06) and the failure of many tow-ing tanks to submit the analysed data severely restricted the size of the data set for both large and small models. Similarly the challenge of moving models between countries and the pos-sibility of damage due to transit could very well have introduced its own age related bias. Any future such activity led by the ITTC should consider following an open approach to ensure the collective community of expertise can en-sure data collected is always to a high standard. Questions that are as yet not possible to resolve are whether the dominant bias is associated with tank blockage or if as in the IIHR tests it is the lack of levelness in the rails which causes the problems with the sinkage and trim com-parisons.

3535

5. SURFACE ROUGHNESS

5.1 Introduction

This section summarizes the state of art in hull surface roughness of actual ship and its influence on the roughness allowance (ΔCF), including experimental and numerical ap-proaches.

5.2 Measurement and evaluation of roughness

For the measurement of roughness of actual

hull surface, the BMT Sea Tech Hull Rough-ness Analyser (a stylus instrument with a sur-face probe) is used in many shipyards as the standard measurement tool. The hull roughness is normally measured in the way that the hull is divided into 10 equal sections with 10 meas-urements each, 5 on the port side and 5 on the starboard side. A total of 50 readings are taken on each side, 30 on the vertical sides and 20 on the flats. From the 100 measuring locations, the average hull roughness is calculated (ITTC 2011).

In ISO-4287:1996, various roughness pa-

rameters are defined. Surface roughness in general is a measure of the texture of a surface, and this is calculated on a profile or on a sur-face. Profile roughness parameters (Ra, Rq….) are more common whereas area roughness pa-rameters (Sa, Sq,….) give more significant values.

There are many different roughness pa-

rameters in use, but Rz, is a useful parameter because it can consider as the BMT roughness parameters. The definition of Rz in 50 mm evaluation length is similar to definition of BMT roughness, and its value is almost same. (Mieno, 2012).

On the other hand, roughness measurements

on ship models are carried out at few model basins (e.g. MARIN, SSPA), but the results of the measurements are used for quality assur-ance and not for further investigation. Most of the model basins do not measure the roughness of the model’s hull. 5.3 Experimental approach of roughness

influence In order to clarify correlation between

properties of coatings on ship hull surface and frictional resistance, experimental studies were carried out by Tanaka et al. (2003), Weinell et al. (2003), and Mieno (2012). A rotating cylin-der type dynamometer is used to measure fric-tional resistance of coatings at higher Reynolds number flow similar to that around a real ship. Measuring the frictional resistance and change of roughness of cylinders coated with self-polishing type paint or water repellent paints, correlation between properties of coating and frictional resistance can be investigated. Fur-ther, a simple method based on these experi-mental results like that shown in Figure 22 can estimate the frictional resistance acting on the surface of the actual ship hull.

Figure 22. Schematic flow of roughness allow-ance estimation by rotating cylinder method.

3636

Tanaka et al (2003) proposed roughness al-

lowance estimation method based on the results of rotating cylinder experiments. Under the assumptions that the flow around the rotating cylinder around becomes turbulent and wall law is established near surface of the cylinder near the surface, velocity profile in boundary layer is as follows.

* 1

lnu y B B

(15)

Estimating roughness function ΔB, fric-

tional resistance including influence of rough-ness can be easily obtained by boundary layer calculation. In Figure 23, it is shown an exam-ple of correlation between surface roughness and frictional resistance.

Figure 23. Correlation between surface rough-ness and resistance coefficient (Tanaka 2003).

When equivalent sand roughness Ks of an

actual ship hull is obtained, roughness allow-ance ΔCF can be estimated by various expres-sions of frictional resistance. An example of

estimated roughness allowance by White's for-mula (White, 1991) is shown in Figure 24.

Figure 24. Example of estimated roughness

allowance for actual ship (Tanaka 2003). The rotating cylinder method is used in or-

der to investigate the effect of damaged hull surface, newly developed paint performance. Weinell et al. (2003) carried out the rotating cylinder tests to investigate the effect of rough-ness on the frictional drag. One smooth cylin-der and two sand roughened cylinders are used for reference, seven roughened cylinders are investigated. In the experiment, torque is meas-ured. Further, ageing effect for fiber and non-fiber containing paints are also examined. Also roughness of simulated weld seam and simu-lated paint remain are also investigated. With respect to frictional drag, the contribution from a modern self-smoothing antifouling or silicone based fouling-release paint is negligible com-pared to the contribution from irregularities found on ship’s hull. In the investigated range of roughness, micro-roughness was found to be much more important than macro-roughness. On the other hand, large-scale irregularities were found to be even more important than both micro-and macro-roughness.

Mieno (2012) investigated the influence of

various roughness parameters to frictional re-sistance increase. In this study the influence of the surface roughness on the friction was measured using rotating cylinder, and rough-ness was measured with a Laser displacement meter and the surface parameters were investi-

104 105

3

4

L/Rz

1000

Cf

at Rn=8,000,000

SF–1

SF–1R

SF–3

SF–5

SF–5R

SF–5RR

5.5kt 10.1kt

Estimated

100 200 3000

0.0002

0.0004

0.0006

0.0008

L(m)

CF

Ke=200m

Ke=100m

Ke=50mKe=25m

ITTC1978

3737

gated by the JIS B 06015 method (similar to ISO-4287). Roughness of an actual ship hull surface was measured by replicar method. The Friction Increasing Ratio FIR(%) is defined in Equation (16). T is equal to the torque meas-ured using a painted cylinder and T0 is the torque measured on a smooth surface cylinder.

100(%)0

0

T

TTFIR (16)

Relation between Rz and FIR is shown as a

graph in Figure 25. Dry spray (DS) are plotted as the symbol, conventional self-polishing coating (C_SPC) as the symbol, new genera-tion self-polishing coating (N_SPC) as the symbol, foul release coating (FRC) as the symbol. Friction increasing was observed ac-cording to increasing of Rz. Even at the same Rz value, FIR differences are observed between FRC, N_SPC, C_SPC and DS. Relation be-tween Sm and FIR is shown in Figure 26. Sm of DS ranges from 2000 to 3000 micron and FIR is more than 25%. Sm of C_SPC ranges from 3000 to 4500 micron of N_SPC from 4000 to 7000 micron and of FRC Sm is more than 8000 micron. FIR for FRC and N_SPC is less than 2%. A lower Sm-value tends to increase FIR more than a higher value. Sasajima (1965) re-ported a correlation between the squared height parameter divided by wavelength as H2/λ and the friction coefficient. When Rz is considered as H, and also Sm is considered as λ, there should be a correlation between FIR and Rz

2/Sm.

Figure 25. Relation between Rz and FIR (Mieno 2012)

Figure 26. Relation between Rz

2/Sm and FIR (Mieno 2012)

As a new experimental technique using a

flat plate, Kawashima (2012) has proposed a new experimental method shown in Figure 27. Aiming to clarify the relationship between fric-tion resistance and roughness parameter (height, period, slope, etc.), the authors carried out tank tests of flat plates that have various types of roughness by painting. According to the results of the tank test, frictional resistance increase becomes smaller as roughness height length ration H/L becomes larger.

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

0 50 100 150 200

Rz (micron)

FIR

(%

)

DS

C_SPC

N_SPC

FRC

y = 2.6195x

r2 = 0.8669

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

0 2 4 6 8 10

Rz2/Sm (micron)

FIR

(%

)

DS

C_SPC

N_SPC

FRC

3838

Figure 27. Schematic view of measurement system (Kawashima, 2012).

5.4 Theoretical and numerical approach of roughness influence

A new theoretical friction factor model for

fully developed turbulent internal flows of smooth and rough pipes and channels has been developed by using a new velocity profile, which is a combination of logarithmic and power law profiles (Atkan et al, 2009). The proposed equation is explicit function of Rey-nolds number and relative roughness. Constants in the derived equation for the friction factor are given by experimental data. The formula recovers Prandtl’s law of friction for smooth pipes well. The model also shows good correla-tion with the available data for turbulent flows in rough pipes for wide ranges of Reynolds number and surface roughness covering the entire Moody chart. The maximum relative error between the published experimental fric-tion factors and those calculated from the de-veloped equation was found to be less than 3%, and the proposed relationship agrees with the Blasius relationship for low Reynolds numbers to within 1%.

Considering roughness influence, Katsui et

al. (2011) proposed a new flat plate friction formula for wide Reynolds number range based on momentum-integral equation and Coles’ wall-wake law. The flat plate frictional coeffi-cient is evaluated by solving a differential equation introduced White’s roughness func-

tion. Roughness allowance ⊿CF by surface roughness is dependent upon roughness height non-dimensionalized by plate length and Rey-nolds number. It is possible to evaluate full scale ship resistance increase caused on surface roughness with the presented method. To esti-mate ⊿CF easily, ⊿CF formula approximated with function of non-dimensional roughness height is also presented in Figure 28.

Figure 28. Added frictional resistance due to surface roughness (Katsui et al, 2011).

Eça and Hoekstra (2010) reported the ef-

fects of hull roughness on viscous flows around ships. These effects are computed by replacing the typically non-uniform roughness of the hull surface by a uniform sand roughness. The cal-culations are performed with the RANS-code PARNASSOS using the SST k−ω model. No wall functions are applied, and the roughness effect is introduced via a change in the ω wall boundary condition. For a tanker, a container ship and a car carrier, the flow is computed at model and full scale Reynolds numbers for sand-grain roughness heights ranging from 0 (smooth wall) to 300 μm. Each case is com-puted on six nearly geometrically similar grids to allow a fair estimate of the numerical uncer-tainty. The results shown, in Figure 29, confirm that an increase of the roughness height leads to an increment of the friction and pressure resis-

106 107 108 109 10100

1

2

ΔC

F×

103

ReL

k/L=1.0×10-5

k/L=9.0×10-6

k/L=8.0×10-6

k/L=7.0×10-6

k/L=6.0×10-6

k/L=5.0×10-6

k/L=4.0×10-6

k/L=3.0×10-6

k/L=2.0×10-6

k/L=1.0×10-6

3939

tance coefficients and the wake fraction. It is clear from the data that there is a significant scale effect, depending not only on the global Reynolds number (based on the ship length) but also on the roughness Reynolds number (based on the roughness height). The combina-tion of the effects observed for CF and CP is reflected in the viscous resistance coefficient, CV. Since CF is dominant, the behavior of CV is similar to that observed for CF. The roughness height does not affect only the near-wall flow. The wake field at the propeller plane is also clearly influenced by hR. The thickness of the “boundary-layer” and the mean wake fraction grow with the increase of hR.

Figure 29. Ratio between friction resistance coefficients predicted with and without sand-grain roughness.

As the first step toward the flow simulation

of a full scale ship with hull surface roughness, Hino (2012) computed simple 2-D channel flows and flat plate flows using the current turbulence models with roughness effect. The results are compared between a smooth wall and a rough wall and the level of applicability of the current roughness models is examined. Flow computations in a 2-D channel and around a flat plate with and without surface

roughness are carried out in order to examine the applicability of the current roughness mod-els in a turbulence closure. 2-D Channel test cases show that κ-ω based roughness model is more robust than SA based model, though both models can simulate effects of sand-grain roughness fairly well. The flat plate simulations also reproduce a reasonable behaviour of fric-tional resistance increase by the roughness ef-fect. For applications to full scale ship flows with a surface roughness, the extension of the current roughness model is required, since the roughness distribution is supposed to be not uniform in the paint surface of an actual ship as shown in Figure 30.

Figure 30. Logarithmic plots of velocity pro-files by SST model at Rn ≈ 108 (Hino, 2012).

5.5 Conclusions ISO-4287 definitions of roughness, which

are widely used in industry, are possible to rep-resent various characteristics of roughness. On the other hand, measurement with BMT rough-ness analyser is general in many shipyards, but it seems to be difficult to understand various characteristics of roughness by the device. Be-cause Reduction of frictional resistance is the essential task for energy saving, it is expected that new types of paints for this purpose will be developed in the future. Therefore, it will be necessary to evaluate more precisely the

4040

roughness influence to hull frictional resistance based on experimental, theoretical, and nu-merical results.

The RC suggests the following items for the

future: (i) Continue to monitor trends and new de-

velopments in measurement techniques of hull roughness.

(ii) Continue to review trends in roughness

definition considering estimation of roughness allowance.

(iii) Continue to monitor new developments

in experimental techniques for roughness al-lowance estimation.

(iv) Continue to monitor new developments

in theoretical and numerical estimation tech-niques for roughness influence to frictional resistance increase. 6. UNSTEADY FREE SURFACE

Experimental tow tank and full-scale meas-

urement techniques have focused on unsteady flows and free-surface phenomena including wave breaking. These techniques have been motivated by interest in wave impact and slamming, spray generation, air entrain-ment/bubble generation, and wave breaking. The experimental work has focused on funda-mental understanding as well as model devel-opment and code validation. Recent examples of full-scale field measurements include Beale et al, (2010), Drazen et al (2010), and Fu et al, (2012), and of laboratory-scale measurements by Masnadi et al (2013), Wang et al (2012), and Andre and Bardet (2014a and 2014b). While work in this area began in earnest back in 2004 with laboratory work utilizing laser induced fluorescence methods, free-surface

flow visualization extends back to the 1990s, Dong et al (1998) and Waniewski (1998) for example (see Figure 31). These laser fluores-cence methods were initially utilized to meas-ure the free-surface of non-breaking flow fields (see Duncan, 1999), but were soon extended to breaking waves, see Kiger & Duncan (2012) and multiphase flows (Fu et al, 2009).

Figure 31. Sample image of the overhead view of the bow wave generated by towing a ship model, from Dong et al (1998).

While standard planar laser induced fluo-

rescence (PLIF) has been used to identify 2-dimension wave profiles, only recently have they been extended to 3-dimensions and cou-pled with PIV measurements. Figure 32 (cour-tesy of Philippe Bardet) shows a conceptual test of multiple simultaneous PLIF planes re-corded with a single camera.

4141

Figure 32. A Multi-plane PLIF sample to demonstrate principle of optical configuration for 3-D surface profile reconstruction (courtesy of Philippe Bardet).

The lower cost of high-resolution digital

cameras and the development light field-imaging which involves sampling a large num-ber of light rays from a scene to allow for scene reparameterization (Isaksen et al, 2000) and synthetic aperture refocusing, Synthetic aper-ture refocusing allows individual planes in the scene to be focused on, while planes not of interest are blurred and has allowed for the development 3-D imaging systems capable of simultaneously measuring a fluid volume and “seeing-through” partial occlusions (see Belden et al, 2010; Belden et al, 2011; and Belden and Techet, 2014).

While development of sophisticated 3-D

techniques are being developed for simultane-ous measurement of the free-surface and veloc-ity field, work also continues on techniques to measure the unsteady free-surface in the field. Scanning LiDAR systems have been mounted on-board ships to document the unsteady free-surface and wave breaking (Terrill & Fu, 2008). More recently airborne LiDAR systems have been used to characterize the open ocean wave field and to validate radar based wave measurement systems. Similarly scanning Li-DAR systems have been used in tow tank fa-cilities (Fu et al, 2009), but their uncertainty and the need for surface roughness to provide

sufficient backscatter limits their usefulness in scale model testing. Figures 33-37 show exam-ples of the scanning LiDAR’s capabilities.

Figure 33. Image of the breaking transom wave generated by NSWCCD Model 5673 towed at 7 knots.

Figure 34. Pseudo-coloured time series of the LiDaR signal return amplitude 1.5 m (5 ft) aft of the transom. Model traveling at 7 knots.

4242

Figure 35. Mean transom wave elevation pro-files of NSWCCD Model 5673 traveling at 7 knots.

Figure 36. Contour plot of the mean free-surface elevation of the transom wave gener-ated by NSWCCD Model 5673 at 7 knots.

Figure 37. LiDAR image of the free-surface transom wave from NSWCCD Model 5673 traveling at 8 knots generated by panning the scanning the LiDAR aft at 3 deg/sec.

So the measurement and simulation of un-steady free surface flows remains and active area of research. Along with this work in de-veloping measurement techniques and funda-mental understanding is the long term need for better comprehension these mechanisms on added resistance. That is, our ability to use un-steady surface fluctuations and relate them back to resistance. 7. MODEL MANUFACTURE

The ability to change the geometry of a

physical model has often required a significant extra expense which has restricted the ability to seek optimal hull form solutions at model scale. The development of new manufacturing tech-niques that can provide a cost-effective way forward for investigation of parametric changes to local hull features or appendage arrangement will allow more effective use of towing tank testing for problems where CFD still has lim-ited applicability due to the need to resolve small computational time steps. The area of rapid manufacturing technology is actively

4343

evolving and even during the duration of the 27th ITTC has both reduced in price and in-creased in capability. In the area of its applica-tion to ship models and the obvious area of ship model appendages there is a lack of published data on the accuracy with which models can be generated. Whereas it is evident that complex features, e. g. generation of turbulence trips can be built into the model or recesses for pressure sensors what needs more effort is in metrology of the finished products to assess the influence of the manufacturing technique on the actual accuracy and crucially the surface finish. For use in larger models there are often size limita-tions on the production of components and so models need to be made from many segments which need joining in a precise manner. Not-withstanding these limitations, it is expected that as material costs drop further many more components will be manufactured. A review by Vaezi et al (2013) considers the next generation devices which allow variable material proper-ties and alternative materials to be generated in the same component.

7.1 Rapid Prototyping Technology Rapid prototyping is an extremely im-

portant technology to both the commercial and military sectors. It is quickly becoming a main-stream technology for the production of models to evaluate fit and form or tooling for low vol-ume manufacturing, see Freitag et al (2003) and Nguyen and Vai (2010) for a more com-plete summary of Rapid Prototyping.

The part to be built is first constructed as a solid model in a 3D modeling system and then exported through a file exchange format, typi-cally the STL (Stereo Lithography) format. In an STL file, the surfaces of a model are repre-sented by triangular polygons. Some rapid pro-totyping systems also accept IGES or DXF formats. A rapid prototyping machine recon-

structs the model from the input file and slices it at relatively small increments, which may vary from 1/1000" (0.025mm) to 1/250" (0.1mm). Each layer is built and stacked on top of the previous layer, until the entire model is generated.

Rapid Prototyping Techniques. Stereo Li-thography: With this method, each layer is gen-erated by exposing the surface of a photosensi-tive liquid polymer, contained in a tank, to a laser beam that traces the section. The exposed area solidifies and is lowered by exactly the thickness of the layer. After all the layers have been generated the part is post-cured to harden the material. The size of the model is restricted by the size of the tank.

Laser Sintering. This process uses a laser beam to solidify particles of a powdered mate-rial. After a layer has been exposed, a new lay-er of powder is applied and exposed. The un-exposed powder also functions as a support for extended and free floating parts of the model. This process may use a variety of powder mate-rials, such as PVC, ABS, nylon, polyester, pol-ypropylene, polyurethane, wax, or powdered metals.

Inkjet and 3D Printing. Unlike Laser Sin-tering, the laser is replaced with an inkjet head that deposits a liquid adhesive onto the powder as it translates across the surface. Key ad-vantages of this process are the potential for increased productivity through the application of multiple inkjet heads and the ability to spa-tially introduce a second phase directly as part of the liquid adhesive.

Masking Process. With this method a black toner mask is generated on a glass plate which is the negative image of the layer to be built. A thin layer of liquid polymer is applied to the plate and is exposed to UV light. The un-

4444

masked area solidifies when exposed and is attached to the previous layer.

Fused Deposition Modeling. With this method a thin plastic or wax like wire filament is fed to a moving head, which traces the area of the layer and deposits the filament on the surface. Just before deposition, the wire is heated above its solidifying temperature. Once deposited, the material solidifies and adheres to the previous layer.

Laminated Object Manufacturing. After a thin sheet of paper like material is positioned on a platform, a laser cuts the outline of the layer. The unwanted pieces of the layer are removed before the next sheet is placed on top. The layers are laminated together with a heat sensitive coating.

Cost of Rapid Prototyping. The price of a rapid prototyping machine currently ranges from $5,000 to $500,000. However, a number of service bureaus specialize in building rapid prototyping models and do it at a relatively low cost. 3D Printers are in general use and its cost depends on the time and material used. While the number of polygons that define a part is a minor factor, the volume and layer resolution of a part affect the production time as well as the quantity of material consumed and ulti-mately determine the cost. Small parts can be built relatively cheap but large parts cost is quite high.

Potential Use of Rapid Prototyping in Mod-el Production. In general, paraffin wax, wood, foam and glass reinforced plastics are materials for manufacture of hull models. Wood is still probably the more commonly used. Rapid pro-totyping technology is quite expensive for model manufacturing purpose for today but appendages such as shaft, barrel, rudder and strut could be produced with extremely high precision. In addition, a shaft, barrel and brack-

et system could be manufactured perfectly in single stage using 3D printers as shown in Fig-ure 38. Figure 39 shows installed system to model after painting phase.

Figure 38. Shaft, barrel, strut and stern tube system (3D Printer used).

Figure 39. Installed shaft, barrel, strut and stern tube systems on a model in ITU Ata Nutku Ship Model Basin, Istanbul, Turkey.

4545

7.2 Example of use of rapid prototyping technology in model testing

Limited publications have detailed the use

of rapid prototyped components. A student project by Cope (2012) provide evidence of possible applications. Cope used a fused depo-sition technique and an ABS-M30 plastic to manufacture a 0.17m diameter propeller for a free running model. The printing process lim-ited the minimum trailing edge thickness to 2 mm and required a modification to the scaled propeller thickness distribution. As ABS is not particularly stiff and has a degree of water permeability the propeller was copper-nickel plated with a thickness of 0.1 mm. An assess-ment was made of the increase of blade stiff-ness as shown in Table 13.

Table 13 Relative Stiffness of ABS Model

Scale Propeller. Blade Material Relative Stiffness ABS FDM 1 Cu+Ni Plated ABS 27 Aluminium Alloy 69

8. SIMULATION BASED DESIGN The development in the computational

power available and the relative maturity of the hydrodynamic analysis tools have significantly advanced simulation based design (SBD). The following section focuses on the different ele-ments of SBD and the associated technological developments. These include developments related to global optimisation strategies, multi-objective optimisation, variable fidelity ap-proaches, meta-models and geometry model-ling. For various examples of the practical ap-plication of SBD the reader is referred to the cited literature and the references therein as

well as the recent proceedings of the PRADS, FAST and IMDC conferences. 8.1 Optimization problem

The optimisation problem is commonly

formulated as a nonlinear programming (NLP) problem (Tahara et al, 2011)

uii

li

j

j

x

xxx

qjg

pjh

,...,1,0)(

,...,1,0)(

X)),(,(min M

xx

xxuxf

(17)

where f is a N-dimensional vector of objective functions, x is a vector of design variables be-longing to a subset X of the M-dimensional real space, u is a vector of the state of the system, hj and gj are the equality and inequality con-straints respectively, and the superscripts l and u refer to the lower and upper bounds of a spe-cific design variable respectively.

Objectives. The optimisation algorithm

tries to minimise or maximise the objective function or functions. Various objectives have been used in literature. The hydrodynamic ob-jectives studied include wave making, total and added resistance, propulsion power, wake qual-ity, wake wash and seakeeping merit functions. However, non-hydrodynamic objectives may also be of interest such as objectives related to structural performance, capacity, manufactur-ing or operating costs.

Depending on the number of objectives of

interest the problem is either of single- or multi-objective type. Real-world design prob-lems are associated with several, often conflict-ing, objectives. Thus, there is a growing inter-est in multi-objective optimisation (see e.g. Tahara et al, 2011; Kuhn et al, 2010). A multi-

4646

objective problem can be a multi-disciplinary problem or a multi-point problem. In the for-mer the objectives are related to different disci-plines (e.g. resistance and seakeeping, Tahara et al, 2008, 2011), whereas in the latter the same objective function is evaluated at differ-ent condition (e.g. resistance at several speeds, Kandasamy et al, 2013). The multi-objective problem can be reduced into a single-objective problem through scalarisation, i.e. by forming a single objective as the weighted sum of the multiple objectives (see e.g. Tahara et al, 2011). For the weighting the knowledge of a designer, builder or owner can be used. How-ever, often it is preferred that the Pareto opti-mality of the problem is maintained. For a Pareto optimal solution the improvement in one objective leads to a decline in one or more of the other objectives. Maintaining the Pareto optimality gives the designer a wider choice of optimal solutions and freedom to choose the weighting of the objectives afterwards.

The fundamental problem related to the ob-

jectives is that in ship hydrodynamics they are often expensive to evaluate. Furthermore, the problem has often multi-modal nature, i.e. the objective function has many local optima. These have a great influence on the choice of the optimisation strategy.

Design variables. The design variables dic-

tate the possible changes to be explored in the optimisation process. The choice of correct design variables is fundamental for the quality of the optimal solution. The number of design variables, which determines the dimensions of the search space, should be as low as possible but still allow sufficient flexibility in the design variations. The hull fairness and limitations of manufacturing should also be considered when making the choice. The knowledge of a de-signer can be used to guide the selection of relevant variables, but also to reduce the di-mensions of the search space. Sensitivity stud-

ies can also be used to support the choice of the variables and to determine dominating varia-tions. Recently Proper Orthogonal Decomposi-tion (POD) has been suggested for reduction of the dimensionality of the design space with the majority of the geometric variability retained. Chen et al (2014) have used POD in the opti-misation of the water-jet propelled Delft Cata-maran. They have managed to reduce the 20-dimensional design space into 4 and 6 dimen-sional spaces depending on the constraints and at the same time maintain 95 percent of the geometric variation.

The design variables are also subject to

various constraints. The constraints can be in the form of equality or inequality constraints, they can be linear or nonlinear and they can constrain the design variables directly (e.g. box constraints) or indirectly (e.g. constraint on displacement). Because of the constraints the search space can be non-convex or even dis-continuous. This limits the set of applicable optimisation algorithms. Furthermore, the way in which constraints are taken into account de-pends on the algorithm. This may be based on direct elimination of infeasible solutions, pen-alty formulation by increasing (or decreasing) the objective function, if constraints are vio-lated or explicitly adjusting the search direction to point back into the feasible space.

Operating conditions. In the most common

case in the literature the optimisation is per-formed for a single operating condition. How-ever, a growing trend in the research is the op-timisation for multiple operating conditions (multi-point optimisation). The variables defin-ing the operating condition include for example ship speed, loading condition, water depth and sea state. These should be included in the statement of the optimisation problem. Opera-tional profiles can be used to weight the differ-ent operating conditions.

4747

Deterministic vs. stochastic problem. In the literature the optimisation problem is often considered as a deterministic problem. How-ever, uncertainties in the real world operating (loading, trim, speed) or environmental condi-tions (sea state, wind, water quality) lead to a stochastic problem. In addition, there are vari-ous other sources of uncertainty such as the deviation between the intended and the manu-factured design and modelling and numerical uncertainties in the evaluation of the objective functions. The practical consequence of this is that a design optimised for the expected values of the uncertain operating parameters may not be the real optimum of the stochastic problem.

Recently the stochastic nature of the real-

life problems has gained more attention. Diez et al (2012) discuss the associated idea of ro-bust design optimisation (RDO) extensively and present a RDO framework combining multi-disciplinary analysis and Bayesian deci-sion making. Here the operating scenario is given as a probability distribution and the op-timisation is based on the minimisation of the expectation of the objective function. They demonstrate RDO for the hydroelastic optimi-sation of the efficiency of a fin keel subject to uncertain yaw angle. Even if the design space is limited and the operating scenario is simple the stochastic and deterministic optima are dif-ferent with the robust design showing a better overall performance. 8.2 Simulation based design framework

A SBD toolbox consists of three elements:

(i) generation of a geometry based on the de-sign variables, (ii) evaluation of the objective functions using the given geometry and (iii) optimisation algorithm which modifies the de-sign variables based on the evaluated objec-tives. These steps are iterated until the optimum

has been found or a set number of iterations has been reached.

Geometry modification. The geometry

modification routine takes as input a set of de-sign variables and produces as output a defini-tion of geometry which can be a surface defini-tion or a computational grid. The approaches for geometry modification can be categorised based on how their operation is related to the Computer Aided Design (CAD) systems:

CAD free: works independently of any CAD

system; might work directly on the compu-tational grid

CAD direct control: controls a real CAD system

CAD emulation: emulates the operations that would normally be done in a CAD sys-tem; uses the same geometry entities and file formats to be compatible with a CAD system Various algorithms have been proposed for

the geometry modifications. In some algo-rithms the design variables are directly related to the points on the hull surface. In this case particular care has to be exercised in order to ensure hull fairness. For example, the hull can be modified by multiplying the hull offsets with smooth functions (Tahara et al, 2011; Zhang and Ma, 2011) or by interpolating the displacements using radial basis functions (Kim et al, 2010), where the parameters of the functions are defined by the displacement of a set of hull surface points.

Alternative approaches have been proposed,

where the design variables are independent of the hull surface definition. Two methods show-ing good performance and great flexibility are the geometry morphing (see e.g. Kang and Lee, 2012) and the free form deformation (FFD, see e.g. Tahara et al, 2011). In morphing two or more hull forms are combined into one as a

4848

weighted sum of the parent forms. The number of weights is usually one less than the number of parent hull forms. By using the weights di-rectly as design variables, an optimisation algo-rithm with very low number of design variables is achieved. In FFD, on the other hand, the hull or a part of it is enclosed in a parallelepiped containing a structured set of control points. The parallelepiped is deformed by moving the control points, and the displacement of any point inside it is interpolated based on the dis-placements of the control points (for details see e.g. Tahara et al, 2008). Several parallelepipeds can be combined to perform global and local modifications of the hull form.

Global modification approaches operating

on the parameterisation of the common ship design curves (e.g. sectional area curve, water-line, profile, sections) have also been proposed (see e.g. Kim et al, 2008). Dedicated, fully ana-lytical approaches have been presented for par-ticular hull forms (e.g. rounded bilge boats by Pérez and Clemente, 2011). The benefit of these methods is that there is a direct link to the classical design office practice, and the modifi-cations are easily related to the changes in the established design parameters. Based on the parameterisation it is also possible to formulate a constrained design approach, in which the geometry will automatically satisfy the con-straints set on the main parameters such as buoyancy, longitudinal centre of buoyancy or waterline area (Pérez and Clemente, 2011).

Analysis tools. The analysis tools take as

inputs the modified geometry and the operating conditions and produce values of the objective functions and possible constraints. The level of detail of the methods used varies a lot. The tool set is a compromise affected by for example the complexity of the design problem (number of design variables and objective functions, multi-disciplinary problems), the time and computa-tional resources available and the requirements

on the accuracy. In concept level design the problem is multidisciplinary, the search space is large and the time to find the optimum is very limited. Here simulation is too time con-suming and the tools may be very simple based on design equations, regression data or correla-tion lines (see e.g. Hart and Vlahopoulos, 2010). When simulations can be afforded, po-tential flow based tools provide more accuracy, but are still relatively efficient. The full range of potential flow based methods ranging from thin ship theory to fully nonlinear boundary element methods have been used in SBD (Kim et al, 2010; Zhang and Ma, 2011; Tahara et al, 2011). The most accurate, but also computa-tionally most expensive methods used so far in SBD are mainly based on the Reynolds-Averaged Navier-Stokes (RANS) equations (see e.g. Kim et al, 2008; Tahara et al, 2008, 2011).

Optimisation algorithms. The optimisation

algorithm of the SBD framework works on the values of the objective functions produced by the analysis tools and tries to find a better set of design variables leading to an improved value of the objective or objectives. In gradient based algorithms the optimisation is driven by the gradient of the objective function, whereas gra-dient free algorithms operate without any knowledge of the gradient. The optimisation algorithms can also be categorised into local and global algorithms based on whether they search for local or global optima. The literature cited in this chapter includes various examples of optimisation algorithms that have been used in SBD and comparisons of common algo-rithms (see e.g. Kim et al, 2008; Campana et al, 2009).

The trend has been towards gradient free

global optimisation algorithms. There are sev-eral reasons for this. (i) The gradient evaluation is problematic due to noisy and non-smooth objective functions or due to unavailability of

4949

derivatives, so that local algorithms could be stuck at local minima. (ii) The geometrical and functional constraints required to make the design realistic result often into nonconvex feasible search space. (iii) In several fields ex-perimental and computational activities have helped the designers to produce near optimal designs, so that finding further improvement with local optimisation is difficult. (Campana et al, 2009) However, local algorithms com-plement global algorithms with different ad-vantages such as faster convergence. Therefore, hybrid algorithms combining global and local algorithms have also been proposed. (see e.g. Campana et al, 2009; Peri and Diez, 2013)

The rapid development of parallel comput-

ing has led to the increasing popularity of population based optimisation algorithms. Many of these draw their inspiration from the processes in nature. These include various forms of evolutionary algorithms (EA), such as evolution strategies (ES) and genetic algo-rithms (GA; Tahara et al, 2008; Kim et al, 2010; Zhang and Ma, 2011; Kandasamy et al, 2013), and particle swarm optimisation (PSO; Kim et al, 2008; Campana et al, 2009; Hart and Vlahopoulos, 2010; Diez et al, 2012; Tahara et al, 2011; Kandasamy et al, 2013). In EAs the main idea is to produce successive generations of designs which exhibit improving perform-ance. The main operations between generations are selection, recombination (crossover) and mutation. The differences between the various EA methods lie in the details of these opera-tions and how the operations are combined. In PSO the global optimum is sought for based on an analogy with the behaviour of a flock of birds. Each individual of the swarm explores the search space with a variable velocity. This is affected by the previous velocity (inertia), by the attraction of the best locations so far for the swarm (social factor) and for the individual (cognitive factor). The original PSO formula-tion has additionally randomness included, but

deterministic variants have also been successful applied (Campana et al, 2009).

Regardless of the type of the algorithm, in

multi-modal problems it is essential that there is a balance between the local and exploring characteristics of the algorithm. A good bal-ance leads to a fast convergence of the algo-rithm and avoids premature convergence to a local optimum. The balance can be changed as the solution approaches the global optimum. For example, in PSO the inertia controls the balance between the local and global character-istic. (Campana et al, 2009)

The computational expense of the evalua-

tion is often a problem in optimisation. The computational cost can be reduced by using meta-models, variable fidelity/physics ap-proach or a combination of these. The idea here is that the number of the most accurate and expensive evaluations is reduced by performing the majority of the evaluations with less expen-sive approach. For example, the expensive method is called only, if the less accurate method shows an improvement in the design. A meta-model is an approximation for the behav-iour of the objective function constructed from the function values at a set of sample points. In variable physics approach the low fidelity solu-tion could be based on low cost potential flow solution and the high fidelity solution on a RANS solver (see e.g. Tahara et al, 2008, 2011; Kandasamy et al, 2013). Alternatively a vari-able resolution or iterative accuracy approach could be used. In the former the low and high fidelity solutions are obtained with a coarse and a fine discretisation resolution, respectively. In the latter, the convergence level of the numeri-cal solution is altered between the fidelities. Meta-models based on the known difference between the high and low fidelity solutions at sample design points can be used to improve the low fidelity estimate, and a trust region

5050

methodology can be used to control the fre-quency of high fidelity evaluations.

The effectiveness of the variable physics

approach was demonstrated by Kandasamy et al (2013). They combined a low-fidelity poten-tial flow code and a high-fidelity RANS code for the resistance optimisation of the water-jet propelled Delft Catamaran. With the variable-fidelity approach the overall CPU time dropped to less than half of the high-fidelity approach, and both approaches converged to the same optimum. A further, and more significant, re-duction in computational effort for the same optimisation problem is achieved by Chen et al (2014). They studied the combination of POD for the dimensional reduction of the design space, multiple meta-models and multiple de-terministic PSO variants. The deterministic PSO gave the same optimum as the original stochastic version of PSO, but with just 2% of the computational cost. Compared to Kan-dasamy et al (2013) the proposed approach provided an additional calm-water resistance reduction of 6.6% with 1/10th of the computa-tional cost.

Verification and Validation. In order to

have confidence in a SBD framework the re-sults of the optimisation process should be veri-fied and validated (V&V), i.e. a simulated im-provement should correspond to a real-life im-provement with a sufficient confidence. It has been proposed that the methodology used for the V&V of single run cases can be extended into a systematic procedure for the V&V of the optimised solution. This V&V process consists of three parts and is based on the difference in performance between a parent and optimised designs. (i) The optimised design is numeri-cally verified, if it can be shown that the mag-nitude of simulated improvement is larger than the numerical uncertainty. (ii) The optimised design is experimentally verified, if the magni-tude of the measured improvement is larger

than the experimental uncertainty. (iii) The optimised solution is validated, if the absolute value of the difference between the simulated and measured improvements is less than the combined uncertainty from the simulations and the measurements (Tahara et al, 2008, 2011).

It should be noted that the methodology is

independent of the V&V of the individual solu-tions for the parent and optimal design and only includes the trend. This is in line with the fun-damental goal of the design problem, i.e. to find the optimal design. The absolute values of individual designs can be verified and validated using a single run procedure.

For practical examples of the application of

the V&V methodology the interested reader is referred to Tahara et al (2008, 2011) and Kan-dasamy et al (2013).

9. RECOMMENDATIONS The 27th ITTC Resistance Committee rec-

ommends the following:

Adopt the updated guideline 7.5-02-02-02 Testing and Extrapolation Methods, Gen-eral Guidelines for Uncertainty Analysis in Resistance Towing Tank Tests

Adopt the updated guideline 7.5-02-02-02.1 Testing and Extrapolation Methods, Exam-ple Uncertainty Analysis of Resistance Tests in Towing Tank which effectively re-places the dropped 7.5-02-02-02(2002, rev.01).

Adopt the new guideline 7.5-02-02-02.2 Testing and Extrapolation Methods, Practi-cal Guide: Uncertainty Analysis of Resis-tance Measurement in Routine Tests.

5151

Remove the procedure 7.5-02-02-03 Test-ing and Extrapolation Methods, Resistance, Uncertainty Analysis Spreadsheet for Re-sistance Measurements.

Remove the procedure 7.5-02-02-04 Test-

ing and Extrapolation Methods Resistance, Uncertainty Analysis Spreadsheet for Speed Measurements.

Remove the procedure 7.5-02-02-05 Test-

ing and Extrapolation Methods Resistance, Uncertainty Analysis Spreadsheet for Sink-age and Trim Measurements.

Remove the procedure 7.5-02-02-06 Resis-

tance uncertainty analysis spreadsheet for wave profile measurements.

10. CONCLUSIONS It is the need to increase energy efficiency

of shipping that drives the need to significantly improve our ability to measure resistance.

10.1 State of the Art The state-of-the-art review captures the

most significant developments. There is an increased need to do higher precision resistance measurements and understand trade-off be-tween resistance components. Trim optimisa-tion requires enhanced precision in resistance test e.g. 1%, improvements need to be able to resolve to greater than this accuracy. There is the capability to acquire more data during test motion e.g. wireless sensors, synchronised video, better documentation for CFD valida-tion, and improve standard of reporting of test conditions. There is some new limited valida-tion data, and preparation for the Tokyo 2015 CFD workshop will give validation for a new ship type. It is noted that there is still a lack of

high quality data for high performance craft e.g. planning/hydrofoil craft.

The increasing availability of computational

resources means that mesh resolution is less of an issue, however noting the recent 32 billion cell alters perspective but need to get better handle on 'real' cost of such analysis. Many challenges remain with breaking, bubbly flow, and spray will have an impact for resistance. It may not change the value but alters detail of flow which may have implications for propul-sion etc. This links into need for new R&D – surface roughness, model construc-tion/precision, aim to reduce uncertainty and better understanding.

With regard to procedures/guidelines, it was

decided to eliminate the spreadsheet as they are were based on the AIAA standard and further-more, not now relevant as they were primarily linked to the world wide campaign. The update to ISO GUM was applied as it is fairly straight-forward and should be widely adopted in rou-tine commercial tests. It should be noted that there is still no procedure for recording wave profile. The surface roughness guideline was not changed, but reviewed in the committee report for better understanding.

The worldwide campaign data should be

made available via the new ITTC website. The previous committee has provided an easily used database for additional studies. Further analysis was conducted by the committee and has shown some greater understanding. No new data was received. We suggest an approach for inter tank bias comparison, established a base line by removing 'outliers' and make accessible the whole database via new ITTC website and will provide a searchable spreadsheet for use when looking at all data. A comparison is made with the corresponding data from the CFD analysis from Gothenburg 2010.

5252

For future such campaigns, the double blind, although a good idea but in actuality, was too much of challenge, with limited par-ticipation and may have bias issues with 'changes' to the model during the campaign due to the extreme time scale of the effort.

With respect to surface roughness, limita-

tions from practical systems available and chal-lenge of getting measurements on full scale ships with vast area and looking at fine scale points resolved to height changes of a few mm. Need for new instrumentation systems perhaps.

Measurement of the unsteady free surface is

still very much needed to support the develop-ment of breaking models and the validation of CFD codes. As the measurement techniques capable of characterising the small scale roughness associated with wave breaking are still very much in development, there is no need for a procedure at this time.

Model based manufacturing has definitely

been impacted by the proliferation of rapid prototyping. The questions of whether large high fidelity physical models can be built from multiple pieces and how strength/stiffness are maintained remain to be answered.

Simulation based design has evolved rap-

idly in the past decade. A main driver of the development is the inherent computational cost of the simulations. There is an apparent trend towards hybrid algorithms, which combine analysis methods of varying fidelity. In these methods, the majority of objective function evaluations is performed with low-cost meth-ods (potential flow, surrogate models) and the accuracy of the optimisation is guaranteed with infrequent high-cost evaluations (e.g. RANS). A careful setup of the design problem is re-quired in order to keep the dimensions of the design space to a minimum. At the same time the geometry manipulation methods should be

able to guarantee a smooth geometry. Geome-try morphing and free form deformation have proven to be favoured choices in this respect. Recently, significant reduction in the computa-tional cost has been obtained by using proper orthogonal decomposition to reduce the num-ber of design variables and at the same time keeping nearly all of the geometric variability. As the approaches for deterministic problems start to mature, it is expected that the stochastic nature of the design problems (e.g. variable environment in terms of seastate/wind, opera-tional profile) will gain more attention.

10.2 Potential Tasks for the 28th ITTC Re-sistance Committee

(i) Develop a new procedure for wave pro-

file measurement and wave resistance analysis, uncertainty analysis for extrapolation can then engage possible alternative scaling techniques in a rational way

(ii) Unsteady free surface dynamics is still

an active area for research – and there remains a long term need for better comprehension of added resistance, that is the ability to use sur-face fluctuations and turbulence and relate them to resistance.

(iii) Resolve differences between ISO 4287

and widely used BMT roughness measurement system.

(iv) Propose an approach for tanks to re-

duce/manage their uncertainty as a follow on from the Worldwide Campaign.

(v) Sensitivity study for which areas of the

ship should you be measuring/modifying roughness

5353

11. REFERENCES Andre, M.A. and Bardet, P.M., 2014a, “High

Resolution Simultaneous Multiphase Time-Resolved Particle Image Velocimetry”, 17th Int. Symposium on Applications of Laser Techniques to Fluid Mechanics.

Andre, M.A. and Bardet, P.M., 2014b, “Veloc-

ity Field, Surface Profile and Curvature Resolution of Steep and Short Free Surface Waves”, Experiments in Fluids, in press.

Atlar,M and Wilczynski, L(eds), 2013, Proc. of

3rd Advanced Model Measurement Tech-nology Conference , Gdansk,Poland

Avci, Atakan and Karagoz, Irfan, 2009, “A

Novel Explicit Equation for Friction Factor in Smooth and Rough Pipes”, Journal of Fluids Engineering, Vol. 22.

Beale, K.C., Fu, T.C., Fullerton, A.M., Drazen,

D. and Wyatt, D., 2010, “An Experimental Characterization of Full-Scale Ship Gener-ated Bow Spray,” Proceedings of the 28th Symposium on Naval Hydrodynamics, Pasadena, USA.

Belden, J. and Techet, A.H., 2014, “Simultane-

ous Quantitative Flow Measurement using PIV on Both Sides of the Air-Water Inter-face for Breaking Waves”, Experiments in Fluids, under review.

Belden, J., Truscott, T.T., Axiak, M., and Te-

chet, A.H., 2010, “Three-dimensional Syn-thetic Aperture Particle Imaging Veloci-metry,” Measurement Science and Tech-nology, 21(12), 2010.

Belden, J., Truscott, T.T., Ravela, S., Techet, A.H., 2011, “Three-Dimensional Synthetic Aperture Imaging and Resolution of Multi-Phase Flows”, Proceedings of the ASME-JSME-KSME Joint Fluids Engineering Conference, Hamamatsu, Japan; 24-29 July.

Campana, E.F., Liuzzi, G., Lucidi, S., Peri, D.,

Piccialli, V., and Pinto, A., 2009, “New Global Optimization Methods for Ship De-sign Problems”, Optimization and Engi-neering, Vol. 10, pp. 533-555.

Chen, X., Diez, M., Kandasamy, M., Zhang, Z.,

Campana, E.F., and Stern, F., 2014, “High-Fidelity Global Optimization of Shape De-sign by Dimensionality Reduction, Meta-Models and Deterministic Particle Swarm”, Engineering Optimization, DOI: 10.1080 /0305215X.2014.895340.

Cope, D., 2012, “Design of a Free-running,

1/30th Froude Scaled Model Destroyer for In-situ Hydrodynamic Flow Visualization”, SM Thesis, MIT

Diez, M., Peri, D., Fasano, G., and Campana,

E.F., 2012, “Hydroelastic Optimization of a Keel Fin of a Sailing Boat: a Multidiscipli-nary Robust Formulation for Ship Design”, Structural and Multidisciplinary Optimiza-tion, Vol. 46, pp 613-625.

Dong, R., Katz, J., and Huang, T.T., 1997, “On

the Structure of Bow Waves on a Ship Model”, Journal of Fluid Mechanics. 346. 77-115.

5454

Drazen, D., Beale, K.L.C., Bhushan, S., Fuller-ton, A.M., O’Shea, T., Brucker, K., Dom-mermuth, D., Wyatt, D., Carrica, P., Fu, T.C., Stern, F., 2010, “Comparisons of Model-Scale Experimental Measurements and Computational Predictions for the Transom Wave of a Large-Scale Transom Model,” Proceedings of the 28th Sympo-sium on Naval Hydrodynamics, Pasadena, USA.

Eça, L. and Hoekstra, M., 2010, “Quantifying

Roughness Effects by Ship Viscous Flow Calculations”, Proceedings of the 28th Symposium on Naval Hydrodynamics, Pasadena, USA.

Eça, L. and Hoekstra, M., 2014, “A Procedure

for the Estimation of the Numerical Uncer-tainty of CFD Calculations Based on Grid Refinement Studies”, Journal of Computa-tional Physics, Vol. 262, pp 104–130.

Freitag, D., Wohlers, T., and Philippi, T., 2003,

“Rapid Prototyping: State of the Art,” Manufacturing Technology Information Analysis Center.

Fu, T.C., Fullerton, A.M. and Drazen, D., 2009,

“Free-Surface Measurements in a Tow Tank Using LIDAR,” ASME 2009 Fluids Engineering Summer Meeting (FEDSM2009), Forum on Fluid Measure-ments and Instrumentation, Paper No. FEDSM2009-78464, Vail, Colorado, USA, August 2-6.

Fu, T.C., Fullerton, A.M., Ratcliffe, T.,

Minnick, L., Walker, D., Pence, M.L., and Anderson, K., 2009, “A Detailed Study of Transom Breaking Waves,” Naval Surface Warfare Center Carderock Division, Hydromechanics Directorate R&D Report, NSWCCD-50-TR-2009/025, May.

Fu, T.C., O’Shea, T.T., Judge, C.Q., Dommer-muth, D., Brucker, K., and Wyatt, D.C., 2013, “A Detailed Assessment of Numeri-cal Flow Analysis (NFA) to Predict the Hy-drodynamics of a Deep-V Planing Hull”, International Shipbuilding Progress, Vol. 60, pp 143-169.

Geerts, S.,Van Kerkove, G, Vantorre, M, and

Delefortrie, G, 2011, “Waterline registra-tion using fluorescent lighting”, Advanced Model Measurement Technology for EU Maritime Industry - AMT’11, Newcastle-upon-Tyne.

Guo, B.J., Deng, G.B., and Steen, S., 2013, “Verification and Validation of Numerical Calculation of Ship Resistance and Flow Field of a Large Tanker”, Ships and Off-shore Structures, Vol. 8, pp 3-14.

Hart, C.G., and Vlahopoulos, N., 2010, “An

Integrated Multidisciplinary Particle Swarm Optimization Approach to Conceptual Ship Design”, Structural and Multidisciplinary Optimization, Vol. 41, pp 481–494.

Hino, T., 2012, “Surface Roughness Models for

CFD Simulations”, Conference Proceedings of the Japan Society of Naval Architects and Ocean Engineers, Vol.15.

Huang, F., Yang, C., and Noblesse, F., 2013,

“Numerical Implementation and Validation of the Neumann–Michell Theory of Ship Waves”, European Journal of Mechanics - B/Fluids, Vol. 42, pp 47–68.

ISO, 1997, “Geometrical Product Specifica-

tions (GPS) - Surface Texture: Profile method - Terms, Definitions and Surface Texture Parameters”, ISO 4287:1997.

5555

ITTC, 1987, “Report of the Advisory Council”, Proceedings of the 18th ITTC Full Confer-ence, Kobe, Japan.

ITTC, 1990, “Report of the Panel on Validation

Procedures”, Proceedings of the 18th ITTC Full Conference, Madrid, Spain.

ITTC, 2005, “Final Report and Recommenda-

tions to the 24th ITTC by the Resistance Committee”, Proceedings of the 24th ITTC Full Conference, 4-10 September, Edin-burgh, UK.

ITTC, 2011a, “Report of Specialist Committee

on Surface Treatment”, Proceedings of the 26th International Towing Tank Conference, Rio de Janeiro, Brazil

. ITTC, 2011b, “Report of Resistance Commit-

tee”, Proceedings of the 26th International Towing Tank Conference, Rio de Janeiro, Brazil.

Isaksen, A., McMillan, L., and Gortler. S.J.,

“Dynamically Reparameterized Light Field”, in SIGGRAPH’00: Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques, Num-ber 1-58113-208-5, New York, NY, USA, 2000. ACM Press/Addison-Wesley Publish-ing Co., pp 297–306.

Kandasamy, M., Peri, D., Tahara, Y., Wilson,

W., Miozzi, M., Georgiev, S., Milanov, E., Campana, E.F., and Stern, F., 2013, “Simu-lation Based Design Optimisation of Water-jet Propelled Delft Catamaran”, Interna-tional Shipbuilding Progress, Vol. 60, pp 277-308.

Kang, J-Y., and Lee, B-S., 2012, “Application of Morphing Technique with Mesh-Merging in Rapid Hull Form Generation,” International Journal of Naval Architecture and Ocean Engineering, Vol. 4, pp 228-240.

Katsui, T., Izumi, T., and Ueno, S., 2011, “The

Effects of Surface Roughness on Flat-Plate Frictional Resistance”, Conference Pro-ceedings of the Japan Society of Naval Ar-chitects and Ocean Engineers, Vol.13.

Kawakita, C., 2013, “Study on Marine Propel-

ler Running in Bubbly Flow”, Proceedings of the 3rd International Symposium on Ma-rine Propulsors, Launceston, Tasmania, Australia.

Kawashima, H., Kodama, Y., Hinatsu, M.,

Hori, T., Makino, M., Ohnawa, M., Take-shi, H., Sakoda, M., Kawashima, H., and Matsuno, F., 2007, “A Research Project on Application of Air Bubble Injection to a Full Scale Ship for Drag Reduction”, Pro-ceedings of FEDSM2007 (5th Joint ASME/JSME Fluids Engineering Confer-ence), San Diego, California USA.

Kawashima, H., Makino, M., Fukasawa, R.,

Takeshi, H., Kawaguchi, Y., Tuji, Y., Iwa-moto, K., Motozawa, M., and Masuda, H., Mieno, H., 2012, “Effect of Geometric Roughness Parameters on Turbulent Fric-tional Resistance”, Conference Proceedings of the Japan Society of Naval Architects and Ocean Engineers, Vol.15.

Kiger, K.T., Duncan, J.H., 2012, “Air-

Entrainment Mechanisms in Plunging Jets and Breaking Waves”, Annual Review of Fluid Mechanics, Vol. 44, pp. 563-596.

5656

Kim, H., Yang, C., Kim, H., and Chun, H.H., 2010, “A Combined Local and Global Hull Form Modification Approach for Hydrody-namic Optimisation”, Proceedings of the 28th Symposium on Naval Hydrodynamics, Vol. 2, pp 1219-1232.

Kim, H.J., Chun, H.H., Peri, D., and Campana,

E.F., 2008, “Optimizing using Parametric Modification Functions and Global Optimi-zation Methods,” Proceedings of the 27th Symposium on Naval Hydrodynamics, Vol. 2, pp 1457-1468.

Kuhn, J., Collette, M., Lin, W.-M., Schlageter,

E., Whipple, D., and Wyatt, D., 2010, “In-vestigation of Competition between Objec-tives in Multiobjective Optimization”, Pro-ceedings of the 28th Symposium on Naval Hydrodynamics, Vol. 2, pp 1233-1244.

Larsen, N.L., Simmonsen, C.L, Neilsen, C.K.,

and Holm, C.R., 2012, “Understanding the physics of trim”. Ninth Annual Green Ship Technology Conference, Copenhagen, March 2012.

Larsson, L., Stern, F., Visonneau, M. (Eds.),

2014, Numerical Ship Hydrodynamics: An Assessment of the Gothenburg 2010 Work-shop, Springer Science+Business Media, Dordrecht.

Larsson, L., Visonneau, M., Stern, F., Hino, T.,

Hirata, N., Kim, J., 2014, Communication with the Steering Committee of the CFD Workshop 2015.

Maki, K.J., Broglia, R., Doctors, L.J., Di Mas-

cio, A., 2013, “Numerical Investigation of the Components of Calm-Water Resistance of a Surface-Effect Ship”, Ocean Engineer-ing, Vol. 72, pp 375- 385.

Masnadi, N., Washuta, N., Wang, A., and Dun-can, J.H., 2013, “The Interaction of a Tur-bulent Ship-Hull Boundary Layer and a Free Surface”, presented at the 66th Annual Meeting of the American Physical Society, Division of Fluid Dynamics, November 24-26, Pittsburgh, Pennsylvania, USA http://meetings.aps.org/link/BAPS.2013.DFD.D28.2.

Mieno, H., 2012, “An Experimental Study for

Friction Resistance of Ship Hull Sprayed with Paints”, Conference Proceedings of the Japan Society of Naval Architects and Ocean Engineers, Vol.14, Japan.

Molland, A.F., Turnock, S.R. and Hudson,

D.A., 2011, Ship Resistance and Propul-sion: Practical Estimation of Ship Propul-sive Power, Cambridge University Press.

Molland, A.F., Turnock, S.R., Hudson, D.A.,

and Utama I.K.A.P, 2014, “Reducing ship emissions: a review of potential practical improvements in the propulsive efficiency of future ships.” Transactions of Royal In-stitution of Naval Architects Part A, 156, 175-188.

Nguyen, Huy and Vai, Michael, 2010, “Rapid

Prototyping Technology”, Vol. 18, No 2, Lincoln Laboratory Journal.

Nishikawa, T., Yamade, Y., Sakuma, M., Kato,

C., 2012, “Application of Fully-Resolved Large Eddy Simulation to KVLCC2 – Bare Hull Double Model at Model Ship Rey-nolds Number”, Journal of the Japan Soci-ety of Naval Architects and Ocean Engi-neers, Vol. 16, pp 1-9.

5757

Nishikawa, T., Yamade, Y., Sakuma, M., Kato, C., 2013, “Fully Resolved Large Eddy Simulation as an Alternative to Towing Tank Resistance Tests – 32 Billion Cells Computation on K computer”, Proceedings of the 16th Numerical Towing Tank Sympo-sium, Mülheim, Germany.

Olivieri, A., Pistani,F., Avanzini,A., Stern, F.

and Penna, R., 2001, “Towing Tank Ex-periments of Resistance, Sinkage and Trim, Boundary Layer, Wake, and Free Surface Flow Around a Naval Combatant INSEAN 2340 Model”, IIHR Technical Report No. 421, September, Iowa City, Iowa, USA

Pérez, F. and Clemente, J.A., 2011, “Con-

strained Design of Simple Ship Hulls with Image-Spline Surfaces”, Computer Aided Design, Vol. 43, pp 1829–1840.

Peri, D. and Diez, M., 2013, “Ship Optimiza-

tion by Globally Convergent Modification of PSO by a Surrogate-Based Newton Method”, Engineering Computations, Vol. 30, pp 548-561.

Rodenhuis, G.S., Morgan, W.B. and Davies,

M.E.H., 1987, “Validation and Uncertainty Analysis – An ITTC Activity Session 1: Report of Advisory Council Working Group on Validation”, Proceedings of the 18th ITTC Full Conference, Kobe, Japan.

Sasajima, H., et.al. 1965, “Experimental Inves-

tigation into Roughness of Hull Surface and Increase of Skin Frictional Resistance”, J. of Zosen Kiokai, Vol. 117, pp 58-71.

SIMMAN 2014, DTMB 5415 Geometry and Conditions, http://www.simman2014.dk/.

Stern, F., Longo, J., Penna, R., Olivieri, A., Ratcliffe, T. and Coleman, H., 2000, “Inter-national Collaboration on Benchmark CFD Validation Data for Surface Combatant DTMB Model 5415”, 23rd ONR Sympo-sium on Naval Hydrodynamics, Val de Reuil, France.

Stern, F., Yang, J., Wang, Z., Sadat-Hosseini,

H., Mousaviraad, M., Bhushan, S., Xing, T., 2013, “Computational Ship Hydrody-namics: Nowadays and Way Forward”, In-ternational Shipbuilding Progress, Vol. 60, pp 3-105.

Tahara, Y., Peri, D., Campana, E.F., and Stern,

F., 2011, “Single- and Multi-objective De-sign Optimization of a Fast Multihull Ship: Numerical and Experimental Results”, Journal of Marine Science and Technology, Vol. 16, pp 412–433.

Tahara, Y., Peri, D., Campana, E.F., and Stern,

F., 2008, “Computational Fluid Dynamics-Based Multiobjective Qptimization of a Surface Combatant Using a Global Optimi-zation Method”, Journal of Marine Science and Technology, Vol. 13, pp 95–116.

Takai, T., Kandasamy, M., Stern, F., 2011,

“Verification and Validation Study of URANS Simulations for an Axial Waterjet Propelled Large High-Speed Ship”, Journal of Marine Science and Technology, Vol. 16, pp 434-447.

Takinaci, A.C., Onen, M., 2013, “Low Engine

Power-Less Emission”, Atmos 2013, 6th Atmospheric Science Symposium, İstanbul Turkey, pp 193-200.

5858

Tanaka, H., Toda, Y., Higo, K., and Yamashita, K., 2003, “Influence of Surface Properties of Coatings to Frictional Resistance”, Jour-nal of the Kansai Society of Naval Archi-tects, Vol. 239, Japan.

Taravella, B.M., and Vorus, W.S., 2012, “A

Wave Resistance Formulation for Slender Bodies at Moderate to High Speeds”, Jour-nal of Ship Research, Vol. 56, pp 207-214.

Terrill, E.J., and Fu, T.C., 2008, “At-Sea

Measurements for Ship Hydromechanics”, Proceedings of the 27th Symposium on Na-val Hydrodynamics, Seoul, Korea, October 5-10.

Vaezi, M, Chianrabutra, S, Mellor, B and

Yang, S., 2013, “Multiple material additive manufacturing – Part 1: a review”. Virtual and Physical Prototyping, 8, (1), 19-50

Wang, D., Dai, D., Liu, X., and Duncan, J.H.,

2012, “An Experimental Study of Droplets Produced by Plunging Breakers”, presented at the 65th Annual Meeting of the Ameri-can Physical Society, Division of Fluid Dy-namics, Nov. 18-20, San Diego, CA, USA, http://meetings.aps.org/link/BAPS.2012.DFD.D28.3.

Waniewski, T.A. 1998. “Air Entrainment by

Bow Waves”, Ph.D. thesis, California Insti-tute of Technology, October.

Weinell, C.E., Olsen, K.N., and Christoffersen

M.W., 2003, “Experimental Study of Drag Resistance Using a Laboratory Scale Rotary Set-Up”, Biofouling, International Congress on Marine Corrosion and Fouling, No.11, San Diego, California, USA.

White, F.M., 1991, “Viscous Fluid Flow,” 2nd Edition, McGrawHill.

Winden, B., Turnock, S.R., and Hudson, D.A.,

2013, “Predicting powering performance changes for ships in offshore conditions from small design modifications,” Twenty-Third International Offshore and Polar En-gineering Conference, Anchorage, Alaska, USA.

Winkel, E.S., Cutbirth, J.M., Ceccio, S.L., Per-

lin, M., and Dowling, D.R. 2012, “Turbu-lence Profiles from a Smooth Flat-Plate Turbulent Boundary Layer at High Rey-nolds Number,” Experimental Thermal and Fluid Science, Vol. 40, pp 140–149.

Yan, H., Liu, Y., 2011, “An Efficient High-

Order Boundary Element Method for Nonlinear Wave-Wave and Wave-Body In-teractions”, Journal of Computational Phys-ics, Vol. 230, pp 402-424.

Zhang, B.-J. and Ma, K., 2011, “Study on Hull

Form Optimization for Minimum Resis-tance Based on Niche Genetic Algorithms”, Journal of Ship Production and Design, Vol. 27, pp 162-168.

Zou, L and Larsson, L, 2014, “A verification

and validation study based on resistance submissions”, Numerical Ship Hydrody-namics: An Assessment of the Gothenburg 2010 Workshop, Springer Sci-ence+Business Media, Dordrecht.

5959

Propulsion Committee

Final Report and Recommendations to the 27th ITTC

1 INTRODUCTION

1.1 Membership and Meetings The members of the Propulsion Commit-

tee of the 27th International Towing Tank Conference are as follows:

• Dr. Didier FRECHOU (Chairman), DGA

Hydrodynamics, France • Tom DINHAM-PEREN (Secretary), BMT

Defence Services Ltd, U.K. • Rainer GRABERT, Schiffbau-

Versuchsanstalt Potsdam GmbH (SVA) Germany

• Valery BORUSEVICH, Krylov State Re-search Center, Russia

• Professor Chen-Jun YANG, Shanghai Jiao Tong University, China

• Professor Emin KORKUT, Istanbul Tech-nical University, Turkey

• Professor Steven CECCIO, University of Michigan, USA

• Takuya OHMORI, Japan Marine United Corporation, Japan

• Professor Moon Chan KIM. Pusan Uni-versity, Korea

Five Committee meetings were held as follows:

• DGA Hydrodynamics, France, 30-31

January 2012 • Krylov Institute, Russia, 8-9-10 October

2012 • Pusan University, Korea, 22nd and 23rd

January 2013 • University of Michigan, USA, 23-25 Oc-

tober, 2013. • BMT Defence Services Ltd, UK, 13-14

March 2014.

1.2 Recommendations of the 26th ITTC The 26th ITTC recommended the follow-

ing tasks for the 27th ITTC Propulsion Com-mittee:

1. Provide an update of the state-of-the-art

for predicting propulsion systems emphasing developments since the 2011 ITTC Conference. The committee report includes discussions of the following top-ics:

a. The potential impact of new techno-

logical developments on the ITTC, in-cluding new types of propulsors, azi-muthing thrusters, and propulsors with flexible blades.

6060

b. New experimental techniques and ex-trapolation methods.

c. New benchmark data. d. The practical applications of computa-

tional methods to the propulsion sys-tems predictions and scaling.

e. New developments of experimental and CFD methods applicable to the prediction of cavitation.

f. The need for R&D for improving methods of model experiments, nu-merical modeling and full-scale meas-urements.

g. A review of new developments regard-ing high-speed marine vehicles

2. Provide a review of ITTC Recommended

Procedures relevant to propulsion. The committee report specifically discusses the following topics:

a. Identification of needed changes in

procedures the light of current practice, and, if approved by the Advisory Council, provision of updated re-quirements.

b. Identification of any needed new pro-cedures, including an outline of their purpose and content.

3. Liaise with the Specialist Committee on

Performance of Ships in Service, espe-cially regarding power prediction and consequences of EEDI.

4. Assess where CFD results can be intro-duced to support experimental model testing by monitoring status of CFD to perform full scale powering, resistance, cavitation and wake simulations and their correlation with full scale data. Identify the needs for hybrid procedures combin-ing experimental and numerical methods.

5. Prepare a state-of-the-art review of model-ling and scaling unconventional propul-sion and wake improving devices.

6. Examine methods of target wake simula-tion, e.g. the “smart dummy” approach.

7. Examine wake fraction scaling for twin-screw ships, and show the consequences on existing procedures.

8. Examine the possibilities of CFD methods regarding scaling of conventional and un-conventional propeller open water data, including initiation of a comparative CFD-calculation project.

9. Develop guidelines for hybrid propulsor testing.

10. Continue monitoring existing full-scale data for podded propulsion, if such data is available.

1.3 General Remarks

All the tasks outlined in the terms of refer-

ence were taken in charge by the present committee. The committee had some difficul-ties liaising with other committees concerning Task 3 and Task 4. The portion of this report regarding procedural reviews has been re-cently reported to the AC, which recom-mended that the procedures be a continuing consideration of the next committee. Con-cerning the CFD comparative benchmark, a joint effort of the Propulsion committee and the CFD committee will continue to gather contributions from all the ITTC organisations.

6161

2 STATE OF THE ART UPDATE

Many major international conferences were held since the 26th ITTC conference in 2011:

• 9th Symposium on Particle Image

Velocimetry, 21-23 July 2011, Kobe. • ICOMIA’s 1st International Hybrid Ma-

rine Propulsion Conference, November 2011, The RAI, Amsterdam.

• SMP11 International Symposium on Ma-rine Propulsors and Workshop, June 2011, Hamburg.

• IWSH 2011: 7th International Workshop on Ship Hydrodynamics 16-19 September 2011, Shanghai.

• MARINE 2011- IV International Confer-ence on computational methods in marine Engineering, 28-30 September, Lisbon.

• IMDC 2012-11th International Marine Design Conference, June 2012, Glasgow.

• ICHD 2012- The 10th International Con-ference on Hydrodynamics, 1 - 4 October, 2012, St Petersburg.

• Voith Hydrodynamic conference, June 2012.

• CAV2012- 8th Symposium on Cavitation, 14-16 August, 2012, Singapore.

• ICMT 2012- International Conference on Maritime Technology, 25-28 June 2012, Harbin.

• ONR 29th symposium on Naval Hydro-dynamics, 24 August 2012, Goteborg.

• Journées de l’Hydrodynamique 2012, 21-22-23 Nov 2012, Paris.

• NAV’2012 - 17th International confer-ence on Ships and Shipping Research 17- 19 October 2012, Naples.

• ICETECH 2012, International Confer-ence and Exhibition on Performance of

Ships and Structures in Ice, September 17-20, 2012, Banff,

• 13th Propeller/Shafting Symposium Sep-tember 11 – 12, 2012, Norfolk.

• ONR Naval S&T Partnership Conference event, October 22-24, 2012, Washington D. C.

• IWSH’2011, The 7th International Work-shop on Ship Hydrodynamics, 16-19 Sep-tember, 2011, Shanghai.

• ISOPE 2012 Conference: 22nd interna-tional Ocean and Polar Eng, 17-23 June, Rhodes.

• HIPER, 28-29 Sept 2012, Duisburg. • ISOPE 2013 Anchorage Conference:

22nd international Ocean and Polar Eng, 30 June – 4 July, Anchorage.

• PRADS 2013: The 12th International Symposium on Practical Design of Ships and Other Floating Structures, 20-25 Oc-tober 2013, Changwon.

• FAST 2013, 12th International Confer-ence on Fast Sea Transportation, 2-5 Dec 2013, Amsterdam.

• AMT 2013, The 3rd International Con-ference on Advanced Model Measure-ment Technology for the EU Maritime Industry, 17-19 September 2013, Gdansk.

• OMAE 2013, The 32nd International Conference on Ocean, Offshore and Arc-tic Engineering, June 9 to 14, 2013, Nantes.

• IWSH 2013: The 8th International Work-shop on Ship Hydrodynamics, 23- 25 September, 2013, Seoul.

• SMP’13, The Third International Sympo-sium on Marine Propulsors, 5 – 8 May, 2013, Launceston. The most relevant papers from these con-

ferences and from other technical journals and conferences were reviewed and reported.

6262

2.1 New technological developments

2.1.1 New types of propulsors There is still a tremendous interest con-

cerning Contra-Rotating Propeller (CRP) concepts based on combination of conven-tional propellers and Pods, either on single or twin shafts (see examples on figures 1, 2, 3 & 4)

Figure 1: The first ferry with a podded CRP

propulsion system (Ueda et al., 2004)

Figure 2: CRP Combination of a Rudder Pod

unit and a single propeller (Sánchez-Caja, et al., 2013)

Figure 3: The main propeller (right) with counter rotating, 360 degree azimuthing, ABB Azipod thruster on 200 TEU Container Feed-

er vessel (Henderson, 2013) One advantage of a CRP system when

compared to a single propeller is that, as the two propellers of the CRP share the total pro-pulsive force, the load on a single propeller is reduced, allowing for a reduction in rotation speed. Thus, increased propulsion efficiency can be obtained compared to a single propel-ler of the same diameter.

Figure 4: CRP Electric Propulsion system

(Hideki, et al. 2011) As pointed out by Hideki et al. (2011), in

an electric propulsion vessel, there is no need to connect a large main engine directly to the propeller shaft. Instead, two electric propul-sion motors much smaller in size than the main engine are connected to the propeller shaft through a CR gear. Since the electric propulsion devices are connected via electri-cal buses, the arrangement in the engine room is more flexible than in conventional vessels (Figure 5).

6363

Figure 5: Flexible engine room arrangement

and development of hull form (Hideki et al., 2011)

Therefore, the hull form from midships to

the stern, which is important to reducing fluid resistance, is can be improved compared to vessels with conventional propulsion (Fig. 6).

Figure 6: Comparison of fuel consumption

between conventional vessel and electric pro-pulsion vessel with IHIMU-CEPS (1,230 m3

type chemical tanker used as an example) (Hideki et al., 2011)

Most cargo vessels have only a single

propeller directly driven by a diesel engine. Single screw propulsion offers an efficient hull form. The resistance is lower and the hull efficiency is high owing to the beneficial wake field behind the skeg. This configura-tion provides the most cost efficient solution for most cargo vessels with modest power,

large drafts and little demands on manoeu-vring performance and redundancy.

Full displacement ferries, on the other

hand, usually have twin screws and multi-engine machinery. There have of course been good reasons for these trends. Ferries are of-ten faster and require increased propulsive power. Their draft is often limited, and the propeller loads becomes higher. These factors favour twin-screw solutions, where the power can be divided between two propellers. Safety aspects and fast turnaround in port favours two propellers.

A range of new propulsion concepts for

ferries have been presented in recent years, such as Podded CRP, Wing Pods and Wing Thrusters. These have some features in com-mon in that they do not use a traditional twin shaft line arrangement but instead employ a propeller mounted on the centreline skeg combined with either one or two azimuthing propulsors.

Several recent papers reveal an increasing

interest in Energy saving devices before or after the propeller or within the propeller it-self. The review of new developments on that topic is largely detailed in Section 6.

A few projects using immersed pump-jet

or water-jet have also been published. Pospiech (2012) presented a design of a pump-jet fully integrated with the ship hull (Figure 7). Giles et al. (2011) presented a de-sign of water-jet fully immersed and also fully integrated within the ship hull (Figure 8).

6464

Figure 7: VOITH’s New Propulsion System:

The Voith Linear Jet (Pospiech, 2012)

Figure 8 : WaterJet: Propulsor

(Giles, et al. 2011)

Although the Hybrid sailing vessel is still at a research and development stage, expected fuel energy savings are very promising. Using 9 rigid sails on a cap-size bulker of 180,000DW, Ouchi et al. (2013) forecast a fuel energy savings of at least 20%. CFD was used to estimate the thrust distribution on eve-ry sail for different apparent wind angles.

Figure 9: Hybrid sailing vessel

(Ouchi, et al. 2013)

Figure 10: Thrust force distribution on sails

(Ouchi, et al. 2013) The potential impact of these new

propulsors are listed below: • New procedures are required for self-

propulsion test of contra-rotating propul-sion system and for pump-jets that are in-tegrated within the hull

• CFD calculation might be required to support EFD to assess the performance of Energy Saving Devices.

• A new procedure for self-propulsion will certainly be required for hybrid sailing vessels.

2.1.2 Azimuthing thrusters Two papers (Palm, et al., 2011; Koushan,

et al., 2011) have shown interests on the effect of ventilation on azimuthing thruster perform-ances.

Palm et al. (2011) present a comparative

study between cycloidal propeller and azi-muthing thruster, investigating the effect ven-

6565

tilation on the thrust losses. The blade thrust force of an azimuthing thruster is subject to large variations when ventilation is occurring. Due to its working principle, the cycloidal propeller is less prone to ventilation than the azimuthing thruster.

Figure 11: Cycloidal propeller and azimuthing

thruster at ventilation conditions (Palm et al., 2011)

Koushan et al., 2011 present a similar

study on ventilated propeller blade loadings and spindle moment of a thruster in calm wa-ter and waves. Experimental results are pre-sented for the thrust, torque, and spindle mo-ment of a single blade of a propeller from a pulling thruster under various ventilated oper-ating conditions and in waves.

Figure 12: Azipull Thruster

(Koushan et al., 2011) For all ventilated conditions, it can be ob-

served that a sudden drop in thrust is meas-ured when the advance coefficient becomes less than a critical advance coefficient, which is J ≈ 0.6 for calm water and J ≈ 0.5 for wave conditions. From the critical advance coeffi-cient down to bollard condition, further reduc-tion of the thrust is occurs, though slight thrust recovery is registered close to the bol-lard condition (J = 0). Dynamic variations are analysed using standard deviation and histo-grams. As histograms approximate the prob-ability distribution, they show that the stan-dard deviation values should be handled with some care as the data shows distributions that can be both highly skewed and non-Gaussian. The effect of waves and ventilation on propel-ler torque follows the same trends as on pro-peller thrust. It is observed that the spindle moment changes sign from positive to nega-tive at high J values.

Amini & Steen (2011) performed a series

of model tests on an azimuth thruster model in oblique inflow conditions for different heading angles and at different advance coef-

6666

ficients in pushing and pulling modes. Tests were performed in ventilating and non-ventilating conditions. A novel shaft dyna-mometer was used to measure all six compo-nent forces and moments on the propeller shaft. It was found that the propeller shaft lat-eral force and bending moment were quite large, and thus, the load at the shaft-bearing positions was about three times larger than when only the propeller weight was consid-ered. The results also showed that oblique in-flow due to steering gives higher bending loads than when the propeller is subject to ventilation in the straight-ahead condition. A basic blade element momentum method (BEMT) was used to predict the forces and moments on the propeller shaft in oblique flow conditions. Fairly good agreement was found between the BEMT results and the ex-perimental results.

The authors finally recommend consider-

ing the shaft side forces and bending moments due to steering and oblique inflow in the me-chanical design of the propeller suspension such as thruster housing and propeller shaft bearings.

2.1.3 Flexible blade propulsors

Composite marine structures are attractive because of their ability to conserve weight, reduce maintenance cost, and improve per-formance via 3-D passive hydroelastic tailor-ing of the load-dependent deformations. Manufacturers are proposing carbon fibre propeller of diameter up to 3m.

Several attempts have been made to manu-

facture full size composite propellers and some trials have been conducted in the recent past to compare composite structure and Nickel Aluminium Bronze (NAB) casting.

In 2000, QinetiQ investigated a 2.9 m di-

ameter composite propeller on the Research Vessel Triton, a triple hull warship, nowadays

used as a patrol vessel by the Australian Cus-toms. This composite propeller consists of five composite blades bolted and bonded to a NAB hub. As mentioned by Kane (2001), it was designed to explore the mechanical prop-erties required in this application include me-chanical performance (stiffness, strength and fatigue) as well cavitation inception speed, reported to be 30% higher than the original NAB propeller.

In another example, Airborne Composites

successfully developed composite propeller blades for the Royal Dutch Navy, supplying them with a composite main propeller for an Alkmaar-class mine hunter (Figure 13). This propeller is for a power of about 1400 kW and has a diameter of 2.5m (Black, 2011).

Figure 13: Composite propeller to the RNLN minehunter. (Black, 2011)

Few experiments have been performed at

model scale, one example is Taketani et al., 2013 where different propeller materials have been tested (Figure 14). The results show that the propeller “C” (sintered nylon powders with a laser heating source) presents larger blade deviations than carbon composite mate-rial. For a same propeller loading (Kt), the ad-vance ratio J is significantly reduced (Figure 15).

6767

Figure 14: Model composite propeller

(Taketani et al., 2013)

Figure 15: Model composite propeller per-

formances (Taketani et al., 2013) Propeller efficiency increased in cases

where this deformation was small (Dry Car-bon), since the loss of torque was greater than the loss in thrust. At a certain point, propeller efficiency begins to decline with greater de-formation, suggesting an optimal level of de-formation. Elastic deformation was dominant at the blade tip. This deformation occurred along the direction of thrust and worked as a forward rake.

The cavitation generated after deformation

indicated that such deformation reduced loads at the blade tip and affected pitch angles. This deformation is expected a reduction of pres-sure fluctuations.

6868

In Fluid-Structure Interaction (FSI) analy-

sis, calculations results and model test results were compared. In the case of small deforma-tions, analysis results were consistent with changes in propeller characteristics. For larger deformations, analysis proved relatively inac-curate in estimating the deformation of the entire propeller. Future efforts should target improvements in this aspect.

Manudha, et al. (2013) presented a valida-

tion study that compares results obtained nu-merically using Fluid-Structure Interaction of Finite Element Analysis and experimental re-sults. This validation has been carried on a twisted-bend-twisted coupled hydrofoil. Al-though several simplifications were made for modelling purposes, the consistency between Finite Element Analysis and experimental re-sults were found in good agreement. The knowledge gained through this validation study is extremely helpful in developing an optimisation scheme and an accompanying numerical model that can accurately predict the performance of optimised designs without the need for extensive experimentations.

Extensive studies have been made by

Young (Young, 2007; Young, 2010; Young, 2012; Motley & Young 2012) on flexible blade propellers. Among all those studies, the impact on similarity to be applied for model tests on flexible blade propellers (Figure 16) in order to scale the fluid structure response is of major interest.

Figure 16: Elastic blade deformation on a

composite propeller (Young, 2012). Young (2010) presents a detailed analysis

of the dynamic hydroelastic scaling of self-adaptive composite marine rotors. The scaling analysis main goal is to define how to achieve the same dynamic load-deformation responses between the model and the prototype. This can be achieved by requiring the model to be geometrically similar to the prototype, by re-quiring the effective structural mass and structural rigidities to be the same and by re-quiring the same flow velocity as at full scale:

FSM VV = The following is a simpler way of present-

ing the implication of the similarity laws. When we have to consider testing at model scale with flexible blades, the strain should be kept the same between model scale and full scale. This is to ensure that the displacement, induced by the elasticity of the blade and which changes the angle of attack, will be scaled between model and full scale (Figure 17).

6969

Rigid material

flexible material

VS

Full scale

Model scale

Rigid material

flexible material

Vm

cS

∆cS

∆cm

cm

Figure 17: Strain at full scale and model scale. This is equivalent to say that the strain on

the blade section, that could be defined as the

ratio cc∆

=ε as shown on the figure, should

be the same at model scale and full scale. The strain is related to the stress σ by the modulus of elasticity E:

Ecc σε =

∆=

c : chord length ∆c : displacement E : is the modulus of elasticity σ : is the stress in the material On the other hand, the stress is a function

of hydrodynamic forces and centrifugal forces which means that we can write:

( )

22

23

2

222

2

21

,

DVF

DDF

RmmF

DVFSVF

FFf

proplCentrifuga

proplCentrifuga

propproplCentrifuga

HydroHydro

lCentrifugaHydro

⋅⋅↔⇔

⋅⋅⋅↔⇔

⋅⋅=⋅↔

⋅⋅↔⇔⋅⋅↔

=

ρ

ωρ

ωγ

ρρ

σ

To be more accurate, added mass in addi-

tion to the mass of the blade should be taken into account. But the blade mass as well as the fluid added mass can both be scaled by a factor.

The similarity between model scale and

full scale implies that the same ratio of hy-drodynamic force to centrifugal forces should be kept the same at model and full scale which means the ratio of water density to propeller density should be kept the same:

propprop DVDV

ρρ

ρρ

=⋅⋅

⋅⋅22

22

Because stress is homogeneous to a pres-

sure, we can write that the stress between model scale and full scale is:

22SS

S

MM

M

VV ⋅=

⋅ ρσ

ρσ

In order to get the same kind of strain on

the blades as at full scale, this leads to the fol-lowing relationship :

m

m

S

SmS EE

σσεε =⇔=

Combining all those similarity rules, we

find that:

7070

SM

SM

SpropMprop

SpropMprop

SS

S

MM

M

M

M

S

SMS

VVEE

VV

EE

===

⇒

=

⋅=

⋅

=⇔=ρρ

ρρ

ρρ

ρσ

ρσ

σσεε

22

This demonstrates that with same blade

material at full scale and with the same full scale speed, the strains and loading will be the same.

However, it is problematic to run model

test following these requirements, as the simi-larity conditions are very difficult to achieve on the model scale for the followings reasons:

1. Manufacture of the model scale propeller

(e.g. 250 mm in diameter) is required to have the same isotropic material proper-ties as that of the full scale (e.g. 2.5 m) propeller; this is difficult to achieve, es-pecially with composite carbon fibre ma-terials at the root and the tip.

2. Testing at full-scale speeds necessitates performing the test at very high static pressures in order to avoid any cavitation. In practice, those conditions cannot be achieved in a towing tank, but they can be achieved in cavitation tunnel.

It seems more reasonable to use CFD

which might use a combination of fluid and structural modelling (Young, 2007; He et al., 2012). Meanwhile special care should be taken for the composite structural characteris-tics, for, as pointed out by Young & Motley (2011), the variations in material parameters and material failure initiation models lead to a much wider spread of propeller performance characteristics, operating conditions, and safe operating envelopes for an adaptive propeller compared to a rigid propeller.

2.1.4 Podded propeller in Ice and bubbly

flow Due to the growing interest of a potential

new northern route induced by the global warming, several studies have been carried on propeller ice blade load impact (Brouwer et al., 2013; Sampson et al., 2013), propeller wash (Ferrieri et al., 2013), and cavitation (Sampson & Atlar, 2013).

Figure 18: Picture from a high-speed cam-era of a propeller entering an ice ridge

(Brouwer et al., 2013).

7171

Figure 19: Propeller Thrust, Torque reduc-

tion as a function of void fraction (Kawakita, 2013)

Brouwer et al. (2013) developed a meas-

urement setup in which a 6 components bal-ance is measuring the 6 degrees of freedom forces and moments on a single blade of the podded propeller. This measurement setup was used to measure the ice impact in a model test at the AARC facility of Helsinki.

A few papers (Kawakita et al. 2011, and

Kawakita, 2013) have discussed the effect of air lubrication of a ship hull on propeller effi-ciency. In recent years, air lubrication systems have been attracting attention as a method of reducing carbon dioxide emissions from ships by reducing the total resistance of the ship. The bubbly flow generated travels to the stern such that the propeller may partially work in two-phase flow modifying the thrust and torque (Figure 19).

2.2 New Experimental Techniques and

Extrapolation Methods

2.2.1 3D flow visualization Only one paper dealing with 3D flow in-

vestigations around the propeller caught the attention of the committee. Pecoraro et al. (2013), present a 3D flow velocity measure-ments, using the LDV technique. The major outcome of the analysis is that the effect of the propeller suction, which increases the ve-locity, extends upstream at a distance about 1 propeller radius, and that the flow fluctuation induced by the blade passage extends up-stream to a distance of about 4 propeller radii. The propeller is able to reduce the size of the detached area longitudinally and transversally but is not able to remove totally the flow separation. The boundary of the separated flow can be identified by using the skewness coefficient, which allows a better identifica-tion of the extension of the separated flow.

7272

Figure 20: Axial velocity at R/R=0.7 with

and without the propeller (Pecoraro et al., 2013)

Manufacture / control of Model Propellers

For measurement of blade geometry, Dang et al. (2012) are presenting a new optical technique based on digital photogrammetry. In order to have control on the accuracy of the blades, the propeller is optically scanned at its design pitch. The results are compared to the theoretical geometry and the deviations are determined and presented in 3D images of the model propeller (Figure 21).

2.2.2 Propeller manufacturing

Only one paper concerning new tech-niques for manufacturing model propeller, (Taketani et al., 2013) was found. A model propeller was manufactured by sintering ny-lon powders with a laser and was then com-pared with aluminum made propeller and car-bon composite propeller (see Figure 14). The-

se results of this work are discussed in Sec-tion 2.1.3.

Figure 21: Optical scan on a CP propeller

(deviation from its theoretical geometry, pres-sure side –suction side) (Dang et al., 2012)

2.2.3 Non-stationary blade force meas-

urements Non-stationary blade forces measurement

on propellers at model scale is still a challeng-ing issue. Funeno et al. (2013) developed a blade spindle torque sensor built in the pro-peller hub (Figure 22) to measure blade torque of a controllable pitch propeller operat-ing in off-design conditions and high propel-ler loading.

Just to mention that there is an increase in-

terest (DNV rules, 2010, 27th ITTC Specialist Committee on Hydrodynamic Noise) for this topic that might have some impact on propulsor design, on the procedure to measure propulsor radiated noise, on the prediction of the cavitation inception point, because cavita-tion is largely increasing the radiated noise of a propeller (Briancon et al., 2013; Bosshers et al., 2013).

7373

Figure 22: Propeller shaft thrust and

torque sensor and the blade spindle sensor (Hagesteijn, 2012; Funeno, 2013)

2.3 New benchmark data A sensitivity study of the testing parame-

ters for Propeller alone and the Podded propulsor open water test have been carried out by CTO (Głodowski et al., 2013). The paper is a summary of the large test campaign that was performed within the framework of the Hydro-Testing Alliance Network of Ex-cellence, Joint Research Program 4 (JRP4), based on the so-call ABB case. Before that, a first benchmark had shown large discrepan-cies of about 5.9% even for the POD propeller open water test (Veikonheimo, 2006). Then a second benchmark testing program has been launched through the Hydro Testing Alliance (HTA) European Project to standardize the testing procedure in order to understand the causes the discrepancies found in the first test and to define recommendations for the testing procedure/setup of Podded propulsors. It was determined that using a same propeller model, a same POD housing, the same aft fairing for the propeller open water test, reduces the dis-crepancies of the results between the different facilities.

The final conclusions of this benchmark-

ing test program were in line with the recom-

mendations given in the 7.5-02-03-01.3 Pro-pulsion, Performance Podded Propulsion Tests and Extrapolation. The authors recom-mend having a aft fairing cone to rotate with the propeller and having a separate pre-test with a dummy hub to correct with the propel-ler open water test results which is a first al-ternative recommended in the 7.5-02-03-01.3 Propulsion, Performance Podded Propulsion Tests and Extrapolation.

2.4 Application of computational meth-ods

With respect to the propulsive perform-

ance, the major interest is still in developing and applying CFD (mainly RANS) models for self-propulsion simulation at model scale, in-cluding different approaches to extract the effective wake field. Meanwhile, such simula-tion at full scale began to appear, which pro-vides a new perspective for studying the Rey-nolds scale effects. On the other hand, there is a pronounced increase in efforts devoted to the research of scale effects on energy saving devices, such as the pre-propeller stators/fins and ducts, the CLT propeller and the PBCF, and on multi-component propulsors, such as ducted, contra-rotating, and podded propellers.

2.4.1 Self-propulsion and effective wake

field In Castro, et al. (2011) the feasibility of

self-propulsion simulation at full scale was demonstrated for the KCS, using a DES model and a dynamic overset approach. The propulsion factors were analyzed from simu-lated full scale resistance, open water, and self-propulsion performances, and compared with those obtained from model scale simula-tions and experiments. The SFC based on EFD and ITTC extrapolation procedure was larger than that based on model and full scale computation results. The computed full scale open water thrust was close to, while the

7474

torque was slightly lower than the model scale EFD data. The simulation results of full scale self-propulsion agreed well with EFD data, except for the torque coefficient (and hence the relative rotative efficiency). Through comparison of simulated stern flows, it was shown that the propeller working behind the hull experiences an inflow of higher axial ve-locity and uniformity due to the thinner boundary layer at full scale than at model scale, which results in favourable effects on propeller and propulsive efficiencies, axial loading fluctuations, as well as the level of bending moment around horizontal axis.

Figure 23: The overset grid system for

KCS (Castro, et al., 2011) Villa, et al. (2012) presented a viscous/

inviscid coupled approach for the simulation of self-propulsion based on a RANS and an unsteady BEM solver. Extraction of the effec-tive wake field was made at a plane 0.2D up-stream of propeller blades by using the time-averaged induced velocities computed by the BEM code. Simulations were conducted for the KCS propelled by KP505 at model scale, Fn = 0.26. It was shown that the iterative ap-proach converged fast and the CFD and EFD results of total resistance, propeller rotation speed, and the velocities at a section in pro-peller slipstream correlated well. In addition the results from an actuator disk model (hav-

ing axial force only) were also presented to compare with those from the more accurate BEM-based body force model.

Sakamoto, et al. (2013a) presented re-

search on a RANS simulation of the resis-tance, open-water, and self-propulsion per-formances for a twin-skeg container ship at model scale, together with towing tank ex-periments. An in-house FVM solver and the Spalart-Allmaras one-equation turbulence model were used. The propeller was modelled by body force distributions computed by a simplified propeller theory. By using three sets of block-structured grids having a re-finement ratio of 2 , the uncertainty analysis for resistance and self-propulsion coefficients was conducted with the V&V method rec-ommended by the 25th ITTC. It was con-cluded that the CFD solver was capable of predicting the resistance and self-propulsion performances for the low L/B twin-skeg ship, though it could be improved by implementing real-geometry propeller computations.

In Rijpkema, et al. (2013) different ways

to extract the effective wake field and their influences on predicted propeller performance behind the hull were studied. Two in-house RANS solvers coupled with a BEM propeller code were used in a comparative investigation of the self-propulsion computations for KCS. The body force field was imposed at the blade positions (instead of the propeller disk posi-tion) by interpolation of BEM output. The ef-fective wake field was obtained by subtract-ing the time-averaged propeller induced ve-locities computed by BEM from the RANS-computed velocity field of the hull-propeller system. The numerical results indicate that the effective inflow accelerates towards the pro-peller, see Figures 24 and 25, hence the axial location where the effective wake is defined has an influence upon the predicted propeller rotation rate.

7575

Figure 24: The axial locations used to

compute the effective wake field. The con-tours represent the body force distribution in

the RANS simulation. (Rijpkema, et al., 2013)

Figure 25: The RANS/BEM coupled re-

sults of effective wake velocities at different locations, where the acceleration of effective

wake flow towards the propeller disk is shown. (Rijpkema et al., 2013)

By using the effective wake field linearly

extrapolated from upstream locations to the propeller reference plane, the accuracy of propeller performance prediction was im-proved. In addition the effective wake fields predicted by a RANS-BEM coupled method and by the more traditional force-field method based on the nominal wake and propeller thrust loading were shown to be quite differ-ent, being the former method more accurate with respect to the propeller performance when the extrapolated effective wake was used.

In the viscous/potential flow coupled ap-proach by Sánchez-Caja, et al. (2014b) a lift-ing line model for the propeller in effective wake was used to compute the body forces which were circumferentially averaged and distributed on the propeller's reference plane, or the actuator disk. However, it was shown from open-water computations that the pro-peller-induced velocities by the lifting-line model were different (and not accurate due to the assumptions made in the model) from those predicted by RANS using the equivalent body force distributions, which would bring about errors in the effective wake so predicted.

For the three components of induced ve-

locity vector, a procedure was proposed to quantify the correction factors for such errors due to the lifting line model through coupled computations for the open-water propeller. Numerical results indicated that the correction factors at a reference thrust loading condition could be applied, with just a little loss of ac-curacy, to another condition where the thrust loading was within ±50% of that at the refer-ence condition. This feature might allow for savings in the computation of the correction factors. The procedure was applied to a hybrid CRP pod configuration, where it was shown that the errors in thrust and torque were about ±5% without corrections for the effective wake. As the interaction between forward and aft propellers was treated as part of the effect wake, the procedure would make it possible to use single propeller design methods for the CRP.

In the naval context, Liefvendahl, et al.

(2012) presented near-wall modelled LES simulation results for the SUBOFF+E1619 configuration, using fine (16M) and coarse (8M) grids respectively. The authors found that the coarse grids resulted in a slightly higher level of unsteadiness in the wake flow, but a higher level of fluctuation in blade thrust, and concluded that much higher grid

7676

resolution would be needed for more accurate simulation of unsteady flow and propeller forces. Zhang et al. (2012) presented RANS simulation results for another submarine hull/propeller configuration to investigate the effects of free surface on resistance and self-propulsion. The CFD and EFD data correlated well, both indicating that the free surface ef-fect on resistance was negligible below a cer-tain centreline submergence (h/L > 1/3), while that on the self-propulsion factors was even smaller.

2.4.2 Energy-saving devices

The scale effects of the Wake Equalizing Duct (WED) and the Vortex Generator Fins (VGF) on propulsion and fluctuating pressure were numerically and experimentally investi-gated by Heinke et al. (2011). The RANS predictions of the nominal wake at model- and full-scale indicated that the propeller in-flow would be largely altered by the scale ef-fect, while the WED, or the VGF alike, could reduce the wake peak distinctly. The CFD re-sults were utilized in designing the VGF. Concerning the scale effects on propulsion factors, discrepancies existed between the RANS and ITTC '78 predictions, especially for the WED, though the hull efficiencies happened to be close to each other.

Huang et al. (2012) presented an investi-

gation of the WEDs for a bulk carrier based on RANS simulations with a body-force pro-peller model and experiments. Both work in-dicated that, for a fixed tilt angle, the asym-metric arrangement of port and starboard half-ducts was quite important for maximizing the energy-saving rate. The RANS-based energy-saving was slightly lower than the model test result.

The effects of symmetrical and asymmet-

rical WEDs for a VLCC were investigated by Yu et al. (2013) based on RANS simulations and a real-geometry propeller model. In this

study, the extrapolated energy-saving rate from the RANS results was somewhat higher than that from the model tests.

In Guiard et al. (2013) the procedure for

designing the Mewis Duct® was presented in brief. The fin setting designed on the basis of model experiments was subject to further ad-justments to make full use of the full scale wake flow. In this final step of design, RANS simulation results at model and full scale pro-vided the designer with a reference. Despite the lack of full scale wake data, it was as-sumed that existing procedure for scaling the effective wake fraction was applicable to scal-ing the nominal wake fraction, too. And it fol-lowed that the full scale nominal wake distri-bution would be deemed as a good prediction if its disk-average was close to the nominal wake fraction predicted by an accepted wake scaling procedure.

For a mid-size tanker, the influences of

grid size, turbulence model, and surface roughness on predicted nominal wake flow and fraction were investigated. For a suffi-ciently fine grid set of high quality, it was shown that the nominal wake fractions were under-predicted with both SST k-ω and RST turbulence model if the hull was treated as a smooth surface. The surface roughness value was shown to have important impacts on the predicted wake distribution and fraction. In terms of the predicted wake fraction, the RST model performed the best for the typical roughness value of 0.188mm, while the SST k-ω model needed a much larger roughness of 0.5mm to yield similar result. Meanwhile, in the two cases the simulated wake flows were quite different (Figure 26).

A numerical investigation was made by

Huang S.-Q, et al. (2012) for the effects of the Pre-Swirl Duct (PSD), a combination of a pre-positive duct and several pre-swirl stators. Four cases having different duct and/or stator

7777

section profiles were simulated. The device's effects of equalizing and pre-swirling the pro-peller inflow were confirmed by the RANS results. It was concluded that, among others, the stator pitch angle was a key parameter for the energy-saving predicted.

Figure 26: The full scale nominal wakes

of a tanker predicted by RANS simulations. Upper: RST model, roughness 0.188mm; lower: SST k-ω model, roughness 0.5mm. The axial wake fractions were 0.313 and

0.318 respectively, against the value of 0.316 estimated by using the model wake fraction obtained from RANS and the RST model.

(Guiard et al., 2013) In Haimov et al. (2011) the scale effects

on ducted and CLT propeller performance in open water were investigated. The RANS computations were conducted with a com-mercial solver, using unstructured tetrahedral grids. In respect to the CLT propeller, the comparison between model and full scale re-sults indicated that the increase in efficiency at full scale was primarily due to the increase

in thrust, in comparison to the decrease in torque. Meanwhile, the RANS-predicted scale effect was smaller when compared with a scaling procedure practically used for the CLT propeller. In addition, the RANS-predicted scale effects under lighter loading condition were more pronounced and much larger than for conventional propellers.

Sánchez-Caja, et al. (2014a) investigated

the influences of endplate geometry on the efficiency of the CLT propeller based on full-scale simulations for two propellers in open water using an in-house RANS solver and Chien's low Reynolds number turbulence model. To reduce numerical uncertainty, a template-based procedure was devised for generating block-structured grids having the same topology and similar grid size distribu-tions for different endplate geometries. The grid dependency was studied by using three successively coarsened grid sets. The largest difference was 1.3% in torque between the coarse and fine grids. Twelve cases were in-vestigated for a 4-bladed propeller with varied endplate geometries. It was shown that the contraction of endplate affects both efficiency and thrust, and lighter loading on endplate improves the efficiency; the forward sweep improves the efficiency, too. From a theoreti-cal viewpoint, the working mechanism of endplate was analyzed based on radial distri-butions of bound and free vortices obtained by integrating the flow velocities. The results indicated that for cases of higher efficiency the tip vortex was weaker, consequently the induced drag was smaller.

The RANS-based investigation was fur-

ther conducted by Sánchez-Caja, et al. (2014c) of scale effects on the 4-bladed CLT propeller. The propeller efficiency at full scale was 10% higher than at model scale, where 2% was from the endplates due to the reduction in torque. The circulation distribution at full scale was higher in magnitude but lower in

7878

slope, confirming the results of higher thrust coefficient as well as efficiency than at model scale. The strong dependency of efficiency scaling on the type of flow regime at model scale was pointed out. In some cases, effi-ciency-based ranking for endplate designs at model scale were different from that at full scale.

2.4.3 Multi-component propulsors

Kinnas, et al. (2013) presented a method based on potential/viscous flow iteration by treating the duct-induced velocity field under propeller action as the effective wake for the propeller, for the purpose of computing pro-peller sheet cavitation under inclined inflow. The comparison of computed and measured open-water performances of a ducted propel-ler indicated that the iterative method was able to predict the hydrodynamic performance with good accuracy, especially for the thrust.

In Bulten, et al. (2011) a RANS approach

was employed to predict and analyze the Reynolds scale effects on the open water per-formance of a ducted propeller and on the nominal wake of a ship hull. In regard to the ducted propeller, RANS results for the Kaplan 4-70 propeller in 19A nozzle indicated that the increase in open-water efficiency at full scale was mainly attributed to the decrease in propeller torque, while the duct and propeller thrust coefficients were quite close between model and full scales. An analysis was made to explain the results by employing the pump theory. It was argued that the reduced viscous loss at full scale had resulted in an increase in the dimensionless flow rate through the noz-zle. Consequently the propeller loading was reduced at full scale, which resulted in de-creases in both thrust and torque of the pro-peller, apart from the traditionally acknowl-edged scale effects on propeller thrust and torque.

The simulation approach was further ex-tended for use with steerable thrusters in Bulten, et al. (2013). In this case the RANS-predicted model and full scale performances of the thruster in a straight course indicated that the efficiency increase at full scale was mainly due to the increase in unit thrust, in which the axial forces on the duct and propel-ler housing made more contributions. For the bollard pull performance, a generic prediction method was proposed by making use of the pump theory and RANS flow data, and the influence factors wherein were discussed. In respect to the transient thrust and torque, it was found from unsteady simulations that their fluctuation amplitudes were larger in free sailing condition than in bollard pull con-dition, and were asymmetrical about the steer-ing angle. The reason for the asymmetry was further analyzed by comparing the contribu-tions from the lateral force and the eccentric-ity of thrust.

To investigate the capability of the RANS

simulation approach for ducted propeller un-der non-cavitating and cavitating conditions, CFD and EFD results were compared by Xia et al. (2012). The RNG k-e model with wall function, and Sauer and Schnerr's mass trans-fer model were employed for turbulence clo-sure and cavitation, respectively. Block-structured and unstructured grids were used for fully wetted flow, where the predicted open water characteristics both agreed well with the measured one, except for lightly loaded conditions where the flow separation occurring near the duct trailing edge was not well simulated. Under developed cavitation conditions, although the numerical approach was able to simulate thrust breakdown, the thrust and torque were over-estimated in gen-eral and there were significant increases in predicted thrust and torque towards the start-ing point of thrust breakdown. It was con-cluded that the unstructured grids were more suitable for modelling the tip-clearance flow.

7979

Sakamoto et al. (2013b) attempted model and full scale viscous CFD computation of a POD propulsor in open water configuration. The propeller was represented by a body-force model. The full scale propulsive effi-ciency estimated by the full scale CFD com-putation was higher than that of ITTC predic-tion method applied to the model scale CFD result (Figure 27). The discrepancy comes from differences in the resistance of the POD drive, caused by the changes of the flow field such as the position of the separation line.

Figure 27 Comparison of propulsor open

water characteristics: unit-based, Exp. vs CFD vs Exp. with ITTC correction, model and full

scale. (Sakamoto et al., 2013b) Fujisawa (2013) discussed the scale effect

on the POC of contra-rotating propeller by CFD. Fore and aft propellers of a CRP system have different trends in scale effect. It is sup-posed that the turbulence caused by the fore propeller hastens the flow transition of aft propeller (Figure 28).

Figure 28: Scale effects on each of fore

(subscript “f”) / aft (subscript “a”) propeller open performance (Fujisawa 2013)

In the ocean engineering context, thruster-

hull interactions are important for DP system design. The numerical modelling of such in-teractions is challenging especially under the bollard condition. Maciel et al. (2013) pre-sented a RANS-based approach to this prob-lem and investigated its feasibility and accu-racy in terms of predicted thrusts and wake flows for three typical cases, i.e., a ducted thruster model working in open water, under a flat plate, and under a barge. The propeller was modelled by an actuator disk where the body force distributions were determined by fitting RANS results. By using a small current speed (J = 0.028) together with careful choice of numerical schemes and parameters, simula-tions under the bollard condition were real-ized with reasonably good agreement with the measured forces on the thruster and the plate/barge, and with the wake flows meas-ured by PIV, see Figure 29 for example.

8080

Figure 29: Comparison of RANS-

simulated axial velocity contours with the wake trajectory measured by PIV for the tilted

thruster working under the barge with bilge keel. (Maciel et al., 2013)

2.4.4 Design & optimization

Vesting et al. (2013) presented a multi-objective optimization study for a cruise ship propeller by considering the open-water effi-ciency as well as blade cavitation and induced pressures in a given wake field. A vortex lat-tice code was used to predict hydrodynamic forces and sheet cavitation, while a boundary element code to calculate the induced pres-sures. The codes were driven by a genetic al-gorithm. The blade geometry was parametri-cally represented in the traditional way, and morphed with B-spline curves in order to re-duce the number of design variables. A sensi-tivity analysis (SA) was carried out for the geometric parameters with regard to their im-pacts on the performance aspects being con-sidered. Starting from the baseline geometry which was already manually optimized and proven to be a good design by model experi-ments, further optimizations were conducted with all and a selected set (according to the results from SA) of the design variables, re-spectively. The latter case proved to converge faster than the former one. The results indi-cated that induced pressures could be further reduced. Satisfaction of the constant-KT con-straint, and the resolution for interaction ef-fects among parameters remain as issues in the method.

2.4.5 Off-design operating conditions Hur et al. (2011) measured propeller shaft

torque and stress for an LPG carrier in a crash stop operation during sea trials. RANS simu-lations were carried out using the RPM and ship speed data from sea trial. Both steady and unsteady RANS results for torque were close to each other at full astern, and also to the sea trial data when the ship wake was ig-nored, suggesting that steady simulation could be used for blade strength analysis at initial design stage.

In Sileo, et al. (2011) RANS simulations

were carried out for a self-propelled chemical tanker model under reversing condition, and the computed forces were compared well with measured data. Based on the numerical results of flow and hull pressure etc. the reasons for a reversing single-screw ship to deviate from straight course were analyzed.

The flow and forces under crash astern

conditions were simulated by using LES for a single propeller, Jang, et al. (2013), and a ducted rotor with upstream stator rows, Jang, et al. (2012), and compared with model test data with good quantitative agreement. The typical flow feature was a highly unsteady vortex ring having its averaged location and strength changing with the advance ratio, Fig-ure 30. Flow separation from the trailing edges resulted in high-amplitude, transient blade loads and the lateral force. For the ducted rotor-stator configuration, the same ring-vortex structure existed and numerical results indicated the lateral force came mostly from the pressures on duct inner surface due to tip-leakage flows. The LES model was fur-ther applied by Verma et al. (2012) to investi-gate the effect of hull on the propeller in crash astern operation. In addition to the ring vortex structure in the vicinity of blade tips, in the presence of the hull there existed a recircula-tion zone upstream of the propeller. The lead-

8181

ing and/or trailing edge flow separation con-tributed to the transient lateral force.

In Amini et al. (2012) the feasibility of

different numerical methods was investigated for predicting the six-component forces acting on the propeller blades of an azimuth pod thruster working in pulling and pushing modes within a ±30° range of the oblique in-flow angle. The computational methods in-clude a blade element momentum theory (BEMT), a boundary element method (BEM), and a RANS method. Being based on Glau-ert's momentum theory, the vortex cylinder model, and the blade element theory, the BEMT for oblique inflow was able to com-pute the blade forces and moments in a quasi-steady manner while the nominal wakes due to the hull and thruster housing were taking into account. Blade forces and moments ob-tained from the three methods were compared with those measured in a towing tank with a six-component force dynamometer embedded in the propeller hub. The MRF-based RANS model performed the best in both pulling and pushing modes, and the prediction accuracy was further improved when unsteady effects were accounted for by using sliding meshes. The two potential flow methods were able to predict the variations of blade forces and moments with oblique flow angle reasonably well, although under and over estimations were seen in the six components made either by the BEMT or by the BEM.

Figure 30: Circumferentially averaged flow around a propeller under crash astern opera-tion. Upper: J = −0.5, lower: J = −1.0. (Jang,

et al., 2013)

2.5 Experimental and CFD methods for the prediction of cavitation

Methods to predict cavitation on marine

propeller blades has been classified by the 26th CFD Committee as interface tracking, discrete bubble dynamics and interface cap-turing methods. The interface tracking meth-od is used to predict steady attached sheet cavitation in inviscid flow. In the discrete bubble dynamics method, cavitation is mod-elled as an interaction between bubble nuclei and pressure field variation. Bubble size gov-erned by Rayleigh-Plesset Equation. This type of method is applied to predict inception, travelling bubble and nuclei effects. The inter-face capturing method assumes that the flow

8282

is a mixture of multi-phase flow and a flow solver and a cavitation model is used to de-termine the vapor volume of fraction.

Sipila and Siikonen (2012) have investi-

gated the numerical simulation of cavitating model size PPTC propeller of SVA Potsdam in uniform inflow. The propeller wake field is calculated with Chien’s k-ε turbulence model in the non-cavitating and cavitating conditions and with Menter’s SST k-ω turbulence model in wetted conditions. The effect of the exper-imental coefficients in Merkle’s mass transfer model on the cavitating tip vortex is studied systematically. The calculations are conduct-ed with FINFLO, a general-purpose CFD code and numerical results are compared with model test results performed by SVA Pots-dam. Particular attention was paid to the grid resolution. The numerical results and the ex-periments show a reasonable correlation with each other.

Figure 31 The surface grid on the blade

and a slice of the grid in the slipstream that has been adjusted to follow the wake of the

blade and the tip vortex at the finest grid level Sipila & Siikonen (2012).

Lu et al. (2012) performed numerical sim-

ulations of the cavitating flow around a typi-cal yacht and Ro-Pax vessel propeller operat-ing in open water but mounted on an inclined shaft. They used Large Eddy simulation (LES) and Unsteady Reynolds-Averaged Navier-Stokes (URANS) in combination with

a Volume-of-Fluid implementation to capture the liquid-vapor interface and a transport equation-based method for the mass transfer between the phases. They compared the nu-merical results with the experiments. Their results indicate that a potential flow solver is not suitable for prediction complex sheet type, and root type of cavitation. RANS has partly captured the dynamic evolution of the sheet close to the tip region as well as the occur-rence of the root cavitation. LES captured the correct location and dynamic behavior of the vortical structure (as was not the case for RANS) as shown in Figure 32. However the grid resolution is still an issue for the LES computation compared to those for the RANS.

(a) (b)

Figure 32: Blade pressure with iso-surface of the second invariant of the vorticity ∇v − ∇×v , indicating vortical structures, as pre-

dicted by RANS (a) and LES (b) Li et al. (2012) also made an attempt to

predict numerically cavitating flows for the INSEAN propeller E779A operating in uni-form and non-uniform wakes. A multiphase mixture flow RANS solver and Zwart’s mass transfer model are used to predict the turbu-lent cavitating flow. Turbulence is modelled by a modified SST k-ω model. In the uniform wake, the predicted sheet cavities are stable and have similar patterns as observed in the experiment. They found that there are unre-solved issues like the cavitation inception or disappearance leading edge cavity position, differences in the maximum cavity area and its location.

8383

2.6 The need for R&D There is still a need for continued R & D

to aid in the improvement of model experi-ments, numerical modelling and full-scale measurement. Specific areas needing im-provement are the following:

• model and full scale meaaurements of

propulsors in off-design conditions • full scale measurements of ship propul-

sive gain due to the use of Energy Sav-ings Devices (ship configurations with and without)

• propulsive performances on composite propeller at full scale and model scale with possible measurement of blade de-formation and torque

• full-scale measurements on Hybrid Con-tra-Rotating Shaft Pod propulsors

• EFD and CFD (e.g. RANS) simulation of the effect of varying Reynolds number on the performace of blade sections.

• full scale measurement of waterjet inlet flow velocity fields

2.7 High-speed marine vehicles Performance prediction of high speed craft

with a view to improve model/ship extrapola-tion techniques and additional investigations into scaling effects of waterjet and surface piercing propeller propulsion tests are focused on in this review.

2.7.1 Powering and performance predic-

tion Numerous studies are related to catamaran

concept including well-known typical high-speed multi-hull model DELFT 372 catama-ran for which new tests were also carried out and a large database is still in construction. Development of CFD tools for high-speed

marine vehicles indicates its wide applicabil-ity in hull optimization processes and ac-ceptable accuracy for power prediction.

Kandasamy et al. (2011) reported on hy-

drodynamic optimization of multihull ships. Simulation based design (SBD) was applied for the resistance optimization of two waterjet propelled high-speed ships, namely JHSS (Joint High Speed Ship) which is monohull and the DELFT catamaran. The adopted SBD explores the concept of variable physics ap-proach for the Delft catamaran which shows strong waterjet hull interaction effects. The design optimization yielded geometries with significant resistance reduction for both JHSS and Delft catamaran. Tahara et al. (2011) also reported on the numerical optimization of the initial design of two waterjet propelled ships JHSS and Delft catamaran.

Figure 33: Axial velocity contours on a

cross-cut inside the waterjet inlets (top figure) and aft view of the powered JHSS with water

exiting the waterjet inlets and free surface colored by wave height (Delaney et al.,

2011). Zaghi et al. (2011) reported on an experi-

mental and numerical test campaign of fast catamarans being done at INSEAN facilities. The CFD models employed second order solver for the unsteady incompressible Navier Stokes equations. The effect of demi-hull sep-aration by means of both experimental and CFD tools is reported. Delaney et al. (2011) performed RANS calculations on the JHSS that was equipped with four waterjets. Fig.33

8484

indicates that the flow through the waterjet inlets is extremely complex including interac-tion with the free surface and these features captured correctly by CFD. Computed pro-pulsion is within 6% of experimental meas-urements which is quite good considering slight differences between the computational model and experimental conditions.

Skejic et al. (2012) theoretically investi-

gated the problem of effective power re-quirements in calm water for high-speed ves-sels (monohulls and catamarans) at prelimi-nary design stage. The effective power re-quirements have been derived from a modi-fied version of Doctor and Day (1997) meth-od which predicts total vessel resistance in calm water, introduced modifications are mainly related to different methods of wave making resistance estimation for deep and shallow water. In particular for deep water the influence of the viscosity effects according to different wave theories is analyzed and demonstrated significant influence to effective power predictions. The wave making re-sistance and effective power are also analyzed in the finite water depth where they depend on the depth Froude number. The results are compared with available published results and show good agreement.

Eslamdoost et al. (2013) developed and

validated the method, which is based on the potential flow assumption with non-linear free surface boundary conditions to model the waterjet-hull interaction. By means of this method, assuming that each of the investigat-ed parameters independently influences the resistance change, the resistance increment of the hull is estimated through a linear expan-sion in a Taylor series, which is a function of the hull sinkage, trim and the flow rate through the waterjet unit. Knowing the mag-nitude of each single parameter separately helps to understand the physics behind the thrust deduction and may aid in the optimiza-

tion of the hull/propulsor configuration. Also it sheds some light on the reason for the nega-tive thrust deduction fractions sometimes found on waterjet driven hulls. Broglia et al. (2011) reported on calm water and seakeeping experimental investigation for a fast catama-ran DELFT 372. The main issue of the paper is the interference effect between the hulls whilst former numerous studies published concerned with catamaran with the nominal separation. A monohull was tested as well. The large measurements collected provide a valuable database for CFD validation. Con-clusions on the interference effects are made. The total coefficient curve shows the presence of two distinct humps one around Fr = 0.3 and one around Fr = 0.5. The peaks in the CT are more accentuated for the catamaran than for monohull. Moreover it has seen that the se-cond hump is strongly dependent on the sepa-ration length.

Figure 34: Calm water tests: CT versus

speed (Broglia, et al., 2011) Broglia, et al. (2012) provided the results

of velocity field measurements around the Delft catamaran 372 model advancing in steady drift course. The purpose of the work is the characterization of the strong vortex structures generated along the keel of each demi-hull and to provide a valuable experi-mental data set for CFD benchmarking in se-vere off design conditions (such as during tight maneuver or when advancing at high drift angles). Zurcher, et al. (2013) discussed experimental set-up, model manufacturing and preparation for the model tests to be car-

8585

ried out using a load-varied self-propulsion testing technique and reported that a set of sea trials database is available to establish a methodology for waterjet self-propulsion test-ing based on the model tests.

2.7.2 High speed vehicle concepts

Besides widely used catamarans other concepts of high speed vehicles were also re-ported, including variations in their design by CFD and model experiment. Bono, et al. (2012) introduced the hybrid hull structure between a catamaran and a monohull (Y-shape) combining the positive characteristics of monohull (better maneuvering and less high frequency roll motions) with those of multihull (less resistance and good stability). Static and dynamic behaviors of the Y-hull model were studied in the towing tank exper-iment. Numerical simulations were carried out to optimize the hull shape. Brizzolara, et al. (2011) reported on hydrodynamic design of a family of hybrid SWATH unmanned sur-face vehicles. Modern CFD automatic para-metric optimization has been developed as an instrument for design and the final design was validated by RANS approach. Eastgate, et al. (2011) discussed the design of a submersible aircraft concept along with the results of some tests.

McDonald, et al. (2011) reported on ana-

lyzing and comparing Tri-SWACH, monohull and trimaran concepts. Fu, et al. (2011) pre-sented experimental and computational results for a Deep-V monohull planing hull. Planing craft model test program focused on collect-ing a wide range of types of measurements. Due to complexity of planning craft hydrody-namics model size was maximized (thus min-imizing scaling errors) while still being able to obtain a wide Froude number range (0.31 to 2.5). Knight, et al. (2011) discussed the methodology of multi-objective particle swarm optimization of a planning craft with uncertainty. Zheng, et al. (2012) presented

hull form design and performance evaluation of a Surface Planing Submersible Ship (SPS) which can sail in planning mode on the sur-face at high speeds and cruse underwater at low speeds. Finite volume based CFD method that took into account dynamic sinkage and trim were used for design of hull form. Pre-liminary model tests are also reported. Mosaad, et al. (2012) presented simple meth-od to predict required power for WIG craft that can be used successfully in the prelimi-nary design stage, using an iterative computer program. The output of this proposed method gives the logic and acceptable performance of the WIG craft compared to the related plan-ning hull.

2.7.3 Propulsors

Hwang, et al. (2011) developed the design procedure for developing trans-velocity pro-pellers (TVP). Trans-velocity (inflow-adopted) foil provided propellers design is that it jumps from non-cavitating condition to the super-cavitating at a very narrow speed range. Thus it operates like subcavitating propeller at low and intermediate speeds and transferred immediately to supercavitating mode at high speeds. Design procedure is ef-fective but time consuming because of RANS. Boundary element method with cavitation model and viscous correction may be more practical for propeller geometry optimization with the RANS application at the last stage for final design. TVP is designed with effi-ciency 0.72 between 20 to 30 knots and 0.67 at 40 knots. Improvement of this TVP with relatively poor efficiency at high speed and large inclined shaft condition is demonstrated by Hsin, et al. (2013). The computational re-sults from the RANS method are compared to the experimental data for both designs.

Propeller with hybrid sections (HB) i.e.

different section geometries at different radii demonstrates better performance than TVP at all speeds and large shaft inclinations up to 10

8686

degrees. A pre-swirl stator (PSS) is designed and as a result efficiency of HB improved by 1.38%. Figure 35 indicates final result of design optimization.

Figure 35: Comparison of performances of

TVP and HB propeller with PSS at 36 knots and 10 degrees inclined shaft condition. Epps, et al. (2011) represents the design

method for high speed propeller blade shape optimization. During the optimization routine the design (ship endurance speed) load distri-bution is optimized, and the off-design (max-imum speed) performance is determined, such that the chord length can be set to a minimum that still prevents cavitation at both condi-tions. Schulze (2011) and Weber (2011) dis-cussed application of improved Z-drive with contra-rotating propellers for high-speed ap-plications.

Dang, et al. (2013) reported on two new

propeller series for Controllable Pitch Propel-lers (CPPs). Following the well knownWageningen B-series and Ka-series, the new C-series comprise open CPPs where-as the new D-series concern ducted CPP’s. These series include 4-bladed CPP with large blade area and high pitch ratios for fast fer-ries. Systematic measurements of the propel-ler and duct thrusts, the torque and also the blade spindle torque have been carried out for

the entire range of operational conditions and pitch-settings of each propeller.

2.7.4 Waterjets

Giles, et al. (2013) designed an advanced submerged type waterjet and reported on its hydrodynamic characteristics, including dif-ferences between powering performance pre-dictions equipped with this propulsor estimat-ed by the ITTC “momentum flux” method and by BMT’s own method based on the pro-prietary software tool Ptool. Notable differ-ences between the two methods were ob-served in the advanced waterjet in calculation of delivered power, with the empirical method giving a higher prediction then the momen-tum flux technique in the low to medium speed range due to higher estimated propul-sive coefficients. The conventionally pro-pelled hullform performance was derived from empirical estimates, with reasonably similar predictions throughout the speed range. The sensitivity study in the calculation of propulsive coefficients highlighted the need for further research to define a robust and mature calculation procedure for sub-merged waterjet technology.

Implementation of the ITTC recommend-

ed test procedures for waterjet systems has been discussed in Dang, et al. (2012) in detail by using a waterjet propelled 15 m Fast River Ferry as an example. Attention has been paid to scale effects of model testing and the method for Reynolds corrections. Self propul-sion tests were conducted accordingly to ITTC procedure with the stock waterjet. In-stalled pump efficiency was found to be 4% lower than uniform free stream efficiency (due to flow distortion to the pump), although in most cases it is typically 2%. Determina-tion of the uniform free stream efficiency may not be necessary if the pump efficiency in in-stalled conditions can be measured correctly.

8787

Significant attention has been paid to waterjet system efficiency tests. From the tests the pump head has been found to be less sensitive to Reynolds number if the flow at the pump inlet is fully turbulent, however im-peller torque is highly Reynolds dependent. This means that a reliable waterjet system performance test should be carried out at high shaft rotational rates. Duct loss is strongly Reynolds dependent. Therefore, to get a good estimation of the duct loss, one must test at high duct Reynolds, a conclusion which is at odds with previous studies.

Figure 36 : Reynolds effect on impeller

torque coefficients. (Dang et al., 2012) ITTC procedure required measurements to

be conducted at more than one Reynolds number to get an appreciation of the Reynolds number dependency. Testing at blade tip chord Reynolds numbers as high as Rn = 5 x 106 is recommended by authors. Strong Reyn-olds dependency of the apparent intake losses have also been found which converges at a duct Rn = 107. A power loading coefficient is proposed for determination of the operating point when the pump characteristics are ex-trapolated to full scale while the existing ITTC procedure defined this point in terms of the towing force.

Chang et al. (2012) discussed a numerical

panel method developed to predict the hydro-dynamic performance of water-jets subject to a uniform flow. Steady potential flow inside the waterjet is calculated using a combined theoretical and numerical algorithm taking into account the interaction between the rotor and stator. The interaction between the rotor and the stator is evaluated using an iterative procedure that considers the effect of circum-ferentially averaged induced velocities from one rotor onto the other rotor. The pressures on the shroud surface inside the waterjet are evaluated by using hybrid scheme that cou-ples the potential flow solver with RANS solver. Satisfactory correlations with the ex-perimental data were observed. The predicted pressure head rise agreed well with experi-mental data and the maximum error is less than 2.5%. The predicted power coefficient is slightly lower (1% error) than those meas-ured.

The predicted performance due to cavita-

tion breakdown is well matched to the meas-urements. The current supercavitating model is to be improved and extended in order to analyze unsteady wetted and cavitating per-formance when the waterjet is subject to a non-uniform flow. The effect of air injection into a water jet is presented earlier by Tsai, et al. (2005) and then by Gany, et al. (2008), Gany (2011) was as much as 15 to30 % in terms of waterjet thrust. Gowing, et al. (2012) and Wu, et al. (2010) demonstrated details of test procedure development as well as optimization of the air injection waterjet. As a result Wu, et al. (2012) reported on net thrust augmentation as high as 70% (compare to 50% reported by Gany, et al. and 12% or 10% in Gowing and Tsai) for an exit void fraction of 50%. It is demonstrated that a well-designed nozzle with a proper air injec-tion scheme can provide significant perfor-mance improvement with high void fraction

8888

air injection. The conclusions are based on numerical and experimental results

2.7.5 Surface Piercing Propellers

Surface Piercing Propellers (SPP) are of-ten employed on high-speed vessels planning to reduce frictional resistance. Thus SPP’s operate in fully or partly ventilated condi-tions, making SPP’s difficult to design with high reliability. Himei, et al. (2013) discussed two theoretical methods for SPP analysis. One is modification of a program code for super-cavitating propellers using the vortex lattice method, and the other is RANS simulation applied the VOF method. The main conclu-sion was that analysis program for fully sub-merged super-cavitating propellers with cor-rection of calculated Kt, Kq values by an im-mersion factor equal to the ratio of submerged propeller disc area to total propeller disc area could provide reasonable results in a short period of time. Also, RANS simulations showed good agreement with experiment, alt-hough the differences of both were larger at higher J (Figure 37). Scherer, et al. (2011) discussed theoretical and experimental results for surface piercing outboard and stern-drive propulsion systems.

Figure 37: Comparison between calculated

and measured KT, KQ and η0. (Experiment: FnD = 6, σ = 2.3) (Himei, et al., 2013).

3 REVIEW ITTC RECOMMENDED PROCEDURES

3.1 Identify any required changes The committee was given a task to review

the procedures relevant to propulsion. In view of this the following procedures were subject-ed for reviewing: • 7.5-02-03-01: Propulsion/Performance

category including five sections. • 7.5-02-03-02 Propulsion/Propulsor cate-

gory including five sections. • 7.5-02-03-03 Propulsion/Cavitation cate-

gory including eight sections. • 7.5-02-04 Ice Testing category including

one section. • 7.5-02-05 High Speed Marine Vehicle

8989

category including one section. • 7.5-02-07-02 Loads and

Reesponses/Seakeeping category includ-ing one section. After discussion with the Advisory Coun-

cil the procedures for the Ice Testing, High Speed Marine Vehicles and Seakeeping cate-gories were left due to a high number of pro-cedures to be reviewed.

The following procedures were reviewed

and updated by the 26th ITTC Propulsion Committee:

• ITTC Procedure 7.5-02 03-01.4 Perfor-

mance, Propulsion 1978 ITTC Perfor-mance Prediction Method

• ITTC Procedure 7.5-02 03-02.3 Propulsor Nominal Wake Measurement by LDV Model Scale Experiments

• ITTC Procedure 7.5-02 03-03.2 Testing and extrapolation Methods Propulsion: Cavitation Description of Cavitation Ap-pearances

• Update to ITTC Procedure 7.5-02 03-03.3 Cavitation Induced Pressure Fluctuations Model Scale experiments

• ITTC Procedure 7.5-02-03-03.4 Cavita-tion Induced Pressure Fluctuations: Nu-merical Prediction Methods

• ITTC Procedure 7.5-02-03-01.2 Propul-sion, Performance Uncertainty Analysis, Example for Propulsion Test

• ITTC Procedure 7.5-02-03-02.1 Testing and Extrapolation Methods Propulsion, Propulsor Open Water Test.

• ITTC Procedure 7.5-02-03-02.2 Propul-sion, Propulsor Uncertainty Analysis, Ex-ample for Open Water Test Minor formatting corrections were made

to Procedures 7.5-02-03-01.2 and 7.5-02-03-

02.2 on uncertainty analysis as also made by the 26th ITTC Propulsion Committee. Proce-dure 7.5–02–03–01.4 on 1978 ITTC Perfor-mance Prediction Method was also modified for minor corrections and formatting. 7.5-02-03-01.5 Testing and Extrapolation Methods, Propulsion, Performance, Predicting Power-ing Margins has been effective since 25th ITTC, 2008 as a guideline. This procedure contains new terms, empirical formulae, etc. In Section 4.1.1 Calm Water Powering Mar-gin, it is not clear how much power margin will be applied to the model tests results with either stock or design propellers. The commit-tee thinks that the power margin requires val-idation and that a review of recommended power margins and power margin policy should be conducted, taking into account data for actual in-service performance in the typi-cal service conditions encountered.

Thanks to a naval engineer, Mikael Huss

who contacted the committee, the propulsive efficiency or quasi-propulsive coefficient, or total efficiency, ηD, equation was corrected in the 7.5-02-03-01.4 1978 ITTC Performance Prediction Method Procedure. In Section 2.4.3 the correct equation is 𝜂𝐷 = 𝑁𝑃

𝑃𝐸𝑃𝐷𝑆

. A small correction was made to the proce-

dure 7.5-02-03-02.1, Open Water Test. In Section 3.1.1, Model, the procedure refers to propeller model accuracy as “The model pro-peller should be manufactured in accordance with Standard Procedure 7.5-01-02-02, Pro-peller Model Accuracy”. Actually the referred procedure is for cavitation tests not for pro-pulsion and open water tests. The open water procedure should refer to 7.5-01-01-01, Ship Model procedure. Therefore this was correct-ed in the open water procedure. In addition 7.5-01-02-02, Model Manufacture, Propeller Model Accuracy is confusing for users. The committee recommends that this procedure should include all tolerances for model manu-facture of propeller in two sections, including

9090

Section 1 for propulsion and open water tests and Section 2 is for cavitation tests.

Some additional information on required

calibration of LDV measurements has been included in the 7.5-02-03-02.3 Propulsor Nominal Wake Measurement by LDV Model Scale Experiments. As a matter of fact, to really follow a Quality System, the calibration of the calculation of the velocity V=fD . df should include both the calibration of the fre-quency processed by the Burst Signal Ana-lyser and the calibration of the fringe spacing with the rotating disk. Then the velocity un-certainty can be determined using both the uncertainty on the Doppler frequency fD and on the fringe spacing df :

f

f

D

D

dd

ff

VV ∆

+∆

=∆

Concerning the 7.5-02-03-03.3 Cavitation

Induced Pressure Fluctuations Model Scale Experiments procedure, the committee rec-ommends distinguishing two types of analysis: harmonic analysis i.e. blade angular position domain analysis [p(θ)=p(ω.t)], and time do-main analysis [p(t)]. The first one is recom-mended to eliminate the potential fluctuation of the shaft revolution rate if the pressure is sampled using a multiple pulses shaft encoder. The second is preferred when examining broadband level of the pressure pulse signal.

Concerning the 7.5-02-03-03.4 Testing

and Extrapolation Methods Propulsion; Cavi-tation Cavitation-Induced Pressure Fluctua-tions: Numerical Prediction Methods, the committee recommends to adopt the follow-ing revisions: • Under Section 2.2.1, three references

were added which reveal the most recent advances in effective wake calculation by the coupled RANS/potential-flow meth-ods.

• Under Section 2.3.1, the status descrip-tion for cavitation prediction was updated. The RANS and two-phase flow methods are now capable of predicting unsteady sheet cavitation though the accuracy needs to be improved, instead of being unable to address the problem.

• Under Section 2.4, a recent publication was added and commented to support the existing description of the more accurate but complicated method for hull pressure calculation.

• Under Section 3.1, “rake” was added as one of the propeller geometric parameters which are to be considered in pressure fluctuation computations. Concerning the 7.5-02-03-03.2 Testing

and Extrapolation Methods Propulsion; Cavi-tation Description of Cavitation Appearances procedure, the committee recommends pro-viding sketches of propeller blade with the cavity extent on the suction side as a function of blade angular position (Figure 38).

0°

126° 108° 90° 72°

54° 36° 18°

198° 180° 162° 144°

270° 252° 234° 216°

342° 324° 306° 288° Figure 38: Suction side cavitation as a func-

tion of blade angular position

9191

3.2 The need for new procedures The committee discussed the need for new

procedures and concluded that ships with multiple shafts and pre-swirl duct concepts with an integrated pre-swirl fin system are the main potential areas to be considered. Since the current procedures deal only with single and twin-screw propulsion, propulsion test

with multiple shafts (mainly three) should be addressed and standardized for more accurate full-scale prediction. The committees recom-mend classification of existing propulsion systems as shown in the next table along with the existing or required self-propulsion proce-dure that should be applied for each class.

9292

Table 1: Propulsion system classification CASE I

Single shaft line Propeller

Twin shaft lines Propellers or Pods

Already existing ITTC self-propulsion proce-dures

CASE II

Center line Propeller + wing conven-tional shafl line propeller

Center line Propeller + wing Pods / Thrusters / Z drives

Need for self-propulsion procedure that should include differentiation of wake fraction and thrust deduction factor for wing and centre propel-lers and issues on power distribution. Possible extension of the existing procedure

CASE III

Single Shaft Line CRP Concept Conventional Propeller / Pod combi-nation

Twin shaft lines CRP Concept Conv. Propeller behind skeg / Pod Combination

Conv. Propeller open shaft / Pod Combination

A new guideline is pro-posed by the present committee for Hybrid Contra-Rotating Shaft Pod Propulsors (HCRSP) Model Test.

CASE IV

Single Forward and aft propulsors (double ended ship)

Twin Forward and aft propulsors (double ended ship)

Water jet(s) combined with conv. propeller / Pods

Need for self-propulsion procedure that should include issues on power distribution optimization

9393

Within this classification, CASE III,

which is related to Hybrid Contra-Rotating Shaft Pod Propulsors, concerns propulsion systems where a high interaction between propulsors occurs. CASE II and CASE IV in-volve configurations where low interaction might exist between propulsors. Interaction between propulsors means that the loading of one propulsor is influenced by the loading of the other propulsors. CASE I is the only pro-pulsion configuration where the interaction is assumed negligible.

The EEDI requirement of the IMO is forc-

ing the ship industry to look for solutions, apart from the conventional ones, to satisfy the new requirements. Mewis ducts are one energy saving device which has been increas-ingly installed on ships. The application of the system requires on a ship-by-ship basis by the help of CFD techniques. Scaling of wake and the angles of the fin system are critical issues. Therefore a combination of CFD methods and experiments should be done in a coordinated way.

The committee highlights the following

issues: • Multiple shaft line (number of shaft >2)

propulsors and the need to address the power distribution in the self-propulsion analysis

• Hybrid propulsion system procedures and guidelines

• Scaling issues on Energy Saving Devices

4 LIAISON WITH THE PERFORM-ANCE OF SHIPS IN SERVICE COMMITTEE

The IMO developed an Energy Efficiency

Design Index (EEDI) which expressed the ratio of total CO2 emission from combustion of fuel, including propulsion and auxiliary engines and boilers, taking into account the carbon content of the fuels in question, with the transport work, calculated by multiplying the ship’s capacity (dwt), as designed, with the ship’s design speed measured at the maximum design load condition and at 75 per cent of the rated installed shaft power.

worktransportemissionCOEEDI

2=

A simplified version of the EEDI formula

is as follows:

∏ fjnj=1 ∑ PMEi. CFMEi. SFCMEinME

i=1 + (PAE. CFAE. SFCAE)fi. fc. Capacity. fw. Vref

Where fj : correction factor for ship spe-

cific design elements fi : capacity facto fc : cubic capacity factor fw : weather factor PMEi : Power of ith Main Engine CFMEi : Conversion factor from Power

to CO2 for fuel of ith Main En-gine

SCFMEi : Specific Fuel consumption for fuel of ith Main Engine

PAE : Power of Auxiliary Engine CFMEi : Conversion factor from Power

to CO2 for fuel of Auxiliary En-gine

SCFMEi : Specific Fuel consumption for fuel of Auxiliary Engine

Capacity : Measure of carrying power, eg deadweight for Tankers

Vref : Ship Design speed

9494

In this formula, only the following items can be determined from model tests (or full size power trails): PMEi, Vref and fw.

The determination of the speed power

curve to determine PMEi, Vref and fw are cov-ered by the following procedures: • Seakeeping Committee; 7.5-02-07-02.2

Testing and Extrapolation Methods Loads and Reponses, Seakeeping Prediction of Power Increase in Irregular Waves from Model Tests

• Specialist Committee on Performance of Ships in Service; 7.5-04-01-01.1, 7.5-04-01-01.2 Speed and Power Trials Parts 1 and 2.

Comments have been made and forwarded to the relevant committees, but too late for a re-sponse to be received within this session.

It is our understanding that the Specilaist

Committee on Performance of Ships in Servi-ce has decided not to produce a procedure for the Fw component in the EEDI formulation.

This committee recommends that the issue

of Power Margins for satisfactory performan-ce and for safety should be reviewed jointly by the Propulsion and the Ships in Service Committee in the next session.

5 IDENTIFY WHERE CFD CAN SUP-PORT EFD AND THE NEED FOR HYBRID CFD/EFD PROCEDURES

5.1 Status of relevant developments Review of the papers published in major

symposiums and journals during the period of the Committee indicates that there is a con-tinuously growing interest in applying viscous CFD tools for predicting the hydrodynamic and cavitation performances of various ma-

rine propulsors and energy-saving devices, in particular the Reynolds scale effects. RANS methods are the most commonly used method; meanwhile, for enhanced resolution of the flow the DES and LES methods began to be applied to more complex configurations or more demanding operating conditions (Castro, et al., 2011; Jang, et al. 2012 & 2013). On some topics, such as the ship wake, ESDs, and propeller at crash astern, fully RANS or combined viscous/inviscid tools are being used as complements to model experiments by providing data that are difficult or impos-sible to measure. Concerning powering per-formance prediction, Verhulst (2012) ex-pected that hybrid procedures would emerge based on suggestions that CFD could be a bet-ter tool than model experiments for predicting resistance scale effect when flow separation is severe at model scale, and for evaluating scale effect on wave-making resistance.

For designing a wake-adapted propeller or

predicting its cavitation behaviour, the effec-tive wake field is needed, which can be pre-dicted from the model-scale nominal wake field by a scaling method, such as the RANS or the Sasajima method as recommended by the Specialist Committee on Wake Scaling of the 26th ITTC.

The effective wake field is not generally

directly measured by model experiments (us-ing e.g. LDV techniques). As already men-tioned in Chapter 2.4.1 , there are a number of researches dedicated to predicting the effec-tive wake field based on coupled vis-cous/potential-flow CFD methods. The hull flow with propeller in action is simulated by the RANS method, while the propeller is re-placed by a body force distribution. The po-tential flow methods are employed to com-pute the propeller working in iteratively de-termined effective inflow, such as the bound-ary element methods (Rijpkema, et al., 2013; Krasilnikov, 2013) or the lifting line method

9595

(Sánchez-Caja, et al., 2014b). The computed propeller loads are converted to a body force distribution that are distributed on the propel-ler disk (Krasilnikov, 2013; Sánchez-Caja, et al., 2014b) or in the actual fluid volume oth-erwise swept by the rotating blades (Rijpkema, et al., 2013). The effective wake distribution defined at the propeller disk is determined by extrapolation from the veloci-ties at planes upstream of the propeller blades (Rijpkema, et al., 2013), or by deducting from the total wake velocities the RANS-computed induced velocities for the actuator disk (with-out hull) in open water (Krasilnikov, 2013; Sánchez-Caja, et al., 2014b).

Potentially these methods are also appli-

cable to full-scale. However, it is not yet clear how the different potential-flow and body-force models for propeller influence the pre-dicted effective wake distribution. A com-parative/benchmark study might be necessary to further assess the methods.

In ITTC recommended procedure 7.5-02-

03-02.5: “Experimental Wake Scaling”, exist-ing approaches have been listed for simulat-ing the full scale wake. RANS simulations are recommended to help with the experimental simulation work, specifically in the scaling approaches using flow liners, water speed ad-justment, and smart dummy models. In fact, the geometric particulars and even shape of the flow liner or smart dummy model are so designed that their nominal wake distributions according to RANS simulation best approxi-mate the target ones which again are some-times predicted by RANS. Hence the proce-dure is already a partially hybrid one. Further work based on viscous CFD seems necessary to investigate how the propulsor interacts with the wake simulating devices.

One category of energy-saving devices

consist of pre-propeller fins, ducts, and the combination of them. They are designed to

produce pre-swirl inflow to the propeller by making use of the swirling flow due to the bilge vortices, and/or to accelerate the high axial wake region. As the direction and speed of stern flow are strongly influenced by the viscous flow around the hull, scale effects are important for the design and performance ex-trapolation of such devices. Being a typical example, Guiard, et al. (2013) presented a de-sign procedure for the Mewis Duct® where the fin setting designed according to model experiments might be subject to final adjust-ments based on full-scale RANS simulation results. Due to the lack of validation data, the full-scale simulation model was fine tuned, interpretation of the results and the final ad-justments were made with care.

It has been known that the CLT and

Kappel type propellers, as well as the propel-ler boss cap fins, are subject to more severe scale effect, and the ITTC'78 procedure origi-nally designed for conventional propellers might be no longer applicable to them. RANS simulations seem to become a routine for the assessment of their scale effects. Besides, RANS tools are widely used in the design process for ESDs, see Section 2.4.2 and Sec-tion 6 for a comprehensive review, since it would be difficult and cost inefficient to im-prove the design by measuring the forces that are usually quite small, or by visualizing the flow experimentally.

For propulsors involving stationary parts,

such as the duct and pod housing, scaling is an issue as the blades and stationary parts ex-perience different flow regimes, which is fur-ther complicated by the change in load shares among the parts at full scale.

For ducted propellers, based on a RANS

simulation and analysis, it was shown that the reduced viscous loss at full scale had resulted in an increase in the dimensionless flow rate through the duct. Consequently the propeller

9696

loading was reduced at full scale, resulting in decreases in both thrust and torque of the pro-peller, apart from the traditionally acknowl-edged scale effects on propeller thrust and torque (Bulten, et al., 2011).

The existing ITTC procedure for podded

propulsor extrapolation still calls for valida-tion by full-scale data. In practice viscous CFD tools are mainly employed to assess the scale effects. In a recent work a scaling pro-cedure for puller pods was proposed, featur-ing a correlation coefficient, β, which is de-pendent on the Reynolds number as well as the thrust-loading coefficient. The correlation coefficient was evaluated with resort to full-scale RANS simulations (Park et al., 2013).

For flexible propellers, the practical diffi-

culties were pointed out in Section 2.1.3 to satisfy both hydro and structural dynamic similarity laws. Instead, numerical simulation involving coupled fluid and structure analysis was recommended.

5.2 Needs for hybrid procedures Based on the brief review in the preceding

section, the Committee finds that, although CFD is being increasingly used to various as-pects of ship propulsion, especially concern-ing the scale effects and helping with design by providing complementary data to experi-ments, it is still too early to recommend a new hybrid procedure mainly for two reasons.

First, the case studies available in the open

domain are based on different modelling ap-proaches which brings about many options and makes it difficult to judge their applicable extent. In this sense benchmark studies are necessary. Second, before a numerical ap-proach can be incorporated into an existing procedure, it needs validation by full-scale data.

However, the Committee recommends that CFD should be gradually integrated into the overall tool set for making predictions in the same way as any other experience or the-ory based method is at present. Potential combinations of CFD and EFD are listed be-low, • CFD-aided scaling of resistance and pow-

ering • CFD-aided simulation of full scale and

effective wake field • CFD-aided performance scaling for duct-

ed propellers, podded propulsors, and en-ergy-saving devices

• Numerical scaling for flexible propellers Another potential use of CFD is that cali-

brated CFD can be used to extend EFD results for items not measured, such as stern flow direction and the side force on propellers, and to give guidance in design modification.

6 MODELLING AND SCALING OF UNCONVENTIONAL PROPULSION AND WAKE IMPROVING DEVICES

Energy saving devices have become an

important issue in recent times due to the in-creased price of oil and EEDI regulations. Many energy saving devices have appeared, however only some of these devices remain after validation of the effectiveness of their performance. In this study, these unconven-tional devices are classified and assessed in terms of their energy saving potential. The unconventional devices are classified into four categories, mostly based upon Carlton’s (2012) criteria, as shown in Table 2. The maximum possible gains in the propulsive efficiency in model tests were recently shown by the HSVA group and are shown in Table 3. The most difficult thing about the compari-son of the efficiency gains is choosing the ba-

9797

sis case against which they are assessed, which could vary by up to 3% or more.

PRI give a pre-rotation to the propeller

inflow IPI improve propeller inflow AFS alleviate flow separation RRE recover rotational energy from

downstream DVP decrease viscous loss after propel-

ler cap DEP decrease eddy after propeller cap DTV decrease tip vortex loss PAT produce additional thrust

Table 2 Classification of unconventional propulsors based on Carlton (2012)

Devices before

the pro-peller

Ducted pro-peller

Mitsui integrated duct: PAT, IPI Hitachi zosen nozzle: PAT, IPI

Wake Equal-izing Duct (partial duct)

Schneekluth duct: IPI, PAT Sumitomo duct:PAT, IPI

Pre-swirl sta-tor

Reaction fin: PRI Asymmetric sta-tor (DSME): PRI

Pre-swirl duct (Mewis duct): PRI, PAT, IPI

Flow regulat-ing front fin

Grothues spoil-ers: IPI, PAT Saver fin (SHI): IPI, AFS

Devices at the

propeller

Unconven-tional tip shape propel-ler

Propellers with End-Plates (CLT): DTV Backward rake tip propeller: DTV Forward rake tip propel-ler(KAPPEL): DTV

Propeller Cone Fins (PBCF): DEP Contra-Rotating Propeller: RRE Rim driven (Hubless propeller): DTV

Devices behind the pro-peller

Grim Vane Wheel: RRE Rudder-Bulb Fins system: DVP Additional thrustor fins: RRE Post-swirl stator: RRE

Renew-able

energy propul-

sion

Sail Kite

Magnus effect

Others Oscillating propulsor Surface piercing propeller

9898

Table 3 Maximum possible gains from measures aimed at increasing the propulsive efficiency (by Hollenbach, and Friesch, in HSVA) Possible

Gain

Model Tests re-quired?

Reducing Separations,/ Improving the Quality of the Wake Field

Grothues wake equal-ing spoiler

3% Yes

Schneekluth wake equal-ing duct

4% Yes

Sumitomo integrated Lammeren duct

(SILD)

6% Yes

Recovering Rotation Losses

Twist rudder without rud-der bulb

(BMS / HSVA) 2% Yes

Single pre-swirl fin

(Peters / Mewis) 3% Yes

Pre-swirl fin system

(DSME, Korea) 4% Yes

Rudder thrust fins

(HHI, Korea) 4% Yes

Reducing Hub Vortex Losses

Divergent propeller boss cap

2% Yes

Rudder with rudder bulb

2% Yes

Propeller boss cap fins

(PBCF) 3% Proposed

Reducing Rotational and Hub Vortex Losses

Twist rudder with rudder bulb

(BMS / HSVA) 4% Yes

High Effi-ciency Rud-ders

(Wartsila, Rolls Royce)

6% Yes

Note: Possible gains are not fully cumulative

6.1 Devices before the propeller

6.1.1 Ducted propeller Recently, the conventional duct has been

modified into several configurations. A partial duct is more popularly used than a conven-tional duct for equalizing the oncoming flow into the propeller as well as for saving energy. The Mitsui integrated duct and the Hitachi Zosen duct might not be classified as partial ducts but rather as a classical ducted propel-ler, because the size is almost the same as for a conventional duct, although the positioning and shape have been changed slightly (see Figures 39 and 40).

Figure 39: Mitsui integrated ducted pro-

peller (Carlton, et al., 2010)

Figure 40 : Hitachi Zosen nozzle propeller

(Carlton et al., 2010) Sumitomo’s SILD has been successfully

applied to a tanker, where the efficiency gain was more than 6% as shown in Table 4. There

Propeller

Rudder

Hz nozzle

B.

9999

may be further room to improve the efficiency at slow speed ship with an optimized duct. The scaling problem for these ducts may dif-fer somewhat from a conventional duct. Fur-ther work on this scaling problem, including pre-swirl in the duct is expected in the near future.

A EFD case study for the effect of a par-

tial wake equalizing duct was carried out for a river-going general cargo ship by Korkut (2006). Analysis of the results indicates that the partial wake equalizing duct concept with an appropriate stern design affects not only the flow characteristics at the aft-end, but also the propulsion characteristics.

Bulten (2011) proposed a full scale nu-

merical towing tank and wake field for analy-ses. The duct was analysed as an axial pump instead of using conventional propulsive coef-ficients within CFD. The well-known 19A nozzle and Kaplan type propeller were used to investigate scale effects as shown in Table 5.

Table 4 Typical efficiency gains from PIDs (Propulsion Improving Devices) from HSVA (Hollenbach & Reinholz, 2011)) Year Ship Type Device Gain in Power

Design Draught

Ballast Draught

2010 ConRo Vessel

DSME Pre-Swirl Stator

3.7% Not investi-gated

2009 Kamsarmax Bulk Car-rier

DSME Pre-Swirl Stator

6.3% 1.4%

2009 7,450 TEU DSME Pre-Swirl Stator

3.6% Not investi-gated

2008 16,000 TEU DSME Pre-Swirl Stator

3.8% Not investi-gated

2008 13,050 TEU DSME Pre-Swirl Stator

4.5% 3.2%

2008 14,000 TEU DSME Pre-Swirl Stator

4.5% 4.7%

2008 4,400 TEU DSME Pre-Swirl Stator

1.0% Not investi-gated

2008 7,090 TEU DSME Pre-Swirl Stator

3.3% 0.4%

2007 VLCC DSME Pre-Swirl Stator

5.6% 5.5%

2007 6,300 TEU DSME Pre-Swirl Stator

3.3% Not investi-gated

2007 8,400 TEU DSME Pre-Swirl Stator

3.5% 1.1%

2005 VLCC DSME Pre-Swirl Stator

4.8% Not investi-gated

2011 158k DWT Tanker

SHI Saver Fins 3.2% Not investi-

gated

2007 8,000 TEU SHI Post Stator 3.9%* Not investi-

gated

2005 8,000 TEU HHI Thrust Fin 4.9% Not investi-

gated

2007 Aframax Tanker

Sumitomo SILD 8.7% Not investi-

gated

2003 Aframax Tanker

Sumitomo SILD 6.0% Not investi-

gated *measured in HSVA’s large cavitation

tunnel HYKAT at higher Reynolds Numbers The research shows that the difference be-

tween the model and full-scale torque is larger for the ducted propeller case than for a con-

100100

ventional propeller for the same thrust. This paper also shows the possibility of full-scale wake predictions using CFD methods for the ducted propeller.

Table 5 Wake fraction comparison for the

ducted propeller Measured CFD

(model scale)

CFD(full scale)

Wake fraction

0.5132 0.5041 0.3477

Heinke et al. (2011) investigated scale ef-

fects for ships with a wake equalizing duct or with vortex generating fins. The CFD calcula-tions at the model and full scale show that the change in the propulsion coefficients, such as the thrust deduction fraction, wake fraction and hull efficiency of ships with a WED or VGF can be predicted with good accuracy using the ITTC 1978 propulsion method. The CFD calculations show a larger scale effect on the effective wake fraction than the predic-tion using the ITTC 1978 method for the WED.

Figure 41 : Appendage profiles

(Heinke et al., 2011) Analysis of the cavitation observations

showed that tip vortex cavitation is only weakly developed if the propeller works in the full-scale wake field (DM69S). This effect

could be an indication of the investigations which are necessary to in order to understand the impact of scale effects on the wake field and cavitation, in particular the tip vortex cavitation (Figure 42).

Figure 42: Comparison of Cavity extent

(Heinke et al., 2011) Yasuhiko et al. (2011) investigated full-

scale design of a semi-circular duct using CFD. The flow field at the front of the duct was analysed at both model and full-scale. From the change in the orientation of the vor-tices, the angle and diameter of the full-scale duct were changed.

Figure 43: Basic energy-saving principles

of the semi-circular duct (Yasuhiko et al. (2011))

101101

6.1.2 Pre-swirl stator A reaction fin has been successfully ap-

plied to high block coefficient ships by Take-kuma et al. (1981) of Mitsubishi heavy indus-try. Research into the pre-swirl stator (see Figure 44) was extended to the development of an asymmetric stator by DSME (Daewoo Ship and Marine Engineering) and also the combination of a stator and wake equalizing duct.

Figure 44: DSME asymmetric pre-swirl

stator (Source unknown) The model test results are normally scaled

using ITTC1999 method (Van et al. 1999) that differs from the ITTC1978 method in the prediction of full scale wake as shown in the equation below.

Although the wake scaling in the 1999 method may be somewhat exaggerated as a result of the axial velocity retardation due to the existence of the stator, there is a good cor-relation between the analysis result and sea trial. This may act as some compensation for having no scaling of the stator drag at full scale.

The new type of stator, the so-called L-J

duct and the pre-swirl stator were recently introduced by Zondervan et al. (2011) as shown in Figure 45.

Figure 45: Illustration of a Bulk carrier fit-ted with an L-J duct and pre-swirl stator

(Zondervan et al., 2011) A similar concept was applied to a twin

shaft vessel to increase the efficiency through the use of struts. There have been many at-tempts at developing different configurations of pre-swirl stator.

6.1.3 Pre-swirl duct

Mewis developed a combination of a par-tial wake equalizing duct and an asymmetric stator that has a very compact configuration from a structural point of view. The model test results were published by HSVA model basin as shown in Table 6 where the effi-ciency gain was about from 2 to 7 percent compared to the conventional propeller. Ma-noeuvrability and cavitation tests were also conducted to compare the performance both with and without the Mewis duct. The Mewis duct has mostly been applied to high block

102102

coefficient ships such as bulk carriers and tankers as the duct is more effective.

The pre-swirl duct was analysed with a

varying stator angle and angle of attack of the duct by Huang et al. (2012). The changes in the wake field were with the variation of pa-rameters in a pre-swirl duct. An efficiency gain of between 2.9% and 4.3% was calcu-lated using CFD computations of these varia-tions.

Guiard et al. (2013) proposed a full-scale

design for a pre-swirl duct through the use of a CFD code. It was stated that a complicated turbulence model is difficult to apply to a compound propulsor such as a Mewis duct. The computed results are expected to be vali-dated by full-scale test data.

Table 6: Model test results with a BMS Mewis Duct from HSVA (Hollenbach and Reinholz, 2011) Year Ship

Type Gain in Power

Design Draught

Ballast Draught

2011 151k DWT Tanker

4.7% Not inves-tigated

2010 75k DWT Tanker

3.9% 7.2%

2010 163k DWT Tanker

4.7% 7.1%

2010 158k DWT Tanker

3.8% Not inves-tigated

2010 57k DWT Tanker

5.4% 7.8%

2010 20,000 DWT MPC

1.5% Not inves-tigated

2009 45k DWT Bulker

6.0% 5.4% *

2008 12,000 DWT MPC

7.7% 7.4%

2008 Aframax Bulk Carrier

6.9% Not inves-tigated

* light loaded draught condition

6.2 Devices at the propeller

6.2.1 Unconventional tip shape propellers Sistemar Company has proposed the Tip

Vortex Free propeller (TVF), though the name has subsequently been changed to the Contracted Loaded Tip propeller (CLT). The concept behind the design is the same as for a winglet on an airplane. This idea has been ex-

103103

tended to both forward (so called KAPPEL) and backward smoothly curved tip rake pro-pellers to mitigate cavitation risks at the pro-peller tip. The reduced strength of the tip vor-tex reduces the pressure fluctuations on the hull surface rather than improving the effi-ciency. The backward tip rake propeller has been applied to most propellers in Korean ship-yards recently.

Anderson (1997) reported three kinds of

extrapolation for tip fins, using a method based on the ITTC78 method. To secure a fair comparison, this procedure was applied to both the tip fin and conventional propellers. The corrections turned out to be bigger for the tip fin propeller, meaning that it is more sensi-tive to scale effects. Unfortunately, no full-scale tests have been undertaken and so no confirmation of this scaling procedure can be made.

Inuakai (2013) conducted a comparative

study on the performance of backward and forward (KAPPEL type) tip rake propellers. It was found that the negative pressure area on the blade can be significantly reduced with backward tip rake propellers. This means that the blade area can be reduced without sacri-ficing cavitation performance, which conse-quently leads to a 2.6% higher efficiency when compared with a conventional propel-ler.

Figure 46: Forward tip rake propeller

(Source unknown)

Figure 47: Backward tip rake propeller

(Source unknown)

Figure 48: Contracted loaded tip propeller

(Source unknown) Bertetta (2012) carried out EFD and CFD

work to analyse the CLT propeller. A panel method and RANS code were used in the computational analysis of the POW perform-

104104

ance and cavitation. There was good agree-ment between the experiments and computa-tions, with an overall error of less than 4%.

Inukai (2011) proposed the concept of a

CRP with a tip rake propeller to improve the propulsive efficiency. The efficiency of the backward tip rake type is slightly better than that of the forward type but there was no gain found in the experiment. Ultimately, a 1.5% efficiency gain was found compared to a con-ventional CRP.

Sanchez-Caja et al. (2012) analysed the

scale effects of the CLT propeller using CFD (RANS Solver). According to their computa-tional results, the difference between the full and model scale is larger than for a conven-tional propeller because the flow separation area is somewhat larger than that of a conven-tional propeller. This means that the standard extrapolation method for a conventional pro-peller may not be applicable to a CLT type propeller.

Cheng et al. (2010) reported on the scale

effects for an end plate effect propeller (KAPPEL) using both numerical computa-tions and experiments. The CFD showed lar-ger scale effect for both thrust and torque as compared with EFD, which only showed scale effect on torque.

Nielson et al. (2012) proposed a combined

system of a KAPPEL propeller and rudder bulb whose efficiency was up to 9% higher than a conventional propeller system. The proportion of this gain from the KAPPEL propeller and rudder bulb were almost even.

Figure 49: Comparison of cavity extent using panel and RANSE numerical computations (Cheng et al., 2010)

Figure 50: Open water propeller

characteristics from RANS / Panel method / Experiments at the model scale

(Nielson et al., 2012)

105105

Figure 51: Pressure distribution and

streamlines for a KAPPEL propeller with a rudder bulb seen from starboard and port

sides (Nielson et al., 2012)

6.2.2 Propeller Boss Cap Fins (PBCF)

The PBCF is a relatively compact and cheap energy saving device. The effects of which are related to the propeller’s radial loading distribution. If the loading around the hub region is large, the rotational energy of the hub vortex can be effectively recovered by the PBCF.

Ouchi (1989) reported the comparative

analyses of the sea trial results of 11 ships and their models. The results indicated consider-able scale effects between the model and ac-tual measurements, such that the efficiency gain at full scale could be two to three times that at the model scale. As far as extrapolation methods are concerned, no dedicated proce-dure for the PBCF has been reported.

Kawamura et al. (2012) reported the dif-

ference between the effects of PBCF in model tests and in the full-scale data using CFD computations. The efficiency of the computed full-scale value was better than the model test results by around 1%, however it was still al-most 2% smaller than the sea trial data.

Hansen et al. (2011) reported the analysed

results of the efficiency improvement from PBCF installation in more than 60 vessels. Improvements in efficiency of between 2% and 10% were shown, with an average im-provement of 5%. Full-scale tests were car-ried out to find the correlation between the model and full scale results. The full scale results, showing an efficiency gain of around 4%, were a slightly less than the model test results which indicate that no large scale ef-fect is present which is a different conclusion from Ouch (1989). The hull condition was also examined to assess the full-scale results sensitivity.

Figure 52: Re Relationship between M/E output and FOC saving using PBCF

(Hansen et al., 2011)

106106

Figure 53: Fitting the PBCF to a full scale

ship (Hansen et al., 2011) Hsin et al. (2011) has carried out compu-

tations on the unsteady forces in contra-rotating propellers using RANS code and the BEM method. Two CRPs were chosen to be examined and the experimental results were compared. Overall there was a reasonable cor-respondence between the two, except for the torque of the aft-propeller

6.2.3 Rim driven

Yakovlev et al. (2011) compared the rim driven propeller of both the hub and hubless types. The thrust and torque of the hubless propeller are higher than those of the propel-ler with the hub, whilst the efficiency is al-most same. Qing-ming et al. (2011) investi-gated the rim driven propeller (hubless pro-peller) with four different pitch distributions to examine the performance variations. It was shown that the vortex at the hub is closely re-lated to the radial loading distribution of the propeller.

Cao et al. (2012) designed and analysed a

rim driven thruster using a CFD code coupled with lifting line theory. The computed results have a good correlation with the experimental data. The computed results also indicated that the correct adjustment of the blade loading distribution can restrain the root and the tip region vortex.

Figure 54: Rim driven propeller

(Superyacht News.com, 2011)

6.3 Devices behind the propeller The most well-known device which can be

located behind the propeller may be the Grimse vane wheel, whose efficiency is known to be around 6%. This device however has been used only reluctantly recently as a result of the possibility of damage due to the free running condition and its large diameter.

Figure 55: Grim vane wheel

(Source unknown) Unconventional rudders have recently

been the focus for energy savings as well as for the reduction of cavitation problems on the rudder surface. Additional thruster fins (including a post-swirl stator) and a rudder-bulb (including a costa bulb) have been fur-ther developed. As the bulb size increases, the efficiency of the propeller can also be higher due to the smaller contraction of the slip stream caused by the rudder bulb. There have

107107

been very few studies recently into devices located behind the propeller, which may be due to the complex flow pattern at the stern.

Figure 56: Thrust fin

(Hyundai heavy industry)

Figure 57: Rudder bulb

(Rolls-Royce brochure)

6.4 Oscillating propulsor. Mattheijssens et al. (2012) reported the

analysis of an oscillating foil with a combined motion of heaving and pitching from experi-ments and a numerical approach. The effi-ciency of the passive horizontal foil is very high at the design frequency and half of one chord length depth.

Politis and Tsarsitalidis (2012) reported

the Flexible Oscillating Duct as a novel pro-pulsor. BEM theory was used for the theoreti-cal analysis of this propulsor and the effi-

ciency gain obtained was between 3% and 10% when compared to a B-series propeller.

7 EXAMINE METHODS OF WAKE SIMULATION

For cavitation tests and measurements of

hull pressure fluctuations, the correct simula-tion of the full-scale wake field is an impor-tant technique for the reduction of scale ef-fects. One possible solution is to use a model that does not have complete geometrical simi-larity but is shaped to produce the full-scale wake field. In this case, the full-scale wake field is calculated by the use of CFD tools. Such models are often called “smart dummy” models.

In Germany the joint research project

KonKav II, “Correlation of Cavitation Effects Under Consideration of the Wake Field” has been initiated. Participants are Flensburger Schiffbau-Gesellschaft (FSG), Hamburgische Schiffbau-Versuchsanstalt (HSVA), Schiff-bau-Versuchsanstalt Potsdam (SVA), Techni-sche Universität Hamburg-Harburg (TUHH) and Universität Rostock (UniHRO). The pur-pose of the research project KonKav II is to develop a more accurate and marketable cavi-tation prognosis. In the project the focus is on scale effects that occur through the interaction between wake field, propeller cavitation and the resulting pressure fluctuations. A deeper understanding of these processes will help to convert model test results to full-scale predic-tions in a reliable way.

One of the objectives is to improve simu-

lation of a full-scale wake field in the context of model tests. It is common practice to use dummy models with attached strainers influ-encing the flow in a way that the wake field of the full-scale version is simulated. A pro-cedure based on an adjoint sensitivity analysis has been developed by Technical University

108108

Hamburg-Harburg to find appropriate dummy model geometry and appropriate mesh pa-rameters of the strainers to simulate the previ-ously calculated full scale wake field. The photographs (Figure 58) show the geometry of the additional grids used on the conven-tional dummy model and a dummy model that is optimised on the base of an adjoint sensitiv-ity analysis.

Figure 58: Additional wire grids for the

conventional dummy model (left) and the op-timised dummy model (right), Photograph from Schiffbau-Versuchsanstalt Potsdam This research project is on-going, and

more results will be available within the next committee.

Schuiling et al. 2011, Bosschers et al.

2012 and Johannsen et al. 2012 report the use of such smart dummy models. Simple short-ening and narrowing the model (Figure 59) did not lead to sufficient wake fields. Good results could be achieved by modifying the gondola only in a way that propeller clearance and shaft height was kept the same. The result is shown in Figure 60.

Figure 59: Examples of intermediate

forms in systematic hull form variations of width and length.

Figure 60: View from behind the Smart

Dummy design (left), and the original geosim hull (right)

Even though the full-scale wake field was

not fully represented, a good similarity in the upper part was achieved (Figure 61).

Figure 61: Axial wake velocities of the

Smart Dummy design (left) compared to those at ship scale (right). The dashed lines

are at 1.1 and 0.6 times propeller radius. The solid line is at the propeller radius

109109

Blade rate order hull pressure fluctuations were close to those found on full scale, but higher order blade rate components were not improved with respect to the full scale. The resulting hull pressure fluctuations for a sen-sor directly above the propeller were com-pared. The full- scale measurements were made available by Lloyd’s Register. The geo-sim results were obtained using the wake peak identity approach while the smart dummy re-sults were obtained using thrust-identity. A very good agreement between model scale tests and full-scale tests was obtained for the first harmonic while for the second harmonics there is still a considerable difference.

Figure 62: Comparison of non-

dimensional full scale (in black) and model scale (in blue and green) pressure amplitudes

for first four blade rate orders. In Heinke, et al. (2011), the scale effects

on cavitation and fluctuating pressure were experimentally investigated for the cases of a container ship without and with the WED or the VGF fitted. Based on RANS simulations, a shortened dummy model was adopted to simulate the “full-scale wake”. In all the cases, the pressure pulses were found to be apparently lower in the “full-scale wake” than in the model wake. Either the WED or the VGF could further reduce the pressure pulses. However, the variations in the fluctuating pressure amplitudes with the attack angle of VGF were found to be different at model- and full-scale, which might provide some hint as to how to optimize the orientation of the VGF.

Further publications deal with adjoint

RANS for hull form optimisation, such as

Stück et al. (2010), Kröger et al. (2011), and Rung et al. (2012). Various objective func-tions are considered, among other things the wake quality. Further developments in this field could also be useful for an effective and fast design of smart dummy models.

8 WAKE FRACTION SCALING FOR TWIN SCREW SHIPS

This is related to the 1978 Performance

Prediction method in which no wake fraction scaling is given for twin skeg cases.

For cases of finer forms where the shaft is

supported by A-brackets using wTS = wTM is still advised and appears to be the general practice.

However, there is an increasing number of

fuller form twin skeg vessels. In these vessels the wake field experienced by the propeller is close to that of a single skeg vessel and it might be that the normal single screw wake-scaling procedure should be used. This proce-dure is used by several establishments.

The use of wake scaling on twin skeg

ships needs to be examined further and the next committee should seek to obtain exam-ples of model and ship data so that the issue of wake scaling can be examined. In particu-lar the issue of whether the full wake scaling as used in the single screw method or some reduced level of scaling should be used.

Ohmori et al. (2013) have studied the scal-

ing of results from a twin skeg model by CFD and by three semi-empirical techniques. They conclude that the method due to Tanaka gave the best result, but also that the axial and tan-gential components of wake may need to be scaled separately. They conclude that more work is needed on the scaling of tangential wake.

110110

Sakamoto et al. (2011) used CFD backed

by model tests to examine wake scaling on twin skeg ships. The work suggests that the ITTC78 single screw scaling method gives reasonably good results, but further work to develop specific twin skeg wake scaling methods is suggested.

9 SCALING OF CONVENTIONAL AND UNCONVENTIONAL PRO-PELLER OPEN WATER DATA

For initiating a comparative CFD-

calculation project two propeller geometries that are in the open domain had to be found. For the conventional propeller the PPTC-propeller (Potsdam Propeller Test Case) from SVA Potsdam could be used. But geometry for an unconventional propeller could not be found.

The PPTC-propeller has 5 blades and a di-

ameter of 250 mm. The following test de-scription was given by the ITTC:

• The propeller shall be tested in a pull con-

figuration. The corresponding hub cap is provided.

• The extent of the shaft behind the propel-ler has to be at least two propeller diame-ters.

• The extent of the solution domain can be chosen arbitrarily, however it is consid-ered necessary to have a radial domain extent which gives a cross sectional area of the domain which is at least 100 times larger than the corresponding propeller disc area. Radial extent Ddomain > 10 DP

• The dimensionless wall distance on the propeller blade shall be chosen such that the viscous sublayer is resolved.

• It is highly recommended to conduct the calculations in model scale under consid-

eration of laminar-turbulent transition on the blade. For the full-scale calculations a fully turbulent inflow can be assumed.

• The calculations shall be carried out ne-glecting cavitation.

• The water characteristics shall be taken for a water temperature of Tw = 15°C as provided in Tab. 1 on page 2.1.

• The propeller is a controllable pitch pro-peller. The gap between hub and blade root is considered unimportant regarding the integral values of the propeller and shall not be taken into account.

• The decision to calculate a single blade passage or the entire propeller (5 blades) is left to the participant. With respect to the test results the follow-

ing evaluation was requested by the ITTC: • Two different scale ratios (λ = 12 and λ =

1). • Five different advance coefficients:

J = 0.6, 0.8, 1.0, 1,2 and 1.4 • The forces on the propeller blade and on

blade sections The thrust and torque of different blade

sections shall be evaluated. The coefficients shall be subdivided into a pressure and a fric-tional component. The intention is to obtain a deeper and more detailed insight into the scale effects of the propeller. The total thrust and torque is obtained via the summation of the different blade section values. The calcula-tions shall be conducted in model and full-scale.

In total 10 calculations were requested.

The results are not linked to the correspond-ing participant, guaranteeing anonymity.

All ITTC members were invited to par-

ticipate on the comparative CFD-calculation by email. On the SVA Potsdam web site

111111

(www.sva-potsdam.de/pptc_ittc_switch.html) the propeller data, an Excel sheet for the re-sults and instructions for the calculations are provided.

In total 9 institutions participated at the

ITTC propeller benchmark. Some institutions provided multiple results, giving in total 12 results. The following the institutions partici-pated: • DGA Hydrodynamics • Hyundai Heavy Industries • Krylov State Research Centre • MARIC • SJTU • SSPA • SSSRI • SVA • Technical Research Centre Japan Marine

United Corporation (JMU) The computational results were evaluated

for the following: • Computational method, approach • Open water curves • Thrust and torque on blade sections

The participants were asked to fill out a

questionnaire and provide details regarding their computations. From this, the following can be drawn: • In general a single blade passage with pe-

riodic side boundaries is used • The side boundaries are in general match-

ing • In general unstructured meshes consisting

of tetrahedral elements with a prismatic boundary layer and local grid refinment are used

• In model scale the dimensionless wall distances ranges between 1 and approx. 50

• In full-scale the dimensionless wall dis-tances ranges in general between 1 and approx. 30

• The number of cells on the blade surface in the range between 9,800 and 80,000

• For the domain extent two groups can be distinguished. One keeping the domain very large with the cross sectional area of the domain being 3600 times as large as the propeller disc area. The other group has the domain extent very small having values of below 16. The same applies for the upstream and downstream extent of the solution domain.

• All participants use 2 equations turbu-lence models The calculated open water curves are

compared with the corresponding measure-ments. The measured model scale open water curves are extrapolated to full-scale according to the ITTC extrapolation method. The ex-trapolated model data is denoted as EFD (ex-perimental fluid dynamics) results. The CFD (computational fluid dynamics) and EFD re-sults are compared.

The thrust and torque generated by differ-

ent blade sections, ranging from the hub to the tip of the propeller, are investigated for the requested advance coefficients, both for model and full-scale. The following two dia-grams show the KT and KQ curves, for the model scale and as recalculated by the ITTC 78 method for ship scale, in comparison with CFD data for full scale. Whilst the KT correc-tion according to ITTC 78 method is very small, the CFD results show bigger correc-tions. For KQ the CFD results show positive corrections for higher advance coefficients. The standard deviation of ∆KT and ∆KQ is greater than the values itself. That may be

112112

caused by the small size of the calculation set. Thus these diagrams are preliminary results. More calculations are necessary.

Figure 63: Mean CFD values with σ-bounds and comparison with open water curves in

model scale.

Figure 64: Mean CFD values with σ-bounds

and comparison with open water curves in full scale.

Figure 65: ∆KT, red acc. to ITTC 78, black

from CFD with σ-bounds.

Figure 66: ∆KQ, red acc. to ITTC 78, black from CFD with σ-bounds

Streckwall, et al. 2013 reported their work

on an advanced scaling procedure for marine propellers. Emphasis is put on propeller de-signs with blade shapes that differ from “con-ventional” type. The work was performed within the European project “PREFUL” with the target to investigate the possibilities of improvements of the scaling calculation in order to consider the differences between blade shapes more precisely. As a result the differences between the several scaling pro-cedures are shown, especially in comparison to the results of a new “strip method”, which was developed within the project. It is under-stood that the enhanced open water correc-tions are to be compared with full-scale ob-servations (trial trips) in future.

10 DEVELOP GUIDELINES FOR HY-BRID PROPULSOR TESTING

10.1 Purpose Social demand on energy saving and

greenhouse gas emission reduction is generat-ing pressure to develop new more efficient propulsors. Remarkable advances in hybrid propulsors (propulsion systems consisting of more than one type of propulsor) has been made in recent years. But the model testing

113113

procedure for such propulsors systems has not yet been established.

Among several configurations of hybrid

propulsor, the committee focused on the Hy-brid Contra-Rotating Shaft Pod propulsor (HCRSP or so-called hybrid CRP concept) and has developed model open water and pro-pulsion test procedure guidelines.

10.2 Definition and Classification There exist many combinations of hybrid

propulsors. They can be classified into two major groups: low interaction group and high interaction group. The high interaction group consists of different propulsors arranged in line in fairly close proximity (e.g. CASE III of Table 1 in Paragraph 3). All other configu-rations (parallel or in line propulsors with more distance in between) are usually classi-fied into low interaction group (e.g. CASE II and IV of Table 1 in Paragraph 3).

As described in the new guideline, model

test of low interaction group can be conducted following the conventional Propulsion Test Procedure 7.5-02-03-01.1 or Podded Propul-sor Test Procedure 7.5-02-03-01.3 or Waterjet Propulsion Performance Prediction – Propul-sion Test and Extrapolation 7.5-02-05-03.1. However load-varying test should be con-ducted for each propulsor separately to deter-mine the self-propulsion point.

10.3 Description of guideline A guideline for the high interaction case

was developed. Although many combinations of propulsor (e.g. conventional propeller, POD, waterjet, Z-drive, CRP and so on) are possible, effective combinations from the viewpoint of energy efficiency are limited. Thus the guideline focuses on the one of the

more significant combinations, HCRSP pro-pulsors. The method is based on the studies by Sasaki (2006/2009), Chang (2011), Quereda (2012), and Sánchez-Caja (2013).

The tank model test consists of propulsor

open water test, resistance test and self-propulsion test. The unique point of hybrid propulsor model test is that load distribution between fore propeller and aft POD is not fixed. Thus the load distribution varying tests are compulsory in both propulsor open test and self-propulsion test.

Another point is the arrangement of the

propeller open boat for the fore conventional propeller. As for the aft POD drive located behind the fore propeller, the propeller open boat must be arranged in reversed configura-tion (in front of the fore propeller). In the re-versed configuration, viscous wake of the propeller open boat flows into the propeller and the measured results are to be appropri-ately corrected (Ohmori, 2013).

Fore Prop.

FLOW

Propeller Open Boat

POD dynamometer

Aft Prop.

Figure 67: Open water test configuration

The final guideline is registered as 7.5-02-03-01.6 Hybrid Contra-Rotating Shaft Pod Propulsors Model Test.

10.4 Discussion The procedure covers only model tests and

the scaling method is not included. The rea-son for this is the lack of full-scale validation data. Although the scaling method for podded propulsors will be helpful, the development of the full-scale prediction method is the subject of future work.

114114

Mitsubishi Heavy Industries (MHI) is the first shipyard that has built a large ship with HCRSP (Ueda, 2004). Their tank test proce-dure is simpler than the new guideline (it is basically based on the conventional propulsor test procedure), but it has already been con-firmed by the full scale trial results. However the committee could not adopt this method as no published paper is available.

Figure 68: HCRSP mounted on a ROPAX

Ferry (Ueda, 2004)

11 MONITORING OF FULL SCALE DATA FOR PODDED PROPULSION.

No new full-scale data has been published,

the only known available example is the “ABB case”. The Propulsion committee con-tacted ABB to establish if such data could be published.

The discussion in the 25th ITTC POD

committee report concluded that the main is-sue is still the scaling of pod housing drag. Taking this into account, the most helpful re-sults for benchmarking would be full-scale data with measurements of propeller thrust, torque and also pod housing drag.

The propulsion committee, with reference

to Honninen, et al. (2007), discussed with ABB if such full-scale data could be made available for the ship “Norilsky Nickel” for which extensive full-scale measurements were

conducted, including simultaneous measure-ments of the loads on the thruster body and propeller. ABB express its willingness to ac-quire and analyze such data and to publish the results in future.

12 CONCLUSIONS

12.1 Recommendations to the Next Com-mittee

12.1.1 Procedure Review/Update

The 27th Propulsion committee has devel-oped a new guideline for HCRSP (Hybrid Contra-Rotating Shaft Pod) Propulsor Model Tests. The model test scaling is not discussed in this guideline as there is a lack of model test and full-scale trials comparison data. For this reason, the committee recommends the continued monitoring of model test and scal-ing procedures used for this kind of device (and for propulsion devices in general) by member institutions of the ITTC. If there is sufficient information on the comparison be-tween model and full scale data indicating changes to test or scaling procedures, the rel-evant guidelines should be updated.

The committee recommends that the mon-

itoring of the existing literature for examples of the Reynolds number scale effects should continue in order to update the 7.5-02 -05-03.2 Waterjet System Performance procedure. Clarification and detailing of the procedure in the part relating to the data acquisition and in the part related to extrapolation is required. Further reviews of the literature should exam-ine the need and use of the blade-tip and chord Reynolds numbers as well the intake duct Reynolds number and update the proce-dure if required.

The committee recommends a review of

the power margins given in the guidelines and

115115

the monitoring of the effect of the introduc-tion of the EEDI on power margins.

12.1.2 New Procedures

The committee recommends developing a new procedure for propulsion performance prediction for triple shaft vessels. Although the existing procedure largely covers this, the procedure needs to be extended to take ac-count of the interaction between centerline and wing propulsors and to allow for the de-termination of the wake fraction and the thrust deduction factors for these propulsors sepa-rately.

12.1.3 Technologies to Monitor

The committee recommends monitoring the model test and scaling procedures for en-ergy saving device technologies. The impact on self-propulsion and cavitation testing is also to be reported to assess the way of taking into account differences in inflow speed and direction between the model and the ship. This is particularly relevant to wake improv-ing devices where the optimum alignment may be different between the model and ship. This raises the question of the relevance of any propulsive gains determined from model tests to the ship. The use of CFD and/or a combination of CFD and EFD should be also considered, as well as full-scale trials results.

The committee recommends monitoring

the “smart dummy” model used for cavitation tests for propulsors behind a skeg. A joint re-search project “Koncav II” has been launched in Germany and preliminary results might be available within the three years term of the next committee.

The committee recommends examining

the existing procedures and assessing where CFD results can be introduced in the propul-sion process to assist EFD e.g. use CFD to determine the target wake field to be modelled in cavitation testing, use of smart dummy model

to model the target wake field, use of CFD in conjunction with EFD for composite propel-lers).

The present committee was not able to

find a suitable model test and CFD study for unconventional propellers. The definition of what would be interesting to work with as an unconventional propulsor is still to be dis-cussed with the CFD committee.

Recent publications suggest that RANS

codes are more and more widely used for propulsor design. The analysis of the data of the benchmark launched by the 27th commit-tee should be able to give some answers to the use and interpretation of RANS methods and procedures. A combine effort with the CFD committee will encourage the continuation of this study with the aim of getting further con-tributions from member institutes. The EFD data used in this study comes from only one institute. The committee recommends that ad-ditional EFD studies on the same propeller design should be performed.

An area to examine is the fluctuating

components of propeller bearing forces, espe-cially on Pod and azimuthing thruster

Further work is still required on the way to

test and analyse the results for composite pro-pellers. The use of CFD in combination with EFD to investigate the fluid-structure interac-tion (static and dynamic hydro-elastic re-sponse) needs to be better understood.

Experimental techniques such as detailed

local flow velocity measurement using PIV, blade strain, cavity surface and volume meas-urement still need to be monitored.

Testing and estimation methods for

propulsors in bubbly flow should be moni-tored. The open-water and self-propulsion characteristics in bubbly flow are relevant to

116116

air lubricated vessels and the effect of void fraction, void type and affected area on the propeller need to be investigated and under-stood.

12.1.4 Scaling for Propulsors

As no full scale data on Pod Propulsor have been found within this session, the committee recommends that the next commit-tee should continue to look for-full scale data.

The prediction of full scale cavitation in-

duced hull pressure is still of interest either by means of CFD directly or CFD and EFD used in combination.

In view of the growing interest in energy

saving devices further work is still required on model test techniques and the prediction of power savings.

12.2 Recommendations to the Conference The Committee recommends to the Full

Conference that they should • Adopt the revised procedure ITTC Proce-

dure 7.5-02 03-01.4 1978 ITTC Perfor-mance Prediction Method

• Adopt the revised procedure ITTC Proce-dure 7.5-02 03-02.3 Propulsor Nominal Wake Measurement by LDV Model Scale Experiments

• Adopt the revised procedure ITTC Proce-dure 7.5-02 03-03.2 Testing and extrapo-lation Methods Propulsion : Cavitation Description of Cavitation Appearances

• Adopt the revised procedure Update to ITTC Procedure 7.5-02 03-03.3 Cavita-tion Induced Pressure Fluctuations Model Scale experiments

• Adopt the revised procedure ITTC Proce-dure 7.5-02-03-03.4 Cavitation Induced Pressure Fluctuations: Numerical Predic-tion Methods

• Adopt the new guideline 7.5-02-03-01.6 HCRSP (Hybrid Contra-Rotating Shaft Pod) Propulsors Model Test

117117

REFERENCES Amini H., Sileo L., Steen S., 2012, “Numeri-

cal Calculations of Propeller Shaft Loads on Azimuth Propulsors in Oblique Inflow”, Journal of Marine Science and Technology,

17(4): 403-421.

Amini H. and Steen S., 2011, “Experimental and Theoretical Analysis of Propeller

Shaft Loads in Oblique Inflow”, Journal of Ship Research, Vol. 55, No. 4, pp. 268-

288.

Barkmann U., Heinke H.J., Lübke L., “Pots-dam Propeller Test Case (PPTC), Test

Case Description”, Second International Symposium on Marine Propulsors smp’11, Hamburg, Germany, June 2011, Workshop:

Propeller performance

Black S. , “Composite propeller for Royal Navy minehunter”,

http://compositesworld.com/articles/, 2011.

Bertetta, D., Brizzolara, S., Canepa, E., Gaggero, S and Viviani, M., 2012, “EFD and CFD characterization of a CLT pro-peller”, International Journal of Rotating

Machinery, Article ID 348939.

Bono,B., Cucinotta,F.,Guglielmino,E.,Zotti,I., 2012, “Experimental and Numerical Evaluation of an Innovative Planning

Hull”, NAV 2012

Bosschers J. and van Wijngaarden E., "Scale effects on hull pressure fluctuations due to cavitating propellers", 10th International

Conference on Hydrodynamics October 1-4, 2012 St. Petersburg, Russia

Bosschers J., Lafeber F.H., de Boer J., Bos-man R. & Bouvy A., “Underwater radiated noise measurements with a silent towing

carriage in the Depressurized Wave Basin”, AMT ’13 17-18 September 2013, Gdansk,

Poland.

Briançon L., Fournier P. & Frechou D., “Ma-rine propeller noise measurements tech-

niques in hydroacoustics tunnel”, AMT ’13 17-18 September 2013, Gdansk,

Poland.

Brouwer J., Hagesteijn G., Bosman R., “Pro-peller-ice impacts measurements with a

six-component blade load sensor”, 2013, Third International Symposium on Marine Propulsors SMP’13, Launceston, Australia.

Bulten, N., Nijland, M., 2011, “On the Devel-

opment of a Full-Scale Numerical Towing Tank Reynolds Scaling Effects on Ducted Propellers and Wakefields”, Second Inter-national Symposium on Marine Propulsors

2011, Hamburg, Germany.

Bulten N., Suijkerbuijk R., 2013, “Full Scale Thruster Performance and Load Determi-nation Based on Numerical Simulations”, Third International Symposium on Marine Propulsors (SMP’13), Launceston, Austra-

lia.

Cao, Q.-M., Hong F.-W., Tang, D.-H., HU, F.-L., Lu, L.-Z., 2012, “Prediction of Load-

ing Distribution and Hydrodynamic Measurements for Propeller Blades in a

Rim Driven Thruster”, Journal of Hydro-dynamics, Vol. 24, Issue 1, pp. 50-57.

Cao, Q.-M., Dong, A., Hong, F. and Tang, D.,

2012 “Numerical Design of Rim Driven Thruster and Hydrodynamic Measurement, International Marine Design Conference

2012, Vol 3, pp. 479-489.

118118

Carlton, J. S., 2012, “Marine propellers and propulsion”, Butterworth Heinemann,

Great Britain.

Castro A. M., Carrica P. M., Stern F., 2011, “Full scale self-propulsion computations using discretized propeller for the KRISO container ship KCS”, Computers & Fluids,

51, pp.35~47.

Chang, S.-H., Kinnas, A., 2012, “Numerical Simulation of Wetted and Cavitating

Flows inside Water-jets”, 29th Symposium on Naval Hydrodynamics, Gothenburg,

Sweden

Chang, B-J., Go, S., 2011, “Study on a proce-dure for propulsive performance predic-

tion for CRP-POD systems”, J. Marine Sci. Technology, Vol. 16 pp.1-7.

Cheng, H., Chien, Y., Hsin, C., Chang, K.,

Chen, P., 2010, “A numerical comparison of end-plate effect propellers and conven-tional propellers”, 9th International Con-ference on Hydrodynamics 2010, Shang-

hai, China, pp.495-500

Dang J., Liu R., Pouw C.,2012,”Waterjet Sys-tem Performance and Cavitation Test Pro-cedures”, Symposium on Marine Propul-

sors SMP’13,Launceston,Australia.

Dang J., Brouwer J., Bosman R. and Pouw C., 2012, “Quasi-Steady Two-Quadrant Open Water Tests for the Wageningen Propeller C- and D-Series”, Twenty-Ninth Sympo-sium on Naval Hydrodynamics, Gothen-

burg, Sweden.

Dang,J., van den Boom, H.J.J., Ligtelijn, J.Th.,2013, The Wageningen C- and D-

Series propellers, FAST ‘13

Delaney, K.,2011, “Use of RANS for analysis of a high speed Sealift Concept Vessel,

MARINE 2011,Lisbon,Portugal

DNV, 2010, “Rules for Classification Of Ships, Silent Class Notation”, Part 6,

Chapter 24. Det Norske Veritas.

Dubbioso G., Muscari R., Di Mascio A., 2013, “CFD Analysis of Propeller Performance

in Oblique Flow”, Third International Symposium on Marine Propulsors (SMP’13), Launceston, Australia.

Epps.Dual Operating point Blade optimiza-

tion for high speed propellers. FAST 2011

Eslamdoost, A., Larsson, L., Bensow, R.,2013, “Contribution of Different Parameters to the Thrust Deduction of a Waterjet Pro-

pelled Hull”, FAST 2013

Ferrieri J.M., Veitch B., and Akinturk A.,”Experimental Study on Ice Manage-ment through the use of Podded Propeller Wash”, Third International Symposium on Marine Propulsors SMP’13, Launceston,

Australia May 2013.

Fujisawa T., 2013, “Study of Scale Effect on CRP by viscous CFD”, Conference Proc.

JASNAOE, 17: 147-148.

Funeno I., Pouw C., Bosman R, “Measure-ments and Computations for Blade Spin-dle Torque of Controllable Pitch Propel-lers in Open Water”, Third International

Symposium on Marine Propulsors SMP’13, Launceston, Australia May 2013.

Fu. Measurements and Computational Predic-

tions of a Deep-V MonohullPlaning Hull FAST 2011

119119

Gany, A., Shemer,A., Gofer,A., Har-Lev,D., 2008, “Testing Air-Augmented Waterjet

Propulsion”, International Conference on Waterjet Propulsion, London, UK

Giles W., Dinham-Peren T., Amaratunga S.,

Vrijdag A., Partridge R., “The Advanced WaterJet: Propulsor Performance and Ef-fect on Ship Design”, 10th IMarEST Na-

val Engineering and Exhibition , May 2010, Portsmouth, UK

Głodowski R., Richards J., Koushan K.,

Alterskjaer S.A., Atlar M. & Saisto I., “Open water performance characteristics of a pod propulsor– the benchmark test”, The 3rd International Conference on Ad-vanced Model Measurement Technology for the EU Maritime Industry, AMT’13,

Gdansk, Poland.

Gowing S., 2011, “Experimental Study of an Air-Augmented Waterjetpropulsor”, Sym-posium on Marine Propulsors’11, Ham-

burg, Germany

Guiard, T., Leonard, S. and Mewis F., “The Becker Mewis Duct-Challenges in Full-Scale Design and new Developments for

Fast Ships”, 2013, Third International Symposium on Marine Propulsors

SMP ’13, Tasmania, Australia.

Haimov H., Vicario J., Del Corral, J., 2011, “RANSE Code Application for Ducted

and Endplate Propellers in Open Water”, Second International Symposium on Ma-rine Propulsors SMP’11, Hamburg, Ger-

many.

Hansen, H. R., Dinham-Peren, T., Nojiri, T., 2011, “Model and Full Scale Evaluation of a ‘Propeller Boss Cap Fins’Device Fitted to an Aframax Tanker”, Second Interna-tional Symposium on Marine Propulsors

2011, Hamburg, Germany.

Heinke H.-J., Hellwig-Rieck K., 2011, “Inves-tigation of Scale Effects on Ships with a

Wake Equalizing Duct or with Vortex Generator Fins”, Second International

Symposium on Marine Propulsors SMP’11, Hamburg, Germany.

He X.D., Hong Y., Wang R.G., “Hydroelastic

optimization of a composite marine pro-peller in a non-uniform wake”, Ocean En-

gineering Journal 39 (2012) 14–23.

Henderson K, “New Design of 2000 TEU Feeder“,

http://articles.maritimepropulsion.com/article/New-Design-of-2000-TEU-Feeder-

6784.aspx, Maritime propulsion, January 2012.

Hideki Y., Hiroaki M., Aiichiro Saeki, 2011, “Energy Saving Technology of the Diesel-Electric Propulsion, System for Japanese

Coastal Vessels”, IHI Engineering Review, Vol 44

Himei K. Numerical Analysis of Unsteady

Open Water Characteristics of Surface piercing Propeller. Third International

Symposium on Marine Propulsors SMP’13, Launceston,Australia

Hollenbach, U. and Reinholz, O., 2011, “Hy-

drodynamics Trends in Optimization Pro-pulsion”, Second International Sympo-sium on Marine Propulsors SMP’11,

Hamburg, Germany

Holtrop J., 2000, “MARIN Training Course on Propulsion Hydrodynamics”, Nether-

lands.

Huang S.-F., Huang G.-F., 2012, “Numerical Investigation for Effectiveness of Wake

Equalizing Duct”, 10th International Con-ference on Hydrodynamics, St. Petersburg,

Russia.

120120

Huang, S., Huang, Z., Branch, S., 2012,” In-vestigation on Energy-Saving Effective-ness of Pre-Swirl Duct Based on CFD”, 10th International Conference on Hydro-

dynamics, St. Petersburg, Russia

Hur J. W., Lee H., Chang B. J., 2011, “Propel-ler Loads of Large Commercial Vessels at Crash Stop”, Second International Sympo-

sium on Marine Propulsors (SMP’11), Hamburg, Germany.

Hwang, J-L.,Chin, S-S.,Chang,K-K., Tu,

Chin-Chin.,2011, “Development of the Trans-Velocity Propellers”, Second Inter-

national Symposium on Marine Propulsors SMP’11, Hamburg, Germany

Inukai, Y., 2011, “Development of Contra-

Rotating Propeller with Tip-Raked Fins”, Second International Symposium on Ma-rine Propulsors SMP’11, Hamburg, Ger-

many.

Inukai, Y., “A Development of a Propeller with Backward Tip Rake Fin”, 2013,

Third International Symposium on Marine Propulsion SMP’13, Tasmania, Australia.

Ishida, S., Fujino, R., 1998, “A Performance

Prediction Method for Contra-rotating Propeller Ships”, Society of Naval Archi-tects of Japan, Vol. 183, pp. 35-44, 1998

(in Japanese).

Jang H., Mahesh, K., 2012, “Large Eddy Simulation of Crashback in Ducted Pro-pellers with Stator Blades”, 29th Sympo-sium on Naval Hydrodynamics, Gothen-

burg, Sweden.

Jang H., Mahesh, K., 2013, “Large Eddy Simulation of Flow around a Reverse Ro-

tating Propeller”, Journal of Fluid Me-chanics, 729: 151-179.

Johannsen, C., Wijngaarden, E., 2012, “Inves-tigation of Hull Pressure Pulses, Making Use of Two Large Scale Cavitation Test Facilities”, Proceedings of the Eighth In-

ternational Symposium on Cavitation (CAV 2012)

Jang, Jin-Yul, 2013, “Study on Performance

of Contra-Rotating Propeller”, 2013, Mas-ter Thesis, Pusan National University, Ko-

rea.

Kandasamy. Optimization of Waterjet Pro-pelled High Speed Ships – JHSS and Delft

Catamaran. FAST 2011

Kane C., http://www.compositecarbonfiberprop.com/qinetiq.htm, “World's Largest Compos-

ite Carbon Fiber Propeller”, 2001

Kawakita C., “Study on Marine Propeller Running in Bubbly Flow”, Third Interna-tional Symposium on Marine Propulsors

SMP’13, Launceston, Australia May 2013.

Kawakita, C., Takano, S., Kodan, Y., & Mizo-kami, S., (2011) ‘Experimental Investiga-tion of the Behavior of Injected Air on the Ship Bottom and its Influence on Propel-

ler’. Journal of the Japan Society of Naval, Architects and Ocean Engineers, Vol.12,

pp.43-50.

Kawamura, T., Ouchi, K., Nojiri, T., 2012, “Model and full scale CFD analysis of

propeller boss cap fins (PBCF)”, Journal of Marine Science and Technology, 0948-

4280.

Kawamura T., Ouchi K., Takeuchi S., 2013, “Model and full scale CFD analysis of

propeller boss cap fins (PBCF)”, Third In-ternational Symposium on Marine Propul-

sors SMP’13, Launceston, Australia.

121121

Kim K.-S., Kim Y.-C., Kim J., Lee Y.-Y., Ahn H., Yim G., Park I.-R., 2012, “A Study on

the Energy Saving Fin Attached on the Ship Stern”, 10th International Conference on Hydrodynamics, St. Petersburg, Russia.

Kim, S.-E., 2009, “A Numerical Study of Un-

steady Cavitation on a Hydrofoil”, Proc., 7th International. Symposium on Cavita-

tion CAV2009, Ann Arbor, Michigan, USA.

Kinnas S.A., Jeon C.-H., Purohit J., TianY., 2013, “Prediction of the Unsteady Cavitat-

ing Performance of Ducted Propellers Subject to an Inclined Inflow”, Third In-

ternational Symposium on Marine Propul-sors SMP’13, Launceston, Australia.

Korkut, E., 2006, “A Case Study for the Ef-

fect of a Flow Improvement Device (a Wake Equalizing Duct) on Ship Powering

Characteristics”, Ocean Engineering, Vol.33, pp.205-218

Koushan K., Spence S., Savio L., 2011, “Ven-

tilated Propeller Blade Loadings and Spindle Moment of a Thruster in Calm

Water and Waves”, Second International Symposium on Marine Propulsors

SMP’11, Hamburg, Germany.

Krasilnikov V. I., 2013, “Self-Propulsion RANS Computations with a Single-Screw Container Ship”, Third International Sym-

posium on Marine Propulsors SMP’13, Launceston, Australia.

Li, D-Q, Grekula,M., Lindell, P. and

Hallander., J. (2012). Prediction of Cavita-tion for the INSEAN Propeller E779A Operating in Uniform Flow and Non-

Uniform Wakes. Proceedings of 8th Inter-national Symposium on Cavitation,

CAV2012, Singapore.

Lu, N. Svennberg,N., Bark, G. and Bensow, R. (2012) Numerical Simulations of the

Cavitating Flow on a Marine Propeller. Proceedings of 8th International Sympo-

sium on Cavitation, CAV2012, Singapore.

Maciel P., Koop A., Vaz G., 2013, “Modelling Thruster-Hull Interaction with CFD”,

OMAE ASME 32nd International Confer-ence on Ocean, Offshore and Arctic Engi-

neering, Nantes, France.

Manudha T. Herath, B. Gangadhara Prusty, G. H. Yeoh, M. Chowdhury and Nigel St.John, “Development of a shape-

adaptive composite propeller using bend-twist coupling characteristics of compos-ites”, Third International Symposium on Marine Propulsors SMP’13, Launceston,

Australia May 2013.

Mattheijssens, J., Marcel, J.-P., Bosschaerts, W., Lefeber, D., 2012, “Oscillating foils for ship propulsion”, 9th National Con-

gress on Theoretical and Applied Mechan-ics, Brussels, 9-10-11.

McDonald. A Tool for Analyzing and Com-

paring Tri-SWACH, Monohull and Trima-ran Hullforms. FAST 2011

Mewis, F., Guiard, T., 2011, “Mewis Duct® -

New Developments, Solutions and Con-clusions”, Second International Sympo-

sium on Marine Propulsors SMP’11, Hamburg, Germany.

Mosaad M.A., Gaafary M.M., El-Kilani H.S.,

Amin I.A., Power prediction of WIG craft. International Shipbuilding Progress 58,

pp219-238, 2012.

122122

Motley M.R. and Young Y.L., “Scaling of the Transient Hydroelastic Response and Fail-ure Mechanisms of Self-Adaptive Com-posite Marine Propellers”, International Journal of Rotating Machinery Volume

2012.

Nielsen, J. R., Shin, K. W., Lundgren, E., Faghani, F., 2012, “Combined Kappel

propeller and rudder bulb system for im-proved propulsion efficiency”, MAN Die-

sel & Turbo, Frederikshavn, Denmark

Ohmori, T., Ochi, F. and Nagaya, S., 2013, “A Study on the Influence of Propeller Open Boat in Reversed Configuration Propeller Open Water Test”, Conference Proc. Japan

Society of Naval Architects and Ocean Engineers, Vol. 17, Nov. 2013, Japan.

Ohmori T., Sudo Y., “Numerical and Experi-

mental Study on Wake Scaling of Twin-Skeg Ships”, The Japan Society of Naval

Architects and Ocean Engineers. 2013

Ouchi, K, 1989, “Research and Development of PBCF –Improvement of Flow from

Propeller Boss”, The Proceedings of the International Symposium on Ship Resis-

tance and Powering Performance ’89 (ISRP), Shanghai.

Ouchi K., Uzawa K, Kanai A. and Katori M.,

“Wind Challenger - the Next Generation Hybrid Sailing Vessel”, Third Interna-

tional Symposium on Marine Propulsors SMP’13, Launceston, Australia May 2013.

Palm M., Jürgens D., Bendl D., “Numerical

and Experimental Study on Ventilation for Azimuth Thrusters and Cycloidal Propel-lers”, Second International Symposium on

Marine Propulsors ,SMP'11, Hamburg, Germany, June 2011 ,

Park H.-G., Choi J.-K., Kim H.-T., 2013, “An Estimation Method of Full Scale Perform-ance for Pulling Type Podded Propellers”, Third International Symposium on Marine Propulsors SMP’13, Launceston, Australia.

Pecoraro A., Di Felice F., Felli M., Salvatore

F., Viviani M., Propeller-hull interaction in a single-screw vessel, Third International

Symposium on Marine Propulsors SMP’13, Launceston, Tasmania, Australia,

May 2013

Politis, G., Tsarsitalidis, V., 2012, “Flexible Oscillating Duct: An approach to a novel propulsor”, Applied Ocean Research, Vol.

36, 99.36-40.

Pospiech P., “VOITH’s New Propulsion Sys-tem: The Voith Linear Jet (VLJ)”,

http://articles.maritimepropulsion.com/, November 23, 2012

Poul, A., 1997, “A Comparative Study of

Conventional and Tip-Fin Propeller Per-formance”, Twenty-First Symposium on Naval Hydrodynamics 1996, Trondheim,

Norway, pp.930-945.

Quereda, R., Veikonheimo, T., Pėrez-Sobrino, M., Ponce, J., Sánchez-Caja, A., Masip, J., González-Adalid, J., Uriarte, A., Nijland, M. and Kokkia, K., 2012, “Model Testing and scaling for CRP POD”, Proc. ICHD10,

Oct. 2012, St. Petersburg, Russia.

Rijpkema D., Starke B., Bosschers J., 2013, “Numerical Simulation of Propeller-Hull Interaction and Determination of the Ef-

fective Wake Field Using a Hybrid RANS-BEM Approach”, Third International

Symposium on Marine Propulsors SMP’13, Launceston, Australia.

123123

Rung T., Wokner K., Manzke M., Brunswig J., Ulrich C., Stuk A., "Challenges and Per-

spectives for Maritime CFD Applications", Jahrbuch der Schibautechnischen Gesell-

schaft, 103:127-139, 2009.

Sakamoto N, Kawanami Y, Uto S., “Compu-tational Study for Single-Screw and Twin-Skeg Container Ships in Full Scale by Un-structured Grid based RANS Solver”, The

Japan Society of Naval Architects and Ocean Engineers, 2011

Sakamoto N., Kawanami Y., Uto S., Sasaki N.,

2013, “Estimation of Resistance and Self-Propulsion Characteristics for Low L/B Twin-Skeg Container Ship by a High-

Fidelity RANS Solver”, Journal of Ship Research, 57(1).

Sampson R., Atlar M., St John J.W. & Sasaki

N., “Podded propeller ice interaction in a cavitation tunnel”, Third International

Symposium on Marine Propulsors SMP’13, Launceston, Australia May 2013.

Sampson R. & Atlar M., 2013, “Prediction of

the effect of cavitation on propeller ice in-teraction”, The 3rd International Confer-ence on Advanced Model Measurement

Technology for the EU Maritime Industry, AMT’13, Gdansk, Poland.

Sanchez-Caja, A., Gonzalez-Adalid, J., Perez-

Sobrino, M. and Saito, I., 2012, “Study of End-Plate Shape Variations for Tip Loaded Propellers Using a RANSE Solver”, 29th Symposium on Naval Hydrodynamics,

Gothenburg, Sweden.

Sánchez-Caja A., González-Adalid J., Pérez-Sobrino M., Saisto I., 2014a, “Evaluation of Endplate Impact on Tip Loaded Propel-ler Performance Using a RANSE Solver”, International Shipbuilding Progress, 61:

103~128.

Sánchez-Caja A., Martio J., Saisto I., Siik-onen T., 2014b, “On the Enhancement of

Coupling Potential Flow Models to RANS Solvers for the Prediction of Propeller Ef-fective Wakes”, Journal of Marine Science and Technology, published online in Feb-

ruary 2014.

Sánchez-Caja A., González-Adalid J., Pérez-Sobrino M., Sipilä T., 2014c, “Scale Ef-fects on Tip Loaded Propeller Perform-

ance Using a RANSE solver”, Ocean En-gineering, http://dx.doi.org/10.1016/

j.oceaneng.2014.04.029.

Sánchez-Caja A, Pérez-Sobrino M., R. Quereda M.R., Nijland N., Veikonheimo

T., González-Adalid J., Saisto I. and Auriarte A., “Combination of Pod, CLT and CRP Propulsion for Improving Ship

Efficiency: the TRIPOD project”, A Third International Symposium on Marine Pro-pulsors SMP’13, Launceston, Australia

May 2013.

Sasaki, N., Kawanami, Y., Ukon, Y., Kano, T. and Tomizawa, S., 2006, “Model test pro-cedure and analysis of hybrid CRP POD system”, Proc. 2nd International confer-

ence on technical advances in podded pro-pulsion (T-POD), Nantes, France.

Sasaki, N., Kuroda, M., Fujisawa, J., Imoto, T.

and Sato, M., 2009, “On the Model Tests and Design Method of Hybrid CRP Pod-ded Propulsion System of a Feeder Con-tainer Ship”, Proc. The First International

Symposium on Marine Propulsors SMP’09, June 2009, Trondheim, Norway.

Scherer., “Hydrodynamics of surface piercing

outboard and sterndrive propulsion sys-tems”. FAST Conference 2011

124124

Schuiling B., Lafeber F. H., van der Ploeg A., van Wijngaarden E., "The Influence of the

Wake Scale Effect on the Prediction of Hull Pressures due to Cavitating Propel-

lers", Second International Symposium on Marine Propulsors SMP’11, Hamburg,

Germany, June 2011.

Schulze, "An improved Z-drive with Contra-Rotating Propellers for High-Speed Appli-

cations", FAST Conference 2011

Sileo L., Steen S., 2011, “Lateral Force and Turning Moment on a Reversing Ship”,

Second International Symposium on Ma-rine Propulsors SMP’11, Hamburg, Ger-

many.

Sipilä, T. and Siikonen, T. (2012). RANS Pre-dictions of a Cavitating Tip Vortex. Pro-ceedings of 8th International Symposium

on Cavitation, CAV2012, Singapore.

Skejic,R.,2012,”Effective Power Require-ments for High Speed Catamarans in Calm

Water During the Preliminary Design Stage”, NAV 2012 - 17th International Conference on Ships and Shipping Re-

search.

Streckwall H, Greitsch L., Scharf M., “An advanced Scaling Procedure for Marine Propellers”, Third International Sympo-

sium on Marine Propulsors SMP’13, Launceston, Australia May 2013.

Stuck A., Kroger J., Rung T., "Adjoint-ased

Hull Desogn for wake optimization”, Ship Technology Research, Vol. 58, No.1, Jan-

uary 2011.

Taketani T., Kimura K., Ando S., Yamamoto K., “Study on Performance of a Ship Pro-peller Using a Composite Material”, Third International Symposium on Marine Pro-pulsors SMP’13, Launceston, Australia

May 2013.

Tomohiro Takai, MamvannanKandasamy, Frederick Stern, Verification and valida-tion study of URANS simulations for an axial waterjet propelled large high-speed ship, J Mar Sci. Tech. (2011) 16: 434-447

Tsai,J.,Tse,C.,Wang,S.,Yeh,D., 2005, “Exper-

imental Study on the Hydrodynamic Performace of Two-Phase Nozzle for

Waterjet propulsion System”, FAST 2005, St-Petersburg, Russia.

Ueda, N., Oshima, A., Unseki, T., Fujita, S.,

Takeda, S. and Kitamura, T., 2004, “The First Hybrid CRP-POD Driven Fast

ROPAX Ferry in the World”, Mitsubishi Heavy Industries, Ltd. Technical Review

Vol.41 No.6.

Van, S.-H., Kim, M.-C., Lee, J.-T., 1993, “Comparison of Powering Performance Prediction Methods for Preswirl Stator

Propulsion System”, Proceedings of SNAK, Spring Meeting.

Veikonheimo T., 2006, “Solving problems of

propeller model scaling”, The naval Archi-tect, pp 51-53.

Verma A., Jang H., Mahesh K., 2012, “The

Effect of an Upstream Hull on a Propeller in Reverse Rotation”, Journal of Fluid

Mechanics, 704: 61-88.

Verhulst M. A., 2012, “Statistical Analysis of Model-Full Scale Correlation for Propul-

sion Tests and the Role of CFD in Ex-trapolation”, 10th International Conference on Hydrodynamics, St. Petersburg, Russia.

125125

Vesting F., Johansson R., Bensow R. E., 2013, “Parameter Influence Analysis in Propeller Optimisation”, Third International Sym-posium on Marine Propulsors SMP’13,

Launceston, Australia.

Villa D., Gaggero S., Brizzolara, S., 2012, “Ship Self Propulsion with Different CFD Methods: From Actuator Disk to Viscous Inviscid Unsteady Coupled Solvers”, 10th International Conference on Hydrodynam-

ics, St. Petersburg, Russia.

Weber, "Application of the New FORTJES Z-drive from REINTJES on Planing Vessels",

FAST 2011

Wu, X., S., Choi, J-K., Hsiao, C-T., Chachine G.L., 2010, “Bubble Augmented Waterjet Propulsion: Numerical and Experimental Studies”, 28th Symposium on Naval Hy-drodynamics, Pasadena, California, USA

Wu, X., Singh, S., Choi, J-K., Chachine G.L.,

2012, “Waterjet Thrust Augmentation us-ing High Void Fraction Air Injection”,

29th Symposium on Naval Hydrodynam-ics, Gothenburg, Sweden

Xia L., Lundberg J., Bensow R. E., 2012,

“Performance Prediction of a Nozzle Pro-peller”, 29th Symposium on Naval Hydro-

dynamics, Gothenburg, Sweden.

Yakovlev, A. Y., Sokolov, M. A., Marinich, N. V., 2011, “Numerical Design and Experi-

mental Verification of a RIM-Driven Thruster”, Second International Sympo-

sium on Marine Propulsors SMP’11, Hamburg, Germany.

Yasuhiko, I, Tadaaki, K., Shigeki, N., Fumito-shi, O., 2011, “Energy-Saving Principle of the IHIMU Semicircular Duct and Its Ap-plication to the Flow Field Around Full Scale Ships”, IHI Engineering Review,

Vol. 44, No. 1, pp.17-22.

Young Y.L., “Hydroelastic behaviour of flexi-ble composite propellers in wake inflow”, 16th International Conference on Compos-

ite material, 2007

Young Y.L.,”Dynamic hydroelastic scaling of self-adaptive composite marine rotors”,

Elsevier Ltd., Composite Structures Jour-nal 92, pp 97–106, 2010.

Young Y. L, 2012, “Analysis & Scaling of

Flexible Composite Marine”, Presentation at the 3rd Propulsion Committee meeting,

Pusan.

Young Y.L., Motley M.R., 2011, “Influence of Material and Loading Uncertainties on the

Hydroelastic Performance of Advanced Material Propellers “, Second Interna-

tional Symposium on Marine Propulsors SMP’11, Hamburg, Germany.

Yu H., Kong W., Cai R., Wang J., Zhang Y.,

2013, “Numerical Study of Energy-saving Mechanism of Duct on a VLCC with

Real-geometry Propeller”, Third Interna-tional Symposium on Marine Propulsors

SMP’13, Launceston, Australia.

Zhang N., Chen X., Xie H., 2012, “Numerical Simulation of Hull/Propeller Interaction of Submarine”, 10th International Conference on Hydrodynamics, St. Petersburg, Russia.

Zheng,Y.,Dong,W.,2012,”Hull Form Design

and Preliminary Evaluation of a Surface Planing Submersible Ship”, ISOPE-2012,

Rhodez, Greece

126126

Zondervan, G.-J., Holtrop, J., Windt, J., Ter-wisga, T., 2011, “On the Design and

Analysis of Pre-Swirl Stators for Single and Twin Screw Ships”, Second Interna-tional Symposium on Marine Propulsors

SMP’2011, Hamburg, Germany

Zurcher K., Bose N.,Binns J.R., Thomas G.,Davidson G. Design and commission-ing tests for waterjet self-propulsion test-ing of a medium speed Catamaran ferry using a single Demihull, Third Interna-

tional Symposium on Marine Propulsors SMP’13, Launceston, Australia.

127127

Manoeuvring Committee

Final Report and Recommendations to the 27th ITTC

1. INTRODUCTION 1.1 Membership

The 27th ITTC Manoeuvring Committee (MC) consisted of: Mr. Frans Quadvlieg (Chairman). MARIN,

The Netherlands. Dr. Guillaume Delefortrie (Secretary).

Flanders Hydraulics Research (FHR), Bel-gium.

Dr. Jonathan Duffy. Australian Maritime College (AMC), Australia.

Prof. dr. Yoshitaka Furukawa. Kyushu Uni-versity, Japan.

Dr. Pierre-Emmanuel Guillerm. Ecole Cen-trale de Nantes (ECN), France.

Dr. Sun-Young Kim. KRISO, South-Korea. Dr. Claus Simonsen. FORCE Technology,

Denmark. Prof. dr. Eduardo Tannuri. Escola

Politécnica da Universidade de São Paulo, Brazil.

Prof. dr. Xiao Fei Mao. Wuhan University of Technology (WHUT), China.

All members except Mr. Quadvlieg and Dr.

Kim were new members in the committee.

In addition to the official members, the MC had significant aid from the representative of the QSQ committee in the area of uncertainty analysis:

Dr. Michael Woodward. University of Newcastle upon Tyne, UK

1.2 Meetings

The committee has met four times during the course of their three years mandate: KORDI (now KRISO), Daejeon, South-

Korea from March 12 to 14, 2012; ECN, Nantes, France from November 19 to

21, 2012; FHR, Antwerp, Belgium from June 5 to 7,

2013, in conjunction with the conference on manoeuvring in shallow and confined wa-ters in Ghent;

WHUT, Wuhan, People Republic of China, from March 3 to March 5, 2014, in conjunc-tion with a seminar on manoeuvrability.

During all meetings, all members were pre-

sent. 2. TASKS AND REPORT STRUC-

TURE

The following lists the tasks given to the 27th MC together with explanation on how the tasks have been executed.

128128

Task 1. Update the state-of-the-art for predicting the manoeuvring behaviour of ships emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. the potential impact of new technologi-cal developments on the ITTC,

b. developments in manoeuvring and course keeping in waves,

c. new experiment techniques and ex-trapolation methods,

d. new benchmark data e. the practical applications of computa-

tional methods to manoeuvring predictions and scaling.

f. the need for R&D for improving meth-ods of model experiments, numerical model-ling and full-scale measurements.

g. the effects of free surface, roll, sinkage, and trim in numerical simulation of manoeu-vring.

This task has been achieved by an extensive discussion of the publications which were is-sued around the world. The particularly inter-esting technique is CFD, which received spe-cial attention in this report. The effects of free surface, roll, sinkage and trim have been dis-cussed.

Manoeuvring and course keeping in waves has received special attention. The criteria proposed by IMO are followed and interpreted. Realising that the present day numerical meth-ods are insufficient, this has also emerged as a separate section on manoeuvrability in waves in the report.

New benchmark data has been pro-actively pursued. These efforts are discussed in the benchmark section.

Task 2. Review ITTC Recommended Procedures relevant to manoeuvring and

a. Identify any requirements for changes in the light of current practice and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

The procedures have been reviewed and updated where needed, as discussed in the sec-tion on procedures.

Task 3. Complete the work on the Pro-cedure 7.5-02-06-04, Uncertainty Analysis; Forces and Moment, Example for Planar Mo-tion Mechanism Test, based on ISO approach. The present procedure 7.5-02-06-04 and the subsection on uncertainty analysis in the Pro-cedure 7.5-02-06-02, Captive Model Test Pro-cedure, prepared by the 23rd ITTC are based on the ASME approach. In view of the work al-ready carried out for the Procedure 7.5-02-06-04, consider to keep the elaborated ASME ex-ample as one of the Appendices to the to-be-renewed 7.5-02-06-04.

The procedure for UA of captive tests has been significantly reviewed. This is discussed in the section on procedures.

Task 4. Based on results of the SIM-MAN workshop held in 2008 and its next edi-tion, continue the already initiated work to generate a guideline on verification and valida-tion of RANS tools in the prediction of ma-noeuvring capabilities. Liaise with the QSG with respect to definitions of Verification and Validation.

A guideline for the use of CFD solutions for manoeuvring predictions is created. This is dis-cussed in the section on procedures.

Task 5. Restricted waters: a. Produce a guideline for experimental

methods. b. Complete the initiated one for numeri-

cal methods which may serve as a basis for

129129

recommended procedures for manoeuvring in restricted waters.

The guideline for experimental methods was integrated with the procedures for free running model tests and captive model tests.

Task 6. Free running model tests: a. Update the Procedure 7.5-02-06-01,

Free Running Model Test (FRMT) Procedure, in particular to include objective statements on the initial conditions of free manoeuvring model tests.

b. Elaborate the already initiated proce-dure on uncertainty analysis for free running manoeuvring model tests, including an exam-ple.

The procedure for FRMT is updated and a guideline on uncertainty for FRMT has been created. Details are provided in the section on procedures.

Task 7. Scale effects in manoeuvring: a. Report on knowledge and collect, ana-

lyse and summarize data on scale effects for manoeuvring predictions.

The work conducted on scale effects is in-cluded in a separate section.

Task 8. Review developments in meth-ods and draft a validation procedure of com-bined manoeuvring and seakeeping with re-spect to simulation. Liaise with the Seakeeping Committee and the Stability in Waves Commit-tee.

The methods are reviewed. It is too early to create a validation procedure for simulations for combined manoeuvring and seakeeping. The Seakeeping Committee and the Stability in Waves Committee did not have tasks to address this.

Task 9. Support the organisation of a second SIMMAN workshop.

The members of the committee actively or-ganise and support this workshop, which will now be held in December 2014.

Task 10. Manoeuvring criteria and rela-tions to IMO:

a. Report on manoeuvring criteria for ships not directly covered by IMO like POD and waterjet driven vessels, naval ships, inland ships, HSMV, etc.

b. Study possible criteria for manoeuvring at low speed and in shallow waters and if war-ranted communicate findings to IMO.

A dedicated section is created on manoeu-

vring criteria and in particular a section is cre-ated to discuss non-IMO related criteria which are in use. 3. USING EXPERIMENTS AS A

TOOL TO ADVANCE THE KNOW-LEDGE IN MANOEUVRING

3.1 In Deep Unrestricted Water

Design Improvements. Recent studies have been undertaken to investigate the influence of ship design and operational aspects on ma-noeuvring characteristics.

Physical model scale experiments were conducted by Park et al. (2011) to measure the running trim of a high speed vessel at zero drift angle. Small drift angle tests were conducted to assess course keeping ability. For the zero drift angle tests vertical motions were measured to investigate the bow down trim at high speeds and how this can be reduced to move the lateral centre of pressure toward the stern to improve course keeping ability. The small drift angle

130130

tests were conducted for the naked hull and with a transom wedge. It was found that the addition of the transom wedge moved the lat-eral centre of pressure toward the stern and im-proved the course stability.

Hirata et al. (2012a, 2012b) presented re-sults from full scale trials and model scale ex-periments to assess the effect that trim has on the manoeuvring performance of the training ship Toyoshio Maru, an azimuth propeller ves-sel. The full scale tests consisted of turning cir-cle and zigzag manoeuvres for three load con-ditions, one even keel condition and two condi-tions trimmed by the stern. The model scale experiments consisted of oblique towing tests and circular motion tests in the even keel and the largest trimmed by the stern load condi-tions. The results showed that the vessel exhib-ited course instability for all load conditions, however trimming by the stern improved the course stability and remarkable improvement was seen in the Y’(β) and the N’(β) derivatives.

Free running physical scale model experi-ments were conducted by Miyazaki et al. (2013) to determine the manoeuvring charac-teristics of a KCS container ship model with a static heel angle. The yaw rate and drift angle during turns with a static heel angle were quan-tified and discussed.

Kang et al. (2011) investigated the ma-noeuvring and powering benefits of aligning twin rudders with the inflow of the propeller stream of a single propeller vessel. They con-ducted free running turning and zigzag physical scale model experiments at Osaka University. They showed that course keeping stability was increased by the non-zero rudder angle; how-ever the turning ability was reduced.

Yasukawa et al. (2011) reported on captive model tests to measure the hydrodynamic force coefficients on a twin screw, twin rudder ferry

hull form with a bow thruster. The force de-rivatives and coefficients were determined ac-cording to the MMG model procedure using the equivalent single rudder method to reduce complexity. The hydrodynamic force coeffi-cients were presented for the hull, propeller and rudder together with the hull force characteris-tics due to bow thruster operation.

Yasukawa et al. (2012a) investigated the hydrodynamic force characteristics of a cata-maran with asymmetrical demi-hulls. Physical scale model experiments were conducted with different demi-hull separations, rudder angles and propeller loads. The demi-hull separation was shown to have little effect on the rudder normal force and the smallest demi-hull separa-tion provided the best course keeping perform-ance. Numerical simulations of a turning circle manoeuvre were conducted and compared to trial results. The steady turning radius showed good correlation, while the advance and tactical diameter were over estimated.

In tight turning manoeuvres involving twin/multi screw vessels, the load in each pro-peller shaft can vary significantly, which can influence the manoeuvring behaviour of the ship. Coraddu et al. (2013) investigated the propeller loads on a twin screw vessel using free running model scale experiments and nu-merical simulations. They investigated the ef-fect of constant propeller RPM, constant power and constant torque on propeller loads. They conducted zigzag, turning circle and Dieudon-née spiral tests and compared the experimental results to numerical manoeuvring simulations, which correlated well and showed the effect of the asymmetrical propeller loading.

Towed Stability. Towed stability receives more and more attention due to the many FPSO’s which are nowadays towed over the oceans. Yang & Wada (2012) have been inves-tigating both numerically and experimentally a

131131

better way to investigate the actual limits of towed stability. They concluded that there is quite a difference between towing in the tradi-tional way and towing using an actual tug in the basin. The numerical model had the capa-bilities to quantify the effect of the environ-mental forces on the towed stability. Nakayama et al. (2011) investigated the towed stability in (head) waves. A mathematical model was pro-posed which was validated using model tests. They show a relationship between peak loads and surge and pitch motions. Zotti (2013) con-ducted a study to investigate the directional stability of a barge being towed by a tug. A physical scale model barge was towed at vari-ous angles of attack up to 6 degrees. Forces on the barge model were measured to perform a directional stability analysis applicable to only small perturbations from the equilibrium condi-tion. The barge was tested in three configura-tions; without appendages, with a rudder and with two side skegs. The barge without ap-pendages demonstrated directional instability, i.e. it had the tendency to move transversely and to rotate on itself when acted upon by an external force. The barge with the central rud-der had little tendency to translate laterally, but a great tendency to rotate. The barge with skegs demonstrated little tendency to rotate and great tendency to translate laterally. Hong et al. (2013) present an overview of two different mathematical models that can be used for towed stability simulations: the MMG model for towed bodies by Fitriadhy & Yasukawa (2011a) and the cross flow drag model accord-ing to Wichers (1988). Coefficients for both models have been derived from captive model tests. Simulations were carried out using both models. By comparison of the simulated trajec-tories to model tests, the authors conclude that the cross flow drag model is easier to use, while giving practically the same results as the MMG model and model tests. Toxopeus et al. (2013b) show how CFD is used to perform vir-tual captive tests to predict the towed stability

of a variety of skeg shapes, and as such CFD is able to balance the resistance and the towed stability in order to achieve good directional stability with minimum barge resistance. 3.2 In Shallow Water

General. It is necessary to validate ship-handling simulation models for use to approve new waterway designs. Böttner et al. (2013) presented experiments with two aims: to detect the influence of under keel clearance on turning and course keeping ability and to sound the limitations of the manoeuvring model imple-mented in a simulator when applied to ma-noeuvring in shallow waterways. A remotely controlled free sailing model was used to per-form IMO standard zigzag manoeuvres in the wave basin of BAW in Hamburg at different initial speeds as well as at a range of water lev-els targeting a representative range of under keel clearances. Data from the manoeuvring trials were proven to be a good base for deter-mination of coefficients. Another finding was the impossibility to find a suitable set of coeffi-cients for a broad range of either water depths or speeds in shallow water.

False Bottom. The use of false bottoms to execute shallow water tests still demands vali-dation and analysis. The flow field at the bor-ders of the false bottom depends on the dimen-sions of the tank and on the size of the structure and the apparatus used to support the false bot-tom. If there is not enough space for the water above the false bottom to flow when the ship is passing, the pressure distribution can be dis-turbed and the shallow water effects will not be accurately measured. Only a few papers dem-onstrated such concern, presenting a validation of the false bottom dimensions and demonstrat-ing that they are properly designed for the ex-periment. An example is the work by Yeo et al. (2013), which describes a false-bottom facility

132132

built at the KRISO towing tank. The tank di-mensions are 200 m x 16 m x 7 m, and the false bottom is 54 m long and 10 m wide. Using this false-bottom facility, captive model tests were conducted with a 1:31 scale model of the KCS hull for three under keel clearances (h/T = 1.2, 1.5 and 2.0). The authors made a preliminary validation of the false bottom concept, aiming to verify the effects of the limited lateral size of the false-bottom. They compared static drift test results conducted along the mid-breadth line of the false-bottom and results from a static drift test conducted along the 1m biased-in-breadth line of the false-bottom. They con-cluded that the limit in the breadth of the false-bottom would not cause a significant effect on test results for cases in which the position bias in breadth (of the model) was within 1 m. Fur-thermore, based on this result, amplitudes of forced motion in dynamic tests of the bench-mark PMM tests were selected to be within 1 m. This kind of verification must be carried out when using false-bottoms to perform shallow water experiments. The benchmark test results obtained in these experiments will be provided to participants of the SIMMAN 2014 confer-ence to add to data for subsequent studies.

Béguin et al. (2013) presented the experi-mental database for three different models (Wigley Hull, Container Carrier and River Barge), with a combination of ship speed and water depth. It focused on additional hydrody-namic forces, as well as squat and vertical mo-tions (trim and sinkage) of hulls sailing straight-ahead in shallow water, as a function of Froude number. The test facility is 138 m x 5 m. A double bottom made of 28 removable plates of 1 m width, firmly fixed to a scaffold structure was used to change the water depth on a 28 m length section of the towing tank. One problem addressed by the authors was re-lated to the time window available to obtain the steady state results in the shallow water sec-tion. This is important for higher speeds and

ship models with large inertia. The authors did not discuss the problems related to the flow at the lateral boundary of the false bottom. This may be a concern due to the small width of the tank, and may play some role in the shallow water effects. 3.3 In Restricted Water

Canal Navigation. Model scale experi-ments were conducted by Iseki & Kawamura (2011) to investigate the rudder angle required to counter ship-bank interaction. The experi-ments were conducted in a circulating water channel and involved adjusting the oblique an-gle of the ship model and the rudder angle close to a lateral bank to find the equilibrium point. The measured values for equilibrium were compared against the theoretical value of the Next Generation Fairway Design Standard, which showed some possibility of underestima-tion for the safety margins of the fairway.

Iseki & Takagi (2013) conducted experi-ments with a propelled scale model to deter-mine the equilibrium position of a ship operat-ing in the vicinity of a bank wall. The propeller RPM, oblique towing angle and rudder angle were varied for a range of water depth to draft ratios and distances off the bank. Ship speed was shown to have little influence on the re-quired rudder angle.

Ibaragi et al. (2012) reported on physical scale model experiments to determine the ef-fect that channel width, drift angle, under keel clearance and distance from a lateral bank has on the sway force and yaw moment of two dif-ferent hull forms in restricted water. The cap-tive model tests were conducted at the Seakeeping and Manoeuvring Basin at Kyushu University. A new empirical formula was pre-sented to predict the sway force and yaw mo-ment due to the drift angle, separation from the

133133

bank and under keel clearance. The formula represented the general trends but showed poor quantitative accuracy.

To investigate the behaviour of a ship in re-stricted water, Sano et al. (2012) conducted physical scale model experiments to quantify the sway force, yaw moment and rudder force acting on a vessel due to the effects of a bank, drift angle and rudder angle. The captive model experiments were conducted in a scale model channel of a Japanese port using a ship model fitted with a propeller operating at the self-propulsion point. The experiments were con-ducted at various water depth to draft ratios. The forces and moments induced by the rudder angle, bank effects and drift angle were exag-gerated at low water depth to draft ratios. New equations were presented to determine whether a vessel is directionally stable when operating in restricted water.

Squat. Delefortrie et al. (2010) presented a mathematical model to predict squat of con-tainer carriers operating in muddy navigation areas. The new squat formulae are based on an extensive experimental research program car-ried out at the Flanders Hydraulics Research Towing Tank over the period 2001 to 2004 to investigate the manoeuvring behaviour of deep drafted vessels in muddy bottom areas. It was found that the sinkage over a muddy bottom is mostly less than a solid bottom, but the trim can be larger when manoeuvring in muddy ar-eas.

An extensive captive model test program was undertaken by Lataire et al. (2012a) to in-vestigate squat with a scale model of the KVLCC2. Tests were carried out for canals with rectangular cross section at different water depths, widths of the canal section, model lat-eral position in the canal and forward speeds (2-16 knots where possible). The measure-ments were used to validate a mathematical

model, which takes into account the forward speed, propeller action, lateral position in the fairway, total width of the fairway and water depth.

Full scale motion measurements of vessels transiting the Columbia River Bar have been obtained by Lesser & Jordan (2013). One of the aims was to quantify under keel clearance in moderate to high seas. Two methods were used to measure the vessel motion: (1) high-precision Trimble GNSS (GPS) units

mounted at the bow and bridge wings with an additional unit mounted to a pilot "chase" boat to measure the sea level;

(2) an iHeave unit in winter to measure the mo-tions due to extreme weather.

Numerical simulations were also conducted us-ing the Delft3/SWAN numerical model and DUKC software. No clear ‘rule of thumb’ was identified to eliminate risky transits; however several aspects affecting the transits were iden-tified.

Briggs et al. (2013) compared full scale Differential Global Positioning System (DGPS) measurements of ship squat for four different vessels in the Panama Canal to predictions us-ing a selection of empirical formulae and nu-merical techniques. They found that the predic-tion techniques provided reasonable results and can be used with confidence in deep draft channel design.

Crabbing. For cruise vessels and ferries, harbour manoeuvring is an important manoeu-vring case. These ships are equipped with bow and stern thrusters, and the main propeller(s) are operating in push-pull model. Usually, berthing (going to the quay) and unberthing (leaving the quay) are investigated. Lee et al. (2011) investigated experimentally a twin screw vessel with bow and stern thrusters. Based on the experiments, a modular mathe-matical model was developed for the complex

134134

flow phenomena for different distances be-tween the ship and quay and also for different water depths. The fine mesh of different dis-tances to the quay at which captive tests were performed is particularly interesting.

Kwon et al. (2013) investigated the limiting operational conditions of a cruise vessel with 3 bow thrusters and 2 pods. Using experiments, the forces generated by the actuators were ob-tained. These were compared to the wind loads obtained by CFD. Model tests were carried out in deep and shallow water. For berthing, the re-sults were similar in deep and shallow water, but for unberthing, there were significant dif-ferences measured.

Locks. During the last couple of years there has been a worldwide growing interest in the study of ship behaviour in locks, mainly due to the construction of new locks or the modernization of existing locks to cope with an ever increasing ship size. The most impressive example is the construction of the Third lane of the Panama Canal (2015) for which several ex-perimental studies have been carried out. For that reason PIANC has started Working Group 155 (Thorenz, 2013) to study the ship behav-iour in locks and approaches to locks. The ship behaviour in locks was also the main topic of the latest International Conference on Ship Manoeuvring Behaviour in Shallow and Con-fined Water (2013). An overview of significant locks and the challenges to enter them is de-scribed in a practical way by Eloot & al. (2013).

Ships are subject to forces during entry and exit manoeuvres, but also during the filling and emptying process while being in the lock chamber. The latter is however not considered to be a manoeuvring topic and is not treated in this report. During a lock entry a ship is sub-jected to an increased resistance, which is well predicted by the six-waves-model described by

Vrijburcht (1988). His model proved to be use-ful for rectangular shapes such as barges, but needed improvement for slender ship’s hulls (Vergote et al., 2013). The improved six-waves-model has been used to calculate the water level elevation at the end of the lock. The results have been compared with measurements for the ship models of a New Panamax con-tainer ship and a bulk carrier.

Locks can be divided into two categories depending on whether an approach structure is present or not. While the latter provides a use-ful aid for alignment, its induced asymmetry must be counteracted by the ship’s available steering aids. A lateral force component and yawing moment also occur when a ship sails eccentrically in a symmetrical lock layout. In-sight into these asymmetries is provided by ex-perimental research, for instance the approach layout for the locks to the Panama canal (Dele-fortrie et al., 2009) or for the lock to IJmuiden (The Netherlands) (Kortlever & de Boer, 2013). In these two cases additional difficulties occur due to the exchange of fresh water with salt water during the levelling process and after the opening of the gates. Model tests and full scale trials for the West lock in Terneuzen (The Netherlands) were described by Verwilligen et al. (2012). The results of lock entry and exit tests can be implemented in a real-time ma-noeuvring simulator to evaluate the nautical qualities of the design of a new lock. An exam-ple of such an approach was discussed by Ver-willigen et al. (2013).

The above mentioned model scale tests were all carried out at FHR (Figure 1), who provided benchmark data to the scientific community (Vantorre & Delefortrie (2013), see section 5.4). During the latest International Conference on Ship Manoeuvring Behaviour in Shallow and Confined Water (2013) several papers were presented focussing on the com-parison between the benchmark data and nu-

135135

merical computations. Wang & Zou (2013) used an unsteady RANS solver with a dynamic mesh method, free undisturbed water level and user defined functions to define the ship motion in the lock. The lateral force and yaw moment were well predicted, while the longitudinal force was under predicted compared to the benchmark data. Lindberg et al. (2013) intro-duced a potential model for nearly real-time ship’s hydrodynamics and linear water waves calculations. The model has been tested with the New Panamax container carrier sailing into the lock, but the interaction with vertical ap-proach and lock walls is not yet well predicted by the model. De Loor et al. (2013) computed the effect of the exchange between fresh and salt water on a moored ship along a lock ap-proach wall and compared the results with the benchmark data. It was concluded that although the application of CFD is not (yet) feasible to predict absolute values with sufficient accu-racy, it can provide more insight in the physical processes.

Figure 1. Lock entry model scale test at FHR.

Other authors also developed numerical or

empirical codes, mainly focussing on lock en-try speed and sinkage. Henn (2013) enhanced an existing code for inviscid flows to enable a

real-time prediction of the ship’s velocity and the squat during lock entry and exit manoeu-vres. A coefficient was added to take account of the lock chamber frictional effects. The code was successfully compared with both experi-mental and full scale results. Spitzer & Soehngen (2013) gave a comprehensive over-view of lock entry and exit manoeuvres. They evaluated existing semi-empirical formulae with model tests and full scale trials. The nu-merous uncertainties of such formulae call for the need of additional physical model tests and CFD research. A specific type of lock entry manoeuvres is an entry in a ship lift. Li et al. (2013) conducted experimental research focus-sing on the squat measurement and the deriva-tion of a squat prediction formula for different ship lifts in China, such as the Three Gorges ship lift. 3.4 Ship-to-Ship Interaction

There has been a growing interest in ship to ship interaction issues, as evidenced by recent work on ship to ship transfer, tug – ship inter-action and ship passing scenarios.

Ship to Ship Transfer. Physical model scale experiments were conducted by Arslan et al. (2011) to investigate the flow around the paral-lel midship sections of two ships in a side by side lightering operation using PIV and dye in-jection. The results from the experiments were used to validate CFD predictions. The numeri-cal predictions generally showed good correla-tion with the experimental results.

Quasi-static and dynamic captive model tests were conducted by Lataire et al. (2012b) to simulate the interaction forces and moment due to a lightering operation of the KVLCC2 and a service ship. Different longitudinal and lateral positions of the service ship relative to the KVLCC2 model were tested. Both models

136136

were fitted with rudders and propellers (run-ning at their open water model self-propulsion point). New formulae were presented to predict the forces and moments experienced by the service ship due to the KVLCC2. The formulae correlated well in surge, sway and yaw.

Yasukawa & Yoshida (2011) investigated a simplified lightering operation by conducting physical scale model experiments using two Wigley parabolic hulls. The lateral separation, drift angle and rudder angle were varied. The tests were conducted with no stagger between the two ships (i.e. midships adjacent). The sway force and yaw moment was measured on each of the models along with the normal force on the rudders. The results from the experi-ments were compared to numerical predictions based on nonlinear lifting surface theory. The numerical predictions correlated reasonably well with the experiments with a few excep-tions.

Sano et al. (2013) reported on physical model scale experiments to investigate a ship to ship transfer manoeuvre. The hydrodynamic in-teraction surge force, sway force and yaw mo-ment were measured on two Wigley parabolic hulls in close proximity with rudders. The rud-der normal force was also measured on both models. The water depth to draught ratio, lat-eral clearance between the hulls, hull drift an-gle and rudder angle were varied during the test program. It was found that when a ship steered the interaction force acted not only on the own ship, but also induced an interaction force and moment on the ship alongside, which varied with water depth. The experimental results were used to validate numerical analyses using a nonlinear lifting body theory.

Tug-Ship Interaction. An investigation into tug-ship interaction was undertaken by Geerts et al. (2011). Physical model scale experiments were conducted to investigate the hydrody-

namic interaction forces experienced by an azimuth stern drive tug sailing in the vicinity of the bow of a Panamax container vessel. The in-teraction forces on the tug model were meas-ured for a range of relative positions and drift angles at multiple forward speeds. The forces were used as input to a fast-time simulation program to assess the required thrust and azi-muth angle to keep the tug at a fixed station. An assessment was made on the most suitable position to pass the tug towline.

Passing Ship Scenarios. Delefortrie et al. (2012) investigated the hydrodynamic forces and moments acting on a berthed ship due to different ship traffic scenarios. Captive physi-cal scale model experiments were undertaken to measure the forces and moments acting on a berthed ship due to a passing ship and due to multiple passing ship interaction, with different dock widths. The effect of a nearby swinging vessel was also investigated. The applicability of superposition theory was assessed for esti-mating the forces and moments experienced by a berthed ship due to multiple passing ships. It was concluded that while applicable in most cases, when under keel clearance or separation ratio is low, the superposition theory is less ac-curate. At low under keel clearances it was found that the forces due to a nearby swinging ship can be significant, even higher than realis-tic passing ship manoeuvres.

Duffy et al. (2011, 2013) and Denehy et al. (2012) reported on investigations into the in-fluence of waterway geometry, around berth geometry, berthed ship size and berth occu-pancy arrangement on the hydrodynamic inter-action forces and moments experienced by a berthed ship due to a passing ship. From cap-tive physical model scale experiments it was found that the different scenarios significantly influenced both the form and magnitude of the interaction forces and moments.

137137

Uliczka et al. (2013) conducted physical model scale experiments to measure the inter-action sway force and yaw moment for two scenarios in a narrow fairway: a containership passing a moored containership and two con-tainerships passing. Both head on and overtak-ing manoeuvres were investigated for both sce-narios, however the overtaking case with both ships moving was conducted with each ship travelling at the same speed sailing parallel. The results were incorporated into ship-handling simulators for the simulation of con-tainership manoeuvres in narrow fairways. 3.5 Special Experimental Techniques

This section focuses on some works that used special or non-conventional experimental techniques and arrangements to study ship ma-noeuvrability. Also, the application of new sys-tem identification (SI) techniques to derive models and coefficients from manoeuvring tests is presented.

Yoshimura et al. (2012) conducted a com-prehensive set of measurements of open water rudder tests in several exposing conditions, us-ing a large scale rudder model. The author’s in-tention was to obtain a better prediction of rud-der lift forces for ballast conditions, when a ship’s rudder may be partially exposed on the water surface. They verified that the actual as-pect ratio used for the prediction of rudder normal force must take into account the water surface at both sides of the rudder. Also, when only a small part of the rudder is above the wa-ter, the stall phenomenon does not appear and the maximum lift coefficient significantly in-creases. The influence of the ship’s loading condition on the manoeuvring characteristics has also been investigated by Hirata et al. (2012a, 2012b). The authors used full scale tri-als of a training ship to verify the influence of trim angle on the manoeuvrability of the ship.

Blendermann et al. (2011) report the results

of a combined numerical and experimental in-vestigation of the wind loads on a scale model of a passenger / car ferry, as well as a full-scale computation. The ship model (scale 1:150) was tested in two wind tunnels. The deviations be-tween the results in the two wind tunnels and the CFD computation were of the same order. Silva (2012) presented a comprehensive set of experimental tests for a supply boat for obtain-ing winds and current loads, in a wind tunnel and towing tank. The author also performed CFD calculations and obtained quite good agreement. The results indicated that CFD is a realistic and reliable alternative to wind tunnel model and towing tank tests for predicting static forces.

The manoeuvrability of an unusual vessel was studied by Ueno et al. (2011) using circu-lar motion tests. The submersible surface ship (SSS) is a new concept ship that avoids rough seas by going underwater using downward lift of wings and keeping residual buoyancy for safety.

The System Identification (SI) technique of

Extended Kalman Filter (EKF) has been used to estimate values of hydrodynamic coeffi-cients for a submarine from its full-scale ma-noeuvring sea trials data in the paper of Ray & Sen (2012). Data from sea trials with two sub-marines were used to identify the hydrody-namic coefficients. The authors provide advice for problems related to the robustness of the SI techniques applied to the identification of hy-drodynamic parameters from noisy full-scale data.

SI based on artificial intelligence was deeply investigated by Chinese researchers. They studied the Support Vector Estimation technique applied to AUV free-running tests (Xu et al. 2011), and obtained hydrodynamic

138138

derivatives similar to those obtained by tradi-tional captive PMM experiments. Extended analysis concerning AUV application is pre-sented by Xu et al. (2013). The technique was also used for ship model identification with good results, as shown by Zhang & Zou (2011c, 2013). More analysis and results for ship model identification is presented by Wang et al. (2013b). The authors also studied the in-fluence of the noise in the estimation of the hy-drodynamic parameters and applied a Wavelet Denoising technique to improve the results (Zhang & Zou, 2011b). The method was also applied to the estimation of a 4DOF mathe-matical model of ships using the roll planar motion mechanism (RPMM) test, adequate for the analysis of ship manoeuvring motion in waves (Wang et al., 2013b).

Neural Networks have also been applied to the SI of manoeuvring models, as presented by Zhang & Zou (2012) and Woo & Kim (2013).

Di Mascio et al. (2011) investigated predic-tion methods for the manoeuvrability of twin-screw naval vessels. Regression analysis, SI method, semi empirical corrections for the in-fluence of appendage and RANSE calculations are applied for analysis of the manoeuvring be-haviour of twin-screw ships. They concluded that the combination of the SI technique and RANSE calculations could be useful for reduc-ing research costs.

Revestido & Velasco (2012) proposed an identification scheme for nonlinear manoeu-vring models based on two steps with a gray box approach. On the first step, a suitable model structure is selected and initial parame-ters are estimated. Estimated parameters are re-fined using a nonlinear prediction error method on the second step.

Luo et al. (2011) applied support vector machines based SI to predict ship manoeu-

vring motion in the proximity of a pier. Ma-noeuvrability indices and the other parameters are identified taking group test results as the training sample.

Ahmed & Hasegawa (2013) conducted free running model tests of automatic ship berthing using an Artificial Neural Network (ANN) trained code. They found that the automatic berthing manoeuvre could be successfully im-plemented up to certain wind speeds once the appropriate teaching variables had been se-lected. 3.6 Improvements in Experimental Meth-

ods

Hexapods have become more common in hydrodynamic laboratories. In the past, these have been used as a tool in the investigation of sloshing and VIV (Vortex Induced Vibrations). The use of hexapods as a replacement of a tra-ditional planar motion mechanism seems an easy step. Up to now, only the work of de Jong & Keuning (2005) was published. Added mass and sway and yaw damping were measured on a segmented model. The results show that the analysis of tests (oscillations tests in waves with a segmented model) is an elaborate job. Nevertheless, the use of a hexapod as a com-plete replacement of a PMM alone implies that the oscillations that can be made are so-called small stroke oscillations: the maximum trans-verse excursions are in the order of ±0.5 m. This would be an important restriction. A better approach is to mount the hexapod under a transverse carriage (with the transverse carriage mounted on the main carriage). As such, the hexapod can be used as part of a large stroke oscillator. Such a set-up is installed in the Marintek facilities (Berget, 2011).

A second observation is the use of false bot-toms, which do not fit the whole basin, to in-

139139

vestigate the behaviour in shallow water. The use of false bottoms has to be considered care-fully. The shallow water PMM results for KVLCC1 and 2 that were obtained using a false bottom in the INSEAN basins were - after long discussion - rejected as benchmark data, as issues were raised concerning the accuracy of results obtained using the false bottom con-structed of removable plates.

The increasing international attention to-wards manoeuvrability encourages several smaller basins to investigate manoeuvrability issues. Yoon & Kang (2013) are reporting the installation of a CPMC in a basin of 20 x 14 m. They performed tests at a scale of 1:223. It is very clear that such techniques should only be used qualitatively for educational purposes and that results obtained in this way are of use to demonstrate that there are indeed scale effects. But besides the considerable effects of block-age, the accuracy of transducers and bottom flatness are of a different level due to the very small forces that need to be measured. Obvi-ously, the main concern with respect to scale effects is that the Reynolds numbers are so small that it is very likely that the flow around the hull is laminar which leads to a different flow pattern around the manoeuvring hull.

The desired increase of knowledge about the manoeuvrability in waves has led to using captive test techniques in waves with the objec-tive to create mathematical models for ma-noeuvring prediction in waves. Sung et al. (2012) reported on the application of this tech-nique to the KCS where PMM tests were per-formed in waves.

Some reported improvements in free sailing

techniques are twofold: the correction of the longitudinal scale effect by adding an air-propeller on the free running model as pro-posed by Ueno & Tsukada (2013). Mauro (2013) reported how the propellers need to be

controlled during free running model tests as a function of the instantaneous propeller load. The impact is considerable and it is indeed rec-ommended to consider the effect of propeller load on the manoeuvrability of the ship.

A new basin to carry out free running model tests was reported by Sanada et al. (2012) and Sanada et al. (2013). The basin at IIHR measures 40x20x3m³ and is equipped with wave makers and a xy carriage with a turntable. The carriage can follow a free ma-noeuvring model to perform free running ma-noeuvring tests in calm water or in waves. The carriage tracking system, the 6 DOF visual mo-tion capture system and the model release and capture system were extensively described. The capability to perform local flow measurements through PIV besides a semi-captive model al-lows the measurement of local flow fields for comparison to CFD results.

An important improvement in the experi-

mental techniques is the application of uncer-tainty analysis. Quadvlieg & Brouwer (2011) are applying this to free running model tests on KVLCC2. Woodward (2013) described how the uncertainty of the measurements of the forces and moments in captive model tests propagates to the manoeuvring derivatives. He applied this on KVLCC1. 4. USING SIMULATIONS AS A TOOL

TO ADVANCE THE KNOWLEDGE IN MANOEUVRING

4.1 In Deep Unrestricted Water

Using Viscous CFD Methods. One of the main advantages of CFD is its ability to pro-vide information about hydrodynamic loads and motions of the vessel together with detailed flow field information, which can help to un-

140140

derstand the flow physics related to manoeu-vring. Another advantage is that this type of simulation does not rely on model testing with physical scale models, which means that for in-stance the hull form or the rudder can be changed relatively easy. This is useful in the early design phase where CFD can help to in-vestigate manoeuvring related issues and help to improve the design. Therefore, CFD is used ranging from detailed flow studies (to learn about the features of the flow field) for predic-tion of hydrodynamic forces and moments to direct simulation of manoeuvres. This applies to both surface ships and submarines. It seems that in addition to the traditional RANS ap-proach, also Detached Eddy Simulation (DES) and Delayed Detached Eddy Simulation (DDES) have started to show up in practical applications.

In terms of flow field investigations Xing et al. (2012) made a very detailed RANS and DES based study of the bare hull of the KVLCC2 with different turbulence models in order to identify and study the generation and breakdown of the vortex structures around the hull in oblique flow at drift angles from 0 to 30 degrees. Comparison with model test results shows that many flow features are captured by the CFD solution. Amin & Hasegawa (2012) also study the flow around the KVLCC2 but find that unstructured grids make it difficult to accurately capture the flow features. Sakamoto et al. (2012a, 2012b) made a comprehensive flow field study covering vortex structures, ve-locities and free surface elevations for the 5415M in static and dynamic PMM conditions. The computed velocities in a number of cross planes along the hull were compared with re-sults from SPIV measurements. Overall level flow features were captured, but vortex core properties were predicted to be too weak. Kim et al. (2012) studied the flow around the DARPA Suboff submarine to investigate the vortex structures from the hull and the fins in

steady turn with drift. Comparison with ex-perimental data in the studies above showed that many of the flow features can be captured, so CFD seems to be a promising tool for learn-ing about the flow physics in manoeuvring.

When it comes to hydrodynamic forces and moments many different applications are cov-ered to gain knowledge about loads on hulls, rudders and propellers. Silva (2012) calculated both hydro and aerodynamic loads on a supply vessel with RANS. Comparison with experi-mental tank and wind tunnel showed both close agreement and deviations depending on the flow angle relative to the heading of the vessel. Xing et al. (2012) computed hull forces and moments for the KVLCC2 and found a rea-sonably good agreement with measurements. Amin & Hasegawa (2012) also computed hull forces for the KVLCC2, but the applied un-structured mesh introduces deviations with the measurements. Arii et al. (2012) calculate rud-der forces for an open water propeller-twin rudder configuration with reaction fins. The forces seem to be difficult to capture for larger rudder angles. In Miyazaki et al., (2011) the KVLCC2 was modified and used for a study of the influence of skeg configurations on the course stability. CFD is used to simulate the CMT test and the computed forces and mo-ments were used to determine the hydrody-namic derivatives and evaluate the course sta-bility index. Compared to experimental data, the results look promising. Shin et al. (2013) performed RANS based CFD computations for the KVLCC1 and KVLCC2 in pure turning and static drift conditions. The computed force and moment coefficients were compared with ex-perimental PMM data. Fukui (2012) performs CFD computations of the forces and moments on a VLCC hull with rudder in order to esti-mate the rudder-hull interaction coefficients used in the MMG model. The overall forces are in reasonable agreement with measured data. Accurate representation of the rudder in the

141141

simulation seems to be important to capture the interaction effects. Simonsen et al. (2012) in-vestigated forces and moments plus hydrody-namic derivatives for the appended KCS con-tainer ship with RANS and a body force pro-peller in a number of PMM conditions. Com-puted forces as well as moments plus hydrody-namic derivatives were compared with model test results. Further, simulations of the standard IMO manoeuvres were made based on both pure experimental PMM data and combinations of experimental dynamic PMM data and static computed PMM data. Results look promising on both force and manoeuvre levels; however, the simplified propeller model may introduce differences. Rajita Shenoi et al. (2013) made numerical simulations of the horizontal PMM conditions by means of RANS in order to de-termine the hydrodynamic derivatives and to perform 3 DOF manoeuvring simulations for the S175 container ship. A combination of measured and computed data is used as input for the simulator, similar to what was done by Simonsen et al. (2012). Computations covered the static drift and pure sway conditions. Other data came from empirical methods and meas-urements. The predicted turning circle com-pared reasonably well with measurements. Mauro et al. (2012) worked with the 5415M to investigate the asymmetric loading on a twin propeller configuration during turning. Saka-moto et al. (2012a, 2012b) performed static and dynamic PMM simulations for the 5415M based on RANS CFD. Thorough V&V was conducted and overall, the CFD solver seems to have the capability of handling static and dynamic PMM simulations, and the resultant forces and moment coefficients as well as hy-drodynamic derivatives show reasonable agreement with measured data. Cheng et al. (2013) also performed RANS CFD computa-tions for the 5415M in the pure yaw and pure sway conditions. When compared with meas-urements the quantitative accuracy of the above studies depends on properties like mesh size,

turbulence modelling, propeller modelling and how extreme the flow condition is in terms of flow separation. Generally it seems that forces and moments plus trends are captured reasona-bly well. Finally, it is possible to use the com-puted forces and moments as input to system based simulators.

Drouet et al. (2011) cover the DARPA Suboff submarine to compute forces and mo-ments in static drift condition. The results gen-erally look good compared to measurements. Though, for a drift angle larger than 12° the configuration of the bare hull including the sail element deviates, possibly due to turbulence modelling. DARPA Suboff is also studied by Kim et al. (2012) to compute the loads in steady turn with drift. Zhang et al. (2013) com-puted the flow around the Series 58, Suboff and DRDC STR submarines with RANS in order to simulate steady turn with and without drift. Pan et al. (2012) used unsteady RANS simulation for captive simulations with the Suboff geome-try, including steady oblique towing and dy-namic pure heave and pure pitch PMM motion. The CFD method is able to provide estimates of the manoeuvring coefficients for the fully appended submarine model, but more studies on application of more advanced turbulence models, finer grid resolution and additional verifications and validations are recommended to improve comparison with data. Zaghi et al. (2012) studied the manoeuvring behaviour of a fully appended submarine in the vertical plane by using CFD based captive data as input for a manoeuvring model. There is no comparison with experimental data for validation. Polis et al. (2013) used CFD to compute the manoeu-vring coefficients for the Suboff in steady con-ditions near the free surface to include the free surface effect in the coefficients. Different submergences and speeds were covered. Com-parison with captive model test data shows rea-sonable agreement. In order to be able to in-clude the coefficients in manoeuvring models,

142142

the computed results are approximated with exponentially fitted expressions.

The final application is direct simulations of the manoeuvres where the CFD tool is used to solve the flow field, compute the hydrody-namic forces and moments and find the trajec-tory of the ship during the manoeuvre. In Bro-glia et al. (2011) and Dubbioso et al. (2012) RANS simulations were performed for a free running twin screw tanker model performing turning circles and 20/20 zigzag. The propeller is modelled as a momentum disk approach where also side forces are accounted for. The results show that reasonable agreement can be obtained with measurements for speed, drift angle and yaw rate during the manoeuvre. In terms of overall manoeuvring characteristics a comparison between measured and computed transfer, advance, tactical and turning diame-ters looks promising. Sadat-Hosseini et al. (2013) performed RANSE based CFD simula-tions for the free running Delft Catamaran with water-jet propulsion during turning and zigzag manoeuvres. Simulations were conducted with two propulsion approaches: 1) bare hull with integral force models for wa-

ter-jet; 2) bare hull with actual water-jet with body

force impeller defined by pump curves.

The CFD results were compared with sys-tem-based predictions and both validated against experimental fluid dynamics (EFD) data. When compared to measured manoeuvres CFD with actual water jet model showed best agreement for turning. For the zigzag manoeu-vre CFD with actual water-jet showed the larg-est errors, while good agreement was shown for CFD bare hull with the system based inte-gral force water-jet model. The authors con-cluded that further works on water-jet charac-teristics and modelling are required.

In Carrica et al. (2013) URANS computa-tions of standard manoeuvres were performed for a surface combatant at model and full scale. Two types of manoeuvres were simulated: steady turn at 35 degrees rudder deflection and 20/20 zigzag both with constant RPM approach and body-force propeller. Results are bench-marked against experimental time series of yaw, yaw rate and roll, and trajectories, and also compared against available integral vari-ables. Comparison between CFD and experi-ments showed reasonable agreement for both manoeuvres, though issues regarding adequate modelling of propellers with side forces remain to be solved. The 20/20 zigzag manoeuvre was also simulated at full scale for one Froude number. The full scale case produces a thinner boundary layer profile compared to the model scale.

Araki et al. (2012a) performed free running CFD simulations for the ONR Tumblehome hull form in order to generate data for SI , which can be used to derive hydrodynamic co-efficients for system-based simulators. The ad-vantage of using free running CFD instead of model testing for this purpose is that both mo-tions and forces on the hull and appendages can be generated in CFD. The results of the ma-noeuvring simulations obtained with coeffi-cients from CFD SI look good when compared to measured standard turning circle and zigzag tests. This approach is an alternative to the one described above where a large set of CFD based PMM simulations are performed to de-termine the hydrodynamic coefficients for the mathematical manoeuvring model. Chase et al. (2012) have performed RANS, DES and DDES simulations for a free running submarine (DARPA Suboff) model performing a horizon-tal overshoot manoeuvre. The propeller was modelled with two different approaches: a body-force approach where the PUF-14

vortex-lattice potential flow code is coupled with the RANS solver;

143143

direct modelling of the propeller in the CFD model.

The horizontal overshoot manoeuvre was simu-lated with both propeller models. In terms of validation, the final free running manoeuvre was not compared with measurements, but both hull and propeller forces were compared with measurements for different conditions.

Using Potential Flow Techniques. Ommani et al. (2012) investigated the hydrodynamic forces on a semi-displacement vessel with a drift angle. The resulting flow asymmetry at the dry transom stern was investigated. The po-tential part was solved using the 3D Rankine source method and the viscous cross flow was calculated using a 2D+t theory. The agreement with experimental results is reasonable for the longitudinal and lateral force, but the neglected nonlinearities and 3D viscous flow are believed to hamper the prediction of the yawing mo-ment.

Ommani & Faltinsen (2013) investigated the dynamic stability performance of an ad-vancing mono-hull, semi-displacement vessel in sway-roll-yaw. A linear Rankine panel method was adopted and various Froude num-bers were analysed. Compared with the ex-periments, the numerical analysis was able to predict the instability of system.

Ichinose & Furukawa (2011) presented an estimation method for hydrodynamic forces acting on a ship hull in oblique motion using a 3D vortex method. A vortex block model and a vortex sheet model were introduced to model the flow in the boundary layer, but the quantita-tive accuracy of the estimated forces is not suf-ficient.

Using Empirical Calculations. In order to predict the hydrodynamic forces acting on a hull, which is necessary to conduct ship ma-noeuvring simulation, Yoshimura & Masumoto

(2012) made a database of manoeuvring hy-drodynamic coefficients for medium speed merchant ships and fishing vessels. The data-base not only contains hydrodynamic deriva-tives but also interaction coefficients. The coef-ficients are arranged by the principal particu-lars of ships and regression formulae are pre-sented. Sugisawa & Kobayashi (2012) pro-posed a correction method for hydrodynamic derivatives estimated by published empirical formulae. Correction factors for derivatives are defined to minimize the difference between simulated and measured turning trajectories. Viallon et al. (2012) investigated the reduction of the order and number of regressors of poly-nomial regression models for manoeuvring forces. A secondary regression which provides practically the same accuracy as the original higher order regression model for moderate manoeuvres is presented. Oh & Hasegawa (2013) evaluated four existing mathematical models for low speed ship manoeuvrability. Sway force and yaw moment predicted by the mathematical models were compared with ex-perimental results. They also conducted a simu-lation study on turning motion and a zigzag manoeuvre to check the influence of each model.

In terms of propeller and rudder force, Shen & Hughes (2012) proposed a computation method for the effective inflow velocity of the rudder. They estimated the axial and tangential flow velocities at the rudder plane separately and the effective inflow velocity was deter-mined based on the axial and transverse flow distributions and the rudder geometry encoun-tered by the propeller slipstream. Hwang (2012) presented a pragmatic 4-quadrant pro-peller-rudder model based on the concept of Thulin (1974) and Chislett (1996). Dubbioso & Viviani (2012) analyzed the effect of stern ap-pendage configurations comprising skegs, fins and rudders on the manoeuvrability of twin-screw ships. Based on extensive experiments

144144

for seven twin-screw models, an empirical cor-rection method for appendages effect is pro-posed.

Fang et al. (2012) developed a real-time simulator based on a 6 DOF mathematical model including seakeeping and manoeuvring characteristics. Hydrodynamic coefficients were estimated with empirical formulae in pub-lished papers. The simulated turning motions of 8,200 TEU container vessels were compared with measured sea trial results for the valida-tion of the simulator. Yuba & Tannuri (2013) investigated the manoeuvrability of pusher-barge systems which have an azimuth or a con-ventional propulsion system with/without an auxiliary bow azimuth thruster. The advantage of each system, depending on the manoeuvring situation, is shown.

Using Experimental Techniques. In order to evaluate the effect of roll motion on ship manoeuvrability, Yoshimura (2011) introduced a rudder to yaw response equation based on a linear mathematical model of hydrodynamic forces acting on a ship. He pointed out that the turning moment induced by roll motion is a key parameter which strongly affects the course-keeping and turning abilities. This tendency becomes remarkable when the roll angle be-comes large. Yasukawa & Hirata (2013) con-ducted oblique towing and circular motion tests with changing heel angle to capture the charac-teristics of hydrodynamic forces acting on a ship hull. The effect of the heel angle on the course stability criterion was evaluated using hydrodynamic derivatives obtained by model experiments. Yasukawa & Yoshimura (2013) investigated the roll-coupling effect on ship manoeuvrability in the framework of linear motion theory. They proposed approximate formulae for the course stability criterion, steady turning index and time constant for steady turning. Simulation results of turning

motion using hydrodynamic derivatives, in-cluding the effect of roll motion, are shown.

A mathematical model for a twin-propeller,

twin-rudder ship was developed by Khanfir et al. (2011) based on captive model tests and free-running experiments. An experiment-based method for estimating rudder-hull inter-action coefficients is proposed. Simulated re-sults based on the proposed mathematical model are compared with free-running test re-sults for validation.

The effect of static and dynamic azimuthing conditions on the propulsive characteristics of a puller podded unit were analyzed by Akinturk et al. (2012) based on model experiments in open water. They conducted a thorough uncer-tainty analysis to assess the uncertainty in their experiments and to identify the major factors influencing measured results. Amini & Steen (2012) also investigated the effect of a dynami-cally changing propeller revolution and azi-muth angle on propeller shaft loads based on model experiments using a model of a pushing azimuth thruster. Song et al. (2013) investi-gated the thrust loss induced by the interaction between an azimuth thruster and a ship hull based on model tests using a model of a wind turbine installation vessel. Comparison be-tween simulation results using a commercial CFD code and measured results is also shown.

Several publications relate to unconven-tional ships. Obreja et al. (2010) developed a simulation code for the manoeuvring character-istics of a Mediterranean fishing vessel. PMM experiments were used for evaluating the hy-drodynamic derivatives. The simulation results for turning motion and zigzag manoeuvres were compared with the model test results. Zhan & Molyneux (2012) developed a simula-tion method for ship motion in packed ice, combining mathematical models for ship mo-tion, ice motion and ship-ice interaction. The

145145

manoeuvring behaviour of an arctic drill ship with ice was simulated by the mathematical model and compared with experimental results.

Avila & Adamowski (2011) carried out forced oscillation and steady-state tests with an open-frame ROV. Analysing the variation of drag and inertia coefficients in Morison’s equa-tion as a function of Keulegan-Carpenter and Reynolds numbers, dependency or independ-ency on the parameters is shown. De Barros & Dantas (2012) presented a comparative study of CFD and ASE (analytic and semi-empirical) methods for the prediction of the normal force and moment coefficients of an AUV with a duct propeller. The advantages of the symbiosis between CFD and ASE methods are suggested.

Towed stability. Fitriadhy & Yasukawa (2011a, 2011b) developed a nonlinear numeri-cal simulation tool to predict course stability and turning ability of a towing system in calm water. The motions of the towing and towed vessels were coupled by a towline. The towline was modelled using a 2D lumped mass method to take into account the dynamic motion of the towline. Linearized equations of motion were also derived to confirm the validity of the nonlinear analysis. The influences of several parameters such as towline length, towed ves-sel’s dimension and tow points on course sta-bility and turning ability of the towing system were investigated.

Fitriadhy et al. (2011) investigated the mechanism of slack towline motion and its in-fluence on towing and towed vessels during manoeuvring. A linearized theory was applied to grasp the basic mechanism of dynamic inter-action between towing and towed vessels. They proposed a formula which gives the appearance limit of slack towline during turning. Further-more Yasukawa et al. (2012b) carried out nonlinear time domain simulations and tank tests to validate the formulae. It is concluded that the slack towline appearance limit qualita-

tively agreed with the results of simulations and tank tests.

Ren et al. (2012) proposed a mathematical model of a tug towage operation for an interac-tive tug simulator. Two kinds of towline ten-sion models were used. The first one is a model with linear strain which can take account of the towline’s own weight. The other one is a model with nonlinear strain which omits the towline’s own weight. The appropriate model was se-lected in their simulation comparing the tow-line strain with the maximal towline strain given by the towline stress-strain diagram. Yoon & Kim (2012) modelled a towline with a finite element model in 5 DOF (roll excluded). The motion of the tow vessel was simulated in 6 DOF but the towed vessel was assumed to solely move in the horizontal plane. In the above papers, only the results of numerical simulations are presented. 4.2 In Shallow Water

Using Viscous CFD Methods. Toxopeus (2011b) performed a comprehensive study of the shallow water effect on the KVLCC2. Computations were performed with fixed sink-age and trim and free-surface effects were not taken into account. Results highlight the ad-verse influence of the water depth on the flow along the aft part of the ship. Kimura et al. (2011) applied CFD to study the manoeuvring forces on a VLCC in shallow water .

Using Potential Flow Techniques. Skejic et al. (2012) investigated the ship manoeuvring performance in calm water with variable finite water depth. A unified seakeeping and ma-noeuvring (MMG based) model was modified with the inclusion of shallow water effects. Simulated results of turning motion for variable sea bottom profile are shown.

146146

Gourlay (2013) applied a modified slender body method to solve the ship‘s squat in a dredged channel and canal. The sinkage in a dredged channel is 20-30% larger than in open water, while in a canal, the squat can increase up to 100% compared to the value in open wa-ter.

A shallow water hydrodynamic coefficient prediction and MMG equation simulation of ship fleets manoeuvring in shallow water (Three Gorges Dam of China) was carried out by Cai, et al. (2012). The added mass was cal-culated by strip theory and empirically cor-rected with the shallow water effect. The simu-lation results showed good agreement with the experiments.

Using Empirical Calculations. A prediction method for linear derivatives in shallow water was proposed by Furukawa et al. (2011). The linear derivatives were obtained by adding cor-rection factors to the deep water derivatives. The correction factors are provided as func-tions of parameters, which consist of principal ship dimensions and so on.

Quadvlieg (2013) presented a method to create mathematical manoeuvring models for the simulation of inland ships based on only the main particulars of hull, rudder and propeller without the need to execute model tests. A modular model is introduced based on slender body theory and cross flow drag theory for hull forces and a parameterised model for rudders of inland vessels based on systematic model tests.

Using Experimental Techniques. The in-herent directional stability of a catamaran was investigated by Milanov et al. (2011) based on a linearized manoeuvring model and model ex-periments covering a wide range of Froude numbers and depth to draft ratios. Maimun et al. (2011) presented an experimental investiga-

tion of the manoeuvring characteristics of a pusher barge system for deep and shallow wa-ter conditions. Comparisons between simulated results using experimental or empirical coeffi-cients with measured results are shown. Rei-chel (2012) developed a mathematical model based on the MMG approach for a twin-propeller, twin-rudder car-passenger ferry. PMM tests were conducted to determine the hydrodynamic derivatives and other parame-ters. Three modes of motion such as ahead, astern and pure drift were considered in the model tests to simulate port operations. 4.3 In Restricted Water

Using Viscous CFD Methods. Zou et al. (2011) compared results obtained with both po-tential and CFD codes with experiments for the KVLCC2 in a canal. Results show the influ-ence of viscous effects on ship behaviour and flow field. The CFD results are in good agree-ment with the experiments for different UKC and lateral clearances. In Zou & Larsson (2012a) the research is extended to provide physical explanations of the flow field. Compu-tations were performed for both 0 RPM and self-propulsion. Results show a strong influ-ence of the bank on stern flow leading to high asymmetrical propeller loadings and yaw mo-ments.

Lou & Zou (2012) performed CFD compu-tations on a KVLCC hull in a canal. Computa-tions were performed in pure sway for symmet-rical and asymmetrical locations of the ship in the canal. Results showed strongly different behaviour of the sway forces and yaw moment for the two cases.

Using Potential Flow Techniques. With the first-order Rankine source panel method, Yao et al. (2011) studied the bank effects of a con-tainer ship sailing along vertical or sloping

147147

banks in shallow channels. The influences of the ship to bank distance , the speed and the water depth on the sway and yaw hydrody-namic forces were discussed.

Using Empirical Calculations. A statistical squat prediction model was proposed by Beaulieu et al. (2012) based on a stepwise re-gression tree algorithm. The prediction model was developed using a database containing 5,141 observations in the St. Lawrence River and produces a relationship between squat and ship speed. Om et al. (2013) evaluated the ma-noeuvrability of a shallow draft ore carrier with twin-propeller and twin-rudder, which is newly designed for inland waterways.

Muto et al. (2011) simulated the motion of a ship running in a non-uniform flow field, mimicking the flow field that a ship may en-counter while sailing near the mouth of a river. Estimated hydrodynamic forces using empiri-cal formula for large drift angles and simulated trajectories were compared with measured re-sults. Hasegawa et al. (2013) investigated the ship manoeuvring behaviour in crossing cur-rent. They pointed out that a mathematical model for low speed should be considered even if the ship speed is not low because the cross-ing current causes a large drift angle.

Carreño et al. (2013) conducted full-scale trials of a riverine support patrol vessel which has a pump-jet propulsion system and a large beam-draft ratio. The standard parameters of turning tests were measured to compare with simulated results based on a mathematical empirical model.

Using Experimental Techniques. Yasu-kawa et al. (2012c) analysed the course stabil-ity and yaw motion of a ship running under steady wind conditions and proposed a course stability criterion including the effect of aero-dynamic force derivatives. Then, Yasukawa et

al. (2013) applied the course stability criterion for a ship running in a channel under steady wind and obtained the check helm angle re-quired for course keeping by solving the steady motion equations. 4.4 Ship-to-Ship Interaction

Using Viscous CFD Methods. Mousavi-raad et al. (2011) use CFDSHIP-IOWA to study interactions between passing ships. Re-plenishment and overtaking computations were performed in both calm water and waves. In-fluence of the spacing between ships and the sheltering effect of one ship was evaluated. Re-sults are compared with experiments.

Fonfach et al. (2011) present a comparative study of potential and CFD computation on the flow past a tug boat close to a large tanker. Computations were performed using free-surface boundary conditions or double body conditions. Results highlighted the influence of the free-surface boundary condition to accu-rately predict the lateral force on the tug boat as the separation distance is reduced.

Simonsen et al. (2011) performed CFD computations on a tug boat next to a tanker for different tugboat drift angles and locations rela-tive to the tanker. The CFD results are in good agreement with the experiments.

Benedict et al. (2011) developed a new and extended mathematical model to solve encoun-tering and overtaking ship operations consider-ing the surge and sway motion. A combined approach with finite volume discretisation and level-set free surface flow was adopted to simulate the hydrodynamic forces. The paper also introduced the safe passing distance based on a reference drift angle.

148148

Sadat-Hosseini et al. (2011b) presented a study on investigating the interaction between two different tankers; Aframax and KVLCC2, free to heave and pitch, advancing in shallow-water with the same speed and with a fixed separation distance using CFDSHIP-IOWA V4.5 URANS simulation. The result was vali-dated and shows good agreement. Several in-fluences such as suction force and asymmetric ship wake on ship-ship interaction and longitu-dinal alignment on yaw moment were dis-cussed. The same problem was investigated by Zou & Larsson (2012b). The paper applied the steady RANS to numerically simulate the hy-drodynamic force between the Aframax and the KVLCC2. Both the RANS and URANS gave good results compared to the experiments.

Zhang & Zou (2011a) used the FLUENT software to calculate the hydrodynamic forces of encountering and passing ship-to-ship inter-action. The influences of boundary conditions such as bank effect and water depth were pre-sented.

Leong et al. (2013) focused on the interac-tion forces and moments acting on an AUV op-erating in close proximity to a moving subma-rine. The influences of longitudinal and lateral distances and a range of speeds were investi-gated through CFD and EFD and a safe path for the AUV to approach or depart from the submarine was suggested.

Zubova & Nikushchenko (2013) investi-

gated ship to ship interaction using the Wigley hull form. Calculated forces and moments us-ing commercial software (FLUENT, FINE/Marine and STAR-CCM+) were pre-sented.

Yang et al. (2011) compared potential flow and CFD results for passing ships at low speed, of which one was the KVLCC2. Potential flow and CFD results are in good agreement.

Using Potential Flow Techniques. Potential

theories are efficient in solving ship-to-ship in-teraction problems. With the manoeuvring model introduced by Skejic (2008) and a 3D boundary element method, Xiang & Faltinsen (2011) simulated the interacting hydrodynamic forces of two ships and carried out verification and validation in infinite water. In this re-search, a low Froude number and a rigid free surface was assumed. Xiang et al. (2011) also predicted the interacting loads of two tankers involved in a typical lightering operation with the 3D panel method. As for the ship to float-ing structure interaction, Skejic et al. (2011) used the STF strip theory and a two time scale manoeuvring model to simulate the process of manoeuvring a ship around a floating object with the assumption of low speed and uniform current.

Sutulo et al. (2012) applied the classic Hess and Smith method, combined with rigid free surface conditions, into the real time interacting forces of two ships. Compared with experi-ment results, the largest discrepancies were discovered for the sway force at a very small horizontal clearance. This effect could be ana-lyzed with viscous flow theory and free-surface boundary condition, see Fonfach et al. (2011).

3D potential flow theories have been ap-plied to the interactions between a moving ship and moored ship. Van der Molen et al. (2011) calculated the hydrodynamic forces of a moored ship in port due to passing ships by means of a 3D source method taking account of the free surface and the finite water depth . Pinkster (2011) gave 3D potential flow results of hydrodynamic forces on a moored vessel due to a passing vessel based on a double-body flow and free surface assumption. He also pointed out that the complexity of geometry, current or drifting angle would lead to inaccu-racies. Based on Pinkster’s double-body

149149

method, Bunnik & Toxopeus (2011) presented a RANS method to compute the effect of pass-ing ships on moored ships. The discrepancy be-tween RANS and the potential methods for large drift angles was analysed. The 3D poten-tial flow method was also applied by Verdugo et al. (2013) who studied the methodology to analyse ship manoeuvres and passing ship ef-fects on moored ships at different berths in the Port of Altamira (Mexico).

De Jong et al. (2013a) applied a newly-developed time-domain model based on the shallow-water flow formulations for continuity and momentum (Xbeach) to simulate the pass-ing ship effect in waterways and ports. The non-linear effects such as shallow water waves, currents and an arbitrary bank condition could be taken into account.

Pinkster & Bhawsinka (2013) introduced a real-time simulation technique which links the program “Delpass” and MARIN’s real-time simulator. This might reflect more precise ship manoeuvring behaviour on the simulator since it uses the real-time force for ship-bank and ship-ship effects instead of the empirical hy-drodynamics.

Ship-to-ship interaction research was car-ried out by Watai et al. (2013). The results based on strip theory, empirical regression and 3D Rankine source boundary element method were compared with the experiments. The 3D-BEM method gave the best agreement with the test on the passing ship effects.

Using Empirical Calculations. An artificial neural network method for predicting the sway force, surge force and yaw moment was studied by Xu et al. (2012). With this ANN technique, the influence of ship speed, water depth and ship dimensions could be immediately trans-lated into ship-to-ship forces to help the pilot quickly judge the navigation environment and

risks. Gronarz (2011) made a so-called hybrid regression to predict the transient behaviour re-lated to forces and moments caused by passing ships. 4.5 Improvement in CFD methods

For the application of CFD for manoeu-vring, simulation of the captive conditions is the most commonly used approach today. It seems that reasonable results can be obtained, Simonsen et al. (2012), but the downside of the approach is that many CFD simulations must be performed to give enough data to provide the required derivatives for simulator models. On the other hand, part of the test matrix can be computed and combined with input from other sources. The CFD based SI approach from Araki et al. (2012a) is currently not used much, but if the CFD code is capable of simulating the free sailing manoeuvres it can be done. The simulations required are complex, but fewer runs are required compared to the captive ap-proach. It should be mentioned that if the free sailing capability is available in the CFD code and one is only looking for the standard IMO manoeuvres they could be directly simulated without going through the system-based model. If more general manoeuvres are to be per-formed the CFD based SI method could be a better option.

Recent works using unsteady Navier-Stokes equations to simulate free-running manoeuvres have been published. Simulations are usually performed using propeller models in order to reduce computational effort. One of the weak points that are currently experienced by many of the CFD applications is the propeller model-ling. It would be good to run the CFD simula-tions with spinning propeller geometry, but this is very time consuming due to the different time scale between propeller physics and ma-noeuvring forces variations. Therefore, many

150150

users apply simplified propeller models which are missing some of the rudder-propeller-hull interaction effects and in some cases also the side force from the propeller. This influences of course the loads on the individual compo-nents and will influence the predicted manoeu-vre. 4.6 Autopilots and other control applica-

tions

This section presents the developments re-lated to the application of control systems to the manoeuvring problem. Besides autopilots’ new technologies, there are some improve-ments related to automatic berthing, optimal route finding, etc.

Bhattacharyya et al. (2012) developed a fuzzy autopilot algorithm for manoeuvring of surface ships and verified the performance us-ing time-domain simulations of a Mariner class vessel. However, this can be considered an in-troductory work, since the analysis assumed an undisturbed environment without any waves, current or wind. Mucha & Moctar (2013) tested different control approaches to design and tune the autopilot applied to a vessel navigation close to a bank.

Luo et al. (2013) proposed a hybrid archi-tecture for the autopilot, with real time identifi-cation of ship dynamics based on support vec-tor machines and robust techniques applied for the controller design. Numerical simulations were used for the performance analysis.

Do (2010) derived a general control algo-rithm for underactuated ships, with no inde-pendent actuator in the sway axis. The trajec-tory control using the rudder is an example of such a problem. The algorithm is based on nonlinear control theory and numerical simula-tions illustrated its effectiveness.

The concept of multi-controller structure

was applied to autopilot design by Saari & Djemai (2012). The ship speed is used to select between different PID control gains, and a simple switching law is adopted. The authors showed that the non-linear behaviour of the system due to the speed can be adequately compensated by the proper switching of PID control gains.

Mizuno & Matsumoto (2013) derived an automatic ship’s manoeuvring system using a sliding mode controller. They demonstrated the advantages of the proposed controller by means of computer simulations and actual sea tests carried out using the small training ship Shioji-Maru under various conditions. The authors emphasized that the control scheme can be eas-ily implemented in the autopilot for small size ships.

The automatic berthing is a marine control related problem, in which the model describing the vessel motion is highly non-linear, espe-cially in the case of low speed and large ma-noeuvring motion. Also, the number of inputs used to control the vessel position and heading may be large, due to the utilization of thrusters and tugboats. Due to the previous characteris-tics of the problem, the definition of the mini-mum-time approaching control for automatic berthing requires a large computer processing capacity. Mizuno et al. (2012) developed an automatic berthing system using GPU, which is able to cope with external disturbances. The method uses the prediction of the future posi-tion of the vessel in order to define the next set of inputs. Numerical simulations and full-scale tests were used to verify the system. Tran & Im (2012) presented an automatic berthing system with an artificial neural network (ANN) con-troller. The controller is designed to use assis-tant devices such as bow thruster and tugboat.

151151

The online prediction of ship roll motion during manoeuvring plays an important role in navigation safety and ship control applications. Yin et al. (2013) derived a method for this task, using neural networks. The results of full-scale sea trials were used to validate the method.

A method for automatic route finding and collision avoidance was presented by Xue et al. (2011). This paper presents an effective and practical method for finding safe passage for ships in possible collision situations, based on the potential field method. Simulations of com-plex navigation situations demonstrated the ef-fectiveness of the method.

Nakano & Hasegawa (2012) proposed a prediction method for manoeuvring indices K and T in Nomoto’s model by analysing AIS (Automatic Identification System) data with an optimisation method. 5. BENCHMARK DATA 5.1 SIMMAN 2014

Goal. In continuation of the Workshop on Verification and Validation of Ship Manoeu-vring Simulation Methods, SIMMAN 2008, a new workshop SIMMAN 2014 will be held in December 2014. Since SIMMAN 2008 some of the deep water data sets used for the workshop has been replaced by new measurements based on the learning from 2008. Further, the scope of SIMMAN 2014 has been extended com-pared to 2008, so shallow water is also a part of the workshop. This has necessitated measure-ments in shallow water.

At the SIMMAN 2008 workshop the focus

was placed on four hull forms selected by the ITTC for benchmark, i.e. the KVLCC1 and KVLCC2 tankers, the KCS container ship and

the 5415M. However, the results from the 2008 workshop showed that there were only minor differences in manoeuvring characteristics be-tween the KVLCC1 and KVLCC2. Therefore, to limit the number of test cases and focus the effort on fewer ships it was decided to only fo-cus on KVLCC2, KCS and 5415M in SIM-MAN 2014. A discussion of the 2008 data is given in Stern et al. (2011).

The main focus of the workshop is on ap-

pended hull tests in deep and shallow water to provide data for simulation of free manoeuvres. Though, bare hull tests for validation of CFD-based methods are also available. Ship, rudder and propeller geometries plus the captive part of the data from the model tests is already available to the public via request from the workshop website www.simman2014.dk. Free running test results will be made available after the workshop, since the free running test cases are blind. An overview of the model test data available for the workshop is given in Table 1. Some test data has not yet been received.

Captive Model Test Data. All test condi-

tions for the workshop are specified in model scale, i.e. appended captive tests are made at model self-propulsion point using constant RPM throughout the manoeuvre. Typical out-put are X- and Y-forces plus yaw and heel moments (4 DOF). In some cases rudder and propeller loads are also measured.

For KVLCC2, new PMM data is available

for both deep and shallow water in both ap-pended and bare hull configurations. Hyundai Maritime Research Institute (HMRI) has pro-vided data for a 3 DOF test in deep water with a model at a scale of 1:46.426. INSEAN is planning on making the same test in deep wa-ter, but with a smaller model at a scale of 1:100. This data will be available in the second half of 2014. The Bulgarian Ship Hydrodynam-ics Centre (BSHC) has contributed with 3DOF

152152

PMM data in shallow water for an appended model at a scale of 1:45.714. Water depths ranging from very shallow to moderate shallow were covered with h/Tm ratios of 1.20, 1.50 and 2.00. In addition to this a number of bare hull conditions were also covered for h/Tm=1.20. Flanders Hydraulics Research (FHR) also exe-cuted shallow water 3 DOF PMM tests for the KVLCC2. In this case the scale was 1:75 and h/Tm ratios of 1.20, 1.50 and 1.80 were cov-ered. A subset of bare hull conditions are also available from FHR. Concerning circular mo-tion tests (CMT) with the KVLCC2, the 3 DOF data set for the appended hull used for SIM-MAN 2008 is still used and available. The scale of the model was 1:110.

For KCS in deep water PMM tests were

performed at FORCE with the appended hull. Since heel plays an important role for the con-tainer ship the test was performed as a 4 DOF

test, which means that heel variation is in-cluded in the test. A limited set of conditions with the bare hull is also covered. The scale of the model is 1:52.667. In shallow water two data sets have been made. One set is made by FHR who considered water depths with h/Tm ratios of 1.20, 1.50 and 2.00. The model used in this case is the same as the one FORCE used, i.e. model scale of 1:52.667. The other data set is made by MOERI. The test was made with a model at a scale of 1:31.6. Data from this test has not yet been released by MOERI.

With respect to CMT tests two data sets are available. The first is from NMRI who made 3 DOF CMT for SIMMAN2008 with a model at a scale of 1:75.5. The other set is made by China Ship Scientific Research Centre (CSSRC). To account for heel, the test was made as a 4 DOF test with the appended hull. The scale of the applied model is 1:52.667.

Table 1. Available data for the SIMMAN 2014 workshop.

Hull KVLCC2 KCS 5415M Cap-tive

PMM app. deep

INSEAN (2014) missing

HMRI (2012)

FORCE (2009)

MARIN (2007)

PMM app. shallow

BSHC (2013)

FHR (2012)

FHR (2012)

MOERI (2013)

PMM bare deep

INSEAN (2014) missing

FORCE (2009)

FORCE (2004)

IIHR (2005)

IN-SEAN (2005)

PMM bare shallow

BSHC (2013)

FHR (2012)

CMT app. deep

NMRI (2006)

NMRI (2005) 3DOF

CSSRC (2013) 4DOF

IHI (2012)

MARIN (2007)

CMT bare deep

Free Free app. deep

HSVA (2006)

MARIN (2007)

CTO (2007)

MARIN (2009)

MARIN (2007)

Free app. shallow

FHR (2012)

MARIN (2013)

BSHC (2008/ 2011)

FHR (2012)

153153

A third set of CMT tests in 4DOF is made

available by JMU and Hokkaido University based on measurements in 2012 at a scale of 1:105.

For the 5415M the PMM test results are available for both bare and appended hulls. For the bare hull three data sets were made by FORCE, INSEAN and IIHR. The three insti-tutes used different model scales: 1:35.480, 1:24.830 and 1:46.588, respectively. These data sets were also available for SIMMAN2008. For the appended 5415M MARIN has provided a set of PMM data for the model with a twin screw-twin rudder arrangement, a centre line skeg, bilge keels and stabiliser fins. The PMM test was conducted as a 4 DOF test and in addi-tion to the traditional overall forces and mo-ments acting on the ship, local force measure-ments on rudders and stabilizers were also per-formed. Concerning CMT test results, MARIN performed this test with the same 5415M model that was used for the PMM test. Both appended 5415M data sets are new compared to the sets used for the workshop in 2008.

Free Model Test Data. The nominal condi-

tions for the free model tests comprised con-stant RPM at the model self-propulsion point as well as a certain speed, rudder rate and GMT for each ship. The typical measurements cover turning circles (full or partial) plus 10/10 and 20/20 zigzag tests.

For the KVLCC2 tanker in deep water free model tests were performed with the same model (1:45.714) at the nominal conditions at three facilities: HSVA, MARIN and CTO for SIMMAN 2008. It can be noted that KVLCC1 was also tested and it was from these results that it was found that the difference in ma-noeuvring characteristics between the two ver-sions of the tanker was quite small. It was de-cided to skip KVLCC1 for SIMMAN2014. In

shallow water two new data sets have been measured. One was made by FHR with the same model (1:75) that they used for the PMM tests. Another data set was measured by MARIN with the FHR model (1:75). The con-sidered water depths covered h/Tm ratios of 1.20, 1.50 and 1.80.

For the KCS container ship in deep water a new set of free model tests has been performed by MARIN with a model at a scale of 1:37.890. It can be noted that this is a somewhat larger model compared to the one used for deep water PMM at FORCE. In shallow water three new data sets have been measured. Two were made by BSHC and one was made by FHR, but they were all made with a model at a scale of 1:52.667. Both BSHC and FHR considered h/Tm ratios of 1.20, 1.50 and 2.00. It can be noted that at BSHC the full turning circles were measured, while the FHR data only contains partial turning circles due to limited width of the towing tank.

During SIMMAN2008, the free model tests from MARIN for the 5415M showed a surpris-ing asymmetry between the port and starboard turning circle manoeuvres, but this has subse-quently been checked and corrected, so data should be ready for SIMMAN2014.

As a final comment to the shallow water

captive and free running test results in shallow water, it should be noted that towing tank blockage may influence the results as indicated in Toxopeus et al. (2013a). In deep water the width of the applied towing tanks does not in-fluence the results significantly. But, when test-ing for instance at h/Tm of 1.20 with very small under keel clearance, the width of the tank has an influence. So, when using the shallow water data for validation of simulation tools this has to be kept in mind.

154154

5.2 Submarine

The DARPA SUBOFF is a recommended submarine hull form for benchmark tests. This is described by Groves et al. (1989). Very de-tailed flow measurements were published by Huang et al. (1992) based on measurements in a wind tunnel. Towing tank experimental re-sults were presented by Roddy (1990). The lat-ter one concerns rotating arm experiments car-ried out in the Carderock Model basin of NSWC.

The DARPA SUBOFF comes in various configurations having different arrangements of aft planes and stern arrangements: it is rec-ommended to work with one of the following configurations. Configuration AFF-1 is an axi-symmetric body without sail, propeller and planes. Measurements are carried out until drift angles of 18°. This is often taken as the base case for many research programs. Configura-tion AFF8 is the fully appended hull with a sail and aft planes. In addition to the results of cap-tive manoeuvring experiments, this set of data also includes the flow field at several locations and the pressures at several locations on the hull measured during the captive manoeuvring experiments.

Many researchers over the world are using this hull form as the study object: A collabora-tive exercise to calculate the manoeuvring forces for DARPA SUBOFF by CFD is re-ported by Toxopeus et al. (2012). Zhang et al. (2013) simulated the flow over the AFF-1 form. Kogishi et al. (2013) have also performed calculations on this hull form, but unfortunately performed experiments on a different subma-rine hull. Ray (2010) used RANS to determine the hydrodynamic coefficients of the DARPA SUBOFF. Vaz et al. (2010) compared the re-sults of two different viscous flow solvers for DARPA SUBOFF.

5.3 Hamburg Test Case

The Hamburg Test Case (HTC) is a 1:24 scale model of a 153.7m container ship built by Bremer Vulkan in 1986. Captive deep water model testing was conducted with the HTC within the VIRtual Tank Utility in Europe (VIRTUE) project by Hamburgische Schibau-Versuchanstalt (HSVA) in order to provide data for CFD validation. The tests covered force measurements for the bare hull, the hull with rudder and the hull with propeller and rudder. In addition, PIV measurements were conducted with the bare hull model while sail-ing in steady turning motion. The experiments were reported in VIRTUE deliverable D3.1.3, Vogt et al. (2007). Further, free running model tests were performed by MARIN to determine the manoeuvring characteristics in connection with measured turning circles and pull out plus 10/10 and 20/20 zigzag manoeuvres. The re-sults are reported in Toxopeus (2011a). 5.4 Restricted Water Cases

Bank Effects. To investigate bank effects and make a public data set to be used for vali-dation of mathematical models and CFD com-putations a comprehensive research project covering captive model testing has been carried out at Flanders Hydraulics Research in Bel-gium in cooperation with the Maritime Tech-nology division of Ghent University, Lataire et al. (2009b). In this study two types of banks were investigated: one covers surface piercing banks, characterised by a constant slope from the bottom up through the free surface and the other covers banks with platform submergence composed of a sloped part from the bottom up to a certain level where it transitions into a horizontal, submerged platform. Further, three different under keel clearances were consid-ered. Three ship models were used during this test: a 8000 TEU container carrier, a LNG-

155155

carrier and a small tanker. Only a limited set of model test results from the study are made available for the container carrier model at dif-ferent loading conditions. The data covers measured hull forces and moments, rudder forces, propeller thrust and torque, dynamic sinkage and trim plus free surface elevations.

Ship to Ship Interaction. In relation to ship-to-ship interaction captive model test results for lightering conditions are presented in Lataire et. al. (2009a). The service ship (SS) is an AFRAMAX and the ship to be lightered (STBL) is the KVLCC2 tanker. Both ships are at a scale of 1:75 and are equipped with rudder and running propeller. Speeds of 2, 4 and 6 knots were covered in shallow water corre-sponding to h/T=1.87 for the STBL. During the static tests the transverse and longitudinal posi-tion of the SS relative to the STBL were varied. Further, different drift angles of the SS were also covered. During dynamic tests both har-monic pure yaw and pure sway conditions were covered. The results of the tests cover propeller thrust and torque, rudder torque and forces plus hull forces and moments for both ships. Fur-ther, the wave elevation at three positions in the basin were recorded to track the wave making of the passing ships.

Lock Effects. The last benchmark data set for restricted water covers model test data for ships approaching and leaving locks. In Van-torre & Delefortrie (2013) model tests with a free running self-propelled 12000 TEU con-tainership at a scale of 1:80 were conducted in model of the new locks in the Panama Canal. Both lock entry and lock exit conditions were covered for under keel clearances of 20% and 10%. In terms of published results the ship’s position, the set speed and the actual speed, the longitudinal forces (propeller thrust and tug force), the propeller rate the lateral force and yawing moment, the absolute running sinkage of the ship’s bow and stern, the height of bow

wave and the water level elevation at the closed lock door were measured. In addition to this data, results for a limited number of captive tests with a bulk carrier at a scale of 1:75 sail-ing in the approach channel to another lock are also provided in Vantorre & Delefortrie (2013). 5.5 Manoeuvring in Waves

Yasukawa (2006) provides benchmark data for manoeuvring in waves. Free running turn-ing circles with a container ship (S-175) at a scale of 1:50 were carried out in regular waves. The ship model always started at Fr = 0.15. The regular waves were tested in both beam and head seas of varying wave length ( /L = 0.5-1.2, H/L = 0.02). Course keeping tests in regu-lar waves were performed for wave directions 0, 30, 90, 150 and 180 and varying wave length ( /L = 0.5-1.5, H/L = 0.02). 6. MANOEUVRING AND COURSE

KEEPING IN WAVES 6.1 Overview

Manoeuvrability in waves is a common name but it gathers many different applications like course keeping in following waves, broaching and “pure manoeuvrability”.

Course keeping in head waves is dealing mainly with forces at wave frequency and small heading deviation. Consequently it is more a seakeeping concern than a manoeuvring issue. In following seas, the encounter fre-quency is significantly lower and ship motions are studied like low-frequency motion. More-over, waves may be jeopardizing ship stability in the horizontal and vertical planes which may result in large heading deviations. Tools to ana-

156156

lyse course keeping in following seas are there-fore derived from manoeuvring tools.

Broaching concerns the loss of stability in

the horizontal plane in following seas. Once broaching occurs, kinematic energy along the velocity axis transfers in the roll motion (Wu et al., 2010) which leads to strong heel angles and loss of heading. Usually models are developed to study broaching inception (early stage corre-sponding to the loss of stability in the horizon-tal plane) since once the ship is broaching, it can hardly be controlled. Small ships are mainly concerned by broaching since sailing speed and ship length have to be close to wave speed and wave length.

Pure manoeuvrability (i.e. turning ability) in waves is concerned with the influence of waves on the manoeuvring criteria of a ship. Turning capability in waves is linked to the IMO “manoeuvring in adverse conditions“.

Ship manoeuvring and course keeping in waves are studied using experimental methods, numerical simulation based on specific numeri-cal models and CFD. 6.2 IMO criteria

In the past few years, some new IMO regu-lations of Energy Efficiency Design Index (EEDI) were carried out in which the ship’s manoeuvrability and course keeping ability in adverse wind and waves are added. It means that the techniques of prediction of manoeu-vring in waves need to be developed urgently.

In May 2011, MEPC 62/5/19 was issued in which a minimum propulsion power line crite-rion is stated and the adverse weather condition is defined. In June 2012, MEPC 64/4/13 and MEPC 64/INF.7 were issued. The approach consists of three levels of assessment: mini-

mum power line method, simplified method and comprehensive method. In May 2013 it re-sulted in an interim guideline, MEPC.65/22, in which the comprehensive approach was dropped. MEPC 66 added in April 2014 EEDI calculations for ships that were not considered in 2012 (LNG carriers, RORO carriers and cruise ships with alternative propulsion). No changes were made to the 2013 interim guide-lines for determining the minimum propulsion power.

The first level of assessment in these guide-lines is an empirical and statistical method to set a minimum power value for the installed power, which correspondent to different ship types (bulk carriers and others) and dead-weights, see for example Figure 2.

The second level of assessment is to evalu-

ate the manoeuvrability empirically based on not only the ship’s size but also the other fac-tors such as windage area and rudder area.

Figure 2. Statistics of minimum propulsion power line of a bulk carrier above 20k DWT.

157157

Table 2. MEPC weather conditions and indices.

Environment and indices

MEPC 62/5/19

MEPC 64/4/13

MEPC 65/22

Sig. wave height (m)

<9.8 <8 <5.5

Mean wind speed (m/s)

<21.4 <25 <19

Course devia-tion (°)

5-10 10 10

Min advance speed (kn)

2-4 4 4

The third level of assessment, which was

dropped in 2013, was to make a comprehensive assessment under specified adverse weather conditions. A ship needed to show the capabil-ity to maintain a minimum speed with a maxi-mum course deviation of 10° in any wind and wave direction. It is clear that more research and tool development is needed before being able to set the limits for the third level ap-proach.

At the same time the weather conditions and criteria for the key indices were lowered from MEPC 62 to 65, see Table 2. 6.3 Overview of methods

There are 4 classes of methods used to con-sider manoeuvring in waves: experimental methods, unified methods, two-time-scale methods and direct calculations by CFD.

Experimental methods. Using a combina-tion of physical model tests and numerical tools, Otzen & Simonsen (2012) developed a mathematical model of a high speed catamaran ferry manoeuvring in waves. The model is able to simulate broaching, as demonstrated by vali-dation against model test results. The aim is to

include the mathematical model in a full mis-sion bridge simulator.

Unified methods. Matusiak & Stigler (2012) presented experiments and simulations of a steady turning manoeuvre in irregular waves. Results show a very unsteady behaviour of the roll angle. The simulations are based on an unsteady manoeuvring model based of infi-nite added mass and Cumming integrals for ra-diation forces.

Two-time scale methods. Skejic & Faltin-sen (2013) applied their two-time scale model to irregular sea states. The effect of varying significant wave heights and varying phase an-gles was applied to the turning circles of the S-175 container ship. Seo & Kim (2011) coupled a potential seakeeping tool with a manoeuvring model. Both models have a different time-scale and coupling is performed at each time step of the manoeuvring model. The coupling con-sisted of adding the drift forces coming from the seakeeping tool to the manoeuvring model while, position and heading coming from the manoeuvring tool were used to update the seakeeping computations. Rankine panels were used with linearized boundary conditions in the seakeeping tool. The manoeuvring model coef-ficients were derived from empirical formulae or from the experimental data in waves (Yasu-kawa, 2006). Nemzer et al. (2012) presented analytical and experimental procedures to as-sess ship manoeuvrability in wind and waves. The procedures were used to find the minimum speed at which test vessels can maintain course in waves and to determine the range of wave encountering angles where the ship can ma-noeuvre at low speeds. Kim & Sung (2012) validated their two-time scale method with PMM-tests in waves on the KCS.

Direct calculations by CFD. Mousaviraad et al. (2012) used CFD simulation software to conduct free running simulations of ships ma-

158158

noeuvring in deep and shallow water in quar-tering waves. The influence of waves on turn-ing circle and zigzag manoeuvres was quanti-fied. De Jong et al. (2013b) performed simula-tions based on a potential method using a tran-sient diffraction-radiation Green function. Re-sistance, seakeeping, forced motion and free-running tests with hydrojet were performed. Extensive simulations were carried out to study broaching and surfriding conditions depending on speed, wave steepness and heading. Araki et al. (2012b) derived improved coefficients for a 6 DOF simulation model from free running CFD simulations. The original 6 DOF simula-tion model was based on captive tests aug-mented with linear FK forces. Sadat-Hosseini et al. (2011a) used CFD-ship IOWA to simu-late 6 DOF ship motions in following seas and to study the broaching instability limits. The CFD results were compared with model tests. Greeley & Willemann (2012) used a weak scat-tered potential flow theory combined with lift-ing line theory with vortex shedding to derive manoeuvring forces in calm water and waves. Simulations of a the 5415M with bilge keels in following and quartering seas were performed and a comparison of relative importance of Froude Krylov (FK) and hydrostatic forces relatively to lift forces. The main results show that the lift forces are of the same order as the FK forces and in phase. Concerning the yaw moment, the results show that the lifting forces are higher than the FK yaw moment. 6.4 Judgement and analysis

Manoeuvring in waves raised new chal-lenges for both experimental and numerical modelling:

Numerical modelling of ship motion and ship stability in steep following waves with low encounter frequency requires the development of new models, different from traditional

seakeeping and manoeuvring models, with spe-cific models for flow-propeller-rudder interac-tions. More and more teams are assessing this problem using CFD. Nevertheless such compu-tations require a tremendous implementation effort and numerical resources. A solver deal-ing with manoeuvring in waves has to include URANS equations with free-surface effects, ship motions, propeller modelling and wave modelling and propagation.

Manoeuvring experiments in waves also re-quire some new background research to ad-dress arising questions, such as: what are the relevant parameters to be measured to study course keeping in stern waves or turning in waves? What methodology (experimental set-up, initial conditions, number of repetitions, analysis procedure, ...) should be used to get converged mean values and standard deviations of the chosen parameters?

For a ship manoeuvring simulator that takes account of wave action, a force based mathe-matical model is needed. If EFD is used this means that captive model tests are needed in waves. Performing PMM tests in waves can be cumbersome because it leads to an exploding test program: each variation of PMM or wave frequency can lead to a different encounter po-sition between the ship and the wave, which can possibly have an effect on the measured forces.

It is clear that numerical methods for the prediction of the IMO third level assessment are not fully developed yet. An experimental verification of the comprehensive approach is so elaborative that it becomes unaffordable. There are many methods used and every ‘prob-lem’ mentioned in 6.1 cannot be dealt with us-ing the same methods. Regarding the complex-ity of the problem, a workshop on manoeuvring in waves should be organized. Possible topics are:

159159

Give input to the IMO MEPC; Propose dedicated guidelines, both for ex-

perimental and numerical methods to verify and validate possible tools;

Define the need for further research on ma-noeuvring in waves;

Stimulate therefore the creation of bench-mark data on manoeuvring in waves.

Define a common understanding of “the re-sult” and if warranted, define a way of ana-lysing time domain results to reach con-verged final results.

7. SCALE EFFECTS 7.1 Correlation data

Effect of model size. At the SIMMAN2008 workshop, PMM and CMT data for KVLCC1 and KVLCC2 with three different size models were submitted. MOERI and INSEAN carried out PMM tests with a 5.5 m model and a 7.0 m model respectively and NMRI carried out CMT tests with a 2.9 m model. INSEAN and NMRI set the propeller rpm to model self-propulsion, but MOERI set the propeller rpm to ship self-propulsion. Bare hull test data are also avail-able for static drift and pure yaw tests.

Figure 3 shows the comparison of side forces and yaw moments with drift angle for the bare hull of the KVLCC2. They show good agreements generally except in the region of large drift angles where the NMRI data have a larger value than the other data. This can be explained by the effects of Reynolds number on the cross flow drag component which be-comes larger as the drift angle increases. This shows that a 3 m model at scale 1:110 is not large enough to avoid scale effects. Whether the difference is due to the scale, the model

size or the Reynolds number achieved during the measurements is unknown.

Similar conclusions can be drawn for the KVLCC1 equipped with rudder and propeller: the NMRI data deviates from other data as drift angle and yaw rate increase.

Figure 3. Comparison of static drift test data for KVLCC2 (Bare Hull)

Scale effects for podded vessels. Specifi-

cally because during the last ITTC period, the correlation between FRMT and full scale trials were questioned, during the course of the ITTC working period, interviews were held with 5

Drift Angle (deg.)

Y'

-20 -10 0 10 20-0.015

-0.01

-0.005

0

0.005

0.01

0.015

MOERINMRIINSEAN

Drift Angle (deg.)

N'

-20 -15 -10 -5 0 5 10 15 20 25-0.004

-0.003

-0.002

-0.001

0

0.001

0.002

0.003

0.004

MOERINMRIINSEAN

160160

shipyards building podded vessels. These ship-yards indicated that the free running model tests used for the prediction of the manoeuvra-bility were satisfactory. The typical model sizes for FRMT were in the range of 4.5-6.5m and the RPM was power controlled (and hence load dependent). 7.2 Recent studies on scale effects

As observed at the SIMMMAN 2008 work-shop, the application of different self-propulsion points during manoeuvring model tests significantly affects the prediction results. Shin et al. (2012) investigated the effects of the choice of the self-propulsion point on the hy-drodynamic coefficients and the predicted ma-noeuvring performance for KVLCC1 and KVLCC2 by PMM tests and simulations. They carried out PMM tests at both ship self-propulsion point (SSPP) and model self-propulsion point (MSPP) and carried out simu-lations with both a whole-ship model and a modular model. When the whole-ship model is used, the hydrodynamic coefficients obtained at the MSPP give a more stable manoeuvring performance than those obtained at the SSPP. Furthermore, the difference of manoeuvring performance between KVLCC1 and KVLCC2 becomes smaller when the hydrodynamic coef-ficients obtained at MSPP are used. In the modular model, the propeller slip stream effect with different propeller loading conditions is taken into account by the rudder inflow model. The manoeuvring performance predicted by hydrodynamic coefficients obtained at MSPP and SSPP is not significantly different. How-ever, the propeller-rudder-hull interaction coef-ficients obtained from tests at MSPP and SSPP show some difference, although they are as-sumed to be independent of the propeller load-ing condition. This means that the selection of the self-propulsion point also could affect the manoeuvring results even when a modular

model is used. Although the modular model can principally consider the effect of the chang-ing propeller loading on the rudder forces, more careful examination on the effects of pro-peller loading on the propeller-rudder-hull in-teraction coefficients is required to assure that the predicted results by a modular model can be completely free from the effects of self-propulsion point.

To apply a self-propulsion point different from MSPP in free running model tests, it is necessary to equip the ship model with an aux-iliary device to apply a towing force. Tsukada et al. (2013) developed a prototype of an auxil-iary thruster that assists free-running model ships’ propellers. The auxiliary thruster can control its forward force and adjusts the model ship propeller load to arbitrarily time varying target values. Free-running tests of a ship were used to study the effect of propeller load on manoeuvrability. The skin friction correction applied to the container ship model demon-strates the auxiliary thruster works well and the effect on manoeuvrability is clear. Theoretical simulation calculation also confirmed the ef-fect. It was observed that the effect on the overshoot angles is marginal, but the effect on the overshoot time is larger.

The optimal self-propulsion point, which makes the ship model’s rudder inflow dynami-cally similar to the full scale ship’s rudder in-flow, lies between MSPP and SSPP, but there has not been a concrete proposal yet on how to determine the optimal self-propulsion point. Ueno & Tsukada (2013) determined the opti-mal self-propulsion point (REC) as the point at which the rudder force of a model is equivalent to the force of a full-scale ship. They carried out free running tests using an auxiliary thruster and performed simulations at MSPP, SSPP and REC. However, the comparisons of free running model test data and simulation re-

161161

sults are not satisfactory, mainly because of the dependency of the rudder force model.

Sun et al. (2012) presented research on the influence of the Reynolds number on the hy-drodynamic coefficients in submarine model tests. A virtual fluid viscosity was introduced and the mesh motion technology based on mesh deformation was used to calculate the hydrodynamic coefficients of a submarine in different orders of Reynolds numbers．They also examined the influence of Reynolds num-bers in submarine manoeuvring hydrodynamic calculation．

7.3 Recommendations for the study of

scale effects

Systematic Method. Since there are many contributors to scale effects, it is not easy to es-tablish a standard full-scale extrapolation method from manoeuvring tests in the near fu-ture, like a full-scale powering prediction method. In this section a systematic method to identify possible scale effects prior to model test is presented.

In the 26th ITTC manoeuvring committee’s

report (ITTC, 2011), several correlation meth-ods to minimize scale effects were reviewed and categorized in-to pre-test methods, post-test methods and during-test methods. Figure 4 and Figure 5 represent a flow chart for free model and captive model tests respectively to-gether with correlation methods applicable at each stage. Each method, however, requires knowledge on scale effects and some tools to be developed.

The first decision in model tests is the size of the model, which is so critical to scale ef-fects that it must be reviewed with available model-ship correlation data and/or some tool to be able to roughly estimate scale effects. CFD

can be used to estimate possible scale effects. The model size is restricted by the dimension of the facility and stock propellers. In this case, the attachment of a flow stabilizer or turbu-lence stimulator can be considered to minimize the scale effects due to a too small model size. A flow analysis in CFD can assist to find a proper size and position of the flow control de-vices.

The determination of the self-propulsion

point is also critical in accurate full-scale pre-diction, especially for free model tests, see for instance the method proposed by Ueno & Tsu-kada (2013). The magnitude of the rudder an-gle can also be adjusted to apply a dynamically equivalent rudder force. It requires information on the effects of the Reynolds number on the rudder force and on the inflow to the rudder.

Post-test methods to correct the test results require an abundant sea-trial database and reli-able mathematical models to describe the ship dynamics.

Before these diagrams and methods can be

matured, much effort will be needed: robust es-timation of hydrodynamic coefficients using SI, established methods to correct hydrody-namic derivatives to full scale and methods to control boundary layer.

162162

Figure 4. Flow diagram for free model test and model-ship correlation method applicable at

each stage.

Figure 5. Flow diagram for captive model test and simulation method and model-ship correla-

tion method applicable at each stage

Necessity of CFD research. Knowledge on scale effects is still limited due to the scarce in-formation on full-scale data. CFD could allow computations to be performed to investigate the similarities (Froude, Strouhal, cavitation) and more specifically the viscous (Reynolds) ef-

fects. Comparison of full scale and model scale CFD computations may then appear as a good candidate to study such effects. However, to study scale effects using CFD, many issues have to be overcome: At full scale, the grid size in the direction

normal to the hull has to be adjusted to full scale boundary layer characteristics which leads to a large mesh size.

Full scale computation of ship manoeuvring requires a huge computational effort and validation data including local flow charac-teristics (boundary layer flow for example), which is scarce, especially for manoeu-vring. There is a strong need for research on scale

effects for knowledge and identification of the limits of present day experimental procedures. CFD is mature enough to be used for specific studies on the different origin of scale effects, such as: influence of the scale on non-linear coefficients, influence of the scale on the wake fraction and propeller loading, influence of the scale on rudder inflow and rudder forces.

For the research on scale effects, more knowledge on propeller-hull-rudder interaction is required. Fukui (2012) has investigated the interaction coefficients between hull and rudder in the MMG model using CFD. This kind of approach is very promising to understand the physics of flow into the rudder during manoeu-vring motion and can easily be extended to un-derstand the mechanism of scale effects. 8. MANOEUVRING CRITERIA

This section gives an overview of criteria that are in use, including those that are com-monly and less commonly used. Apart from the commonly known IMO criteria for ship ma-noeuvrability, the heel angles, the guidelines

163163

for naval vessel ship manoeuvrability, the SO-LAS rudder tests, criteria for inland ships, fast ships and dedicated low speed manoeuvres are studied. 8.1 Overview of Existing Manoeuvring

Criteria

IMO Criteria for Manoeuvrability. At the 25th ITTC (2005), a review was given based on the experience with the (at that time new) IMO criteria MSC.137(76). The 2005 ITTC-MC report describes the history of the devel-opment of the IMO manoeuvring criteria. Turn-ing ability, initial turning, course keeping and stopping ability were at that time considered the manoeuvring criteria that were to be en-compassed. This was the first ITTC conference taking place after the IMO criteria for ship ma-noeuvrability had become mandatory in 2003. In 2005, it was discussed how many institutes were considering the code as mandatory and how they assured compliance with the criteria. The interpretation was quite diverse. The MC believes at present that the manoeuvring crite-ria are less ambivalent, and considered more widely accepted by the shipbuilding commu-nity. Moreover, currently many researchers know the actions to be undertaken to assure that the ships are able to meet the requirements of IMO MSC 137(76).

Criteria for Heel Angles during Turn for Passenger Vessels. In the international code on intact stability IS2008, issued by the IMO, it is stated that for passenger vessels, the angle of heel on account of turning shall not exceed 10° when calculated using the following formula as heeling moment due to turning:

0.2 ∆ (1)

The RINA has proposed amendments on

the code, amongst others because it was not

clear whether this code was related to the con-stant heel angle during a turn, or related to the more critical and larger outward initial heel an-gle in a turn. RINA proposed 15° as criterion for the maximum outward heel angle in a ma-noeuvre and 10° as criterion for the maximum constant heel angle in a turn. The objective of the criterion is not to prevent capsizing, but to ensure passenger safety. RINA recommends to use simulations or model tests or full scale measurements to demonstrate compliance with these criteria.

It is not the mandate of the ITTC-MC to

come up with a level value for the maximum heel angle, it is the mandate to have an opinion on the applicability and realism of the proce-dures to achieve the level. The MC has investi-gated the applicability of the rule and compared the actual measured heeling angle due to turn-ing with the IMO rule. The opinion of the MC is that it may be the maximum angle which is more representative for the passengers’ safety than the constant heel angle. Furthermore, the formula originally proposed by IMO is not rep-resentative for the maximum heel angle.

SOLAS Test. A SOLAS test is often used

(considered mandatory) to demonstrate at full scale that the rudder engine has enough capa-bility. The aspect to prove is that at full speed, the rudder should be able to move from 0° to +30° to -35° and back to zero. The objective of the manoeuvre is to verify that the rudder movement from +30° to -35° should take place in 28 seconds or less. Care should be taken that the heel angle during such test does not become critical. 8.2 Inland Ships

Europe. In Europe the inland ships are as-signed to a class based on their length and beam. The classes vary from I (38.5 m x 5.05

164164

m) to VII (285 m x 34.2 m). While the fist con-sist of a small self-propelled barge, the latter represents a push convoy of one pusher and 6 barges in 3 by 2 or 2 by 3 configuration. The inland waterways receive the same classifica-tion, for instance a ship of class III can sail on waterways of class III and higher.

The Central Commission for Navigation on the Rhine (CCNR) has issued manoeuvring cri-teria for vessels sailing on the river Rhine. These criteria concern speed, stopping and turning abilities and evasive capabilities. The trials have to be carried out with a minimal loading condition of 70% in calm water in a channel of sufficient width and minimum 2 km straight. The minimal under keel clearance is 20% of the draft, but never lower than 0.5 m.

Every inland ship, including convoys needs to be able to reach a speed of 13 km/h ahead and 6.5 km/h astern. Any ship needs to be able to reach 6.5 km/h with its installed emergency power (e.g. a bow thruster). Ships that are up to 110 m x 11.45 m need to be able to stop from 13 km/h within 305 m. Larger ships have to stop within 350 m.

A specific kind of test for inland ships is the so-called evasive manoeuvre, also performed at 13 km/h, that is comparable to a zigzag ma-noeuvre, but the rudder checking is performed based on the yaw rate instead of the heading deviation. The yaw rate to be checked depends on the ship’s size and the rudder angle, which depends on the under keel clearance. The crite-rion depends only on the period of the evasive manoeuvre, which is a function of ship size and under keel clearance, see Table 3.

Table 3. Evasive manoeuvre: maximal period.

UKC (% of draft) 40 40 >40 >100

Used rudder angle (°) 20 20

Size (LxB m²)

Yaw rate check-ing (°/min) Maximal period (s)

110x11.45 20 28 150 110 110 193x11.45 110x22.9

12 18 180 130 110

193x22.9 8 12 180 130 110 270x22.9

193x33.35 6 8 Expert judgement

China. Manoeuvring standards were issued

for the Yangtze river because both the dimen-sion and the speed of the vessels increase and the fact that hazardous goods are being trans-ported along the river.

The maximum length of the vessels or con-voys is 150 m. According to hydrological con-ditions the river is divided in several navigation areas, namely, in increasing order of difficulty, A, B, C and J (J1: very turbulent, J2: turbulent). Like in Europe each ship (type) can be as-signed to a limit class. Sometimes due to changing hydrological conditions (which can also be a consequence of operational decisions) a section of the river can have a more restricted class, for example near the Three Gorges Dam the class can be restricted to J2. Typical ships are: A: large dimension (> 130 m); B: large B/D ratio: B > 20 m, T: 3 to 5 m; C: twin propeller.

The following manoeuvring indices are regulated (JT/T 258-2004): Stability; Turning; Stopping; Astern stability.

165165

Table 4 shows the requirements to be met for each manoeuvre and each navigation class. In this table the following variables are used: ΔC0: the allowable course variation at δ =

0°, measured over 3 min. δ0: the allowable rudder variation to keep a

prescribed course during 5 min. y0-15: the minimal allowable yaw rate when

moving towards 15°/min with a rudder an-gle of 15°

D0 and Ah represent the dimensionless tac-tical diameter and track reach;

δA: the allowable rudder variation to keep a prescribed course astern during 3 min.

The manoeuvres have to be carried out at a

steady speed, the value of which is not speci-fied. Due to water level variations in the Three Gorges dam, the navigation conditions can vary significantly. In deep conditions navigation needs to occur in the vicinity of flooded banks, while in shallow conditions 180° turning is im-possible. The strong current (~3m/s) of the river challenges both downstream navigation (less rudder efficiency) and upstream naviga-tion (power lacking).

Table 4. Yangtze river manoeuvring require-ments (Standard Ship Type Index System of

Inland Transportation Vessel).

Applicability of the criteria. In practice modern inland ships do not have significant problems to comply with the CCNR criteria. On the other hand there is a tendency to in-crease the class of the European waterways. A lot of research is going on to investigate whether an inland waterway can accept a larger class inland vessel. This research consists of analysing a wide range of scenarios and is typi-cally performed on a ship manoeuvring simula-tor, Eloot & Delefortrie (2012), which of course requires the availability of realistic ma-noeuvring models in restricted waters. Hase-gawa (2013) also sums up the difficulties and challenges of river transportation in Asia. 8.3 Waterjet/Fast Ships

Whereas the manoeuvring characteristics and criteria of displacement vessels are well understood and documented, the same informa-tion regarding high speed craft is not so readily available. Some seminal works discussing spe-cific manoeuvring criteria for high speed ves-sels (HSVs) are presented in this section.

The stopping manoeuvre for HSV was in-vestigated by Varyani & Krishnankutty (2009). The stopping abilities of vessels ranging from medium speed containership to high-speed ves-sels have been estimated using analytical mod-els, verified with known results and checked for the actual stopping criteria. The authors verified that the stopping ability of high-speed vessels with waterjet propulsion has been found to be far better than the IMO manoeu-vring criteria, which are based on stopping tests performed on conventional vessels. This result is coherent with the fact that a HSV must stop in a smaller distance for safety reasons, since if there is traffic around, the other vessels do not have sufficient time to avoid collision with HSVs. This paper is an indication that a more

166166

stringent stopping criterion must be defined for HSVs.

The turning capability of HSVs was studied by Lewandowski (2004), who derived a regres-sion equation based on full-scale data. The work of Bowles (2012) examined various as-pects of the turning capabilities of a high speed monohull craft and based on the previous stud-ies, tried to define a set of criteria adequate for the turning ability of HSVs. The first criterion defined by Bowles (2012) is that a high speed monohull should be capable of a predictable, controllable hard over turn at maximum speed while rolling inboard to the turn. The author demonstrated several problems associated with outboard rolling angles related to safety and comfort. Furthermore, a high speed monohull should be able to manoeuvre within a turning circle diameter not larger than 110% of the predicted diameter based on the regression equation developed by Lewandowski (2004). Finally, a high speed monohull (recreational craft passenger vessel) should not be able to execute turning manoeuvres if the horizontal accelerations developed exceed 0.35g to avoid being hazardous to occupants. A method for calculating the minimum recommended turning circle diameter is also derived in that work. 8.4 Naval ships

An initiative of several NATO countries has led to the development of proposed manoeu-vring standards for naval vessels. This process is described by Örnfelt (2009). Since about 2002, the specialist team on seagoing mobility formed under NATO Maritime Capability Group 6 on Naval ship design has been pro-gressing significantly in the development of new mission-oriented criteria, which include a large envelope of operational requirements. This work has resulted in several Allied Engi-neering Publications (ANEP) like NATO

ANEP 70 (2003), ANEP 78 (2007) and ANEP 79 (2007). Based on the experience obtained from these ANEPs, definitive criteria in the format of a NATO STANAG have been devel-oped, and is at present under ratification (NATO STANAG 4721). Justification for the need of a common naval manoeuvrability stan-dard is given by Örnfelt (2009). Examples to get experience on how to apply the manoeu-vring criteria to naval vessels are described by Armaoglu et al. (2010) and Quadvlieg et al. (2010). Armaoglu et al. (2010) explain the draft criteria, Quadvlieg et al. (2010) explain an up-date of the criteria and a practical application on the 5415M (the ITTC benchmark vessel). The main objective was not to judge if the 5415M would meet the criteria, but to judge if the tools that are available have the capabilities to predict whether the performance could be met or not.

The key of these developments is that the manoeuvring criteria are related to the general profile of a naval ship (the safety) and to mis-sion abilities (for example, for mine hunting, different manoeuvrability may be required than for replenishment at sea).

To quantify the “safety”, the following ba-sic capabilities are distinguished. Transit and patrol; Harbour manoeuvring.

To quantify the “mission ability”, the fol-lowing missions are distinguished: Anti-submarine warfare (pro-active); Anti-submarine warfare (re-active); Mine warfare (hunting); Mine warfare (sweeping); Mine warfare (avoiding); Anti-air warfare (pro-active); Anti-air warfare (re-active); Vehicle interaction (replenishment at sea);

167167

Vehicle interaction (air vehicle); Vehicle interaction (sea vehicle); Vehicle interaction (sea vehicle LPD/

Dock).

A minimum amount of manoeuvring abili-ties are required to fulfil the missions. The fol-lowing are the manoeuvring abilities: Course keeping (where the maximum al-

lowed course deviation (95% probability) in a sea state has to remain below a criterion level);

Track keeping (where the maximum al-lowed track deviation (95% probability) in a sea state has to remain below a criterion level);

Turning (quantified by the tactical diame-ter);

Initial turning (quantified by the time it takes to reach 20 degrees heading change after setting the rudder to 20 degrees. This can be obtained from a 20/20 zigzag test.);

Yaw checking (quantified by the first over-shoot time in a 20/20 zigzag test);

Turning from rest (quantified by the time needed to turn to 90 degrees from rest);

Stopping (quantified by the track reach from a stopping test);

Acceleration (measured by the maximum acceleration during a manoeuvre from 0 to maximum speed);

Astern course keeping (where the maxi-mum allowed course deviation (95% prob-ability) in a sea state has to remain below a criterion level while sailing astern);

Station keeping (showing the ability to maintain a position with environmental dis-turbances, quantified by a heading/position deviation that the ship is not to supersede during 95% of the time);

Lateral transfer (quantified by the crabbing velocity);

Turning from rest (quantified by the time needed to turn to 90 degrees at rest using all manoeuvring aids);

SDNE (standard deviation of navigational error), this involves not only the hydrody-namic capabilities of the ship, but also the accuracy of navigational aids, including navigational sensors and autopilot. This is quantified by the standard deviation from a predefined earth fixed track).

For every mission or for safety, a different

speed is to be selected at which the manoeu-vring ability needs to be demonstrated. Fur-thermore, for the requirements of course keep-ing, track keeping, astern course keeping and station keeping, a target sea state needs to be selected.

The required levels for every manoeuvring ability, (for example a minimum tactical di-ameter of 3.5 ship lengths) have a minimum level (i.e. the level that at least needs to be met) and a target level (the vessel that meets that level shows superior performance).

Apart from the NATO development, the Korean Navy also employed a similar structure to quantify the manoeuvring performance of their naval vessels together with the IMO crite-ria. Rhee et al. (2013) established the relation-ship among ship types, missions and manoeu-vring tests based on naval experts’ opinions, and finally proposed manoeuvring criteria for Korean naval ships with respect to ship types, referring to the criteria of NATO, Lloyd regis-ter (2006) and Korean naval ship’s trial data.

168168

8.5 Pod-Driven Ships

A question that is often raised is whether the manoeuvring criteria of IMO would be valid for pod-driven ships of over 100 m length as well. IMO manoeuvring criteria were devel-oped for conventionally propelled and steered ships. This is augmented by a discussion about the large heel angles that podded vessels may encounter when sailing at full speed and apply-ing 35 degrees of helm. Also the crash stop test was under discussion as the loads on the bear-ings during the full scale crash stop test are not desired.

To answer these questions it is important to address each manoeuvre separately. The MC made a couple of mini-interviews with ship-yards regularly building podded vessels and in-stitutes having experience with the podded ves-sels.

To demonstrate adequate turning ability, the turning circle test is used. On full scale trials a common approach among the interviewed shipyards is that it is considered acceptable to carry out the turning circle test with a lower pod angle than 35 degrees, as long as with this lower pod angle, it is also demonstrated that the criteria of advance and tactical diameter can be met.

For course keeping, yaw checking and ini-tial turning, the zigzag test is used. Investiga-tions of Woodward et al. (2009) have revealed that the application of the same criteria for the overshoot angle of the 10/10 zigzag test and the 20/20 zigzag test are realistic and valid. The zigzag test is still a measure for directional sta-bility and also a measure for the steering diffi-culty. So, for course keeping, yaw checking and initial turning, the 10/10 and 20/20 zigzag tests are to be carried out and the results judged in the same way as for the conventionally pro-pelled ships. Kobyliński (2012) warns that the

overshoot angles of ships with podded propul-sion may be larger than for ships with conven-tional twin screw twin propeller arrangements.

For crash stop tests, it is considered accept-able to perform the crash stop test in such a way that it can be demonstrated that the ship can stop within 15 ship lengths. 8.6 Manoeuvres in Restricted Conditions

Initiatives to develop criteria in restricted conditions. The restrictions can have different sources, namely speed limitations, shallow or restricted water or harsh weather conditions.

SNAME Panel H-10 performed a study of the issues of characterising slow ship manoeu-vring performance (Hwang et al., 2003). They surveyed senior mariners, simulator operators and other relevant professionals to collect in-formation on the characteristics of slow speed manoeuvring. They also considered that the test procedure should not be complex and the per-formance indices should be easy to derive, in-tuitive, quantifiable, and of practical use to both operational people and technical people. Based on the survey results and the require-ment of tests, they proposed eleven basic slow speed manoeuvres.

Abramowicz-Gerigk (2005) evaluated the manoeuvres proposed to characterize the ship performance in constrained waters previously proposed by Hwang et al. (2003). The investi-gations used full mission simulators and a training vessel of Gdynia Maritime University, and considered the back & fill - fill first to starboard manoeuvre. The slow speed manoeu-vres involve rather complex hydrodynamic phenomena, large drift angles, big propeller loadings, strong interaction between ship hull and control devices. There are frequent piloting commands and the vessels are mainly in transi-

169169

tory motion (not steady state), and the opera-tion involves different combinations of vessel moving and propeller thrust directions (four quadrant operation). Due to this complexity, the investigations have concluded the necessity of full scale trials since the accuracy of mathe-matical models in such cases are not always satisfactory. The author also concluded that it was still too early to define standards for slow speed manoeuvrability.

In Europe several joint-industry projects have started that focus on the validation of ma-noeuvring models, including scale effects and manoeuvring in waves.

An on-going R&D project sponsored by Research Council of Norway, Norwegian and international partners named "Sea Trials and Model Tests for Validation of Ship-handling Simulation Models" aims to continue this effort to define standards for slow speed manoeuvra-bility (2013 to 2016). The main objective is to develop and apply a method for validation of numerical ship models used in engineering tools for studies of ships' manoeuvring per-formance in deep and restricted waters and ship handling training simulators. This will be done by comparing outcomes of numerical simula-tion models to measured responses from sea trials of selected case vessels. It also aims to establish benchmark datasets for validation of simulation models. Some preliminary informa-tion can be found at Marintek, (2014).

Figure 6. Project layout of MAROFF KPN.

The project SHOPERA "Energy Efficient Safe Ship Operation" also runs from 2013 to 2016 and started from the concerns on suffi-cient propulsion and steering power in harsh weather conditions due to the EEDI. The aims of the project are: Further development and refinement of ex-

isting hydrodynamic simulation software tools for the efficient analysis of the seakeeping and manoeuvring performance;

Performing seakeeping/ manoeuvring model tests in combined seaway/wind envi-ronment by use of a series of prototypes of different ship types to validate the numeri-cal tools. Full scale trials will also be used as a validation tool.

Develop new guidelines for the required minimum propulsion power and steering performance to maintain manoeuvrability in adverse conditions.

Development of criteria. In a general point

of view, to select manoeuvring criteria, the fol-lowing sequence is to be followed: 1. The selection of an important characteristic

(for example turning ability) 2. The selection of a representative measure

(for example turning radius) 3. The selection of a limiting value (for exam-

ple 5 ship lengths)

Regarding the first item, the MC considers that, just as in deep water, there could be re-quirements for turning ability and yaw check-ing. A minimum amount of turning should be considered, related to the turning radii that a ship has to make in shallow water as well, when approaching a harbour. A minimum level of course keeping and initial turning is required as well, such that the ship should be able to not turn too drastically so that the rate of turn can be sufficiently counteracted.

170170

Regarding the second point, it is essential to define the speed at which the manoeuvres are to be carried out. It needs to be representative for the ship. It is considered to be the speed at which ships are approaching the harbour, but at which the ships are not yet under tug/bow thrusters assistance. Considering that this slow-ahead will have different speeds for all ships, this means that there is some variability of the speed allowable for this.

The turning ability could be typically ex-pressed by a turning circle test or a test at which maximum rudder is given and a constant rate of turn is achieved. In shallow water, this rate of turn converges much quicker to a con-stant value than in deep water, so perhaps al-ready only a partial turning circle could be suf-ficient. The course keeping ability is in shallow (and/or restricted) water often evaluated in an evasive type of manoeuvre like applied to inland ships (see section 8.2). The rudder is ap-plied to an angle (maximum angle). A rate of turn builds up, and at a certain value, the rudder is swung over to the opposite side. This is simi-lar to a classical zigzag test, but now with the rate of turn as lead signal.

The international guidelines and rules for port and navigation channels design such as PIANC (MarCom Working Group 121, 2014) and ROM (2000) are intrinsically related to the definition of standards for slow-speed manoeu-vres. Those guidelines take into account “aver-age” vessels navigating to or from the berth, and design the port/channels dimensions ac-cordingly. A more accurate definition of the requirements for the vessels during the port manoeuvres will directly result in a more accu-rate definition for the dimensions of the ports and channels.

9. PROCEDURES 9.1 Overview

The MC reviewed the procedures and guidelines under its responsibility and made updates as follows:

7.5-02-06-01 Free Running Model Tests: descriptions on the parts of the procedures which are common in captive model tests and free running model tests were unified. A sec-tion on restricted water was added. The defini-tion of deep, shallow and restricted water was included. Specific test types in shallow and re-stricted water have been added for free running model tests, e.g. evasive (avoidance) tests are different in shallow and deep water. The as-pects which require special considerations when performing manoeuvring tests in shallow and restricted water were specifically outlined.

7.5-02-06-02 Captive Model Tests: descrip-tions on the parts of the procedures which are common in captive model tests and free run-ning model tests were unified. The SIMMAN 2008 tests were added to the benchmark list. The definition of deep, shallow and restricted water was included. The explanation of multi-modal tests was added. Special considerations for shallow and restricted water were added. Because there is now a section related to uncer-tainty analysis for captive model tests, a large part of UA was deleted from this procedure, and reference is given to the procedure for un-certainty analysis of captive model tests, which received a very significant update.

7.5-02-06-03 Validation of Manoeuvring Simulation Methods: more precise definitions of deep, shallow and restricted water are in-cluded. References for benchmark data for shallow and restricted water manoeuvres have

171171

been added. A general revision on the nomen-clature was also carried out.

7.5-02-06-04 Force and Moment Uncer-tainty Analysis on Captive Model Tests: the procedure has been very significantly updated. The text was adapted to ISO GUM and the ex-ample was rewritten for clarity. Furthermore, as the previous procedure provided just an ex-ample of an uncertainly analysis towards the measured force during captive tests, the pre-sent procedure describes how the uncertainty in the measured force can be used to determine the uncertainty of a characteristic derived from a manoeuvre based on simulations which are based on captive tests. The description of how this ‘from-begin-to-end’ uncertainty chain is working is fully elaborated. An example from beginning to end is not yet included.

7.5-02-05-05 Manoeuvrability of HSMV: the year of the sources has been updated and minor English corrections have been applied. The procedure reflects that the worldwide ex-perience to HSMV is limited and that the ITTC recommends to perform free running tests or CMT tests in 6 DOF, not in 3 or 4 DOF.

The MC also developed two new guide-lines, with the following topics:

7.5-03-04-02 - A new guideline named

"Validation and Verification of CFD Solutions in the Prediction of Manoeuvring Capabilities" has been made. The guideline describes how Validation and Verification (V&V) can be per-formed for CFD based simulation of captive and free-running conditions. The verification covers the assessment of the numerical uncer-tainty and hereby gives an indication of the un-certainty related to the simulated results. The validation concerns the comparison between computation and measurements in order to quantify how well the computation agrees with the measurement, taking both numerical and

experimental uncertainty into account. More details about this new guideline are given in Section 9.2.

7.5-02-06-05 Guideline on Uncertainty Analysis on Free Model Tests. The purpose of the guideline is to provide guidance for ITTC members to perform uncertainty analysis (UA) of a model scale free-running model test fol-lowing the ITTC Procedures 7.5-02-06-01, ‘Free Running Model Tests’. It is a guideline until it has proved itself for at least one 3-year period of the ITTC so that more institutes can elaborate this and become familiar with the concept of uncertainty analysis for free running model tests. More details about this new guide-line are given in the Section 9.3 9.2 New guideline on V&V of CFD Solu-

tions in the Prediction of Manoeuvring Capabilities

Captive PMM type CFD simulations are

becoming more widely used, therefore a V&V guideline for this type of simulation has been created. The captive part of the guideline cov-ers stationary straight-line motions (static drift, static rudder etc.), dynamic harmonic motions (pure sway, pure yaw etc.) and stationary circu-lar motions.

Static simulations are typically treated as steady computations and the hydrodynamic forces and moments will in this case be con-stant. Dynamic simulations are treated as tran-sient computations, since the flow is not steady due to the dynamic motion of the ship and the hydrodynamic forces and moments will be rep-resented as time series. V&V in the guideline is therefore focused on single value forces or moments for the static conditions, while for the dynamic simulations the focus is put on time series for forces and moments, either in the

172172

time domain or in the frequency domain (Fou-rier coefficients).

In the guideline the numerical error covers contributions from the iterative solution proce-dure and the grid for all kinds of simulations. The time step size is also concerned for dy-namic simulations.

The free running part of the guideline cov-ers V&V of free running simulations, where the trajectory of the manoeuvring ship is pre-dicted directly by CFD. The focus is on classi-cal IMO manoeuvres like ±35° turning circle and 10/10 or 20/20 zigzag tests and the goal is to make V&V representative for the trajectory instead of the force level.

In reality it is quite difficult to make a for-mal V&V on time level for the trajectories, so a more practical approach is to consider the global parameters representing the trajectory. This means that for turning circles it is recom-mended to consider the following global pa-rameters for V&V: tactical diameter, advance, transfer, yaw rate once steady in turn, peak yaw rate, drift angle once steady in turn, speed loss and heel angle (4 DOF). For zigzag tests, rele-vant parameters are: first and second overshoot angles, first and second overshoot time, peak yaw rate and period.

For these global parameters the guideline suggests that the numerical error estimate cov-ers contributions from the iterative solution procedure, the grid and the time step size.

Assuming that the numerical uncertainties are estimated during the verification procedure described in the guideline and that model test data with experimental uncertainties is avail-able the guideline finally gives a procedure on how the validation should be made in order to check how well the CFD simulation captures the manoeuvre of interest.

9.3 New guideline on UA in free running manoeuvring tests

This newly developed guideline is based on

ideas proposed by Quadvlieg & Brouwer (2011). The ideas were sparked by discussions during the SIMMAN2008 workshop, because it was deemed that the initial conditions at the start of a manoeuvre were significantly deter-mining the outcome of a manoeuvre such as the first overshoot angle. A methodology is de-scribed that takes into account these effects, and is based on the uncertainty propagation technique. The methodology uses the sensitiv-ity of the final outcome to the initial condition. It is important to note that this sensitivity coef-ficient may be determined based on simula-tions, as long as the simulations are adequate enough to capture the desired effect. The guideline comes with an example. In the light of the comparison between the manoeuvring predictions made by different prediction meth-ods in the frame of the SIMMAN2014 ma-noeuvring workshop, the determination of the uncertainties of free running manoeuvring tests will gain importance. 10. CONCLUSIONS 10.1 Using Experiments as a Tool to ad-

vance the Knowledge in Manoeuvring

As in previous years, work has been con-ducted to investigate standard manoeuvres in deep unrestricted water. However, there is a growing trend towards research in shallow and restricted water. For example, a significant amount of research into vessel behaviour in locks, ship-ship interaction and ship-bank in-teraction can be observed. Experiments have been carried out with false bottoms in towing tanks and basins to study the behaviour of ships in shallow water. Further work is required to

173173

establish the length of the false bottom needed to ensure the flow around the model is not ad-versely influenced by the ends of the false bot-tom. The rigidity of the false bottoms is also a large concern.

There is a trend towards more detailed spe-cialized manoeuvring research, such as investi-gating propulsion system operation settings, asymmetrical propeller loading effects, ap-pendage configurations and the effect of static trim and heel angles. Also, a significant quan-tity of work has been conducted on SI, includ-ing the use of artificial intelligence. 10.2 Using Simulations as a Tool to advance

the Knowledge in Manoeuvring

The viscous CFD methods have not evolved that much over the last three years, but have become more widely used. The most used ap-proach is the simulation of captive deep water conditions to provide input for manoeuvring simulations. The propeller modelling however remains a weak point.

In restricted water the use of CFD is mainly focussing at ship-bank interaction or ship-ship interaction. The latter has been tackled thor-oughly, also with potential flow models. In any case more emphasis should be put on verifica-tion and validation of the simulation models. 10.3 Benchmark Data

Concerning generation of new benchmark data most work has been performed with sur-face ships. The upcoming SIMMAN2014 workshop on manoeuvring has facilitated much new deep and shallow water data for both KCS and KVLCC2. Further, it seems that both of these ships plus the naval combatant 5415M have been adopted by the community as

benchmark ships. In addition, HTC, S175 and DARPA SUBOFF became benchmark cases. Also data for more complex restricted water cases are made available. So, it appears that there is focus on benchmark data generation in the community and that people are using it. This is positive and valuable in order to support the validation of the numerical simulation methods, which are being used widely. 10.4 Manoeuvring and Course Keeping in

Waves

Concerning manoeuvring in waves, the IMO criteria are currently defined, which has been discussed in the report. The title “ma-noeuvring in waves” may cover very different topics (broaching, course-keeping, manoeuvres at sea). For each of these topics, different methodologies are used. The MC grouped the methodologies that are in use in logical groups. FRMT are still giving the most complete pic-ture of reality including events like for example propeller ventilation. Simulations are however strived at for obvious reasons. There is no con-sensus yet on the preferred simulation method per topic.

10.5 Scale Effects

Some researches were carried out to inves-tigate the effect of the self-propulsion point on the manoeuvrability. However, research on scale effects is hampered by the absence of good quality open full scale data that can serve as benchmark. As an alternative CFD can be used as a tool to assess geosim conditions. 10.6 Manoeuvring Criteria

An overview is given for criteria for ship manoeuvrability.

174174

The IMO criteria for ship manoeuvrability

are in place and well established and used. They are valid for podded vessels and ships with flap rudders as well.

The criteria for heel angles initiated by turning are not very well established and lack some realism. They need further improvement.

For naval vessels and inland vessels, ma-noeuvring standards are in place. For planing vessels and manoeuvres at slow speed and shal-low water, proposals for criteria are made and summarised in this section.

It is not the mandate of the ITTC-MC to generate criteria, but the ITTC-MC will have an opinion about the realism, practicality and applicability and can, as such, contribute to the development of criteria. 10.7 Procedures

The MC reviewed the procedures and guidelines under its responsibility. Major up-dates and improvements were done in 7.5-02-06-04 Force and Moment Uncertainty Analysis on Captive Model Tests. Additional restricted water recommendations have been added to captive and free running procedures.

The MC also developed two new guide-lines. The guideline "Validation and Verifica-tion of CFD Solutions in the Prediction of Ma-noeuvring Capabilities" (7.5-03-04-02) de-scribes how Validation and Verification (V&V) can be performed for CFD based simulation of captive and free-running conditions. The veri-fication covers the assessment of the numerical uncertainty and hereby gives an indication of the uncertainty related to the simulated results. The validation concerns the comparison be-tween computation and measurements in order

to quantify how well the computation agrees with the measurement, taking into account both numerical and experimental uncertainty.

The Guideline on Uncertainty Analysis on Free Running Model Tests (7.5-02-06-05) pro-vides guidance for ITTC members to perform uncertainty analysis (UA) of a model scale free-running model test following the ITTC Procedures 7.5-02-06-01, ‘Free Running Model Tests’. Amongst others, this guideline uses the uncertainty propagation techniques to quantify the effect of the initial conditions on the final result. 11. RECOMMENDATIONS

Continue work in order to have a full set of benchmark data for each of the benchmark hulls (KVLCC2, KCS, 5415M, HTC, SUBOFF and S175 – manoeuvring in waves). Ideally add real vessels to the benchmark set.

Capitalize the momentum created by SIM-MAN2014 and the conference on shallow and confined water to continue the development of V&V of ship manoeuvring simulation methods, including CFD.

Extend the UA for captive model tests from measurements towards the mathematical mod-els and the predicted manoeuvres. Elaborate with an example.

Issue a new questionnaire concerning the procedure of captive tests (7.5-02-06-02), with particular attention to the use of PMM and hexapod, and have the procedure of captive test (7.5-02-06-02) revised, including 6 DOF con-siderations.

Revisit the full scale manoeuvring trials procedure (7.5-04-02-01). Monitor the full scale measurement campaigns starting up in the

175175

joint industry projects to use this as a starting point for scale effects research, supported by CFD.

Investigate the effect of movable bottoms to study the behaviour of ships in shallow water.

Stimulate the use of proposed low speed manoeuvres (full scale, free running, simula-tion model). Share the results and build up a database to identify possible manoeuvring cri-teria.

Manoeuvring in waves needs specialist knowledge from various fields and has a vari-ety of applications and goals. It is therefore recommended to work either with a specialist committee on manoeuvring in waves or to or-ganize a workshop on manoeuvring in waves or to have a dedicated member both in the seakeeping and the manoeuvring committee to address the topic. Liaise with IMO or IACS to address manoeuvring in waves in the future.

The Manoeuvring Committee recommends to the Full Conference to: Adopt the revised procedure 7.5-02-06-01

Free running model tests Adopt the revised procedure 7.5-02-06-02

Captive model tests Adopt the revised procedure 7.5-02-06-03

Validation of manoeuvring simulations models

Adopt the revised procedure 7.5-02-06-04 Uncertainty analysis on captive model tests

Adopt the revised procedure 7.5-02-05-05 Manoeuvrability of HSMV

Adopt the new guideline 7.5-03-04-02 Verification and validation of CFD solu-tions in the prediction of manoeuvring ca-pabilities

Adopt the new guideline 7.5-02-06-05 Un-certainty analysis on free model tests

12. REFERENCES 26th ITTC MC final report, Proceedings of the

26th ITTC – Vol. I; Rio de Janeiro, Brazil, 2011

Abramowicz-Gerigk, T., 2005, "Investigations

on Standards for Ship Manoeuvring Performance at Slow Speed in Constrained Space", 11th International Congress of the International Maritime Association of the Mediterranean, Lisbon, Portugal.

Ahmed, Y. and Hasegawa, K., 2013, "Im-

plementation of Automatic Ship Berthing Using Artificial Neural Network for Free Running Experiment", 9th IFAC conference on Control Applications in Marine Systems (CAMS). Part 1, Osaka, Japan, pp. 25–30.

Akinturk, A., Islam, M., Veitch, B. and Liu, P.,

2012, "Performance of Dynamic Azi-muthing Podded Propulsor", International Shipbuilding Progress, Vol. 59, pp. 83–106.

Amin, O. and Hasegawa, K., 2012, "Numerical

Prediction of Flow Field around a Ship Drifting with Various Angles", MARSIM 2012, Singapore.

Amini, H. and Steen, S., 2012, "Theoretical

and Experimental Investigation of Pro-peller", International Shipbuilding Progress, Vol. 59, pp. 55–82.

Araki, M., Sadat-Hosseini, H., Sanada, Y.,

Tanimoto, K., Umeda, N. and Stern, F., 2012, "Estimating Maneuvering Coeffi-cients Using System Identification Methods with Experimental, System-Based and CFD Free-Running Trial Data", Ocean Enginee-ring, Vol. 51, pp. 63–84.

176176

Araki, M., Sadat-Hosseini, S., Sanada, Y., Umeda, N. and Stern, F., 2012, "Study of a System-based Mathematical Model Using System Identification Technique with Experimental, CFD, and System-based Free Running Trials in Following Waves", 11th International Conference on the Stability of Ships and Ocean Vehicles , Athens, Greece, pp. 171–185.

Arii, T., Kobayashi, A. and Hasegawa, K.,

2012, "Performance Analysis of Single-Propeller / Twin-Rudder System by Com-putational Fluid Dynamics", MARSIM 2012, Singapore.

Armaoglu, E., Eggers, R., Quadvlieg, F. and

van Coevorden, P., 2010, "On the Manoeuvring of Naval Surface Ships with Respect to a New STANAG", WARSHIP 2010, London, UK.

Arslan, T., Pettersen, B., Visscher, J.,

Muthanna, C. and Andersson, H., 2011, "A Comparative Study of PIV Experiments and Numerical Simulations of Flow Fields around Two Interacting Ships", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 31–37.

Avila, J. and Adamowski, J., 2011, "Ex-

perimental Evaluation of the Hydrodynamic Coefficients of a ROV through Morison’s Equation", Ocean Engineering, Vol. 38, No.17-18, pp. 2162–2170.

Beaulieu, C., Gharbi, S., Ouarda, T., Charron,

C. and Aissia, M., 2012, "Improved Model of Deep-Draft Ship Squat in Shallow Waterways Using Stepwise Regression Trees", Journal of Waterway, Port, Coastal, and Ocean Engineering, Vol. 138, No.2, pp. 115–121.

Béguin, L., Guillerm, P., Rousset, J., Weber, M., Lambert, S., Toularastel, J. and Da-voust, L., 2013, "An Experimental Investi-gation of Behavior and Hydrodynamics Forces Acting on a Ship in Shallow Water", Third international conference on ship manoeuvring in shallow and confined water, Ghent, Belgium, pp. 199–207.

Benedict, K., Köpnick, W., Baldauf, M.,

Gronarz, A., Friedhoff, B. and De Mello Petey, F., 2011, "Improved Simulation Mo-del and Methods for Determination of the Required Passing Distance between Ships in Restricted Fairways", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 39–50.

Berget, K., 2011, "A New Multi-Purpose Car-

riage in the MARINTEK Towing Tank", Marintek review no. 2. Retrieved from http://www.sintef.no/upload/MARINTEK/PDF-filer/Review/Review_2_2011_Web.pdf

Bhattacharyya, S., Rajesh, G. and Gupta, D.,

2012, "Fuzzy Autopilot for Ship Maneuvering", International Shipbuilding Progress, Vol. 58, No.2011, pp. 191–218.

Blendermann, W., Hellwig, K. and Schuckert,

E., 2011, "Wind Loads on a Passenger/car Ferry by CFD Computations and Wind Tunnel Tests", Ship Technology Research, Vol. 58, No.2, pp. 82–88.

Böttner, C., Uliczka, K., Carstens, D. and

Sponholz, J., 2013, "Free Sailing Manoeu-vres for Tuning of Analytical Shallow Water Models", Third international confe-rence on ship manoeuvring in shallow and confined water, Ghent, Belgium, pp. 177–183.

177177

Bowles, J., 2012, "Turning Characteristics and Capabilities of High-Speed Monohulls", The Third Chesapeake Power Boat Symposium , Annapolis, MD, USA.

Briggs, M., Kopp, P., Ankudinov, V. and Sil-

ver, A., 2013, "Comparison of Measured Ship Squat with Numerical and Empirical Methods", Journal of Ship Research, Vol. 57, No.2, pp. 73–85.

Broglia, R., Durante, D., Dubbioso, G. and Di

Mascio, A., 2011, "Turning Ability Characteristics Study of a Twin Screw Vessel by CFD", International Conference on Computational Methods in Marine Engineering, Lisbon, Portugal.

Bunnik, T. and Toxopeus, S., 2011, "Viscous

Flow Effects of Passing Ships in Ports", OMAE 2011, Rotterdam, The Netherlands.

Cai, C., Cai, X., Zhao, C. and Chen, M., 2012,

"Numerical Simulation on Ships Maneu-verability in Shallow Water", Journal of Chongqing University, Vol. 35, No.8, pp. 116–121.

Carreño, J., Mora, J. and Pérez, F., 2013, "Mathematical Model for Maneuverability of a Riverine Support Patrol Vessel with a Pump-Jet Propulsion System", Ocean Engineering, Vol. 63, pp. 96–104.

Carrica, P., Ismail, F., Hyman, M., Bhushan, S.

and Stern, F., 2013, "Turn and Zigzag Maneuvers of a Surface Combatant Using a URANS Approach with Dynamic Overset Grids", Journal of Marine Science and Technology, Vol. 18, No.2, pp. 166–181.

Chase, N., Michael, T. and Carrica, P., 2012, "Overset Simulation of a Submarine and Propeller in Towed , Self-Propelled and Maneuvering Conditions", 29th symposium on Naval Hydrodynamics, Gothenburg, Sweden.

Cheng, J., Zhang, Z., Jiang, F., Wang, X. and

Feng, D., 2013, "Numerical Simulation of the Planar Motion Mechanism Tests", Chinese Journal of Hydrodynamics, Vol. 28, No.4, pp. 460–464.

Chislett, M., 1996, "A Generalized Math

Model for Manoeuvring", Marsim 1996, Copenhagen, Denmark.

Coraddu, A., Dubbioso, G., Mauro, S. and

Viviani, M., 2013, "Analysis of Twin Screw Ships’ Asymmetric Propeller Beha-viour by Means of Free Running Model Tests", Ocean Engineering, Vol. 68, pp. 47–64.

Dand, I. and Ferguson, A., 1973, "Estimating

the Bow and Stern Sinkage of a Ship under Way in Shallow Water", Naval Architect, No.238.

De Barros, E. and Dantas, J., 2012, "Effect of a

Propeller Duct on AUV Maneuverability", Ocean Engineering, Vol. 42, pp. 61–70.

De Jong, M., Roelvink, D., Reijmerink, S. and

Breederveld, C., 2013, "Numerical Model-ling of Passing-Ship Effects in Complex Geometries and on Shallow Water", Smart Rivers 2013, Liège, Belgium and Maas-tricht, The Netherlands, pp. 95.1–95.7.

De Jong, P. and Keuning, J., 2005, "Tests with

a Six Degrees of Freedom Forced Oscillator to Investigate the Non-Linear Effects in Ship Motions", FAST 2005, St. Petersburg, Russia.

178178

De Jong, P., Renilson, M. and van Walree, F., 2013, "The Broaching of a Fast Rescue Craft in Following Seas", FAST 2013, Amsterdam, The Netherlands.

De Loor, A., van der Hout, A., Weiler, O. and

Kortlever, W., 2013, "The Use and Vali-dation of Openfoam to Determine the Late-ral and Longitudinal Forces Exerted on a Vessel in the Lock and in the Lock Approach", Third international conference on ship manoeuvring in shallow and confined water, Ghent, Belgium, pp. 157–165.

Delefortrie, G., Vantorre, M., Cappelle, J. and

Ides, S., 2012, "The Effect of Shipping Traffic on Moored Ships", 10th Inter-national Conference on Hydrodynamics, St. Petersburg, Russia.

Delefortrie, G., Vantorre, M., Eloot, K., Ver-

willigen, J. and Lataire, E., 2010, "Squat Prediction in Muddy Navigation Areas", Ocean Engineering, Vol. 37, No.16, pp. 1464–1476.

Delefortrie, G., Willems, M., Laforce, E., Van-

torre, M., De Mulder, T., De Regge, J. and Wong, J., 2008, "Tank Test of Vessel Entry and Exit for Third Set of Panama Locks", International Navigation Seminar following PIANC AGA 2008, Beijing, China, pp. 517–523.

Denehy, S., Duffy, J., Renilson, M. and Ran-

muthugala, D., 2012, "The Effect of Ber-thed Ship Size and Berth Occupancy on Berthed Ship - Passing Ship Interaction Forces and Moments", Hydronav 2012/ Manoeuvring 2012, Ilawa, Poland, pp. 70–78.

Di Mascio, A., Dubbioso, G., Notaro, C. and Viviani, M., 2011, "Investigation of Twin-Screw Navai Ships Maneuverabiiity Beha-vior", Journal of Ship Research, Vol. 55, No.4, pp. 221–248.

Do, K., 2010, "Practical Control of Under-

actuated Ships", Ocean Engineering, Vol. 37, No.13, pp. 1111–1119.

Drouet, A., Reliquet, G., Bardin, A., Jacquin,

E., Gentaz, L. and Alessandrini, B., 2011, "Simulation of Submarine Manoeuvring Using Navier-Stokes Solver", International Conference on Computational Methods in Marine Engineering , Lisbon, Portugal.

Dubbioso, G., Durante, D., Broglia, R. and Di

Mascio, A., 2012, "CFD Maneuvering Pre-diction of a Twin Screw Vessel with Dif-ferent Stern Appendages Configuration", 29th symposium on Naval Hydrodynamics, Gothenburg, Sweden.

Dubbioso, G. and Viviani, M., 2012, "Aspects

of Twin Screw Ships Semi-Empirical Maneuvering Models", Ocean Engineering, Vol. 48, pp. 69–80.

Duffy, J., Denehy, S., Ranmuthugala, D. and

Renilson, M., 2013, "The Effect of Berth Blockage on Berthed Ship - Passing Ship Interaction", Third international conference on ship manoeuvring in shallow and con-fined water, Ghent, Belgium, pp. 209–214.

Duffy, J. and Renilson, M., 2011, "The Impor-

tance of the Form of Time Domain Forces and Moments on Berthed Ship Interaction", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 107–116.

179179

Eloot, K. and Delefortrie, G., 2012, "Inland Navigation: Assessing the Manoeuvring Behaviour for Real-Time Simulation Purposes", MARSIM 2012, Singapore.

Eloot, K., Vantorre, M., Verwilligen, J., Bus,

D., Cleeren, R., Gheyle, E., … Detienne, R., 2013, "Lock Manoeuvres: Practical Cases in Europe and Panama", Third Inter-national Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 309–323.

Fang, C., Lin, C. and Tsai, J., 2012, "Sea Trial

Validation on the Manoeuvring Perfor-mance for the 8200 TEU Container Ship", MARSIM 2012, Singapore.

Fitriadhy, A. and Yasukawa, H., 2011, "Course

Stability of a Ship Towing System", Ship Technology Research, Vol. 58, No.1, pp. 4–23.

Fitriadhy, A. and Yasukawa, H., 2011, "Tur-

ning Ability of a Ship Towing System", Ship Technology Research, Vol. 58, No.2, pp. 112–124.

Fitriadhy, A., Yasukawa, H. and Masaki, Y.,

2011, "Slack Towline on Tow-Towed Ships Interaction during Maneuvering", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 127–135.

Fonfach, J., Sutulo, S. and Guedes Soares, C.,

2011, "Numerical Study of Ship-to-Ship Interaction Forces on the Basis of Various Flow Models", 2nd International Con-ference on Ship Manoeuvring in Shallow and Confined Water , Trondheim, Norway, pp. 137–146.

Fukui, Y., 2012, "A Study on Interaction Coefficients between Hull and Rudder in Maneuvering Model Using CFD", MARSIM 2012, Singapore.

Furukawa, Y., Nakiri, Y. and Kijima, K., 2011,

"Prediction of Linear Hydrodynamic Deri-vatives in Shallow Water", 2nd Internatio-nal Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 147–152.

Geerts, S., Vantorre, M., Eloot, K., Huijsmans,

R. and Fierens, N., 2011, "Interaction Forces in Tug Operations", 2nd Internatio-nal Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 153–163.

Gourlay, T., 2013, "Ship Squat in Non-Uni-

form Water Depth", Coasts and Ports 2013 , Sydney, Australia.

Greeley, D. and Willemann, S., 2012, "Surface

Ship Maneuvering Forces in Calm Water and Waves", 29th symposium on Naval Hydrodynamics, Gothenburg, Sweden.

Gronarz, A., 2011, "A New Approach in

Modelling the Interaction Forces", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 165–174.

Groves, N., Huang, T. and Chang, M., 1989,

"Geometric Characteristics of DARPA SUBOFF Models (DTRC Model Nos. 5470 and 5471)", Report No. DTRC/SHD-1298-01.

180180

Hasegawa, K., 2013, "River Utilization and Transportation for 21st Century : Expe-riences and Recommendations for a Sustai-nable Development", International Con-ference IDS2013 - Amazonia, Iquitos, Peru, pp. 18.1–18.12.

Hasegawa, K., Oh, K., Ahmed, Y. and Rigo, P.,

2013, "Ship Manoeuvring Behaviour in Crossing Current", Third international con-ference on ship manoeuvring in shallow and confined water, Ghent, Belgium, pp. 301–307.

Henn, R., 2013, "Real-Time Simulation of

Ships Manoeuvring in Locks", Third international conference on ship manoeu-vring in shallow and confined water, Ghent, Belgium, pp. 215–220.

Hirata, N., Ukita, H. and Yasukawa, H., 2012,

"Trim Effect on Maneuverability of a Training Ship “Toyoshio-Maru”", Journal of Japan Institute of Navigation, Vol. 126, pp. 81–89.

Hirata, N., Yasukawa, H. and Ukita, H., 2012,

"Maneuvering Fullscale Trial of a Ship with Azimuth Propellers to Capture the Trim Effect", MARSIM 2012, Singapore.

Hong, S., Nam, B., Kim, J. and Park, J., 2013,

"A Study on Towing Characteristics of a Transportation Barge in Waves", ISOPE 2013, Anchorage, Alaska, USA, Vol 9, pp. 777–783.

Huang, T., Liu, H., Groves, N., Forlini, T.,

Blanton, J. and Gowing, S., 1992, "Mea-surements of Flows over an Axisymmetric Body with Various Appendages in a Wind Tunnel: The DARPA SUBOFF Experimen-tal Program", 19th Symposium on Naval Hydrodynamics, Seoul, Korea, pp. 312–346.

Hwang, W., 2012, "On a Pragmatic Analytical 4-Quadrant Propeller-Rudder Model", MARSIM 2012, Singapore.

Hwang, W., Jakobsen, B., Barr, R., Ankudinov,

V., Fuller, N., Vest, L., … Landsburg, A., 2003, "An Exploratory Study to Charac-terize Ship Manoeuvring Performance at Slow Speed", MARSIM 2003, Kanazawa, Japan.

Ibaragi, H., Nakiri, Y. and Furukawa, Y., 2012,

"Experimental Study on Bank Suction", MARSIM 2012 , Singapore.

Ichinose, Y. and Furukawa, Y., 2011, "Funda-

mental Investigation on Estimation of Hy-drodynamic Forces Acting on a Ship in Ob-lique Motion by Using Three Dimensional Vortex Method", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 13, pp. 1–9.

Iseki, T. and Kawamura, D., 2011, "Experi-

mental Study on Ship-Handling Limits under Ship-Bank Interaction Forces", Jour-nal of Japan Institute of Navigation, Vol. 124, pp. 257–263.

Iseki, T. and Takagi, T., 2013, "Experimental

Study on Ship-Handling Limits under Ship-Bank Interaction Forces -Ⅱ- Effects of Pro-peller Slipstream", Journal of Japan Insti-tute of Navigation, Vol. 128, pp. 161–166.

Kang, D., Nagarajan, V., Gonno, Y., Uematsu,

Y., Hasegawa, K. and Shin, S., 2011, "Installing Single-Propeller Twin-Rudder System with Less Asymmetric Maneuve-ring Motions", Ocean Engineering, Vol. 38, No.10, pp. 1184–1196.

181181

Khanfir, S., Hasegawa, K., Nagarajan, V., Shouji, K. and Lee, S., 2011, "Manoeuvring Characteristics of Twin-Rudder Systems: Rudder-Hull Interaction Effect on the Manoeuvrability of Twin-Rudder Ships", Journal of Marine Science and Technology, Vol. 16, No.4, pp. 472–490.

Kim, S., Rhee, B. and Miller, R., 2012,

"Anatomy of Turbulent Flow around DARPA SUBOFF Body in a Turning Maneuver Using High-Fidelity RANS Computations", 29th symposium on Naval Hydrodynamics, Gothenburg, Sweden.

Kim, S. and Sung, Y., 2012, "Numerical Simu-

lations of Maneuvering and Dynamic Stabi-lity of a Containership in Waves", 11th In-ternational Conference on the Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 215–225.

Kimura, Y., Kobayashi, E., Tahara, Y. and

Koshimura, S., 2011, "A Study on Esti-mation of Hydrodynamic Forces Acting on a Ship Hull in Shallow Water by CFD", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 193–201.

Kobyliński, L., 2012, "Problems of Handling

Ships Equipped with Azimuthing Propulsion Devices", Hydronav 2012/Man-oeuvring 2012, Ilawa, Poland, pp. 168–177.

Kogishi, A., Correa da Silva, H., Brinati, H., de

Conti, M. and Domiciano, V., 2013, "Ex-perimental Evaluation of Submersible Drift Forces", PRADS 2013, Changwon, Korea.

Kortlever, W. and de Boer, G., 2013, "Manoeu-vring at Transit of the Existing North Lock and the Future Sea Lock at IJmuiden, The Netherlands; Pilot Procedures and Informa-tion from Scale Model Tests", Third Inter-national Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 147–156.

Kwon, C., Lee, S., Lee, Y., Yang, H., Song, G.,

Kim, J. and Seo, J., 2013, "Experimental Study on the Crabbing Ability of a Cruise Ship Equipped Three Bow Thrusters and Two Pods in Open and Shallow Water", PRADS 2013, Changwon, Korea, pp. 1112 –1119.

Lataire, E., Vantorre, M. and Delefortrie, G.,

2009, "Captive Model Testing for Ship-to-Ship Operations", MARSIM 2009 , Panama City, Panama.

Lataire, E., Vantorre, M. and Delefortrie, G.,

2012, "A Prediction Method for Squat in Restricted and Unrestricted Rectangular Fairways", Ocean Engineering, Vol. 55, pp. 71–80.

Lataire, E., Vantorre, M., Delefortrie, G. and

Candries, M., 2012, "Mathematical Model-ling of Forces Acting on Ships during Ligh-tering Operations", Ocean Engineering, Vol. 55, pp. 101–115.

Lataire, E., Vantorre, M. and Eloot, K., 2009,

"Systematic Model Tests on Ship-Bank Interaction Effects", International Confe-rence on Ship Manoeuvring in Shallow and Confined Water: Bank Effects, Antwerp, Belgium, pp. 9–22.

182182

Lee, S., Park, J., Rhee, K. P., Yoo, W. and Kim, S., 2011, "An Experimental Research on the Hydrodynamic Forces Acting on a Twin-Screw Ferry during Unberthing", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 223–231.

Leong, Z., Saad, K., Ranmuthugala, D. and

Duffy, J., 2013, "Investigation into the Hydrodynamic Interaction Effects on an AUV Operating Close to a Submarine", Pacific 2013: International Maritime Con-ference, Sydney, Australia.

Lesser, G., Jordan, D., River, C. and Pilots, B.,

2013, "Under-Keel Clearance at the Columbia River Bar", Coasts and Ports 2013, Sydney, Australia.

Lewandowski, E. M., 2004, "The Dynamics Of

Marine Craft – Maneuvering And Seakee-ping", Advanced Series on Ocean Engi-neering, Volume 22 World Scientific.

Li, Z., Hu, Y., Guo, C. and Wang, J., 2013,

"Research on Ship Squat Calculation Method for Ships Entering and Leaving the Shiplift Chamber", Smart Rivers 2013, Liège, Belgium and Maastricht, The Netherlands, pp. 162.1–162.4.

Lindberg, O., Glimberg, S., Bingham, H.,

Engsig-Karup, A. and Schjeldahl, P., 2013, "Real-Time Simulation of Ship-Structure and Ship-Ship Interaction", Third Inter-national Conference on Ship Manoeuvring in Shallow and Confined water, Ghent, Belgium, pp. 257–264.

Lloyd Register, 2006, “Rules and Regulations

for the Classification of Naval Ships,” Vol-ume 3, Part 1, Chapter 6.

Lou, P. and Zou, Z., 2012, "Numerical Cal-culation of the Viscous Hydrodynamic Forces on a KVLCC Hull Sailing along Channel Bank", OMAE 2012, Rio de Ja-neiro, Brazil.

Luo, W., Fu, B., Guedes Soares, C. and Zou,

Z., 2013, "Robust Control for Ship Course-Keeping Based on Support Vector Machi-nes, Particle Swarm Optimization and L2-Gain", OMAE 2013, Nantes, France.

Luo, W., Zou, Z. and Xiang, H., 2011, "Si-

mulation of Ship Manoeuvring in the Proxi-mity of a Pier by Using Support Vector Machines", OMAE 2011, Rotterdam, The Netherlands.

Maimun, A., Priyanto, A., Muhammad, A.,

Scully, C. and Awal, Z., 2011, "Manoeu-vring Prediction of Pusher Barge in Deep and Shallow Water", Ocean Engineering, Vol. 38, No.11-12, pp. 1291–1299.

MarCom Working Group 121, 2014, "Harbour

Approach Channels Design Guidelines." Brussels, Belgium, pp. 1–328.

MARINTEK, 2014, "Sea Trials and Model

Tests for Validation of Shiphandling Simulation Models". Retrieved from http://www.sintef.no/Projectweb/SimVal/

Matusiak, J. and Stigler, C., 2012 "Ship Roll

Motion in Irregular Waves during a Turning Circle Maneuver", 11th International Con-ference on the Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 291–297.

Mauro, S., 2013, "Experimental Study of the

Asymmetrical Propeller Loads Occurring in Unmanned Self-Propelled Free Running Surface Model Tests", AMT ’13: 3rd Inter-national Conference on Advanced Model Measurement Technology, Gdansk, Poland.

183183

Mauro, S., Dubbioso, G., Broglia, R. and Muscari, R., 2012, "Investigation of Asymmetrical Shaft Power Increase during Ship Maneuvers by Means of Model Tests and CFD", 29th symposium on Naval Hydrodynamics , Gothenburg, Sweden.

Milanov, E., Zlatev, Z., Chotukova, V. and

Stern, F., 2011, "Analysis of Inherent Cour-se Stability of a High-Speed Catamaran in Deep and Shallow Water", International Shipbuilding Progress, Vol. 58, pp. 83–96.

Miyazaki, H., Sawada, H., Tukada, Y. and

Kuroda, T., 2013, "Experimental and Nu-merical Study about the Static Heel Effect on the Turning Ability", ISOPE 2013, An-chorage, Alaska, USA, Vol 9, pp. 840–845.

Miyazaki, H., Ueno, M. and Tsukada, Y., 2011,

"Numerical Study about Effects of Stern Skeg on Course Stability", ISOPE 2011, Maui, Hawaii, Vol 8, pp. 964–969.

Mizuno, N., Kakami, H. and Okazaki, T., 2012,

"Parallel Simulation Based Predictive Con-trol Scheme With Application To Approa-ching Control For Automatic Berthing", MCMC 2012, Arenzano, Italy.

Mizuno, N. and Matsumoto, S., 2013, "Design

and Evaluation of Simple Ship ’s Automatic Maneuvering System Using Sliding Mode Controller", 9th IFAC conference on Control Applications in Marine Systems (CAMS). Part 1., Osaka, Japan, pp. 67–72.

Mousaviraad, S., Bhushan, S. and Stern, F.,

2012, "CFD Prediction of Free-Running SES/ACV Deep and Shallow Water Maneuvering in Calm Water and Waves", MARSIM 2012, Singapore.

Mousaviraad, S., Sadat-Hosseini, S., Carrica, P. and Stern, F., 2011, "URANS Studies of Ship-Ship Interactions in Calm Water and Waves for Replenishment and Overtaking Conditions", 2nd International Conference on Ship Manoeuvring in Shallow and Con-fined Water, Trondheim, Norway, pp. 253–264.

Mucha, P. and Moctar, O., 2013, "Ship-Bank

Interaction of a Large Tanker and Related Control Problems", OMAE 2013, Nantes, France.

Muto, H., Furukawa, Y., Ibaragi, H. and Habu,

I., 2011, "Prediction of Ship Manoeuvring Motion in Non-Uniform Flow Field", ISOPE 2011, Maui, Hawaii, Vol 8, pp. 951–957.

Nakano, T. and Hasegawa, K., 2012, "An

Attempt to Predict Manoeuvring Indices Using AIS Data for Automatic OD Data Acquisition", MCMC 2012, Arenzano, Italy, Vol 3.

Nakayama, Y., Yasukawa, H. and Watanabe,

J., 2011, "Dynamic Interaction between Tow and Towed Ships in Head Seas", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 265–272.

NATO ANEP-70, 2003, "Volume I – Guidance

for Naval Surface Ships Mission Oriented Manoeuvring Requirements. Volume II – Guidance for the Preparation of Onboard Manoeuvring Information.", NATO Allied Naval Engineering Publication 70.

NATO ANEP-78, 2007, "Naval Surface Ships

Mission Oriented Manoeuvring Require-ments. Specification and Verification Tem-plates.", NATO Allied Naval Engineering Publication 78.

184184

NATO ANEP-79, 2007, "Controllability and

Safety in a Seaway", NATO Allied Naval Engineering Publication 79.

Nemzer, A., Sergeev, V. and Polischuk, D.,

2012, "Investigation of Maneuverability under Wave and Wind Conditions", 10th International Conference on Hydrodyna-mics, St. Petersburg, Russia.

Obreja, D., Nabergoj, R., Crudu, L. and

Păcuraru-Popoiu, S., 2010, "Identification of Hydrodynamic Coefficients for Manoeu-vring Simulation Model of a Fishing Ves-sel", Ocean Engineering, Vol. 37, No.8-9, pp. 678–687.

Oh, K. and Hasegawa, K., 2013, "Low Speed

Ship Manoeuvrability: Mathematical Model and Its Simulation", OMAE 2013, Nantes, France.

Om, P., Satish, P., Nagarajan, V. and Suresh,

C., 2013, "Feasibility Study and Design of Shallow Draft Ore Carriers for Inland Waterways", International Conference IDS 2013 - Amazonia, Iquitos, Peru, pp. 08.1–08.18.

Ommani, B. and Faltinsen, O., 2013, "An In-

vestigation on the Calm-Water Linear Dynamic Stability of Semi-Displacement Vessels in Sway-Roll-Yaw", FAST 2013 , Amsterdam, The Netherlands.

Ommani, B., Faltinsen, O. and Lugni, C., 2012,

"Hydrodynamic Forces on a Semi-Displa-cement Vessel on a Straight Course with Drift Angle", 10th International Conference on Hydrodynamics, St. Petersburg, Russia.

Örnfelt, J., 2009, "Naval Mission and Task

Driven Manoeuvrability Requirements for Naval Ships", FAST 2009, Athens, Greece.

Otzen, J. and Simonsen, C., 2012, "Develop-

ment of a Mathematical Model of a High Speed Catamaran Ferry for Simulations in Hard Weather during Arrival Based on PMM and Seakeeping Model Tests", MARSIM 2012, Singapore.

Pan, Y., Zhang, H. and Zhou, Q., 2012,

"Numerical Prediction of Submarine Hydrodynamic Coefficients Using CFD Simulation", Journal of Hydrodynamics, Vol. 24, No.6, pp. 840–847.

Park, H., Kim, D., Lee, S. and Rhee, K., 2011,

"The Effect of Appendages on the Course Keeping Ability of a Semi-Displacement Round Bilge Ship", ISOPE 2011, Maui, Hawaii, Vol 8, pp. 958–963.

Pinkster, J., 2011, "Prediction of Hydrodyna-

mic Interaction Effects of Vessels in Ports", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water , Trondheim, Norway, pp. 279–287.

Pinkster, J. and Bhawsinka, K., 2013, "A Real-

Time Simulation Technique for Ship-Ship and Ship-Port Interactions", 28th Internatio-nal Workshop on Water Waves and Floa-ting Bodies, L’isle sur la Sorgue, France.

Polis, C., Ranmuthugala, D., Duffy, J. and

Renilson, M., 2013, "Enabling the Predic-tion of Manoeuvring Characteristics of a Submarine Operating Near the Free Sur-face", Pacific 2013: International Maritime Conference, Sydney, Australia.

Quadvlieg, F., 2013, "Mathematical Models for

the Prediction of Manoeuvres of Inland Ships; Does the Ship Fit in the River", Smart Rivers 2013 , Liège, Belgium and Maastricht, The Netherlands, pp. 187.1–187.9.

185185

Quadvlieg, F., Armaoglu, E., Eggers, R. and van Coevorden, P., 2010, "Prediction and Verification of the Maneuverability of Naval Surface Ships", SNAME 2010, Bellevue, USA.

Quadvlieg, F. and Brouwer, J., 2011,

"KVLCC2 Benchmark Data Including Un-certainty Analysis to Support Manoeuvring Predictions", International Conference on Computational Methods in Marine Engi-neering, Lisbon, Portugal.

Rajita Shenoi, R., Krishnankutty, P., Panneer

Selvam, R. and Kulshrestha, A., 2013, "Pre-diction of Manoeuvring Coeffcients of a Container Shup by Numerically Simulating HPMM Using RANSE Based Solver", Third international conference on ship manoeuvring in shallow and confined water, Ghent, Belgium, pp. 221–229.

Ray, A., 2010, "Analysis of Manoeuvring

Hydrodynamics of Underwater Vehicles" IIT Delhi.

Ray, A. and Sen, D., 2012, "Identification of

Submarine Hydrodynamic Coefficients from Sea Trials Using Extended Kalman Filter", 10th International Conference on Hydrodynamics, St. Petersburg, Russia.

Refsnes, J. and Sorensen, A., 2012, "Hydrody-

namic Modelling for Dynamic Positioning and Maneuvering Controller Design", OMAE 2012, Rio de Janeiro, Brazil.

Reichel, M., 2012, "Application of MMG Mo-

del Simulations to Twin-Propeller, Twin-Rudder Car-Passenger Ferry in Port Ma-noeuvring Conditions", MARSIM 2012, Singapore.

Ren, J., Zhang, X. and Sun, X., 2012, "Mathe-matical Modeling of Towage Operation for Interactive Tug Simulator", MARSIM 2012, Singapore.

Revestido, E. and Velasco, F., 2012, "Two-

Step Identification of Non-Linear Manoeuvring Models of Marine Vessels", Ocean Engineering, Vol. 53, pp. 72–82.

Rhee, K., Kim, D. and Kang, D., 2013, “A

Study on the Development of Korean Reg-ister (KR) Manoeuvrability Rules for Naval Ships,” 2013 Naval Ship Technology & Weapon Systems Seminar, Pusan, Korea.

Roddy, R., 1990, "Investigation of the Stability

and Control Characteristics of Several Configurations of the DARPA SUBOFF Model (DTRC Model 5470) from Captive-Model Experiments", Report No. DTRC/ SHD-1298-08.

ROM, 2000, "Recommendations for the Design

of the Maritime Configuration of Ports, Approach Channels and Harbour Basins", Madrid, Spain.

Saari, H. and Djemai, M., 2012, "Ship Motion

Control Using Multi-Controller Structure", Ocean Engineering, Vol. 55, pp. 184–190.

Sadat-Hosseini, H., Carrica, P., Stern, F.,

Umeda, N., Hashimoto, H., Yamamura, S. and Mastuda, A., 2011, "CFD, System-Based and EFD Study of Ship Dynamic In-stability Events: Surf-Riding, Periodic Mo-tion, and Broaching", Ocean Engineering, Vol. 38, No.1, pp. 88–110.

186186

Sadat-Hosseini, H., Chen, X., Kim, D., Mila-nov, E., Georgiev, S., Zlatev, Z. and Stern, F., 2013, "CFD and System-Based Predic-tion of Delft Catamaran Maneuvering and Course Stability in Calm Water", FAST 2013, Amsterdam, The Netherlands.

Sadat-Hosseini, H., Wu, P., Toda, Y., Carrica,

P. and Stern, F., 2011, "URANS Studies of Ship-Ship Interactions in Shallow Water", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 299–308.

Sakamoto, N., Carrica, P. and Stern, F., 2012,

"URANS Simulations of Static and Dyna-mic Maneuvering for Surface Combatant: Part 1. Verification and Validation for For-ces, Moment, and Hydrodynamic Derivati-ves", Journal of Marine Science and Tech-nology, Vol. 17, No.4, pp. 422–445.

Sakamoto, N., Carrica, P. and Stern, F., 2012,

"URANS Simulations of Static and Dyna-mic Maneuvering for Surface Combatant: Part 2. Analysis and Validation for Local Flow Characteristics", Journal of Marine Science and Technology, Vol. 17, No.4, pp. 446–468.

Sanada, Y., Tanimoto, K., Takagi, K., Gui, L.,

Toda, Y. and Stern, F., 2013, "Trajectories for ONR Tumblehome Maneuvering in Calm Water and Waves", Ocean Engineering, Vol. 72, pp. 45–65.

Sanada, Y., Tanimoto, K., Takagi, K., Sano,

M., Yeo, D., Gui, L., … Stern, F., 2012, "Trajectories and Local Flow Field Measurements around ONR Tumblehome in Maneuvering Motion", 29th symposium on Naval Hydrodynamics, Gothenburg, Sweden.

Sano, M., Yasukawa, H. and Hata, H., 2012, "Experimental Study on Ship Operation in Close Proximity to Bank Channel", MARSIM 2012, Singapore.

Sano, M., Yasukawa, H., Kitagawa, K. and

Yoshida, S., 2013, "Shallow Water Effect on the Hydrodynamic Interaction between Two Ships with Rudder in Close Pro-ximity", Third International Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 113–121.

Seo, M. and Kim, Y., 2011, "Numerical Ana-

lysis on Ship Maneuvering Coupled with Ship Motion in Waves", Ocean Engi-neering, Vol. 38, No.17–18, pp. 1934–1945.

Shen, Y. and Hughes, M., 2012, "Effective

Inflow Velocity for Rudder Calculations", International Shipbuilding Progress, Vol. 59, pp. 107–127.

Shin, S., Ahn, K., Sung, Y. and Oh, S., 2012,

"A Study on Effect of the Self-Propulsion Points in Pmm Tests for KVLCC’s Ma-noeuvrability", MARSIM 2012, Singapore.

Shin, S., Lee, T. and Ahn, K., 2013, "A Study

on the Numerical Analysis of Maneuvera-bility at Hull Form Design Stage", PRADS 2013, Changwon, Korea, pp. 1106–1111.

Silva, D., 2012, "CFD Virtual Testing for Re-

sistance , Wind and Current Loads on a Supply Boat", OMAE 2012, Rio de Janeiro, Brazil.

Simonsen, C., Nielsen, C., Otzen, J. and Ag-

drup, K., 2011, "CFD Based Prediction of Ship-Ship Interaction Forces on a Tug beside a Tanker", 2nd International Confe-rence on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 329–338.

187187

Simonsen, C., Otzen, J., Klimt, C., Larsen, N. L. and Stern, F., 2012, "Maneuvering Pre-dictions in the Early Design Phase Using CFD Generated PMM Data", 29th sympo-sium on Naval Hydrodynamics, Gothen-burg, Sweden.

Skejic, R., 2008, "Maneuvering and Seakee-

ping of a Single Ship and Two Ships in Interaction" Norwegian University of Science and Technology.

Skejic, R., Breivik, M. and Berg, T., 2011, "In-

vestigating Ship Maneuvers around a Floa-ting Structure under the Influence of a Uni-form Current in Deep and Calm Water", 2nd International Conference on Ship Man-oeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 339–350.

Skejic, R. and Faltinsen, O., 2013, "Maneu-

vering Behavior of Ships in Irregular Waves", OMAE 2013, Nantes, France.

Skejic, R., Kirimoto, K., Berg, T. and Peder-

sen, E., 2012, "Maneuvering Performance of Ships in Calm Water with Variable Finite Water", MARSIM 2012, Singapore.

Song, G., Kim, H., Park, H. and Seo, J., 2013,

"The Investigation for Interaction Pheno-menon of Azimuth Thruster on Ship", PRADS 2013 , Changwon, Korea, pp. 823–828.

Spitzer, D. and Soehngen, B., 2013, "On the

Longitudinal Dynamics of Ship Entry and Exit at Locks", Smart Rivers 2013, Liège, Belgium and Maastricht, The Netherlands, pp. 44.1–44.13.

Stern, F., Agdrup, K., Kim, S., Cura Hoch-baum, A. C., Rhee, K., Quadvlieg, F., … Gorski, J., 2011, "Experience from SIM-MAN 2008 — The First Workshop on Ve-rification and Validation of Ship Maneu-vering Simulation Methods", Journal of Ship Research, Vol. 55, No.2, pp. 135–147.

Sugisawa, M. and Kobayashi, H., 2012,

"Estimation on Ship Maneuverability Based on the Correlation between Hydrodynamic Force and Moment", MARSIM 2012, Singapore.

Sun, M., Wang, Y. and Yang, Q., 2012,

"Analysis of the Reynolds Number Influence on Hydrodynamic Coefficients in Numerical Simulation of Submarine Maneuverability", Journal of Harbin Engi-neering University, Vol. 33, No.11, pp. 1334–1340.

Sung, Y., Lee, H., Lee, T. and Kim, S., 2012,

"Captive Model Test and Numerical Simu-lation on the Manoeuvring Forces in Wa-ves", 11th International Conference on the Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 865–876.

Sutulo, S., Guedes Soares, C. and Otzen, J.,

2012, "Validation of Potential-Flow Esti-mation of Interaction Forces", Journal of Ship Research, Vol. 56, No.3, pp. 129–145.

Thorenz, C., 2013, "Activities of PIANC

Working Group 155: Ship Behaviour in Locks and Lock Approaches", Third Inter-national Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 99–102.

188188

Thulin, S., 1974, "Discussion of “Large Ampli-tude PMM Tests and Manoeuvring Predic-tion for a Mariner Class Vessel”", 10th ONR Symposium on Naval Hydrodyna-mics, Boston, USA.

Toxopeus, S., 2011, "Practical Application of

Viscous-Flow Calculation for the Simulation of Manoeuvring Ships" Delft University of Technology.

Toxopeus, S., 2011, "Viscous-Flow Calcula-

tions for KVLCC2 in Deep and Shallow Water", International Conference on Com-putational Methods in Marine Engineering, Lisbon, Portugal.

Toxopeus, S., Atsavapranee, P., Wolf, E.,

Daum, S., Pattenden, R., Widjaja, R., … Gerber, A., 2012, "Collaborative CFD Exercise for a Submarine in a Steady Turn", OMAE 2012, Rio de Janeiro, Brazil.

Toxopeus, S., Simonsen, C., Guilmineau, E.,

Visonneau, M., Xing, T. and Stern, F., 2013, "Investigation of Water Depth and Basin Wall Effects on KVLCC2 in Manoeuvring Motion Using Viscous-Flow Calculations", Journal of Marine Science and Technology, Vol. 18, No.4, pp. 471–496.

Toxopeus, S., Stroo, K. and Muller, B., 2013,

"Optimisation of Resistance and Towed Stability of an Offshore Going Barge With CFD", OMAE 2013, Nantes, France.

Tran, V. and Im, N., 2012, "A Study on Ship

Automatic Berthing with Assistance of Auxiliary Devices", International Journal of Naval Architecture and Ocean Engineering, Vol. 4, No.3, pp. 199–210.

Tsukada, Y., Ueno, M., Miyazaki, H. and Takimoto, T., 2013, "An Auxiliary Thruster for Free-Running Model Ship Test", OMAE 2013, Nantes, France.

Ueno, M. and Tsukada, Y., 2013, "Numerical

Study on Rudder Effectiveness Correction of a Free-Running Model Ship", PRADS 2013, Changwon, Korea, pp. 1120–1127.

Ueno, M., Tsukada, Y. and Sawada, H., 2011,

"A Prototype of Submersible Surface Ship and Its Hydrodynamic Characteristics", Ocean Engineering, Vol. 38, No.14-15, pp. 1686–1695.

Uliczka, K., Böttner, C. and Carstens, D., 2013,

"Head-on Traffic at the Approach Channel to the Port of Hamburg", Third Inter-national Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 185–190.

Van der Molen, W., Moes, J., Swiegers, P. and

Vantorre, M., 2011, "Calculation of Forces on Moored Ships due to Passing Ships", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 369–374.

Vantorre, M. and Delefortrie, G., 2013,

"Behaviour of Ships Approaching and Leaving Locks: Open Model Test Data for Validation Purposes", Third International Conference on Ship Manoeuvring in Shal-low and Confined water, Ghent, Belgium, pp. 337–352.

Varyani, K. and Krishnankutty, P., 2009,

"Stopping Manoeuvre of High Speed Vessels Fitted with Screw and Waterjet Propulsion", Journal of Marine Engineering and Technology, Vol. 8, No.1, pp. 11–19.

189189

Vaz, G., Toxopeus, S. and Holmes, S., 2010, "Calculation of Manoeuvring Forces on Submarines Using Two Viscous-Flow Solvers", OMAE 2010, Shanghai, China.

Verdugo, I., Iribarren, J., Atienza, R., Cal, C.,

Pecharroman, L. and Trejo, I., 2013, "Passing Ship Interaction Study in Altamira Port (Mexico)", Third International Conference on Ship Manoeuvring in Shal-low and Confined water, Ghent, Belgium, pp. 191–198.

Vergote, T., Eloot, K., Vantorre, M. and

Verwilligen, J., 2013, "Hydrodynamics of a Ship While Entering a Lock", Third International Conference on Ship Manoeu-vring in Shallow and Confined Water, Ghent, Belgium, pp. 281–289.

Verwilligen, J., Maes, E. and Eloot, K., 2013,

"The Seine-Scheldt Project: Nautical Accessibility of a New Lock in Harelbeke", Third International Conference on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 291–299.

Verwilligen, J., Richter, J., Reddy, D., Van-

torre, M. and Eloot, K., 2012, "Analysis of Full Ship Types in High-Blockage Lock Configurations", MARSIM 2012, Sin-gapore.

Viallon, M., Sutulo, S. and Guedes Soares, C.,

2012, "On the Order of Polynomial Regres-sion Models for Manoeuvring Forces", MCMC 2012, Arenzano, Italy.

Vogt, M., Cura Hochbaum, A., Schneider, M.

and Dautel, J., 2007, "Report on Model Tests. Virtue Deliverable D3.1.3", Ham-burg, Germany.

Vrijburcht, A., 1988, "Calculations of Wave Height and Ship Speed When Entering a Lock", Delft Hydraulics Publication, No. 391.

Wang, H. and Zou, Z., 2013, "Numerical Study

on Hydrodynamic Behaviour of Ships Sailing in Locks", Third International Conference on Ship Manoeuvring in Shal-low and Confined Water, Ghent, Belgium, pp. 129–134.

Wang, X., Zou, Z. and Xu, F., 2013, "Model-

ling of Ship Manoeuvring Motion in 4 Deg-rees of Freedom Based on Support Vector Machines", OMAE 2013, Nantes, France.

Watai, R., Rudderi, F., Tannuri, E. and Weiss,

J., 2013, "Evaluation of Empirical and Nu-merical Methods on the Prediction of Hy-drodynamic Loads Involved in the Passing Ship Problem", Third International Confe-rence on Ship Manoeuvring in Shallow and Confined Water, Ghent, Belgium, pp. 167–175.

Wichers, J., 1988, "A Simulation Model for a

Single Point Moored Tanker", MARIN Publication, No. 797.

Woo, J. and Kim, N., 2013, "System Identifica-

tion of an Autonomous Surface Vehicle Using Neural Networks", ISOPE 2013, An-chorage, Alaska, USA, Vol 9, pp. 402–405.

Woodward, M., Atlar, M. and Clarke, D., 2009,

"Application of the IMO Maneuvering Criteria for Pod-Driven Ships", Journal of Ship Research, Vol. 53, No.2, pp. 106–120.

190190

Woodward, M., 2013, Propagation of experimental uncertainty from force measurements into manoeuvring derivatives. AMT ’13: 3rd Inter-national Conference on Advanced Model Measurement Technology, Gdansk, Poland.

Wu, W., Spyrou, K. and McCue, L., 2010,

"Improved Prediction of the Threshold of Surf-Riding of a Ship in Steep Following Seas", Ocean Engineering, Vol. 37, No.13, pp. 1103–1110.

Xiang, X. and Faltinsen, O., 2011, "Maneuve-

ring of Two Interacting Ships in Calm Water", Marine systems & Ocean Techno-logy, Vol. 6, No.2, pp. 65–73.

Xiang, X., Skejic, R., Faltinsen, O. and Berg,

T., 2011, "Hydrodynamic Interaction Loads between Two Ships during Lightering Ope-ration in Calm Water", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway.

Xing, T., Bhushan, S. and Stern, F., 2012,

"Vortical and Turbulent Structures for KVLCC2 at Drift Angle 0, 12, and 30 Degrees", Ocean Engineering, Vol. 55, pp. 23–43.

Xu, F., Zou, Z. and Song, X., 2011, "Parame-

tric Identification of AUV ’ S Maneuvering Motion Based on Support Vector Machines", Journal of Ship Mechanics, Vol. 15, No.9, pp. 981–987.

Xu, F., Zou, Z., Yin, J. and Cao, J., 2013,

"Identification Modeling of Underwater Vehicles’ Nonlinear Dynamics Based on Support Vector Machines", Ocean Engin-eering, Vol. 67, pp. 68–76.

Xu, Y., Sun, Y., Wei, Y., Guan, H., Liu, M. and Cai, W., 2012, "Study on Ship-Ship Hydrodynamic Interaction by ANN Optimization", MARSIM 2012, Singapore.

Xue, Y., Clelland, D., Lee, B. S. and Han, D.,

2011, "Automatic Simulation of Ship Navigation", Ocean Engineering, Vol. 38, No.17-18, pp. 2290–2305.

Yang, H. and Wada, Y., 2012, "Theoretical and

Experimental Study on the Towing Stability of the Large Offshore Floater System under the Various Environmental Forces", MAR-SIM 2012, Singapore.

Yang, H., Wu, B., Miao, Q., Xiang, X., Berg,

T. and Kuang X., 2011, "Study of the Effects of Unsteady Ship to Ship Interaction by CFD Method", 2nd International Con-ference on Ship Manoeuvring in Shallow and Confined Water, Trondheim, Norway, pp. 393–398.

Yao, J., Zou, Z. and Wang, H., 2011,

"Numerical Study on Bank Effects for a Ship Sailing in Shallow Channel", Journal of Shanghai Jiaotong University (Science), Vol. 16, No.1, pp. 91–96.

Yasukawa, H., 2006, “6-DOF Motion Simula-tions of a Turning Ship in Regular Waves”, MARSIM 2006, Terschelling, The Nether-lands.

Yasukawa, H. and Hirata, N., 2013,

"Maneuverability and Hydrodynamic Derivatives of Ships Traveling in Heeled Condition", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 17, pp. 19–29.

191191

Yasukawa, H., Hirata, N., Ikezoe, S. and Hirata, Y., 2012, "Experimental Study on Ship Performaces of a Catamaran with Asymmetric Demi-Hulls", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 15, pp. 101–110.

Yasukawa, H., Hirata, N., Tanaka, S. and Ito,

S., 2011, "Hydrodynamic Force Charac-teristics on Maneuvering of a Twin-Propeller Twin-Rudder Ship with Bow Thruster", Journal of Japan Institute of Navigation, Vol. 125, pp. 209–219.

Yasukawa, H., Hirata, N., Yokoo, R. and

Fitriadhy, A., 2012, "Slack Towline of Tow and Towed Ships during Turning", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 16, pp. 41–48.

Yasukawa, H., Hirono, T., Nakayama, Y. and

Koh, K., 2012, "Course Stability and Yaw Motion of a Ship in Steady Wind", Journal of Marine Science and Technology, Vol. 17, No.3, pp. 291–304.

Yasukawa, H., Sano, M. and Amii, H., 2013,

"Wind Effect on Directional Stability of a Ship Moving in a Channel", Third International Conference on Ship Manoeu-vring in Shallow and Confined water, Ghent, Belgium, pp. 83–91.

Yasukawa, H. and Yoshida, S., 2011,

"Hydrodynamic Interaction of Two Thin Ships with Rudder in Close Proximity", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 399–406.

Yasukawa, H. and Yoshimura, Y., 2013,

"Investigation of Roll-Coupling Effect on Ship Maneuverability", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 17, pp. 54–64.

Yeo, D., Yun, K., Kim, Y. and Kim, S., 2013, "Benchmark HPMM Tests for KCS in Shallow Water", Third International Con-ference on Ship Manoeuvring in Shallow and Confined Water , Ghent, Belgium, pp. 249–255.

Yin, J., Zou, Z. and Xu, F., 2013, "On-Line

Prediction of Ship Roll Motion during Maneuvering Using Sequential Learning RBF Neuralnetworks", Ocean Engineering, Vol. 61, pp. 139–147.

Yoon, H. and Kang, S., 2013, "Experimental

Investigation on the Depth Effect of Hydrodynamic Coefficients Obtained by PMM Test in Square Tank", Third International Conference on Ship Manoeu-vring in Shallow and Confined Water, Ghent, Belgium, pp. 209–214.

Yoon, H. and Kim, Y., 2012, "Coupled

Dynamic Simulation of a Tug-Towline-Towed Barge Based on the Multiple Element Model of Towline", Journal of Korean navigation and port research, Vol. 36, No.9, pp. 707–714.

Yoshimura, Y., 2011, "Effect of Roll Motion

on Ship Manoeuvrability by a Rudder to Yaw Response Equation", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 13, pp. 11–18.

Yoshimura, Y. and Masumoto, Y., 2012,

"Hydrodynamic Force Database with Medium High Speed Merchant Ships Including Fishing Vessels and Investigation into a Manoeuvring Prediction Method", Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 14, pp. 63–73.

192192

Yoshimura, Y., Nakao, I., Kanamoto, M. and Nakamura, M., 2012, "Ship Manoeuvrabili-ty in Ballast Condition (Open Water Cha-racteristics of Rudder)", Journal of Japan Institute of Navigation, Vol. 126, pp. 99–104.

Yuba, D. and Tannuri, E., 2013, "Analysis of

Pusher-Barge System with Different Maneuvering and Propulsion Devices", OMAE 2013, Nantes, France.

Zaghi, S., Dubbioso, G. and Broglia, R., 2012,

"Numerical Simulation of Submarine Rising Maneuver", 10th International Con-ference on Hydrodynamics, St. Petersburg, Russia.

Zhan, D. and Molyneux, D., 2012, "3-Dimen-

sional Numerical Simulation of Ship Mo-tion in Pack Ice", OMAE 2012, Rio de Janeiro, Brazil.

Zhang, C. and Zou, Z., 2011, "Numerical

Investigation on Ship-Ship Hydrodynamic Interaction in Restricted Waters", 2nd International Conference on Ship Manoeu-vring in Shallow and Confined Water, Trondheim, Norway, pp. 407–412.

Zhang, J., Maxwell, J., Gerber, A., Holloway,

A. and Watt, G., 2013, "Simulation of the Flow over Axisymmetric Submarine Hulls in Steady Turning", Ocean Engineering, Vol. 57, pp. 180–196.

Zhang, X. and Zou, Z., 2011, "Application of

Wavelet Denoising in the Modeling of Ship Manoeuvring Motion", Journal of Ship Mechanics, Vol. 15, No.6, pp. 616–622.

Zhang, X. and Zou, Z., 2011, "Identification of Abkowitz Model for Ship Manoeuvring Motion Using Ɛ-Support Vector Regres-sion", Journal of Hydrodynamics, Vol. 23, No.3, pp. 353–360.

Zhang, X. and Zou, Z., 2012, "Black-Box

Modeling of Ship Manoeuvring Motion Based on Feed-Forward Neural Network with Chebyshev Orthogonal Basis Function", Journal of Marine Science and Technology, Vol. 18, No.1, pp. 42–49.

Zhang, X. and Zou, Z., 2013, "Estimation of

the Hydrodynamic Coefficients from Captive Model Test Results by Using Support Vector Machines", Ocean Engi-neering, Vol. 73, pp. 25–31.

Zotti, I., 2013, "River Transport by Barges :

Resistance and Directional Stability Problems", International Conference IDS 2013 - Amazonia , Iquitos, Peru, pp. 15.1–15.20.

Zou, L. and Larsson, L., 2012, "Confined

Water Effects on the Viscous Flow around a Tanker with Propeller and Rudder", 29th symposium on Naval Hydrodynamics, Got-henburg, Sweden.

Zou, L. and Larsson, L., 2012, "Investigation

of Ship-to-Ship Interaction during a Lightering Operation in Shallow Water Using a RANS Solver", Marsim 2012, Singapore.

Zou, L., Larsson, L., Delefortrie, G. and

Lataire, E., 2011, "CFD Prediction and Validation of Ship-Bank Interaction in a Canal", 2nd International Conference on Ship Manoeuvring in Shallow and Confined Water , Trondheim, Norway, pp. 413–422.

193193

Zubova, A. and Nikushchenko, D., 2013, "Ship to Ship Interaction Investigations with the Use of CFD Methods", Third International Conference on Ship Manoeuvring in Shal-low and Confined Water, Ghent, Belgium, pp. 213–236.

194194

Seakeeping Committee

Final Report and Recommendations to the 27th ITTC

1. GENERAL 1.1. Membership and meetings

The Committee appointed by the 26th ITTC consisted of the following members :

• Yonghwan Kim (Chairman), Seoul Na-tional University, Korea

• Dan Hayden (Secretary), Naval Surface Wafare Center, West Bethesda, USA

• Dariusz Fathi, Norwegian Marine Technology Research Institute (MARINTEK), Trondheim, Norway

• Greg Hermanski, Institute for Ocean Technology, St. John’s, Canada

• Dominic Hudson, University of South-ampton, United Kingdom

• Pepijn de Jong, Delft University of Technology, The Netherlands

• Katsuji Tanizawa, National Maritime Research Institute, Tokyo, Japan

• Giles Thomas, Australian Maritime College, University of Tasmania, Tas-mania, Australia

• Wu Chengshen, China Ship Scientific Research Center, Wuxi, China (Re-placed Dr. Quanming Miao in 2012)

Four committee meetings were held at:

• University of Southampton, Southamp-ton, United Kingdom, January 2012

• National Maritime Research Institute, Tokyo, November 2012.

• David Taylor Model Basin, West Be-thesda, USA, July 2013

• Delft University of Technology, Delft, Netherlands, February 2014

In addition, a joint ISSC/ITTC workshop on

uncertainty modelling for ships and offshore structures was held in Rostock, Germany in September 2012.

1.2. Terms of Reference Given by the 26th ITTC

The list of tasks recommended by the 26th

ITTC was as follows:

1. Update the state-of-the-art for predicting the behaviour of ships in waves emphasis-ing developments since the 2011 ITTC Conference. The committee report should include sections on: a. the potential impact of new technologi-

cal developments on the ITTC b. new experiment techniques and ex-

trapolation methods, c. new benchmark data

195195

d. the practical applications of computa-tional methods to seakeeping predic-tions and scaling.

e. the need for R&D for improving meth-ods of model experiments, numerical modelling and full- scale measure-ments.

2. Review ITTC Recommended Proce-dures relevant to seakeeping and a. Identify any requirements for changes

in the light of current practice, and, if approved by the Advisory Council, up-date them.

b. Identify the need for new procedures and outline the purpose and content of these.

c. Introduce a definition of slamming.

3. Liaise with ISSC, the Ocean Engineering Committee, The Stability in Waves Com-mittee and the Specialist Committee on Performance of Ships in Service.

4. Update existing ITTC Recommended Pro-

cedure 7.5-02-07-02.5, Verification and Validation of Linear and Weakly Non-Linear Seakeeping Codes, to reflect the outcomes of the Verification and Validation workshop held in 2010.

5. Investigate methodology for Verification and Validation of fully non-linear seakeep-ing viscous flow codes.

6. Develop a guideline for the verification andoutline further developments required for validation of hydroelastic seakeeping codes.

7. Jointly organize and participate in the joint ISSC/ITTC workshop on uncertainty in measurement and prediction of wave loads and responses.

8. Establish a numerical and experimental process for estimating fw, in the EEDI cal-culation. Liaise with the Specialist Com-mittee on Performance of Ships in Service.

9. Develop a unified method for sloshing ex-periments drawing on the methods devel-oped by the classification societies. Identify benchmark data for sloshing in LNG car-riers.

10. Review and update the Procedure 7.5-02-05-04, Seakeeping Tests for High Speed Marine Vehicles.

2. REVIEW OF STATE-OF-THE-ART

2.1. New Experimental Facilities

2.1.1. Actual Sea Model Basin, National Maritime Research Institute

The Actual Sea Model Basin (Figure 1) is a

very advanced indoor facility for the simulation of the actual sea environment, including wind and waves, constructed at the National Mari-time Research Institute and completed at the end of August 2010. The length, width and depth of the basin are 80m, 40m and 4.5m, respectively. A total of 382 segmented flap-type absorbing wave makers are installed on all peripheries of the basin. By numerical control of individual segments, realistic wave field of the actual seas can be reproduced in the basin. For model tests, a three degree of freedom tow-ing carriage is available. The main carriage, which has a rail span of 41m, travels up to 3.5m/s, and the sub-carriage installed below the main-carriage runs up to 3.0m/s and is equipped with a turntable. In addition to multi-functional towing capability, auto-tracking function is available for free running tests in waves. For wind generation, removable blow-

196196

ers are available and a fluctuating wind up to 10m/s can be generated. The basin has a central control system of the wave makers, towing carriages and the wind generators. All func-tions of this basin are controlled synchronously. As a result, a high level of accuracy and repro-ducibility are achieved.

The Actual Sea Model Basin is a rectangu-

lar tank with rounded corners. Dimensions of the basin and its trimming tank are given in Table 1 and Table 2. For the installation about 2 meters of space is required at the backside of the flap. As a result, size of the water surface is about 76m x 36m. The four corners radius of curvature is 7.70m.

The Actual Sea Model Basin has 382 flap type absorbing wavemakers along the entire periphery except in front of the trim tank. The flap boards of the unit are connected to neigh-bors by watertight fan-like connection plates to avoid discontinuity. Each unit is numerically controlled both for generation and absorption and the entire water surface can be used for uniform wave field even in the case of short crested irregular wave generation.

The Actual Sea Model Basin has a X-Y-φ towing carriage. Main carriage travels the lon-gitudinal X direction, sub-carriage installed blow the main carriage travels the transverse Y direction and the turntable installed on the sub-carriage rotate φ direction around vertical axis.

Figure 1. New actual sea model basin at NMRI

Table 1. Dimension of Actual Sea Model Basin

Length Between Wall 80.0 m Water Surface 76.2 m

Width Between Wall 40.0 m Water Surface 36.0 m

Depth --- 4.95 m Water Depth Standard 4.50 m

Table 2. Dimension of Trimming Tank.

Length 6.0 m Width 1.2 m Depth 0.95 m

Water Depth 0.65 m

197197

2.1.2. Seoul National University Sloshing Facility

Recently Seoul National University (SNU)

installed three hexapod motion platforms with different payloads: 1.5 tonne, 5 tonne and 14 tonne (Figure 2). Each platform has six linear motors of different capacities, and all platforms are capable of simulating the 6-DOF motions of ships in a seaway. The small and midsize platforms of 1.5 tonne and 5 tonne capacity were installed in 2009, but the large platform of 14 tonne capacity including mount base was installed in 2012 and upgraded twice in 2013. The large platform height is 4.0m at rest condi-tion, 4.9m in stand-by condition, and about 5.7m in maximum heave motion. This platform consumes 140kW in normal/average excitation condition and 270kW in peak excitation. The detailed kinematic performance is summarized in the Table 3.

The small platform of 1.5 tonne payload is

suitable for the 3D model tests of about 1/70 scale, and the midsize platform of 5 tonne pay-load can be used for the 3D model tests of 1/60~1/40 scale. The large platform of 14 tonne payload can be used for the 3D model of 1/40~1/20 scale, but the experiment becomes more expensive as the size of model increases. Figure 3 shows the relative scale of the three motion platforms.

The facility at SNU is the world’s largest sloshing experimental facility, with 500 dy-namic, high quality pressure sensors, associated DAQ system and about 160TB storage for data acquisition and storage. 2D and 3D PIV sys-tems are available in this facility. The heavy-gas test using SF6 and N2 is also carried out in this test facility.

(a) 1.5 tonne platform (b) 5 tonne platform (c) 14 tonne platform

Figure 2. Three hexapod motion platforms in SNU

198198

Table 3. Performance of 14 tonne Hexapod Platform

Displacement Speed Acceleration @1500 rpm @2000 rpm Surge ±144 cm 155 cm/s 200 cm/s > 0.9G Sway ±138 cm 138 cm/s 180 cm/s > 0.9G Heave ±84 cm 84 cm/s 110 cm/s > 0.9G Roll ±33° 34°/s 45 °/s > 250°/s

2

Pitch ±33° 37°/s 49 °/s > 250°/s2

Yaw ±33° 56°/s 74 °/s > 250°/s2

Figure 3. Scaled model tanks on the large and midsize platforms and a 2D tank on the small plat-

form

199199

Figure 4. New Wavemaking Facility in Maneuvering and Seakeeping Basin (MASK), CDNSWC

2.1.3. New Wavemaker – MASK Basin,

Naval Surface Warfare Center

A wavemaker replacement for the Maneu-vering and Seakeeping Basin, of Carderock Division, Naval Surface Warfare Center was publicly completed in December 2013 (Figure 4). The wavemaker machine consists of 216 paddles at a 0.658 m spacing from centreline to centreline. There are 108 paddles along the long wall of the tank, 60 paddles in the curve, and 48 paddles along the short wall. The pad-dles have a hinge depth of 2.5 meters. The wavemaker is of a dry-back design with gusset material connecting each paddle. The paddles integrate a force feedback design where forces are measured at the lower hydrostatic assist location and at the upper motion control at-tachment. The components of the wavemaker are illustrated in Figure 5. The new wavemaker is capable of regular and irregular seas, multi-component long and short crested seaways, and other superposition events as required. No changes in the beaches along the opposite sides of the basin were required.

Figure 5. Rendering of 4 MASK Wavemaker

Paddles Showing Components 2.1.4. New Wavemaker – Depressurized

Wave Basin, MARIN

A wavemaker replacement for the depres-surized basin of the MARIN facility was pub-licly completed in March 2012. In tandem with the wavemaker upgrade, several new sub car-riages were built due to the improvement of

200200

having wavemakers installed in the depressur-ized wave basin as shown in Figure 6. The wavemaker installation includes 24 dry-back paddles with a 2.5m hinge depth and a 0.6m width along the short wall; and 200 dry-back paddles with a 1.8m hinge depth and 0.6m width along the long edge. The junction of the short and long walls is shown in Figure 7. Both banks of paddles were similar in design concept to the components shown in Figure 5 for the MASK basin as both designs were pro-vided by the same company. Deployable beaches were installed as required on opposite walls since wavemakers had not previously been installed. The wavemakers had to be de-signed and built to satisfy the unique chal-lenges of a depressurized facility.

The wavemaking capability in a depressur-ized basin will allow for the investigation of air-water phenomena not previously possible. These areas of investigation could include damaged stability, cavitation, designed air cavi-ties, and air cavities during slamming and wave impacts.

Figure 6. New wavemakers and sub carriages

at MARIN depressurized basin

Figure 7. New wavemakers at junction of short

and long walls of depressurized basin 2.1.5. New Towing Tank, University of Sou-

thampton A new towing tank is under construction at

the University of Southampton, UK, due for completion in September 2014. The new facil-ity is 138m long with a breadth of 6m and a depth of 3.5m. The tank is equipped with a cable-driven carriage having a maximum speed of 12 m/s. The Wolfson Unit for Marine Tech-nology and Industrial Aerodynamics, part of the University of Southampton, and the Uni-versity’s Ship Science degree programme will be the primary users of the facility. The tank is designed to allow all types of hydromechanic experiments for the shipping, offshore and yacht and small craft industries. The towing tank will be used for a mix of activities includ-ing education, research and consultancy. Seakeeping experiments will be performed with hinged-flap wavemakers, which are capa-ble of generating regular and irregular waves as well as transient breaking and focused waves. The maximum wave height for regular waves is 0.5m and waves with a period between 0.8s and 3.5s can be generated. Both standard and user-defined sea-states can be used. The tank is also to be equipped with a motion-tracking camera system and PIV for fluid flow diagnos-tics. A small coastal wave basin (5m x 5m), narrow flume with wavemakers and three wind

201201

tunnels are also included in the purpose-built fluid dynamics laboratory building.

2.2. Development in Experimental, Analy-tical, and Numerical Techniques

2.2.1. Experimental Techniques

This section contains reviews of work con-

cerning developments in experimental tech-niques, which include model scale and full scale experiments.

2.2.1.1. Model Scale Experiment

Added resistance / speed loss in waves The prediction of added resistance or speed

loss of a ship in waves is essential to evaluate the ship performance in a seaway. In the past several decades, experimental techniques on added resistance in waves have been well de-veloped, especially for ships in long and inter-mediate-length waves. However, experiments for added resistance in short waves are still a challenge to many researchers.

Some of the modern ships are very large, for example, a VLCC (Very Large Crude-oil Carrier) will exceed 320m in length. That means when the VLCC is travelling in normal sea states, most of the waves encountered can be considered as short waves. So the prediction of added resistance for ships in short waves is an important topic.

One of the major challenges is the genera-tion of short waves with high quality in wave basin. Waves with high steepness are unstable (called the Benjamin-Feir instability effect), and short waves with low steepness are subject to more spatial variation than long waves due to the variation in their transversal amplitude across the basin. The generation of short waves

is also restricted by the characteristics of the wave generator.

Guo and Steen (2011) carried out an ex-perimental study on the added resistance of KVLCC2 in short waves. The shortest wave length for model test is about 0.18Lpp. A unique feature of this experiment is that the ship model is divided into three segments: fore- segment, aft-segment, and parallel mid-body. An aluminium frame is used to keep the three segments together. The fore- and aft- segments are connected to the frame through springs and force transducers. The springs only absorb ver-tical forces, whereas the force transducers measure the longitudinal forces. The added resistance distribution with respect to the hull segments can be explored through this method. Before the experiment, a detailed wave calibra-tion was carefully performed. A new data proc-essing method was proposed to eliminate the effect of low-frequency noise in the measured force to achieve more accurate results.

The experimental results show that the added resistance is concentrated at the fore segment and that it is small at the aft segment. In the mid segment, the increase of frictional resistance due to short waves is very small (Figure 8). The non-dimensional added resis-tance coefficient measured by the experiment is fairly independent of wave amplitude, which confirms that the added resistance for short waves is roughly proportional to the square of the wave amplitude.

202202

Figure 8. Added resistance with respect to hull

segments The effect of oblique waves on ocean-going

vessel behaviour in realistic sea states was studied by Chuang and Steen (2013). Seakeep-ing model tests were carried out with a free running model in oblique waves in the ocean basin laboratory. Calm water resistance, azi-muth propulsion system, ship machinery, seakeeping, steering and automatic control were all included in the model tests. In order to compensate for the relatively higher frictional resistance of the model, a tow rope force was applied by an air fan mounted on the model. Due to the limitation of the experimental envi-ronment, converged speed in waves could not be achieved in all runs. A correction method was also proposed to find converged speed from non-converged model tests.

The experimental results show that in oblique waves, the speed loss increase with the added wave resistance (Figure 9). When wave length approaches ship length, the speed loss reaches its peak value. For a fixed heading an-gle, speed loss is increasing roughly linearly with increasing wave elevation for tests with constant propulsive power. When the power is kept constant in head sea and bow sea condi-tions, the higher the initial calm water speed, the less will the speed drop in waves.

Figure 9. Experimental speed loss in waves

Tanizawa, K. (2012) and Kitagawa, Y

(2014) introduced an experimental methodolo-gy for free running test to measure the nominal speed loss in waves. They developed two de-vices. One is a marine diesel engine simulator, MDES. Based on the mathematical model of a marine diesel engine, MDES controls the pro-peller rotational speed of model ships by real time simulation of engine response to the pro-peller loading oscillation. With MDES, engine characteristics could be considered in the mod-el test. The other is an auxiliary thruster system, ATS. This is a duct fan working in the air to add thrust to the model ship in order to correct for differences in skin friction. With ATS, the propeller loading condition of model ship could be adjusted to that of the full-scale ship at the same Froude number. They conducted a free running model experiment in waves using the MDES and ATS and measured not only ship motion responses but also the realistic dynamic responses of a ship propulsion system in waves such as propeller load and rotating speed oscil-lation, fuel supply rate and nominal speed loss in waves.

Influence of abnormal waves

Abnormal wave encounters can result in significant damage to or loss of a vessel. Sig-

203203

nificant attention should be paid to identifying the risks to a vessel when encountering abnor-mal waves.

Clauss and Klein (2011) investigated the generation, propagation, kinematics and dy-namics of extreme waves in a seakeeping basin. The measurements were conducted in the seakeeping basin of the Ocean Engineering Division, Technical University of Berlin. The spatial development of the extreme wave was measured in a range from 30.9m ahead of, to 21.0m behind the target position for a total of 520 registrations. The towing carriage was equipped with 13 wave gauges installed at an interval of 0.2m and the seakeeping basin was subdivided into 20 measurement sections. Fig-ure 10 shows the experimental set-up sche-matically, with a side view on the set-up (top) describing the measurement orders as well as a top view on the arrangement of the wave gauges installed on the towing carriage (bot-tom).

Figure 10. Schematic sketch of the extreme

wave experimental set-up

The impact of the extreme wave on a ship was also investigated, in particular the vertical bending moment. A Ro/Ro vessel in the ex-treme wave in head seas was studied (Figure 11). The wooden model was subdivided into three segments intersected at Lpp/2 and 3/4Lpp (measured from the A.P.). The segments were connected by three force transducers at each

cut. The force transducers registered the longi-tudinal forces during the model tests. The ver-tical wave bending moment superimposed by the counteracting vertical bending moment caused by the longitudinal forces can be deter-mined based from the measured forces. Figure 12 shows an example of experimental vertical bending moment (VBM) time traces.

Figure 11. Model test of a Ro/Ro vessel in an

extreme wave

Figure 12. Experimental results of VBM for

vessel in abnormal sea states The analysis of the registrations reveal ex-

treme waves occurring at three different posi-tions in the seakeeping basin, emerging from a

204204

wave group, which propagates almost con-stantly along the wave tank. The analysis of the total energy propagation shows that the wave crest velocity of the three waves in the wave group, i.e. the celerity is almost twice the ve-locity of the mean energy (group velocity). The investigations on wave-structure interaction between such an extraordinarily high wave and a segmented wooden Ro/Ro ship model reveal that the impact is severe and results in high global loads.

Bennett et al. (2012) carried out an experi-

mental investigation of global symmetric wave-induced loads, as well as motions, experienced by a naval ship (a frigate) in abnormal waves. Experiments were conducted using a seg-mented flexible backbone model in regular and irregular (random and abnormal) sea states at forward speed. Abnormal sea states were gen-erated using a previously developed optimisa-tion technique. Measurements were made of symmetric motions and the vertical bending moment at various locations along the ship. The influence of slamming on severity of ab-normal wave encounters was discussed.

Water on deck and slamming Green-water events are well recognised as

dangerous circumstances for marine vehicles in general. They are characterized by compact masses of liquid shipped onto the vessel deck due to the ship interactions with sufficiently severe sea states and their consequences can affect stability, structural integrity, operations on board and safety, depending on the vessel type and operational conditions. Slamming is another phenomenon of concern for ships and may occur in connection with water-shipping events, complicating the wave-ship interaction scenario. It is associated typically to small spa-tial and temporal scales, with location and fea-tures depending on the vessel geometry and operational conditions.

A synchronic 3-D experimental investiga-

tion was conducted by Greco et al. (2012) for wave-ship interactions involving the water-on-deck and slamming phenomena. The experi-ments examined a patrol ship at rest and with forward speed that was free to oscillate in heave and pitch in regular and irregular waves (Figure 13). In the study, the head-sea regular wave conditions were examined in terms of (1) RAOs and relative motions, (2) occurrence, features and loads of water-on-deck, bottom-slamming and flare-slamming events and (3) added resistance in waves. A systematic and comprehensive analysis of the phenomena was made available in terms of the Fr, incoming wavelength-to-ship length ratio and wave steepness. The main parameters that affect the global and local quantities were identified and possible danger in terms of water-on-deck se-verity and structural consequences were deter-mined. Different slamming behaviors were identified, depending on the spatial location of the impact on the vessel: single-peak, church-roof and double-peak behaviors. A bottom-slamming criterion was assessed.

Figure 13. Model test of water on deck and

slamming

Thomas et al. (2011), Lavroff et al. (2013) investigated slam events experienced by high-speed catamarans in irregular waves through experiments using a hydroelastic segmented

205205

model (Figure 14). It was tested in irregular head seas at two speeds relating to Fr of 0.32 and 0.60. Nearly 300 slams were identified in the test data and analyzed with respect to ki-nematic parameters. Slams were found to have a large range of magnitudes; however, the ma-jority of events were of relatively low severity. Differences in slam characteristics were found for two model speeds tested.

Figure 14. Slamming on the centre bow of the

catamaran model

Sloshing

Wave-impact in sloshing flows is an impor-tant issue for the safety of the LNG carriers. Ji et al. (2012) carried out experiments on non-resonant sloshing in a rectangular tank with large amplitude lateral oscillation. A sequence of experiments was performed to investigate large amplitude sloshing flows at off-resonant condition far from the system natural frequency. Through PIV measurement, it showed that the flow physics on nonlinear off-resonant sloshing problem can be characterized into a combina-tion of three peculiar sloshing motions: stand-ing wave motions, run-up phenomenon and gradually propagating bore motion from one sidewall to the opposite wall.

Bardazzi et al. (2012) carried out an ex-perimental study on the kinematic and dynamic features of a wave impacting a rigid vertical

wall of a 2D sloshing tank in the shallow water condition. The strain distribution along a verti-cal aluminium plate inserted in a rigid vertical wall of a sloshing tank was measured to char-acterize the dynamic features of the local loads. To assess the effect of the hydroelasticity, the same phenomenon was reproduced on the op-posite fully rigid wall of the tank. The experi-mental results show that although the overall kinematical evolution of the phenomenon is quite well reproduced, strong differences were observed in the dynamical features between elastic and rigid case.

Loads due to sloshing in LNG tanks not only act on tank walls as inner loads, but also affect the global wave loads by coupling with general motions of the carrier. Wang et al. (2012) investigated sloshing and its effects on global responses of a large LNG carrier. In their experiments, the interactions of sloshing motions and the global wave loads were stud-ied by seakeeping model tests of a self-propelled LNG ship with a liquid tank (Figure 15). The results show that the existence of liq-uid in tank will affect the vertical natural fre-quencies of the hull girder and natural rolling period of the ship. The motion period of liquid in the tank depends on the inner shape of the tank and the filling level, and on the wave heading and ship speed. The general effects of sloshing on global wave loads are not very re-markable, though the wave direction and ship speed are the sensitive parameters of the LNG carrier relative to sloshing.

Figure 15. Model test of LNG carrier with liq-

uid tank

206206

Other issues

Tiao (2011) carried out an experimental in-vestigation of nonlinearities of ship responses in head waves. The experimental program con-sisted of tests in both regular and irregular head waves, and the measured quantities included wave elevation, vertical motions, and hull pres-sures. By contrasting these results to the quasi-linear behaviours of heave motion, the nonlin-ear behaviours of pressure were highlighted and presented. Three nonlinear assessments, the probability density function, and the vari-ance spectra were provided.

Hashimoto et al. (2011) carried out the broaching prediction of a wave-piercing tumblehome vessel with twin screws and twin rudders (Figure 16). In their study, a series of captive model tests were conducted to measure the resistance, the manoeuvring forces, the wave-exciting forces, the heel-induced hydro-dynamic forces, and the roll restoring variation for the vessel.

Figure 16. Captive model test for broaching

2.2.1.2. Full Scale Experiment

Full-scale measurements are an extremely effective mechanism for investigating seakeep-ing behaviour, although they are complex and expensive to conduct.

Jacobi et al. (2013) investigated the slam-ming behaviour of large high-speed catamarans through full-scale measurements. The US Navy conducted the trials in the North Sea and North

Atlantic region on a 98m wave piercer catama-ran. For varying wave headings, vessel speeds and sea states the data records were interro-gated to identify slam events. An automatic slam identification algorithm was developed. This has allowed the slam occurrence rates to be found for a range of conditions and the in-fluence of vessel speed, wave environment and heading to be determined. The slam events were further characterized by assessing the relative vertical velocity at impact between the vessel and the wave.

Koning and Kapsenberg carried out a measurement campaign on board a 9,300 TEU container vessel. The measurements comprised ship performance parameters, cross section loads on two locations, local stresses in the bow area and accelerations on five longitudinal locations on deck. The wave environment was monitored by wave radar analyzing the back scatter from the waves and by two height level radars on the bow. Figures 17 and 18 show sample full scale time traces.

Figure 17. Sample full scale raw strain gauge

data showing slam events

207207

Figure 18. Full scale rigid body motions: roll

and pitch

2.2.2. Numerical Methods

Frequency Domain Methods for Motions and Loads

Due to the advances that have been made in the development and validation of time domain methods in recent years there is a visible shift in the literature from frequency based methods towards time domain methods. This shift in focus has reached the point where in applica-tion time domain methods are now superseding frequency domain methods to a large extent.

Nonetheless, in the early design stage fre-

quency domain methods prove more efficient in providing quick solutions, allowing for the evaluation of a large amount of design alterna-tives at a lower level of detail and complexity. Also for the analysis of typical zero or slow speed applications such as moored floating structures in waves and current and in particu-lar for multi-body problems as side-by-side moored systems, the frequency domain method

still proves to be a reliable and efficient solu-tion.

There has been recent work done on im-proving the numerical properties of frequency domain methods. Du et al. (2012) studied the occurrence of irregular frequencies for zero and for forward speed problems. They found that for most applications irregular frequencies oc-cur outside the range of practical interest for rigid body motions. However difficulties can occur in the analysis of large offshore struc-tures and in hydro-elastic problems of flexible bodies. They implemented a lid method to sup-press the occurrence of irregular frequencies at zero speed. Their work also shows that while irregular frequencies may not occur with for-ward speed, the disturbances can be caused by inaccurate treatment of the waterline integral terms and the solution method as the forward speed tends to smaller values.

Nan and Vassalos (2012) discuss the treat-ment of the m-terms in a forward speed fre-quency domain method. M-terms are second order derivatives of the steady flow potential that appear in the body boundary condition. In their study they evaluated the m-terms explic-itly with a numerical scheme in a frequency domain Rankine panel method. They showed agreement between the predictions from their method and model experiments.

As an example of the application of fre-

quency domain methods in design applications, Tello et al. (2011) presented a study of the seakeeping performance of a set of fishing ves-sels applying a linear three-dimensional fre-quency domain method. Maximo et al. (2012) used a linear frequency domain panel code to evaluate the seakeeping performance of a high speed trimarans vessel in a parametric design tool for rapid evaluation of various design solu-tions.

208208

As an illustration of the usage of frequency domain methods for zero speed applications, Wang and Xie (2012) combined a linear fre-quency domain method to compute the first order wave induced motions with mean and low frequency drift motions estimated from pre-computed drift design curves for a floating offshore unit. For the pre-computed drift mo-tion design curves use was made of a nonlinear coupled time domain analysis.

Zhao et al. (2011) investigated the interac-tions between the motions and inner-tank sloshing of a FNLG using a frequency domain method. They included the interior wetted sur-face of the tanks as conventional outer wetted surface and evaluated the effect of sloshing on the global response by comparing responses with and without the effect of sloshing.

Time Domain Methods for Motions and Loads

Time domain methods have gained increas-ing interest and many alternative methods have been developed over the last few decades. At this moment, time domain methods seem to be displacing the more traditional frequency do-main methods for many practical applications. The advantage of time domain methods lies in the more intuitive extension towards nonlinear motions and loads and the relative ease of in-corporating external forces, such as propulsion and control forces or coupling with for flexible structural modes and sloshing problems. This usually comes at the cost of an increased com-putational demand compared to frequency do-main methods. Especially for the more nonlin-ear approaches dealing with the geometry, for instance generating a panelization on the time dependent wetted surface can be a significant task.

There are many alternative time domain so-

lutions being developed and used for practical

applications. These range from two dimen-sional linear or nonlinear strip theory to three dimensional transient Green Function Methods (GFM) and Rankine Panel Methods (RPM). Emerging alternative potential flow based techniques are Higher Order Boundary Element Methods (HOBEM) and nonlinear potential flow Finite Element Methods (FEM). In some cases hybrid methods are being proposed.

(i) 2D time domain techniques

Two dimensional time domain methods are relatively efficient and less complex in devel-opment compared to three dimensional time domain approaches. Often they are based on frequency domain methods that are extended to the time domain by using retardation functions. Time domain based solutions exist and are of-ten applied for high speed planing problems.

Chuang and Steen (2013), for example, computed the speed loss of a vessel in oblique waves by combining linear strip theory using retardation functions to obtain a two dimen-sional time domain solution with second order wave forces, a thrust model and a nonlinear maneuvering model. The outcomes were com-pared with experimental data of a freely run-ning model.

Mortola at al. (2011) proposed a more complex time domain solution employing a two dimensional nonlinear radiation solution on the actual wetted surface below the undis-turbed waves combined with nonlinear restor-ing and wave exciting forces. They presented a comparison of the proposed method and two and three dimensional linear approaches ap-plied to the S-175 container ship.

For motions and loads of high speed plan-ing craft time domain methods based on two dimensional time domain theory are often ap-plied. Faltinsen and Sun (2011) computed the dynamic response of planing vessels in regular

209209

head seas using a 2D+t methodology. They introduced three dimensional corrections at the transom stern assessing the influence of the flow around the transom on the vertical plane motions.

Rijkens (2013) used a nonlinear semi-empirical strip theory method for high speed craft in an real-time active control scheme for reducing vertical acceleration levels in head waves. Continuous ship response predictions are made based on the incident wave to esti-mate the vertical acceleration level, leading to interventions by the control system when a threshold value is exceeded by means of thrust reduction or control device actuation.

(ii) 3D transient Green Function Methods

Three dimensional transient Green Func-

tion Methods only require panelling of the wet-ted hull geometry, relying on a linearized free surface boundary condition that is automati-cally satisfied by the transient Green function, as well as the radiation condition. Typically, the approach used allows for direct incorpora-tion of forward speed effects at the cost of a relatively complicated numerical scheme.

Time domain GFM approaches come in various degrees of complexity, ranging from fully linear time domain approaches that only require setting up the influence matrix once for the entire time domain simulation to body exact approaches that require re-panelling and re-computation of the influence matrix at each time step. There are many intermediate possi-bilities, by using nonlinear restoring forces and nonlinear Froude-Krylov forces on the actual wetted body. These approaches are often loosely termed ‘blended methods’ or ‘semi-nonlinear methods’.

Datta et al. (2013) used a linear time do-main GFM for the analysis of radiation forces on a ship advancing with forced heave and

pitch motions. They compared their outcomes to experimental data and the results obtained with conventional strip theory. They concluded that forward speed has a significant effect on the coupling effects between heave and pitch and stressed the importance of taking into ac-count the linear interactions between steady and unsteady flows.

The application of semi-nonlinear GFM to high speed semi-displacement vessels was studied by van Walree and de Jong (2011) and Hughes and Weems (2011). Van Walree and de Jong validated their body linear time domain method with nonlinear restoring and incident wave forces by deterministically comparing with the motions obtained with model experi-ments in stern-quartering seas of a fast patrol boat. To achieve this, they reconstructed the wave train from the experiments as an input for their simulations.

Hughes and Weems (2011) used a compa-rable method (LAMP) with an active ride con-trol system to simulate the motions of a high speed wave piecing catamaran and validated against data obtained from full scale sea trails. They also compared their outcomes with the results of linear frequency domain simulation and stressed the necessity of time domain simu-lation to enable nonlinear aspects of the ride control system.

A body exact GFM was presented by Zhang et al. (2011) using a more sophisticated version of LAMP. They introduced the pre-corrected Fast Fourier Transform (pFFT) method in their solution scheme to improve computational effi-ciency in terms of both CPU time and the re-quired core memory for (linear and nonlinear) problems with a very large number of un-knowns.

Van Walree and Turner (2013) presented

the development and validation of a body exact GFM. Based on the weak scattered assumption,

210210

they transformed hull surface vertically to ap-ply the linear free surface condition in a nonlinear way on the incident wave surface. They validated their results against motions and pressures obtained with model experiments with a patrol boat in head seas. Their method was shown to be able to capture the pressure peaks occurring during slam events.

(iii) 3D time domain Rankine Panel Methods

The Rankine Panel Method (RPM) uses a distribution of singularities of much simpler form compared to the GFM. However, in order to satisfy the free surface condition also panels need to be distributed over the free surface and the radiation condition requires an additional numerical method such as a numerical beach. The distribution of singularities over the free surface enables the relatively easy extension to nonlinear analysis. The RPM has gained sig-nificant popularity over the past decade. Also the RPM comes in multiple forms, ranging from fully linear to body exact and a nonlinear free surface condition.

Zaraphonitis at al. (2011) performed seakeeping analysis of a medium speed con-tainer vessel with a linear RPM. They also ap-plied linear strip theory and a frequency do-main GFM and compared the relative merits of the three computational methods of varying degree of complexity.

Ommani and Faltinsen (2011) applied the linear time domain RPM for the hydrodynam-ics of semi-displacement vessels. They incor-porated transom effects by modeling a hollow behind the transom based on an analytical ap-proach and the unsteady flow is linearized about the steady flow including the hollow. They showed results in good overall agreement with experimental data obtained in literature.

Kim and Kim (2013) combined a linear RPM with a numerical wave tank generating

linear waves and Boussinesq-type shallow wa-ter waves to evaluate the influence of nonlinear behaviour. They did not find significant differ-ences between linear and nonlinear waves. They performed an analysis of the hydrody-namic coefficients, wave loads, and motion responses for a LNG carrier and observed the influence of varying bathymetry.

Song et al. (2011) validated a weakly

nonlinear RPM consisting of a linear RPM combined with nonlinear restoring and incident wave forces for ship motions and structural loads on a container ship. They recommended that to control the non-restoring horizontal plane motions in steep stern quartering seas they carefully considered soft springs for better computational accuracy.

You and Faltinsen (2012) developed a fully nonlinear RPM combined with a numerical wave tank and numerical damping zone to simulate the interaction between moored float-ing bodies and waves in six degrees of freedom. After presenting verification and validation results they present a simulation of a moored Wigley hull in regular waves in shallow water.

Xu and Duan (2013) used a multi-transmitting formula with artificial wave speed to eliminate wave reflection on the artificial boundary, demonstrating that their method is capable of performing stable long time simula-tions of floating bodies. Nan and Vassalos (2012) included the m-terms in the body boundary condition of a RPM with a double body linearization. (iv) Higher Order Boundary Element Methods (HOBEM)

In higher order BEMs the boundary sur-

faces are discretized with higher order bound-ary elements avoiding some of the problems introduced by the stepwise discretization of the

211211

traditional constant panel methods. The higher order elements allow for much smoother repre-sentation of the velocity potential and its de-rivatives and therefore require much less ele-ments compared to traditional panel methods and allowing for much easier evaluation of spatial flow derivatives.

He and Kashiwagi (2013) developed a

higher-order BEM within the frame of linear potential flow theory to predict the radiation forces of a Wigley forced heave and pitch at forward speed. They used the Rankine source as the kernel function. The results were com-pared to model experiments and other numeri-cal solutions.

Shao and Faltinsen (2012) presented an al-

ternative formulation of the boundary value problem in a body-fixed coordinate frame, avoiding the numerical difficulties associated with the mj-terms and their derivatives. They used a higher order BEM with cubic shape functions as solution scheme. They applied the method to second order sum frequency excita-tion of ship springing.

(v) Finite Element Methods

An alternative to Boundary Element Meth-

ods is the application of the Finite Element Method to solve the potential flow problem. Hong and Nam (2010) used a FEM method to analyze second-order wave forces on side-by-side moored floating bodies. Yan and Ma (2011) used the quasi arbitrary Langrangian-Eulerian Finite Element Method based on fully nonlinear potential flow theory to investigate the nonlinear interaction between two floating structures.

(vi) Hybrid methods

Usually hybrid methods consist of a sophis-

ticated inner domain solution matched with a

more efficient outer domain solution. Tong et al. (2013) presented a matched Rankine Panel Method with a Green Function Method in the outer domain.

Kjellberg et al. (2011) developed a nested approach that combines a two-dimensional numerical wave tank with a three-dimensional fully nonlinear body exact boundary element method using constant strength source panels that only resolves the 3D flow in vicinity of the hull.

Guo et al. (2012) presented a coupled nu-merical wave model using a Volume Of Fluid (VOF) method to resolve the extreme wave motions near a structure while using a BEM further upstream.

Weymouth and Yue (2013) developed physics-based learning models for ship hydro-dynamics. This approach uses a very limited amount of high fidelity data points obtained from experiments or CFD computations com-bined with a large amount of intermediate data points for the same problem obtained from less accurate but far more efficient methods such as linear potential flow methods. The approach then uses both data sets to generate an im-proved prediction over the entire data range. The aim is to achieve far more accurate simula-tions, while spending a minimum amount of computational effort.

Maneuvering in Waves and Dynamic Stability

There is a growing interest in the assess-ment of the dynamic stability of ships operating in waves, due to IMO activity regarding the update of intact stability criteria. This devel-opment has led to an increased demand for numerical methods capable of dealing with the problem of a ship maneuvering in waves.

212212

Skejic and Faltinsen (2013) analyzed ship maneuvering in waves by using a unified seakeeping and maneuvering two-time scale model. They used an approximated method for slow drift second order drift forces combined with a maneuvering model based on nonlinear slender body theory.

Yu and Ma (2012) considered a frequency domain strip theory solution transferred to the time domain with nonlinear restoring forces incorporating rudder control and propeller forces. They applied the method to parametric roll of container vessels.

Belenky and Weems (2012) used a linear GFM combined with nonlinear restoring and incident wave forces to determine the interde-pendence of roll angles and rates. Van Walree (2012) used a very similar approach for the behaviour of a destroyer in steep stern-quartering seas.

Kim and Sung (2012) extended a nonlinear time domain seakeeping panel method by add-ing resistance, propulsion and maneuvering force models. They calibrated the maneuvering force model with captive model tests and car-ried out numerical simulations for a container vessel in waves.

2.2.3. Rarely Occurring Events

Slamming Slamming is defined as an impact between

the hull of a vessel and the water surface. Keel, stern, flare or wet deck slamming can impart significant global and local structural loads onto vessels. The impacts can also induce vi-bration within the ship (known as whipping) and can ultimately lead to an increase in struc-tural fatigue.

Developing techniques to accurately predict the magnitude of slamming events is still a key focus for researchers. Yang et al. (2013) pre-sented a technique to estimate slamming im-pact loads and dynamic structural responses of containerships at an initial design stage using a direct analysis method based on fluid-structure interaction. The method is based on using a commercial CFD program (STAR-CCM+) and a structural analysis program (ABAQUS), re-spectively. Bow and stern slamming loads were calculated, but the authors undertook no valida-tion. Rahaman and Akimoto (2012) used a RANS based motion simulator to model slam-ming of a modern container ship. The numeri-cal method was successfully validated in regu-lar head waves and mechanism of slamming on the bow flare region analyzed based on visuali-zation of flow field.

Full-scale measurements are an extremely effective mechanism for investigating slam-ming behavior, although they are complex and expensive to conduct. Ogawa et al. (2012) ex-amined the relationship between the occurrence probability of a slamming induced vibration and sea state based on the full-scale measure-ment data of two large container ships.

Jacobi et al. (2014) investigated the slam-ming behaviour of a 98m high-speed catamaran through the analysis of extensive full-scale trials data. Slam occurrence rates were found for a range of conditions and the influence of vessel speed, wave environment and heading determined. Since the ship was equipped with a ride control system its influence on the slam occurrence rates was also assessed. Identifying slam events in full-scale trials data can be chal-lenging; however Amin et al. (2012) introduced, described, applied and recommended the con-tinuous wavelet transform as an effective means to identify and investigate the wave in-duced hull vibrations in both the time and fre-quency domains simultaneously.

213213

Using experimental data for a hydroelastic model of a high-speed ferry, Dessi and Chiappi (2013) analyzed the statistical properties of the slamming impact process. One of their major findings was that the impact statistics are largely affected by the grouping of slams into clusters, thus violating the hypothesis of mutual independence between successive impacts that is at the basis of most of the statistical models. They also proposed a new criterion for slam-ming identification based on the evaluation of the whipping bending moment.

Chen et al. (2012) having performed model experiments on a segmented hydroelastic model concluded that in larger sea states the influence of whipping has a major influence on the magnitude of the longitudinal bending mo-ment. They also found that a linear hydroelastic theory can accurately predict the bending mo-ment in small sea states.

A hydroelastic model was used by Lavroff et al. (2013) and French et al. (2013, 2014) to examine the slamming behaviour of large high speed catamarans. Lavroff et al. (2013) per-formed towing tank tests in regular seas to measure the dynamic slam loads acting on the centre bow and vertical bending moments act-ing on the demihulls of the catamaran model as a function of wave frequency and wave height. Peak slam loads measured on the centre bow of the model were found to approach the total mass of the model. French et al. (2013, 2014) investigated slamming behaviour in irregular waves finding that encounter wave frequency and significant wave height are important pa-rameters with regard to centrebow slamming, but that relative vertical velocity is a poor indi-cator of slam magnitude.

Water Entry

The ability to accurately predict the loads and pressures on a body entering the water is fundamental to the slamming problem.

Korobkin (2011, 2013) continues to work on this fundamental problem. Korobkin (2011) presented a numerical method to solve the problem of symmetric rigid contour entering water at a given speed based upon the so-called Generalized Wagner Model (GWM). The solu-tion derived predicts accurately the hydrody-namic force similar to Modified Logvinovich Model (MLM) but additionally it gives access to the pressure distribution, which is not avail-able within MLM. This method was extended by Korobkin (2013) to accurately account for the second stage of the flow, when the wedge is already completely wetted and a cavity is formed behind the wedge.

Drop tests provide the ability to obtain ex-perimental results for the water entry problem. Alaoui et al. (2012) conducted drop tests on cones (with and without knuckles) and hemi-spheres at constant velocity. The experimental set up enabled impacts at high-speeds with small velocity deviations. Good agreement between numerical results using Impact++ ABAQUS, ABAQUS/Explicit and FLUENT codes and available experimental measure-ments were obtained.

Panciroli (2012) conducted a series of drop test experiments on flexible wedges and found that large structural deflormations generate two fluid-structure interaction phenomena that never occur in rigid-bodies impact: (i) the repe-tition of impacts and separation between the fluid and the structure in the region character-ized by the fluid jet generated during the water entry and (ii) an underpressure region with a cylindrical wavefront in the underwater fluid/structure interface. Yamada et al. (2012) used LS-DYNA whereby the fluid structure

214214

interaction (FSI) is taken into account by cou-pling fluid analysis and structural analysis in each time step of time domain simulations. Comparisons were made of the pressure distri-bution and slamming impact water entry of a rigid wedge, with those determined by conven-tional Wagner theory.

Reynolds-Averaged Navier-Stokes Equa-tions (RANSE) appear to be able to satisfacto-rily model the water entry problem. Swidan et al. (2013) used quasi-2D drop test experimental measurements to validate the simulation of symmetric wedge water impacts using RANSE, with close agreement found between the ex-perimental and numerical results.

Green Water Green water on deck can result in signifi-

cant loads that are significant with respect to the safety of forward stowed cargo and deck equipment. Kim et al. (2013) provided an analysis procedure to calculate the design pres-sure on ship’s breakwaters using the Computa-tional Fluid Dynamics (CFD) method and pro-vided the technical background of the newly proposed rule requirements for breakwaters. Zhang et al. (2013) used a Moving Particle Semi-implicit (MPS) method to simulate green water on deck scenarios and successfully vali-dated the technique with experimental data available in the literature. A similar MPS method was used by Bellizi et al. (2013) to investigate the effect of bow shape on green water on deck.

Buchner and van den Berg (2013) studied green water on deck emanating from the side of the vessel using experiments. They concluded that this is a very complex process that will need CFD for the prediction of important non-linear effects. Their model tests can be used as important validation material in this process.

Extreme Accelerations on Small High-Speed Craft

When operating in waves, small high-speed craft can experience extreme accelerations if the hull exits the water and slams upon re-entry. Modelling the wave impacts is a current indus-try challenge and as such Rose et al. (2011) used a vibro-impact oscillator to model non-linear planing hull accelerations and predict extreme events in variable environments. Whilst Riley et al. (2011) presented a simpli-fied approach to quantifying the comparison of acceleration responses of small high-speed craft in rough seas and proposed the use of a Ride Quality Index (RQI).

An effective method of reducing the likeli-hood of these extreme events is through a ride control system. Rijkens et al. (2011) developed a computational tool for the design and optimi-sation of these ride control systems for high speed planing monohulls. Hydrodynamic char-acteristics of both transom flaps and intercep-tors were determined by a systematic series of model test experiments.

2.2.4. Hydroelasticity

Understanding the hydroelastic response of a ship is an important part of the overall struc-tural response. This is true for both extreme ship structural responses and the fatigue loads of some structural details. The challenges are both in model test techniques as well as devel-opment and verification/validation of numeri-cal methods. The applicability of the methods for design applications are also addressed.

K.-H. Kim et al. (2013) presented a fully coupled BEM-FEM analysis for ship hydroe-lasticity in waves. For the analysis of fluid-structure interaction problems, a partitioned method was applied. The fluid domain sur-rounding a flexible body was solved using a B-spline Rankine panel method, and the structural

215215

domain was handled with a three-dimensional finite element method. The two distinct meth-ods were fully coupled in the time domain. The numerical results of natural frequency and the motion responses of simple and segmented barges were computed to validate the method. The study extended to the application to two real ships, 6500 TEU and 10,000 TEU contain-erships, for more validation and also observa-tion on the practicality of the method. It was found that the method provides reliable solu-tions to linear ship hydroelasticity problems.

J.-H. Kim et al. (2013a) introduced an analysis of ship hydroelasticity for a fatigue assessment of ship structural design. In this study, the hydroelastic analysis for springing and whipping was carried out by using a fully coupled three-dimensional BEM-FEM ap-proach with two-dimensional slamming theo-ries, and a sequential fatigue assessment is per-formed. The fatigue damage was decomposed to wave frequency and high frequency compo-nents. Furthermore, the high frequency compo-nent was again decomposed to 1st harmonic springing, super harmonic springing and whip-ping contributions. The amount of the contribu-tions was compared in irregular sea states.

J.-H. Kim et al. (2013b) applied different numerical methods for the coupled hydroelas-ticity analysis of ship structures in regular and irregular waves. For the hydrodynamic analysis of flexible body motion, a time domain Rankine panel method was applied. For the structural analysis, three different approaches were considered: beam approximation, modal approach by using the eigenvectors of three-dimensional (3-D) finite element (FE) model, and full 3-D FE analysis. For the computation of slamming force, wedge approximation and generalized Wagner model (GWM) were ap-plied for 2-D slices of the ship. The computa-tional results were compared with experimental results for the validation of the methodology

and numerical scheme. The hydroelastic mo-tions and loads on ship structures were com-pared for segmented models of large container-ships.

He and Kashiwagi (2012) developed a hy-droelastic simulation method based on BEM with MEL for fully nonlinear water waves and FEM for elastic deflection. A hybrid wave-absorbing beach was installed to prevent wave reflection from the end of the wave tank. Using this simulation method, they simulated the in-teraction of a surface-piercing plate with non-zero initial free surface and compared the result with the corresponding linear analytical solu-tion. They also simulated hydroelastic response of a surface-piercing vertical plate due to a solitary wave.

Das and Cheung (2011) proposed a hydroe-lasticity model to couple the hydrodynamic load, elastic deformation, and rigid-body mo-tion for marine vessels advancing in ocean waves. Small amplitude assumptions of the surface waves and body surface motions lead to linearization of the mathematical problem in the frequency domain. The formulation adopted a translating coordinate system with the free surface boundary conditions account-ing for the double body flow around the vessel and the radiation condition taking into account the Doppler shift of the wave field. A boundary element model, based on the Rankine source distribution, described the potential flow and the hydrodynamic pressure on the vessel. A finite element model relates the hull motion to the hydrodynamic pressure through a kinematic and a dynamic boundary condition. This direct coupling of the structural and hydrodynamic systems leads to a matrix equation in terms of the body surface displacement. The model was verified with published data from the modal superposition method without forward speed effects and applied to examine the characteris-

216216

tics of a flexible Wigley hull advancing in waves.

Piro and Maki (2011) studied hydroelastic impact together with the exit of simple ship sections. The method used a loosely coupled fluid-structure interaction (FSI) solver to cou-ple a finite element model to a computational fluid dynamics (CFD) model. The structure was represented using beam and plate finite elements and decomposed into its dry mode shapes. The motion of the structure was applied to the boundary of the CFD simulation using either the exact or approximate body boundary condition. The fluid pressure on the structure was expanded in the structural modes and ap-plied in the force term of the structural equa-tions of motion. The system was solved itera-tively in each time step to ensure time accuracy. The hydroelastic impact of a wedge was stud-ied to validate the numerical method and the exit of the wedge from the water was investi-gated.

Paredes and Imas (2011) investigated the three-dimensional fluid-structure interaction between a free-surface disturbance and a de-formable membrane as a canonical problem representative of the interaction between a sur-face-effect ship (skirt) advancing with forward speed in waves. The numerical study was per-formed using a hydrodynamic solver developed around an SPH algorithm that was used to si-multaneously model both the fluid dynamics and structural dynamics with two-way fluid-structure coupling. Results from this study were presented along with validation examples and as well as a discussion of their SPH algo-rithm, in particular their methodology for treatment of boundary conditions, FSI, and fluid viscous effects.

Iijima et al. (2011) evaluated the effect of the wave-induced vibrations on long-term fa-tigue damage in various types of ships is evalu-ated by using a series of numerical simulations.

A bulk carrier, a VLCC, and a container carrier were employed as subject ships. A fully three dimensional numerical method was employed for evaluating the load effects. The pressure obtained by three-dimensional potential theory was integrated over the instantaneous wet sur-face to account for linear and nonlinear wave loads. Slamming loads were separately mod-elled by using momentum theory. The calcula-tions were performed for the respective short-term sea states. The characteristics of the fa-tigue damage by the wave-induced vibrations were clarified. It was shown that the amount of the increase in fatigue damage depends on the wave loading properties of the ships in waves as well as the structural properties such as natu-ral frequencies of flexible modes.

Stenius et al. (2011) discussed challenges in modeling and quantifying hydroelastic effects in panel-water impacts and summarised results from numerical and experimental studies. Ki-nematic and inertia related hydroelastic effects were discussed and exemplified in relation to pressure distributions and structural responses. Hydroelastic effects were quantified by com-paring hydroelastic results with rigid/quasi-static reference results. The formulation of non-coupled reference solutions in experimental studies is particularly challenging and the paper addressed this problem by outlining a semi-empirical approach to reach such solutions. For those impact situations where the hydroelastic interaction seemed to have a significant effect, it was found both numerically and experimen-tally that the hydroelastic effects were amplify-ing the structural responses in comparison to the rigid/quasi-static reference solutions. Two approaches for characterization of impact situa-tions regarding the involved hydroelastic ef-fects in relation to panel properties and impact conditions were discussed and exemplified. These approaches can tentatively be used to evaluate the hydroelastic effects in design situations.

217217

White et al. (2012) presented some methods

to determine values of dynamic bending mo-ments considering the effects due to whipping and springing which are suitable for design application. Examples of the use of these methods were also presented.

Senjanović et al. (2011, 2013) discussed treatment of the restoring stiffness, which cou-ples displacements and deformations, playing a very important role in hydroelastic analysis of marine structures. The problem of its formula-tion is quite complex and is still discussed in relevant literature. Different numerical formu-lations were implemented and compared.

Begovic et al. (2011) presented an experi-mental investigation to obtain motion and load measurements of an intact and damaged frigate model in waves. The experimental measure-ments showed the changes in motion and hull girder loading when a ship hull is damaged. The obtained data were compared with numeri-cal predictions from non-linear time domain motion code (strip theory) implemented in ShipX.

Chen et al. (2012) carried out segmented ship model experiments on bow slamming and whipping of a ship. A nonlinear hydroelasticity method considering slamming loads was pro-posed ． Variable cross-section beams were used to improve the simulation of the stiffness of the hull．Severe bow slamming was ob-served when the model was in head-following regular waves. Experimental results showed that when the wave height increased from 5.6m to 21m the mean value of the total moment increased from 25% to 92% compared with that of the wave moment because of severe whip-ping．The measured results on the central hull in different sea states were compared with cal-culations based on linear and nonlinear hydroe-lasticity theory showing that their present

method and program can predict the wave loads properly.

Matsubara et al. (2011) performed model tests on a segmented model of a wave-piercing catamaran to obtain experimental values of global motions and loads as well as slamming loads, with a particular focus on the influence of the centrebow configuration. The motions were found to be distinctly non-linear with respect to wave height; this was due to the im-mersion of the centrebow in larger waves tend-ing to reduce the heave and pitch motions. The wave loads were found to be dominated by the slam load on the centrebow, varying in magni-tude and location with respect to wave condi-tions.

Wu and Stambaugh (2013) presented a comparative study carried out for a 45m long high-speed vessel. The time history of the ver-tical bending moments (VBM) and the standard deviations of both wave-frequency and high-frequency components in the VBM were com-pared between model tests and numerical simu-lations. A comparison of the probability of ex-ceedance derived from the hydroelastic hog-ging and sagging vertical bending moments was also presented. Different aspects of model testing and numerical simulation were dis-cussed. The paper concludes that an integrated approach, that uses the advantages of both model testing and numerical simulation while overcoming the drawbacks of either method applied alone, is the best way forward in the near future.

Halswell et al. (2011) discussed each area of hydroelasticity found in an inflatable boat; defining each problem and possible methods of investigation. Anecdotal evidence has shown that this flexibility or hydroelasticity of an in-flatable boat improves its performance, espe-cially in waves.

218218

Besten et al. (2011) developed an analytical, 2D, mathematical model for the local structural response of a hydrodynamic impact loaded sandwich structure with vibration isolation and structural damping properties. The structural response was determined by solving semi-analytically a hydro-elastic coupled sandwich flexible core model and a hydrodynamic im-pact model in modal space, verified by results found in literature and FEM calculations.

3. PROCESS FOR THE ESTIMATION OF SHIP SPEED REDUCTION COEFFICIENT FW IN WAVES

3.1. Introduction The speed reduction coefficient fw is intro-

duced in the 2012 Guidelines on the method of calculation of the attained energy efficiency design index for new ships (EEDI), adopted by MEPC.212(63). fw is a non-dimensional coef-ficient indicating the ship speed reduction in a representative sea condition of wave height, wave frequency and wind speed. As the repre-sentative sea condition, Beaufort scale 6 was adopted by MEPC considering mean sea condi-tion of north Atlantic and north Pacific. fw can be determined by conducting the ship specific simulation on its performance at representative sea condition.

In the following review of the state of the art for the fw estimation process, ship resis-tance as well as brake power in a calm sea con-dition (no wind and no waves) is assumed to be evaluated by tank tests, which means model towing tests, model self-propulsion tests and model propeller open water tests. Numerical calculations can be used as equivalent to model propeller open water tests or used to comple-ment the tank tests conducted to evaluate the

effect of additional hull features such as fins, etc., on ship's performance.

For the estimation of fw to evaluate EEDI,

the design parameters and the assumed condi-tions in the simulation to obtain the coefficient fw should be consistent with those used in cal-culating the other components in the EEDI. 3.2. Basic Conditions in the Prediction of

Ship Speed Reduction

Symbols for ship performance (also refer to Figures 19 and 20)

: Brake power : Total resistance in a calm sea condition (no wind and no waves) : Design ship speed when the ship is in op-eration in a calm sea condition (no wind and no waves)

: Design ship speed when the ship is in op-eration under the representative sea condi-tion

: Added resistance due to waves

: Added resistance due to wind : Propulsion efficiency : Transmission efficiency

Subscript refers to wind and wave sea con-

ditions.

Symbols for representative sea conditions

: Angular distribution function : Directional spectrum : Significant wave height

: Frequency spectrum : Mean wave period : Angle between ship course and regular

waves (angle 0(deg.) is defined as the head waves direction)

BP

TR

refV

wV

waveR∆

windR∆Dη

Sη

w

DEHSTα

219219

: Mean wave direction ( = 0 (deg.)) : Circular frequency of incident regular waves

Figure 19. Relationship between power and ship speed reduction.

Figure 20. Flow chart of the calculation of ship speed reduction The representative sea conditions for ships

have to be determined first. The sea condition for the prediction of ship speed reduction is

dependent on marine area. Larger ships are operated in relatively shorter wave length and lower wave height waves than smaller ships.

θ θω

220220

Therefore, even in the same sea condition, ship speed reduction can be dependent on ship di-mension, i.e. capacity of cargo, and ship type. The direction of wind and waves are defined as heading direction, which has the most signifi-cant effect on the speed reduction. As ocean waves are characterised as irregular, the direc-tional spectrum should be considered. To ob-tain the mean wave period from the Beaufort scale, the following formula derived from a frequency spectrum for fully-developed waves is used.

(1)

where H is the significant wave height in me-tres and T is the mean wave period in seconds.

The directional spectrum E is composed of

frequency spectrum S and angular distribution function D .

(2)

(3)

where

, , ,

(4)

Ships are assumed to be in steady navigating conditions on a fixed course with constant main engine output. The current effect is not consid-ered.

The total resistance in the representative sea condition, , is calculated by adding the added resistance due to wind and waves to the total resistance in a calm sea condition

. The ship speed is the value of

where the brake power in the representative sea condition equals to , which is the brake power required for achieving the speed of in a calm sea condition. Where can be derived from the total resistance in the repre-sentative sea condition , the properties for propellers and propulsion efficiency should be derived from the formulas obtained from tank tests or an alternative method equivalent in terms of accuracy, and transmission effi-ciency should be the proven value as verifi-able as possible. The brake power can also be obtained from the reliable self-propulsion tests.

(5)

The coefficient of the ship speed reduction fw is calculated by

( 6) at the point where

at at . (7)

Total Resistance In A Calm Sea Condition: RT The total resistance in a calm sea condition (no wind and no waves) is evaluated by tank tests, which means model towing tests, model self-propulsion tests and model propeller open water tests. Numerical calculations may be accepted as equivalent to model propeller open water tests or used to complement the tank tests conducted (e.g. to evaluate the effect of addi-tional hull features such as fins, etc., on ship's performance).

Total resistance in the representative sea condition: RTw The total resistance in the repre-sentative sea condition, , is calculated by adding , which is the added resistance due to wind, and , which is the added

HT 86.3=

( , ; , , ) ( ; , ) ( ; )E H T S H T Dω α θ ω α θ=

4e),;( 5ω

ωω

SBSATHS

−=

42 24

=

zS T

HA ππ

421

=

zS T

B ππ

TTz 920.0=

( ) ( )

( )

≤−−

=others 0

2cos2

,2

　　

　　παθαθ

πθαD

TwR

wR∆

TR wV V

BwP BPrefV

BwP

TwR

Dη

Sη

( )SDTB VRP ηη=

refww VVf /=

BP Bwref PV = wV

TwR

windR∆

waveR∆

221221

resistance due to waves, to the total resistance in a calm sea condition .

(8)

Added resistance due to wind: Added

resistance due to wind can be calculated by the following typical formula on the basis of the mean wind speed and wind direction.

(9)

CDwind should be calculated by a formula with considerable accuracy, which has been con-firmed by model tests in wind tunnel. More general formula can be applied when wind di-rection is not longitudinal, e.g. Fujiwara and Ueno (2006), Blendermann (1994). The verti-cal profile of wind can be also considered. There are a few different models of vertical variation for ocean waves such as models based on power law (Blendermann, 1994) and loga-rithmic approximation (DNV, 2010). These models can be applied for the more accurate prediction of CDwind.

Added resistance due to waves: ∆Rwave. Ir-

regular waves can be represented as linear su-perposition of the components of regular waves. Therefore added resistance due to waves is also calculated by linear superposition of the direc-tional spectrum E and added resistance in regu-lar wave.

2

20 0

( , ; )2 ( , ; , , )wavewave

a

R VR E H T d dπ ω α ω α θ ω α

ς∞

∆ = ∫ ∫(9)

Added resistance in irregular waves Rwave

should be determined by tank tests or a formula equivalent in terms of accuracy. In cases of applying the theoretical formula, added resis-tance in regular waves, Rwave, is calculated from the radiation and diffraction components of added resistance primary induced by ship mo-tion and wave diffraction in regular waves, Rwm, and the reflection component due to wave re-flection for the correction of added resistance in short waves, Rwr.

(10)

TR

wavewindT

wTTw

RRRRRR

∆+∆+=∆+=

windR∆

( ) 2 212wind a T Dwind wind w refR A C U V Vρ∆ = + −

wrwmwave RRR +=

222222

Table 4. Methods for added resistance prediction

3.3. Calculation Methods for Added Resis-tance in Regular Waves

Added resistance can be obtained either by

using numerical computation or towing-tank experiments. Since added resistance is the sec-ond-order mean quantity which can be obtained by linear solution of the seakeeping problem, linear seakeeping programs can be applied. The method of added resistance prediction in regu-lar waves can be summarized as in Table 4. The comparison of added resistance obtained by different methods has been recently intro-duced by Seo et al. (2013).

3.4. Correction of added resistance in short waves, Rwr .

Symbols

: Ship breadth : Bluntness coefficient, which is de-

rived from the shape of water plane and wave direction

β : Wave incident angle (defined in Figure 19)

: Coefficient of advance speed, which is determined on the basis of the guidance for tank tests

Approaches Numerical method

Experiment Slender-body theory 3D panel method CFD

Added resistance

computation

Direct pressure integration (e.g. Faltinsen et al, 1980, Kim & Kim, 2011) Direct pressure integration:

Added resistance = (Total Resistance in waves) –

(Resistance in cal water)

Momentum conservation method (e.g. Maruo, 1960, Joncquez, 2009)

Radiated energy method (e.g. Salvesen, 1978)

Methodology

Strip method, (enhanced)

unified theory

Green-function method, Rankine panel

method

Commercial or in-house

codes

Surge-fixed or surge-free

tests

Linear formulation for seakeeping. Fully

nonlinear formulation.

Fully nonlinear Short-Wave

Approximation Faltinsen’s approximation, NMRI’s

empirical formula

Remarks

Quick computation

Different formulations for time-domain and frequency-domain

methods.

A lot of computationa

l time Expensive

In shot waves, empirical or asymptotic

formula should be combined.

Grid dependency should be observed in

short waves.

Strong grid dependency

in short waves.

Scale dependency and

repeatability should be observed.

BfB

UC

223223

: Ship draft : Froude number (non-

dimensional number in relation to ship speed)

: Gravitational acceleration : Modified Bessel function of the first

kind of order 1 k : Wave number of regular waves

: Modified Bessel function of the sec-ond kind of order 1

Iζ : Incident wave elevation 1 2 3( , , )n n n n=

: Normal vector on ship surface L : Ship length ρ : Water density U : Ship speed (x0, y0) : Position of body surface

eω : Encounter wave frequency To overcome the difficulty of computing

added resistance in short waves several formu-lae can be used:

Ray theory formulation: Faltinsen et al.

(1980) The integration in Eq. (11) is performed

over the non-shaded part (A-F-B) of the wa-terline as shown in Figure 21.

Figure 21. Coordinate system for the added

resistance calculation in the short wave range

[ ]

2

2

1 2 6 0 0

12

2 sin ( ) 1 cos cos( )

( sin , cos , cos sin )

wave I

L

R g

U ndLg

n n n x y

ρ ζ

ωθ β θ θ β

θ θ θ θ

=

× − + + −

= = = −

∫

(11)

Semi-empirical formulae

21(1 ) ( )2wave d U I fR g BBα α ρ ζ β = +

(12)

where 2 21( ) sin ( )sin sin ( )sinf I II

B dl dlB

β θ β θ θ β θ = − + + ∫ ∫

(13)

- Fujii and Takahashi (1975) 2 2

12 2 2

1 1

( ) , 1 1 5( ) ( )d U n

I kd FI kd K kdπα α

π= + = +

+ (14)

- NMRI (Tsujimoto et al. 2008, Kuroda et al.

2008)

Added resistance in regular waves for cor-recting is calculated as follows.

(15)

where

,

, ,

,2

2

1( ) sin ( )sin

sin ( )sin

f I

II

B dlB

dl

β θ β θ

θ β θ

= −

+ +

∫

∫

is a line element along the water plane,

is the slope of line element along the waterline, and domains of integration are shown in Figure 22. Unified definition of the heading angle of

dgLVF ppn =

g1I

1K

wmR

dnUfawr FCBBgR αζρ )1(21 2 +=

)()()(

21

21

2

21

2

dKKdKIdKI

ee

ed +=π

πα

( )2cos1 αΩ+= KKe gVωΩ =

dl wβ

224224

ship to wind and wave is used to prevent con-fusion in MEPC, i.e. α = 0 for head sea.

Figure 22. Coordinate system for

Effect of advance speed is determined as follows:

(16)

The coefficient of advance speed in oblique

waves is calculated as follows:

(17)

where

(i) or : ,

(ii) and :

,

and , .

3.5. A Practical Estimation of fw from Standard Curve

The design parameters in the calculation of

fw from the standard fw curves should be consis-tent with those used in the calculation of the other components in the EEDI. Three kinds of standard fw curves are provided for bulk carri-ers, tankers and containerships, and expressed as a function of Capacity defined in the 2012

Guidelines on the method of calculation of the attained Energy Efficiency Design Index for new ships (EEDI), adopted by MEPC.212(63). Ship types are defined in regulation 2 in Annex VI to the International Convention for the Pre-vention of Pollution from Ships, 1973, as modified by the Protocol of 1978, as amended by resolution MEPC.203(62).

The Japanese delegation suggested a

method to estimate the coefficient fw from the standard fw curves. When real ship data for speed reduction are known, this method can be an alternative method, which does not require computation or experiment. When this is the case, the accuracy of real ship measurement is essential. Otherwise, this approach can provide inaccurate prediction of the coefficient fw.

Example

Each standard fw curve has been obtained

on the basis of data of actual speed reduction of existing ships under the representative sea con-dition in accordance with procedure for deriv-ing standard fw curves. Each standard fw curve is shown from Figure 23 to Figure 25, and the standard fw value is expressed as follows:

fw = a ln(Capacity)+ b (18)

where a and b are the parameters given in Ta-ble 5.

Table 5. Parameters for determination of standard fw value

Ship type a b Bulk carrier 0.0429 0.294 Tanker 0.0238 0.526 Containership 0.0208 0.633

wβ waves I

II

Y

XGfore aft

α

wrR

Uα

nUU FC )(αα =

)(αUC

[ ]CSU FFC ,Max)( =α

fcf BB <= )0(α fsf BB <= )0(α )0()(310)0( =−−== ααα ffUS BBCF

[ ]10),0(Min == αUC CF

fcf BB ≥= )0(α fsf BB ≥= )0(α)(31068 αfS BF −= )0( == αUC CF

31058

=fcB310

)0(68 =−=

αUfs

CB

×

225225

Figure 23. Standard fw curve for bulk carrier

Figure 24. Standard fw curve for tanker

Figure 25. Standard fw curve for containership

4. CFD-BASED ANALYSIS ON SEAKEEPING PROBLEMS : STA-TE OF THE ART REVIEW AND SUMMARY OF METHODOLOGY

During the past two decades, thanks to the

rapid development of computer power, compu-tational fluid dynamics (CFD) has been applied to some seakeeping problems. In the broadest sense, ‘CFD method’ refers to all computa-tional methods for fluid flow, including bound-ary element methods (BEM), finite element methods (FEM), finite difference, or volume, methods (FDM/FVM), spectral methods, etc. However, it is now generally understood that the term ‘CFD method’ concerns only the field equations, i.e. the continuity equation and the Navier-Stokes, or the Euler equation. There are several criteria for the taxonomy of CFD based methods for seakeeping analysis as follows: - Grid system: grid based method (FDM,

FVM, FEM) vs. particle method (SPH, MPS) - Characteristics of flow I: inviscid vs. vis-

cous (RANS, LES) - Characteristics of flow II: incompressible

(SIMPLE, fractional step) vs. compressible (artificial compressibility)

- Treatment for interface: interface tracking vs. interface capturing (VOF, Level-Set)

- Treatment for moving body: boundary-fitted (re-mesh, overlapping) vs. immersed bound-ary

- Domain of problem : global flow vs. local flow

This is graphically summarised in Figure 26.

Current numerical methods can be catego-rised largely into two groups: grid methods and gridless methods. The former is known as an Eulerian approach, which discretizes a fluid volume in structured or unstructured grids and solve the field equations defined on these spa-tial grids. On the other hand, gridless methods

0.700 0.750 0.800 0.850 0.900 0.950 1.000

0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000

f w

Capacity

Container ship

Observed fw of existing ships Draft standard fw curve regression

0.700 0.750 0.800 0.850 0.900 0.950 1.000

0 100,000 200,000 300,000 400,000

f w

Capacity

Tanker

Observed fw of existing ships Draft standard fw curve regression

0.700 0.750 0.800 0.850 0.900 0.950 1.000

0 50,000 100,000 150,000 200,000 250,000

f w

Capacity

Bulk carrier

Observed fw of existing ships Draft standard fw curve regression

226226

have seen increased applications recently. These methods, e.g. SPH (smoothed particle hydrodynamics) and MPS (moving particle semi-implicit method), define a finite number of fluid mass (basically, they are volume frac-tions) and solve the field equations by using their interactions.

In most cases in classical seakeeping prob-lems the effects of viscosity are limited to roll motions or flow around appendages. That is, most problems related to free surface flows in seakeeping problems are inertia-dominant problems and therefore diffusion effects are

relatively smaller than convection effects. In fact, this is the reason why potential flow the-ory is valid in the ship motion problem and is capable of reasonable accuracy. In many cases, the more important physical phenomenon is the interaction between the free surface flow and air flow. This is the case particularly when the hydrodynamic pressure due to local impacts is of primary interest. As the related problems of ship propulsion, or manoeuvring, in waves be-come of more interest then the importance of viscous effects will increase in comparison to classical seakeeping problems.

Figure 26. Overall status of the art of CFD schemes: Field equation solvers

227227

Table 6. Summary of CFD methodology for seakeeping analysis

The key technology in the application of CFD methods to seakeeping problems, includ-ing ship motion and local free-surface flows, is how to obtain or trace the dynamic free-surface profile. When grid methods are applied, there are several candidates to choose for the imple-mentation of dynamic and kinematic free-surface boundary conditions. For ship motion problems, VOF (Volume of Fluid) and level-set approaches are popular, but there has also been recent work done using other methods. A good example is CIP (constrained interpolation pro-file) method. In contrast to grid methods, the numerical treatment of the free surface in parti-cle methods is more straightforward. Most of them adopt a Lagrangian method, i.e. particle tracking with time-marching. Along with the simulation of particle motions inside a fluid volume, particle movement on the free surface can be used to trace its profile. At present commercial programs and the open source pro-gram OpenFOAM are commonly applied and it is likely that the application of these programs will be more popular in the future.

The main reason for applying a CFD based method, as opposed to potential flow, to seakeeping analysis is for calculation of prob-lems which contain strongly nonlinear phe-nomena such as breaking waves, large-amplitude ship motions and wake flows, etc. Besides the accuracy of physical modeling and computational results, the colourful post-processing of results and capability of simulat-ing strongly nonlinear free surface flows are appealing to researchers and engineers. Up-to-date numerical methods such as volume-of-fluid (VOF), level-set methods or particle methods provide reliable results even for the violent flow problem in which the topology of the free-surface boundary is largely distorted, fragmented and merged. Recent turbulence modeling such as RANS and LES become quite popular and they provide reasonable nu-merical results for an engineering purpose. The major difficulty in the numerical simulation of strongly nonlinear wave-body interaction prob-lems using a field equation solver is that a rigid body can move arbitrarily without coincidence of the grid lines and body boundary, so that

C. Hu et al.

(Kyushu

Univ.)

D.G.

Dommermut

h et al.

(SAIC)

J. Yang et al.

(Univ. of

Iowa)

P. Queutey et

al.

(ECN)

R. Löhner et

al. (George

Mason

Univ.)

H. Miyata et

al.

(Univ. of

Tokyo)

Y. Kim et al.

(Seoul

National

Univ.)

Discretization

for convective

term

CIP 3rd QUICK 3rd QUICK /

WENO

Improved

Gamma Galerkin QUICK MC Limiter

Body motion IBM

Particle

IBM

Triangle

panel

IBM

Triangle

panel

Mesh

Deformation ALE

Overlapping

Grid

IBM

Triangle

panel

Free surface THINC

(VOF) CLSVOF CLSVOF VOF VOF

Density

Function

(QUICK)

THINC

(VOF)

Remark

LES

LES

Ghost Fluid

Method

RANS

RANS

228228

some special treatment is required, such as re-meshing, moving mesh or embedded (overset) meshing techniques. Each scheme has its own strengths and weakness and recent studies clearly show a diversity of method applied with no significant dominance of any one numerical scheme. Furthermore, in spite of the improve-ment of computational resources, there are still doubts over the accuracy of CFD based meth-ods due to the sensitivity of the solution to grid spacing and time step size. For a three-dimensional full-scale ship calculation CFD methods still require very large computational effort, which limits their application as a prac-tical ship design tool.

Many computational results for ship mo-tions using CFD methods were produced in the last few years (refer to Table 6 for an overview of CFD methods used for seakeeping). Orihara and Miyata (2003) solved the ship motions problem in regular head wave conditions and evaluated the added resistance of a series of different bow-form for a medium-speed tanker in regular head waves using a CFD simulation method called WISDAM-X. The Reynolds-averaged Navier-Stokes (RANS) equations were solved by the finite-volume method with an overlapping grid system.

Figure 27. Overset grid system (Sadat-Hosseini

et al., 2013)

The group at Iowa University has led many research projects on the ship resistance prob-lem using CFD methods. Based on their past experience in CFD computations, their work has extended to manoeuvering and seakeeping problems in recent years. For example, Carrica et al. (2007) solved RANS equation with sin-gle-phase level set method for surface ships free to heave and pitch in regular head waves. The overset grid system which is shown in the Figure 27 was used for a rigid body movement. More recently, Sadat-Hosseini et al. (2013) validated CFD Ship-Iowa V4.5 for the ship motions and added resistance of KVLCC2 tanker advancing at Fn=0.142 with fixed and free surge in head waves.

Dommermuth et al. (2007) simulated break-

ing waves around ships and prescribed the mo-tion problem by Numerical Flow Analysis (NFA) code based on a combination of Carte-sian-grid methods and volume-of-fluid meth-ods. A ship hull was represented on a Cartesian grid by an immersed boundary generated from the panelled ship hull surface data. They used a Smagorinsky turbulence model, which is an LES scheme for computation of turbulence phenomena in the flow field, while a free slip

229229

boundary condition was adopted for the body boundary condition and an empirical model for shear stress was used for friction of body.

Hu and Kashiwagi (2007) developed a CFD-code named Research Institute for Ap-plied Mechanics, Computation Method for Ex-tremely Nonlinear hydrodynamics (RIAM-CMEN) which adopted a constrained interpola-tion profile (CIP) based Cartesian grid method. In the CIP-based formulation, the wave-body interaction problem is considered as a multi-phase problem. Different phases are recognized by a density function that has a definition simi-lar to the volume fraction function in the VOF method. To calculate the volume fraction of the solid phase, virtual particles were used. They compared the THINC scheme and the CIP scheme as an interface capturing method and showed the possibility that a CIP-based method could be applied to simulate strongly nonlinear wave-body interaction problems for modified Wigley models. Hu et al. (2008) conducted computation for green water effects in large amplitude ship motion of S-175 containership as shown in Figure 28.

Figure 28. S-175 containership advancing in

large amplitude head waves (Hu et al., 2008)

Visonneau et al. (2010) conducted analysis

for ship motion problems using their CFD pro-

gram called ISISCFD. This program used im-proved gamma differencing scheme for discre-tization of the convection term, and the RANS solver was applied to computation of the turbu-lence effect. One of the main characteristics of this program is using an unstructured hexahe-dral grid and an analytical weighting mesh de-formation approach for a moving body. This program was also validated by Guo et al. (2012) for calculating the added resistance of KVLCC2 in head waves.

Monroy et al. (2009) validated a spectral

wave explicit Navier-Stokes equation (SWENSE) method to solve the ship motion problem in irregular head waves. In the SWENSE method, incident wave terms are calculated by a potential flow model and dif-fracted wave fields are solved based on the RANSE equation under a structured body-fitted grid system. Due to the potential based theory, this program can have the capability for simu-lating ship motions in irregular waves. They carried out computation for heave and pitch motion in irregular waves using this approach.

Yang et al. (2013) simulated large-

amplitude ship motions by using a finite-volume based method on a non-uniform Carte-sian grid. Viscous effects were ignored and the wave-body interaction problem was considered as multi-phase problem with water, air, and solid. The volume fraction of a solid body em-bedded in a Cartesian grid system was calcu-lated by a level-set based algorithm and sys-tematic numerical simulations for Wigley III hull and S-175 containership in regular head waves were conducted.

Particle methods have also been applied to

wave-body interaction problems. Sueyoshi (2004) and Doring et al. (2004) conducted computations for motion analysis of two di-mensional floating bodies with a hole using a particle based method such as moving particle

230230

semi-implicit (MPS) and smoothed particle hydrodynamics (SPH). These efforts may be a useful foundation for damaged ship analysis.

Figure 29 shows some sample results of the pierced box case.

Figure 29. Pierced box test case (Doring, 2004)

As well as the above applications of pro-

prietary codes, there have also been applica-tions of open source and commercial CFD software to wave-body interaction problems. Moctar et al. (2010) calculated the ship mo-tions in regular head waves for λ/L=0.6, 1.1, and 1.6 by using Comet and OpenFOAM based on the RANS equations with finite-volume approach. Test ships were a containership (KCS) and an oil tanker (KVLCC2). Recently, the same group has continued to simulate vio-lent ship motion by using OpenFOAM and STAR-CCM+. The commercial software Star-CCM+ developed by CD-adapco is becoming popular and Kim et al. (2013) showed the CFD simulations of ringing response of a gravity based structure in extreme sea states using this technique.

A comparative study for various seakeeping tools was conducted by Bunnik et al. (2010). A container ship and a ferry were chosen for model ship. For the container ship, rigid body motions including hydrodynamic coefficients, added resistance, internal loads and relative vertical motions all calculated for 24.5 knots in head seas while for the ferry, rigid body mo-tions, internal loads and relative vertical mo-tions were compared for 25.0 knots in head seas. All the numerical results were compared with experimental data. In this comparative study, the participants based on CFD methods were as follows:

- ECN-CFD : CFD based method using RANS solver, “ISISCFD” (Ecole Centrale de Nantes)

- GL-CFD : CFD based method using un-structured FVM RANS solver, “COMET” (Germanischer Lloyd)

- KU-OU-CFD : CFD based method using CIP and THINC scheme, “RIAM-CMEN” (Kyushu University and Osaka University)

For these test models, there was no clear

advantage of any particular CFD based method compared with potential flow based methods, as long as there are no strong nonlinearities or viscous effects. Also, numerical codes using nominally the same method can produce differ-ent results meaning that the choice of numeri-cal scheme and the procedure of implementa-tion are both of critical importance for seakeep-ing problems.

Another comparative study of CFD meth-ods for seakeeping was conducted by Larsson et al. (2010). In this comparative study, the performance of various CFD based methods was compared. Although most test cases were for steady wave problems such as prediction of ship resistance, in some cases, the ship motions, added resistance and roll decay were compared with experimental data. Test cases were for the KVLCC2, KCS and DTMB 5415.

231231

Larsson et al. (2011) analyzed the results of the comparative study and pointed out that the number of grid points has an obvious effect on both motions and resistance results. The predic-tion error is around 16 %D (standard deviation) for 1st harmonic motion amplitude and the smallest error averaged over amplitudes and phase for motions is 2.66%D for CFDShip-Iowa with the largest number of grids, 4.73M grid points. A comprehensive analysis of all results is published in Larsson et al. (2014).

A detailed study of both steady and un-steady ship motions is considered in Simonsen et al. (2013), who compare experimental results for the KCS to CFD predictions using both Star CCM+ and CFDSHIP-IOWA and a potential flow method. Attention is paid to the uncer-tainty of both the measured and predicted quan-tities. Overall agreement of the CFD with the experimental data is good, with the steady-flow quantities better predicted than the unsteady motions. In waves, the mean resistance was accurately predicted by the CFD, but the ampli-tude of the resistance variation with time is underpredicted. This is consistent with other studies of the same phenomena using CFD.

A further comparison of the accuracy of CFD methods to predict added resistance in waves is found in Soding et al (2012) where a comparison to a potential flow Rankine Panel Method and experiments is made for a con-tainer ship advancing in head waves. Predic-tions from the CFD method are close to ex-perimental results in the long wave region, but less accurate in shorter waves.

An example of the application of an over-lapping grid method applied to large amplitude motions predicted using the Star CCM+ code is found in Peric and Schreck (2012), where cases of a free-fall lifeboat entering the free surface and the KRISO container ship advancing in oblique waves are addressed.

Although CFD based methods can be ap-plied to wave-body interaction problems, they generally require massive computational time and thus offer few advantages unless violent flows or highly nonlinearity are involved. Thus, many studies have focused on CFD computa-tion to simulate violent local flows rather than three-dimensional wave-body interaction prob-lems. Sueyoshi et al. (2005) have applied the MPS method for sloshing problem of a two dimensional tank. Nam and Kim (2006) intro-duced the application of SPH, and Kishev et al. (2006) have applied a CIP scheme for violent sloshing problems. Level-Set and SPH methods have been applied by Colicchio (2007) for flip-through phenomena during sloshing flows and compared with experimental results. Kim (2007) described experimental and numerical issues in sloshing analysis, and the comparison between the SPH and SURF schemes has been introduced. Wemmenhove et al. (2009) solved three-dimensional violent sloshing problems by using ComFLOW code. Typical results of fluid configuration are shown in Figure 30.

For the slamming problem, CFD methods

are not generally useful because the impact pressure is quite sensitive to grid resolution and time step. The water entry problem with impact occurrence is strongly nonlinear and regarded as a non-memory problem, where the impulsive pressure variation is involved in a similar man-ner to sloshing-induced impact. This problem has been tackled by using SPH. Good examples can be found in the work of Oger et al. (2006, 2007) which solved 2D and 3D water entry impact problems. Kim et al. (2007) also applied the SPH method for the water entry of wedges, and free surface evolutions have been com-pared with experimental results. Particularly, SPH has been applied for simulating both the non-cavity and cavity flows during impact. Recently, Oger et al. (2009) extended their SPH method to simulate hydroelastic impacts

232232

with strong fluid-structure coupling. An exam-ple of their results is shown in Figure 31.

Figure 30. Comparative study of sloshing simulation (ISOPE, 2009)

Figure 31. Visualisation of pressure field in water and Von Mises equivalent stress in structure at

various instants, Oger et al. (2009) 5. OVERVIEW OF SLOSHING EXPE-

RIMENTS 5.1. Introduction

Liquefied natural gas carriers (LNGCs) with capacities of 138,000–145,000 m3 were the most popular in the market from the 1970s to the 1990s. Starting in 2000, though, con-struction of larger LNGCs increased dramati-

cally, and LNGCs with capacities greater than 180,000 m3 appeared in the late 2000s (Figure 32). Although the capacity of LNG carriers has been increased dramatically, the size of the loads has remained nearly unchanged. Such unbalance can result in the significant increase in sloshing loads in liquefied gas tanks.

<SPH> <Exp.> <MPS> <VOF1> <Level Set> <CIP2>

233233

Figure 32. Recent trend of LNGC capacity

The two major concerns in sloshing prob-

lems are the prediction of impact loads and coupling with floating-body motion. The latter concern is related to the motion dynamics of ships or offshore structures, but the former is the main interest in LNG carrier design. De-spite many previous theoretical and computa-tional efforts to predict sloshing pressure, model scale testing is still considered as the most reliable approach for practical purposes. Analytic approaches cannot simulate violent

sloshing flows with strong nonlinear phenom-ena, and computational fluid dynamics (CFD)-based computation is not yet an appropriate tool to replace experimental methods. For this reason, in the last decade, highly systematic methodologies or concepts for the experimental assessment of sloshing loads have been studied (e.g., Graczyk et al., 2006; Kuo et al., 2010), and a few large experimental facilities have been built for practical model tests. Such large facilities with capacities of more than 3- or 4-tonne payloads were installed at GazTransport and Technigaz (GTT), Marintek, Pusan Na-tional University, and Seoul National Univer-sity (SNU) (Figure 33). In particular, very re-cently, a hexapod with a payload of more than 10 t was introduced by SNU. This trend is mostly due to the demand for larger-scale model tests, which implies that the importance of and interest in sloshing are increasing among not only naval architects but also ocean engi-neers.

(a) Marinrek (b) SNU Figure 33. Practical model-scale sloshing experiment (Marintek and SNU)

Many studies were conducted in the 1970s

and 1980s, which were mostly limited to small scale-model tests and/or 2D experiments, to understand the physics of sloshing phenomena and determine the magnitude of sloshing-induced impact pressure on LNG containment systems. Based on this foundation, larger-scale

and 3D experiments have become more popu-lar since the late 1990s and 2000s. Nowadays, the typical model scale of sloshing experiments for practical LNG carrier design is in the range of 1/60–1/40, and the 1/50 scale has become a sort of standard size for model tanks.

234234

Recently, high-performance data acquisi-tion and large data storage systems have al-lowed the capture of sloshing impact simula-tions with a high sampling rate. Many studies have been conducted based on an experimental approach (Lugni et al., 2006; He et al., 2009; Maillard and Brosset, 2009; Yung et al., 2009). A real-scale impact test was carried out at the Maritime Research Institute Netherlands (MARIN) (Brosset et al., 2009; Kaminski and Bogaert, 2009). Previous experimental studies were focused on sloshing phenomena and in-vestigation of the scale effect on sloshing. Many research activities were highlighted in the Sloshing Dynamics Symposium of the In-ternational Society of Offshore and Polar Engi-neers (ISOPE) conference. Very recently, an ISOPE sloshing benchmark test was carried out (Loysel et al., 2012), and the differences be-tween the experimental results of various ex-perimental facilities were observed.

In spite of the considerable efforts ex-pended in experimental analysis, there are many uncertainties in these sloshing experi-ments. Recently, Souto-Ielesias et al. (2011) discussed uncertainty analysis of the experi-mental setup. In terms of experimental instru-ments, Choi et al. (2010) tested two piezoelec-tric sensors and discussed the effects of thermal shock, sensing diameter, and improper mount-ing on the sloshing pressure. Pistani and Thiag-arajan (2012) thoroughly examined a motion platform, a pressure sensor, and a data acquisi-tion system and observed the characteristics of instruments. Except for those papers, it is diffi-cult to find studies on errors analysis of ex-perimental instruments.

In sloshing experiments, in addition to un-certainty, there are many technical barriers to the accurate measurement of impact pressure, e.g., the sensitivity of pressure sensors, scale effects, and appropriate media to simulate LNG-NG flows. Because there is no experi-mental technique on which everyone agrees organizations with large sloshing experimental facilities and classification societies have their own procedures for sloshing experiments. Some procedures or techniques are common, but there are some differences in the detailed methodology. However, it should be mentioned that while some procedures/techniques are common, it does not mean that they are the best or most appropriate. That is, there are still many uncertainties in sloshing experiments, which are not clear or validated. Therefore, it is not appropriate to develop or suggest a unified procedure for sloshing experiments at this time. Instead, the committee would like to summa-rize the current status of model-scale sloshing experiments and the guidance /recommendations of classification societies.

5.2. Sloshing Experiment: Overview

Figure 34 shows a typical schematic dia-gram of a measurement system for sloshing experiments. A motion platform, which is con-trolled by a motion controller, provides a model tank with six degrees of motion. Then, pressure sensors installed in the tank measure the dynamic pressure on the tank walls. A data acquisition system converts electric pressure signals into digital data. The acquired data is monitored in real time and saved to a data stor-age server.

235235

Figure 34. Schematic diagram of a measurement system for sloshing experiments

5.2.1. Motion Platform

In the model-scale sloshing experiment, a

motion excitation bed is essential to simulate the motion of the tank (i.e., motion of a ship or offshore structure). There are a few types of excitation bed. In the case of MARINTEK (Figure 33), a moving table with rotating axes is used to simulate motion. However, the most typical type is the hexapod-type platform shown in Figure 35. A hexapod platform com-prises six actuators that can move vertically and transversely. Linear actuators are typically equipped to minimize the time lag between the controller and the actuators.

There are about 10 facilities with hexapod platforms with payload capacities of 1~2 tonne. Such small platforms can be used for 1/100–1/60-scale tests for 3D model tanks and up to 1/50-scale tests for 2D models of typical LNG carriers or LNG floating production storage and off-loading (FPSO) facilities. For practical experiments, i.e., for predicting sloshing loads or the certifying classification societies, a 1/50–1/40 scale experiment should be carried out. In this case, a hexapod platform for a payload simulation of 2–6 tonne is needed. At present, only a few facilities have this capacity. In the case of GTT, a platform with a 6-ton capacity is being used. Very recently, SNU installed three motion platforms with payload capacities of 1.5, 5, and 14 tonne that can conduct ex-periments of up to 1/20 scale with a 3D model tank.

236236

(a) SNU (3 platforms of different sizes) (b) GTT (c) PNU Figure 35. Hexapod platforms for sloshing experiments (over 4-tonne dynamic payloads)

The greatest technical difficulty in the de-sign and fabrication of a large platform is the severe requirements of the motion characteris-tics. Since violent sloshing flows typically oc-cur in harsh environments, all the motion prop-erties, i.e., displacement, velocity, and accel-eration, must be large enough to simulate the severe motion responses of ships and offshore structures. Furthermore, the accuracy of motion signals should be carefully checked. The accu-racy of motion displacement and phase shift can be observed by using motion sensors such as optical sensing devices, accelerometers, and/or potentiometers. To this end, it is desir-able to use multiple sensing devices to cross-check accuracy. If the error in the motion am-plitude is larger than 3%–5%, the platform mo-tion sensors should be calibrated to increase their accuracy.

5.2.2. Model Tank

A model tank is generally made of acrylic so that the detailed flow can be visually observed. Figure 36 shows typical 2D and 3D models for sloshing experiments. The model tank should be water-tight and the wall surface should be very flat and smooth if there is no particular reason to make it rough, so that sloshing flow

is not affected by surface roughness. It is also important for the thickness of the acrylic layer to be sufficient to minimize the hydroelastic behavior of a model tank. When the wall thick-ness is not sufficient, the sloshing impact loads can cause hydroelastic vibration of a model tank, consequently resulting in unreliable measurement of pressure and flow.

Before an experiment with partial filling, it is desirable to carry out a hammering test. The results of the hammering test can be used to predict the natural frequency of tank wall vi-bration, and the period of this natural mode should be much smaller than the typical dura-tion of sloshing-induced impact pressure, so that the effect of hydroelastic vibration will not have any effect on the impact process.

When heavy gas is used in sloshing experi-ments in order to match the density ratio be-tween LNG and NG, rather than that between water and air, the model tank should be gas proof. It is very important to ensure that the heavy gas does not leak during the experiment. Heavy gas (SF6 is typically used) can be harm-ful to humans, so safety should be guaranteed during the experiment.

237237

(a) 2D tank (b) 3D tank

Figure 36. 2D and 3D model tanks

5.2.3. Pressure Sensor Pressure sensors can be the most important

of all experimental instruments. The motion platform can be calibrated by measuring the displacement of the input and output. The error of a data acquisition system is relatively lower than that of other instruments. A model tank can be the source of error, but that error can be minimized by the manufacturer. However, the error from the pressure sensors in the sloshing experiment has yet to be accurately estimated. Linearity, hysteresis, and resolution of a pres-sure sensor can be evaluated, and calibration can be performed using a reference sensor or an impact test in air. However, those cannot guarantee the accuracy of sloshing pressure, because sloshing impact occurs within a very short time, and the medium contacting the sen-sor suddenly changes from gas to liquid. The pressure sensors are typically not calibrated in that situation.

There are various types of pressure-sensing

technologies, such as piezoresistive, capacitive, electromagnetic, piezoelectric, optical, and potentiometric. For measurement of sloshing load, piezoelectric sensors are mainly applied, and pressure sensors from the three manufac-turers Kistler, Kulite, and PCB are popular as shown in Table 7. The sensors by Kulite are mainly piezoresistive (Kulite, 2004), while those of Kistler and PCB are mainly piezoelec-

tric. Many pressure sensors used in previous studies have small sensing diameters of about 2.5–5.5 mm. The pressure sensor should be small as possible and have a high natural fre-quency because large sloshing impacts occur in a very small region within a very short time. Moreover, the pressure sensor needs to be ca-pable of measuring in two-phase flows over a large pressure range.

Piezoresistive sensors are not affected by

temperature differences between the sensor and the medium. Furthermore, they are effective in measuring slowly varying pressure. However, piezoelectric sensors are regarded as a mature technology with outstanding inherent reliabil-ity. Piezoelectric materials typically have a high modulus of elasticity and thus nearly zero deflection and extremely high natural frequen-cies. Moreover, they have excellent linearity over a wide amplitude range. Therefore, piezo-electric sensors are appropriate for sloshing experiments. However, it is known that an ad-ditional signal can be generated when the sen-sor contacts a medium with a different tem-perature. This can be a problem when measur-ing sloshing pressure because there can be a temperature difference between the gas and the liquid. Therefore, this sensor is not effective for measuring static pressure, which produces a constant loss of electrons, resulting in signal drift.

238238

Piezoelectric sensors for sloshing experi-ments can be categorized into two types. The first is charge-mode-type sensors, which re-quire an amplifier to measure pressure signals. The second is integrated electronics piezoelec-tric (IEPE) or integrated circuit piezoelectric (ICP) sensors, which have an amplifier built into the sensor. The charge-mode-type sensor is good for high temperatures, and the sensitivity of the sensor can be changed. However, they take up a huge amount of space when a large number of measuring points are required. ICP

sensors have fixed sensitivity, but the measur-ing system is relatively simple. Therefore, ICP sensors are mainly used in many sloshing fa-cilities. In sloshing experiments, it has not yet been determined which type of pressure sensor is best to be used for measuring the sloshing impact pressure. The piezoelectric sensor is regarded as being better than the piezoresistive sensor for capturing impact pressure changes that occur within 1~10 ms.

Table 7. Main features of pressure sensors for sloshing experiments

Recently, Ahn et al. (2013) conducted a comparative study on several pressure sensors in sloshing experiments. They used one pie-zoresistive sensor and three piezoelectric sen-sors, including two ICP sensors, in 2D tank tests, and tested and compared the sensitivity to temperature differences between the sensors and the medium by exposing the sensors to hot

and cold water. Sloshing pressures during the regular and irregular motions were also meas-ured. Figure 37 shows an example of results from their comparative study.

Pressure measurement can be performed by using not only a single pressure sensor but also a cluster of sensors. Pressure sensors in 2 × 2, 3

Group Maker Model Diameter (mm) Reference

Ecole Centrale Marseille PCB 112A21 5.5 Loysel et al.

(2012)

Exxon Mobile Kulite XCL-8M-100-3.5BARA 2.6 Yung et al. (2009)

GTT PCB 112A21 5.5 Loysel et al. (2012)

MARINTEK Kulite ~2.5 Loysel et al. (2012)

Pusan National Univ. Kistler 211B5 5.5 Choi et al. (2010)

Seoul National Univ. Kistler 211B5 5.5 Kim et al. (2011)

Technical Univ. of Madrid Kulite XTL-190 ~2.5 Souto-Iglesias et

al. (2012) Univ. of Duisburg-

Essen Kulite XTM-190 3.8 Loysel et al. (2012)

Univ. of Rostock PCB M106B 11 Mehl and Schreier (2011)

Univ. of Western Australia Kulite XCL-8M-

100-3.5BARA 2.6 Pistani and Thiagarajan (2012)

239239

× 3, 4 × 4, or any other n × m combination can be installed to measure local pressure in a cer-tain area. Figure 38 shows two clusters sensors with 3 × 3 and 2 × 2 configurations. These can be used to analyze the spatial distribution of pressure and observe the averaged local pres-sure or force in the measured area.

(a) Piezoresistive and piezoelectric sensors

(b) ICP sensors Figure 37. Time histories of pressure signals

measured in a 2D tank under surge motion with 20% H filling (Ahn et al., 2013)

(a) Metal adaptors for a 3 × 3 cluster around a

corner

(b) Installed 2 × 2 cluster of sensors

Figure 38. Examples of cluster sensors

The following tests are recommended before the selection of pressure sensors for sloshing tests: - Slowly varying pressure test - Test of sensitivity to temperature differ-

ences between liquid and sensor - Test of sensitivity to the test medium, e.g.,

water or other liquid - Drift test for long measurement time - Motor noise test

Metal adaptors are commonly employed to increase the reliability of pressure measurement by pressure sensors. Bronze is the typical mate-rial for adaptors. This type of adaptor can give more reliable and stable pressure signals. Fur-thermore, it is very important to maintain the

240240

same temperature in the sensor and fluid. This can be achieved by exciting fluid motion for a certain time and allowing the temperatures of the contacted fluid and sensor surfaces to equalize.

5.2.4. Sampling Rate and Time Window

It is known that the sampling rate in slosh-

ing experiments should be high in order to cap-ture spikes in sloshing pressure. In general, it is agreed that 20 kHz or greater is acceptable for most sloshing experiments (Kim et al., 2012; Maillard et al., 2009; Ryu et al., 2009).

The size of the experimental time window is still under discussion. Since impact pressures occur randomly and the magnitudes of peak pressures are also random, the size of the time window can be a critical parameter in the sta-tistical analysis of impact loads. Thus far, a 5-h time window in real scale has been popular for irregular experiments, but recent studies have shown that this may be insufficient for practical LNG cargo containment system (CCS) design (e.g., Ahn et al., 2013). It is not yet clear what the optimum time window should be, but a minimum measurement time of 50 h has been recommended by SNU and a measurement time of 200 h been suggested by Bureau Veri-tas.

5.2.5. Test Conditions

For the prediction of design loads due to sloshing, the selection of the appropriate ocean (i.e., motion) condition is a critical element in sloshing experiments. It is strongly recom-mended to carry out prescreening tests to de-termine irregular wave conditions. However, in practice, such prescreening tests incur a large cost and require a long time. Therefore, the type and number of the prescreening tests should be carefully chosen. For the ocean con-ditions to be used for main experiments, re-

peated tests are strongly recommended. These repeated tests with different phases of wave components, i.e., motion components, are de-sirable to reduce the error or uncertainty of random signals.

When a prescreening test cannot be con-ducted owing to cost and/or time limitations, a typical set of conditions for sloshing experi-ments is listed in Table 8.

Table 8. Typical experimental conditions for

irregular motion (real scale) Test

condition Description

Filling levels 15%, 30%, 70%, and 95% of tank height

Ship speed 5 knots

Heading angles 150° and 90°

Sea states

Tz (modal period): 9.0 s and 11.0 s Hs (significant wave height) of 40-year return period for a 150° head-ing, and 1-year return period for a

90° heading

Measurement time 5 hours for each case

Test repetiti-ons At least 2 times

5.2.6. Measurement Area

It is obvious that sloshing pressure varies in space. Therefore, the pressure sensors should be installed in areas where largest impacts oc-cur. In general, large sloshing pressures are measured around the still-water level in low filling conditions and around the upper cham-ber or the tank top in high filling conditions (see Figures 39 and 40). Therefore, more sen-sors should be installed in these areas.

241241

In a practical experiment, e.g., for the de-sign of an LNG CCS, more sensors are better in order to cover more areas. In particular, for areas of high impact pressure, the installation of cluster sensors is highly desirable. It is also important to understand that the magnitudes of impact pressures can differ between the weather and lee sides; therefore, the locations of the sensors should be carefully chosen.

(a) 180° wave heading

(b) 90° wave heading

Figure 39. Sloshing impact areas (Pastoor et al., 2004)

Figure 40. Example of sensor locations for a 3D model

5.3. Statistical Analysis of Sloshing Impact Pressure

5.3.1. Peak Sampling

In statistical analysis, peak pressure signals need to be sampled for the entire pressure time history. Sampled sloshing peaks, or global peaks, are chosen by imposing a set of thresh-old pressure and sampling time windows (Fig-ure 41).

Figure 41. Methodology of peak sampling

Within a moving time window, the largest

peak signal is sampled as the global peak, and others are disregarded in the analysis. The maximum pressures collected from all the

242242

segments become a set of sampled peaks for statistical analysis. Therefore, the set of sam-pled data is dependent on the threshold pres-sure and the sampling time interval. The threshold pressure plays a key role in this se-lection process. However, the criteria for se-lecting these parameters have yet to be clearly defined. Therefore, the moving window size and the threshold are varied to determine the reliability of the results.

5.3.2. Peak Modeling

Sampled peak pressure signals can be mod-eled as simple triangular shapes, and thus, the characteristics of the peaks can be determined. Figure 42 shows an example of peak modeling and the main characteristics of a peak: peak pressure (

maxP ), rise time (riseT ), decay time (

decayT

), and total time (totalT ). Peak pressure is defined

as the maximum pressure value of the peak. However, definitions of rise time and decay time are different in many studies. According to existing studies and guidance notes from classification societies, rise time and decay time can be categorized as follows:

- Type 1: Absolute thresholding:

max thresholdrise up-crossingP PT t t= − , (19)

threshold maxdecay down-crossingP PT t t= − . (20)

- Type 2: Relative thresholding:

max rise max( )up-crossingrise

rise1P PT

t t α

α⋅=

−

−, (21)

decay max max

decay

( )down-crossing

decay1P P

Tt tα

α⋅

=−

−. (22)

where maxpt is the time when the peak pressure

maxP occurs; the subscript indicates the time when pressure becomes rise maxPα , decay maxPα . The up-crossing time is considered for the rise time and the down-crossing time is considered for the decay time. Type 1 thresholding applies the time when a certain absolute pressure is found, regardless of the peak value. Conversely, type 2 thresholding measures the rise and decay times at the instants when the pressure crosses the up and down percentages (100 α× ) of the peak pressure, respectively. This method, based on a relative-pressure concept, defines the times at which the rise and decay times should be measured. Table 9 presents the current mod-eling method used by test facilities and classifi-cation societies. These different peak modeling methods may predict different impact proper-ties.

Figure 42. Definition of characteristics of a

modeled sloshing peak

243243

Table 9. Current modeling methods used by test facilities and classification societies

Rise Time Decay Time ABS Type 2 (α = 0.5) Type 2 (α = 0.5)

DNV Type 1 Type 2 (α = 0.5)

Type 1 Type 2 (α = 0.5)

LR Type 2 (α = 0.5) Type 2 (α = 0.5) GTT Type 2 (α = 0.5) Type 2 (α = 0.5)

MARINTEK Type 2 (α = 0.2) Type 2 (α = 0.3) 5.3.3. Statistical Distribution

Two methods are popular for statistical

analysis for sloshing impact pressures: the three-parameter Weibull distribution and the generalized Pareto distribution. The cumulative probability functions of the two distributions take the following forms:

- Weibull distribution:

[ ]( )( ) 1 exp ( ) /F x xγ

δ β= − − − (23)

- Generalized Pareto distribution:

( ) 1/( ) 1 1 / cF x cx λ −= + + (24) In the Weibull distribution function, δ is the

location parameter, β is the scale parameter, and γ is the shape parameter. Here, x should be larger than the location parameter. To esti-mate these three parameters, the method of moments can be applied, which matches the first three model moments—mean, variance, and skewness—with their corresponding sam-ple moments. Figure 43 shows an example of a Weibull distribution fitted on sloshing impact pressure data. In the generalized Pareto distri-bution function, λ is the scale parameter and c is the shape parameter, both of which can also be estimated by using the method of moments

Figure 43. Example of Weilbull distribution of

sloshing impact pressure

6. COLLABORATION WITH ISSC 6.1. Collaboration with ISSC

The committee has liaised with ISSC, the Ocean Engineering (OE) Committee, and the Specialist Committee on Performance of Ships in Service. Particularly, the committee has been collaborating strongly with the Loads Commit-tee of ISSC. G. Hermanski plays an important role as the liaison of ITTC and ISSC.

244244

6.2. The First Joint ISSC/ITTC Internatio-nal Workshop

The first joint meeting of ITTC and ISSC

was held on 8th September, 2012, at Rostock, Germany, with the title of Uncertainty Model-ling for Ships and Offshore Structures (UM-SOS). (Figure 44) Two ITTC committees, Seakeeping Committee and Ocean Engineering Committee, participated and gave two plenary presentations. Also two ISSC Committees par-ticipated in the joint workshop. A panel session followed the plenary presentations and fruitful discussion was made among panellists and par-ticipants. A few ideas were proposed to strengthen the collaboration between ITTC and ISSC.

Figure 44. Flyer of 1st ITTC-ISSC Joint Work-

shop As a follow-up of this joint workshop, four

committees submitted technical papers to Ocean Engineering. Seakeeping Committee

submitted the paper titled “Uncertainties in Seakeeping Analysis and Related Load and Response Procedures”. Y. Kim and G. Har-manski contributed to complete this paper, and the paper was accepted for publication. 6.3. The Second Joint ITTC-ISSC Interna-

tional Workshop

The second joint workshop of ITTC and ISSC will be held in Copenhagen, as a part of ITTC Conference. Like the first joint workshop, the four committees, two of ITTC and two of ISSC, will contribute to the organization and presentation. Lloyd’s Register and Seoul Na-tional National University are supporting strongly the joint workshop, and DNV-GL and MARIN are also supporting the organzation. (Figure 45)

Figure 45. Flyer of 2nd ITTC-ISSC Joint

Workshop

245245

In this workshop, a benchmark test for mo-tion and loads on a containership will be car-ried out. The model ship is a real ship designed and tested in Korea. The benchmark test is a blind test, in which the participants do not know the results of experiment. Several repre-sentative results will be presented at the joint workshop. 7. ITTC RECOMMENDED

PROCEDURES 7.1. ITTC Procedure 7.5-02-07-02.1,

Seakeeping Experiments This procedure is well written and mature.

Therefore, no significant revision was consid-ered. There were proposed changes on sections of the regular and irregular wave sections. It was also proposed that blockage and depth issues should be reviewed. There are several figures without references. Additionally it was considered if there is a better way to look at uncertainty of random processes for the appen-dix of the procedure.

Based on these suggestions between mem-

bers, the sections for regular and irregular waves are revised. Also the appendix for uncer-tainty analysis is revised. The Seakeeping Committee unsuccessfully tried to find the source of Fig.3 - the original document men-tions about the ‘non published work’ of Fer-nandez. However, the committee members agreed that Fig.3 should be kept since it con-tains useful information.

7.2. ITTC Procedure 7.5-02-07-02.2, Predicting Power Increase in Irregular Waves from Model Experiments in Regular Waves

It was suggested that the biggest change in

procedure should be the inclusion of a section to address directional spectrum with short crested components. It was concluded that oth-er aspects of procedure would essentially re-main the same. There was a discussion with regards to applicability of various simulation efforts to calculate added resistance. The thought was whether there would be a future area of the procedure that might incorporate simulation combined with experimental results to determine added resistance. Based on this discussion, some sentences are revised, particu-larly for the wave spectrum.

7.3. ITTC Procedure 7.5-02-07-02.3, Expe-riments on Rarely Occurring Events

This procedure was discussed in the general

context as to how it should be approached. Ochi’s formulae had principally looked at slamming velocity. It was thought that bow flare and hull shape should also be an included factor. In the revision, the definition of slam-ming has been included.

In the future ABS, ISSC and other classifi-

cation rules should be reviewed for applicabil-ity to slamming and rarely occurring events.

7.4. ITTC Procedure 7.5-02-07-02.5, Verification and Validation of Linear and Weakly Non-Linear Seakeeping Computer

After the review of the procedure and the

papers of ITTC Seakeeping Workshop held in Seoul, no changes were recommended by the

246246

committee. However, there was an important comment that the current state of art shows that most authors do not include details of their V&V activities in publications other than straightforward comparison between experi-mental and computed data, be it RAOs, signal statistics, or direct time trace comparison. This issue should be considered for any future revi-sion.

In the 27th term, the committee could not

provide the final draft which includes the de-scription about hydroelasticity computation. The computation procedure for ship structural hydroelasticity can be included in the future or can be a separate procedure.

7.5. ITTC Procedure 7.5-02-07-02.6, Prediction of Global Wave Loads

This procedure was not revised in the 27th

term. However the committee discussed com-bining it with the computational procedure for ship hydroelasticity, but it was recommended not to combine with computational procedure at this stage. 7.6. ITTC Procedure 7.5-02-05-04, HSMV

Seakeeping Tests It was recommended to rewrite data acquisi-tion and data sampling rates. There were only a few paragraphs which need to be addressed, and the references needed to be included. Some revisions were made as follows: - References were included (there were none

in the previous version) - A paragraph on placement of ‘free to

pitch’ fitting for catamaran vessels was added

- A requirement to measure pitch inertia was added

- Planing craft testing was updated to in-clude a requirement to consider a appro-priate sample rate for human factors meas-urements

- Free-running model testing was updated to recognise that onboard digital storage is now possible and commonly used. The use of small inertial measurement units for ac-celerations/motion measurements was rec-ognised

- A minor comment was added on the diffi-culty of determining the number of wave encounters for planing craft where ‘skip-ping’ from wave crest to crest may occur

- The S175 was removed from the suggested benchmark/database of ship. This hull cannot be considered as an HSMV.

8. CONCLUSIONS

8.1. General Technical Conclusions

A few experimental facilities were newly introduced for seakeeping experiment and sloshing. Although numerical schemes are heavily being developed, the importance of seakeeping experiments is still evident through the need to validate numerical codes and to evaluate the seakeeping performance of uncon-ventional ships, e.g. high-speed vehicles and multi-hull ships. The demand to observe very nonlinear phenomena such as nonlinear wave-induced loads, slamming-whipping and green water, is also increasing. Generation of severe ocean environments and investigation of corre-sponding seakeeping performance is of interest, particularly for offshore structures.

Thanks to the increase of tank size in LNG carriers and offshore structures, the capacity of sloshing experimental facilities is getting big-ger. This trend makes it possible to observe larger scaled-model tests than ever. Experimen-

247247

tal skills to measure local impact loads have been developed, but there are many technical issues in order to utilise the pressure measured in a model tank for the design of a real-size tank of a ship or offshore structure. At the pre-sent stage, it is very desirable to develop an appropriate experimental procedure for model-scale tests and application to ship design. This technical demand is very strong nowadays, particularly for the design of safe LNG cargo tanks of large offshore structures such as FLNG and FSRU.

The Energy efficiency Design Index (EEDI) and Energy Efficiency Operation Index (EEOI) are critical issues for the shipping and ship-building industry. The procedures of estimating and verifying CO2 emission from ships are under intensive discussion at IMO/MEPC and ITTC should cooperate with the IMO. From the viewpoint of seakeeping, the most important parameter is power increase or speed loss in waves. For calculating EEDI, power increase or speed loss in an actual seaway has to be pre-dicted by model tests or theoretical calculations. There is a coefficient fw in the calculation of EEDI that describes the ratio of ship speed in waves and in wind to that in calm water. A reliable simulation procedure to compute fw is not yet available..

The most crucial element in the calculation of fw is to predict added resistance in waves. Besides towing-tank experiments, there are several computational methods, including slen-der-body theory, 3D panel methods, and CFD application for seakeeping analysis, and direct pressure integration, momentum conservation, and radiated energy methods for added resis-tance. To date, the most popular method in the shipbuilding field is the combination of slen-der-body theory and momentum conservation formula, specifically Maruo’s formula. How-ever, ascomputational resources continue to increase 3D panel methods become a strong

candidate to replace slender-body theory and the application of CFD is slowly increasing. For practical ship design, the prediction of added resistance in short waves is crucial. So far empirical formula, such as NMRI’s formula seem to be useful up to a certain level, but a practical method to consider nonlinear effects should be developed in the near future.

Ship structural hydroelasticity is an emerg-ing problem in the design of very large ships, such as ultra large containerships. Strip-based approaches combined with a modal approach have been popular in the past, but recent re-search has focused on the application of 3D panel methods combined with beam approxi-mations. Instead of a beam approximation, a whole ship FE analysis is also considered al-though it requires significantly larger computa-tional effort. Both towing-tank experiments using segmented models and numerical compu-tation are being used in recent years. Not only for springing but also slamming and resultant whipping are main topics of recent researches.

Seakeeping analysis based on frequency-domain formulation still represents the chosen approach when considering rapid evaluation of prototype designs. However, the popularity of time-domain methods for seakeeping analysis has increased in recent years. This trend is due to the advantages of the time-domain analysis in the extension to nonlinear motion and struc-tural loads, and coupling with external or inter-nal forces. Also the demand for the analysis of ship structural hydroelasticity including slam-ming and whipping makes the time-domain approach more popular. CFD application is in use in the field of seakeeping , but its robust-ness and computation efficiency are not yet to a mature state. However, the application of CFD programs, particularly commercial software, is rapidly increasing.

248248

8.2. Recommendation To The Full Confe-rence

Adopt the updated procedure No. 7.5-02-

07-02.1 Seakeeping Experiments.

Adopt the updated procedure No. 7.5-02-07-02.2 Prediction of Power Increase in Irregu-lar Waves from Model Tests.

Adopt the updated procedure No. 7.5-02-07-02.3 Experiments on Rarely Occurring Events.

No modification of the procedure No. 7.5-02-07-02.5 Verification and Validation of Lin-ear and Weakly Non-linear Seakeeping Com-puter Codes.

No revision of the new procedure No. 7.5-02-07-02.6 Global Loads Seakeeping Proce-dure.

Adopt the updated procedure No. 7.5-02-05-04 HSMV Seakeeping Tests.

8.3. Proposals For Future Work It is recommended that ITTC has a combi-

nation of pure technical committees and special committee(s) for external needs. ITTC has been a technical organization to create and update the procedures for experiments and computa-tion in the marine hydrodynamics field. In the 27th term, the role of ITTC was extended to provide professional comments to IMO and/or ISO and it is desirable that such external need is handled by a special committee(s) which takes charge of non-technical issues. By split-ting the committees and their roles, most ITTC committees can remain as pure technical com-mittees.

It is recommended to survey and/or collect benchmark data for seakeeping problems, such

as motions, loads, sloshing, slamming and full-scale measurements. The benchmark data can be very useful to validate the results of experi-ments and computation. In particular it is rec-ommended to collect the reliable benchmark data of added resistance. The prediction of added resistance is the key element of the pre-diction of the power increase in waves. To validate and understand the accuracy of com-putational codes, the reliable benchmark data is necessary.

It is recommended to write a new section for the V&V of ship hydroelasticity codes in the procedure 7.5-02-07-02.5, Verification and Validation of Linear and Weakly Non-linear Seakeeping Computer Codes. If it is too lengthy, it can be a separate procedure. It is recommended that the developed sec-tion/procedure is reviewed by the ISSC Loads and Responses Committee.

It is recommended to strengthen the col-laboration with ISSC committees, including, Loads and Responses and Environment Com-mittees. ITTC Seakeeping Committee and Ocean Engineering Committee, and ISSC Loads and Responses and Environment Com-mittees can share the information relating to nonlinear motion and structural loads and to understand the impact of projected changes in the sea wave environment and the influence the types of wave spectra have in seakeeping ex-periments. Where there is such overlap with these committees, then collaboration will be valuable. The collaboration can be achieved by the liaison(s) of the committees, but a new working group can be organized for more sys-tematic and active collaboration between ITTC and ISSC.

It is recommended to liaison with Propul-sion and Manoeuvring Committees for seakeeping/motion effects. When the ship mo-tion becomes large, the propulsion and ma-noeuvring performance can be influenced by

249249

motion effects. Also it is recommended to liaise with the Ship Stability in Waves Com-mittee for nonlinear ship motions and statistical analysis of large roll motions.

It is recommended to organize a special

committee for sloshing to create a procedure for sloshing model experiments. Due to the high demand of LNG in the world energy mar-ket, the construction of LNG carriers and LNG offshore platforms is increasing rapidly. Slosh-ing is a critical problem of LNG ships and off-shore platforms, and hence the number of sloshing experimental facilities has increased over the last decade. However, the procedure for sloshing experiments is not yet fully estab-lished. ITTC should create a general procedure for sloshing experiments, particularly focusing on model-scale tank test.

9. REFERENCES American Bureau of Shipping (ABS), 2006,

"Strength assessment of membrane-type LNG containment systems under sloshing loads", American Bureau of Shipping (ABS), Guidance Notes.

Advanced Marine Technology, 2010, “Devel-

opment of large scale sloshing experiment facilities”, AMEC Project Report, Seoul National University, Seoul, Korea.

Ahn, Y., Kim, J.H., Kim, S.Y., Kim, K.H.,

Kim, Y., 2012, “Particle Image Veloci-metry Measurement on the Oscillating Characteristics of Sloshing-induced Internal Flow Fields”, Proceedings Of the 6th Asia-Pacific Workshop on Marine Hydrodynam-ics, Johor, Malaysia.

Ahn, Y.J., Kim, K.H., Lee, S.W., Kim, Y., 2013, “Experimental Study on the Effects of Pressure Sensors and Time Window in Violent Sloshing Pressure Measurement”, 23rd ISOPE, Alaska, USA.

Ahn, Y., Kim, S.Y., Kim, K.H., Lee, S.W.,

Kim, Y., Park, J.J., 2012a, “Study on the Effect of Density Ratio of Liquid and Gas in Sloshing Experiment”, 22nd ISOPE, Rhodes, Greece.

Alaoui, A.E.M., Nême, A., Tassin, A., Jacques,

N., 2012, “Experimental study of coeffi-cients during vertical water entry of axi-symmetric rigid shapes at constant speeds”, Applied Ocean Research, Vol. 37, pp 183–197.

Amin, W., Davis, M., Thomas, G. Holloway,

D., 2012, “Analysis of Wave Induced Hull Vibrations using Continuous Wavelet Transforms”, Ocean Engineering, Vol. 58, pp 154–166.

Bardazzi, A., Lugni, C., Faltinsen, O.M.,

Graziani, G., 2012, “Hydroelastic study of the impact phenomena in sloshing flow”, Hydroelasticity in marine technology, To-kyo, Japan.

Begovic, E., Day, A.H., Incecik, A., 2011,

"Experimental Ship Motion and Load Measurements in Head and Beam Seas", Proceedings of the 9th HSMV Symposium, Naples, Italy.

Belenky, V. and Weems, K.M., 2012, “De-

pendence of roll and roll rate in nonlinear ship motions in following and stern quarter-ing seas”, Proceedings of the 11th Interna-tional Conference on the Stability of Ships and Ocean Vehicles, Athens, Greece.

250250

Bellezi, C.A., Cheng, L-Y., Nishimoto, K., 2013, “A Numerical Study of the Effects of Bow Shape on Green Water Phenomenon”, 23rd ISOPE, Alaska, USA.

Bennett, S.S., Hudson, D.A., Temarel, P.,

Price, W.G., 2012, “The influence of ab-normal waves on global wave-induced loads”, Hydroelasticity in marine technol-ogy, Tokyo, Japan.

Besten, J.H. den, Huijsmans, R.H.M., Vrede-

veldt, A., 2011, "Hydrodynamic impact of sandwich structures with vibration isolation and structural damping properties", Interna-tional Shipbuilding Progress, Vol. 58(1), pp 1-57.

Blendermann, W., 1994, “Parameter identifica-

tion of wind loads on ships”, Journal of Wind Engineering and Industrial Aerody-namics, Vol. 51, pp 339-351.

Bogaert, H., Brosset, L., Kaminski, M.L., 2010,

“Interaction between wave impacts and cor-rugations of MarkIII Containment System for LNG carriers: findings from the Sloshel project”, 20th ISOPE, Beijing, China.

Boom, H. van den, Hout, I. van der, Flikema,

M., 2008, “Speed-Power Performance of Ships during Trials and in Sevice”, SNAME Greek Section 2nd International Symposium on Ship Operations, Manage-ment & Economics, Athens, Greece, pp 1-7.

Bouscasse, B., Broglia, R., Stern, F., 2013,

“Experimental investigation of a fast cata-maran in head waves”, Ocean Engineering, Vol. 72, pp 318-330.

Buchner, B. and van den Berg, J., 2013, “Non-

linear Wave Reflection along the Side of Ships leading to Green Water on Deck”, PRADS 2013, Changwon, Korea.

Bunnik, T., Daalen, E. van, Kapsenberg, G., Shin, Y., Huijsmans, R., Deng, G., Del-hommeau, G., Kashiwagi, M., Beck, B., 2010, “A comparative study on state-of-the-art prediction tools for seakeeping”, 28th Symposium on Naval Hydrodynamics, California, USA, pp 1-13.

Bureau Veritas, 2011, “New Guidance Note NI

554 Design Sloshing Loads for LNG Mem-brane Tanks”, Bureau Veritas (BV), Classi-fication Notes.

Carrica, P.M., Wilson, R.V., Noack, R., Stern,

F., 2007, “Ship motions using single-phase level set with dynamic overset grids”, Computer & Fluids, Vol. 36, pp 1095–1111.

Chen, Z.Y., Ren, H.-L., Li, H., Zhang, K.-H.,

2012, "Experimental and numerical analy-sis of bow slamming and whipping in dif-ferent sea states", Journal of Ship Mechan-ics, Vol. 16(3), pp 246-253.

Chiu, F.C., Tiao, W.C., Guo, J., 2007, “Ex-

perimental study on the nonlinear pressure acting on a high-speed vessel in regular waves”, Journal of Marine Science and Technology, Vol. 12, pp 230-217.

Choi, H.I., Choi, Y.M., Kim, H.Y., Kwon,

S.H., Park, J.S., Lee, K.H., 2010, "A study on the characteristic of piezoelectric sensor in sloshing experiment", 20th ISOPE, Bei-jing China.

Chuang, Z.J. and Steen, S., 2013, “Speed loss

of a vessel sailing in oblique waves”, Ocean Engineering, Vol. 64, pp 88-99.

251251

Clauss, G.F. and Klein, M., 2011, “The New Year Wave in a seakeeping basin: Genera-tion, propagation, kinematics and dynam-ics”, Ocean Engineering, Vol. 38, pp 1624-1639.

Colicchio, G., Colagrossi, A., Lugni, C., Broc-

chini, M., Faltinsen, O.M., 2007, “Chal-lenges in the numerical investigation of the flip-through”, 9th International Conference in Numerical Ship Hydrodynamics, Michi-gan, USA.

Das, S. and Cheung, K.F., 2011, "Hydroelastic-

ity of Marine Vessels Advancing in a Sea-way", Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp. 192-200.

Datta, R., Fonseca, N., Guedes Soares, C.,

2013, “Analysis of the forward speed ef-fects on the radiation forces on a Fast Ferry”, Ocean Engineering, Vol. 60, pp 136-148.

Det Norske Veritas (DNV), 2006, "Sloshing

analysis of LNG membrane tanks", Det Norske Veritas (DNV), Classification Notes.

Det Norske Veritas (DNV), 2010, “Environ-

mental conditions and environmental loads”, Recommended Practice DNV-RP-C205.

Dommermuth, D.G., O'Shea, T.T., Wyatt,

D.C., Ratcliffe, T., Weymouth, G.D., Hendrikson, K.L., Yue, D.K.P., Sussman, M., Adams, P., Valenciano, M., 2007, “An Application of Cartesian-Grid and Volume-of-Fluid Methods to Numerical Ship Hy-drodynamics”, 9th International Conference in Numerical Ship Hydrodynamics, Michi-gan, USA.

Doring, M., Oger, G., Alessandrini, B., Ferrant, P., 2004, “SPH Simulations of Floating Bodies in Waves”, 19th IWWWFB, Cor-tona, Italy.

Du, S.X., Hudson, D., Price, W.G., Temarel, P.,

2012, “An investigation into the hydrody-namic analysis of vessels with a zero or forward speed”, Journal of Engineering for the Maritime Environment, Vol. 226(2), pp 83-102.

Faltinsen, O.M., Minsaas, K.J., Liapis, N.,

Skjørdal, S.O., 1980, “Prediction of resis-tance and propulsion of a ship in a seaway”, 13th Symposium on Naval Hydrodynamics, Tokyo, Japan, pp 505-529.

Faltinsen, O.M. and Sun, H., 2011, “Dynamic

motions of planing vessels in head seas”, Journal of Marine Science and Technology, Vol. 16, pp 168-180.

Fonseca, N., Ricardo, P., Guedes, S.C., Clauss,

G., Schmittner, C., 2010, “Numerical and experimental analysis of extreme wave in-duced vertical bending moments on a FPSO”, Applied Ocean Research, Vol. 32, pp 374-390.

French, B., Thomas, G., Davis, M., 2013,

“Slam Occurrences and Loads of a High-Speed Wave Piercer Catamaran in Irregular Seas”, Proceedings of the Institution of Me-chanical Engineers, Part M: Journal of En-gineering for the Maritime Environment, August, 2013.

French, B., Thomas, G., Davis, M.R., 2014,

“Slam Characteristics of a High-Speed Wave Piercing Catamaran in Irregular Waves”, International Journal of Maritime Engineering, Vol. 156 (1), pp A-25 - A-36.

252252

Fujii, H. and Takahashi, T., 1975, “Experimen-tal study on the resistance increase of a ship in regular oblique waves”, Proceedings of the 14th ITTC, Ottawa, Canada, pp 351-360.

Fujiwara, T. and Ueno, M., 2006, “Cruising

performance of a large passenger ship in heavy sea”, 16th ISOPE, San Francisco, USA.

Gavory, T., 2005, "Innovative tools open up

new prospects for liquid motion model tests", Proceedings of the Gastech, Bilbao, Spain.

Graczyk, M., Moan, T., Rognebakke, O., 2006,

“Probabilistic Analysis of Characteristic Pressure for LNG Tanks”, Journal of Off-shore Mechanics and Arctic Engineering, Vol. 128, pp 133-144.

Gran, S., 1981, “Statistical distributions of lo-

cal impact pressures in liquid sloshing”, Norwegian Maritime Research, Vol. 9(2), pp 2-12.

Greco, M., Bouscasse, B., Lugni, C., 2012, “3-

D seakeeping analysis with water on deck and slamming. Part 2: Experiments and physical investigation”, Journal of Fluids and Structures, Vol. 33, pp 148-179.

Grin, R., 2012, “On the Prediction of Wave

Added Resistance with Empirical Meth-ods”, Proceedings of 11th IMDC Confer-ence, Glasgow, UK, pp 313-324.

Guo, B.J. and Steen, S., 2010, “Experiment on

added resistance in short waves”, 28th Symposium on Naval Hydrodynamics, Pasadena, California.

Guo, B.J. and Steen, S., 2011, “Evaluation of added resistance of KVLCC2 in short waves”, Journal of Hydrodynamics, Vol. 23(6), pp 709-722.

Guo, B.J., Steen, S., Deng, G.B., 2012,

“Seakeeping prediction of KVLCC2 in head waves with RANS”, Applied Ocean Research, Vol. 35, pp 56-67.

Guo, L.-D., Sun, D.-P., Wu, H., 2012a, “A new

numerical wave flume combining the 0-1 type BEM and the VOF method”, Journal of Hydrodynamics, Vol. 24(4), pp 506-517.

Halswell, P.K., Wilson, P.A., Taunton, D.J.,

Austen, S., 2011, "Design and performance of inflatable boats: Flexibility and environ-mental considerations", Proceedings of the 9th HSMV Symposium, Naples, Italy.

Hashimoto, H., Umeda, N., Matsuda, A., 2011,

“Broaching prediction of a wave-piercing tumblehome vessel with twin screws and twin rudders”, Journal of Marine Science and Technology, Vol. 16, pp 448-461.

He, G. and Kashiwagi, M., 2012, “Numerical

analysis of the hydroelastic behavior of a vertical plate due to solitary waves”, Jour-nal of Marine Science and Technology, Vol. 17, pp 154-167.

He, G. and Kashiwagi, M., 2013, “Higher-

Order BEM for Radiation Forces of a Modified Wigley Hull with Forward Speed”, 23rd ISOPE, Anchorage, Alaska.

Hong, S.Y. and Nam, B.-W., 2011, “Analysis

of 2nd-Order Wave Force on Floating Bod-ies Using FEM in Time Domain”, Interna-tional Journal of Offshore and Polar Engi-neering, Vol. 21(1), pp 22-28.

253253

Hu, C. and Kashiwagi, M., 2007, “Numerical and Experimental Studies on Three-Dimensional Water on Deck with a Modi-fied Wigley Model”, 9th International Con-ference in Numerical Ship Hydrodynamics, Michigan, USA.

Hu, C., Kashiwagi, M., Sueyoshi, M., 2008,

“Improvement towards High-Resolution Computation on Strongly Nonlinear Wave-Induced Motions of anActual Ship”, 27th Symposium on Naval Hydrodynamics, Seoul, Korea.

Hughes, M.J. and Weems, K.M., 2011, “Time-

Domain Seakeeping Simulations for a High Speed Catamaran with an Active Ride Con-trol System”, Proceedings of the 11th Inter-national Conference on Fast Sea Transpor-tation, Hawaii, USA.

Ichinose, Y., Tsujimoto, M., Shiraishi, K.,

Sogihara, N., 2011, “Decrease of Ship Speed in Actual Seas of a Bulk Carrier in Full Load and Ballast Conditions”, Journal of the Japan Society of Naval Architects and Ocean Engineers, Vol. 15, pp 37-45.

Iijima, K., Itamura, N., Fujikubo, M., 2011,

"Comparison of Long-term Fatigue Dam-age in Bulk Carrier, VLCC and Container Carrier Subjected to Wave-induced Vibra-tions", Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp. 547-554.

IMO MEPC.1/Circ.796 12, 2012, “Interim

Guidelines for the Calculation of the Coef-ficient fw for Decrease in Ship Speed in a Representative Sea Condition for Trial Use”.

IMO MEPC64/4/15 29, 2012a, “ITTC Rec-

ommended Procedure 7.5-04-01-01.2,

"Speed/power trials, part 2, analysis of speed/power trial data"”.

IMO MEPC64/INF.6 29, 2012b, “Additional

information on ITTC Recommended Pro-cedure 7.5-04-01-01.2, "Speed/power trials, part 2, analysis of speed/power trial data"”.

Jacobi, G., Thomas, G., Davis, M.R., David-

son, G., 2014, “An insight into the slam-ming behaviour of large high-speed catama-rans through full-scale measurements” Journal of Marine Science and Technology, Vol. 19(1), pp 15-32.

Ji, Y.M., Shin, Y.S., Park, J.S., Hyun, J.M.,

2012, “Experiments on non-resonant slosh-ing in a rectangular tank with large ampli-tude lateral oscillation”, Ocean Engineer-ing, Vol. 50, pp 10-22.

Joncquez, S.A.G., 2009, “Second-Order Forces

and Moments Acting on Ships in Waves,” PhD Thesis, Technical University of Den-mark, Copenhagen, Denmark.

Kashiwagi, M. and Ohkusu, M., 1993, “Study

on the Wave-Induced Steady Force and Moment”, Journal of The Society of Naval Architects of Japan, Vol. 173, pp 185-194.

Kim, B., Ryu, M.C., Jung, J.H., Shin, Y.S.,

2010, “Identification of critical sea states for sloshing model tests”, Proceedings of the Society of Naval Architects & Marine Engineers, Washington, USA.

Kim, H.I., Kwon, S.H., Park, J.S., Lee, K.H.,

Jeon, S.S., Jung, J.H., Ryu, M.C., Hwang, Y.S., 2009, “An Experimental Investigation of Hydrodynamic Impact on 2-D LNGC Models”, 19th ISOPE, Osaka, Japan.

254254

Kim, J.-H., Kim, K.-H., Lee, D.-Y., Jung, B.-H., Kim, Y., 2013, "A Fully Coupled BEM-FEM Analysis on Ship Structural Hydroe-lasticity and Experimental Validation", PRADS 2013, Changwon, Korea, pp. 603-611.

Kim, J.-H., Kim, Y., Kim, Y., Kim, K.-H.,

2013a, "Study on Ship Structural Hydroe-lasticity and Fatigue Assessment in Irregu-lar Seaways", OMAE 2013, Nantes, France.

Kim, J.W., Tan, J.H.C., Magee, A., Wu, G.,

Paulson, S., Davies, B., 2013b, “Analysis of ringing response of a gravity based struc-ture in extreme sea states”, ASME 2013, Nantes, France.

Kim, K.-H., Bang, J.-S., Kim, J.-H., 2013c,

"Fully coupled BEM-FEM analysis for ship hydroelasticity in waves", Journal of Ma-rine Structures, Vol. 33, pp 71-99.

Kim, K.H. and Kim, Y., 2011, “Numerical

study on added resistance of ships by using a time-domain Rankine panel method”, Ocean Engineering, Vol. 38, pp 1357-1367.

Kim, S., Kim, C-Y., Cronin, D., 2013d, “Green

Water Impact Loads on Breakwaters of Large Container Carriers”, PRADS 2013, Changwon, Korea.

Kim, S.Y., Kim, K.H., Kim, Y., 2012, “Com-

parative Study on Model-Scale Sloshing Tests”, Journal of Marine Science and Technology, Vol. 17(1) pp 47-58.

Kim, S. and Sung, Y.J. 2012, “Numerical simu-

lations of maneuvering and dynamic stabil-ity of a containership in waves”, Proceed-ings of the 11th International Conference on the Stability of Ships and Ocean Vehi-cles, Athens, Greece.

Kim, T. and Kim, Y., 2011a, “Numerical analysis on floating-body motion responses in arbitrary bathymetry”, Ocean Engineer-ing, Vol. 62, pp 123-139.

Kim, Y., 2007, “Experimental and numerical

analyses of sloshing flows”, Journal of En-gineering Mathematics, Vol. 58, pp 191-210.

Kim, Y., Kim, S.Y., Yoo, W.J., 2010a, “Statis-

tical Evaluation of Local Impact Pressures in Sloshing”. 20th ISOPE, Beijing, China.

Kim, Y., Kim, S.Y., Kim, K.H., Chun, S.E.,

Suh, Y.S., Park, J.J., Hwang-Bo, S.M., Lee, Y.J., 2011, “Model-Scale Sloshing Tests for an Anti-Sloshing Floating Blanket System”, 21st ISOPE, Hawaii, USA.

Kim, Y., Kim, Y., Liu, Y., Yue, D.K.P., 2007,

“On the Water-Entry Impact Problem of Asymmetric Bodies”, 9th International Conference in Numerical Ship Hydrody-namics, Michigan, USA.

Kishev, Z.R., Hu, C., Kashiwagi, M., 2006,

“Numerical simulation of violent sloshing by a CIP-based method”, Journal of Marine Science and Technology, Vol. 11 pp 111-122.

Kitagawa,Y., Tanizawa,K., Tsukada,Y., 2014,

“Development of an Experimental Method-ology for Self-Propulsion Test with a MarineDiesel Engine Simulator, Third Re-port – Auxiliary Thruster System -”, Pro-ceeings of the 24th ISOPE conference, Vol.3, pp 691-696.

255255

Kjellberg, M., Contento, G., Janson, C.E., 2011, “A Nested Domains Technique for a Fully-Nonlinear Unsteady Three-Dimensional Boundary Element Method for Free-Surface Flows with Forward Speed”, 21st ISOPE, Hawaii, USA.

Koning, J. and Kapsenberg, G., 2012, “Full

scale container ship cross section loads - First results”, Hydroelasticity in marine technology, Tokyo, Japan.

Korobkin, A.A., 2011, “Semi-analytical ap-

proach in Generalized Wagner Model”, 26th IWWWFB, Athens, Greece.

Korobkin, A.A., 2013, “Wagner-type models

of water impact with separation for a finite wedge”, 28th IWWWFB, Marseille, France.

Kuo, J.F., Campbell, R.B., Ding, Z., Hoie,

S.M., Rinehart, A.J., SandstrÖm, R.E., Yung, T.W., Greer, M.N., Danaczko, M.A., 2009, “LNG Tank Sloshing Assessment Methodology – The New Generation”, 19th ISOPE, Osaka, Japan.

Kuroda, M., Tsujimoto, M, Fujiwara, T., Oh-

matsu, S., Takagi, K., 2008, “Investigation on Components of Added Resistance in Short Waves”, Journal of the Japan Society of Naval Atchtects and Ocean Engineers, Vol. 8, pp 141-146.

Kuroda, M., Tsujimoto, M., Sogihara, N.,

2012, “Onboard Measurement for a Con-tainer Ship in view of Container Load Con-dition”, Journal of the Japan Society of Na-val Atchtects and Ocean Engineers, Vol. 15, pp 29-35.

Larsson, L., Stern,F., Visonneau, M., 2010,

“Gothenburg 2010: A Workshop on Nu-

merical Ship Hydrodynamics”, Gothenburg, Sweden.

Larsson, L., Stern, F., Visonneau, M., 2011,

“CFD in Ship Hydrodynamics – Results of the Gothenburg 2010 Workshop”, MA-RINE 2011 – IV International Conference on Computational Methods in Marine En-gineering, Lisbon, Portugal.

Larsson, L., Stern, F., Visonneau, M., 2014,

“Numerical Ship Hydrodynamics – an as-sessment of the Gothenburg 2010 Work-shop”, Springer.

Lavroff, J., Davis, M., Holloway, D., Thomas,

G., 2013, “Wave Slamming Loads on Wave-Piercer Catamarans Operating at High-Speed Determined by Hydroelastic Segmented Model Experiments”, Marine Structures, Vol. 33, pp. 120-142.

Lloyd’s Register (LR), 2009, “Sloshing As-

sessment Guidance Document for Mem-brane Tank LNG Operations”, Lloyd’s Register (LR), Classification Notes.

Loysel, T., Chollet, S., Gervaise, E., Brosset,

L., De Seze, P.-E., 2012, "Results of the First Sloshing Model Test Benchmark", 22nd ISOPE, Rhodes, Greece.

Maillard, S. and Brosset, L., 2009, “Influence

of Density Ratio between Liquid and Gas on Sloshing Model Test Results”, 19th ISOPE, Osaka, Japan.

Maruo, H., 1960, “The drift of a body floating

on waves,” Journal of Ship Research, Vol. 4, pp 1-10.

Mathiesen, J., 1976, “Sloshing loads due to

random pitching”, Norwegian Maritime Re-search, Vol. 4(3), pp 2-13.

256256

Matsubara, S., Thomas, G., Davis, M.R., Hol-loway, D.S., Roberts, T., 2011, "Influence of Centrebow on Motions and Loads of High-Speed Catamarans", Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp 661-668.

Maximo, R., Ruggeri, F., Mueller, C., 2012,

“Parametric Design Tool Considering Inte-grated Seakeeping, Stability and Wave Re-sistance of a High Speed Trimaran Vessel”, Proceedings of the 8th International Con-ference on High-Performance Marine Vehi-cles (HIPER), Mülheim an der Ruhr, Ger-many.

Moctar, B., Kaufmann, J., Ley, J., Oberhage-

mann, J., Shigunov, V., Zorn, T., 2010, “Prediction of ship resistance and ship mo-tions using RANSE”, Gothenburg 2010: A workshop on CFD in Ship Hydrodynamics, Gothenburg, Sweden, pp 495-505.

Monroy, C., Ducrozet, G., Roux de Reilhac, P.,

Gentaz, L., Ferrant, P., Alessandrini, B., 2009, “RANS Simulations of Ship Motions in Regular and Irregular Head Seas using the SWENSE Method”, 19th ISOPE, Osaka, Japan.

Mortola, G., Incecik, A., Turan, O., Hirdaris, S.,

2011, “Large Amplitude Motions and Loads Using a Non-linear 2D Approach”, 26th IWWWFB, Athens, Greece.

Nam, B.W. and Kim, Y., 2006, “Simulation of

Two-Dimensional Sloshing Flows by SPH Method”, 16th ISOPE, San Francisco, Cali-fornia, USA.

Nan, X. and Vassalos, D., 2012, “A three di-mensional numerical study for ship motion in waves with forward speed”, Proceedings of the 11th International Marine Design Conference, Glasgow, UK.

Nippon Kaiji Kyokai (NK), 2010, “Guideline

for the Technical Appraisal of Ship Per-formance in Actural Seas”, Nippon Kaiji Kyokai (NK), Classification Notes, pp 1-27.

Nitin, R., Tam, T., Krish, T., Dominique, R.,

Robert, K.M., Timothy, F., 2010, “The Ef-fect of Sampling Rate on the Statistics of Impact Pressure”, OMAE 2010, Shanghai, China.

Ogawa, Y., Kitamura, O., Toyoda, M., 2012,

“A study for the statistical characteristic of slamming induced vibration of large con-tainer ship”, Hydroelasticity in Marine Technology, Tokyo, Japan.

Oger, G., Brosset, L., Guilcher, P.-M., Jacquin,

E., Deuff, J.-B, Le Touzé, D., 2009, “Simu-lations of Hydro-elastic Impacts Using a Parallel SPH Model”, 19th ISOPE, Osaka, Japan.

Oger, G., Doring, M., Alessandrini, B., Ferrant,

P., 2006, “Two-dimensional SPH Simula-tions of Wedge Water Entries”, Journal of Computational Physics, Vol. 213, pp 803-822.

Oger, G., Rousset, J.M., Le Touzē, D., Ales-

sandrini, B., Ferrant, P., 2007, “SPH simu-lations of 3-D slamming problems”, 9th In-ternational Conference in Numerical Ship Hydrodynamics, Michigan, USA.

257257

Ommani, B. and Faltinsen, O.M., 2011, “Study on Linear 3D Rankine Panel Method for Prediction of Semi-Displacement Vessels’ Hydrodynamic Characteristics at High Speed”, Proceedings of the 11th Interna-tional Conference on Fast Sea Transporta-tion, Hawaii, USA.

Orihara, H. and Miyata, H., 2003, “Evaluation

of added resistance in regular incident waves by computational fluid dynamics motion simulation using an overlapping grid system”, Journal of Marine Science and Technology, Vol. 8, pp 47-60.

Panciroli, R., 2012, “Water entry of flexible

wedges: Some issues on the FSI phenom-ena”, Applied Ocean Research Vol. 39, pp 72–74.

Paredes, R., Imas., L., 2011, "Application of

SPH in Fluid-Structure Interaction Prob-lems Involving Free-surface Hydrodynam-ics", Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp 200-208.

Pastoor, W., Tveitnes, T., Valsgård, S., Sele,

H.O., 2004, “Sloshing in Partially filled LNG Tanks – an Experimental Survey”, Offshore Technology Conference, Houston, Texas, USA.

Peric, M. and Schreck, E., 2012, “Simulation of

extreme motion of floating bodies using overlapping grids”, NuTTS2012: 15th Nu-merical Towing Tank Symposium, Cortona, Italy.

Piro, D.J. and Maki, K.J., 2011, "Hydroelastic

Wedge Entry and Exit", Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp 653-660.

Pistani, F., and Thiagarajan, K., 2010, "Set-up of a Sloshing Laboratory at the University of Western Australia", 20th ISOPE, Beijing, China.

Rahaman, M. and Akimoto, H., 2012, “Analy-

sis of the Mechanism of Slamming on the Bow Flare Region of a Container Ship Us-ing RaNS CFD Method”, 22nd ISOPE, Rhodes, Greece.

Rijkens, A.A.K., 2013, “Improving the sea

keeping behaviour of fast ships using a pro-active ride control system”, Proceedings of the 12th International Conference on Fast Sea Transportation, Amsterdam, Nether-lands.

Rijkens, A.A.K., Keuning, J.A., Huijsmans,

R.H.M., 2011, “A computational tool for the design of ride control systems for fast planing vessels” International Shipbuilding Progress, Vol. 58, pp 165–190.

Riley, M.R., Coats, T., Haupt, K., Jacobson,

D., 2011, “Ride Severity Index – A New Approach to Quantifying the Comparison of Acceleration Responses of High-Speed Craft”, Proceedings of the 11th Interna-tional Conference on Fast Sea Transporta-tion, Hawaii, USA.

Rose, C.J., Weil, C.R., Troesch, A.W., 2011,

“Planing Craft Acceleration Prediction Us-ing Vibro-Impact Methodology”, Proceed-ings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA.

Ryu, M.C., Jung, J.H., Jeon, S.S., Hwang, Y.S.,

Han, S.K., Kim, Y.S., Cho, T.I., Kwon, S.H., 2009, “Reference Load for a Conven-tional 138K CBM LNG Carrier in a Com-parative Approach”, OMAE 2009, Hono-lulu, Hawaii, USA.

258258

Sadat-Hosseini, H., Wu, P., Carrica, P.M., Kim, H., Toda, Y., Stern, F., 2013, “CFD verification and validation of added resis-tance and motions of KVLCC2 with fixed and free surge in short and long head waves”, Ocean Engineering, Vol. 59, pp 240-273.

Salvesen, N., Tuck, E.O., Faltinsen, O.M.,

1970, “Ship Motions and Sea Loads,” Transactions of Society of Naval Architects and Marine Engineers, Vol. 78, pp 250-279.

Senjanović, I., Hadžić, N., Bigot, F., 2013,

"Finite element formulation of different re-storing stiffness issues in the ship hydroe-lastic analysis and their influence on re-sponse", Ocean Engineering, Vol. 59, pp 198-213.

Senjanović, I., Hadžić, N., Tomić, M., 2011,

"Investigation of Restoring Stiffness in the Hydroelastic Analysis of Slender Marine Structures", Journal of Offshore Mechanics and Arctic Engineering, Vol. 133(3).

Seo, M.G., Lee, J.H., Yang, K.K., Kim, K.H.,

Kim, Y., 2013, “Analysis of added resis-tance: comparative study on different meth-odologies”, OMAE 2013, Nantes, France.

Seo, M.G., Park, D.M., Kim, K.H., Kim, Y.,

2012, “Computation on Added Resistance Based on Near-Field and Far-Field Meth-ods”, 22nd ISOPE, Rhodes, Greece.

Shao, Y-L. and Faltinsen, O.M., 2012, “A nu-

merical study of the second-order wave ex-citation of ship springing in infinite water depth”, Journal of Engineering for the Maritime Environment, Vol. 226(2), pp 103-119.

Simonsen, C.D., Otzen, J.F., Joncquez, S., Stern, F., 2014, “EFD and CFD for KCS heaving and pitching in regular head waves”, Journal of Marine Science and Technology, Vol. 18, pp 435-459.

Skejic, R. and Faltinsen, O.M., 2013, “Maneu-

vering Behavior of Ships in Irregular Waves”, OMAE 2013, Nantes, France.

Soding, H., Shigunov, V., Schellin, T.E., el

Moctar, O., Walter, S., 2012, “Computing added resistance in waves – Rankine panel method vs RANSE method”, NuTTS2012: 15th Numerical Towing Tank Symposium, Cortona, Italy.

Sogihara, N., Ueno, M., Fujiwara, T., Tsuji-

moto, M., Sasaki, N., 2011, “Onboard Measurement for Verification of a Calcula-tion Method on Decrease of Ship Speed for a RoRo Cargo Ship and an Oil Tanker”, 21st ISOPE, Hawaii, USA.

Sogihara, N., Ueno, M., Hoshino, K., Tsuji-

moto, M., Sasaki, N., 2010, “Verification of Calculation Method on Ship Performance by Onboard Measurement”, 20th ISOPE, Beijing, China.

Song, M-J., Kim, K.-H., Kim, Y., 2011, “Nu-

merical analysis and validation of weakly nonlinear ship motions and structural loads on a modern containership”, Ocean Engi-neering, Vol. 38(1), pp 77-87.

Specialist Committee on Performance of Ships

in Service 27th ITTC, 2012, “Speed and Power Trials Analysis of Speed/Power Trial Data”, ITTC – Recommended Procedures and Guidelines.

259259

Stenius, I., Rosén, A., Battley, M., Allen, T., Pehrson, P., 2011, "Hydroelastic Effects in Slamming Loaded Panels". Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA, pp. 644-652.

Sueyoshi, M., 2004, “Numerical Simulation of

Extreme Motions of a Floating Body by MPS Method”, Proceedings of the Joint In-ternational Conference of OCEANS’04 and TECHNO-OCEAN, Kobe, Japan.

Sueyoshi, M., Kishev, Z.R., Kashiwagi, M.,

2005, “A Particle Method for Impulsive Loads Caused by Violent Sloshing”, 20th IWWWFB, Spitsbergen, Norway.

Swidan, A., Amin, W., Ranmuthugala, D.,

Thomas, G., Penesis, I., 2013, “Numerical prediction of symmetric water impact loads on wedge shaped hull form using CFD”, World Journal of Mechanics, Vol. 3(8).

Tanizawa, K., Kitagawa, Y., Takimoto, T.,

Tsukada, Y., 2012, “Development of an Experimental Methodology for Self-Propulsion Test With a Marine Diesel En-gine Simulator”, International Journal of Offshore and Polar Engineering, Vol.23, No.3, pp. 197–204

Tello, M., Ribeiro e Silva, S., Guedes Soares,

C., 2011, “Seakeeping performance of fish-ing vessels in irregular waves”, Ocean En-gineering, Vol. 38(5-6), pp 763-773.

Thomas, G., Matsubara, S., Davis, M.R., French, B., Lavroff, J., Amin, W., 2012, “Lessons learnt through the design, con-struction and testing of a hydroelastic model for determining motions, loads and slamming behaviour in severe sea states”, Hydroelasticity in marine technology, To-kyo, Japan.

Thomas, G., Winkler, S., Davis, M.R., Hollo-

way, D., Matsubara, S., Lavroff, J., French, B., 2011, “Slam events of high-speed cata-marans in irregular waves”, Journal of Ma-rine Science and Technology, Vol. 16, pp 8-21.

Tiao, W.C., 2011, “Experimental investigation

of nonlinearities of ship responses in head waves”, Applied Ocean Research, Vol. 33, pp 60-68.

Tong, X.-W., Li, H., Ren, H.-L., 2013, “A hy-

brid approach applied to fast calculating the time-domain ship motions”, Journal of Ship Mechanics, Vol. 17(7), pp 756-762.

Tsujimoto, M., Kuroda, M., Shibata, K., Sogi-

hara, N., Takagi, K., 2009, “On a Calcula-tion of Decrease of Ship Speed in Actual Seas”, Journal of the Japan Society of Na-val Atchtects and Ocean Engineers, Vol. 9, pp 79-85.

Tsujimoto, M., Kuroda, M., Shiraishi, K., Ichi-

nose, Y., Sogihara, N., 2012, “Verification on the Resistance Test in Waves Using the Actual Sea Model Basin”, Journal of the Japan Society of Naval Atchtects and Ocean Engineers, Vol. 16, pp 33-39.

Tsujimoto, M., Sasaki, N., Takagi, K., 2011,

“On the Evaluation Method of Ship Per-formance for Blunt Ships”, Journal of the Japan Society of Naval Atchtects and Ocean Engineers, Vol. 15, pp 21-27.

260260

Tsujimoto, M., Sasaki, N., Takagi, K., 2012a, “On the Evaluation Method of Ship Per-formance for Blunt Ships - Extension of 10 mode Index for Ships”, Journal of the Japan Society of Naval Atchtects and Ocean En-gineers, Vol. 15, pp 21-27.

Tsujimoto, M., Shibata, K., Kuroda, M., Ta-

kagi, K., 2008, “A Practical Correction Method for Added Resisitance in Waves”, Journal of the Japan Society of Naval Atchtects and Ocean Engineers, Vol. 8, pp 147-154.

Visonneau, M., Queutey, P., Leroyer, A.,

Deng, G.B., Guilmineau, E., 2008, “Ship Motions in Moderate and Steep Waves with an Interface Capturing Method”, 8th ICHD, Nantes, France, pp.485–492.

Walree, F. van, 2012, “Development and vali-

dation of a time domain seakeeping code for a destroyer hull form operating in ex-treme sea states”, Proceedings of the 11th International Conference on the Stability of Ships and Ocean Vehicles, Athens, Greece.

Walree, F. van and de Jong, P., 2011, “Valida-

tion of a Time Domain Panel Code for High Speed Craft operating in Stern Quartering Seas”, Proceedings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA.

Walree, F. van and Turner, T., 2013, “Devel-

opment And Validation Of A Time Domain Panel Code For Prediction Of Hydrody-namic Loads On High Speed Craft”, Pro-ceedings of the 12th International Confer-ence on Fast Sea Transportation, Amster-dam, Netherlands.

Wang, J. and Xie, B., 2012, “A simplified method for predicting global motion of moored semi-submersible platforms”, 22nd ISOPE, Rhodes, Greece.

Wang, X.L., Gu, X.K., Hu, J.J., Xu, C., 2012,

“Investigation of sloshing an its effects on global responses of a large LNG carrier by experimental method”, Journal of Ship Me-chanics, Vol. 16(12), pp 1394-1401.

Wemmenhove, R., Iwanowski, B., Lefranc, M.,

Veldman, A.E.P., Luppes, R., Bunnik, T., 2009, “Simulation of Sloshing Dynamics in a Tank by an Improved Volume-of-Fluid Method”, 19th ISOPE, Osaka, Japan.

Weymouth, G.D. and Yue, D.K.P., 2013,

“Physics-Based Learning Models for Ship Hydrodynamics”, Journal of Ship Research, Vol. 57(1), pp 1-12.

White, N., Wang, Z., Lee, Y., 2012, "Guidance

Notes on Whipping and Springing Assess-ment", Proceedings of the 11th Interna-tional Marine Design Conference (IMDC 2012), Glasgow, UK.

Wu, M.K., Lehn, E., Moan, T., 2012, “Design

of segmented model for ship seakeeping tests with hydroelastic effects”, Hydroelas-ticity in marine technology, Tokyo ,Japan.

Wu, M.K. and Stambaugh, K., 2013, "Experi-

mental And Numerical Study Of Hydroe-lastic Responses In A High-Speed Vessel", Proceedings of the 12th International Con-ference on Fast Sea Transportation, Am-sterdam, Netherlands.

Xie, N. and Vassalos, D., 2012, “Evaluation of

the m-terms and 3D ship motions in waves”, Journal of Ship Mechanics, Vol. 16(9), pp 971-979.

261261

Xu, G., Duan, W.-Y., 2013, “Time-domain simulation of wave–structure interaction based on multi-transmitting formula cou-pled with damping zone method for radia-tion boundary condition”, Applied Ocean Research, Vol. 42, pp 136-143.

Yamada, Y., Takami, T., Oka, M., 2012, “Nu-

merical Study on the Slamming Impact of Wedge Shaped Obstacles considering Fluid-Structure Interaction (FSI)”, 22nd ISOPE, Rhodes, Greece.

Yan, S. and Ma, Q.W., 2011, “Fully Nonlinear

Hydrodynamic Interaction Between Two 3D Floating Structures in Close Proximity”, International Journal of Offshore and Polar Engineering, Vol. 21(3), pp 662-669.

Yang, J., Kim, S., Park, J.S., Jung, B-H., Lee

T-L., 2013, “Numerical Analysis for Slamming Impact Loads and Dynamic Structural Responses of a Containership” PRADS 2013, Changwon, Korea.

Yang, K.K., Nam, B.W., Lee, J.H., Kim, Y.,

2013a, “Numerical Analysis of Large-Amplitude Ship Motions Using FV-based Cartesian Grid Method”, International Journal of Offshore and Polar Engineering, Vol. 23(3), pp 186-196.

You, J. and Faltinsen, O.M., 2012, “A 3D Fully

Nonlinear Numerical Wave Tank with a Moored Floating Body in Shallow Water”, 22nd ISOPE, Rhodes, Greece.

Yu, L., Ma, N., Gu, X., 2012, “Ship dynamic

stability in rough seas”, Proceedings of the 11th International Conference on the Stabil-ity of Ships and Ocean Vehicles, Athens, Greece.

Yung, T.W., Ding, Z., He, H., Sandström, R.E., 2009, "LNG Sloshing: Characteristics and Scaling Laws," International Journal of Offshore and Polar Engineering, Vol. 19(4), pp 264-270.

Zaraphonitis, G., Grigoropoulos, G.J., Damala,

D.P., Mourkoyannis, D., 2011, “Seakeeping Analysis of a Medium-Speed Twin-Hull Containership”, Proceedings of the 11th In-ternational Conference on Fast Sea Trans-portation, Hawaii, USA.

Zhang, S., Weems, K., Lin, W.-M., 2011,

“Solving Nonlinear Wave-Body Interaction Problems with the Pre-Corrected Fast Fou-rier Transform (pFFT) Method”, Proceed-ings of the 11th International Conference on Fast Sea Transportation, Hawaii, USA.

Zhang, Y., Wang, X., Tang, Z., Wan, D., 2013,

“Numerical Simulation of Green Water In-cidents Based on Parallel MPS Method”, 23rd ISOPE, Anchorage, Alaska, USA.

Zhao, W.-H., Hu, Z.-Q., Yang, J.-M., Wei, Y.-

F., 2011, “Investigation on sloshing effects of tank liquid on the FLNG vessel re-sponses in frequency domain”, Journal of Ship Mechanics Vol.15(3), pp 227-237.

Zhao, X.Z. and Hu, C.H., 2012, “Numerical

and experimental study on a 2-D floating body under extreme wave conditions”, Ap-plied Ocean Research, Vol. 35, pp 1-13.

262262

1

Ocean Engineering Committee

Final Report and Recommendations to the 27th ITTC

1. GENERAL

1.1 Membership and Meetings The members of the Ocean Engineering

Committee of the 27th International Towing Tank Conference were as follows:

• Prof. Wei Qiu (Chairman), Memorial

University of Newfoundland, Canada • Mr. Halvor Lie (Secretary), MARINTEK,

Norway • Dr. Jean-Marc Rousset, Ecole Centrale de

Nantes, France • Dr. Dong-Yeon Lee, Samsung Ship Model

Basin, Korea • Prof. Sergio H. Sphaier, Federal University

of Rio de Janeiro, Brazil • Prof. Longbin Tao¸ University of

Newcastle upon Tyne, United Kingdom • Prof. Xuefeng Wang, Shanghai Jiao Tong

University, China • Dr. Takashi Mikami, Akishima Laboratory

(MITSUI ZOSEN) Inc., Japan • Dr. Viacheslav Magarovskii, Krylov

Shipbuilding Research Institute, Russia Four Committee meetings were held

respectively at:

• Samsung Heavy Industries, Geoje Shipyard, Korea, December 2011.

• MARINTEK, Trondheim, Norway, September 2012.

• Ecole Centrale de Nantes, France, June 2013.

• Shanghai Jiao Tong University, China, February 2014.

1.2 Tasks Based on the Recommendation of 26th ITTC

• Update the state-of-the-art for predicting

the behavior of bottom founded or stationary floating structures including moored and dynamically positioned ships emphasizing developments since the 2011 ITTC Conference. The committee report should include sections on: - The potential impact of new

technological developments on the ITTC - New experimental techniques and

extrapolation methods - New benchmark data - The practical applications of

computational methods to prediction and scaling

- The need for R&D for improving methods of model experiments, numerical modeling and full scale measurements.

263263

• Review ITTC Recommended Procedures relevant to ocean engineering - Identify any requirements for changes in

the light of current practice, and, if approved by the Advisory Council.

Identify the need for new procedures and outline the purpose and content of these.

• Complete the VIV and VIM guidelines and

benchmark study initialized by the Specialist Committee in Vortex Induced Vibrations of the 26th ITTC. The report on the benchmark test shall include clear definition of all the test parameters.

• Complete and report on the wave run-up benchmark study for a single cylinder.

• Carry out a wave run-up benchmark study for cases of four columns using the experimental data from MARINTEK.

• Investigate and report on the thruster-thruster interaction, ventilation and their scaling for DP systems.

• Investigate and report on physical and numerical modeling of vessels in side-by-side operations with an emphasis on wave elevation in the gap.

• Investigate and report on motions of large vessels and floating structures in shallow water.

• Jointly organize and participate in the joint ISSC/ITTC workshop on uncertainty in measurement and prediction of wave loads and responses.

1.3 Structure of the Report The work carried out by the Committee is

presented as follows:

2. State of the Art Reviews

• Section 2.1: Predicting the Behaviour of Stationary Floating Structures and Ships

• Section 2.2: Predicting the Behaviour of Dynamically Positioned Structures

• Section 2.3: Highly Nonlinear Effects on Ocean Structures

• Section 2.4: Predicting Vortex Induced Vibrations and Vortex Induced Motions

• Section 2.5: New Experimental Techniques • Section 2.6: New Extrapolation Methods • Section 2.7: Practical Applications of

Computational Methods to Prediction and Scaling

• Section 2.8: Improving Method of Model Experiments, Numerical Methods and Full-Scale Measurements

3. Review of Existing Procedures Section 3 reviews the procedures, 7.5-02-

07-03.1 Floating Offshore Platform Experiments, 7.5-02-07-03.2 Analysis Procedure for Model Tests in Regular Waves and 7.5-02-07-03.3 Model Tests on Tanker-Turret Systems, and addresses the need for new procedures.

New Documentation

• Section 4 discusses the development of

guideline for VIV and VIM model tests. • Section 5 presents numerical benchmark

studies of VIV.

264264

3

• Section 6 presents benchmark studies of wave run-up for cases of single and four columns.

• Section 7 discusses the investigation of thruster-thruster interaction, ventilation and their scaling effect for dynamic positioning (DP) systems.

• Section 8 presents the study of physical and numerical modeling of vessels in side-by-side operations.

• Section 9 discusses the motions of large ships and floating structures in shallow water.

• Section 10 summarizes the outcome of the first joint ISSC/ITTC workshop on uncertainties in measurement and prediction of wave loads and responses.

Conclusions and Recommendations Sections 11 and 12 present the conclusions

and recommendations, respectively.

2. STATE OF THE ART REVIEWS

2.1 Stationary Floating Structures and

Ships

2.1.1 Spar Platforms Majority of the new field developments

using Spar platforms have been in deepwater offshore regions. There are many technical challenges with deployment and operation in deep or ultra-deep water, typically including the design and construction of drilling and production facilities to withstand the harsh deepwater environment and regulatory issues that arise from operations at these depths.

Research has been carried out particularly

to address the global motions of spar hulls in

waves, current and wind including vortex-induced-motion (VIM). The wave and current interactions is also an important issue for the spar platform.

VIM of spars has been studied by many

researchers using numerical and experimental methods. Gonçalves et al. (2012) applied the Hilbert-Huang Transform Method to analyse VIM of a mono-column platform and showed a good agreement compared to that from the traditional analysis. Gonçalves et al. (2012a) presented an overview of relevant aspects on VIM of spars and mono-column platforms and showed that the loading condition had the largest impact on VIM responses because the low aspect ratio promotes large 3D effects on the vortex shedding.

Figure 2.1.1.1 Sketch of S-Spar (Sun and

Huang, 2012) Lefevre et al. (2013) presented CFD studies

on the VIM of a spar using STAR-CCM+. The numerical solutions were compared with model test results. Good agreement was found. Guidelines on computing spar motions, the use of turbulence models, mesh resolutions and the choice of time step were given for CFD simulations of spar VIM. Constantinides and

265265

Oakley (2013) simulated the VIM of a truss spar using AcuSolve. A cylindrical domain and a specialized boundary condition were used to avoid the creation of separate setups for each heading in the spar design phase.

A JIP has recently been set up to

specifically address VIM. The project started at summer 2013 and will be completed by summer 2016. MARIN and USP will carry out model tests as well as CFD computations. Comprehensive benchmark data will be produced from the model tests for the validation of numerical simulations.

Efforts have been made to investigate the

responses of spars in waves and current. Murray et al. (2012) conducted a model test campaign on a 1:50 Radial Wellbay Spar (RAW Spar) at the OTRC, Texas A&M University. The model test results were compared with numerical simulations by ABAQUS based a semi-empirical model by Muehlner et al. (2012b). Kurian et al. (2013) conducted experimental and numerical study on the truss spar subjected to long and short crested waves. The numerical solutions agreed well with the experimental results. Lower responses were also observed for short crested waves. Zhang et al. (2012) conducted model tests and investigated the added mass coefficients of a truss spar. In the model tests, the truss spar was subjected to uniform current. It was found that the added mass coefficients decrease with the reduced velocity increased. Hong et al. (2013) presented an experimental study on the motion of a 1/100 scaled model of a spar-type floating platform. The effect of test conditions, e.g., the center of gravity, mooring stiffness and the fairlead location were investigated. Rodriguez and Neves (2012) studied the nonlinear instabilities of spar platforms in waves with a focus on the parametric resonance phenomenon. Parametric Amplification Domains (PADs) were computed, showing the boundaries of the

instability regions and the maximum roll amplitudes.

New design concepts have also been

developed. Sun and Huang (2012) developed a new spar concept called "S-Spar" (Figure 2.1.1.1). The "S-Spar" combined the features of classic spars and truss spars. Numerical predictions were performed using the panel method. The new concept led to smaller wave forces and motions than those of the classic spars.

2.1.2 TLPs

In the past three years, research has been

carried out on TLPs using experimental and numerical methods focusing on motions and loads on tethers due to air-gap and wave impact on deck.

Some of the results of the Cooperative

Research on Extreme Seas and their impacT (CresT) JIP are presented and discussed in the work of Hagen (2011), Bitner-Gregersen (2011), Hennig et al. (2011), Forristal (2011), and Forristal and Aubult (2013) in the 30th and 32nd OMAE Conferences. Figure 2.1.2.1 shows the TLP model used in the CresT JIP.

Figure 2.1.2.1 The TLP Model Used in

CresT (Henning et al., 2011) Hagen (2011) discussed the wave

nonlinearities that might lead to unrealistically low estimates of the extreme tether tension for

266266

5

the TLP, especially those related to wave-in-deck events. The 100-year return period value was shown to increase considerably if the nonlinearities beyond the second order are included.

Bitner-Gregersen (2011) presented the

reliability assessment of air-gap of the TLP in extreme waves. The study showed the effects of wave nonlinearity, diffraction and radiation by the TLP, spatial variations of crest statistics, deck heights and sea water level variations. Based on a stochastic model, sensitivity studies were carried out to identify the importance of parameters on the probability of failure. Uncertainties related to the analyses were identified and ranked.

Hennig et al. (2011) reported some results

of extreme wave loads and responses observed in the model tests of the TLP. It was concluded that the wetted deck area, depending on the type of wave impact, wave-in-deck event and design variation, affect significantly the actual responses of the TLP. The effect of the wave short-crestedness on extreme loading was also assessed.

Based on the JIP experimental results,

Forristal (2011) showed that the maximum crest heights in an area are greater than those at a single point. This work also stated that Piterbarg’s theory (Piterbarg, 1996) can accurately predict this behavior. Model tests showed that the short-term statistics of the diffracted waves under a TLP have the same form as that of the incident waves. Based on these evidences, the author proposed a method for the calculation of air-gap. More recently, also in the context of the CresT JIP, Forristal and Aubult (2013) analyzed the effect of wave diffraction on the measured wave heights under the deck of the TLP. The results demonstrated that the first-order diffraction theory can be used to find the wave heights under the deck of the TLP, and it should also be used to correct

the wave measurements for a TLP. The second-order theory gave marginal improvements and is therefore not recommended.

Based on extensive model tests, Gaidai et

al. (2012) proposed another method for estimating the extreme value statistics of air-gap for a TLP subjected to random events. The method used only the area extreme value at each point to obtain a robust identification of the crossing rate function that determines the extreme value distribution. It was shown that this method can lead to an accurate prediction by using much less data in comparison with the conventional statistical procedure.

Johannessen (2011) investigated the high-

frequency loading and the response of a TLP in irregular steep waves. By comparing the model test results of tether loadings, it was concluded that the weakly nonlinear methods seem to be incapable of estimating the excitation at very high frequencies, while a much simpler impulse formulation gave a better estimate of horizontal excitations at these high frequencies.

Muehlner et al. (2012a) investigated the

effect of high-frequency oscillations of a TLP on the fatigue of its tendons by direct calculations in time domain. The coupled analysis of the TLP with tendons and risers was carried out by considering several nonlinearities, including large displacements, finite wave height, viscous drag, higher-order wave effects, and variable added mass of the TLP columns. The analysis results showed that the fatigue damage due to high-frequency oscillations in the tendons was significant.

Mansour et al. (2013) investigated the

design aspects of TLP tendon and tendon foundation systems. The study involved the numerical simulation of progressive failure of tendons in cyclonic events. The TLP responses during the transition from the restrained condition (TLP with all tendons) to the free-

267267

floating condition have been numerically simulated and verified against model test results. The paper presents two approaches concerning the design of TLPs. The first approach involves the design the TLP hull in case of progressive failure of the tendon system. The second approach concerns the design of a TLP Gravity Base Foundation (GBF) system so that the TLP is less reliant on the soil suction even in the survival condition.

Relatively new TLP concepts have also

been proposed and studied. Chandrasekaran et al. (2011) investigated a relatively new platform concept for ultra deepwater offshore exploration by using an experimental approach. The platform, Triceratops, combines the characteristics of TLP and spar, and consists of deck structure supported by three buoyant leg structures (BLS) connected through ball joints. Model tests in regular waves showed that the compliancy of ball joints affects significantly the responses and tether tensions. Only surge motions are transferred from the BLS to the deck. Figure 2.1.2.2 shows a typical triceratops.

Figure 2.1.2.2 The Triceratops Concept (Chandrasekaran et al., 2011) Rao et al. (2012) performed a

hydrodynamic analysis of a relatively new concept of a TLP, namely Tension Based Tension Leg Platform (TBTLP) (see Figure 2.1.2.3), which was proposed for much deeper water than the conventional TLPs. Time series of free vibration and response amplitude operators (RAOs) have been obtained and compared for three different cases of TLPs with and without tension base in various water depths. The efficacy of the provision of a tension base has been proved by comparing the RAOs.

More recently, Srinath and Chandrasekaran

(2013) investigated the influence of perforated members on the dynamic response of TLPs through model testing. Experiments in regular unidirectional waves showed that surge and pitch response amplitudes decrease in the presence of retrofitting perforated cylindrical members. Depending on the wave period, the reduction may vary from 4% to 25%.

268268

7

Figure 2.1.2.3 The TBTLP Concept (Rao et al., 2012)

Figure 2.1.2.4 The TLP with Perforated Members (Srinath and Chandrasekaran, 2013)

New installation methodologies regarding

TLP and its tendons have also been proposed. For tendon installation, Li et al. (2012) proposed an innovative approach, which involved the horizontal assembly of TLP tendon segments on a construction barge, instead of the typical vertical installation using

expensive heavy lift vessels. The partially assembled tendon is then incrementally pulled out through a stinger at the barge stern and secured with a holdback clamp so that the next tendon joint can be connected. The process repeats itself until the whole tendon is assembled and deployed. The tendon is then upended to a vertical configuration and connected to a TLP or a foundation pile.

Rijken (2013) provided two engineering

solutions for the installation of TLPs under swell conditions. These methodologies aim to reduce heave motions either by installing heave plates or by temporarily decreasing the waterplane area. Both methods reduce the heave RAOs when the wave period is greater than 12 s, and they may be applicable to situations where the installation window may contain prolonged periods of persistent swell.

In terms of hydrodynamic behavior of

TLPs, some interesting work has been published. Cruz et al. (2012) reported the parametric yaw motions of a TLP in close proximity to a moored FPSO. It was also observed in the experiments that as the TLP yaw motion amplifies, the TLP sway amplitudes reduce, revealing a strongly non-linear coupling between these modes. A nonlinear mathematical model that takes hydrodynamic interactions of two bodies and nonlinear restoring into account was also proposed for investigating the occurrence of parametric instabilities of this type of system. Rudman and Cleary (2013) applied the Smoothed Particle Hydrodynamics (SPH) method to the fully-coupled simulation of the impact of a highly non-linear breaking rogue wave on a moored semi-submersible tension leg platform. They showed the detailed effect of wave impact angle on the subsequent platform motions and determined how the cable tension varied with wave impact angle and time after impact. The application of the method and the presented results highlighted

269269

how the simulations could be used for practical design purposes and in assessment of operating conditions, especially in extreme wave conditions.

TLPs with direct and indirect applications

as offshore wind energy devices have also been investigated. Bachynski and Moan (2012) analyzed four TLP wind turbine (TLPWT) concepts using a fully-coupled nonlinear time-domain method and a linear frequency-domain method. The designs included a wide range of displacements. The wind-induced responses were found to be significant and dependent on the TLPWT hull design according to the nonlinear simulations. The nonlinear time-domain results for coupled wind and wave simulations indicated that wind loads were important for both operational and survival cases. In the operational cases, the operating turbines provided low-frequency excitation as well as some damping of the pitch motions. The wave-induced motions tended to become more important in more severe sea states. In the parked condition, the aerodynamic torque was found to be quite strong, and proved to be a critical force component for the smallest TLPWT design.

The Tension-Leg-Buoy (TLB), a concept

developed based on the TLP for offshore wind turbine applications, was investigated by Myhr and Nygaard (2012). They addressed the effects of the Excess Buoyancy (EB) and mooring lines layout. Other offshore wind energy applications related to TLPs can be found in Copple and Capanoglu (2012), Ren et al. (2012) and Stewart et. al (2012).

Bae and Kim (2013) presented an analysis

method for a mono-column TLP-type floating offshore wind turbine (FOWT) designed for 200m water depth. The proposed method integrates rotor dynamics and control, aero-dynamics, tower elasticity, floater dynamics, and mooring line dynamics to investigate the

full dynamic coupling among them along with sum-frequency wave-excitation effects in time domain. The sum-frequency wave loading effects can be significant in the coupled analysis when blades are fixed (not rotating) at minimal angle like the survival condition. Therefore, there are significant differences between uncoupled and coupled analyses, and care needs to be taken when applying the conventional dynamic analysis methods, which are typically used for floating offshore oil and gas platforms, to the design of FOWTs. There exist complicated coupling effects among blade rotation, tower flexibility, blade-control mechanism, platform and mooring dynamic characteristics, and they should be fully considered for effective and robust design of future FOWTs.

2.1.3 Semi-Submersibles

Semi-submersibles are a subject of

continuing interest studied by a number of authors using a variety of methods.

DaSilva and Knecht (2011) introduced the

practical implementation of a calculation methodology covering all environmental aspects that affect the airgap for semi-submersibles. The effects considered include the first-order vessel motions under an undisturbed wave field, diffracted wave elevation along the free surface under the platform, slow drift quadratic transfer function (QTF), vessel set-down and the heel effects due to mooring stiffness. The results showed good correlation between the model test results and the two analytical methods. The proposed analytical approach allows the designers to verify the airgap results in the early stages of the design.

VIM of semisubmersibles has been

addressed in various degrees. For example, Xu (2011) introduced a new semisubmersible design (NexGen) as a wet-tree floater which

270270

9

achieves significantly improved heave motion and vortex-induced-motion (VIM) performance through hull-form optimization while maintaining the simplicity of a conventional semisubmersible design. The difference between the NexGen semi-submersible design and a conventional semi-submersible design is in the blisters attached to the columns, the distribution of pontoon volume, and the pontoon cross-section shape. In the NexGen semi-submersible design, the pontoon volume is re-distributed to minimize heave loading while maintaining sufficient structural rigidity, a long heave natural period and adequate quayside buoyancy. The blisters attached to the columns effectively break the vortex shedding coherence along the column length and therefore suppress VIM.

Kyoung et al. (2013) conducted model tests

and numerical simulations to validate the Heave and VIM Suppressed (HVS) semisubmersible's global performance. The performance of the HVS semisubmersible was verified and validated. Xu et al. (2012) validated the HVS semisubmersible's VIM responses by model tests and CFD computations. Both the model test and the CFD analysis showed better performance of the HVS design than an equivalent conventional semisubmersible. Gonçalves et al. (2012) presented new experimental results on VIM of a large volume semisubmersible platform. The wave effects were the main focus. According to the results, regular and irregular waves led to considerable differences in responses. Bai et al. (2013) conducted model tests in a towing tank to study the VIM response of a Deep Draft Semisubmersible (DDS) with four rectangular columns and four pontoons. CFD computations based on RANS model were also carried out to investigate the problem. The experimental results showed that the VIM responses of the DDS mainly include horizontal motions (surge, sway and yaw), among which sway is dominant. The numerical results gave

confidence on the prediction of VIM using the CFD method.

Gonçalves et al. (2012) experimentally

studied the Vortex Induced Yaw (VIY) motion on a large volume semisubmersible platform. The yaw motion showed a resonant behavior with considerable amplitudes.

Mansour and Kumar (2013) presented the numerical results for the motion response of a Free Hanging Solid Ballast (FHSB) Semisubmersible in extreme hurricane. The new feature was proved to improve the performance of a conventional semisubmersible.

Figure 2.1.3.1 The Free Hanging Solid

Ballast (FHSB) Semi (Mansour and Kumar, 2013)

Kurian et al. (2012) conducted model test

on a moored semisubmersible with one failure mooring line. Results showed that the platform migrated to new mean position with a considerable transient response after the line failure.

Shan et al. (2012) conducted model tests

and studied the wave run-up phenomenon of

271271

column arrays. The leg spacing was found to be a factor that affects the wave run-ups.

2.1.4 FPSO Vessels

FPSO vessels have been operated in a

variety of water depth due to their flexibility, reliability, and low cost. Efforts have been made to investigate the responses of FPSO vessels in waves. Studies have been addressing low frequency motion, the effect of internal liquid cargo, and the hydrodynamic interaction when operated in a close proximity, the shallow water effects, and fully nonlinear analysis.

Minnebo et al. (2012) investigated the

response of FPSO systems subjected to squalls and developed a robust approach for estimating the design value. It was shown that the governing squall parameter concerning FPSO offset is the peak wind speed, both for spread moored and turret moored vessels. The method of dynamic amplification limitations showed great potential to be used as a Design Value Estimating method, as it combines the physical correctness of the squalls and the response characteristics of the FPSO system.

A procedure for selecting the best heading

for FPSOs in Santos basin considering the wave induced motions was introduced by Oliveira (2012), particularly for FPSOs with spread mooring pattern. A search algorithm has been implemented to enable the comparison of a large number of statistical results and to determine an adequate heading for the FPSOs. The optimum ranges concerning the roll motion, the vertical displacement and the vertical acceleration at the riser connection point don’t occur at the same heading. Considering the restraints, an approximation of the best range can be selected, allowing for some deviations.

Zhang et al. (2012) performed a study on SPM mooring system for side-by-side two vessels. A new side-by-side mooring bay designed by Keppel Offshore & Marine Technology Centre was investigated, and its global performance and dynamic stability were compared against those of the traditional SPM-mooring system. The multi-body systems include a SPM buoy with a turntable and mooring system, a VLCC FPSO, oil tanker, and the hawsers/fenders and yokes between them. The paper clearly showed that the newly-designed SPM mooring system experienced smaller relative motions between vessels and was more stable in the same environment compared to the traditional SPM mooring system.

Van’t Veer et al. (2012) introduced a

validated methodology to calculate the oscillatory loads on bilge keels of ships operating at zero forward speed in irregular sea states. To calculate these loads, the local relative fluid velocities acting normal to the bilge keel were combined with a KC dependent drag coefficient. The local relative velocity to the bilge keel was obtained from 3D potential flow calculation. The KC dependent drag coefficient of the bilge keel geometry was calculated by 2D CFD simulations in harmonic flow oscillations utilizing a rectangular fluid domain. With the present approach it is possible to quantify the ultimate load on the bilge keel in design extreme conditions and to obtain the long term load distribution necessary for fatigue analysis. Model tests for several FPSO vessels have been used to validate and calibrate the methodology.

272272

11

Figure 2.1.4.1 Bow connection details of

new concept SPM mooring system (Zhang et al., 2012)

2.1.5 Floating LNG Production Storage

and Offloading Vessels Kaminski and Bogaert (2010) presented the

progress made in the full-scale tests of real membrane containment systems subjected to the action of breaking waves, which were used to model the sloshing impacts in LNG tanks of LNG carriers or Floating LNG terminals (FLNGs). The waves were generated in a water flume using a wave focusing method. The tests were carried out through the Sloshel project. Their paper explains steps undertaken to improve the test repeatability, and to collect data for the analysis of scaling laws, hydro-structural interaction, and effects of membrane corrugations.

The motions and mooring loads of a turret

moored Floating Storage and Regasification Unit (FRSU) and an Liquefied Natural Gas Carrier (LNGC) including sloshing were studied by Cho et al. (2011). The turret moored FSRU weathervanes on a turret, and the side-by-side LNGC moves and interacts with FSRU. It was concluded that the longitudinal sloshing considerably affects the surge motion and mooring tensions. The paper showed that sloshing need to be considered simultaneously for the analysis of side-by-side moored FSRU and LNGC.

Kim et al. (2012) introduced improved

methods on offloading operability of side-by-side moored FLNG. The operational envelop of loading arm is a function of relative motion and wave drift force between two vessels. In the proposed methods, the concept 1 involved an articulated type reduction device with oil and spring as a damper in the cylinder which stroke is 0.15m (9.0m in real scale). This motion device can be installed at bow and stern of FLNG and LNGC to reduce the relative motion between FLNG and LNGC. The waves inside the gap are the main reasons for the sway drift forces in head seas. For the reduced gap wave between FLNG and LNG carrier, the wave absorber type device, concept 2, was designed. Using this device, a reduction of the second order drift forces can be expected. This device can be installed at side of FLNG between fenders. From the experimental study, it was found that the proposed motion reduction devices reduce the relative motion between two vessels significantly, and finally, improved offloading operability is expected.

2.2 Dynamically Positioned (DP) Structures

Xu et al. (2013) presented a new control

strategy considering roll-pitch motion control. Traditionally, DP systems only deal with horizontal motions without considering vertical ones including roll and pitch. However, large roll and pitch motions may be induced by thruster actions, which obviously should be avoided. The main idea in the new control strategy is to consider roll-pitch velocity and acceleration feedback in the horizontal control law in order to avoid large roll-pitch motions. The time-domain simulation results showed that the new control strategy can suppress roll and pitch motions. However, it will reduce the positioning accuracy in the horizontal-plane in some degree. Moreover, the energy

273273

consumption with the new control strategy was lower than that with the conventional one.

Smit et al. (2011) investigated to what

extent the current feed forward control improves the positioning performance of dynamically positioned FPSO vessels in varying currents. Tidal current reversals and so called ‘internal soliton’ currents are examples of varying currents, in which the current feed forward control is expected to improve the positioning performance. The DpSim software developed by MARIN was extended with a module containing the current feed forward control. When the current feed forward control was applied in DP in varying currents, the mean and standard deviation of the control point excursion were reduced. The heading performance and power usage did not change significantly while achieving this reduction.

For dynamically positioned crane barges

operating in a close proximity to FPSO during lifting operations, the hydrodynamic interactions are important and must be considered in the analysis. Tannuri et al. (2012) presented a large set of experimental tests considering a DP Barge in a close proximity to a FPSO. Results include the hydrodynamic interactions and their effect on DP performance in terms of station-keeping and thrust demand.

van Daalen et al. (2011) presented a generic

optimization algorithm for the allocation of dynamic positioning actuators, such as azimuthing thrusters and fixed thrusters. The algorithm is based on the well-known Lagrange multipliers method. In their work, the Lagrangian represents not only the cost function (the total power delivered by all actuators), but also all constraints related to thruster saturation and forbidden zones for azimuthing thrusters. The Newton-Raphson method was recommended to solve the thruster allocation problem. Depending on the configuration, it may lead to significant power

(energy) savings. An iterative process has also been studied by taking the limitations of actuators into account.

2.3 Highly Nonlinear Effects on Ocean Structures

2.3.1 Slamming

Slamming is a complex nonlinear problem.

It has been continuously studied by many researchers using experimental and numerical methods. In the numerical methods, methods based on the potential-flow theory, and CFD methods such as SPH, VOF and CIP, have been employed.

Figure 2.3.1.1 High-Speed Shock

Apparatus (Alaoui and Neme, 2012) Alaoui and Nême (2012) carried out an

experiment to study fluid-structure interactions during the slamming impacts. The vertical impact velocities were maintained constant by using a specially designed high-speed shock machine. Three rigid structures, including a cone, a square pyramid and a wedge-cone, were tested. Good repeatability of impact velocities and slamming loads was observed, and empirical formulas for dimensionless slamming coefficients were obtained. The predicted slamming coefficients of the cone by

274274

13

the ABAQUS/Explicit code are in good agreement with the experimental results.

Constantinescu et al. (2011) proposed three

numerical methods to study 2D slamming (wedge) and pseudo-3D slamming (cone) problems. The first method, Impact++ABAQUS, was based on Wagner's theory and the displacement potential formulation. The second one used the Arbitrary Lagrangian-Eulerian (ALE) analysis and a commercial finite element software program, ABAQUS/Explicit. The third method was based on the Coupled Eulerian-Lagrangian (CEL) approach and the VOF method. Experiments were also carried out by using a hydraulic shock machine for cones with varying deadrise angles.

Damblans et al. (2012) carried out model

tests to investigate the process of lowering a mud mat (plate with shirts) into water. The model tests were conducted by lowering a large scale mud mat with different porosities into calm water, regular wave, and irregular wave. Constant velocities were assured by using an electric jack. Impact loads were measured during the tests. Effects of porosity, impact velocity, and inclination angles of the plate on the impact coefficient were studied. Further, a numerical method, based on RANS with VOF for free surface capturing, was applied to simulate the slamming phenomena and to predict the impact loads.

Huera-Huarte et al. (2011) conducted a

series of experiments to study slamming forces for the water entry of a rigid flat plate. A novel test apparatus, named Slingshot Impact Testing System (SITS), was developed. The tests were conducted with high impact velocity up to 5 m/s and a wide range of deadrise angles from 0.3° to 25°. The cushion effect due to trapped air with small deadrise angles (<4°) was confirmed from the tests. An empirical formula

for non-dimensional impact coefficients was proposed in their work.

Figure 2.3.1.2 Slingshot Impact Testing

System (Huera-Huarte and Gharib, 2011) Jiang et al. (2012) validated two CFD

methods for slamming problems by comparing the predicted impact loads with experimental data. The two CFD methods were a RANS method with STAR-CCM+ and a Lagrangian-Eulerian Fluid-Structure Interactions (FSI) method with DYSMAS. The pressure peak, the pressure time history, and the pressure-area relationship were investigated, and reasonable agreements between numerical predictions and experimental results were reported.

Korobkin and Khabakhpasheva (2013)

investigated the effect of water depth on the first peak of bending stresses during the entry of an elastic body into water. Wagner's model was applied to solve deepwater impact while the leading-order solution was presented for shallow water impact. Two typical shapes, including a wedge with small deadrise angle and a cylindrical shell of elastic structures were considered. Computed bending stresses were compared to experimental data, and good agreement was observed.

275275

Nuffel et al. (2011) conducted free-drop

tests to study water entry of a rigid cylinder into calm water. Improvements were made on the accuracy and the repeatability of the pressure measurements. In their study, pressures and accelerations were recorded and further investigated. Parametric studies were carried out to examine the effects of sensor mounting, data sampling rate, temperature shock, the object surface conditions, and the water surface conditions on the measured pressure peak. Recommendations for experimental set-ups were provided. In 2012, they continued the study with the same apparatus (Nuffel et al., 2012). Global forces were also measured in the tests. Relationships for pressure-speed and force-speed were investigated.

To investigate the slamming load

distribution and its relationship with the impact velocity, Peng et al. (2011) conducted a free drop slamming test with a scaled trimaran model. Pressures were measured at the main hull, the side hull, and the cross structure of the trimaran. Comparisons were made between the experimental data and the simulation results based on the finite element method.

Rahaman and Akimoto (2012) studied the

slamming at the bow flare region by using a RANS-based CFD method. They investigated the pitch and heave motion as well as the pressure distribution at the bow flare region of a 3D container ship model travelling in regular head waves. Two dimensional simulations were carried out for selected bow flare sections based on the VOF method and FLUENT. Predicted slamming loads agreed reasonably well with the experimental data by Zhao et al. (1996).

Stansberg et al. (2012) conducted

experiments to investigate the breaking-wave induced slamming loads on vertical offshore structures. Time series of wave elevation,

pressure distribution, and the integrated slamming loads were measured in the experiments. An empirical formula for slamming coefficient was also presented.

Veen and Gourlay (2011) conducted

parametric studies to investigate the effects of sectional shape and time-varying impact velocity on slamming loads by using the 2D SPH method. They used three section shapes, including a wedge, a bow flare, and a catamaran, in their studies. Veen and Gourlay (2012) further carried out numerical studies on 2-D bottom slamming and bow flare slamming problems. In their work, the solid wall boundary conditions were modelled by using the ghost particle technique. The numerical method was applied to the free drop impact with prescribed velocity profile, and the numerical results were compared with the experimental data (Aarsnes, 1996). Furthermore, a linear strip-theory code was combined with the SPH algorithm to compute the impact loads on hull sections. The numerical solutions were also compared to the experimental data from Ochi (1958).

Vepa et al. (2011) carried out comparative

studies of slamming loads on cylindrical structures with three methods: a mesh-based implicit method, a mesh-based explicit method, and the SPH method combined with the finite element (FE) model. The explicit method and the SPH-FE method were applied by using LS-DYNA while the implicit method was applied by using FLUENT-ABAQUS. Rigid and deformable cylinders were included in the computations. A significant pressure peak reduction was observed in the deformable cylinder cases. They also concluded that the SPH method had better convergence than the mesh-based methods.

Wang and Soares (2013) investigated the

2D water entry of a bow flare section and the effects of roll angle on slamming loads by

276276

15

applying an explicit finite element method in combination with a multi-material Eulerian formulation and a penalty coupling method. In the previous work by Wang et al. (2012), computations were carried out by using LS-DYNA/Explicit. The predicted slamming load and pressure were compared with experimental data by Aarsnes (1996) and the numerical results by using other methods, including BEM and CIP.

Yamada et al. (2012) applied an explicit

finite element code LS-DYNA to simulate the slamming loads of water-entry of a full-scale wedge. They first validated the method with a small-scale rigid wedge and an elastic cylinder. Further, the method was employed to investigate slamming of a full-scale elastic-plastic wedge. Results from numerical simulations were compared to those by the conventional Wagner method.

Yan and Liu (2011) proposed a fully

nonlinear numerical method based on the potential flow theory to investigate water entry of axisymmetric bodies. The method was based on an axisymmetric linear boundary element method (BEM) and a mixed Eulerian-Lagrangian (MEL) approach. A jet cutting technique, which was effective and robust, was developed. They applied the numerical method to study the effect of gravity and body geometry as well as the flow separation location on the continuous body surface. Two representative bodies, an inverted cone and a sphere, were included in the study. They concluded a formula with a single similarity parameter for evaluating the contribution of the gravity of the total impact force on the cones. For the sphere impact, they observed the gravity effect was unimportant in the initial stage of impact, but slightly increased the impact pressure in the later stage when Froude number is less than 2.0. The flow separation location remained at a fixed location at the

angle of 62.5 deg when Froude number is larger than 1.0.

Yang and Qiu (2012) continued their

studies of 2D and 3D slamming problems based on a constrained interpolation profile (CIP) method. The compressible air was considered in the simulations. Validation studies of the numerical methods were carried out for 2D wedges with large and small deadrise angles, a 3D catamaran, and 3D cylinders. Numerical results were compared with solutions by other numerical methods and experimental data.

2.3.2 Sloshing

The problem of the sloshing of liquid cargo

in tanks is especially important in the case of Liquefied Natural Gas (LNG). The liquid is stored at atmospheric pressure in insulated tanks at -161° Celsius. Due to the insulation system, tanks cannot be partitioned. As a result, important liquid motions in the tanks, excited by the vessel motions, may be observed. The design of LNG ships or storage units is thus very complex. The state-of-the-art methodology is based on the use of seakeeping computer codes to estimate ship or platform motions. Experiments on tank models and CFD simulations have been performed in order to estimate global and local fluid loadings in the tanks.

Experimental assessments of many

parameters affecting the fluid motion and pressure are presented in Loysel et al. (2012) and (2013). These results were acquired during two rounds of Sloshing Model Test (SMT) benchmark studies.

The first round of benchmark studies,

involving nine participants, were based on a simple tank geometry (2D rectangular tank with clear water), 14 different excitation conditions, and a measurement setup. It aimed

277277

to compare the laboratory measurements. From the comparison of experimental data, some preliminary conclusions were given: the repeatability of single impact waves seems to be acceptable, however notable discrepancies in event rates and probabilities of pressure exceedance were clearly observed for harmonic and irregular waves. These differences led to the next round of benchmark studies in 2012-2013.

The focus in the second round was on the

accurate control of three parameters: the water filling level, the positioning of the tank and the rig motion. Many of motion rigs were hexapods (Loysel et al., 2013, Baudin et al. 2013). The results for single wave impacts with large gas pockets showed good agreement. This resulted in considering this setup as a reference configuration to validate methodologies. Differences can be found when the impact location, the gas pocket location and its size are not be accurately controlled. Discrepancies in the results for irregular motions still existed and they are comparable to those in the precedent benchmark studies. Temperature effects were highlighted and further investigations regarding this aspect were proposed.

Figure 2.3.2.1: Representative Pressure

Time Histories by Six Participants for A Single Impact Wave (Loysel et al., 2013)

Souto-Iglesias et al. (2011) presented a description of an experimental setup for sloshing tests involving angular harmonic motions. Details on data acquisition and synchronisation schemes were given. An uncertainty analysis was presented, focusing on the measurements of the first peak pressure.

Pistani and Thiagarajn (2012) carried out

sloshing tests using a hexapod with two 2D model tanks. The maximum pressures were measured for 1-DOF motions. An analysis of the experimental setup was presented in their paper. A thermal artefact on the pressure transducers was observed when the water hit their sensitive surfaces. This effect was also reported by Loysel (2012). They also checked the motions of the excitation rig. The steps of the data collection and analysis, as well as corrections to experimental shortcomings, were described.

Figure 2.3.2.2 Full-Scale Air Pocket Impact

on MarkIII (Kaminski and Bogaert, 2010) Hydroelasticity in sloshing experiments

was studied by Choi et al. (2012) using a hexapod and rectangular tank models. Surge motions for four different filling levels were tested using a regular rigid tank and a tank with a flexible stainless steel wall. The experiments

278278

17

confirmed the impact pressure are higher in the case of the flexible wall.

Figure 2.3.2.3 Rigid and Flexible Models

and Locations of Pressure Transducers (Choi et al., 2012)

Lugni et al (2013) presented similar

findings from their hydroelasticity experiments on a flexible tank wall. They also investigated the pressure effects on air cavities formed by the impact waves.

Wang et al. (2012) developed a new reliability-based methodology for the sloshing assessment of a membrane-type LNG cargo containment system (CCS) in LNGCs and FLNGs. For each individual sloshing impact event, the dependency of two parameters (the magnitude and the rising time) was taken into account in this new methodology. Based on sloshing model test data, the equivalent static pressure for each individual impact event was calculated using the magnitude and the dynamic capacity factor (DCF) through the associated rising time. In the reliability analysis, the limit state function was used by applying the Load and Resistance Factor Design (LRFD) approach. Then, the sloshing load (equivalent static pressure) and structural resistance (static capacity) distributions were employed to determine the partial safety factors in the CCS design formula at a target reliability index.

Fillon et al. (2013) applied the extreme values theory to sloshing pressure samples in

order to improve the model for predicting sloshing pressure maxima. Two different statistical fitting methods were used for sloshing pressure measurements in one sea state which is equivalent to a 480-hour duration sloshing test at full scale. This long duration sloshing test was in fact generated by using 96 five-hour individual experimental tests. The two methods led to a correct estimation of the maximum sloshing pressure. Graczyk et al. (2012) investigated local pressure effects based on low filling level tests of a 2D LNG tank model (scale 1:35). The tank has a smooth wall surface and a wall with horizontal protrusions similar to Invar edges that disturb the local flows, inducing either pressure amplifications or cancellations. The authors indicated the need of advanced instrumentation in combination with high-speed cameras to explain the measurements of local pressure.

Using the same motion rig, Bouscasse et al.

(2013) measured the free surface elevations in a 2D rectangular swaying tank. The experimental data were used to check the weakly-compressibility assumption in the SPH simulations.

Clauss et al. (2012) presented the

comparison of model test results for a moored LNG model (scale 1:100) and numerical solutions. The study focused on the water motions within the prismatic tanks (30% filled) in beam seas.

Flow velocity measurements were

performed by Ji et al. (2012) using a PIV system in a small swaying rectangular tank excited by a crank motorised arm. The harmonic motions in non-resonant conditions were compared to the published results, and the velocity flow fields were processed to obtain the properties of travelling waves. Water run-up and run-down on walls were analysed with respect to the flow regimes.

279279

Progress has been made to investigate the

scale effect using model and full scale tests. Lafeber et al. (2012) reported the scale effect on wave impact using an instrumented wall adopted in the Sloshel JIP. The first test was carried out at scale 1:6 in 2009 and at the full scale in 2010 at the same ambient conditions. The comparison of the two tests results showed the effect of the liquid properties and the air. As the compressibility of the gas was not scaled, the loading processes, ‘building jets along the wall from the impact area’ and ‘compression of entrapped air’, were not Froude-similar. The full-scale wave impact was found after the loading process of ‘compression of entrapped air’ at scale 1:6 was corrected using the one-dimensional model of Bagnold (1939).

Pasquier and Berthon (2012) compared the

sloshing impact measurements at full scale and at model scale. The actual ship motions were used as input for the model test (scale 1:40). A good correlation was found between the model test and full-scale results. At small scale, the experimental simulations of LNG sloshing represented an accurate global flow inside the tanks in terms of impact frequency. However, the authors suggested the need of studies for a wider range of conditions.

Karimi et al. (2013) also investigated the

effect of scaling on the sloshing pressure. Two sets of model tests of 2D tanks at scales of 1∶10 and 1∶40 were carried out by GTT with a fill level of 20%. The effect of the gas-liquid density ratio and the speed of sound in the gas on measured pressures was studied.

The sloshing pressures were also measured

by Kim et al. (2012) for a 1:50-scale model and a 1:70-scale model. The sloshing pressures were recorded at the same location for the excited model tanks with irregular motions at the same Froude scale. The comparison showed the differences in the statistical results for the

two models. However, when the Froude scaling was applied, a good agreement was found.

2.3.3 Wave Run-up

Research has been carried out in the past

years on the study of the wave run-up on circular cylinders, monopiles, barges and columns of large semisubmersibles using CFD methods. The outcome of the studies indicates the importance of high-order nonlinearities and the need of computational efforts for accurate predictions.

Ramírez et al. (2011) presented a CFD

model (NS3), which solves Navier-Stokes equations and uses the VOF method to treat the free surface. NS3 was applied to simulate the wave run-up on a vertical circular cylinder and the numerical results were compared to the experimental data from the large-scale tests performed at the Large Wave Channel (GWK) in Hannover, Germany.

Cao et al. (2011) conducted simulations of

the wave run-up on a fixed vertical cylinder. The finite volume method was employed to solve Navier-Stokes equations based on OpenFOAM. The wave elevations were computed at several locations within a radial distance around the cylinder. The computed wave run-ups, velocities and pressures at various locations were compared with the published experimental data from MARINTEK.

Kim et al. (2011) developed a numerical

wave tank model by matching the far-field wave solution based on the potential-flow theory and the near-field CFD solution. This model was implemented in a CFD code, based on the Arbitrary Langrangian Eulerian (ALE) finite element method. The developed method was applied to a truncated vertical cylinder exposed to nonlinear regular waves with wave length much greater than the diameter of the

280280

19

cylinder. Comparison with the theoretical and experimental data showed that the proposed method predicts wave run-up accurately with a small computational domain confined near the cylinder.

Li et al. (2012) investigated wave run-ups

on a vertical circular cylinder in an extreme wave environment. The waves were generated in a 3D wave basin using a focused wave method with different frequency and directional components. A practical method based on the velocity stagnation head theory was calibrated to calculate the wave run-up. The maximum wave kinematics at wave crest was calculated using the second-order theory. Wave run-ups on the weather side of the cylinder were calculated and compared with measurements in experiments with multi-directional focused wave groups. The relations between the defined parameters and the wave parameters, such as model scale, directional spreading index and wave steepness, were discussed.

Peng at al. (2012) investigated wave run-

ups on a monopile foundation in regular and irregular waves using ComFLOW. The numerical solutions are in good agreement with experimental measurements. It was showed that the wave run-up is dependent on the wave nonlinearity. A set of dimensionless and simple formulae have been derived to relate dimensionless wave run-up on the structure to the diameter of the structure and the Ursell number. The proposed formulae included the effect of the diameter of structure on the wave run-up. It led to similar results in comparison with other formulae.

Watai et al. (2011) reported some of the

results of a cooperative project that investigates the wave run-up on a large moving semi-submersible platform. Wave elevations at various locations below the deck were measured and compared to the predictions by

WAMIT and by the improved ComFLOW code. Results showed that ComFLOW was able to predict the relative wave elevations at different positions below the deck.

Priyanto et al. (2012) investigated the wave

run-ups on a large semi-submersible. Tests of a small-scale moored model in irregular waves were carried out in Marine Technology Centre (MTC)'s towing tank at Universiti Teknologi Malaysia. Significant wave run-up occurrences on the square-sectioned columns were observed.

Shan et al. (2012) presented an

experimental investigation on three model configurations including a four-column array, a two-column array and a single column, aiming to reveal the relationship between wave run-up and leg spacing, the relationship between wave interaction and model configurations, as well as the wave run-up distributions around columns. The wave environment was restricted to monochromatic progressive waves with different wave steepness. For the three tested configurations, wave run-up reached the maximum on the front sides of the fore columns, and decreased gradually as wave propagated. It was also found from the tests for the four- and two-column arrays that wave run-up decreases gradually as the leg spacing increases, which indicates the leg spacing would have important effect on the wave interaction among columns, and eventually affect the wave run-up on the columns.

2.4 VIV/VIM

2.4.1 Empirical VIV Prediction Programs Slender structures subjected to VIV often

vibrate in both in-line (IL) and cross-flow (CF) directions. The in-line motion of VIV can be a major contributor to the fatigue damage due to its higher frequencies and response modes even

281281

that the IL displacement normally is less than the CF one. It also triggers higher-order harmonic responses in both IL and CF direction which further increase the fatigue damage. Passano et al. (2012) reported the latest development of the VIV prediction program, VIVANA, with its new IL prediction model. The modelling of risers with partially covered strakes in Shear7 was presented by Resvanis and Vandiver (2011). The hydrodynamic force model of the strake section was generalized from forced motion tests with a rigid straked cylinder.

Efforts have been made to develop a

general methodology to calibrate Factors of Safety (FoS) for fatigue damage due to VIV. Fontaine et al. (2013) presented a reliability based method which accounts for uncertainties in S-N behaviour, metocean conditions and software prediction of VIV. Tognarelli et al. (2013) also presented similar methods, in which the prediction uncertainty is based on the measured flow and the response data for full-scale drilling risers in the field.

2.4.2 VIV Prediction Based on CFD

Huang and Larsen (2011) presented the 2-D

numerical simulation results for an elastically mounted circular cylinder subject to vortex induced vibrations. Reynolds-Averaged Navier–Stokes (RANS) equations and k–ω turbulent equations were solved by a finite volume method. The predicted response amplitudes, hydrodynamic forces and wake patterns were compared with the measured data in the equivalent experiments.

Zhao et al. (2012) simulated the one-

degree-of-freedom (1-DOF) VIV of a circular cylinder in oscillatory flow. The vibration of the cylinder was confined in the cross-flow direction only. RANS equations and k–ω turbulent equations were solved by a Petrov–Galerkin finite element method. The same

method was applied by Zhao et al. (2013) to study VIV responses of a cylinder in the combined steady and oscillatory flow.

Bourguet et al. (2011, 2012) performed

direct numerical simulations on a tensioned beam with a length to diameter ratio of 200, subject to vortex-induced vibrations in linear varying shear flow at three different Reynolds numbers, from 110 to 1,100. The energy transfer between the structure and the fluid was studied and the presence of mono- and multi-frequency responses was investigated. Similar study was also carried out for the tensioned beam subject to the exponential flow (Bourguet et al., 2013). The mechanism of the broadband VIV responses was studied.

2.4.3 New VIV Prediction Methods

A time-domain finite element analysis

method using a local hydrodynamic force model has been developed by Mainçon et al. (2011). In this model, the recent history of the velocity is used to enter a database of velocity and force measurements obtained from rigid cylinder tests, and thus to determine the force and advance the dynamic FEM analysis. Preliminary results are encouraging. The objective was to create models that can capture higher harmonics and can be used in the analysis of risers with seafloor contact or time-varying currents and waves.

Campbell et al. (2013) proposed a new

random vibration method with a band-limited white-noise lift-force model to predict the VIV responses of a fully straked flexible cylinder.

Ma et al. (2012) developed a time-domain

analysis tool for VIV prediction of marine risers based on a forcing algorithm and by making full use of the available high Reynolds number experimental data. In the formulation, the hydrodynamic damping is not treated as a special case but simply an extension of the

282282

21

experimentally derived lift curves. The forcing algorithm was integrated into a mooring analysis program based on the global-coordinate based finite element method. At each time step, the added mass, lifting force and drag force coefficients and their corresponding loads are computed for each element. Validation studies have been carried out for a full-scale rigid riser segment and a model-scale flexible riser. The numerical results were compared with experimental data and solutions by other programs. The validation studies have shown the proposed method is promising in VIV prediction.

2.4.4 Experiments

A. 2D Tests

The semi-empirical VIV prediction

programs rely on hydrodynamic force coefficients generalized from forced motion tests of rigid cylinders. In the work by Zheng et al (2011), extensive forced in-line and combined in-line and cross-flow experiments were employed to provide the hydrodynamic coefficient databases, in addition to the existing CF hydrodynamic coefficients. In these tests, the IL and/or CF motions are harmonic.

It is known that the cross-section motions

of a flexible beam subjected to VIV can be far from harmonic motions. The motion amplitude can also vary in time. To investigate the VIV response subjected to the non-harmonic motions, forced motion tests for rigid risers using observed orbits extracted from flexible beam were carried out by Yin and Larsen (2011). The tests results were compared with CFD solutions. Using the same experiment technique, Yin and Larsen (2012) further compared the hydrodynamic coefficients obtained from the forced motion tests with observed motion orbits extracted from flexible beam experiments and from periodic approximations. Aglen et al. (2011) studied the

added mass coefficients from forced motion tests with extracted orbits from flexible beam tests. The influence of added mass on the IL and CF interactions has been studied for tests with mode one dominating the responses in both directions.

Raghavan and Bernitsas (2011) performed

free oscillation tests of a rigid cylinder to study the Reynolds number effect within the range of 8.00×l03 to 1.50×l05. The objective of their work was to design a power generation unit based on VIV that can absorb energy from the fluid. It was found that VIV is significantly different at different flow regimes. An amplitude ratio of 1.9 was achieved for a smooth cylinder in VIV even with high damping imposed.

To further investigate the effect of

Reynolds number on VIV, a new innovative VIV test rig was designed and built at MARINTEK to test a rigid full-scale riser model (Lie et al., 2013). The rigid riser model was mounted vertically and can either be elastically mounted or be given a forced CF motion. The bare cylinder was tested in both sub-critical and critical Reynolds number regimes. The effect of Reynolds number on the amplitude of VIV displacement was found to be significant and further research was recommended to explore the subject.

Yiannis et al. (2013) performed the tandem

riser tests at the prototype Reynolds numbers. The tests were carried out utilizing two full-scale cylinders fitted with actual VIV suppression devices and towed either in fixed or spring supported configurations. The results revealed significant differences from those by today's design practices and industry codes.

283283

B. 3D Tests

Several VIV tests with flexible beam have

been carried out during 2011-2013. Strain gauges are mostly used in these tests. Accelerometers are also used in some of the tests to provide redundancy in the measurements. All of the tests were carried out in sub-critical Reynolds numbers due to the limitation in the test facility and the cost.

An extensive hydrodynamic test program of

riser models subjected to vortex-induced vibrations was carried out in the MARINTEK Offshore Basin Laboratory on behalf of Shell Oil Company (Lie et al. 2012). Three different riser models were towed horizontally at various speeds, simulating uniform and linearly varying sheared current. The test program included approximately 400 tests with different riser configurations. VIV responses of risers with/without suppression devices as well as the effect of Reynolds number and marine growth were investigated.

Huera-Huarte et al. (2013) presented the

experiment results of a long flexible cylinder with low mass ratio subject to a stepped current. The test pipe is 3m long with an external diameter of 19 mm. The effect of low mass ratio on VIV was investigated.

Song et al. (2010) performed VIV tests with

a long flexible riser towed horizontally in a wave basin. The riser model has an external diameter of 16 mm and a total length of 28.0 m. The asymmetrical distribution of displacement was mainly resulted from the modal composition.

Fu et al. (2013) performed VIV tests of a

flexible cylinder in an oscillatory flow. A flexible test cylinder was forced to harmonically oscillate at various combinations of amplitude and period. The test cylinder is 4m long with an outer diameter of 24mm.

Unique features of VIV in an oscillatory flow were presented.

Huera-Huarte and Bearman (2011)

performed model tests to study the interference between two identical risers. In these tests, two flexible risers were arranged in tandem and side-by-side positions. The test pipe is 1.5m long with an outer diameter of 16mm. The dynamic responses of the two interfering risers were presented.

Efforts have also been made to further

analyze the existing VIV test data. Larsen et al. (2012) applied wavelet analyses to reveal the frequency components in the measured signals, using Hanøytangen and NDP high-mode VIV test data. This study characterized the frequency components of VIV measured in flexible beams subjected to sheared current in order to establish a general model for use in the empirical VIV prediction programs.

The presence of higher-order harmonic

frequency components and chaotic responses has been observed in many flexible beam tests. Price et al. (2011) studied the impact of higher-order harmonic stress components and the broad band responses on fatigue damage using NDP high-mode VIV test data. The study indicated that both factors can lead to significant fatigue damages.

Modarres-Sadeghi et al. (2011) also

analyzed the NDP high-mode VIV experimental data. The stationary and chaotic VIV responses were characterized. Their influences on the fatigue damage were discussed.

Vandiver (2012) proposed a dimensionless

damping parameter to describe the cylinder VIV response, which overcomes the limitations in existing "mass-damping" parameters.

284284

23

Swithenbank and Larsen (2012) calculated the energy in the system from measured responses of a flexible beam and associated energy levels with the duration of the high VIV amplitudes.

Rao et al. (2013) studied the excitation

competition between the bare and buoyant segments of flexible cylinders using the Shell high-mode VIV test data.

McNeill and Agarwal (2011) proposed an

efficient method for modal decomposition and reconstruction of riser responses due to VIV. The travelling wave responses and the fatigue damage along the riser can be estimated accurately by this method using a limited number of measurements. McNeill (2012) further proposed an alternative way of estimating the fatigue damage, which is based on Dirlik's method to obtain rain-flow damage for Gaussian random stress.

2.5 New Experimental Techniques Song et al. (2013) presented a velocity

measurement method derived from the PIV technique using a high speed camera, called Bubble Image Velocimetry (BIV). It directly uses air-water interfaces in the image without the use of a laser for illumination. The measurement plane is controlled by minimizing the depth of field within which objects (i.e., air bubbles and water droplets in this case) are in focus and sharp, and therefore carrying more weight (i.e., higher intensity) in the correlation process for the velocity determination.

A subsea imaging technique was described

by Embry et al. (2012) for in-situ measurements, using a high resolution 3D laser imaging unit. The optical head is mounted on a 2D-scanning device. The equipment and the preliminary tests in a basin were described showing the ability of high spatial accuracy at a

relatively low scanning frequency. Experimental data (clouds of points) can be processed by regular dedicated software. Shapes of objects can be measured within a large volume at a millimeter precision. The system was initially developed for surveys and maintenance purposes, and it could be used in wave tanks for underwater measurements, for example, the scours around foundations.

Chabaud et al. (2013) developed the

concept of real-time hybrid testing (RTHT), defined as a hardware-in-the-loop (HiL) simulation, and applied it to scaled model testing. The authors admitted this method is not a standard and accurate method in offshore studies. In order to generalize its use, they described the global scheme and presented details on the different stages of calculations and data processing, at least on numerical and theoretical aspects.

The modeling of fenders in experiments

was presented by Cole et al. (2012). The design and development of model-scale fenders and their application in float-over topsides installation experiments are provided. Improvements were shown in term of versatility and robustness.

It should be mentioned the openings of two

new facilities in UK with wave and current capacities, mainly for ocean engineering and testing of marine energy devices. The Plymouth Ocean Wave Basin, established in 2013, is 35m long and 15m wide fitted with a movable floor (0-3m depth) and multidirectional wave generator. FLOWAVE-TT, opened in 2014 and located at Edinburgh, is a 25m diameter circular tank with a rising floor. Pumps around the tank allow to generate a water current up to 1.6 m/s at any direction in the tank.

285285

2.6 New Extrapolation Methods

During the period of 2011-2013, limited

investigations have been carried out on the development of extrapolation methods. However, challenges and issues in scaling of model test results to full scale have been indicated in problems related to sloshing, dynamic position systems, and mooring and risers.

Figure 2.6.1 Air Pocket Impact on A

Corrugated Wall (Upper: 1:6 Scale; Lower: Full Scale) (Bogaert et al., 2011)

For example, in the Sloshel project, Bogaert

et al. (2011) discussed the uncertainties in model tests of sloshing due to scaling biases which are associated with the Froude-scaled excitations. Based on the results of experiments at 1:6 and 1:1 scales (see Figure 2.6.1), it was concluded that gas pocket pressures are greatly affected by the gas compressibility bias due to the un-scaled properties of the gas.

2.7 Practical Applications of Computational Methods to Prediction and Scaling

Koop and Bereznitski (2011) calculated

current coefficients for the JBF-14000 semi-submersible using MARIN's in-house code ReFRESCO. Full-scale CFD computations were carried out to investigate possible scale effects using five subsequently refined grids for three different headings and ten different grids of different type for a scaled model. The numerical results were compared with the experimental data obtained from wind tunnel experiments and tests in the offshore basin. Approximately 15-20% lower values were found than those from the model-scale tests.

Ottens and Dijk (2012) studied the thruster-

hull interaction of a semi-submersible crane vessel in a current. CFD computations were compared with the model test data for the assessment of the thrust efficiency of the DP thrusters. From the comparison between the CFD and model test data, it was observed that the CFD method was able to predict the relevant force components within a sufficient accuracy for engineering purposes. To assess the CFD prediction in case of full scale, numerical results were compared with the sea trial data for the vessel with different thrust combinations. The comparison suggests the improvement in CFD code.

2.8 Improving Method of Experiments, Numerical Methods and Full-Scale Measurements

Huera-Huarte (2012) used the Defocusing

Digital Image Particle Velocimetry (DDPIV) method to measure vortex-induced vibrations of long flexible cylinders in wind/water tunnel. The concept of the proposed method was given by Willert and Gharib (1992). The author suggested the method, as a better alternative to other traditional vibration response

286286

25

measurement techniques, could be used to study VIV in the laboratory. The good agreement of measured data with known results confirmed the effectiveness of the above technique.

The flooding process of a tank in a

damaged ship was studied by Ruponen et al. (2012) at full scale. A decommissioned ship was used for the full-scale tests. Bernoulli’s equation for compressible fluid was used for air flows in the time-domain flooding simulations. For numerical simulations, The NAPA flooding simulation tool was used. In general, the comparison between experimental and simulated results showed a good agreement with small inaccuracies in the calculation of transient phenomena in the beginning of the flooding process.

3. REVIEW OF THE EXISTING PROCEDURES

The Committee reviewed three existing

procedures: 7.5-02-07-03.1 Floating Offshore Platform Experiments, 7.5-02-07-03.2 Analysis Procedure for Model Tests in Regular Waves and 7.5-02-07-03.3 Model Tests on Tanker-Turret Systems.

Only very minor revisions were identified

for 7.5-02-07-03.1 and 7.5-02-07-03.2. The Committee however found that there is little information in 7.5-02-07-03.3 and the limited information in 7.5-02-07-03.3 is very similar to that in 7.5-02-07-03.1. The Committee recommends to move the contents of 7.5-02-07-03.3 to 7.5-02-07-03.1.

The Committee also identified that there is

no existing procedure dealing with the result analysis of model tests in irregular waves. The Committee recommends to develop a new procedure on this aspect.

4. GUIDELINES FOR VIV AND VIM TESTS

Bluff marine structural bodies such as the

risers, free spanning pipelines and offshore platforms with cylindrical members (e.g., spars and semi-submersible) can undergo vortex shedding in ocean currents. The vortex shedding process and vortices induce periodic forces on the body which can cause the body to vibrate in both in-line (IL) and cross-flow (CF) directions. If the vortex induced response mainly causes elastic deformation in marine structures, such as risers, cables and free spanning pipelines, this phenomenon is known as Vortex Induced Vibrations (VIV). If the vortex induced response mainly causes rigid body motions such as a sway motion of a platform, this response often is denoted as Vortex Induced Motion (VIM).

The Committee focused on the

development of guideline for VIV testing (7.5-02-07-03.10). The purpose of this guideline is to ensure that laboratory model tests of VIV responses of marine structures are adequately performed according to the best available techniques and to provide an indication of improvements that might be made. The guideline is also to ensure that any comprises inherent in VIV tests are identified and their effects on the measured results are understood.

The Committee has also drafted the

guideline for VIM testing (7.5-02-07-03.11). It is recommended to be completed by the Ocean Engineering Committee of 28th ITTC.

5. NUMERICAL BENCHMARK STUDIES OF VIV

In the previous benchmark studies

organized by the 26th ITTC Ocean Engineering Committee, all participants selected two-dimensional unsteady RANS methodology.

287287

Various turbulence models were used with the assumption that the flow is fully developed to the turbulent status.

It was concluded from the study that the

drag crisis phenomenon on the stationary smooth cylinder was not predicted in the numerical studies. It is well known that the drag crisis is caused by the instability of separated shear layer in critical range (3x105< ReD <3.5x105). At the critical Reynolds numbers, the transition point is located very close to the point of flow separation. As a result, the shear layer eddies cause the mixing of flow in boundary layer so that the flow is energized and the flow separation is delayed. The delay of separation point leads to the reduction of the drag coefficient. The methodology based on two-dimensional, unsteady RANS with turbulence models, is not sufficient to simulate the physical phenomenon (ITTC Ocean Engineering Committee Report, 2011). It is necessary to extend the benchmark studies by including other CFD methods.

A Workshop for benchmark studies on VIV

and wave run-up was held in Nantes, France, October 17-18, 2013. Six participations presented their results of VIV studies. The results of benchmark studies are summarized below.

5.1 Benchmark Data As reported in the ITTC Ocean Engineering

Committee Report (2011), the benchmark data for the VIV of a circular cylinder was provided by MARIN. The rigid circular cylinder is 200mm in diameter and 3.52m in length (Figure 5.1.1). The cylinder was suspended from the carriage about 1.7m below the calm water surface. The VIV test apparatus is shown in Figure 5.1.2. The towing tank is 4m deep, 4m wide and 210m long. The cylinder was kept fixed in the flow and towed by the carriage at various speeds. Details of the tests can found in

de Wilde and Huijsmans (2001) and de Wilde et al. (2003, 2004 and 2006).

Figure 5.1.1 Smooth Cylinder of MARIN

Figure 5.1.2 High Reynolds VIV Test

Apparatus For the numerical computations, six (6)

Reynolds number are selected as follow: 6.31E+04, 1.26E+05, 2.52E+05 3.15E+05, 5.06E+05, 7.57E+05 The measured drag coefficient for the

smooth stationary cylinder is presented in Figure 5.1.3.

288288

27

Figure 5.1.3 Drag Coefficient for Smooth

Cylinder

5.2 Participants and Numerical Methods Various organizations and individuals have

been invited to participate in the benchmark studies. A list of participants is given in Table 5.2.1. The numerical methods and computational details are summarized in Table 5.2.2.

Table 5.2.1 Participants for the VIV

Benchmark Studies

Affiliation Nationality China Ship Scientific

Research Centre China

Seoul National University

Korea

Samsung Ship Model Basin

Korea

Memorial University Canada Inha University Korea

University of Iowa USA University of Southampton

UK

Shanghai Jiao Tong University

China

Table 5.2.2 Numerical Methods Used by Participants

Code name 2D/ 3D

Steady/ Unsteady

RANS /DES LES

A FLUENT (Commercial Code) 2D Unsteady RANS

B SNUFOAM (In-house Code) 2D Unsteady RANS

C FLUENT (Commercial Code) 2D Unsteady RANS

D CFDShip-IOWA (In-house Code) 3D Unsteady LES

E Code-S (In-house Code) 3D Unsteady LES

F OpenFOAM (Open Source Code) 3D Unsteady LES

G Naoe-FOAM-SJTU (In-house Code) 2D Unsteady RANS

H1 STAR-CCM+ (Commercial Code) 2D Unsteady RANS

H2 STAR-CCM+ (Commercial Code) 2D Unsteady RANS

H3 STAR-CCM+ (Commercial Code) 3D Unsteady DES

H4 STAR-CCM+ (Commercial Code) 3D Unsteady LES

Number of Grid Type of Grid Convection

Term ∆t

A 87,223 Structured Upwind 0.001/0.00

05

B 32,280 Structured Upwind 0.001/0.00

02 /0.0001

C 43,820 Structured Upwind 0.001

D 67,000,000 Structured QUICK/WENO 0.00008 /0.0001

E 11,300,000 Unstructured (Cartesian) Upwind (CFL=0.5)

F Max 4,000,000 Unstructured

Hybrid (Central + Upwind)

0.005

G 100,000 Chimera Upwind 0.00017 ~ 0.0015

H1 592,478 Hybrid Upwind 0.0001 ~0.002

H2 592,478 Hybrid Upwind 0.0001 ~0.002

H3 12,400,000 Structured Upwind 0.002 ~0.02

H4 12,400,000 Structured Upwind 0.002 ~0.02

289289

y+

Wall Function (Used/No

t Used)

Turbulence Model

Transition Model

(Used/ Not Used)

A 59 U k-w SST N B 2 N k-w SST N C 10 N k-w SST N

D 0.03 ~0.15 N Dynamic model N

E - N Dynamic model N

F 1 N Dynamic model N

G 1~4.9 U k-w SST N H1 0.06~0.56 N k-w SST N H2 0.06~0.56 N k-e (Standard) N H3 0.06~0.56 N - N H4 0.06~0.56 N - N

5.3 Numerical Results

In the benchmark studies, URANS, detached eddy simulation (DES) and large eddy simulation (LES) methods were used. In term of overall trend, results by DES and LES are generally in better agreement with the experimental data than those by URANS. The steep drop of mean DC was captured by LES. In addition, the LES results agree better with the experimental data at most points than those by URANS. Some URANS methods gave reasonably good results at high Reynolds numbers. The mean DC , the mean LC , the RMS of LC , and the Strouhal number are compared with experimental data in Figs. 5.3.1-5.3.4, respectively, and are also presented in Tables 5.3.1-5.3.4.

Figure 5.3.1 Mean Drag Coefficient

Figure 5.3.2 Mean Lift Coefficient

290290

29

Figure 5.3.3 RMS of Lift Coefficient

Figure 5.3.4 Strouhal Number Table 5.3.1 Mean Drag Coefficient

Mean CD Reynolds Number (E+05) 0.631 1.26 2.52 3.15 5.06 7.57

Exp. 1.16 1.10 0.77 0.27 0.26 0.26 A 0.70 0.57 0.46 0.45 0.44 0.42 B 0.87 0.69 0.61 0.61 0.54 0.54 C 1.05 0.89 0.82 0.79 0.71 0.62 D 1.37 1.37 0.56 0.27 0.25 0.21 E 1.10 1.08 0.88 - - - F 1.49 1.10 0.41 0.38 - - G 1.14 1.02 0.74 0.68 0.61 0.60 H1 1.28 1.16 1.04 1.06 1.00 0.95 H2 0.54 0.57 0.59 - 0.54 0.50 H3 1.38 1.02 0.84 0.81 0.69 0.52 H4 1.70 - 1.48 - - 0.30

Table 5.3.2 Mean Lift Coefficient

Mean CL Reynolds Number (E+05)

0.631 1.26 2.52 3.15 5.06 7.57 Exp. 0.01 -0.02 -0.17 -0.02 -0.04 -0.11

A 0.000 0.000 0.001 0.000 0.002 0.000 B 0.000 0.000 0.000 0.000 0.000 0.000 C -0.002 0.003 -0.002 0.000 0.001 0.001 D 0.038 -0.052 0.079 0.007 0.003 -0.005 E -0.030 -0.020 0.000 - - - F 0.000 0.040 0.000 0.000 - - G 0.000 0.012 0.000 0.000 0.001 -0.001 H1 0.130 -0.120 0.000 -0.080 -0.160 -0.180 H2 0.010 -0.010 0.010 - 0.010 0.010 H3 -0.040 -0.020 0.020 -0.030 -0.010 0.010 H4 -0.010 - -0.010 - - -0.040

Table 5.3.3 RMS of Lift Coefficient

RMS CL Reynolds Number (E+05)

0.631 1.26 2.52 3.15 5.06 7.57 Exp. 0.24 0.26 0.07 0.03 0.05 0.04

A 0.53 0.46 0.26 0.22 0.25 0.21 B 0.47 0.28 0.17 0.17 0.16 0.16 C 0.83 0.70 0.64 0.61 0.50 0.34 D 0.60 0.62 0.12 0.06 0.06 0.04 E 0.18 0.16 0.08 - - - F 0.51 0.34 0.06 0.06 - - G 0.86 0.72 0.24 0.19 0.15 0.12 H1 1.08 1.04 0.95 1.00 1.02 0.95 H2 0.10 0.12 0.13 - 0.13 0.12 H3 0.58 0.38 0.32 0.32 0.21 0.13 H4 0.99 - 0.55 - - 0.10

Table 5.3.4 Strouhal Number

Strouhal Number

Reynolds Number (E+05) 0.631 1.26 2.52 3.15 5.06 7.57

Exp. 0.19 0.20 2.00 0.46 0.79 0.45 A 0.28 0.29 0.31 0.31 0.31 0.32 B 0.25 0.26 0.27 0.27 0.22 0.21 C 0.25 0.26 0.27 0.27 0.27 0.28 D 0.18 0.18 0.19 0.41 0.41 0.34 E 0.28 0.27 0.53 - - - F 0.20 0.21 0.28 0.30 - - G 0.24 0.24 0.25 0.26 0.26 0.28 H1 0.29 0.29 0.29 0.30 0.31 0.31 H2 0.35 0.34 0.35 - 0.36 0.37 H3 0.22 0.25 0.28 0.27 0.28 0.31 H4 0.21 - 0.20 - - 0.31

5.4 Summary of Presentations at the Workshop

In the Workshop, six papers were presented

on VIV benchmark studies. A summary of some papers related to the benchmark studies is given below.

291291

Yeon et al. (2013) studied drag crisis with

the LES method and the computations were carried out at various Reynolds numbers. It was indicated that the solutions are strongly affected by the domain size. The mean drag coefficients were compared to experimental data by MARIN in Figure 5.4.1.

Figure 5.4.1 Drag Coefficient versus Re

(Yeon et al., 2013) Lee and Yang (2013) also employed LES

for the benchmark studies and carried out simulations at three Reynolds numbers, Re = 6.31E+04, 1.26E+05, and 2.52E+05. The 3D LES code was developed in-house based on a finite-volume method. A dynamic sub-grid scale model, in which the model coefficient is dynamically determined by the currently resolved flow field rather than by assigning a prefixed constant, was implemented for accurate turbulence modelling. Figure 5.4.2 presented time-averaged statistical data in comparison with those by MARIN. Note that Grids #1, #2 and #3 consist of 4.4, 8.7 and 11.3 millions of cells, respectively. As shown in the figure, LES captured the trends of the mean drag coefficients near the drag crisis.

Figure 5.4.2 Simulation Results of 3D LES ( Lee and Yang, 2013) James and Lloyd (2013) studied the flow

around the circular cylinder at high Reynolds numbers using LES. They found that unstructured grids provide better resolution of key flow features, when a ‘reasonable’ grid size is maintained. A blended upwind-central scheme, unique in OpenFOAM, was used, avoiding unnecessarily high numerical dissipation as well as removing artificial wiggles observed in the full central scheme. Figure 5.4.3 presents an example of vortical structures.

Re = 3.15 x 105 Figure 5.4.3 Vortical Structures in terms of

the Second Invariant of the Velocity Gradient Tensor (James and Lloyd, 2013)

Wen and Qiu (2013) simulated the two-

dimensional unsteady turbulence using a RANS solver, Star-CCM+, and various turbulence models. The studies showed that turbulence models have significant effects on the solutions (see Figure 5.4.4) and RANS is inadequate to address the “drag crisis” phenomenon.

292292

31

Figure 5.4.4 Horizontal Velocity Contours

at Re = 6.31E+04 (top: SST k- ω model, middle: k-ε model, bottom: RSTM model) (Wen and Qiu, 2013)

Ye et al. (2013) used a RANS solver,

pimpleFoam in OpenFOAM, coupled with an overset grid technique. The k-ω SST turbulence model was employed. Numerical results without overset grid approach were also presented for comparison study. An example of predicted velocity contour is presented in Figure 5.4.5.

Re = 7.57× 105 Figure 5.4.5 Velocity Contour (Ye et al.,

2013)

5.5 Conclusions and Recommendations

In this benchmark study, URANS, DES and LES were used. In terms of overall trend, numerical predictions by DES and LES are generally in better agreement with the experimental data than those by URANS. It can be concluded that the LES method captured the

drag crisis phenomenon and the LES solutions agree better with experimental data at most points than those by URANS. At high Reynolds numbers, some solutions by the URANS method agree reasonably well with the experimental data.

Some of the participants are still working

on completing the simulations at all the Reynolds numbers using DES and LES. More comparisons will be made in a journal paper, which is being prepared by the Committee.

The Committee recommended to continue

the benchmark studies based on LES and DES.

6. BENCHMARK STUDIES OF WAVE RUN-UP

6.1 Introduction

The Committee conducted benchmark studies of wave run-ups on single truncated cylinder and on four truncated cylinders. A Workshop was held at Nantes, France on October 17 and 18, 2013 and provided opportunities for participants to present and discuss the results of benchmark studies. Numerical solutions based on various methods were compared with experimental data. Note that six organizations of Korean Towing Tank Conference (KTTC) also carried out the comparative studies on wave run-ups using MOERI's benchmark data. A Workshop on the benchmark studies was held at Daejeon, Korea on September 12, 2013. Note that this Section will focus on the outcome of the Workshop hosted by the Committee.

6.2 Benchmark Data

The experiments for wave run-ups on a single truncated circular cylinder were carried out by both MOERI and MARINTEK.

293293

MARINTEK conducted the model tests for wave run-ups on four truncated cylinders. The benchmark data are summarized in the following sections.

6.2.1 Single Truncated Circular Cylinder

Model tests were carried out by MOERI for

the single truncated circular cylinder for six (6) wave periods and four (4) wave steepness. Table 6.2.1.1 presents the test matrix, in which the shaded cases were also tested by MARINTEK (Kristiansen et al., 2004). Note that the experimental data provided by MOERI was used in the benchmark studies. The diameter of the prototype cylinder is 16.0m and its draft is 24.0m. The model scale is 1:50.3 and the model diameter is 31.8cm. Figure 6.2.1.1 shows the locations of wave probes. The experimental set-up used by MOERI is presented in the Figure 6.2.1.2.

In the benchmark studies, the first-order,

second-order harmonics and mean results of following measured items were compared with numerical solutions at various wave frequencies in terms of kR ( k is wave number and R is the radius of the cylinder):

1) Horizontal force, xF , vs. kR

2) Vertical force, zF , vs. kR 3) Wave elevations at 10 wave probe

locations. Table 6.2.1.1 Test Matrix for the Single

Truncated Circular Cylinder

Figure 6.2.1.1 Locations of Wave Probes

(MOERI)

Figure 6.2.1.2 Experimental Set-up

(MOERI)

6.2.2 Four Truncated Cylinders

Model tests were conducted by MARNTEK for four truncated cylinders of two different cross-section geometries (circular and squared). The locations of wave probes and four truncated cylinders as well as wave headings are shown in Figure 6.2.2.1. The coordinates of wave probes are given in Table 6.2.2.1. Test conditions are summarized in Table 6.2.2.2.

294294

33

Note that MARINTEK has also carried out model tests for single circular and squared cylinders, as shown in Table 6.2.2.2. Only experimental data for four truncated cylinders were used in the benchmark studies.

(a) Four Circular Truncated Cylinders

(b) Four Squared Truncated Cylinders Figure 6.2.2.1 Experimental Set-up for

Four Truncated Cylinders (MARINTEK)

Table 6.2.2.1 Locations of Wave Probes (in Prototype Scale)

Circular Cylinder Squared Cylinder

X(m) Y(m) X(m) Y(m) a1 34.0000 25.9500 a1 34.0000 25.9500 a2 34.0000 24.5300 a2 34.0000 24.5300 a3 34.0000 21.2500 a3 34.0000 21.2500 a4 34.0000 18.0000 a4 34.0000 18.0000 b1 28.3078 28.3078 b1 27.1362 27.1362 b2 27.3037 27.3037 b2 26.1321 26.1321 b3 24.9844 24.9844 b3 23.8128 23.8128 b4 22.6863 22.6863 b4 21.5147 21.5147 c1 25.9500 34.0000 c1 25.9500 34.0000 c2 24.5300 34.0000 c2 24.5300 34.0000 c3 21.2500 34.0000 c3 21.2500 34.0000 c4 18.0000 34.0000 c4 18.0000 34.0000

Note: a1,b1 and c1 are the wave probes on the cylinder surface

Table 6.2.2.2 Test Matrix

I) single circular column, II) single squared column III) four circular columns, IV) four squared columns

6.3 Participants

Eleven organizations participated in the benchmark studies. The list of participants is given in Table 6.3.1. Some participants participated in the benchmark studies by carrying out numerical simulations and others conducted model tests. Table 6.3.2 presents the cases studied by each participant.

(0,0)

(0,0)

295295

Table 6.3.1 Participants for Wave Run-up

Benchmark Studies No. Affiliation 1 ECN, France 2 Hyundai Heavy Industries 3 Inha University 4 University of Iowa 5 MOERI(KRISO) 6 University of Bath 7 MARINTEK 8 Pusan National University 9 Samsung Heavy Industries with CD-Adapco Korea

10 Seoul National University 11 Shanghai Jiao Tong University

Table 6.3.2 Benchmark Studies by

Participants

Participant Single

Circular Cylinder

Single Squared Cylinder

Four Circular

Cylinders

Four Squared

Cylinders

Wave

heading: 0deg

0 deg 0 deg 45 deg 0 deg 45 deg

ECN, France O O O O Hyundai Heavy

Industries O

Inha University O University of Iowa O

MOERI O University of Bath O

MARINTEK* O O O O O O Pusan Nat. University O

Samsung Heavy Industries with

CD-Adapco Korea O O O

Seoul National University * O

Shanghai Jiao Tong University O O O

* Note that MARINTEK and Pusan National University participated in the benchmark studies by carrying out model tests.

6.4 Numerical Results

6.4.1 Grid Topology and Numerical Scheme

Nine participants employed CFD methods

to simulate wave run-ups. The numerical methods, the computational domains, boundary conditions and time steps employed by participants are summarized in Table 6.4.1.1.

The grid topologies are presented in Figure 6.4.1.2.

Table 6.4.1.1 Numerical Methods and

Schemes

Participants A B C D E G-1 G-2 H

Turbulence model RKE SGS - - K-omega SKE

Free-surface scheme VOF-Implicit MMD-ExplicitVOF with

local height function

- - VOF-Explicit

Wave theoryStoke 5th

orderStoke 5th order

Stoke 5th

orderStoke 1st & 2nd order

Stoke 2nd

orderAiry theory,

LinearStoke 2nd rder

Boundary conditions

Inlet Velocity Velocity Velocity

Damping zone

w/ wall B.C

Wave pressure & Velocity

Not applicable

Patch with relaxation

zoneVelocity

OutletPressure outlet

Pressure outletPressureoutlet

ExitNot

applicable

Patch with relaxation

zoneVelocity

Side Symmetry Symmetry Wall Zero gradientNot

applicableSlip-wall Velocity

Top Symmetry Symmetry - Far-fieldNot

applicablePatch Atmosphere

Bottom Wall Symmetry Wall Slip-wallNot

applicableNo-slip wall Slip wall

Time step size T/250 1/1000s

Variable timestep by courant number

0.005T/250~T/100

Variable time step

controlled by Courant number

T/200

Grid size

Inlet 1~2λ 2λ 1λ

±6λ

2λ 3λ 3D

Outlet 4λ 15λ 1λ 3λ 3λ 5D

Side 8D 12.5D 3D 3λ 16D 8D

Numberof Cells

Per length 150EA 75EA Min. 60EA 20~30EA 82EANot

applicable50~70EA 70EA

Per height 20EA 20EA Min. 6EA - At least 15Not

applicable12~22EA 10EA

CodeCommercialStar-CCM+

In-HouseINHAWAVE-II

Comflowver 3.1

In-HouseFEDIF

In-HouseCFDShip-Iowa

V4.5

In-houseDIFFRACTVer.2009

In-houseOpenFoamVer.2.2.1

In-house

296296

35

(a) A (b) B

(c) C

(d) D

(e) E (f) F

(g) G-1 (h) G-2

(i) H

Figure 6.4.1.1 Grids for the Single Truncated Circular Cylinder

(a) A

(b) E

(c) F

Figure 6.4.1.2 Grids for four Truncated Circular Cylinders

6.4.2 Single Circular Cylinder

The time series of computed wave

elevations at 10 wave probe locations (WPB#01-WPB#05, WPO#01-WPO#05) and forces on the single truncated circular cylinder ( xF and zF ) are compared with the time histories of experimental data by MOERI in Figure 6.4.2.1 for two test conditions (T15S110 and T09S116). It can be observed that the predicted patterns of wave run-ups on the single truncated circular cylinder, obtained by all the participants, are very similar to those of experimental ones.

[ Surface grid distribution around the cylinder ]

[ Volume grid distribution ]

297297

Figure 6.4.2.2(a) shows the 1st harmonic

results at WPB#01. The trends of experimental results by MOERI and MARINTEK are similar at WPB#01. The trends of numerical results by participants A, B and F agree well with the experimental results by MOERI at WPB#01. Figure 6.4.2.2(b) shows the 2nd harmonic results at WPB#01. The trends of experimental results by MOERI and Pusan National University (denoted as EXP-A) are similar at WPB#01. However, the trend of those by MARINTEK is somehow different from the other two.

(a) T15S110

(b) T09S116 Figure 6.4.2.1 Predicted Wave Elevations

and Forces with Experimental Results

1st harmonic components

(b) 2nd harmonic components Figure 6.4.2.2 Wave Run-ups at WPB#01

for the Single Truncated Circular Cylinder

298298

37

(a) Four Circular Cylinders,

T07S130,Wave Heading =0 deg

(b) Four Squared Cylinders, T07S130,

Wave Heading =0 deg

(c) Four Squared Cylinders, T07S130,

Wave Heading =45deg Figure 6.4.2.3 Comparison of Time

Histories for Four Truncated Circular and Squared Cylinders

6.4.3 Four Truncated Cylinders Figure 6.4.3.1 shows the time series of

wave elevations at 12 locations (a1~a4, b1~b4, c1~c4) for four truncated columns. They are compared with experimental results by MARINTEK. The predicted patterns of wave run-ups on the four truncated circular and squared cylinders are similar to those of experimental data. The 1st and 2nd harmonic components of wave elevations at a1, b1 and c1 are compared with experimental data in Figure 6.4.3.2 and Figure 6.4.3.3, respectively, for the four truncated circular cylinders with wave heading of 0 degree.

299299

(a) Location - a1

(b) Location - b1

(c) Location - c1 Figure 6.4.3.1 1st Harmonic Values for Four

Truncated Circular Cylinders at Locations of Three Wave Probes (Wave Heading =0 deg)

(a) Position - a1

(b) Position - b1

(c) Position - c1 Figure 6.4.3.2 2nd Harmonic Values for

Four Truncated Circular Cylinders at Locations of Three Wave Probes (Wave Heading =0 deg)

6.5 Summary of Presentations at Workshop

In the Workshop, six papers were presented

on wave run-up benchmark studies. A summary of papers related to the wave run-up benchmark studies is given below.

300300

39

Yoon et al. (2013) computed wave elevations at several locations around a truncated circular cylinder using CFDShip-Iowa. Time series of wave elevations and wave forces were compared with experimental data. Figure 6.5.1 presents the comparison of predicted wave elevations with experimental ones.

Figure 6.5.1 Experimental Wave Profiles

(Stansberg and Kristiansen, 2005) and Numerical Predictions for the Case of T=15s, H=35m (Yoon et al., 2013)

Sun et al. (2013) predicted wave elevations

around a single truncated circular cylinder using a potential-flow solver (DIFFRACT) and a viscous-flow solver, OpenFOAM. Results were compared with measured time series in experiments and the solutions by WAMIT. Spectral analyses were carried out. RAOs and QTFs of wave elevations were compared with the results obtained by Kristiansen et al. (2004).

(a) 1st Harmonic (b) 2nd Harmonic Figure 6.5.2 RAOs and QTFs (Sun et al.,

2013)

Kristiansen and Stansberg (2013) studied the wave diffraction (upwelling) and run-up on vertical columns in steep wave conditions by reviewing a data set from the scaled model tests with single and multiple fixed columns in deep water. Measurements clearly show higher wave crests than those by the linear modelling for steep waves. The second-order modelling can be used to improve this. However, there are still deviations in steep waves, especially at short wave periods (see Figure 6.5.3).

Figure 6.5.3 Wave Crest Heights for the

Case of T=15sec, 4 Columns and Heading = 45 deg. (Kristiansen and Stansberg, 2013)

Cao et al. (2013) computed wave run-ups

on a fixed single truncated circular cylinder and four circular cylinders using in-house CFD naoe-FOAM-SJTU solver. Favourable wave elevations were obtained in comparison with experimental data (see Figure 6.5.4).

Figure 6.5.4 Wave Elevations for the Case

of Four Circular Cylinders (T = 9 sec and Heading=45 deg) (Cao et al., 2013)

6.6 Conclusions and Recommendations

Eleven organizations participated in the

benchmark studies on the cases of the single

301301

truncated cylinder, and four organizations participated in the benchmark studies on the cases of four truncated cylinders. Nine participants employed CFD methods in their studies. The FFT analysis was performed to predict the harmonic values, and the results of harmonic values were compared with the experimental results.

It was concluded that the values and trends

of the computed wave elevations and forces by CFD methods are in good agreement with the experimental results for the cases of single and four cylinders.

Due to time constrains and limited

computing resources, most of participants have focused on the single circular cylinder cases. It is recommended that more studies be extended to the four-column cases. Some of the participants are still working on completing the simulations. More comparisons will be made in a journal paper, which is being prepared by the Committee.

7. THRUSTER INTERACTION AND SCALE EFFECT IN DP TESTS

7.1 Introduction Dynamic positioning (DP) systems and

azimuthing thrusters are widely used in the offshore industry for station-keeping. The effective force generated by thrusters can be significantly smaller than those obtained from their open-water characteristics. This is a result of thruster interactions with the hull, current and the wake of neighbouring thrusters. These phenomena are often referred as thruster-thruster and thruster-hull interactions. The understanding and quantification of thruster interaction (or thrust degradation) effects is essential for the evaluation of the station-keeping capabilities of DP vessels. Figure 7.1.1

presents some typical scenarios of thruster-thruster interactions.

Figure 7.1.1 Scenarios for Thruster-

Thruster Interactions (Ruiz et al., 2012)

7.2 Literature Review

A lot of today’s knowledge on thruster interaction effects was due to the development in the late 1980s and early 1990s (Nienhuis, 1992). In 1983, a semi-empirical calculation procedure was developed by MARIN to estimate the thruster-thruster interaction (Nienhuis, 1986). The underlying assumption of this method was that the propeller slipstream behaves similar to a swirling turbulent jet. The good correlation between model test results and calculations indicated that this assumption was correct. However, this conclusion was not confirmed since the extent of the thruster-thruster interaction was largely determined by two factors: the decrease of the velocity in the slipstream and the width of the slipstream. These two factors are related by the conservation of momentum. In 1986/1987, the first detailed thruster slipstream was measured by MARIN using the 2D LDV. These measurements were carried out on a thruster

302302

41

mounted under a simple-shaped barge. Some of the conditions were however very similar to the open-water condition. The test results showed a jet spread and a velocity decay different from that predicted by the simple calculation procedure mentioned above.

One of the oldest data available for thruster-

thruster interaction in open water were published by Lehn (1980, 1981) which are for zero-speed conditions only and cover variations in relative thruster position and thruster angle.

Nienhuis (1992) used the calculated

velocity field downstream of a simplified thruster (which delivered the same thrust and the same power as Lehn's thruster) and calculated the thruster-thruster interaction. The average inflow velocity over the propeller disk of the second thruster was calculated, and the resulting advance ratio was used in conjunction with the estimated open-water diagram of the thruster used by Lehn (1980). Moreover, Nienhuis (1992) investigated the interaction of thrusters below a flat plate. By using the calculated velocity field for a simplified thruster close to a flat plate, it is possible to calculate the thruster-thruster interaction using the same approach as for the open-water case. Both measurements and calculations showed that the interaction in open-water persists for larger distances between the thrusters. In the work of Nienhuis (1992), the thruster-hull interaction was also investigated.

API (1996) provided guidelines for the

determination of available thrust, and particularly gives guidance on how the thrust varies with the inflow velocity. It indicates that the thrust reduction for dynamic positioning systems due to oblique inflow cross-coupling effects is not well researched. API states that the propeller thrust decreases with increasing inflow, which is caused by the current speed, movement of vessel or the slipstream from another thruster. A 5-15% correction factor is

suggested to account for the Coanda effect. If there are support struts to the propeller in the flow, the reduction in thrust is approximately 10%.

Det Norske Veritas (DNV, 1996) outlined

rules for thruster assisted (TA) mooring systems. Depending on whether manual TA or automatic control (ATA) is employed, 70% or 100% of the net thrust can be used. It is assumed that azimuthing thrusters can provide thrust in all directions, unless specific restrictions are defined.

Brandner (1998) investigated the interaction

between two closely spaced ducted azimuthing thrusters through a series of experiments. Forces acting on a single thruster as well as on two thrusters were measured for a range of operating conditions and relative positions. The results showed that forces from the trailing thruster were heavily affected by interaction due to impingement of the race from the leading thruster, whereas forces from the leading thruster essentially remain unaffected despite its proximity to the trailing thruster.

van Dijk and Aalbers (2001) showed that

degradation effects on a thruster in model scale may occur due to inflow and cross flow, and due to waves if they cause ventilation effects. They stated that as the thrust is generated based on the principle of accelerating water, there is a suction flow and a jet flow. The suction flow is characterized by relatively low flow velocity over a wide area, while the jet flow is high speed and concentrated in a relatively small cross-section area. Furthermore, the jet may induce other flow patterns, depending on the local hull form and the intensity and direction of the jet. These flows, together with the current flow and waves may cause interaction effects leading to degradation of thruster performance. The following types of interaction were considered: thruster-thruster,

303303

thruster-hull (including the Coanda effect), thruster-current, and thruster-waves.

Figure 7.2.1 Thruster Configurations

Reported in Nordveit (2007) Ekstrom and Brown (2002) addressed the

influence of two thrusters in close proximity. The experiments were carried out at the wave tank in the Department of Mechanical Engineering at University College London. They recognized that additional research into the thruster cross-coupling effects at a range of in-flow velocities is needed, as these effects are likely to be significant at the current speeds appropriate to manoeuvring situations using dynamic positioning systems. In their work, they drew some interesting conclusions:

• The thrust in open water tests was 8 to 15%

more than that in the tests with a thruster attached to a vessel.

• Thrust losses were up to 5% for the thruster closer to the vessel.

• The loss of thrust is more likely due to the hull influence rather than the Coanda effect for thrusters placed close to the stern of the vessel.

• A reduction or an increase in thrust can occur when one thruster’s slipstream is influenced by the other thruster’s slipstream.

• A reduction in thrust (up to 40%) was recorded when one thruster’s slipstream is pointed into the other thruster’s slipstream. Nordtveit et al. (2007) presented the results

of model tests carried out in MARINTEK. The tests were to assess the thrust degradation in DP operations of an Aframax DP2 shuttle tanker, operating in rough environmental conditions in the North Sea. Investigations were conducted for tunnel thrusters, azimuth thrusters and main propellers with rudders. The results showed that thrust loss or thrust degradation effects due to thruster-thruster and thruster-hull interactions and the dynamics effects are of significant importance for the design, analysis and operation of a DP vessel. The magnitude of thrust degradation depends on the vessel type, design, operation and environmental forces. Thrust degradation coefficients are recommended for the DP capability analysis of the Aframax shuttle tanker.

Bosland et al. (2009) proposed a numerical

method to predict the interaction effects. The developed propeller interaction model is based on the panel method. At the second thruster the distorted flow field due to the first thruster was modeled by means of two wake field models; a linear potential wake model and an empirical turbulent jet model. Due to the intersection of wake and the body panels at the second thruster, numerical instabilities occurred at the collocation points. These instabilities were removed by applying a realistic vortex model instead of the analytical vortex model. It was concluded that the thrusters interaction propeller model coupled with the turbulent jet wake field yield an accurate prediction of thruster interaction. Although results based on the linear potential wake field model are promising, the prediction of the divergent and subsiding characteristics of the physical wake field needs to be improved, since the linear

304304

43

wake model does not correctly represent the physical properties of the wake.

Palmer et al. (2009) assessed thruster-hull

and thruster-thruster interactions on for autonomous underwater vehicles (AUVs). The interactions were investigated using an experimental approach. The induced longitudinal force (thrust loss) was less than 10% of the desired thrust force and the induced lateral force was between 8% and 20% of the desired thrust.

Figure 7.2.2 Possible Thruster Interactions

on AUVs (Palmer et al., 2009) de Wit (2009) and van Daalen et al. (2011)

investigated the interaction effects on the optimization of DP allocation algorithms. In their work, the effect of the thruster-thruster interaction on the reduction of delivered power was studied.

Det Norske Veritas (DNV, 2012) outlined

the guidelines for the design of dynamically positioned vessels. In the design process of DP systems, the thruster interaction effect is recognized as an important issue in the determination of the desired capability. The

magnitude of losses associated with the interaction effects is a function of the hull shape, the locations of thrusters, the degree of tilt of the propeller or nozzle axis. ‘Barred zones’ that prevent thrust in defined sectors can be created in the DP control system software to address issues associated with the thruster wash for azimuthing thrusters. Such barred zones may result in reduced capability. Furthermore, in order to minimize negative effects caused by thrusters interacting hydrodynamically with each other, DNV recommends that the distance between thrusters should be maximized to the feasible extent.

American Bureau of Shipping (ABS, 2013)

also published a Guide for Dynamic Position Systems. It recommends that the thruster-thruster interaction effect should be included in the station-keeping performance assessment, and that the results from full-scale or suitable model tests for thruster-thruster interaction effects can be used whenever possible. If such results are not available, Appendix 1 of the ABS Guide provides guidelines for the assessment of the interaction effect on the available thrust.

Song et al. (2013) investigated the thrust

loss by interactions between azimuth thrusters and ship hull based on the model tests and the numerical simulations. In the DP condition, two thrusts need to be considered: one is the thrust of the azimuth thrusters and the other one is the resultant thrust of the ship. The difference between these two thrusts denotes a thrust loss due to the thruster-hull interaction. In the model tests, the thrust and torque of an azimuth thruster were measured at 15° interval between 0° and 360°. The resultant thrust and moment were obtained by measuring the force using the dynamometer in the towing carriage. A Wind Turbine Installation Vessel (WTIV) was used in the studies. Based on the model tests, the thrust loss due to thruster-hull interaction was up to 30% of the pure thrust. In

305305

the numerical simulations, two methodologies, MRF (Moving Reference Frame) and SM (Sliding Mesh), were applied. Although both numerical methodologies showed good agreements with experimental data, it was suggested that the MRF method is time-saving and therefore more practical to predict the thrust loss.

Figure 7.2.3 Ship Model for Thruster-

Thruster/Hull Interaction Tests at Samsung Ship Model Basin (Song et al. , 2013 )

Figure 7.2.4 Polar Plot of Thrust Loss due

to Thruster-Thruster Interaction (Song et al., 2013 )

The work described above demonstrates

progress made in the study of the effects of thruster-thruster interaction on the performance of DP vessels. Thrust degradation effects can be quantified using data available from

literature, or by carrying out dedicated model tests. Published data can give valuable insights, but it is often too general, or not applicable to a specific design. Model tests, on the other hand, do provide detailed results but they are relatively expensive. In addition, model test results often become available relatively late in the design process, making it difficult to incorporate the results in the design.

The CFD simulation could be an alternative

method but there is little experience in the application of CFD as an engineering tool for thrust degradation effects. With the rapidly increasing capabilities of CFD models and computer hardware, the time is right for the development of new tools to analyze the thruster interactions (Cozijn, 2010).

In offshore heavy lift or pipe-laying

operations, the station keeping capabilities of a DP-vessel affect the operability limits of these operations. The efficiencies of DP thrusters of these vessels have been assessed by comparing the CFD solutions with model test results (Ottens et al., 2011). Numerical studies using CFD were performed to assess thruster-hull interactions on a semi-submersible vessel. The CFD results were validated against results of a series of model tests, including an open-water thruster, the single thruster-hull interactions without current, and full thruster-hull interactions with all thrusters active without current. The CFD results show good agreement with the model test data. The computed forces on the semi-submersible as well as on the individual floater with active thrusters are in 10% difference in comparison with the model test data. The largest discrepancies are in the bow quartering conditions where the thruster-hull interactions show the most complex flow pattern due to the location and shape of the stern keel. The comparisons between the CFD and model test results demonstrates that CFD is able to predict

306306

45

the relevant force components well with a sufficient accuracy for engineering purposes.

7.3 Measurement of Thruster Wake

Research has been carried out to understand the thruster interaction effects by measuring the detailed wake flow using PIV systems.

Figure 7.2.5 Downwash of Active

Thrusters, Azimuth 270 deg (Ottens et al., 2011)

Figure 7.2.6 Visualization of the

Impingement of Wakes on the Portside of the Floater, Azimuth 270 deg. (Ottens et al., 2011)

Figure 7.3.1 Measured Velocities in the

Wake of an Azimuth Thruster (Cozijn et al., 2010)

Figure 7.3.2 Flow Field of A Single Sweep

Measurement with the PIV System ( Cozijn et al., 2010)

Cozijn et al. (2010) investigated the wake

flow behind a ducted azimuthing thruster in open water and under a barge. Model tests were carried out in stationary conditions. The propeller thrust and torque were recorded and the flow velocities in a large number of cross-sections at various distances from the thruster were measured using a PIV system (see Figure 7.3.1). In addition, velocities were measured in a longitudinal plane at the thruster centre line (see Figure 7.3.2). The PIV measurements for the thruster under a barge show the thruster wake deformed by the presence of the barge as well as by its bilge. The bottom of the barge forms a flat plate above the thruster, clearly flattening the cross-section of the thruster wake. In addition, the wake flow along the bottom and the bilge of the barge resulted in a low pressure region, causing the wake flow to diverge up as it flows from under the barge into

307307

the open water. This phenomenon is known as the Coanda effect and was clearly visible in the PIV measurements.

Cozijn and Hallmann (2013) reported on

thruster-interaction model tests carried out in MARIN's Deepwater Towing Tank. The wake flow at a large number of cross-sections at different distances from the thrusters was measured with a PIV system for two different DP vessels, a semi-submersible and a drill ship. The PIV measurements provided a detailed image of the flow velocities in the thruster wake, showing the axial velocities, as well as the transverse and vertical velocity components (see Figure 7.3.3).

Figure 7.3.3 Measured Wake Velocity Field (Cozijn and Hallman, 2013)

7.4 Recommendations The Joint Industry Project on the

hydrodynamics of thruster interaction (TRUST JIP) was initialized to investigate the thruster interaction effects using both experimental and CFD methods. As an outcome, guidelines are also expected to be developed on how to use model tests and CFD computations in the analysis of thruster interaction effects and for the optimization of thruster configurations on DP vessels.

Although the application of CFD methods

for thruster interactions is still largely unexplored, suitable modeling methods should be investigated and developed in the near future. Thorough validation studies of CFD

models against measurement results, both at model-scale and at full-scale, are required.

Research into CFD computations for

thruster interactions should first focus on the computation of the velocities in the wake of a thruster in open water. The accurate computation of the velocities, especially at large distances from the thruster, is crucial for the accurate prediction of thruster interaction effects in a later stage. Different modelling options should be investigated. Subsequently, complex configurations should be considered in the computation by introducing additional physics, such as friction forces on the hull and the deflection of the thruster wake (Coanda effect).

8. MULTIPLE-BODY INTERACTION IN WAVES

8.1 Introduction

When two vessels are in a close proximity, the large resonant elevation of free surface occurs in the gap. Most of the linear sea-keeping programs currently used by the industry over-predict the free surface elevation between the vessels and hence the low-frequency loadings on the hull. This leads to problems in the design of the fenders, hawsers and loading arms and causes unsafe operations.

To overcome the problems, the lid

technique (Huijsmans et al., 2001), in which the free surface in the gap is replaced by a flexible plate, has been developed to suppress the unrealistic values of low-frequency forces. A linear dissipation term has also been proposed by Chen (2004) to modify the free-surface equation. Newman (2003) used the generalized mode technique to model the free surface. However, these methods require to input the artificial damping factors. For example, Chen (2005) computed the drift

308308

47

forces and wave elevations in the gap for two side-by-side barges and for a barge adjacent to a Wigley hull and compared the results to measured values. The numerical model was based on the linear potential theory with the addition of a damping term to the free surface boundary conditions in the gap region. Numerical simulations showed wave height and drift force in the resonance band was over-predicted using a damping coefficient ε =0; better agreement to measured data was achieved with ε = 0.016. Cheetham et al. (2007) presented numerical results by using AQWA software for side-by-side ship hydrodynamics and validation studies. A linearized damping lid boundary condition was used in the gap region. Simulations were performed for a ship-barge case where it was determined that a value of damping factor α = 0.01 was the most appropriate for the boundary condition.

Since these methods are inadequate to give

reasonable predictions without providing the experimental data beforehand, it may not be practical to apply them for design and analysis. It is desirable to determine the damping contribution due to viscous flow.

8.2 State-of-the-Art Review Many researchers have contributed to the

studies of interactions of side-by-side bodies based on the potential-flow theory in the frequency domain and using lower-order and high-order panel methods. Pauw et al. (2007) performed a comparison of measured data and numerical analysis of two side-by-side LNG carriers. The numerical analysis was performed with a panel method code using a flexible damping lid in the gap region. A variety of gap widths were used in head seas in an attempt to obtain rationale for predicting suitable damping factors. It was concluded that no unique value for the damping factor could fully cover all the

measured cases. It was noted that the damping factor should be tuned via the second-order drift force and not first-order quantities, such as wave height. The damping factor was said to have the greatest effect on the second-order drift force.

Bunnick et al. (2009) performed a

numerical simulation using the damping lid method, as described in Chen (2005), to compare to the model tests results of two side-by-side LNG carriers in head seas. The damping lid was also extended to the surface inside the vessel but not just the free surface gap. It was reported that the damping lid method worked better than the rigid lid method since the former showed better comparison with experimental results over the frequency range of interest.

Molin et al. (2009) used DIODORE, which

is based on the potential-flow theory, to analyze two side-by-side fixed barges. A set of massless plates were added to the gap between the barges and a quadratic damping force was applied to the plates. The numerical results were compared to the model tests of two rectangular barges in irregular waves. A drag coefficient, DC = 0.5, for determining the quadratic damping force, led to good agreement with measured data. It was recommended that an investigation of freely floating ships should be performed in the future.

Kawabe et al. (2010) examined water

surface response in a moon pool for a freely floating vessel. There was a comparison of numerical results with a damped moon pool free surface and measured data. This investigation showed a damping factor of α = 0.05 resulted in a good agreement between numerical and measured results.

Zhang et al. (2013) conducted numerical

calculations of the hydrodynamic interactions

309309

of two bodies in arbitrary arrangements, in terms of the different gap distances, relative sizes and the arbitrary relative angles. On the basis of the potential theory, the hydrodynamics software HYDROSTAR was utilized. The results of different cases showed that with the gap distances reducing, the resonance phenomenon became more dominant than shielding effect. For cases of different relative angles, the results suggested that sway and heave response were sensitive in head sea. For parallel arrangement, with the size of barges turn smaller, the motion responses were larger.

Ten et al. (2012) performed a semi-

analytical method to predict the characteristics of the resonance of viscous fluid in narrow gap. Meanwhile, dissipation was introduced in the form of pressure loss. This method was validated by comparing with the existing analytical, numerical and model test results. It was found that the classical BEM was inapplicable for very small gap size. The semi-analytical model was reported suitable both when the ratio of gap distance to the ship body breadth is small and large.

Clauss et al. (2013) investigated the gap

effects between side-by-side LNGs with numerical methods in the frequency domain. The free surface elevation was adapted by a damping lid. WAMIT that is based on the linear and second-order potential theory was used. The numerical and experimental investigations were conducted with a fixed terminal. Wave propagation in terms of wave height and regions of cancellation and amplification were examined. They reported that at a frequency around 0.81 rad /s, the surface elevation inside the gap was not overestimated by WAMIT without a numerical damping lid.

Xu et. al. (2013) calculated the second-

order mean drift force and moment on three

side-by-side barges during float-over operation, using the potential flow code WAMIT. The numerical results were validated by model tests. It was reported that numerical simulation could obtain a satisfactory result by adding viscous damping rectification.

Kashiwagi and Shi (2010) obtained the

pressure distribution for multiple bodies in a close proximity. They solved the integral equation of the diffraction potential by the Higher-order Boundary Element Method (HOBEM). It was found that when the separation distance between bodies becomes smaller, there would be a larger deviation of the pressure distribution.

Hong et al. (2013) studied the gap

resonance between the bodies in close proximity by two methods, in terms of a nine-node discontinuous higher order boundary element method (9dHOBEM) and a constant boundary element method based on the boundary matching formulation (BM-CBEM). The results showed that using BM-CBEM combined with the free surface damping or 9dHOBEM combined with a tuned value of the wetted surface damping parameter could largely reduce the over-predicted first-order hydrodynamic coefficients and successfully estimate the time-mean drift forces of two side by side floating structures.

Efforts have been made to address the

interaction problem in the time domain. Xiang and Faltinsen (2011) developed a time-domain solution for the linear loads and motions of two tankers paralleled in calm and deep water in lightering operation by using 3D Rankine source method. The numerical solution was validated by comparing with existing analytical, numerical and model test results. Zhu et al. (2008) presented a time-domain solution for two side-by-side floating structures using the potential flow theory. Two side-by-side hull-shape boxes with a narrow gap were

310310

49

fixed and the body forces due to the incoming waves and diffracted waves were computed. This time-domain analysis showed good agreement for the narrow gap resonant phenomena with the frequency-domain analysis.

Numerical methods based on nonlinear

potential-flow theory, such as the finite element method, have also been developed to solve the interaction problem.

Wang et al. (2011) applied fully nonlinear

potential theory to study 2D resonant waves in the gap between two floating structures. A higher-order finite element method was used to analyze the fully nonlinear resonant oscillations of the liquid in the gap. They compared the second-order time-domain results with corresponding fully nonlinear results and concluded that the second-order theory might overestimate the wave amplitude in the gap and the wave loads on the structures.

Ma et al. (2013) applied the fully nonlinear

potential theory to study the 2D resonant waves in the gap between two floating barges by using the Quasi Arbitrary Lagrangian-Eulerian Finite Element Method (QALEFEM). The computed free surface elevations and the forces acting on barges suggested the use of nonlinear models for such cases, in particular, the 4th- order or higher-order component was recommended. These investigations provided a basis for future 3D studies.

Attempts have also been made to determine

the viscous effect by solving Reynolds Averaged Navier–Stokes (RANS) equations. Lu et al. (2010a) performed numerical simulations on two identical bodies and three identical bodies at a close proximity. Numerical studies were completed by using potential flow theory and viscous fluid theory without artificial damping force. The viscous flow model was solved by three-step finite

element solver, and the CLEAR-VOF method was applied to capture the free surface in the gap region. Experimental results were used to assess the performance of each model. Both potential and viscous models performed well for predicting frequencies outside the resonance band, while the potential flow model over-predicts the wave height around resonant frequencies. The viscous flow model showed good agreement with measured values for all frequencies. To improve the potential flow method, Lu et al. (2010b) extended the previous work and applied artificial damping to the free surface. The potential flow model with a damping coefficient µ = 0.4 showed good agreement with the viscous flow results and measured values for two body cases and for both gaps in three body cases. In 2011, they conducted an investigation on the effects of gap width, body draft, body width and number of bodies of multi-bodies at close proximity (Lu et al., 2011).

Lu and Chen (2012) examined the energy

dissipation around resonant frequencies between two bodies by CFD computations. The dissipation was said to be relatively constant over frequencies near the resonant frequency. The dissipation rate was examined over various zones. The outcome showed that using the dissipation coefficient to assimilate the friction force could reduce the over-prediction of resonant wave elevation compared with conventional potential model. An explicit formula to achieve dissipation coefficient was recommended.

Zou and Larsson (2013) investigated

interactions of two side-by-side ships in shallow water. They completed a systematic numerical investigation of the ship-to-ship interaction during a lightering operation, using a steady-state Reynolds Averaged Navier–Stokes (RANS) solver. The numerical results were compared with benchmark experimental data. A good agreement was found between

311311

measured and computed wave heights, indicating that the predicted pressure distribution on the free surface was appropriate.

8.3 Existing Model Tests Many model tests of two side-by-side

bodies have been identified in the review, including captive and floating tests with and without mooring lines between two bodies. Besides those listed in Section 8.2, some recent model tests on side-by-side vessels are presented below.

Figure 8.3.1 Side-by-Side Moored Vessels (Kim et al., 2012) Kim et al. (2012) performed a series of

model tests and investigated the effect of the heading control on the offloading operability of side-by-side moored vessels, LNGC and LNG FPSO (FLNG), in multidirectional environments. In the tests, hawser tensions, fender loads, and relative motions between two vessels were measured, which are the key factors defining the offloading operability. The heading control was designed to maintain the FLNG’s heading. In the model tests, several heading angles were selected to investigate the impact of the heading control on the offloading operability, which includes the heading angles aligned with swell, and between swell and wind wave. The loading conditions of the FLNG and LNGC were chosen to have a similar roll natural period, and the period of swell was also selected close to the roll natural period, which realizes an worst situation. The model tests proved that the heading control

improves the offloading operability in the multi-directional environments. However, in the test, as vessel’s heading angle approaches to the direction of swell, the LNGC was exposed to wind wave as much as to increase the relative motion between the two vessels and deteriorate the offloading operability. In the model test campaign, the motion RAOs and horizontal drift forces/moment due to waves for the side-by-side moored vessels were measured and compared with the analytical calculations, which show the strong shielding effect on the wind wave by the FLNG.

Cho et al. (2011) carried out experimental

studies of motions and drift forces of side-by-side moored FSRU and LNGC including sloshing effect. Both FSRU and LNGC have LNG cargo tanks. The sloshing of LNG can affect the motions and drift forces due to the coupling between sloshing and motions of the floating bodies. The effect of coupling may vary with the filling level of LNG. The effect of filling level was investigated in their work. The effect of gap flow was also investigated. The horizontal motions and drift forces were analyzed and it was confirmed that the gap flow is affected by the sloshing. It was found that the sway motion, sway drift force and gap flow are influenced by sloshing in head sea even the sloshing is weak.

8.4 Potential Experimental Data for Benchmark Studies

To investigate the wave elevation in the gap

between two side-by-side bodies and the effect of viscous effect on the prediction, it is of importance to identify benchmark data for validation studies. The Committee recommended model tests of two floating bodies without mooring lines and fenders in between. The experimental data should include at least measured wave elevations in the gap and motions of bodies and/or mean drift forces

312312

51

at various gaps, wave frequencies and headings. The following experimental data have been identified for potential use in the proposed benchmark studies.

Hong et al. (2005) carried out model tests

of side-by-side LNG FPSO and LNGC with wired springs as mooring at the KRISO Ocean Engineering Basin. The gap was set as 4 m in full scale. Model tests were performed in both regular and irregular waves. For regular waves, the wave frequencies were from 0.25 rad/s to 1.2 rad/s in full scale and the headings were 150, 180, 240 and 270 degrees. Six DOF motions of each vessel were measured with photo sensors, relative waves at 3 locations (midship and +/- 0.3L from the midship, portside) of LNG FPSO were measured by capacitance type probes. Strain gauge type accelerometers were used for measuring horizontal and vertical accelerations. Drift forces were measured using the tension load cells at the end of spring wire moored to the ships.

The Committee have also initialized a

model test program for two identical bodies with simplified geometry in regular waves (wave steepness is 1/30). The tests were carried out at Ocean Engineering Research Centre of Memorial University's 60 m towing tank. The test matrix included three gaps and three wave headings. Motions of each body and wave elevations at three locations in the gap were measured. Figure 8.4.1 shows the simplified models in the tank. Model tests have also been carried out for the single body in regular waves. The tests are planned to be repeated in an ocean engineering basin. It is anticipated that the experimental results can be used in benchmark studies.

Figure 8.4.1 Simplified Models

8.5 Recommendations The determination of the viscous effect on

the prediction of wave elevations in the gap and the drift forces on two bodies in a close proximity remains as a challenge. As the next step, the Committee recommended to collect the experimental data available for benchmark studies. With an objective to investigate the viscous effect on the predictions by potential-flow methods, numerical tools based on the CFD methods should be included in the studies.

9. MOTIONS OF LARGE SHIPS AND FLOATING STRUCTURES IN SHALLOW WATER

The shallow water wave problem has

become one of the important issues in offshore hydrodynamics as the need for floating LNG terminals increases. The amplitude of the long period resonant motion of moored structures in shallow water is greatly influenced by the low frequency part of the incident waves, which themselves are a result of interactions of the component waves of the incident wave spectrum (ITTC Ocean Engineering Committee, 2008). The Committee was tasked to report on the motions of large ships and floating structures in shallow water. A literature review was first conducted to identify

313313

the progress made in model tests and numerical simulations of large floating structures.

In terms of the prediction of slowly varying

motions and loads, Stansberg and Kristiansen (2011) conducted model tests with a large LNG carrier moored in shallow water. Quadratic Transfer Function (QTF) for slowly varying surge force was presented using the cross-bi-spectral analysis. The off-the-diagonal QTF values have a tendency to increase significantly as the difference frequency is increased in shallow water condition, especially for smaller incident wave frequencies. Newman’s approximation was claimed to underestimate QTF for shallow water condition.

Pessoa et al. (2013) applied a second-order

boundary element method to an axisymmetric floating body in bi-chromatic waves. The occurrence of large low-frequency motions was shown in experiments and in numerical results when the difference frequencies between bichromatic waves were close to natural frequencies of surge motion and even pitch motion.

Progress has also been made in simulations

of nonlinear shallow water waves by solving Boussinesq equations. Lee et al. (2010) simulated nonlinear waves in shallow water based on the Boussinesq equations. The simulated waves well represented the wave deformations such as shoaling, refraction and non-linear wave interactions among wave components as they approach to the coastal region from the far field. By using the computed wave field, motion responses of two moored floaters were computed. The wave excitation and radiation force were estimated by a constant panel method. In order to estimate the wave excitation forces including shallow water effects, the wave height and velocity components obtained by the wave simulation were utilized. The computations

were applied to a floating storage and regasification unit (FRSU) and an LNGC in shallow water waves of varied depth. The numerical results were compared with those obtained for the equivalent constant water depth condition. The comparison shows that the motion responses are in general larger than those for even bottom cases. In particular, the horizontal motions are significantly large because of the wave deformations due to the bottom topography and the low-frequency waves. The enlarged horizontal surge motion is certainly important for the mooring design for floaters in shallow water.

Figure 9.1 Tower Yoke Mooring System (Kim et al., 2011 )

The motions of a LNG carrier in various

bathymetric conditions were studied by Kim (2013) using a linear Rankine panel method and by solving the Boussinesq equations. The depth was assumed to be shallow (15-30m), constant or with a slope. The numerical results were compared with those with infinite depth conditions, in term of hydrodynamic coefficients and motion responses. The authors concluded that the nonlinear effects are not noticeable except for a very steep slope, and that the linear methods can be used to evaluate the hydrodynamics of floating bodies over varied bathymetry.

Model tests have also been carried out to

study the responses of large floating structures

314314

53

such as FRSU in shallow water waves. For example, Kim et al. (2012) carried out model tests to study the responses of a FSRU with a mooring system in shallow water. In their work, two different mooring systems, the turret catenary system and the tower yoke system, were compared.

Zeng et al. (2012) presented a shallow

water mooring system for FPSO systems. Based on a turret mooring with self adjusting stiffness system (TUMSAS), the mooring systems were developed using numerical approaches and experimental validations. The design case was for a 24m deep location and a 30,000-ton FPSO. The self adjusting stiffness was obtained by using a weight module and catenary chains. The results show a larger offset of the FPSO and a drastic reduction of the forces acting on the mooring system. Vertical motions were also reduced in comparison with a regular turret.

The shallow water wave problem and the

motions of large floating structures in shallow water remain as challenging topics. The Committee recommended to identify benchmark data to validate numerical methods including those based on the potential flow theory, CFD and those based on solving the Boussinesq equations.

10. ISSC/ITTC WORKSHOP The first ISSC/ITTC joint workshop on

uncertainty modeling for ships and offshore structures has been successfully organized by ISSC, ITTC Ocean Engineering Committee and ITTC Seakeeping Committee. The Committee presented the uncertainties related to predictions of loads and responses for offshore structures at the Workshop on September 8, 2012 at Rostock, Germany. The joint effort between ISSC and ITTC has led to the publication of a special issue on uncertainty

modeling for ships and offshore structures in the journal of Ocean Engineering.

The second ISSC/ITTC joint workshop to be held in August 30, 2014 will focus on the wave-induced motion and structural loads on ships and offshore structures, including a computational benchmark test for a large modern ship.

11. CONCLUSIONS

11.1 State of the Art Review

Stationary Floating Structures and Ships Experimental and numerical procedures for

predicting motions of floating structures are in general well established. There is still a need of research on vortex induced motions of spars and semisubmersibles, and on the platform responses in extreme seas. Studies have been carried out on novel TLPs, spars and semisubmersible structures.

Relative motions between two floating

bodies remain very important research topics, especially for the safe operation of floating LNG production and storage and offloading vessels.

Highly Nonlinear Effects on Ocean Structures

Slamming, sloshing and wave run-ups,

representing the highly nonlinear effects, remain as important issues for the design/operation of offshore structures in extreme sea conditions. CFD methods such as VOF, SPH and CIP, along with experiments, are the primary tools to address these highly nonlinear phenomena.

For sloshing, the state-of-the-art

methodology is based on the use of seakeeping

315315

computer codes to estimate ship or platform motions. Experiments on sloshing tank models and CFD simulations have been performed in order to estimate global and local fluid loadings in the tanks. Benchmark studies of LNG sloshing have been carried out to assess the uncertainties in measurement of pressures, to investigate scale effects, and to validate the numerical tools. Research has also been focused on hydroelasticity using experimental studies. There is still a need of research in these areas using experimental and numerical methods.

VIV and VIM

Progress has been made in the prediction of

VIV and VIM using empirical prediction programs, CFD methods and experimental methods. A few new prediction programs have been developed based on the time-domain methods. Further research is required in this area.

New Experimental Techniques

A couple of new experimental techniques

have been identified, including the BIV technique for velocity measurements and a subsea imaging technique with a potential to be used in a tank for underwater measurements.

New Extrapolation Methods

Limited investigations have been carried

out on the development of extrapolation methods. Challenging issues in scaling of model tests results to full scale have been indicated in various applications throughout the report, particularly in sloshing tests.

11.2 Review of the Existing Procedures The Committee reviewed three procedures.

Very minor revisions were identified for 7.5-

02-07-03.1 and 7.5-02-07-03.2. The Committee however found that there is little information in 7.5-02-07-03.3 and the limited information in 7.5-02-07-03.3 is very similar to that in 7.5-02-07-03.1. The Committee recommended to move the contents of 7.5-02-07-03.3 to 7.5-02-07-03.1. The Committee also identified that there is no existing procedure dealing with the result analysis of model tests in irregular waves. The Committee recommends to develop a new procedure on this aspect.

11.3 Benchmark Studies on VIV

In this benchmark study, URANS, DES and LES were employed by six participants. In terms of overall trend, numerical predictions by DES and LES are generally in better agreement with the experimental data than those by URANS. It can be concluded that the LES method captured the drag crisis phenomenon and the LES solutions agree better with experimental data at most points than those by URANS. At high Reynolds numbers, some solutions by the URANS method agree reasonably well with the experimental data. The Committee recommended to continue the benchmark studies based on LES and DES.

11.4 Wave Run-Up Benchmark Studies Eleven organizations participated in the

benchmark studies on the cases of the single truncated cylinder and four organizations participated in the benchmark studies on the cases of four truncated cylinders. Nine participants employed CFD methods in their studies. It was concluded that the values and trends of the computed wave elevations and forces by CFD methods are in good agreement with the experimental results for the cases of single and four cylinders. Due to time constrains and limited computing resources, most of participants have focused on the single

316316

55

circular cylinder cases. It is recommended that more studies be extended to the four-column cases.

11.5 Thruster-Thruster Interactions A literature review has been conducted for

thruster-thruster interactions. Great progress has been made in investigating the interactions using experimental and CFD methods. Research has been carried out to understand the thruster interaction effects by measuring the detailed wake flow using PIV systems. Although the application of CFD methods for thruster interactions is still largely unexplored, suitable modeling methods should be investigated and developed in the near future. Thorough validation studies of CFD models against measurement results, both at model-scale and at full-scale, are required.

11.6 Side-by-Side Body Interaction A literature review has been conducted for

side-by-side body interactions in waves with an emphasis on the prediction of wave elevation in the gap and the drift forces. Progress has been made in investigating the damping effect using model tests and CFD simulations. However, the determination of wave elevations and drift forces using the potential-flow based methods remains as a challenge. The Committee recommended to collect the available experimental data for benchmark studies. With an objective to investigate the viscous effect on the predictions by potential-flow methods, numerical tools based on the CFD methods should be included in the studies.

11.7 Motions of Large Ships and Floating Structures in Shallow Water

A literature review has been carried out for

motions of large ships and floating structures in

shallow water. The focus was on the LNG ships and terminals as well as FPSO and their mooring systems. The shallow water wave problem and motions of large floating structures in shallow water remain as challenging topics.

In shallow water, the low-frequency

component induced by nonlinear wave interactions is important for the low-frequency motions of two floating bodies. The hydrodynamic effects of sloshing tank and the gap phenomena for two floating bodies in shallow water need to be studied further.

The Committee recommended to identify

benchmark data to validate numerical methods including those based on potential flow theory, CFD and those based on solving the Boussinesq equations.

12. RECOMMENDATIONS The Ocean Engineering Committee would

like to make the following recommendation to the 27th ITTC:

- Adopt the new guideline 7.5-02-07-03.10,

"Guideline for VIV Testing"

13. REFERENCES Aarsnes, J.V., 1996, “Drop Test with Ship

Sections - Effect of Roll Angle”, Report 603834.00.01, Norwegian Marine Technology Research Institute, Trondheim.

Aglen, I.M. and Larsen, C.M., 2011,

“Importance of Added Mass for the Interaction between IL and CF Vibrations of Free Spanning Pipelines”, Proc. OMAE2011, Rotterdam, The Netherlands.

317317

Alaoui, A.E.M. and Nême, A., 2012,

“Slamming Load during Vertical Water Entry at Constant Velocity”, Proc. ISOPE2012, Rhodes, Greece.

American Petroleum Institute (API), 1996,

“Recommended Practice for Design and Analysis of Station-keeping Systems for Floating Structures”, App. C, API, 2SK, 2nd edition.

American Bureau of Shipping (ABS), 2013,

“Guide for Dynamic Positioning Systems”, Houston.

Bachynski, E.E. and Moan, T., 2012, “Linear

and Nonlinear Analysis of Tension Leg Platform Wind Turbines”, Proc. ISOPE2012, Rhodes, Greece.

Bae, Y.H. and Kim, M.H., 2013, “Rotor-floater-

tether Coupled Dynamics Including Second-order Sum-frequency Wave Loads for a Mono-column-TLP-type FOWT (Floating Offshore Wind Turbine)”, Ocean Eng., Vol. 61, pp.109-122.

Bagnold, R.A., 1939, "Interim Report on

Wave-Pressure Research", Report, TU Delft. Bai, Z., Xiao, L., Kou, Y. and Yang, L., 2013,

“Research on Vortex Induced Motion of a Deep Draft Semisubmersible with Four Rectangular Columns”, Proc. ISOPE2013, Anchorage, USA.

Baudin, E, Diebold, L., Pettinotti, B, Rousset,

J.M. and Weber, M., 2013, “Bureau Veritas Sloshing Model Tests & CFD Calculations within ISOPE Benchmark”, Proc. ISOPE2013, Anchorage, USA.

Bitner-Gregersen, E.M., 2011, “Reliability

Assessment of TLP Air-Gap in Nonlinear Waves”, Proc. OMAE2011, Rotterdam, The Netherland.

Bogaert, H., Kaminski, M.L.M. and Brosset, L., 2011, “Full and Large Scale Wave Impact Tests for a Better Understanding of Sloshing: Results of the Sloshel Project”, Proc. OMAE2011, Rotterdam, The Netherland.

Bosland, R., Dijk, J.M. and Huijsmans, R.H.M.,

2009, “Numerical Prediction of Thruster-Thruster Interaction”, Proc. OMAE2009, Honolulu, USA.

Bourguet, R., Karniadakis, G.E. and

Triantafyllou, M.S., 2013, “Multi-Frequency Vortex-Induced Vibrations of a Long Tensioned Beam in Linear and Exponential Shear Flows”, J. Fluid and Structures, Vol. 41, pp.33-42.

Bourguet, R., Triantafyllou, M.S., Tognarelli,

M. and Beynet, P., 2011, “Fluid-Structure Energy Transfer of a Tensioned Beam Subject to Vortex-Induced Vibrations in Shear Flow”, Proc. OMAE2011, Rotterdam, The Netherlands.

Bourguet, R., Triantafyllou, M.S., Tognarelli,

M. and Beynet, P., 2012, “Distributed Wake-Body Resonance of a Long Flexible Cylinder in Shear Flow”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Bouscasse, B., Colagrossi, A., Antuono, M. and

Lugni, C., 2013, “A Classification of Shallow Water Resonant Sloshing in a Rectangular Tank”, Proc. OMAE2013, Nantes, France.

Brandner, P.A., 1998, “An Experimental

Investigation into the Interaction between Two Closely Spaced Azimuthing Thrusters”, Report AME CRC C 98/4, Australian Maritime Engineering CRC Limited.

318318

57

Bunnik, T., Pauw, W. and Voogt, A., 2009, “Hydrodynamic Analysis for Side-by-Side Offloading”, Proc. ISOPE2009, Osaka, Japan.

Campbell, B. and Slocum, S., 2013,

“Prediction of VIV for Risers With Effective Strakes”, Proc. OMAE2013, Nantes, France.

Cao, H., Liu, Y. and Wan, D. 2013,

“Benchmark Computations of Wave Run-Up on Single Cylinder and Four Cylinders by naoe-FOAM-SJTU Solver”, Proc. ITTC Workshop on Wave Run-up and Vortex Shedding, Nantes, France,

Cao, H., Zha, J. and Wan, D., 2011, “Numerical

Simulation of Wave Run-up Around a Vertical Cylinder”, Proc. ISOPE2011, Maui, USA.

Chabaud, V., Steen, S. and Skjetne, R., 2013,

“Real-Time Hybrid Testing for Marine Structures: Challenges and Strategies”, Proc. OMAE2013, Nantes, France.

Chandrasekaran, S., Sundaravadivelu, R.,

Pannerselvam, R., Madhuri, S. and Varthini, D.S., 2011, “Experimental Investigations of Offshore Triceratops under Regular Waves”, Proc. OMAE2011, Rotterdam, The Netherlands.

Cheetham, P., Du, S., May, R. and Smith, S.,

2007, “Hydrodynamic Analysis of Ships Side By Side in Waves”, Proc. International Aerospace CFD Conference, Paris, France.

Chen, X.B., 2004, "Hydrodynamics in Offshore

and Naval Applications - Part I", Keynote lecture of 6th Intl. Conf. Hydrodynamics, Perth (Australia).

Chen, X.B., 2005, “Hydrodynamic Analysis for Offshore LNG Terminals”, Proc. International Workshop on Applied Offshore Hydrodynamics, Rio de Janeiro, Brazil.

Cho, S., Sung, H., Hong, S., Hong, S., Ha, M.,

Choi, Y., Yu, B., Jang, R. and H. Choi, 2011, “Experimental Study on the Effect of Sloshing on Side-by-Side moored FSRU and LNGC”, Proc. ISOPE2011, Maui, USA.

Choi, H.I., Park, J.H., Kwon, S.H., Lee, K.H.,

Lee, S.B. and Yang, Y.J., 2012, “An Experimental Study on Hydro-Elasticity in Sloshing”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Clauss G.F., Dudek M. and Testa D., 2013,

“Gap Effects at Side-by-Side LNG-Transfer Operations”, Proc. OMAE2013, Nantes, France.

Clauss, G.F., Sprenger, F., Dudek, M. and Testa,

D., 2012, “Sloshing: From Theory to Offshore Operations”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Cole, J.C., Islam, M., Garvin, M. and

Herrington, P., 2012, “Modelling Details of Fenders in Float-Over Installation Experiments”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Constantinescu, A., Alaoui, A.E.M., Nême, A.

and Jacques, N. and Rigo, P., 2011, “Numerical and Experimental Studies of Simple Geometries in Slamming”, Inter. J. Offshore and Polar Eng., Vol. 21, pp.216-224.

Constantinides, Y. and Oakley, O.H., 2013,

“Prediction and Screening of Truss Spar VIM with CFD”, Proc. OMAE2013, Nantes, France.

319319

Copple, R.W. and Capanoglu, C., 2012,

“Tension Leg Wind Turbine (TLWT) Conceptual Design Suitable For a Wide Range of Water Depths”, Proc. ISOPE2012, Rhodes, Greece.

Cozijn, H., 2010, "New Initiatives Probe:

Thruster Interaction Effects", Maritime Report, Num. 99, pp.10-12.

Cozijn, J.L. and Hallmann, R., 2013, "Thruster-

Interaction Effects on a DP Semi-Submersible and a Drill Ship: Measurement and Analysis of the Thruster Wake Flow", Proc. OMAE2013, Nantes, France.

Cozijn, J.L., Hallmann, R. and Koop, A.H.,

2010, “Analysis of the Velocities in the wake of an Azimuthing Thruster Using PIV Measurements and CFD Calculations”, Proc. Dynamic Positioning Conference, Houston, USA.

Cruz, R.E., Neves, M.A.S., Rivera, L.A. and

Esperança, P.T.T., 2012, “An Investigation on the Excitation of Yaw Parametric Resonance of a Tension Leg Platform”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Damblans, G., Berhault, C., Marcer, R., Cinello,

A. and Pétrié, F., 2012, “CFD and Experimental Investigations of Slamming Load Predictionon Subsea Structures in Splash Zone”, Proc. ISOPE2012, Rhodes, Greece.

DaSilva, O. and Knecht, H., 2011, “Airgap on

Semi Submersibles, A Practical Guide on the Implementation of a Stochastic Approach”, Proc. OMAE2011, Rotterdam, The Netherlands.

Det Norske Veritas (DNV), 1996, “Rules for

Classification of Mobile Offshore Units, Position Mooring”, Part 6 Ch. 2.

Det Norske Veritas (DNV), 2012, “Dynamic Positioning Vessel Design Philosophy Guidelines”, Recommended Practice DNV-RP-E306.

de Wilde, J.J. and Huijsmans, R.H.M, 2001,

“Experiments for High Reynolds Numbers VIV on Risers”, Proc. ISOPE2001, Stavanger, Norway.

de Wilde, J.J. and Huijsmans, R.H.M., 2004,

“Laboratory Investigation of Long Riser VIV Response”, Proc. ISOPE2004, Toulon.

de Wilde, J.J., Huijsmans, R.H.M. and

Triantafyllou, M.S., 2003, “Experimental Investigation of the Sensitivity to In-line on Vortex-induced Vibrations”, Proc. ISOPE2003, Hawaii, USA.

de Wilde, J.J., Huijsmans, R.H.M. and Tukker,

J., 2006, “Experimental Investigation into the Vortex Formation in the Wake of an Oscillating Cylinder Using Particle Image Velocimetry”, Proc. ISOPE2006, San Francisco, USA.

de Wit, C., 2009, "Optimal Thrust Allocation

Methods for Dynamic Positioning of Ships", M.Sc. Thesis, Delft Univ. of Technology.

Ekstrom, L. and Brown, D.T., 2002,

"Interaction Between Thrusters Attached to a Vessel Hull", Proc. OMAE2002, Oslo, Norway.

Embry, C., Hardy, M., Nickerson, B., Manning,

N., Goodyear, D., Richardson D. and Pappas, J.M., 2012, “SS MTS: Subsea Monitoring – High Resolution 3D Laser Imaging for inspection, Maintenance, Repair and Operations”, Proc. OTC2012, Houston, USA

320320

59

Fillon, B., Henry, J., Diebold, L. and Derbanne, Q., 2013, “Extreme Values Theory Applied to Sloshing Pressure Peaks”, Proc. ISOPE2013, Anchorage, USA.

Fontaine, E.R., Rosen, J., Marcollo, H.,

Vandiver, J.K., Triantafyllou, M., Resvanis, T.L., Larsen, C.M., Tognarelli, M.A., Oakley, O.H., Constantinides, Y. and Johnstone, D.R., 2013, “Using Model Test Data to Assess VIV Factor of Safety for SCR and TTR in GOM”, Proc. OMAE2013, Nantes, France.

Forristal, G.Z. and Aubault, A., 2013,

“Improving Offshore Platforms Wave Measurements”, Proc. OMAE2013, Nantes, France.

Forristal, G.Z., 2011, “Maximum Crest Heights

under a Model TLP Deck”, Proc. OMAE2011, Rotterdam, The Netherlands.

Fu, S., Wang, J., Baarholm, R., Wu, J. and

Larsen, C.M., 2013, "VIV of Flexible Cylinder in Oscillatory Flow", Proc. OMAE2013, Nantes, France.

Gaidai, O., Naess, A. and Stansberg, C., T.,

2012, “Airgap Statistics for a Tension Leg Platform”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Gonçalves, R.T., Franzini, G.R., Rosetti, G.F.,

Fujarra, A.L.C. and Nishimoto, K., 2012, “Analysis Methodology for Vortex-Induced Motion (VIM) of a Monocolumn Platform Applying the Hilbert–Huang Transform Method”, J. Offshore Mech. and Arctic Eng., Vol. 134.

Gonçalves, R.T., Rosetti, G.F., Fujarra, A.L.C.

and Nishimoto, K., 2012a, “An Overview of Relevant Aspects on VIM of Spar and Monocolumn Platforms”, J. Offshore Mech.and Arctic Eng., Vol. 134.

Gonçalves, R.T., Rosetti, G.F., Fujarra, A.L.C., Nishimoto, K. and Oliveira, A.C., 2012, “Wave Effects on Vortex-Induced Motion (VIM) of a Large-Volume Semi-submersible Platform”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Gonçalves, R.T., Rosetti, G.F., Fujarra, A.L.C.,

Nishimoto, K. and Oliveira, A.C., 2012, “Vortex-Induced Yaw Motion (VIY) of a Large-Volume Semi-Submersible Platform”, Proc. ISOPE2012, Rhodes, Greece.

Graczyk, M., Berget, K. and Allers, J., 2012,

“Experimental Investigation of Invar Edge Effect in Membrane LNG Tanks”, J. Offshore Mech. and Arctic Eng., Vol. 134.

Hagen, Ø., 2011, “Reliability Assessment for

TLP Tether Overload in Nonlinear Waves”, Proc. OMAE2011, Rotterdam, The Netherlands.

Hennig, J., Scharnke, J., Buchner B. and van

den Berg, J., 2011, “Extreme Load-Response Mechanisms of a Tension Leg Platform due to Larger Wave Crests – Some Results of the ‘Crest’ JIP”, Proc. OMAE2011, Rotterdam, The Netherlands.

Hong, D.C., Hong, S.Y., Nam, B.W. and Hong,

S.W., 2013, “Comparative Numerical Study of Repulsive Drift Forces and Gap Resonances between Two Vessels Floating Side-By-Side in Proximity in Head Seas Using A Discontinuous HOBEM and A Constant BEM with Boundary Matching Formulation”, Ocean Eng., Vol. 72.

Hong, S.Y., Kim, J.H., Cho, S.K., Choi, Y.R.

and Kim, Y.S., 2005, "Numerical and Experimental Study on Hydrodynamic Interaction of Side-by-Side Moored Multiple Vessels", Ocean Eng., Vol. 32.

321321

Hong S., Lee, I., Park, S.H., Lee, C. and Chun,

H., 2013, “Scale Model Experiments of the SPAR-Type Floating Offshore Platform”, Proc. ISOPE2013, Anchorage, USA.

Huang, Z. and Larsen, C.M., 2011, "Numerical

Simulation on Vortex-Induced Vibration of an Elastically Mounted Circular Cylinder With Two Degrees-of-Freedom", Proc. OMAE2011, Rotterdam, The Netherlands.

Huera-Huarte, F. J., 2012, “An Optical Tool for

Measuring Vortex-Induced Vibrations of Long Flexible Cylinders”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Huera-Huarte, F.J. and Gharib, M., 2011,

"Vortex and Wake-Induced Vibrations of a Tandem Arrangement of Two Flexible Circular Cylinders with Far Wake Interference", J. Fluid and Structures, Vol. 27, pp.824-828.

Huera-Huarte, F.J. and Gharib, M., 2011,

"Flow-Induced Vibrations of a Side-by-Side Arrangement of Two Flexible Cylinders", J. Fluid and Structures, Vol. 27, pp.354-366.

Huera-Huarte, F.J., Jeon, D. and Gharib, M.,

2011, “Experimental Investigation of Water Slamming Loads on Panels”, Ocean Eng., Vol. 38, pp.1347-1355.

Huijsmans, R.H.M., Pinkster, J.A. and de

Wilde, J.J., 2001, "Diffraction and Radiation of Waves Around Side by Side Moored Vessels", Proc. ISOPE2001, Oslo, Norway.

James, M. and Lloyd, T., 2013, “Large Eddy

Simulations of Circular Cylinders at a Range of Reynolds Numbers”, Proc. ITTC Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Ji, Y.M., Shin, Y.S., Park, J.S., and Hyun, J.M., 2012, “Experiments on Non-Resonant Sloshing in a Rectangular Tank with Large Amplitude Lateral Oscillation”, Ocean Eng., Vol. 50, pp.10-22.

Jiang, M., Lien, V., Lesar, D., Engle, A. and

Lewis, R., 2012, “A Validation of Various Codes Using Hydrodynamic Wedge Impact Data”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Johannessen, T.B., 2011, “High Frequency

Loading and Response of Offshore Structures in Steep Waves”, Proc. OMAE2011, Rotterdam, The Netherlands.

Kaminski, M.L. and Bogaert, H., 2010, “Full

Scale Sloshing Impact Tests – Part 2”, Proc. ISOPE2010, Beijing, China.

Karimi M.R., Kosinski, C. and Brosset, L.,

2013,"Comparison of Sloshing Model Test Results at Scales 1:10 and 1:40", Proc. ISOPE2013, Anchorage, USA.

Kashiwagi, M. and Shi, Q., 2010, “Pressure

Distribution Computed by Wave-Interaction Theory for Adjacent Multiple Bodies”, J. Hydrodynamics, Vol. 22(5), pp.526-531.

Kawabe, H., Kwak, H. U., Park, J. and Ha, M.,

2010, “The Numerical and Experimental Study on Moonpool Water Surface Response of a Ship In Wave Condition”, Proc. ISOPE2010, Beijing, China.

Kim, J., Jaiman, R., Cosgrove, S. and

O’Sullivan, J., 2011, “Numerical Wave Tank Analysis of Wave Run-Up on a Truncated Vertical Cylinder”, Proc. OMAE2011, Rotterdam, The Netherlands.

322322

61

Kim, M.S., Jeong, H.S., Kwak, H.W., Kim, B.W. and Eom, J.K., 2012, “Improvement Method on Offloading Operability of Side-by-side Moored FLNG”, Proc. ISOPE2012, Rhodes, Greece.

Kim, T. and Kim, Y, 2013, “Numerical

Analysis on Floating-Body Motion Responses in Arbitrary Bathymetry”, Ocean Eng., Vol. 62, pp. 123-139.

Kim, H.J., Kwak, H.U., Lee, J.H., Kim, S.E.

and Seo, J.S. 2012, “Offloading Operability with Heading Control of Side-by-Side Moored FLNG and LNGC”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Kim, S., Kim, K. and Kim, Y., 2012,

"Comparative Study on Model-Scale Sloshing Tests", J. of Marine Sci. and Tec., Vol. 17, pp. 47–58.

Kim, H.T., Lee, D.Y. and Kwon, J.W., 2013,

“Numerical Simulations of Wave Run-Up on Single cylinder and Four Cylinders Using CFD Tools”, Proc. KTTC Daejeon Workshop on Wave Run-up.

Kim, Y.S., Sung, H.G., Kim, J.H, Hong, S.Y.

and Hong, J.P. 2012, “An Experimental Study on the Response of FSRU in Shallow Water in Comparison of Mooring Systems”, Proc. ISPOE 2012, Rhodes, Greece.

Koop, A. and Bereznitski, A., 2011, "Model-

Scale and Full-Scale CFD Calculations for Current Loads on Semi-Submersible", Proc. OMAE2011, Rotterdam, The Netherlands.

Korobkin, A.A. and Khabakhpasheva, T.I.,

2013, “Impact of Elastic Body on the Deep and Shallow Water”, Proc. OMAE2013, Nantes, France.

Kristiansen, T., Barrholm, R. and Stansberg, C.T., 2004, “Validation of Second-order Analysis in Predicting Diffracted Wave Elevation around a Vertical Circular Cylinder”, Proc. ISOPE2004, Toulon, France.

Kristiansen, T. and Stansberg, C.T., 2013,

“Empirical Prediction of Nonlinear Wave Diffraction and Run-Up on Vertical Columns”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Kurian, V.J., Ng, C.Y. and Liew, M.S., 2013,

"Effect of Short-crested Waves on the Dynamic Responses of Truss SPAR Platforms", Proc. ISOPE2013, Anchorage, USA.

Kyoung, J. and Yang, C.K., O’Sullivan, J. and

Miliante, T., 2013, “Validation of the HVS Semisubmersible Global Performance by Model Tests”, Proc. OMAE2013, Nantes, France.

Kurian, V.J., Yassir, M.A. and Harahap, I.S.H.,

2012, “Dynamic Response of Semisubmersibles with Damaged Mooring Lines”, Proc. ISOPE2012, Rhodes, Greece.

Lafeber, W., Bogaert, H. and Brosset, L., 2012,

"Comparison of Wave Impact Tests at Large and Full Scale: Results from the Sloshel Project", Proc. ISOPE2012, Rhodes, Greece.

Larsen, C.M., Zhao, Z.G. and Lie, H., 2012,

"Frequency Components of Vortex Induced Vibrations in Sheared Current", Proc. OMAE2012, Rio de Janeiro, Brazil.

Lee, H.W., Lee, D.Y., Kim, B.K., Shin Y. S.

and Choi, H.S., 2010, “A Motion Analysis of Two Floaters in Shallow Water using Boussinesq Equations”, Proc. ISPOE 2010, Beijing, China.

323323

Lee, K. and Yang, K.S., 2013, “Large Eddy

Simulation of Turbulent Flow Past a Circular Cylinder”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Lefevre, C., Constantinides, Y., Kim, J.W.,

Henneke, M., Gordon, R., Jang, H. and Wu, G., 2013, “Guidelines for CFD Simulations of Spar VIM”, Proc. OMAE2013, Nantes, France.

Lehn, E., 1980, “Thruster Interaction Effects”,

NSFI, Marintek, R102-80, Trondheim, Norway.

Lehn, E., 1981, “Thruster-hull Interaction

Effects”, NSFI, Marintek, R119-81, Trondheim, Norway.

Li, G., Kipp, R. and Leverette, S., 2012,

“Horizontal Installation of TLP Tendons”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Li, J., Wang, Z. and Liu, S., 2012,

“Calculations of Wave Run-up on Cylinder in Multi-Directional Focused Waves”, Proc. ISOPE2012, Rhodes, Greece.

Lie, H., Braaten, H., Jhingran, V.G., Sequeiros,

O.E. and Vandiver, K., 2012, “Comprehensive Riser VIV Model Tests in Uniform and Sheared Flow”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Lie, H., Braaten, H., Szwalek, J., Russo, M.

and Baarholm, R., 2013, “Drilling Riser VIV Tests with Prototype Reynolds Numbers”, Proc. OMAE2013, Nantes, France.

Loysel, T., Chollet, S., Gervaise, E., Brosset, L.

and De Seze, P.E., 2012, “Results of the First Sloshing Model Test Benchmark”, Proc. ISOPE2012, Rhodes, Greece.

Loysel, T., Gervaise, E., Morean, S. and Brosset, L, 2013, “Results of the 2012 – 2013 Slohing Model Test Benchmark”, Proc. ISOPE2013, Anchorage, USA.

Lu, L. and Chen, X., 2012, “Dissipation in the

Gap Resonance between Two Bodies”, Proc. IWWWFB2012, Copenhagen, Denmark.

Lu, L., Cheng, L., Teng, B. and Sun, L., 2010a,

“Comparison of Potential Flow and Viscous Fluid Models in Gap Resonance”, Proc. IWWWFB2010, Harbin, China.

Lu, L., Cheng, L., Teng, B. and Sun, L., 2010b,

“Numerical Simulation and Comparison of Potential Flow and Viscous Fluid Models in Near Trapping of Narrow Gaps”, J. Hydrodynamics, Vol. 22, pp.120–125.

Lu, L., Teng, B., Cheng, L., Sun, L. and Chen,

X., 2011, “Modelling of Multi-Bodies in Close Proximity under Water Waves - Fluid Resonance in Narrow Gaps”, Science China Physics, Mechanics & Astronomy, Vol. 54, pp.16–25.

Lugni,C., Bardazzi, A., Faltinsen, O. and

Graziani, G., 2013, “Hydroelastic Challenges for Wave-Impact Phenomena in Sloshing Flow”, Proc. OMAE2013, Nantes, France.

Ma, P., Qiu, W. and Spencer, D., 2012, "Time-

Domain VIV Prediction of Marine Risers", Proc. OMAE2012, Rio de Janeiro, Brazil.

Ma, Q., Yan, S. and Zhou, J., 2013, “Fully

Nonlinear Simulation of Resonant Wave Motion in Gap between Two Structures”, Proc. ISOPE2013, Anchorage, USA.

Mainçon, P. and Larsen, C.M., 2011, "Towards

a Time-Domain Finite Element Analysis of Vortex Induced Vibrations", Proc. OMAE2011, Rotterdam, The Netherlands.

324324

63

Mansour, A.M., Gordon, B.J., Ling, Q. and Shen, Q., 2013, “TLP Survivability against Progressive Failure of Tendon and Foundation Systems in Offshore Western Australian Harsh Environment”, Proc. OMAE2013, Nantes, France.

Mansour, A.M. and Kumar, D., 2013, “High

Performance Semisubmersible Design for Dry Tree Applications in Harsh Environment”, Proc. OMAE 2013, Nantes, France.

McNeill, S.I., 2012, “Spectral Formulations for

Vortex-Induced Vibration Modal Decomposition and Reconstruction”, J. Offshore Mech. and Arctic Eng., Vol. 134.

McNeill, S.I. and Agarwal, P., 2011, “Efficient

Modal Decomposition and Reconstruction of Riser Response due to VIV”, Proc. OMAE2011, Rotterdam, The Netherlands.

Minnebo, J., Izadparast, A.H., Duggal, A. and

Huijsmans, R.H.M., 2012, “Response-Based Analysis of FPSO Systems for Squall Loadings”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Modarres-Sadeghi, Y., Chasparis, F.,

Triantafyllou, M.S., Tognarelli, M. and Beynet, P., 2011, “Chaotic Response is Generic Feature of Vortex-Induced Vibrations of Flexible Risers”, J. Sound and Vibration, Vol. 330, pp.2565-2579.

Molin, B., Remy, F., Camhi, A. and Ledoux, A.,

2009, “Experimental and Numerical Study of the Gap Resonances in-between Two Rectangular Barges”, Proc. IMAM2009, Instanbul, Turkey.

Muehlner, E., Banumurthy, S. and Murray, J.,

2012a, “Effect of High-Frequency Response on TLP Tendon Fatigue”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Muehlner, E., Murray, J., Banumurthy, S. and Lakhotia, C., 2012b, "Comparison Between Numerical Prediction and Model Tests Measurements on the Centerwell Tank of a Radial Wellbay Spar", Proc. OMAE 2012, Rio de Janeiro, Brazil.

Murray, J., Muehlner, E., Banumurthy, S. and

Choi, G., 2012, "Model Tests on a Radial Wellbay Spar in Gulf of Mexico, Norwegian Sea and Offshore Brazil Environments", Proc. OMAE 2012, Rio de Janeiro, Brazil.

Myhr, A. and Nygaard, T.A., 2012, “Load

Reductions and Optimizations on Tension-Leg-Buoy Offshore Wind Turbine Platforms”, Proc. ISOPE2012, Rhodes, Greece.

Newman, J.N., 2003, "Application of

Generalized Modes for the Simulation of Free Surface Patches in Multi Body Hydrodynamics", WAMIT Consortium Report.

Nienhuis, U., 1986, "Simulations of Low

Frequency Motions of Dynamically Positioned Offshore Structures", RINA Spring Meeting, London.

Nienhuis, U., 1992, "Analysis of Thruster

Effectivity for Dynamic Positioning and Low Speed Manoeuvring", Ph.D. Thesis, Delft University of Technology.

Nordveit, R., Nygård, B. and Jullumstrø, E.,

2007, "Thrust Degradation in DP Operations: DP Model Test of an Aframaz Shuttle Tanker - Methods, Results, Operations", Dynamic Positioning Conference.

325325

Nuffel, D.V., Vepa, S., Baere, I., Degrieck, J.,

Rouch, J. and Paepegem, W.V., 2011, “Pressure Measurement on the Surface of a Rigid Cylindrical Body during Slamming Wave Impact”, Proc. OMAE2011, Rotterdam, The Netherlands.

Nuffel, D.V., Vepa, S., Baere, I., Degrieck, J.,

Rouch, J. and Paepegem, W.V., 2012, “Experimental Study on the Impact Loads Acting on a Horizontal Rigid Cylinder during Vertical Water Entry”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Ochi, M.K., 1958, “Model Experiments on

Ship Strength and Slamming in Regular Waves”, Trans. Royal Society of Naval Architects and Marine Engineers, Vol. 66, pp. 345-383.

Oliveira, M.C., 2012, “Selecting the Best

Heading for FPSOs in Santos Basin Considering the Wave Induced Motions”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Ottens, H. and van Dijk, R., 2012, "Benchmark

Study on Thruster-Hull Interaction in Current on a Semi-Submersible Crane Vessel", Proc. OMAE2012, Rio de Janeiro, Brazil.

Ottens, H., van Dijk, R. and Meskers, G., 2011,

“Benchmark Study on the Thruster-Hull Interaction on a Semi-Submersible Crane Vessel”, Proc. OMAE2011, Rotterdam, The Netherlands.

Palmer, A.R., Hearn, G.E. and Stevenson P.,

2009, “Thruster Interactions on Autonomous Underwater Vehicles”, Proc. OMAE2009, Honolulu, USA.

Pasquier, R. and Berthon, C., 2012, "Model Scale Test Vs. Full Scale Measurement: Findings from the Full Scale Measurement of Sloshing Project", Proc. ISOPE2012, Rhodes, Greece.

Passano, E., Larsen, C.M. and Lie, H., 2012,

“Comparison of Calculated In-Line Vortex Induced Vibrations Response to Model Tests”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Pauw, W. H., Huijsmans, R. H.M. and Voogt,

A., 2007, “Advances in the Hydrodynamics of Side-By-Side Moored Vessels”, Proc. OMAE2007, San Diego, USA.

Peng, S., Sun, H., Pan, J. Xia, Z. and Wu, W.,

2011, “Experimental Study and Numerical Simulation on Slamming of Trimaran”, Proc. ISOPE2011, Hawaii, USA.

Peng, Z., Wellens, P. and Raaijmakers, T., 2012,

“3-D Numerical Modeling of Wave Run-Up on Monopiles”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Pessoa, J, Fonseca, N. and Soares, C.G., 2013,

“Analysis of the First Order and Slowly Varying Motions of an Axisymmetric Floating Body in Bichromatic Waves”, J. Offshore Mechanics and Arctic Eng., Vol.135.

Pistani, F. and Thiagarajan, K.,

2012,”Experimental Measurements and Data Analysis of the Impact Pressures in a Sloshing Experiment”, Ocean Eng., Vol. 52, pp. 60-74.

Piterbarg, V.I., 1996, "Asymptotic Methods in

the Theory of Gaussian Processes and Fields", AMS Trans. of Math. Monographs, Vol. 148, Providence, R.I.

326326

65

Price, R., Zheng, H., Modarres-Sadeghi, Y. and Triantafyllou, M.S., 2011, "Effect of Higher Stress Harmonics and Spectral Width on Fatigue Damage of Marine Risers", Proc. OMAE2011, Rotterdam, The Netherlands.

Priyanto, A., Maimun, A., Ghani, M.P.A.,

Nasrudin, N. and Kader, A.S.A., 2012, “Recurrence Wave Run Up on a Large Volume Semisubmersible”, Proc. ISOPE2012, Rhodes, Greece.

Raghavan, K. and Bernitsas, M. M., 2011,

“Experimental Investigation of Reynolds Number Effect on Vortex Induced Vibration of Rigid Cylinder on Elastic Supports,” Ocean Eng. Vol. 38, pp.719-731.

Rahaman, M.M. and Akimoto, H., 2012,

“Analysis of the Mechanism of Slamming on the Bow Flare Region of a Container Ship Using RaNS CFD Method”, Proc. ISOPE2012, Rhodes, Greece.

Ramirez, J., Frigaard, P., Andersen, T.L. and

Christensen, E.D., 2011, “Numerical Modelling of Wave Run-up: Regular Waves”, Proc. ISOPE2011, Hawaii, USA.

Rao, D.S.B., Selvam, R.P. and Srinivasan, N.,

2012, “Hydrodynamic Analysis of Tension Based Tension Leg Platform”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Rao, Z., Vandiver, J.K. and Jhingran, V., 2013,

“VIV Excitation Competition between Bare and Buoyant Segments of Flexible Cylinders”, Proc. OMAE2013, Nantes, France.

Ren, N., Li, Y. and Ou, J., 2012, “The Wind-

wave Tunnel Test of a New Offshore Floating Wind Turbine with Combined Tension Leg-mooring Line System”, Proc. ISOPE2012, Rhodes, Greece.

¨

Resvanis, T.L. and Vandiver, J.K., "Modelling Risers with Partial Strake Coverage", Proc. OMAE2011, Rotterdam, The Netherlands.

Rijken, O., 2013, “Installation Methodologies

for a Tension Leg Platform under Ocean Swell Conditions”, Proc. OMAE2013, Nantes, France.

Rodriguez, C.A. and Neves, M.A.S., 2012,

"Nonlinear Instabilities of Spar Platforms in Waves", Proc. OMAE2012, Rio de Janeiro, Brazil.

Rudman, M. and Cleary, P.W., 2013, “Rogue

Wave Impact on a Tension Leg Platform: The Effect of Wave Incidence Angle and Mooring Line Tension”, Ocean Eng., Vol. 61.

Ruponen, P., Kurvinen, P., Saisto, I. and Harras,

J., 2012, “Air Compression in a Flooded Tank of a Damaged Ship”, Ocean Eng., Vol. 57, pp.64-71.

Shan, T., Li, X., Chen, G., Xiao, L., Lu, H. and

Li, J., 2012, “Leg Spacing Effect on Wave Run-up and Non-linear Wave Disturbance along Semi-submersible Columns”, Proc. ISOPE2012, Rhodes, Greece.

Smit, M.R., Tjallema, A.R. and Huijsmans,

R.H.M., 2011, “Current Feed Forward Control in Dynamic Positioning”, Proc. OMAE2011, Rotterdam, The Netherlands.

Song, Y.K., Chang, K.A., Ryu, Y. and Kwon,

S.H., 2013, “Flow Velocity and Impact Pressure in Liquid Sloshing”, Proc. OMAE2013, Nantes, France.

Song, J.S., Kim, H.J., Park, H.G. and Seo, J.S.,

2013, “The Investigation for Interaction Phenomenon of Azimuth Thruster of Ship”, Proc. PRADS 2013, Changwan City, Korea.

327327

Song, J., Teng, B., Tang G., Wu, H., Park, H.,

Lu, L. and Zhang, J., 2010, "Experimental Investigation on VIV Responses of a Long Flexible Riser Towed Horizontally in a Wave Basin", Proc. ISOPE2010, Beijing, China.

Souto-Iglesias, A., Botia-Vera, E., Martin, A.

and Pérez-Arribas, F., 2011, “A Set of Canonical Problems in Sloshing. Part 0: Experimental Setup and Data Processing”, Ocean Eng., Vol. 38, pp.1823–1830.

Srinath, V. and Chandrasekaran, S., 2013,

“Experimental Investigation of Dynamic Response of Tension Leg Platform with Perforated Members”, Proc. OMAE2013, Nantes, France.

Stansberg, C.T., Berget, K., Graczyk, M.,

Muthanna, C. and Pakozdi, C., 2012, “Breaking Wave Kinematics and Resulting Slamming Pressures on a Vertical Column”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Stansberg, C.T. and Kristiansen, T., 2011,

“Experimental Study of Slow-drift Ship Motions in Shallow Water Random Waves”, Proc. OMAE2011, Rotterdam, The Netherlands

Stewart, G., Lackner, M., Robertson, A.,

Jonkman, J. and Goupee, A., 2012, “Calibration and Validation of a FAST Floating Wind Turbine Model of the DeepCwind Scaled Tension-leg Platform”, Proc. ISOPE2012, Rhodes, Greece.

Sun, L., Chen, L., Zang, J., Taylor, R.E. and

Taylor, P.H., 2013, “Nonlinear Interactions of Regular Wave with Truncated Circular Column”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Sun, M. and Huang, W., 2012, “A New Concept of Spar in Deep Water and Its Hydrodynamic Performance under Internal Wave”, Proc. ISOPE 2012, Rhodes, Greece.

Swithenbank, S.B. and Larsen, C.M., 2012,

"Occurrence of High Amplitude VIV With Time Sharing", Proc. OMAE2012, Rio de Janeiro, Brazil.

Tannuri, E.A., Simos, A.N., Sparano, J.V. and

Matos, V.L.F., 2012, "Motion-Based Wave Estimation: Small-Scale Tests with a Crane-Barge Model", Marine Structures, Vol. 28.

Ten, I., Lu, L. and Chen, X., 2012, “A Semi-

Analytical Method with Dissipation for Fluid Resonances”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Tognarelli, M.A., Panicker, N., Campbell, M.

and Winterstein, S.R., 2013, "VIV Safety Factors for Drilling Risers: Propagating Model Uncertainty in a Long-Term Reliability Analysis", Proc. OMAE2013, Nantes, France.

Van’t Veer, R., Pistidda, A. and Koop, A., 2012,

“Forces on Bilge Keels in Regular and Irregular Oscillating Flow”, Proc. ISOPE2012, Rhodes, Greece.

van Daalen, E.F.G., Cozijn, J.L., Loussouarn, C.

and Hemker, P.W., 2011, “A Generic Optimization Algorithm for the Allocation of DP Actuators”, Proc. OMAE2011, Rotterdam, The Netherlands.

van Dijk, R.R.T. and Aalbers, A.B., 2001,

"What Happens in Water: The use of Hydrodynamics to Improve DP", Dynamic Positioning Conference, Houston, USA.

Vandiver, J.K., 2012, "A Damping Parameter

for Flow Induced Vibration", J. of Fluids and Structures, Vol. 35, pp.105-119.

328328

67

Veen, D.J. and Gourlay, T.P., 2011, “A 2D Smoothed Particle Hydrodynamics Theory for Calculating Slamming Loads on Ship Hull Sections”, Proc. High Speed Marine Vessels Conference, Fremantle.

Veen, D.J. and Gourlay T.P., 2012, “A

Combined Strip Theory and Smoothed Particle Hydrodynamics Approach for Estimating Slamming Loads on a Ship in Head Seas”, Ocean Eng., Vol. 43, pp.64-71.

Vepa, K.S., Nuffel, D.V., Paepegem, W.V.,

Degroote, J. and Vierendeels, J., 2011, “Comparative Study of Slamming Loads on Cylindrical Structures”, Proc. OMAE2011, Rotterdam, The Netherlands.

Wang, S., Luo, H. and Soares, C.G., 2012,

“Explicit FE Simulation of Slamming Load on Rigid Wedge with Various Deadrise Angles during Water Entry”, Maritime Tech. and Eng., pp. 399-406.

Wang, B., Shin, Y. and Wang, X., 2012,

“Reliability-based Sloshing Assessment of Containment Systems in LNGCs and FLNGs”, Proc. ISOPE2012, Rhodes, Greece.

Wang, S. and Soares, C.G., 2013, “Slam

Induced Loads on Bow-flared Sections with Various Roll Angles”, Ocean Eng., Vol. 67, pp. 45-57.

Wang, C., Wu, G. and Khoo, B.C., 2011, “Fully

Nonlinear Simulation of Resonant Motion of Liquid Confined Between Floating Structures”, J. Computers & Fluids, Vol. 44(1), pp. 89-101.

Watai, R.A., Matsumoto, F.T., Sparano, J.V., Simos, A.N. and Ferreira, M.D.A.S., 2011, “Wave Run-up Simulations with a Moving Large Volume Semi-submersible Platform”, Proc. OMAE2011, Rotterdam, The Netherlands.

Wen, P. and Qiu, W., 2013, “Numerical Studies

of VIV of a Smooth Cylinder”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Willert, C. and Gharib, M., 1992, "Three-

Dimensional Partical Imaging With a Single Camera", Experiments In Fluids, Vol. 12(6), pp. 353-358.

Xiang, X. and Faltinsen, O. M., 2011, “Time

Domain Simulation of Two Interacting Ships Advancing Parallel in Waves”, Proc. OMAE2011, Rotterdam, Netherlands.

Xu, Q., 2011, “A New Semisubmersible Design

for Improved Heave Motion, Vortex-induced Motion and Quayside Stability”, Proc. OMAE2011, Rotterdam, The Netherlands.

Xu, Q., Kim, J., Bhaumik, T., O’Sullivan, J.

and Ermon, J., 2012, “Validation of HVS Semisubmersible VIM Performance by Model Test and CFD”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Xu, S., Wang, X., Wang, L. and Li, J., 2013,

“Dynamic Positioning with Roll-pitch Motion Control for a Semi-Submersible”, Proc. ISOPE2013, Anchorage, USA.

Xu, X., Yang, J., Li, X. and Lu, H., 2013,

“Wave Drift Forces on Three Barges Arranged Side by Side in Floatover Installation”, Proc. OMAE2013, Nantes, France.

329329

Yamada, Y., Takami, T. and Oka, M., 2012,

“Numerical Study on the Slamming Impact of Wedge Shaped Obstacles Considering Fluid-structure Interaction (FSI)”, Proc. ISOPE2012, Rhodes, Greece.

Yan, H. and Liu, Y., 2011, “Nonlinear

Computation of Water Impact of Axisymmetric Bodies”, J. Ship Research, Vol. 55, No. 1, pp. 29-44.

Yang, Q. and Qiu, W., 2012, “Numerical

Simulation of Water Impact for 2D and 3D Bodies”, Ocean Eng., Vol. 43, pp. 82-89.

Ye, H., Zhong, Q. and Wan, D., 2013,

“Benchmark Computations of Flows around a Stationary Cylinder with High Reynolds Number by RANS- Overset Grid Approach”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Yeon, S.M., Yang, J. and Stern, F., 2013,

“Large Eddy Simulation of Drag Crisis in Turbulent Flow Past a Circular Cylinder”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Yiannis, C., Raghavan, K., Karayaka, M. and

Spencer, D., 2013, "Tandem, Riser Hydrodynamic Tests at Prototype Reynolds Number", Proc. OMAE2013, Nantes, France.

Yin, D. and Larsen, C.M., 2011, “Experimental

and Numerical Analysis of Forced Motion of a Circular Cylinder”, Proc. OMAE2011, Rotterdam, The Netherlands.

Yin, D. and Larsen, C.M., 2012, “Forced

Motion Experiments with Measured Motions from Flexible Beam Tests under Uniform and Sheared Flows”, Proc. OMAE2012, Rio de Janeiro, Brazil.

Yoon, S.H., Kim, D.H., Sadat-Hosseini, H., Yang, J. and Stern, F., 2013, “High Fidelity CFD Simulation of Wave Run-up around Vertical Cylinders in Monochromatic Waves”, Proc. ITTC Nantes Workshop on Wave Run-up and Vortex Shedding, Nantes, France.

Zeng, H., Li, X., Chen, J., Chen, M., Tan, J.,

Jing, Y., Shi, S., Li, Z. and Li, X., 2012, “TUMSAS: a New Solution for Shallow Water FPSO Mooring”, Proc. OTC2012, Houston, USA

Zhang, Z., Kim, M.H., Kang, H.Y., Zong, Y.

and Wu, J.F., 2012, “A Study on SPM Mooring System for Side-by-side Two Vessels”, Proc. ISOPE2012, Rhodes, Greece.

Zhang, Z., Sun L., Ma, Q.W. and Xu, G., 2013,

“A Numerical Investigation on Hydrodynamics of Two Floating Bodies of Arbitrary Arrangements in Regular Waves”, Proc. ISOPE2013, Anchorage, USA.

Zhang, H., Yang, J., Peng, T. and Lu, H., 2012,

"Experimental Investigation on Added Mass Coefficient of a Truss Spar Subjected to Vortex-Induced Motions", Proc. OMAE2012, Rio de Janeiro, Brazil.

Zhao, M., Cheng, L. and An, H., 2012,

"Numerical Investigation of Vortex-Induced Vibration of a Circular Cylinder in Transverse Direction in Oscillatory Flow", Ocean Eng., Vol. 41, pp. 39-52.

Zhao, R., Faltinsen, O.M. and Aarsnes, J.V.,

1996, “Water Entry of Arbitrary Two-Dimensional Sections With and Without Flow Separation”, Proc. 21st Sym. on Nav. Hydro., pp. 408-423

330330

69

Zhao, M., Kaja, K., Xiang, Y. and Yan, G. 2013, "Vortex-Induced Vibration (VIV) of a Circular Cylinder in Combined Steady and Oscillatory Flow", Ocean Eng., Vol. 73, pp. 83-95.

Zheng, H., Price, R., Modarres-Sadeghi, R.,

Triantafyllou, G. and Triantafyllou, M., 2011, "Vortex-Induced Vibration Analysis (VIVA) Based on Hydrodynamic Databases", Proc. OMAE2011, Rotterdam, The Netherlands.

Zhu, H., Zhu, R. and Miao, G., 2008, “A Time

Domain Investigation on the Hydrodynamic Resonance Phenomena of 3-D Multiple Floating Structures”, J. Hydrodynamics, Vol. 20, pp. 611–616.

Zou, L. and Larsson, L., 2013, “Numerical

Predictions of Ship-to-Ship Interaction in Shallow Water”, Ocean Eng., Vol. 72, pp. 386-402.

331331

Stability in Waves Committee

Final Report and Recommendations to the 27th ITTC

1. INTRODUCTION

1.1. Membership and Meetings

Membership. The Committee appointed by the 26th ITTC consisted of the following members:

Dr. A. M. Reed (Chairman)

Carderock Division, Naval Surface Warfare Centre (NSWCCD), USA

Mr. A. Peters (Secretary) QinetiQ, Haslar, UK

Professor W. Y. Duan Harbin Engineering University, China

Assoc. Professor P. Gualeni University of Genoa, Italy

Assoc. Professor T. Katayama Osaka Prefecture University, Japan

Dr. G. J. Lee Korea Research Institute of Ships & Ocean Engineering (KRISO), S. Korea

Dr. F. van Walree Maritime Research Institute Netherlands (MARIN), the Netherlands

The committee would like to acknowledge the valuable contributions of wave data to the reviews from MARIN; and the work from Joel Park and John Telste from NSWCCD for their

contributions on uncer- tainty and extreme waves, respectively.

Meetings. Four Committee meetings were held as follows: Osaka, Japan - February 2012 Athens, Greece - September 2012 Washington, D.C, USA - June 2013 Daejeon, Korea - March 2014

1.2. Tasks from the 26th ITTC Update the state-of-the-art for predicting ship stability in waves, emphasizing develop-ments since the 2011 International Towing Tank Conference (ITTC). The committee re-port should include sections on:

a. Definition of loss and survival of a ship (particularly damaged ships);

b. The amount of detail required for modelling the internal geometry of a ship;

c. Leak and collapse pressures for water tight doors and bulkheads; and

d. Importance of taking air pressure into account (how open or closed compart-ments are in ships ties into item b above)

332332

e. Modelling of extreme wave conditions.

Review ITTC Recommended Procedures relevant to stability and

a. Identify any requirements for changes in the light of current practice, and, if approved by the Advisory Council, up-date them.

b. Identify the need for new procedures and outline the purpose and content of these.

Investigate uncertainty analysis for intact and damaged model tests to complement cur-rent procedures (Uncertainty in making meas-urements, and technical means that are used). Investigate the criteria for modelling wave spectra in the determination of dynamic instability of intact vessels [Stability failures in the International Maritime Organization (IMO) sense: pure loss of stability, parametric roll, broaching, dead ship condition (resonant roll in beam seas)], i.e., wave steepness, non- linearity, frequency contents of the spectrum, statistical distribution of wave and crest height and spatial behaviour of the waves and non-linear wave kinematics.

Develop better understanding of uncer-tainties associated with the results from experi-ments and simulations of extreme motions of intact vessels in realistic irregular seaways and develop quantitative techniques which reflect the nature and magnitude of the phenomena. Review vulnerability criteria (including long term probability of loss of the ship) for intact and damaged ships, and outline further developments that are required. [Directly tied to on-going IMO Sub-committee on Sta-bility, Load Lines & Fishing Vesseal Safety (SLF) actions]

Update ITTC Recommended Procedure 7.5-02-07-04.2, Model Tests on Damage Sta-bility in Waves, paying specific attention to:

a. Investigate the significance of scale ef-fects in air pressure on flooding-model tests under atmospheric conditions. Comment on the need to perform flooding-model tests under scaled air pressure conditions.

b. Investigate how to deal with inertia due to the floodwater mass.

Investigate roll damping for large-amplitude roll motions in irregular seas. Re-view suitable data for future benchmarking of time-domain computer codes.

a. Time-domain roll damping in irregular waves´

b. Modelling of hydrodynamics of large-amplitude roll motion (regular and irregular seas)

Cooperate with the IMO SLF subcommit-tee correspondence group and the ITTC Sea-keeping Committee. 2. STATE-OF-THE-ART REVIEW 2.1. Review

During the past few years major efforts

have been on-going in ship stability research. The most well known references in this area are the International Stability Conference and Workshops. The last Stability Conference occurred in Athens, Greece in 2012 (STAB, 2012) and the last two stability workshops occurred in Washington, D.C, USA, in 2011 (ISSW, 2011) and another in Brest, France in

333333

2013 (ISSW, 2013) The focus of this state-of-the-art review is enumerated in Steps a–e of Section 1.2.

2.2. Definition of Loss and Survival of a

Ship

In any structured framework or methodol-ogy aimed at assessing ship safety, an accurate definition of the boundary between survival and loss is necessary.

The whole assessment methodology is yet

to be properly defined and validated (Peters, et al. 2012) in order to recognise in a reliable way the possible ship-specific weakness in that term.

The term “ship loss” is commonly used as

a statement of an undesired event, but the same expression can be used to describe many different scenarios. The opposite statement “ship survival” is also regularly used, but also suffers from the same problem—lack of a pre-cise definition of the situation.

In current literature, a trend has been ob-

served to mention the concept of “ship loss” when dealing with an intact ship, while “ship survival” is more likely to be used when dis-cussing the safety of a damaged ship.

2.3. Relationship between Loss and

Survival

Detailed examination is required of the definitions and relationship between “ship loss” and “ship survival” in order to avoid redundancies, overlapping concepts or contra-dictions.

In terms of probabilistic definition, surviv-ability, PS, is the combination of susceptibility, PH, the inability to avoid an undesired event or a related initiating event and vulnerability, PK/H, the inability to withstand the effect of an undesired event (Ball & Calvano, 1994). Therefore, survivability is defined as :

PS 1 PH PK /H .

If susceptibility and vulnerability are the inability to avoid or withstand, respectively, the effect of a certain situation, their combina-tion is defined as the mathematical comple-ment to survivability, i.e., the ability to sur-vive.

Susceptibility in and of itself is a complex

concept to fully understand and model. In the case of a damaged ship, for example, it might correspond to the probability that a ship will be hit by another ship. For an intact ship, this might correspond to the probability that a ship is caught in a severe storm.

Vulnerability represents the probability of

severe consequences or even total loss of a vessel when an undesired initiating event has occurred.

We can assume that “loss” is an extreme

negative consequence given a certain unde-sired initiating event. In this perspective it can be considered the mathematical comple-ment to survivability.

2.4. Definition of Loss

The loss of a ship is an expression that, in addition to an explicit negative connotation, can be used to indicate many different levels of severity of a situation. One approach would be to decide to focus only on sinking

334334

and/or capsizing, the latter defined as the transition to another stable equilibrium, other than upright, which is intrinsically unsafe. It is evident that from a safety point of view, some other intermediate levels of undesired severe situations should be taken into account.

During the recent IMO activities regarding

the development of second- generation in-tact-stability criteria, a new terminology was identified, i.e., “intact-stability failure” (IMO, 2008). This is defined as “a state of inability of a ship to remain within design limits of roll (heel, list) angle and a combination of rigid body accelerations”. A “total stability failure” and a “partial stability failure” are defined below:

Total Stability Failure — Capsizing, being

the total loss of a ship’s operability with likely loss of lives.

Partial Stability Failure — The occurrence of very large roll (heel, list) angles and/or excessive rigid body accelerations, which will not result in loss of the ship, but which would impair normal operation of the ship and could be dangerous to crew, passen-gers, cargo or ship equipment.

It is immediately evident that besides the

well-known concept of ship loss coinciding with ship capsize, it is important to discuss scenarios where the roll angles exceed a pre-scribed limit; and where the combinations of lateral and vertical accelerations exceed pre-scribed limits.

The so-called prescribed limits of roll an-

gles can be fixed in absolute terms (e.g., 45 degrees, 30 degrees) or other less precise terms (e.g., deck-edge immersion or immersion of some defined critical point like the down- flooding openings) (Bačkalov, 2012).

The adoption of a fixed absolute roll-angle value as a limit to define a capsizing event is very common in literature, even if it is well recognised that this, in principle, might change from ship to ship due to the different dynamics of each ship. Beaupuy, et al. (2012) suggest that this aspect should be investigated by as-suming that the critical threshold is a percent-age of the angle of vanishing stability of each ship. Another possible event of partial failure, cargo shift, is mentioned in Kubo, et al. (2012).

In Kobyliński (2006), the concept of a

loss-of-stability accident (LOSA) was intro-duced as a better description of the situation that occurs in reality, instead of talking about just a capsizing event. Kobyliński referred to a prolonged discussion on the definition of capsizing during the second International Con-ference on Stability of Ships & Ocean Vehi-cles (STAB) conference in 1982. He pro-posed that capsizing be defined as a situation where amplitudes of rolling motion or heel exceed a limit that makes operation or han-dling a ship impossible for various reasons (loss of power, loss of manoeuvrability, neces-sity to abandon the ship). Kobyliński’s pro-posed definition of capsizing did not neces-sarily assume the ship taking the inverted posi-tion. Therefore, capsizing might be better defined as LOSA and the definition might also be suitable for use in assessing the risk of cap-sizing.

LOSA can be divided into subcategories to

cover the different types and severity of loss, i.e., sudden capsizing, large heel with loss of power and manoeuvrability, large heel with progressive flooding and eventually capsizing or foundering.

In the case of a damaged-ship scenario, a

reference is often made to a critical limiting- heel angle to define loss of a vessel. A 45-

335335

degree mean angle was used by Spanos & Pa-panikolaou (2012). Alternatively, for Roll On-Roll Off (RO-RO) passenger ships, the procedure derived from the Directive 2003/25/EC is used where a ship is regarded as capsized if the roll angle exceeds 30 degrees instantaneously or if the steady (mean) heel angle is greater than 20 degrees for a period longer than three minutes (Kwon, et al., 2012). In the same paper, the importance of the ship structures condition for a damaged ship is dis-cussed as well: the rapid deterioration and degradation of the structural integrity might become important as much as stability defi-ciency for some types of ships.

The concept of critical limiting heel angle

was also discussed by Montewka, et al. (2013) where the loss of the Roll On Passenger (ROPAX) is expected if two consecutive limit-ing states are exceeded, namely crashworthi-ness and stability. In application, ship cap-sizing is assumed to occur when 60 degrees of roll angle is exceeded.

It should be recognised that intact and

damaged ships have some basic analogies when defining ship loss.

When considering the dynamic behaviour

of a damaged ship in a seaway, the threshold definition should be treated in line with the intact-ship approach, i.e., recognising the con-cepts of total loss or partial loss dealing with roll angles, accelerations, and immersion of critical points. These concepts need to be discussed within specific restrictions in rela-tion to the residual operational capability re-quired for a ship.

As regards the sinking phenomenon, this is

generally applicable to a ship with damage to her hull, leading to a significant ingress of water and a consequent reduction of the re-serve of buoyancy. For an intact ship, water

might enter from unprotected openings, which could be just as critical.

In line with the treatment of stability fail-

ure, it might be possible in principle to con-sider a “total loss of buoyancy” and a “partial loss of buoyancy”. Partial loss of buoyancy can be defined as a situation that will jeopard-ize the normal operations of a ship and its crew, or present a possibly critical situation for passengers, cargo or ship equipment.

Therefore, a situation other than the total

sinking of a ship should be read in terms of residual buoyancy and equilibrium waterline characteristics.

A possible combination of different

measures of various safety elements synthe-sised in a Relative Damage Loss Index (RDLI) is applied by Peters & Wing (2009) allowing a more comprehensive evaluation of a ship’s damage performance. 2.5. Loss of Functional Capability

The rule-making framework for ship safety

is currently focused towards goal-based standards. With goal-based standards, func-tional requirements must be complied with in order to meet the overall goal. IMO has al-ready agreed in principle with the following goal, valid for all kinds of new ships: “Ships are to be designed and constructed for a speci-fied design life to be safe and environmentally friendly.” (IMO, 2005).

This implies that a ship must have char-

acteristics adequate to minimise the risk of loss of the ship.

This new approach tends to avoid pre-

scriptive standards in favour of rules referring

336336

to safety goals, with an identified level of per-formance, but without specifying the means of achieving that level (Kobyliński, 2012).

It is, therefore, important to focus on the

functional capabilities that are vital for a ship so that the “loss of ship” definition can coin-cide with the loss of such functional capabili-ties.

For a ship in the intact condition, it is rea-

sonable to define the main list of functional capabilities as:

Buoyancy Watertight integrity Stability Navigation Some specific operational and systems

activity

Traditionally, for a ship in a damaged condition, reserve buoyancy and stability are the key desirable functions while possibly ac-cepting a degraded level of performance. The most important issue is the ability to per-form the evacuation and the emergency proce-dures, but some other key activities might also be required, for example returning to port un-der your own power.

IMO has recently introduced the regulatory concept of safe return to port (Spanos & Papanikolaou, 2012) through the International Convention for the Safety of Life at Sea (SO-LAS) Regulation II-1/8-1, where a passenger ship shall be designed so that key specified systems remain operational when a ship is sub-ject to flooding of any single watertight com-partment.

A passenger ship is deemed capable of re-

turning to port, when key functions and sys-tems such as propulsion, navigation, and es-sential hotel services remain operational.

The orderly evacuation and abandonment

of a ship, therefore, becomes a secondary op-tion only to be employed if the casualty thresh-old is exceeded. In this case the issue of en-ergy production and distribution is another functional capability that should be considered as a key to safe abandonment, when defining the concept of ship loss or survival. 2.6. Internal Geometric Modelling

The European Union (EU) Integrated Flooding Control & Standards for Stability & Crisis Management Project (FLOODSTAND) (Naar & Vaher, 2010) was a European research project which set out to derive detailed data on flooding mechanisms to validate numerical simulation tools and to help develop a standard for damaged-ship stability, focussing on the risk of flooding.

The modelling of internal geometry and

effects on stability modelling have been stud-ied by Karlberg, et al. (2011) as part of the FLOODSTAND project. As described in this report, the routes floodwater takes as it pro-gresses inside a ship and the order in which compartments fill can have a significant effect on the consequent motions and events onboard, and in some cases on the final flooded state of a vessel.

In large passenger ships the internal layout

of a vessel is typically characterised by water-tight subdivisions such as double bottoms, watertight bulkheads and bulkhead decks. More specifically, it also includes the decks and significant non-watertight subdivisions, which make up the corridors and cabins, etc. This complex internal structure makes it chal-lenging to model the damage stability of a pas-senger vessel both numerically and physically.

337337

The progress of flooding is typically char-

acterised by the amount of incoming floodwa-ter and how it is distributed in a ship. The distribution of floodwater affects sinkage, heel and trim of a ship, and consequently the stabil-ity, which are the most important factors gov-erning the survivability of a vessel. With complex subdivisions and multiple damage openings, very small changes in floodwater flow can result in various possibilities for a flooding sequence. Therefore, it is not al-ways straightforward to predict with certainty what the final flooded state of a vessel will be.

The use of time-domain flooding simula-

tion tools has expanded in recent years with the increase in available computer capability. It is well known that simulation results depend on applied input data for ship openings. The leakage and collapsing of non-watertight struc-tures, such as closed fire doors, can have a very remarkable effect on the time-to-flood calculations.

The main objective of the Work Package 2

in the FLOODSTAND project was to provide data for more accurate and realistic modelling of progressive flooding in time- domain simulations. In the study, both experimental and numerical studies were performed in order to develop guidelines on modelling leaking and collapsing structures for use in flooding simulation. Furthermore, discharge coefficients for water flow through typical openings were evaluated.

It was clear that the exact values for dis-

charge coefficients for leakage through a closed door cannot be evaluated for each open-ing in a large passenger ship. Therefore the discharge coefficients that are used have to be based on approximations and estimates.

The effect of variations in the input data on the results of a flooding simulation was studied through systematic sensitivity analysis with three different damage scenarios. The results indicate that the effect of these flow coefficients and collapse pressures on transient heeling in the beginning of flooding is mini-mal. However, the parameters were found to have a notable effect on the time-to-flood. A higher critical collapse pressure was found to significantly slow down the flooding process. The leakage area ratio also was found to have a significant effect on the time-to-flood, espe-cially in a flooding case where closed doors do not reach collapse. In a flooding case where most of the flooding was simulated as leakage through closed doors, the applied leakage area ratios had a dominant effect on time-to-flood. Underestimation of this coefficient by 50% was shown to potentially lead to a 50% over-estimation in the time-to-flood according to the FLOODSTAND report.

It was also found during the study that

during simulations, variations of critical pres-sure head for collapse had a significant effect on the way the flooding progressed, and thus had an effect on the overall flooding rate and the time-to-flood. These also affected the resulting vessel-heeling behaviour, and in turn, the flooding rate. In the early flooding phases, leakage modelling was shown to have a clear effect on the time-to-flood. 2.7. Leak and Collapse Pressures of

Water- tight Doors and Bulkheads

Part of the FLOODSTAND Project fo-cussed on work to investigate flooding through watertight doors and hatches.

Utilising full-scale testing for the leakage

and collapse of watertight doors and bulkheads,

338338

real-time data was collected to help develop simulation tools. Full-scale bending, tensile strength and compressive tests were carried out on individual wall-panel materials. Nu-merical studies and simulation analyses were carried out using Finite Element Methods and Computational Fluid Dynamics in order to give a comparison between experimental and numerical data.

Through extensive simulations for different

damage scenarios, the FLOODSTAND research established some guidelines for mod-elling these structures during progressive flooding.

During the FLOODSTAND physical tests

undertaken by the Centrum Techniki Oretowej (CTO) in Gdansk, Poland, water-pressure head was gradually increased at 0.5 m increments until the test object was damaged, the water- flow-rate value exceeded a critical value of 90 litres/second, or the critical pressure was above 220 kPa. Measurements included leakage-flow rate, deflection of the test object at six points and pressure head at structural failure.

Twenty different types of doors, windows,

walls and hatch configurations were tested, including;

Class A-60 double leaf-hinged-marine

fire door, Class B joiner door — hinged, Steel frames for Class B wall and cabin

wall, Steel frames for cabin wall, Cool Room sliding door, Semi Watertight Door--sliding steel

frames, Cross flooding hatch, Sliding door and Hinged door.

The point of collapse or maximum flow rate was found to be dependent on the type, material and construction of a door and frame. Due to the fact that pressure was at its highest at the bottom of a door panel, structural defor-mation and structural leakage to the lower door hinges and sills occurred.

A key finding from these experiments was

that for many doors, the leakage-area ratio in-creased almost linearly as a function of the pressure head. For example, the results ob-tained for a light watertight door showed that the leakage through the test door started at a water height of about 2 m and had leakage of less than 1.0 litre/second until structural dam-age occurred at a pressure head of about 8.0 m. Even after significant structural failure, the leakage through the door was approximately 40 litres/seconds, which corresponded to a leakage-area ratio of 0.017.

A Class A60 double leaf-hinged door, a

Class A60 sliding door and a cabin wall were also modelled. The panel bending tests showed a good correlation with the Finite Ele-ment Method (FEM) analysis. Generally, the ability to compare FEM results with physical tests was limited. The doors tested had a leakage rate too high to be comparable with FEM results. It was found in the study that the use of analytical methods is not always justified, as door failure often depends on the strength of the hinges as opposed to the strength of the main body of the door.

For many doors it was found that the as-

sumption that the leakage area is evenly dis-tributed vertically is not valid as there is often a gap between the bottom of the door and the sill.

The FLOODSTAND study concluded that

different categories of doors behave very dif-ferently under flooding conditions and even

339339

the same door within the same category can behave very differently as the gap between the sill and the door can vary considerably. Gen-eral guidelines were presented for both Class A and B doors, but it was noted that significant further work is needed, including further phys-ical testing. With Class A structures, sen-sitivity analysis was recommended to consider the variability in the results, while Class B structures were found to fail at water levels lower than expected.

3. UNCERTAINTY ANALYSIS FOR IN-

TACT AND DAMAGED MODELS

USED IN SEAKEEPING AND

EXTREME MOTION TESTS1

The results of seakeeping and extreme motions testing are the characteristics of sto-chastic processes in random seas. As such, there is no uncertainty to be reported in the results, but rather confidence bands on the sta-tistics characterizing the results of the experi-ment. The statistical uncertainty of seakeep-ing and extreme motions in a seaway will be discussed in Chapter 5, which follows.

In seakeeping and extreme-motions exper-

iments, the area where traditional deterministic uncertainty analysis applies is in determining the mass properties of the model being tested. Documentation of surface-ship-model tests usually includes tables of the results but does not explicitly include the equations in the bal-lasting process or the instrumentation.

Given the uncertainty range on the mass

properties of a seakeeping or extreme-motions test, the ideal approach would be to repeat the experiment with the model ballasted to the

1 This section is based largely on unpublished notes by Dr. Joel Park of DTMB (NSWCCD).

extremes of uncertainty of the mass properties to determine the impact of this uncertainty on the experimental results. Technically the above approach is impossible as there would again be uncertainties associated with the mass properties for these new tests. The only fea-sible approach to determining the impact of the uncertainties in mass properties on the uncertainties of experimental results appears to be computational, although there is no estab-lished procedure.

From a practical perspective, the use of a

validated linear seakeeping code is the most realistic approach to solving the above prob-lem, as it will allow rapid assessment of the impact of the various mass-properties un-certainties in various combinations on the measured motions. Although a linear code will have its own accuracy issues, it will pro-vide a consistent metric against which the im-pact of mass-properties uncertainties can be judged. Also, as a linear code provides a low-est common denominator, it will allow realistic comparisons between various experimental facilities and organizations without introduc-ing many computational tool variables into the assessment.

The material that follows outlines the

equations typically used to determine the mass properties of a model for seakeeping and extreme-motions testing, and derives the uncertainty equations for ballasting based on ISO GUM (JCGM, 2008) and ITTC (2008).

3.1. Model Weight and Mass

The formulation is from the Archimedes principle; that is, a ship’s weight is equal to its buoyancy force. In that case, the equation for a ship or model weight is given by:

340340

W g (1)

where is water density, g is local acceleration due to gravity, and is the displaced volume. From (1), the model weight is then computed as:

)/( ssmmmsm gmW (2)

where m is mass and the subscripts m and s are for the model and ship, respectively. How-ever, the displaced volume is related to the scale ratio by:

3ms / (3)

The scale ratio is defined as the ratio of the

ship length to the model length:

Ls / Lm

From (2) and (3), the model weight is then

Wm msmgm / (s3) (4)

In mass units, (4) becomes

mm msm / (s3) (5)

The calculation of the model weight and

mass from (4) and (5) should be computed on the basis of the standard values for s

(1026.021 kg/m3 for salt water at 15° C) and gc (9.80665 m/s2) for full scale, and the

values of m (generally fresh water) and gm

appropriate for the experimental facility. Standard gravity is fixed at an internationally accepted value of 9.80665 m/s2 from Thompson & Taylor (2008).

The values for freshwater and seawater for

standard field and laboratory conditions can be

found in international standards. The sea-water values are in TEOS-10 (IOC, SCOR & IAPSO, 2010), and freshwater values are in Harvey, et al. (2008) and IAPWS (2008). The uncertainty in density may be computed from the measured temperature and salinity.

From (4) and (5), the expanded relative

uncertainty in weight and mass is as follows:

UWm / Wm (Um / m )2 (3ULm / Lm )2

(Ugm / gm )21

2 (6)

2mm

2mmmm )/3()/(/ LUUmU Lm (7)

After the model weight and mass are ad-

justed to the values from (4) and (5), the model must be weighed. After the model is weigh-ed, the combined uncertainty in model mass includes the result of the measured weight and the computed weight. The combined uncer-tainty is then:

2m

2measc mUUU (8)

The final measured weight and mass as

computed from (4) and (5) should be within the uncertainty of (8). The uncertainty esti-mate in density for (6) and (7) should be the maximum difference between the value ap-plied during ballasting and value measured during testing. The uncertainty in the model length should be obtained from direct meas-urements of the model dimensions while the uncertainty in g is from an internationally recognized standards organization. 3.2. Longitudinal Centre of Gravity

The remaining procedures described here

require suspension of the model from a struc-

341341

turally rigid frame. The process includes a beam to which the model is attached.2 The beam is attached to the frame by a pivot point. The determination of the longitudinal centre of gravity (LCG) is a two-step measurement pro-cess:

The CG of the beam is measured. The CG of the beam and model is

measured. The CG of the model is then computed

from the previous two steps.

One method in the determination of the CG is simply to move the beam, or the beam and model combination until the beam is level. In this case, the CG is directly below the pivot. However, the uncertainty in the CG location by levelling may be unreliable. A more di-rect method is suspension of the model at two points: one near the bow and the second near the stern. The location of the CG is then computed from the moments and the uncer-tainty is easily established. The load at the bow and stern is measured with electronic load cells attached to the suspension cables.

3.2.1. Levelling Method The simplest method for locating the LCG

may be by moving the beam alone under the pivot until it is level, and then moving the model on the beam until the model on the beam is level. When the model is level, the LCG is located directly below the pivot point. Any deviation of the LCG is given by:

zx /tan .

where is the pitch angle, x is deviation from the true LCG, and z is the vertical distance

2 The beam is not required for all procedures, but is required for some, those for which the beam is not necessicarily required will be noted.

from the pivot point to the CG. For a pitch angle near zero, the result is

zx / (9)

Calculation of the location of the vertical CG is described in the following section.

From (9), uncertainty in the displacement

from the true LCG is:

Ux (Uz)2 (zU )2 (10).

For a pitch angle near zero, (10) becomes:

zUU x .

The combined uncertainty relative to the

model reference point for the LCG is then

2m

2)( xc UzUU .

If the instrument for measurement of level

is removed from the fixture, an equivalent weight should be located at the measurement point.

3.2.2. Two-point Suspension Method

The LCG location can be measured by supporting the beam at two points (1, near the stern, and 2, near the bow), suitably far apart with the LCG somewhere in-between (the beam is not necessary for this). At both points there should be a load cell measuring the weight at that point. The model should also be levelled. The weights from the load cells are as follows:

For the beam

21b FFW (11).

342342

For the beam and model

43bh FFW (12).

From (11) and (12), the weight of the hull

is

)()( 2413h FFFFW (13)

where F1 and F3 are the load cell readings at the stern and F2 and F4 are at the bow.

Similarly, the moments are as follows: For the beam

2211bb FxFxWx (14).

and x1 is the distance from the reference loca-tion (say amidships) to the aft suspension ca-ble (negative aft) and x2 is the distance from the reference location to the forward suspen-sion cable (positive forward). For the beam and hull

4231bhbh FxFxWx (15).

From (13)–(15), the CG of the model

relative to the reference location is

xh (x1F31 x2F42 ) / (F31 F42 )

(x1F31 x2F42 ) / Wh

(16)

where

1331 FFF

2442 FFF For the uncertainty estimates, the sensitiv-

ity coefficients from (16) are as follows:

hWFFFFxxc /)/(/ 314231311h1

h424231422h2 /)/(/ WFFFFxxc

2h214231h3 /)(/ WxxFFxc

2h213142h4 /)(/ WxxFFxc .

The distances x1 and x2 are likely measured

with the same device, and the uncertainty in the distance will be the same and correlated. Similarly, the load cells for the measurement of the aft and forward locations may have the same uncertainty. If they are calibrated at the same time with the same equipment, then the load measurements are also correlated. The uncertainty in the location of the LCG is as follows:

22

4322

21m )()( Fxx UccUccU .

In tests where equipment will be added to

the model later, the LCG of the model hull is determined by either of the previous methods, and components are added. Mass properties of the smaller components are measured with a mass properties instrument. The total weight of the model then is

n

iiWW

1m (17),

where n is the number of components. The LCG is given by

n

iii WWxx

1mm )/( (18).

The uncertainty in the weight is

n

iWiW UU

1

22m (19),

and the uncertainty in the LCG is

343343

Uxm2 (WiUxi / Wm )2

i1

n

(xiUWi / Wm )2

i1

n

(xiWiUWm / Wm2)2

i1

n

(20)

The previous formulation assumes the

measurements are independent; however, some of the measurements are correlated. The hull weight and LCG are independent of the meas-urements by the mass-properties instrument, although the measurements of the components by the mass-properties-instrument are corre-lated. In any case, the uncertainty from the hull measurements will be the dominant term in the estimate.

If both the levelling and two-point suspen-

sion methods are applied in the determination of LCG, the result should be within the uncer-tainty estimates of both methods. Estimates using both methods indicate a discrepancy in LCG location by the levelling method in com-parison to the two-point load method. The difference is larger than the uncertainty esti-mates on the location of the LCG. This illus-trates the difficulty in getting accurate results using the levelling method.

3.3. Vertical Centre of Gravity

The vertical centre of gravity (VCG) is

determined by the added weight or inclining method in air. In added weight or inclining method, a weight is added or moved trans-versely, respectively, resulting in a heeling moment. The added weight or inclining method is as follows:

)tan/)(/( m ww zyWwz (21)

where w is the added or shifted weight, yw is the lateral location, zw the vertical location be-

low the pivot point on the frame, and the heel angle. The instrument for measuring the heel angle should be located on the model where the instrument replaces a mass of equal weight. A fixture may be added that is in-cluded in the added weight so that weight may be moved to multiple locations. Then the (yw/tan ) term can be computed as the slope from yw versus tan by regression analysis as a better estimate.

From (21), the sensitivity coefficients are

as follows:

)tan/)(/1(/ m1 ww zyWwzc

)tan/)(/(/ 2mm2 ww zyWwWzc

)tan/(/ m3 Wwyzc w

)sin/(/ 2m4 Wwyzc w

m5 // Wwzzc w

The uncertainty in the VCG is then

Uz (c1Uw)2 (c2UWm )2 (c3Uyw)2

(c4U )2 (c5Uzw)21

2 (22)

If the slope method for (yw/tan ) is applied,

the uncertainty in the slope must be added to (22). The sensitivity coefficient for the slope is from (21)

m6 /Wwc

For the uncertainty with the slope, (22) be-

comes

Uz (c1Uw)2 (c2UWm )2 (c3Uyw)2

(c4U )2 (c5Uzw)2 (2c6ub)21

2 (23)

344344

where ub is the standard uncertainty in the slope b from linear regression analysis.

The VCG of the hull is computed from the

hull-beam combination from the following:

hbbbhbhh /)( WzWzWz

The sensitivity coefficients from (23) are as

follows:

hbhbhh1 // WzWzc

hbhbhh2 // WWzzc

hbbh3 // WzWzc

hbbh4 // WWzzc 2

hbbbhbhhh5 /)(/ WzWzWWzc

The uncertainty of the VCG for the hull is

then

Uzh (c1UWbh )2 (c2Uzbh )2 (c3UWb )2 (c4Uzb)2 (c5UWh )21

2

For a model assembled from several parts,

the VCG is

n

1imm /WWzz ii

where the assembled model weight is given

by (17) and x is replaced with z in (18). The uncertainty in model weight is given in (19).

From (20) the uncertainty in the VCG is:

Uzm2 (WiUzi / Wm )2

i1

n

(ziUWi / Wm )2

i1

n

(ziWiUWm / Wm2)2

i1

n

3.4. Moment of Inertia The moment of inertia of a model is com-

puted from the oscillation of the model about the pivot point on the frame. The moment of inertia (MOI) in pitch is:

2)2/( TmgdI (24)

where d is the distance from the pivot point to the CG, and T is the period of oscillation. The period of oscillation is determined by at-taching a precision electronic inclinometer to the model and collecting a time series of its signal with a digital-data-collection system. The sensitivity coefficients from (24) are as follows:

2

1 )2/(/ TgdmIc 2

2 )2/(/ TmgdIc 2

3 )2/(/ TmdgIc

)2/()2/(/ 224 TmgdTIc

The uncertainty in the MOI is

UI (c1Um)2 (c2Ud )2

(c3Ug)2 (c4UT )212

The period in (24) is obtained by linear re-

gression analysis of the time series with a damped sine wave of the following form:

y aexp(bt)sin(2t / c d) e (25) where a, b, c, d, and e are constants that are computed from linear regression analysis. In this equation, c is the period, T. Regression analysis also provides the standard deviation. The combined uncertainty in the period is computed from:

345345

22)2( tcT UuU

where uc is the standard deviation or standard uncertainty of the period from linear regres-sion analysis and Ut is the uncertainty in the time traceable to an internationally recognized standards organization for the electronic de-vice.

In some cases, damping may be low, and

the damping term in (25) may be dropped. That is, a sine wave curve fit may work better. Then, (25) becomes

y asin(2t / c d) e

Calculations for the MOI indicate that a

time standard traceable to an internationally recognized standards organization is critical in the measurement of the oscillation period.

The hull MOI is separated from the beam

and hull MOI by the following:

bhbh III (26)

The uncertainty in the hull alone from (26)

is

2b

2hbh UUU

where Ihb is the MOI of the beam and hull as-sembly and Ib is the MOI of the beam only.

The MOI about the CG and its uncertainty

are given by

2mmmcg dmII

2242 )(4 dmmmmmIcgI UmdUdUU

The radius of gyration is defined as

mcg mIk / (27)

From (27), the sensitivity coefficients are

as follows:

c1 k / Icg 1

2 mmIcg

c2 k / mm 1

2Icg / m3

The uncertainty in the radius of gyration in

roll is 2

22

1 )()( mmcgIk UcUcU

The previous equations in this section for

the MOI of pitch are also applicable to roll where the subscript is replaced with .

3.4.1. Composite Pitch MOI For a model assembled from a number of

pieces, the MOI and its uncertainty for an assembled model are as follows in pitch:

n

iiicici ImzxI

1

22m ])[(

where xci, yci, and zci are the Cartesian coordi-nates relative to the model CG. The sensitiv-ity coefficients are as follows:

22m1 / ciciii zxmIc

icicii mxxIc 2/m2

icicii mzzIc 2/m3

1/m4 ii iIc

The composite uncertainty for the model in

pitch is then

346346

UIm2 (c1iUmi )

2 (c2iUxci )2

i1

n

(c3iUzci )2 UIi

2

For the MOI of components measured on

the same mass-properties instrument, the uncertainties may be considered correlated. In that case, the uncertainties may be summed for those parts.

3.4.2. Composite Roll MOI

For a model assembled from a number of pieces, the MOI and its uncertainty for an assembled model are as follows in roll where x is replaced by y in the pitch equation,

n

iiicici ImzyI

1

22m ])[( .

The sensitivity coefficients are as follows:

22

m1 / ciciii zymIc

icicii myyIc 2/m2

icicii mzzIc 2/m3

1/m4 ii IIc

The composite uncertainty for the model in

roll is then

UIm

2 (c1iUmi )2 (c2iUyci )

2i1

n

(c3iUzci )2 UIi

2

(28)

For a symmetric model where yci = 0, (28)

becomes

n

iiIzciicimiciI UUmzUzU

1

22222m ])2()[( .

3.5. Transverse Metacentric Height The transverse metacentric height is deter-

mined by performing an inclining experiment in water, floating the model in calm water and adding weight to the model in the transverse direction. The method and equations are sim-ilar to those for the VCG. The result for the metacentric height is as follows:

GMT w yw tan zw wWm .

The sensitivity coefficients are then

c1 GMT / w

Wm(yw / tan zw) / (wWm )2

c2 GMT / Wm

w(yw / tan zw) / (wWm )2

)](/[tan/ m3 WwwyGMc wT

)](/[sin/ m2

4 WwwyGMc wT

)/(/ m5 WwwzGMc wT

The uncertainty in GMT is

UGMT c1Uw 2 c2UWm 2

c3Uyw 2 c4U 2

c5Uzw 21

2 (29)

A better estimate of (yw/tan ) may be

computed by linear regression analysis of the slope of the yw versus tan curve at . (Linear curve fit requires model inclination that does not exceed 5 degrees.) In that case, the sensitivity coefficient is

)/( m6 Wwwc

The uncertainty in GMT is then from (29)

with the addition in the uncertainty of the slope is:

347347

UGMT c1Uw 2 c2UWm 2

c3Uyw 2 c4U 2

c5Uzw 2 2c6ub 2

12

where ub is the standard uncertainty of the slope b from linear regression analysis. 4. MODELING OF WAVE SPECTRA

4.1. Extreme-wave Modelling Related to

Stability Research When nonlinear or extreme wave model-

ling is considered with respect to ship- stability research, the following related ques-tions can be raised:

How often do extreme waves occur and

how relevant are they, What are their typical shapes and kine-

matics, How can we model extreme waves. These questions shall be treated in the fol-

lowing, looking both at state-of-the-art meth-ods and at recent research. This section is organized accordingly.

4.2. Probability of Occurrence and Rele-

vance of Extreme Waves

From the numerous data sets investigated during the Cooperative Research on Extreme Seas and their ImpacT Joint Industry Project (CresT JIP), on the effect of extreme- wave impacts on offshore structures, it was concluded that a second-order wave-crest-

distribution function is a good basis for the estimation of a design-wave crest, Buchner, et al. (2011). However, depending on parame-ters such as directional spreading, sea- state steepness and propagation distance, crests may exceed the second-order distribution in some severe seas by around 10%. On the other hand, the very highest crests may be limited by breaking and even fall below a second-order model. 4.3. Effect of Directional Spreading

For three different sea states at the same

peak period, the effect of spreading is illus-trated in Figure 4-1. Three spreading factors are shown, increasing from top to bottom. The three sea states were measured in the MARIN Offshore Basin during the CresT pro-ject. The waves were steep, with a nominal significant wave height of 12 m and a peak period of 12 seconds. The model scale was 50. The theory, Provosto & Forristall (2002), shows that the deviation from second-order theory is much less in short-crested waves. The measured crest-height distribution lies above both the Rayleigh distribution and the standard second- order distribution for the long-crested and the low-spreading case. It should be noted that the figures correspond to one phase seed per sea state. In on-going projects, corresponding investigations concern a large number of seeds.

348348

Figure 4-1 Wave-crest distribution depending

on spreading, from top to bottom: Long-crested, low-spreading (s=15) and strong-spreading

(s=4), measurements by MARIN for the CresT JIP.

4.4. Effect of Sea State Steepness

The effect of sea state steepness is illus-

trated in Figure 4-2 showing the measured crest distributions for 4000 hours of field data, increasing from top to bottom. The sea state steepness is defined on the basis of the mean spectral period, T1:

It can be seen that the wave crests become larger with increasing sea-state steepness,

starting from below the second-order theory and increasing up to a significant deviation beyond second order. For the largest crests, wave breaking as a counteracting effect limits a further increase and the wave-crest distribu-tions fall even below second order. This ef-fect of wave breaking as a limiting process is considered an important observation.

349349

Figure 4-2 Wave-crest distribution depending on sea-state steepness, increasing from top to

bottom: Analysis of 4000 hours of field measurement by Shell for the CresT JIP.

Figure 4-2 (Continued).

4.5. Effect of Distance (from a Wave

Maker)

In order to investigate the effect of wave evolution with distance on wave-crest distribu-tions, measurements at several locations along MARIN’s Offshore Basin length were carried out. Figure 4-3 shows the distribution of wave probes over the basin length.

Following the evolution of the wave with

increasing distance from the wave generator, it can be observed that breaking does not stop the possible further development of extreme crests. Figure 4-4 shows crest-height distri-butions for the same test, but at greater dis-tances from the wave generator. These meas-urements show that in long-crested waves, it may take a few wavelengths to modify the crest-height distribution. The observed growth may be due to third-order resonant interactions, or Benjamin-Feir instabilities, accompanied by a shift of spectral energy in the frequency band, and observed growth

350350

seems somewhat faster here than has been re-ported in some other studies—at scale 1:50, the MARIN Offshore Basin has a length of 5–10 wavelengths.

In summary, for the wave statistics, the

following can be concluded from the research undertaken in CresT:

1. Use the Forristall distribution for the

wave height. 2. Use second order distribution as basis

for the crest height.

Figure 4-3 Distribution of wave probes along

MARIN’s Offshore Basin.

Figure 4-4 Crest-height distribution observed for long-crested seas in the MARIN Offshore

Basin, 100 m from wave flap, approximately 2 wavelengths from the wave generator (649 m)

and approximately 5 wavelengths from the wave generator (1930 m), scale 1:50.

3. Correct for observed deviations from

second order. (This is the subject of ongoing research.)

Understanding the processes described previously and giving useful recommendations demands an effort in defining the correct wave spectrum, understanding wave amplification and breaking, and generating fully nonlinear crest statistics in a scheme useful for engineer-ing applications.

351351

4.6. Shape and Kinematics of Extreme

Waves

In order to answer this question, the fol-lowing aspects are considered:

How to model the most realistic wave directionality

Wave loading and response in short-crested waves

In case of short-crested waves the direc-

tional distribution of the wave energy has to be defined. The directional spectrum, S(ω,θ), is a combination of a frequency-dependent spec-trum, S(ω), and a frequency- and direction-dependent spreading function, D(ω, θ):

ω,θ ω . ω,θ / ω,θS S D G

2π

0

ω,θ ω,θ θG D d

The frequency dependent, S(ω), can be de-

scribed using a JONSWAP formulation, for example. For the spreading function, D(θ), a number of formulations that do not depend on ω are commonly used, amongst others:

An illustration of this type of spreading function is given in Figure 4-5.

Figure 4-5 Formulation of the spreading function θ / θD G with s = 7.

By using an s-parameter that is frequency-

dependent, each of these formulations can be used to describe a D(ω,θ) function. For exam-ple, in the Park, et al. (2001) spreading func-tion, the exponent in the cos2s formulation is frequency dependent:

5

max

2.5

max

ω ω, for 1

ω ωω

ω ω, for 1

ω ω

p p

p p

s

s

s

where ωp denotes the peak frequency of the S(ω) spectrum.

An example of a frequency-dependent

spreading is given in Figure 4-6.

4.7. Calibration of Directional Waves To improve the quality of waves in a model

basin, a calibration loop can be used. For a target wave spectrum the wave-maker- control

352352

software determines the theoretical flap motions, leading to a wave realization in the basin. Depending on the quality of the wave-maker theory used, the resulting wave in the basin can differ from the target spectrum. In a typical calibration loop the generated wave is measured and analyzed. The result-ing spectrum is compared against the target spectrum. Next, the target spectrum sent to the wave maker can be adjusted in an attempt to obtain a better-quality basin wave.

Figure 4-6 Frequency-dependent directional-

spreading function D (ωp = 0.80 [rad/s]).

For long-crested waves the calibration pro-cedure is well-established and included in common wave-generation software. For short-crested waves a similar approach was implemented and tested at MARIN: First the directional spectrum, S(ω,θ), was defined as a combination of a frequency-dependent spec-trum, S(ω), and a frequency- and direction-dependent spreading function D(ω,θ); in the correction procedure, S(ω,θ) and D(ω,θ) are treated separately. In a global overview the calibration worked as follows:

1. A wave was generated in the basin for the theoretical spectrum, St(ω), and the spreading function, Dt(ω,θ).

2. The results were measured and ana-lysed to determine the measured spec-

trum, Sm (ω), and the measured spread-ing function, Dm(ω,θ).

3. The corrections, CS(ω) and CD(ω,θ), were computed.

4. A new wave attempt based on CS(ω)St (ω) and CD(ω,θ)Dt (ω,θ) was gener-ated.

5. The calibration process was repeated from point 2 until satisfactory results were obtained.

To measure the waves, resistance-type, wave-elevation probes were used. The probe layout consisted of a number of small footprint arrays distributed over a larger area of the ba-sin. To determine the wave spectral density, a combination of two methods was used: Ex-tended Maximum Likelihood Method (EMLM), Waals, et al. (2002) and Maximum Entropy Method (MEM), Briggs (1982)] which were both implemented and tested for typical probe arrays. For frequencies above 2.5 rad/s (18 s prototype), a slope-based MEM method was used on each of the small foot-print arrays to obtain local information on the Dm(ω, θ). At lower frequencies, i.e., longer waves, the slope fell within the resolution/ measurement accuracy of the wave probes within a small footprint array. As an alter-native, a phase-difference-based EMLM met-hod was used, based on single-wave probes distributed over a larger area in the basin. Combining the two methods gave a reliable analysis for a wide range of frequencies. The correction factor, CD(ω, θ), was computed using: CD(ω,θ) = Dm(ω, θ)/Dt(ω, θ). The correction was only computed for the range of ω and θ with sufficient spectral energy.

4.8. Extreme Wave Modeling in Model

Basins

To model extreme waves accurately in both in test basins and in numerical simulations,

353353

different approaches are required which are addressed briefly in the following sections. Numerical wave tanks are addressed else-where.

Deterministic-wave generation means to

reproduce a predefined target wave train at a given position in a basin. For the generation of deterministic-wave sequences in a model basin, different types of wave makers are available. The wave generation process, as illustrated in Figure 4-7 (an example of a dou-ble-flap wave maker), can be divided into four steps:

1. Definition of the target wave train: the target position in time and space is selected—for example, the position where a ship encounters the wave train at a given time. At this location, the target wave train is designed—based on defined parameters or a wave rec-ord.

2. Upstream transformation: the target wave train is transformed upstream to the position of the wave maker, e.g., by means of a nonlinear wave propagation model.

3. Calculation of control signals: the corresponding control signals are calculated using adequate transfer functions of the wave generator.

4. Performing the model tests: the control signals are used to generate the speci-fied wave train, which is measured at selected positions in the tank.

Figure 4-7 Process of deterministic-wave generation: Calculation starts from the desired

target wave train, defined by particular parameters (1). Modelling wave propagation properly, the wave train at the position of the wave maker (2) as well as the corresponding

wave-maker-control signals (3) are calculated. The resulting wave train can be measured at the

target position (4) and compared to the given target wave (5).

4.9. Optimization of Wave Realisations

Furthermore, the target wave can be achieved by optimization applied both to a nu-merical and a physical wave tank. In the fig-ure below, (an example of the well-known “New-Year Wave” as an extreme directional wave), this optimization process is illustrated. The “New-Year Wave” was measured on 01/01/95 in the Norwegian sector of the North Sea (Draupner) by a down-looking radar, Ha-ver & Anderson (2000). It is a 20-min wave record, with TP = 10.8 s, HS = 11.92 m, HMAX = 25.6 m HMAX / HS = 2.15, Crest height 18.5 m, water depth = 70 m. The directional-wave generation based on optimization works as fol-lows:

Combining target wave train (time do-

main) and directional spectrum (fre-quency domain) to “fronts” as an unique parameter set of wave fre-quency, heading, amplitude and phase

354354

Transferring wave fronts upstream us-ing linear theory

Calculating the motion of the first-wave board, and then of neighbor-ing boards

Generating, measuring and analyzing waves

Optimizing wave-board motions, based on comparisons with the target wave

Figure 4-8 shows the result of the opti-

mized-basin realization of the short-crested New-Year Wave.

Figure 4-8 “New-Year Wave” modelled in the

basin by using an optimization method.

4.10. Focused Waves

Focused wave techniques (Clauss, 2008) can be used to deterministically generate ex-treme waves in model tests such as capsizing tests and is based on the in-phase superposi-tion of component waves at a target location (or at a focusing point), at a given target time. Another input can be the spectral shape of a single wave and/or the underlying sea state. Focused-wave techniques can be applied to the determination of response amplitude operators (RAOs) (linear focusing waves), the simula-tion of extreme events and embedding extreme waves in sea states. The advantages of these techniques are a short test duration, smooth transfer functions and extreme waves con-trollable in space and time. Figure 4-9 gives an example of a focused-wave generated in a model basin.

Figure 4-9 Focused waves generated for model

tests. 4.11. Numerical Methods

Modelling of extreme waves requires a

nonlinear wave-propagation model for both physical and numerical wave generation. Numerical wave tanks can be based on poten-tial (e.g., WAVETUB) or viscous-flow solvers,

355355

which can be coupled with motion- simulation tools. Also a coupling between potential and viscous flow solvers is a good approach to limit the calculation domain and save simula-tion time. Wave-structure interaction can be simulated in a wave field introduced via a pressure distribution (requires coupling with a wave model) or wave-velocity inlet (requires a wave-maker model in CFD).

Such advanced methods are required to

model extreme-wave properties such as: wave- propagation speed increasing with wave steep-ness; vertical asymmetry of wave crest and trough; mass transport; interaction between wave frequencies; and Benjamin-Feir instabili-ties (cf. Green & Naghdi, 1986, 1987; Dom-mermuth & Yue, 1987; Webster, 2009).

For use in simulation methods for stability

investigations, advanced CFD-based methods are still too “central processing unit (CPU)- intensive” for practical use. Higher order theories based on potential flow can be used for such purposes as described in the next sec-tions.

4.12. Pressure Modelling For a nonlinear environmental representa-

tion, the selection of the proper hydrodynamic pressure model is an important issue. The model must be able to represent the pressure in directional sea states, account for the increased steepness of wave crests, allow for the accu-rate representation of the wave kinematics in both the surf zone and the fluid domain, con-tain a statistical structure consistent with that observed in nature, and allow fast simulations of ambient sea-state pressures for use in the prediction of vessel and platform responses. Second-order theory meets most of these re-quirements.

For a single monochromatic wave, exact

second-order theory gives very close to zero total (hydrostatic plus hydrodynamic) pressure on the free surface up to wave steepness’s ap-proaching 1/7. However, in a steep sea state composed of many waves of different frequen-cies, the sum- and the difference-frequency exponential terms can contribute unrealisti-cally large terms that result in free-surface pressures that are far from zero. Therefore, we shall adopt an approximation that uses a two-term Taylor series expansion of the first-order term, which the literature suggests is one of the best compromises for dealing with these issues. This can be thought of as a two-term expansion of the exponential term that gets very large in the exact second-order solution. 4.12.1. Coordinate System

An earth-fixed Cartesian X,Y,Z coordi-nate system is used where Z points upward and the plane, Z 0 , lies in the mean free-surface level. The horizontal X and Y axes are such that the coordinate system is right-handed; otherwise, the orientations of the horizontal axes are arbitrary. 4.12.2. Determining Linear-Wave

Amplitudes Given a two-sided linear spectrum, S1( ) ,

such that the quadratic spectrum

S2 ( ) S1( )

2 d S1( )S1( )Z2 ,

is a good approximation to some desired two-

sided target spectrum, ST ( ), linear-wave

amplitudes are determined from the equation

356356

aj a( j ) 2 S1 j .

It is assumed that S1(0) 0 and a0 0 ,

but it is not assumed that is uniform. Therefore, the user may provide wave periods Tj 2 / j in decreasing order for j 1,

2,…,N. The frequencies, j , might also be

determined so that S( j )( j1 j1) is ap-

proximately constant. In either case, the increments are defined from the j as

follows:

j1 j1 2 if j 1,...,N 1

N N1 if j N

For all j between 1 and N , a j aj .

For positive j , phase angles, j , are

chosen so that they are random numbers uniformly distributed between 0 and 2 . For negative j , the phase angles satisfy the equation j j . The phase angle for

j 0 is irrelevant since the magnitude, a0 ,

vanishes and thus may be defined as 0. 4.12.3. Perturbation Series

The velocity potential, wave elevation, and pressure are written as a perturbation series

where I() , I

() , and pI() are O( ) . The

perturbation parameter, , is often taken to be the wave steepness or the wave amplitude. 4.12.4. Equations to Obtain First- and

Second-Order Pressures The perturbation series are substituted into

the Bernoulli equation to obtain the equation

The sum of all terms of order, , on the

left side of the equation must equal the sum of all terms of order, ,on the right side of the equation for all ℓ. Therefore, the following equations are obtained:

pI(0) gZ,

pI(1) I

1

t,

pI(2) I

2

t

1

2I

1 I1

.

4.12.5. Zeroth-Order Pressure

The zeroth-order velocity potential and the zeroth-order wave height vanish, but the

357357

zeroth-order pressure is nonzero and equals the linear hydrostatic pressure:

pI(0) gZ.

4.12.6. First-Order Pressure

There are N linear wave components with associated positive frequencies, j , and

positive wave numbers, kj j2 / g, for

j 1,2,..., N. The j-th wave component prop-agates in the direction that makes the angle, j , with respect to the positive X -axis where

j is measured counter-clockwise about the

Z -axis as seen from a point on the positive Z -axis. It has amplitude, aj , and a phase

angle, j . The phase angles are random num-

bers uniformly distributed between 0 and 2 . To compute the first-order pressure, additionalN wave components, j N,...,1, are defined with negative frequencies and wave numbers as:

j j

k j kj

kj j j g

a j aj

j j

j j

Where it is assumed that a0 0. Using

this notation, the linear pressure is a sum of components

pI

(1)(X,Y,Z,t) g ajekj Z

j1

N

cos j t kj Xcos j Y sin j j

4.12.7. Second-Order Pressure

To shorten the equation, it is helpful to define the quantity, kjℓ

, by the equation

kjℓ kj

2 kℓ2 2kjkℓ cos j ℓ

The second-order pressure, pI

(2) , can be determined from the equation

pI(2) (X,Y,Z,t)

2

aℓ2 ℓ

2e2kℓZ

ℓ1

N

2

(1 jℓ)ajaℓ j ℓ 1 cos j ℓ ℓ1

N

j1

N

j ℓ 2

ekjℓ Z

gkjℓ j ℓ 2

ekj kℓ Z

2

cos jℓ

2

(1 jℓ)ajaℓ j ℓ 1 cos j ℓ ℓ1

N

j1

N

j ℓ 2

ekjℓ Z

gkjℓ j ℓ 2

ekj kℓ Z

2

cos jℓ

where j is the Kronecker delta and jℓ is

defined as

jℓ

j ℓ t X kj cos j kℓ cosℓ Y kj sin j kℓ sinℓ j ℓ

The single sum and the first double sum

are the contributions due to sum frequencies. The second double sum is the contribution from difference frequencies.

358358

4.12.8. Pressure Above the Mean Free-

Surface Level One could evaluate the pressure above the

mean free-surface level just as given in the equations for Z 0. However, as is pointed out by Gudmestad (1993), this leads to unre-

alistic results as terms involving ekj Z become very large near the crests of waves. There-fore, the approach of Stansberg, et al. (2006) is used here. Of the various methods consid-ered by them, their second-order model has provided computed data closest to measured data. The pressure given by this second-order wave model for

0 Z I(1) (X,Y;t) I

(2) (X,Y;t) is

pI (X,Y,Z,t) pI

(1)(X,Y,0,t) pI(2)(X,Y,0,t )

ZZ

pI(1)(X,Y,Z,t)

Z0

4.12.9. Computational Methods

Unidirectional- and multidirectional-wave systems are treated separately since the computational methods for the two cases are significantly different. Unidirectional Waves

If waves travel in the direction that makes the angle , measured counter clockwise from the positive earth-fixed X -axis as viewed from above, then one can change to a primed coordinate system with coordinates X , Y , Z such that

X ' X cos YsinY ' Xsin YcosZ ' Z

The direction of wave propagation then

coincides with the positive X -axis.

First-Order Sums. The first-order

pressure is given by the equations

pI

1 X,Y,Z,t g ajekj Z

j1

N

cos j t kj X ' j

where it is assumed that a0 0 holds.

Second-Order Sums. The second-order correction to the pressure for unidirectional-wave systems is entirely due to difference frequencies:

pI

2 X,Y,Z,t pI2 X,Y,Z,t

4

ajaℓZjℓpe

i j ℓ t kj kℓ X ' j ℓ

ℓN

N

jN

N

where Zj( p) is defined by the equation

Zjℓp kj ,kℓ

0 if kjkℓ 0

j ℓ ekj kℓ Z

j ℓ max j , ℓ e

kj kℓ Zif kjkℓ 0

.

Multidirectional Waves

An efficient method for calculating first- and second-order pressures is not known for situations in which wave directions and wave frequencies are irregular. Therefore, it is assumed that wave amplitude is supplied on a topologically rectangular grid of points in the ( ,)-plane so that

apℓ a p,ℓ

359359

for p N ,..., N and ℓ 1,..., N where

a a( ,) is a real-valued function whose domain is a subset of( ,) : and 0 2 . The frequencies, p , satisfy

the equation, p p, and the discrete

amplitudes, apℓ, satisfy the equation,

a pℓ apℓ. For each p and there is a

phase angle, p . For positive p, the phase

angles are uniformly distributed random numbers between 0 and 2 radians. For negative p, the phase angles are chosen so that pℓ pℓ. It is assumed that the wave

numbers are equally spaced so that kp pk

for some k . The discrete wave numbers and angular frequencies are related by the equation kp p | p | /g.

The first-order pressure is given by the

equation

pI1 X,Y,Z,t g ' apℓe

i pℓei pt p0

N

ℓ1

N

ekpZ

cos kp X cosℓ Ysinℓ

' apℓei pℓei pt ekpZ

p0

N

sin kp X cosℓ Ysinℓ

The primed summation symbol indicates

that the first term in the summation should be halved. The inner sums can be evaluated with the aid of Clenshaw's algorithm (Goertzel, 1960; Luke, 1976; Newman, 1987; Press, et al., 1986) if the wave numbers are equally spaced.

The second-order correction to the pressure

isgiven by the equation

pI2 X,Y,Z,t

4apℓ

pN

N

m1

N

ℓ1

N

ei pℓei pt

e ikp X cosℓYsinℓ

aqmeiqmeiqtZpℓqmp

pN

N

e ikq X cosmYsinm .

The function Zpqm( p) Z( p) ( p,;q ,m)

is defined by the equation

Zpℓqmp

pq 1 cos ℓ m

p q 2

gkpℓqm p q 2 ekpℓqmZ

1

2e

kp kq Z

if pq 0

pq 1 cos ℓ m

p q 2

gkpℓqm p q 2 ekpℓqmZ

1

2e

kp kq Z

if pq 0

If p q and m , there is a

removable singularity. In this case, the

transfer function equals p2e2|kp|Z . The sum

over q can be obtained with the aid of Clenshaw's algorithm after which the sum over p can be obtained with the same algorithm.

If the second-order pressure due to the sum

frequencies is desired, then the definition of

Zpqm( p) should be modified so that it is 0 if

360360

pq 0. Similarly, if the second-order

pressure due to the difference frequencies is

desired, then the definition of Zpqm( p) should

be modified so that it is 0 if pq 0.

4.13. Linear Spectrum from a Nonlinear

Spectrum

In extreme nonlinear seas, one cannot di-rectly use the measured spectra, ST ( ) , from these seas in an analysis, or to derive a sea- keeping prediction, but rather one must derive the underlying linear spectrum to describe the waves that should be simulated. This is be-cause nonlinear interactions between the linear waves will provide second-order, nonlinear contributions through the physics capturing wave-wave interactions.

At extreme wave heights theoretical spec-

tra such as the Joint North Sea Wave Observation Project (JONSWAP) spectrum have nonlinear tails that are unrealizable in an experimental facility due to the breaking of high frequency waves. The underlying realizable spectrum may be derived as the corresponding linear spectrum by the tech- niques to be described.

The derivation of the linear spectrum

underlying the nonlinear spectrum requires the solution of an integral equation describing the measured spectrum by either direct or indirect methods. This section will introduce two possible methods of solving this problem, with the assumption that the process involves only first- and second-order processes, a reasonable assumption in most circumstances.

4.13.1. Determining a Linear Spectrum Only the case of unidirectional waves is

considered here since an integral equation sim-ilar to the one that exists for unidirectional waves is not known for the case of multidirec-tional waves. A two-sided target spectrum, ST ( ), is assumed to have been provided by the user. A two-sided linear spectrum S1( ),is sought which approximately satisfies the equation

ST S1

2 d

S1 S1 Z2 , (30)

for real where

2 2

2 2

2 if > 0

2 if < 0

σ ω / g ωσZ σ ,ω

σ ω / g ωσ

(31)

The details of the derivation are presented

in Sclavounos (1992). The spectral density, S1( ) , is that of the linear model and is defined as follows:

21

1

8 j ja S ω ω

Therefore, the statistical inference of a

second-order model reduces to the determination of the wave amplitudes, aj , so

that the second-order spectral density best matches the measured spectrum, ST ( ) . The linear spectral density, S1( ) , may be selected from any of the standard families with parameters such that the equality (30) is satisfied in a least squares sense.

361361

For example, the ITTC spectrum may be used for the representation3, S1( ) :

42 5 0 4401 1 3 1

1

0 110

4

2

. λ/

.S ω H T λ e ,

πωT

λπ

(32)

In (32) an accurate estimate of the modal

period, T1, may be available from full-scale measurements. Significant wave height on the other hand must be selected so that (30) is satisfied as accurately as possible, given ST ( ). The amplitudes of the regular wave components then follow from (31) and are used in equations for the representation of the linear- and second-order velocity potentials which then yield all desired quantities in the second-order wave-kinematics model.

An alternative numerical approach such as

the following might be considered. Using the definition of Z (31) and assuming that the spectra, S1( ) and ST ( ), are even func-tions of , the integral equation can be re-written as

1 1

2 21 1

2

TS ω S ω dσ S σ

S ω σ Z σ ,ω σ S ω σ Z σ ,ω σ

The integral equation has no solution if the

target spectrum has content of higher than the second order in the wave amplitude. This subsection describes how a least-squares approximation to the desired linear spectrum,

3 This representation can be obtained from equations on page 38 of Beck, et al. (1989) if three significant digits are retained.

S1( ) , may be obtained and thus avoids the issue of whether a solution exists or not.

The numerical scheme that follows requires that discrete frequencies be equally spaced. If this is not the case, then in the discretized integral equation will not be one of the discrete frequencies, j , and any

numerical scheme becomes complicated. The discrete frequencies in this subsection are therefore not necessarily those for which linear wave amplitudes, aj , are chosen in the next

subsection, and the N used in the description of the numerical scheme is not necessarily the number of positive wave frequencies used in the next subsection. It is assumed that j

are given by the equation

jω j ω

for j 0,1,2,... and some increment of frequency .

If S1,0 0 , the integral equation can be

discretized as

ST ,ℓ S1,ℓ

2 S1,n S1,ℓnZn,ℓn2 S1,ℓn Zℓn

2 n1

Where Z

pq Z( p ,q ). Here

S1,p S1(p ) and ST ,p ST (p ). The

series is truncated and the equations are written as

362362

12

1 1 11

2 21 1 1 1

1 1

2

0

, T , ,n , n n, nn

N N

,n ,n n, n ,n ,n n, nn n

f S S ω S S Z

S S Z S S Z

for 1,2,..., N . The frequency, ,

and the number, N are provided by the user. The objective is to minimize the sum

2 2

1

NΧ f

An initial guess, S1,ℓ(0) , for the discrete

linear spectrum is provided by the equation

S1,ℓ0 ST ,ℓ for ℓ 0,1,...,N

All iterates for the linear spectrum are

assumed to vanish at 0 rad/sec:

S1,0p 0 for p 0,1,...

It is now assumed that the p-th iterate,

say S1,m( p) , is known. For m 1,2,..., N ,

S1,m( p1) is chosen between (1 )S1,m

( p) and

(1 )S1,m( p) such that

1 1 1211 1 1 1 1 1 1

0

Np p p p p

, ,m ,m ,m ,Nf S ,...,S ,S ,S ,....,S

is approximately minimized. The number is somewhat arbitrary and can be provided by the user; it only serves to bound the interval in which a minimum of 2 is sought. Numerical tests for some spectra indicate that 0.1 is acceptable for those spectra. To

minimize 2 , we can check the sum at

several, say 10, evenly spaced points, S1,m( p1) ,

in the interval, [(1 )S1,m( p) ,(1 )S1,m

( p) ], and

make the change based on the 10 evaluations of 2. The number 10 is arbitrary and can be replaced by another value supplied by the user. Furthermore, the points do not have to be evenly spaced. The whole process is repeated for a specified number of iterations. The sum 2 can be monitored and the iterative process can be truncated when the fractional change in the sum is less than a user-specified tolerance or no longer decreases.

The desired values, S1,ℓ, for the discrete

linear spectrum are given by S1,m( p) where p

is the number of the most recent iterate. Interpolation is required if the spectral density function is desired at frequencies other than m m .

5. STATISTICAL UNCERTAINTIES

ASSOCIATED WITH (EXTREME)

SHIP- MOTIONS IN WAVES

5.1. Introduction

Measured data of physical phenomena can be classified as either deterministic or random. Often repeated measurements show variations due to the inability to control experimental conditions and/or due to the randomness of the physical phenomena considered. For exam-ple, the results from a standard resistance experiment are a deterministic quantity, which can be affected by small flow disturbances cre-ated by previous test runs.

363363

A seaway, the loads on a vessel and the responses of a ship are all random processes. The results of scale-model experiments and numerical simulations of ships in waves de-pend on the duration of the test runs or numerical simulations. This is a key factor in determining the number of test runs for scale- model experiments and numerical simulations. Furthermore, in analysing test or simulation data, it is important to assess the statistical reliability of motions and events.

Quantities such as incident waves and the

resulting first-order ship motions can be re-garded as “linear” signals for which straight-forward formulas are known that describe its probability-distribution function as a function of the standard deviation of the signal. The distribution of individual oscillations (“local” extremes) is known to satisfy distribution functions which depend on the bandwidth of the frequency spectrum of the signal. The “most probable” extreme value of a signal is then characterised by the number of oscilla-tions and the standard deviation.

In the case of “nonlinear” phenomena such

as wave-impact pressures, parametric- roll motions, water ingress on open-top container ships, and broaching, an estimate of the most probable extreme value cannot be solely based on the standard deviation and number of oscillations in the signal. In such cases it is customary to sort the peak values and to plot these as a function of the frequency of exceed-ance. Fitting a distribution function and extrapolating to the required number of events yields the most probable value. The reliabil-ity of such a procedure depends heavily on the number of samples, for instance the number of slams encountered during a certain time period. The intention of this chapter is to provide methods for determining the duration of scale-model experiments or numerical simula-tions such that linear motions can be obtained

with a given uncertainty margin. At the same time, methods are provided to predict the statistical uncertainty related to the occurrence of extreme motions.

5.2. Linear Signals

Incident waves and “linear” ship motions

satisfy a Gaussian (or Normal) distribution function (Ochi, 1973). This distribution function is characterised by the standard devia-tion of the signal, qs :

21

1 N

q ii

s q qN

Here N is the number of samples, qi is the

sample value and q is the mean value of the signal:

1

1 N

ii

q qN

The probability density function of a

Gaussian distribution is:

2

221

2q

q q

s

q

p q es

The probability that a value, q q , ex-

ceeds a certain value, qm, is obtained from the integral:

mq

P q q q p y dy

(33)

Based on (33), Table 5.1 shows exceedance

probabilities for several values of qm.

364364

Additionally, the stochastic variable, q, can

be described by the distribution of amplitudes (peak values) of q. When q has a Gaussian distribution, its amplitudes follow a Rayleigh or Rice distribution, depending on the bandwidth of the frequency spectrum (see Section 5.1). Amplitudes are often the most interesting quantities in ship-motion analysis.

qm mP q q mP q q

3 qq s 99.9 0.13

2 qq s 97.7 2.28

qq s 84.1 15.9

qq s 15.9 84.1

2 qq s 2.28 97.7

3 qq s 0.13 99.9

Table 5.1 Exceedance probabilities.

When qa = q q is the amplitude of a

Gaussian process then the mean of the highest one-third of the maximum to minimum values of qa is known as the significant double ampli-tude of q.

The most probable maximum value, 2qa,max,

of the variable, q, depends on the number of oscillations, n, as shown by Longuet-Higgins (1957):

a,max2 2s 2qq

With

1ln ln 1 (1 )

2n e

For large values of n it can be shown that

2qa,max 2sq 2ln n .

Figure 5-1 shows a schematic view of the

main quantities of interest.

5.3. Nonlinear Signals In case of nonlinear quantities like large-

amplitude roll motions or wave-impact loads, the estimate of the most probable extreme value cannot be based solely on the standard deviation and number of oscillations of the signal. In this case it is customary to sort the peak values and to plot these as a function of the frequency of exceedance, i.e., the fraction of the amplitudes exceeding a certain value. Fitting a distribution function and extrapolat-ing to the required number of events yields the most probable value. In this procedure the highest value with zero “frequency” is actually not accounted for.

Figure 5-1 Schematic view of a test signal; sigma represents the standard deviation.

The cumulative, 3-parameter, Weibull-

probability-density function is often used to fit the data. The governing parameters in this distribution function are the scale parameter, , shape parameter, and offset, :

365365

mq

mP q q e

If the fit yields a shape parameter of 2 ,

the results resemble a Rayleigh distribution. For processes which are governed by quadratic values of the underlying motions (like the relative velocity which governs a wave impact pressure), 1 , which corresponds to a negative exponential distribution.

The most probable maximum value is de-

fined by:

βa,max

12 θ α lnq

n

Ochi (1990) describes several other types

of distribution functions and how to derive ex-treme values with a certain adopted exceed-ance risk level. The frequency of exceedance is the number of exceedances of a certain amplitude divided by the total number of amplitudes. Figure 5-2 shows a typical fre-quency of exceedance plot.

Figure 5-2 Frequency of an exceedance plot.

Highly nonlinear and rare processes like

capsizing are difficult to fit by means of a distribution function, and prediction of the

capsize probability requires special techniques. Naess & Moan (2012) and Wang & Moan (2004) describe and compare methods for ex-treme-value estimation such as the Peak over Threshold (POT) method. This method is based on peak values that exceed a certain threshold level; sample values that are below the threshold are not considered. Using the POT method gives allows better modelling of the tail of the peak-value distribution.

The opposite approach is to use only the

less nonlinear part of the distribution function to make predictions of a threshold value. In the case of capsizing for instance, the thresh-old heel angle would be one where the righting moment arm (GZ) curve is at a maximum. A variation on this approach is by Belenky, et al., (2012a) which describes a split-time method with separate approaches for the linear and nonlinear parts.

5.4. Statistical Reliability

Seakeeping tests are generally designed to

obtain a fair estimate of the standard deviation of linear quantities. As a rule of thumb, the standard deviation of linear signals obtained from realisations with a typical duration of 180 wave encounters will show scatter (i.e., a standard deviation of the standard deviations of multiple runs of about 5%.

For the analysis below it is assumed that

the observed processes are stationary and er-godic. For stationary processes the true mean value of a quantity is time independent and the auto-covariance function is a function of time only. A stochastic process is said to be ergodic if its statistical properties (such as its mean and variance) can be deduced from a sin-gle, sufficiently long sample (realisation) of the process.

366366

For a given system under evaluation, the

question is what is a sufficiently long measure-ment duration? In general this depends on two properties of the spectral density function of a signal: the frequency where the spectrum has its peak value, ωP, and the bandwidth of the spectrum, b. This is illustrated in Figure 5.3.

Figure 5-3 Plot of spectral density vs

frequency.

Peak frequency is usually easily recognised (note that more than one peak may exist). The response bandwidth is small (narrow-banded) for lightly damped resonant- roll motions. A more broad-banded response is observed for heave and pitch motions in head seas. Low frequency responses due to wave- drift forces and course keeping enlarge the bandwidth of the frequency spectrum and have a profound influence on the statistical error as shown below.

An estimate of the statistical error in the mean value is given by Pierce (1992) as:

2 2

14

10lf lf

qP

A A

T

(34)

where T is the run duration in seconds and Alf is the low-frequency-area ratio in the spectrum. This ratio is defined as:

10 /

0

0

T

lf

S d

A

S d

(35)

For large durations (10) reduces to:

2

qPT

(36)

Multiplication of q with the standard

deviation of the sample yields the error in physical quantities.

Equation (12) can be used to determine the

required duration given a certain error:

2

P qT

(37)

For a Gaussian process, an estimate for the

statistical error of the standard deviation is given by Pierce (1992):

3

5 2qsbT

(38)

where the bandwidth, b, is defined at half the peak spectral density. Vice versa, the re-quired duration given a certain error follows from:

2

3

5 2qs

Tb

(39)

In summary, the variability of the standard

deviation decreases when the bandwidth of the response spectrum increases and reduces with one over the square root of the duration. The variability of the mean value depends on the low-frequency content and the peak frequency

367367

of the spectrum and is independent of the bandwidth of the signal. It reduces with one over the duration, provided the low-frequency content is very low.

It should be noted that for forward-speed

cases, peak frequency and bandwidth of the encounter spectrum should be used in (34)– (39).

5.5. Nonlinear Signals and Extreme Events

The number of extreme events is generally much smaller than the number of wave encounters. Due to this and the statistical scatter of nonlinear phenomena, the statistical reliability of this information may be quite limited. Extrapolation of the probability of exceedance of measured extreme values to larger extreme values further increases the scatter; a reliable assessment of extreme and/or rare values requires a long test or a numerical simulation procedure.

To illustrate the above problem, Figure 5.4

shows the results of a numerical experiment in which a large number of time series (batches) were generated. Each time series contained N peaks. The function values followed a prescribed Weibull distribution.

For 180 events (wave encounters or oscil-

lations) and 2 , a Rayleigh Distribution, the standard deviation of the batch mean amplitude is around 4%. For N=180 the standard deviation of the most probable ex-treme is about 8%. When extrapolating smaller batches with N=20 to the 1/180 probability level, the uncertainty in the most probable extreme increases to between 10 and 20%.

Considering the results for a nonlinear process 1 and a batch size of 20, the

standard deviation of the most probable ex-treme is about 30%; the standard deviation of the extrapolated most-probable extreme value with a 1% exceedance probability is some 40%.

Figure 5-4 Sample size and reliability.

5.6. Confidence Intervals for Mean and

Standard Deviation

When performing scale-model tests or numerical simulations, one may use the number of wave encounters to determine the required duration of a time series of linear motions and loads such that the results are accurate within an adopted confidence interval. For nonlinear motions and rare and/or extreme events, the number of encounters is usually unknown a priori, and statistical accuracy can only be determined after a certain test or numerical-simulation duration has been obtained. Statistical accuracy can be assessed when uncertainty estimators are derived from the time signals. The procedure below outlines derivation of such uncertainty estimators for single- and multiple-time records.

368368

5.7. Multiple Uncorrelated Time Records

Multiple uncorrelated time records are obtained when a number of test runs or simulations are obtained under identical conditions but with a different wave sequence for each run or simulation.

The mean value can be estimated using an

averaged quantity of a single realisation over a time interval. If the time series, ( )jq t , is the

jth realisation of a stationary random process with time average, jq , and N samples per

realisation,

1

1 N

j jii

q q tN

An ensemble average, nq , is an average

quantity of a set of n realisations:

1

1 n

n jj

q qn

The time-average, jq , and the ensemble

average, nq , are estimators of the true mean, q . Due to practical restrictions, the signal length, T, is often limited causing a difference between estimated averages and the true mean. When a finite set of n repeated time series,

( )jq t , is available, the variance, Vn = 2ns , and

the standard deviation, ns , of the mean values are defined as:

22

1

1

1

n

n j nj

s q qn

(40)

For uncorrelated sample mean values, the first-order estimate of the random uncertainty,

1,estu , follows from:

1,n

ests

un

The 95% confidence interval for the mean value is then obtained from:

95 1,1.96U n estq q u

The factor 1.96 stems from a normal

distribution for a 5% probability of exceedance.

The variance of a single time trace is given by:

22

1

1 N

j i ji

s q qN

The mean variance, 2s , is given by

2 2

1

1 n

jj

s sn

The variance of the variance for the

ensemble of time records is

22 2 2

1

1

1

n

v jj

s s sn

(41)

The random uncertainty, 1,estu , follows

from

u1,est

svn1

/22

and u1,est

svn1

1/22

where 2 is the asymmetrical Chi-squared distribution-function value. The 95% confidence interval for the standard deviation is then obtained from

95 1,U v ests s u

369369

In its report to the 26th ITTC, the

Specialist Committee on Stability provides an example of the application of the methodology for calculating variance of the mean and variance of the variance (ITTC, 2011a, Sect. 5).

If there is only a single time record

available, n = 1, this procedure does not work due to the factor, n–1, in the denominator of (40) and (41). For single-time records a different procedure can be adopted as outlined in the next section.

5.8. Single-time Records

As explained by Bendat & Piersol (2010), the auto-covariance function of a signal enables the computation of the expected variance of the mean and variance of the variance. The auto-covariance function shows the dependence between current (at time, t) and previous (time shift, τ) values of quantities in a stochastic process. For sta-tionary processes the true mean value of a quantity, μx, is time independent and the auto-covariance function is a function of time only. The mean value and the auto- covariance function can be calculated using temporal averages for an ergodic, stationary random process:

xxC lim

T

1

T0

T

q t q q t q dt

(42)

where is the time shift and T is the duration of the time series. The mean value is defined as:

0

1lim

T

Tq q t dt

T

When = 0, the auto-covariance value,

xxC(0), is equal to the variance of the signal. It is noted that the auto-covariance function, xxC (), is related to the auto-correlation function, xxR(), by

2C Rxx xx q

with

0

1T

Rxx q t q t dtT

The variance of the mean value is given by

Bendat & Piersol (2010):

2

0

21

TCs xx d

T T

while the first-order estimate of the random uncertainty, 1,estu , now equals the standard

deviation, s s2 .

The variance of the variance is given by Bendat & Piersol (2010) as

sv2

2

T0

T

1T

xxC 2 d.

The first-order estimate of the random

uncertainty, 1,estu , is the standard deviation.

The uncertainty of the signal variance, 2

vs , is presented by the confidence interval

2 2

1, 1,, vv est ests u s u

370370

Here and are the confidence factors to be obtained from a Chi-square distribution, 2

, where ν equals the degrees of freedom. It is noted that this is an asymmetrical distribution with . For

more information see Bendat & Piersol (2010) or other statistical handbooks like Ochi (1973).

Several methods can be found in literature for the computation of the auto-covariance function (Brouwer, et al., 2013). The direct calculation according to (42) is a time-consuming process. A more efficient ap-proach is to use Fourier transforms. The Fourier transform of the auto-covariance func-

tion, xxC(), equals the spectral density func-

tion, q(t) : Sxx( f ). The inverse transform

yields

2( )C ifxxxx S f e df

where f denotes the frequency, 2f .

This computation is not without numerical problems; repetition and noise amplification can occur. Belenky, et al. (2007) proposed to smooth the spectrum to prevent numerical problems. Brouwer, et al. (2013) proposed to use a biased auto-covariance function to prevent such problems. Brouwer, et al. (2013) also proposed an alternative method to determine the uncertainty of the mean and variance by using the covariance of correlated segments. These segments are consecutive parts of a single, sufficiently long time record. Sufficiently long is defined here as

Ts n

fL and Ts

n

b

where sT is the length of a segment, Lf is the lowest frequency component present in the signal,b is the bandwidth of the spectrum and

n is the number of segments. They show that the estimator for random uncertainty for the mean value in the segment method is

1,n

ests

un

Apparently, splitting a single measurement

into several segments shows a much faster decrease of uncertainty than taking several uncorrelated measurements with the same total length. A similar estimator for the variance is under development.

6. REVIEW OF VULNERABILITY

CRITERIA

The review of vulnerability criteria, in-

cluding long-term probability of loss of a ship, is carried out both for intact and damaged ships. Further development of vulnerability criteria that are required is outlined in Section 6.4. 6.1. What is a Vulnerability Criterion?

The concept of a vulnerability criterion has

a very clear definition when dealing with an intact ship. In IMO documents, vulnerability criteria are intended as tools to assess whether a ship is susceptible to different modes of stability failures. If a ship is susceptible to a stability failure that is neither explicitly or properly covered by the existing in-tact-stability regulations, the ship is regarded as an “unconventional ship” for that particular stability-failure mode.

An intact-stability failure occurs when a

ship cannot remain within the design limits of the roll (heel, list) angle and a combination of rigid-body accelerations (IMO, 2008).

371371

The phenomena in waves which may cause

large roll angles and/or accelerations have been identified in the 2008 Intact Stability (IS) Code, Section 1.2, Part A as follows:

1. Restoring-arm-variation events such as

parametric excitation and pure loss of stability;

2. Critical behaviour under dead-ship conditions (i.e., loss of steering ability or propulsion, and possible endanger-ment by resonant roll while drifting freely.

3. Manoeuvring-related problems in waves (e.g., broaching-to in following and quartering seas when a ship may not be able to maintain a constant course, which in turn may lead to ex-treme angles of heel).

Therefore, under the specific agenda item devoted to “second-generation intact-stability criteria,” the activity at the IMO is focussed on the development of specific vulnerability crite-ria for parametric roll, pure loss of stability, dead-ship conditions, and broaching. Re-cently, attention has also been given to the is-sue of excessive accelerations.

“Second-generation intact-stability criteria”

are based on a multi-tiered assessment ap-proach: for a given ship design, each stabil-ity-failure mode is evaluated using two levels of vulnerability assessment. The two levels of vulnerability assessment criteria at the dif-ferent tiers are characterized by different levels of accuracy and computational effort.

A ship which fails to comply with the first

level is assessed by the second-level criteria. In a case of unacceptable results, the vessel must then be examined by means of a direct assessment procedure based on tools and methodologies corresponding to the best

state-of-the-art prediction methods in the field of ship-capsizing prediction. This third-level criteria should be as close to the physics of capsizing as practically possible.

Direct assessment procedures for stability

failure are intended to employ the most ad-vanced technology available, yet be suffi-ciently practical so as to be uniformly applied, verified, validated, and approved using cur-rently available infrastructure. Ship motions in waves, used for assessment on stability per-formance, can be reproduced by means of nu-merical simulations or model tests (IMO, 2013a).

At present, a great deal of attention is paid

to specifying the characteristics of numerical simulations that adequately replicate ship mo-tions. This field will attract the interest and efforts of researchers and the rule-making community for the next several years (IMO, 2014).

In recent years the activity at the IMO has

focussed on the development of first- and second-level criteria. The first level of crite-ria is designed to be a simple procedure based on geometry/hydrostatics, load conditions, and basic operational parameters, thus having low complexity but a higher safety margin. The second level of criteria relies on simplified physics-based calculations with reasonable computational efforts and straight-forward applications following suitable guidelines. This second level is characterised by a moder-ate level of complexity coupled with the appropriate safety margin. It is important to point out that this second level vulnerability criterion should be able to eliminate any suspicion of vulnerability and if this is not the case it should confirm vulnerability and justify the application of direct stability assessment for this mode (IMO, 2010).

372372

For a damaged ship, it is uncommon to find explicit reference to the term “vulnerability criteria” in the literature. There currently appears to be no structured reference frame-work for damaged ships as there is for intact ships with functions and purposes.

In the warship context, the word vulner-

ability may be defined as an antonym of the term survivability (see Paragraph 2.3) since vulnerability is the conditional probability of being ‘lost’ given a certain scenario. In a situation where susceptibility (probability of being damaged) is equal to 1, survivability and vulnerability can be considered mathematical “opposites” for the purpose of this review.

6.2. Second-generation Intact-stability

Vulnerability Criteria

There is a need to properly balance the as-sessment of the probability of capsizing be-tween a specific sea state and an average of sea conditions. This need is well illustrated by Reed (2009) where the criticalities due to predictions based on linear superposition of a phenomenon claimed universally as nonlinear are discussed. 6.2.1. Review of Vulnerability Criteria

As already discussed, vulnerability criteria for specific stability failures are under devel-opment at the IMO. In the last ten years re-search communities have been very active in this subject area. At STAB conferences and ISSW workshops there have been dedicated sessions on related in the area of vulnerability criteria.

Two levels of vulnerability criteria and

standards for parametric-roll resonance, for pure loss of stability, and for broaching are

going to be finalised very soon. Further de-velopment is needed in relation to dead-ship conditions and excessive accelerations (IMO, 2014).

For the pure loss-of-stability failure mode,

the vulnerability criteria are expressed in IMO (2013a). The first vulnerability level is fo-cussed on the transverse metacentric height, GM, which is calculated when a longitudinal wave passes a ship. In this calculation the moment of inertia of the water plane is consid-ered at a draft corresponding to the level of the wave trough. The wave height that is used in this calculation is described in this method. The criterion is very simple and straightfor-ward and is based on the traditional hydrostat-ics of a vessel.

As an alternative at the first level, the met-

acentric height (GM) can be determined as the minimum value calculated for a ship balanced on a wave crest. The wavelength is selected equal to the ship length and with a specific wave height. The wave crest is then centred at different longitudinal positions along the vessel and the hydrostatics are calculated.

The second level of vulnerability criteria

takes into account characteristics of the right-ing arm, GZ, in longitudinal waves and then weighted averages of these stability parame-ters are calculated. As in the level one method, the calculations are conducted with the vessel balanced on a wave with the wave crest at different longitudinal positions along the vessel.

The selection of wave heights and

wavelengths used in the calculations are still under discussion, with two main options. The first option is based on sixteen representative wave cases. The second option is based on Grim’s (1961) effective wave height calculated for all possible

373373

significant wave heights and zero-crossing wave periods in the wave scattering diagram of the North Atlantic, but with the wavelength equal to the ship length (Umeda, 2013).

In the case of parametric-roll-stability fail-

ure mode, reference is made to IMO (2013a, 2013b). The first level vulnerability criteria is based on the ratio between variations of am-plitude of the GM when a longitudinal wave passes a ship, and, the GM of loading condi-tions in calm water. Variations of the GM amplitude are evaluated by considering half the difference between the moment of inertia of the water plane calculated at the draughts corresponding to the height of the wave crest and the wave trough. Wave height is again described in the methodology.

Another alternative in determining the

variation of GM may be calculated as half the difference between the maximum and mini-mum values of the GM calculated, assuming a ship to be balanced on a series of waves with the wavelength equal to ship length and pre-scribed wave height, with the wave crest cen-tred at the longitudinal centre of gravity and at each 0.1L forward and aft from the longitudi-nal centre of gravity.

Second-level vulnerability criteria consists

of two stages. Evaluation of the first stage employs the calculation of the ratio of GMs from the first level of vulnerability, but uses a statistical average of the results from multiple wavelengths and wave heights in the computa-tions instead of using a single wavelength and single wave height. The ratio in the first stage of second-level vulnerability also as-sumes the ship to be balanced on a set of waves defined in terms of prescribed wave-lengths and wave heights.

In the second stage of the second-level of

vulnerability criteria, a weighted average-roll

amplitude in head and following seas is also evaluated. Roll response is calculated using the equation for uncoupled roll motion while accounting for the influence of pitch and heave quasi-statically. A range of speeds is consid-ered and the environment is described by a specified set of waves. Grim’s effective wave height is calculated for all possible significant wave heights, and for zero-crossing wave periods appearing in the wave-scatter diagram of the North Atlantic, with wavelength equal to ship length. With this procedure the roll amplitude for all possible short-term sea states in the North Atlantic is obtained. The probability of encountering critical sea states where the roll amplitude is greater than the critical angle can be calculated and compared with the required standard (Umeda, 2013).

For broaching stability failures (IMO

2013a), the first vulnerability level is very sim-ple and only considers the Froude number and ship length.

For the second level of vulnerability, the

critical Froude number (i.e., corresponding to the susceptible threshold of surf-riding), is evaluated for a regular wave with a specific steepness and a specific ratio between the wave and ship length. The short-term prob-ability of surf riding can be calculated with Longuet-Higgins’s theoretical formula for the joint-probability-density function of local wave height and length. The long-term prob-ability of surf riding needs to be calculated with the wave scatter diagram of the North At-lantic and compared with an acceptable stand-ard (Umeda, 2013).

The issue of dead-ship conditions (IMO,

2013a) at the first vulnerability level is dealt with by the adoption of the IMO weather crite-ria, and amended in the specific area of wave steepness.

374374

For the second level of vulnerability crite-

ria, a weighted average representing a compre-hensive failure index, is evaluated considering different combinations of possible environ-mental conditions (IMO, 2012). The refer-ence exposure time is one hour. Calculation of a possible critical-roll angle is repeated for several sea states according to the relevant wave-scattering diagram. The necessary calculations can be made using one of two methodologies, both of which are based on the same underlying one degree of freedom (DoF) model, but are slightly different in their cal-culation details (IMO, 2013c). One method uses the linearization of the GZ curve in the vicinity of the equilibrium heel angle under the action of mean wind, and estimating the failure probability by means of the equivalent-area concept. The second method approximates the original GZ curve with piece-wise linear curves. More details about the two method-ologies are available in IMO (2009). Bassler, et al. (2009) provides a critique of the two ap-proaches from a theoretical point of view.

For the problem of excessive accelerations,

proposals for the first and second vulnerability levels are still under development at the IMO. The most recent version of these criteria is given in IMO (2012).

Based on the work described above, espe-

cially for the second-level vulnerability criteria, it is evident that the assessment of a ship is structured in terms of ship-environment inter-action. While formulating the criteria, ship characteristics are given as defined by a design team and fixed in terms of geometry and speed. The loading condition is defined as the “load-ing condition under investigation.”

Attention, therefore, is very much focussed

on the issue of including environmental conditions in the methodology of assessing

ship vulnerability. In general, this inclusion is made by means of a weighted average using a large number of wave cases. This approach seems to be sufficiently appropriate to measure, with a certain level of accuracy, the vulnerabil-ity of a ship. The adequacy of the assessment tools requires further examination when com-bined with standard values.

Notwithstanding the robust and efficient

theoretical and methodological approaches as the basis of the present vulnerability criteria, consistency with the use of other possible sources of wave statistics (on the discretion of various nations’ Administrations) needs to be taken into consideration.

As an extrapolation, the so-called “direct

assessment” can be considered a vulnerability criterion also. In this case the approach con-sists of two major parts: identification of a tool/methodology that adequately predicts ship motions in waves; and development of a pro-cedure that determines ship safety based on the likelihood or risk of stability failure. IMO (2013a) provides a description of the capabili-ties of a methodology which is used for direct assessment presented by different stabil-ity-failure modes. In the same document measurement of stability-failure likelihood is described as a probabilistic performance-based criteria.

Validated numerical tools are necessary,

but not sufficient by themselves to complete a direct stability assessment. There should be a prescribed procedure of applications of the tools, and following such a procedure, multiple applications should reach the same conclu-sions on a subject vessel. The procedure should also describe how to choose loading and environmental conditions. The measure of likelihood of stability failure is the main result of a direct stability assessment.

375375

When using validated numerical tools, the following issues must be addressed: time of exposure; the problem of rarity (see also IMO, 2011, Annex 1; and IMO, 2007b); statistical uncertainty; a set of loading conditions (rea-sonably selected from a vessel-stability book-let); and environmental conditions (in terms of the type of wave spectrum and its characteris-tics).

The proper selection of wave conditions is

a key issue (Belenky, et al., 2009a). In order to provide practical and consistent vulnerabil-ity criteria, stability failures must be evaluated for reasonable environmental and operational conditions. It is usually possible to find a combination of these conditions which results in a stability failure. While excluding unre-alistic operational conditions is relatively obvi-ous, determination of appropriate wave condi-tions is more difficult, due to their stochastic nature.

For the intact-ship condition the biggest

issues of vulnerability assessment, in addition to the environmental context, are the proper prediction of the physical behaviour of an in-tact ship in her interaction with a seaway, cou-pled with the statistics of ship conditions (dis-placement, center of gravity (CG), speed, etc.). All of these factors need to be taken into ac-count in an overall capsize-probability assess-ment (Ypma & Harmsen, 2012). In this per-spective an Insufficient Stability Event Index (ISEI) has been defined and applied to several full-scale capsizing events with appropriate numerical methods and procedures in order to establish appropriate threshold values (Krue-ger & Kluwe, 2010).

The attention to ship vulnerability is evi-

dent also in the field of naval ships (Beaupuy, et al., 2012; Gu, et al., 2012) and is expressed in terms of capsizing probability.

6.2.2. The Problem of Rarity For the treatment of the problem of rarity,

several techniques have been investigated: envelope peaks over threshold (EPOT) (Ypma & Harmsen, 2012); critical wave groups (Shi-gunov, et al., 2012); split time for dead-ship conditions and split time for surf riding (Belenky, et al., 2012b).

The split-time method is proposed as a

possible way to simplify the approach for pre-dicting the probability of ships capsizing in irregular waves, and separating the prediction process into a rare problem and a non-rare problem. The non-rare problem is treated through direct statistical processing of the time-domain motion data so the intermediate threshold is expected to be low enough that up-crossing statistics may be evaluated directly. The rare problem is solved by using the roll rate at the instant of up-crossing in order to find the value that leads to the specified stability failure (Belenky, et al., 2013; Belenky, et al., 2009b). A very interesting discussion of potential applications of POT and EPOT approaches is given in Belenky & Campbell (2012).

In Themelis & Spyrou (2006) an interest-

ing alternative use of a short-term or long-term prediction is postulated. Given a particular ship, the methodology can be deployed for short- or long-term assessments, depending on the intended period of exposure to the weather. In the current context, “short-term” is de-scribed as an assessment for a single trip, with a time window of a “few hours” forecast of weather parameters. Such an assessment could serve as a decision-making tool in an operational situation. Long-term assessments are performed for a variety of reasons on an annual basis or projected on a ship’s lifespan. The use of a long-term assessment is most common during the design phase of a ship.

376376

6.3. Damaged-ship Survivability Criteria There are several degrees of increased

complexity involved in damaged-ship dynamic-stability studies compared with intact-ship dynamic-stability studies (Peters & Wing, 2009).

Additional issues involved in developing

survivability criteria for a damaged ship versus an intact one include the damage scenario it-self, the flooding process, and the presence of water on-board after damage.

The damage itself introduces further sta-

tistical and probability issues into the problem. Flooding, especially in the progressive transi-ent phase is characterised as stochastic in na-ture, while water on-board enhances the nonlinear implications in the behaviour of a ship.

Because of the uncertainty and stochastic

nature of flooding, the identification and discussion of vulnerability criteria are further complicated with regard to the intact-ship problem. Therefore, dealing with a damaged ship will require a more comprehensive tool for the prediction of the physical behavior of the ship, inclusive of the damage scenario and flooding phenomenon (Ruponen, et al. 2012; Dankowski, 2012).

Harmsen (2006) presented a study about

the impact on stability of progressive flooding through small openings. However, additional studies are required on how to deal with these effects in static-stability calculations.

In damaged-ship scenarios, time is a cen-

tral issue in vulnerability investigations (Spanos & Papanikolaou, 2014), and often represents the most important factor in many situations: e.g., time-to-flood and time-to-

capsize (Spanos & Papanikolaou 2007; Spanos & Papanikolaou, 2014; Jalonen, et al., 2012); time-to-sink (Van’t Veer, et al., 2002; Ruponen, 2007); survival time (Jasionowski, et al., 2004; Pawłowski, 2008). In Ran, et al. (2012) the importance of the proper modelling of water ingress is pointed out because of its influence on time-to-capsize.

The strong influence of time on ship

survivability is emphasised, especially for Ro- Ro ships, in Spanos & Papanikolaou (2010) where time-dependant survivability is analysed. The time issue for passenger ships is also dis-cussed in Spanos & Papanikolaou (2012) where time-to-capsize for a given ship is as-sumed as a random variable depending on: random environmental conditions during a flooding casualty; the random shape and loca-tion of the hull breach; and the ship’s loading and local (e.g., arrangements and permeability) details of the flooded spaces. In the case of passenger ships, the statistical probability distribution (when capsizing is a possible event), can be approached with a basic Monte Carlo simulation. In this method time-to-capsize is sampled from a deterministic time- domain simulation for ship flooding and for a sufficiently large number of damage cases to meet statistical convergence of the results.

A comprehensive approach to possible

passenger ship loss must consider both the is-sue of time-to-sink together with an evacuation model (Skjong, et al., 2006; Spyrou & Roupas, 2006).

Determination of the time required to carry

out emergency procedures for a damaged ship is the result of a survivability assessment.(Spanos & Papanikolaou, 2012; and Jasionowski, et al., 2010). This problem of the time needed to carry out emergency procedures is implicit in the safe-return-to-port concept (IMO, 2007a). Within a certain

377377

damage threshold, it is assumed that the ship will survive indefinitely, whereas if the threshold is exceeded and abandonment becomes a possible event, then sufficient time is needed to carry out emergency procedures.

In investigations where “time” is a fixed

parameter (30 minutes, one hour, etc.), the out-come of the assessment process is given in terms of: capsize probability (Jasionowski, et al., 2004); probability of survival (Tagg & Tuczu, 2002; van’t Veer, et al., 2004; IMO, 2005); capsizing risk, capsizing index, and capsizing band (Papanikolaou, et al. 2010; Tsakalakis, et al. 2010). In these investiga-tions with “time” a fixed parameter, attention is paid to environmental conditions, in particu-lar to significant wave height. It is recog-nized that predicting ship survival is not a well-defined process, but there is a range of conditions within which the transition from “safe” to “unsafe” takes place. By conven-tion this range has been named the “capsize band.” This band begins at highest wave height where no capsizes are observed and ends at the lowest wave height where all predictions result in loss of ship. In order to better describe the capsize band, the term, rate-of-capsize, has been introduced.

For example, a capsize band is created by

reporting the rate-of-capsize as a function of different, significant wave heights, creating a sigmoid distribution. The rate-of-capsize is the probability of capsizing (PF) given a particular sea state.

Figure 6-1 Capsize-rate values for different Hs and different loading conditions

(Tsakalakis, et al. 2010).

Therefore PF will be 0 at the lower end of

the capsize band and 1 at the upper end. The point of the capsize band where PF = 0.5 is the critical wave height and it is this value that is used by convention when referring to ship survivability.

It is also important to analyse the influence

of a specific damage/flooding scenario as shown by Tellkamp & Cramer (2002). In Vassalos (2012) and in Vassalos & Ja-sionowski (2013), a definition of vulnerability is given as “the probability that a ship may capsize within a certain time when subjected to any feasible flooding case.”

The above definition is applied in Jalonen,

et al., (2012), where expressions like “vulner-ability to flooding” and “vulnerability to open watertight doors” are used in relation to a rapid capsize.

For the survivability of a ship in a damaged

condition, safety rules have recently shifted from a deterministic approach to a probabilistic one (IMO, 2005), where a comprehensive procedure is carried out in or-der to attain an A Index representative of the

378378

global performance of a ship in case of dam-age. The term 1-A is the probability of capsizing/sinking and is applied in risk evalua-tion procedures (Zaraphonitis, et al., 2013). A strong correlation between ship survivability and wave height is presented in Peters & Wing (2009) where a global, relative damage-loss index is formulated and applied. 6.4. Further Developments in a

Survivability Definition

In the past decade there has been a trend to create principles for a move from prescriptive-based to performance-based ap-proaches in the field of ship safety and in particular, in the fields of both intact-ship stability and damaged-ship stability (Peters, et al., 2013; Vassalos, et al., 2005; Kobyliński, 2007).

In developing a new approach in terms of

risk assessment, it is assumed that the safety rules will be restructured. In general risk assessment relies on a physics-based assessment of ship behaviour, given some physical and environmental conditions, and the proper treatment of the statistics involved in order to get to a strong probability of occurrence (or non occurrence) of an undesired event.

A large number of experimental studies

have been carried out in order to support the possible theoretical approaches and the studies have been shown to be of great importance, particularly in the field of damaged-ship stabil-ity.

The assessment of ship vulnerability in

terms of ship loss is the result of a comprehen-sive methodology where the following points are identified:

1. Loss mode 2. Loss threshold 3. Ship operational conditions 4. Environmental conditions 5. Time of exposure 6. Methodology for short-term prediction 7. Methodology for long-term prediction

(taking into account the problem of rar-ity for an intact ship).

The extension of the meaning of “loss” is

already considered (Peters & Harrison, 2006) when applied to naval ships. For a naval ship, the concept of mission continuity needs to be part of the meaning of “loss.” Instead of only the physical damage to a naval vessel being considered, mission continuity, which is con-cerned with the ability of a vessel and crew to both defend herself and perform its required mission, must also be considered. Mission interruption is one example of a mission-continuity loss and can be described as an “indirect loss,” contrasted with “direct loss” from damage to ship systems due to structural and flooding damage.

Validation of the individual steps of a

methodology and of an assessment framework as a whole is vital to build confidence in the final outcomes (Smith & Campbell, 2013; Montewka, et al., 2013). The importance of defining the relationship between capsize probability and general ship properties is dis-cussed in Ypma & Harmsen (2012).

In Bassler, et al. (2009) some fundamental

issues are raised in relation to the selection of realistic environmental conditions. This pa-per highlights the fact that an unrealistic envi-ronmental condition may lead to incorrect re-sults, even if the criteria are technically correct. In Bassler, et al. (2009) some possible options for using realistic environmental conditions are listed: e.g., an equivalent wave for life-

379379

time risk; a short-term sea state deemed “representative” of a specific ship-operational profile; and a long-term approach using a scat-ter diagram for a representative part of the World Ocean.

Consideration must be given to the

appropriate application of classical metho- dologies developed for the evaluation of ex-treme values of structural loads to stability prediction. A typical scheme for the cal- culation of extreme loads is based on long- term statistics, so a number of sea states needs to be considered. An operational profile is usually assumed based on existing experience. It includes the fraction of time that a ship is expected to spend in each sea state. Short- term probability of exceedance is calculated for each sea state; then the formula for total probability is used to determine the life-time probability of exceedance of the given level. This level is typically associated with signifi-cant wave height and a zero-crossing or mean period.

In calculating extreme loads, actual physi-

cal failure and the implied possible nonlineari-ties are not considered. The discussion in Bassler, et al. (2009) highlights the relevance of what is discussed above, specifically in the short-term phase of the evaluation. Consider-ing a regular wave as the equivalent of a spe-cific sea state is attractive because of its simplicity. However, the physics of some stability failures may be quite different in regular and irregular waves.

When vulnerability criteria are probabilis-

tic in nature, then the next important parameter to examine is the time scale, whether long- term or short-term. Short-term, as already mentioned earlier in this report, refers to a time interval where quasi-stationary statistics are assumed. A long-term scale covers a

larger time interval such as a season, a year, or the life-time of a vessel.

A short-term description of the environ-

ment can be characterised by one sea state or wave spectrum. However, if either of the above are chosen for use in a vulnerability criteria, justification will be required as to why a particular sea state or wave spectrum is used. Justification of the choice is important because sea states which are too severe may make the criteria too conservative and diminish its value. Special research is needed in order to choose a sea state “equivalent” or “representative” for a ship’s operational profile. This may result in a ship-specific sea-state to use for assessment.

An alternative to the selection of a limited

set of environmental conditions may be the use of long-term statistics considering all the combinations of weather parameters available from scatter diagrams or appropriate analytical parametric models.

In the traditional literature of naval

architecture and ship design, long-term predic-tion is usually performed with a statistical model composed of a short-term probability distribution of ship responses obtained with the linear superposition principle and a long- term occurrence-probability distribution of sea states provided in an ocean-wave statistics table.

A difficult issue for finding a shared vision

is the identification of a representative, if not realistic, environmental and operational context (Perrault, 2013). It has been proven that proper representation of the wave environ-ment is key to correctly evaluating dynamic- stability-related risks (Rosén, et al., 2013).

For an intact-ship assessment, the non-

ergodic nature of capsizing is incompatible with the linear hypothesis of the traditional

380380

statistical procedures used to assess the risk of capsizing for an intact ship. Further develop-ment of a proper theoretical approach and/or acceptable approximated methodology is needed.

It must be decided in the case of a damaged

ship whether the assessment should be posed in terms of probability of survival or in terms of survival time.

The introduction of the human-factor ele-ment is beyond the scope of this Committee but it is an important element in the process of assessing ship vulnerability. Evaluation of ship behaviour should remain in the design domain; however; when moving towards an operational context, the human-factor influ-ence cannot be disregarded (Kobylinski, 2012).

Once a satisfactory process is identified for

assessing ship vulnerability, additional effort will be required to evaluate the acceptable level (Sheinberg, et al., 2006).

7. DAMAGE-STABILITY-IN-WAVES

PROCEDURE

Procedure 7.5-02-07-04.2, “Model Tests on

Damage Stability in Waves,” provides a test procedure for carrying out model tests on a damaged ship in irregular waves to determine the probability of capsizing, or the significant wave height that will cause a model to capsize in a fixed time period. The Committee investigated the significance of scale effects in air pressure on flooding-model tests under atmospheric conditions, and also how to deal with inertia due to floodwater mass. Based on these investigations, the Committee up-dated two ITTC recommended procedures: Procedure 7.5-02-07-04.2, “Model Tests on Damage Stability in Waves,” and Procedure

7.5-02-07-04.4, “Simulation of Capsize Be-haviour of Damaged Ships in Irregular Beam Seas.” 7.1. Scale Effects in Air Pressure

There are some cases in which the flooding

of a ship is affected by the air pressure inside the vessel. The main contribution of air pressure takes place in the “trapped-air case” and in the “vented-air case with small vent area.” In a model test of a damaged ship, if the air pressure is maintained at atmospheric pres-sure, then scale effects in air pressure occur.

Let be the ratio of ship length to model

length. The model-scale pressure should be scaled by 1 in order to maintain dynamic similitude. That is, if the model is small, then the pressure of the air should be reduced proportionally. This is possible only in a depressurised tank facility. However, most model basins can only test under atmospheric air conditions, not under scaled-air pressure conditions. Figure 7.1 reveals, conceptually, the difference in pressure head between scaled air-pressure model test and an atmospheric pressure model test.

Figure 7-1 Concept of a scaled-model test.

For a trapped-air case, the pressure of the

model in atmospheric conditions is higher than in scaled pressure. Therefore, flooding to

381381

that compartment is restricted as shown in Fig-ure 7-2.

Figure 7-2 Flooding in a trapped-air case.

For a vented-air case, air is compressed

and the internal pressure increases. The pres-sure under atmospheric conditions is higher than under scaled-air pressure, so the flooding speed will be slower than under scaled-air pressure, and the following situation will occur, Figure 7-3.

Figure 7-3 Flooding in a vented-air case.

The above situation can be simulated by

using the state equation of air,

.PV const

where P is the absolute pressure of the air, V is the volume under consideration, and is the ratio of specific heats; in the case of air is 1.0 for an isothermal process and 1.4 for an adiabatic process. The flow through an opening can be estimated by the orifice equa-tion.

Figures 7-4, 7-5 and 7-6 show the water- height behaviour along with scaled time in the case of a trapped-air case for both small and opening large openings in a compartment bot-tom.

Figure 7-4 Schematic drawing for flooding in a non-vented air case.

Figure 7-5 Flooding in a non-vented air case

for a small opening. The above two figures are exactly the same

except for the time scale. This time scale difference comes from the opening-area ratio. As one over the scale ratio becomes small, the final water height is reduced also. In this case, the scale effect of air pressure is significant, regardless of the size of the open-ing.

10m

5m 5m

Scaled ModelScaled Air Pressure

Scaled ModelAtmospheric PressureFull Scale

0 10 20 30time(scaled)

0

0.1

0.2

0.3

0.4

h/D

Scale Factor

1/1

1/2

1/4

1/10

1/201/301/401/50

iso-thermal process Adamage/Abottom=0.001

382

Figure 7-6 Flooding in a non-vented air-case

for a large opening. For a vented case, Figures 7-7 to 7-10

show the density ratio of air and water height during the flooding process.

The ratio of vent area to damage area plays

an important role in the flooding process. When this ratio is large, i.e., for a large-vent area, the scale effect turns out to be small. For the small-vent area, the scale effect is large during the initial stage, and as time passes the scale effect becomes small.

In order to reflect the damaged-model-test

procedure in which a model is initially set in equilibrium condition, the effects of assuming the air-compression process to be isothermal or adiabatic can be simulated after setting the inner air pressure to be equal to the outside water pressure at the position of the damaged opening. For this purpose, the pressure of the compartment is initially set to the outside wa-ter pressure for the vented case. Figures 7.11 and 7.12 show the flooding process of the iso-thermal and adiabatic processes, respectively.

Figure 7-7 Schematic drawing for flooding

in a vented-air case.

Figure 7-8 Flooding in a vented-air case for a

large air-vent area.

0 1 2 3time(scaled)

0

0.1

0.2

0.3

0.4

h/D

Scale Factor

1/1

1/2

1/4

1/10

1/201/301/401/50

iso-thermal process Adamage/Abottom=0.01

0 2 4 6 8 10time(scaled)

0

0.4

0.8

1.2

1.6

h/D

, rho

/rho_

0

density

water height

Scales 1/11/21/41/101/201/301/401/50

1/1

1/21/4

1/10

iso-thermal processAairvent/Adamage=1.0Adamage/Abottom=0.01

1/1

1/50

383

Figure 7-9 Flooding in a vented-air case for a medium air-vent area.

Figure 7-10 Flooding in a vented-air case for

a small air-vent area.

If the flooding speed is slow, the air- compression process will be isothermal; if the flooding speed is fast, the air-compression pro-cess will be adiabatic. When a damaged ship with a large damage opening floats in waves, the flooding due to wave and ship motion is relatively fast, so an adiabatic process takes place in the air-compression process. Figures 7.11 and 7.12 show that the scale effect is not large.

Figure 7-11 Flooding for the isothermal process when air pressure was initially

balanced.

Figure 7-12 Flooding for the adiabatic process when air pressure was initially

balanced.

In line with the above discussion, it can be concluded that the scale effect is large for the trapped air/small-vent area case. For other cases, the scale effect is small and can there-fore be ignored in model tests of a damaged ship.

Under atmospheric conditions, it is possi-

ble to use alternative methods to reduce the scale effect of air pressure. For the case of a small-vent area, the vent opening can be en-larged to an appropriate size in order to reflect the inflow and outflow of a full-scale situation. For the case of trapped air, a simple solution would be to attach a balloon to the compart-

0 20 40 60 80 100time(scaled)

0

0.4

0.8

1.2

1.6

h/D

, rho

/rho0

density

water height Scales 1/11/21/41/101/201/301/401/50

1/1

1/21/4

1/10

isothermal process Adamage/Abottom=0.01Aairvent/Adamage=0.1

1/1

0 200 400 600 800 1000time(scaled)

0

0.4

0.8

1.2

1.6

h/D

, rho

/rho_

0

density

water height Scales 1/11/21/41/101/201/301/401/50

1/1

1/21/4

1/10

isothermal process Adamage/Abottom=0.01Aairvent/Adamage=0.01

1/1

0 20 40 60 80 100time(scaled)

0

0.4

0.8

1.2

1.6

2

h/D

, rho

/rho0

density

water height

Scales 1/11/21/41/101/201/301/401/50

1/1

1/2

1/41/10

isothermal process initial pressure equilibrium

1/1

Aairvent/Adamage=0.1Adamage/Abottom=0.01

0 20 40 60 80 100time(scaled)

0

0.4

0.8

1.2

1.6

2

h/D

, rho

/rho0

density

water height

Scales 1/11/21/41/101/201/301/401/50

1/1

1/2

1/41/10

adiabatic process initial pressure equilibrium

1/1

Aairvent/Adamage=0.1Adamage/Abottom=0.01

384

ment in order to lessen the scale effect of air pressure, and to obtain realistic flooding re-sults in a test condition.

In summary, if a damage opening is large and the compartment is well vented, scale ef-fects due to air pressure will be small, and model tests in atmospheric conditions are suitable. The scale effects will be large in a trapped-air or a small-vent area case. If pre-cise and accurate test results are required, the use of pressure regulation values on compart-ments to control the internal pressure may be a viable solution in the former situation, or in either case, model tests may be conducted in a depressurised model basin. At a minimum, when model tests are conducted under atmos-pheric conditions, modifications are recom-mended to reduce scale effects.

Procedure 7.5-02-07-04.2, “Model Tests on

Damage Stability in Waves,” was updated to reflect the above discussion. 7.2. Inertia Due to Floodwater Mass

Floodwater inertia has two main effects on

a ship’s behaviour; one is the inflow/outflow effect, and the other effect is the inertia of the flood water itself.

7.2.1. Floodwater Domain

There is a problem of which region of a ship should be treated as floodwater if the damage opening is large enough. First a more reasonable and clear definition of flood-water in the analysis of a damaged ship is needed. If the focus is on the inertial proper-ties of water, floodwater can be determined by looking at whether or not the water is moving together with the ship. If the focus is on the hydrodynamics of floodwater, this may be determined by investigating whether pressure of the floodwater is strongly related to the out-

side water level or not, and whether the hydrodynamic problem of the floodwater can be analysed separately or not. Provided that a boundary condition is given for the matching of inner- and outer-flow domains, the problem can be separated into one of flow in inner- (in-side the ship) and outer- (outside the ship) flow domains.

The following questions can be used as criteria to determine how the floodwater should be treated: What is the amount of water and is it or is

it not moving with the ship? What, if any, is the significant pressure

jump across the compartment boundary? Can the dynamics of the water be solved

separately or not? The above criteria also provide clues as to

what to consider as floodwater when examin-ing damaged ships. 7.2.2. Partially-flooded Compartments

The hydrodynamics of floodwater and its force on a compartment partially filled with floodwater can be calculated by theory or by a numerical scheme such as: resonant-mode analysis; potential-flow theory; computational fluid dynamics (CFD) with a free surface; etc. In these methods, the force generated by the floodwater is treated as an external force which affects the motion of a ship. An addi-tional problem to consider is whether the mass of the floodwater should be included in the ship’s mass or not. Since quasi-static analy-sis considers only the centre of gravity of the floodwater, the mass of floodwater should be included in the ship’s mass for this type of analysis. However, in a fully dynamic analy-sis, the pressure includes both static and dy-namic pressures. The force derived from integrating these pressures on the surface of a compartment includes all the effects of flood-water inertia and flow properties. The force

385

of the floodwater from a fully dynamic analy-sis assumes that the body force includes the actual acceleration of the floodwater, i.e., both gravitational acceleration and floodwater ac-celeration. In this case, the mass of the floodwater should not be included in a ship’s mass. 7.2.3. Fully-flooded Compartments

Floodwater in a fully filled compartment is often treated as a solid and thus is considered part of the ship. In rectilinear acceleration, floodwater acts like a solid. In rotational acceleration, the moment of inertia of floodwater in a compartment is smaller than that of a solid, because part of the water does not rotate with the ship. Lee (2014) showed the ratio of the moment of inertia of floodwa-ter and that of solids for various shapes of compartments.

SolidLiquidR IIC / where LiquidI and SolidI are the moment of inertias of floodwater when treated as a liquid and a solid, respectively.

Figure 7-13 shows the shapes of compart-

ments treated in his study.

Figure 7-13 Various shapes of tanks useful

for application (Lee, 2014).

The inertias of fluid in tanks of different

aspect ratios and shapes, Figure 7-14, become small as the aspect ratio goes to unity. The solid lines of Figure 7-14 are analytical or nu-merical results while the dashed lines are from an estimation formula that provides accurate results.

7.2.4. Inertia of Floodwater Entering a

Ship Newton’s Second Law of Motion states

that the force (moment) on a body is equal to its time rate-of-change of momentum (angular momentum). For a body of constant mass (moment of inertia) this translates to

F m

a

(

M I d dt ). However, for a body such as

a rocket which is burning fuel and ejecting gas, or a damaged ship in a seaway taking on and possibly discharging water, the

F m

a anal-

ogy is incorrect, because the time-rate-of- change of mass must be taken into account. Since the force of a body must remain independent of the coordinate system, a simple application of the rule for differentiation of the product of two functions does not apply. The contribution from the term for time-rate-of- change of mass belongs on the left-hand side of the equation with the force.

386

Figure 7-14 Moments-of–inertia of water in

fully filled tanks of various shapes, calculated and estimated from Lee (2014).

If we represent the momentum of the ves-

sel as and the angular momentum as , where

p mv and then, with m

the mass of the ship, v the velocity, I the moment of inertial tensor and

the angular

velocity, Newton’s Second Law of Motion can be written as:

F m

dv

dt,

M I

ddt

. (43)

When the mass, and hence the moment of

inertia are constant, these equations can be re-duced to the original

F m

a equation.

However, in the damaged condition, the ves-sel’s mass and moment of inertia vary with time and the equations of motion must be writ-ten as in (43). Rewriting (43) to account for the intake and/or discharge of floodwater as for a closed system yields:

(44)

where v ' and

' are the velocity and angu-

lar velocity of the flooding (discharging) water relative to the vessel, respectively4. All of the quantities dm dt , and dI dt in (44) can be determined from analysis of the flow at the damaged opening. However, if there is flow between flooded compartments, then the force due to the flow of floodwater between compartments must be accounted for in a similar manner. The evaluation of dI dt is also somewhat more complex as it involves the actual shape of the compartment.

The above material dealing with the

change of inertia due to floodwater was in-cluded Procedure 7.5-02-07-04.4.

8. IMO LIAISON

ITTC Specialist Committee on Stability in Waves (SiW) has reviewed draft reports of the Intercessional Correspondence Group (ISCG) as well as IMO documents including the SLF54, SLF55 and SDC1 sub-committee re-ports. The reports discuss methodologies for vulnerability criteria and direct stability as-sessment for the following stability failures:

Quasi-steady stability variation in waves in following/stern quartering seas;

Parametric resonance due to stability variation in waves;

Dead-ship conditions; 4 Note that these velocities are positive in the same direction as that of the ship, which is opposite the convention often used in rocket propulsion, where the positive velocity of the exhaust gases is opposite the positive velocity of the rocket.

0 0.2 0.4 0.6 0.8 1h/b or b/h

0

0.2

0.4

0.6

0.8

1

Rat

io o

f mom

ent o

f ine

rtia

C_R

CalculatedProposed Formula

Rectangle

HexagonOctagonEllipse

p

L

387

on the quality of direct calculation methodolo-gies. The ITTC SiW Committee reiterated the availability of technical specifications for numerical tools for direct assessment of vulnerability criteria that were contained in the Committee’s report to the 26th ITTC (ITTC, 2011a). 9. PREDICTING ROLL MOTION AND

DAMPING

Roll motion is one of the most critical re-sponses of a ship in waves, and the roll re-sponse of a ship is an important consideration in its design. Roll motion limits ship oper-ability, affects crew performance and ship habitability, and affects dynamic stability and ship capsize. The roll motion of a ship can be determined by analysing the various mo-ments acting on the ship: virtual and actual moments of inertia of mass; roll-damping mo-ment; restoring moment; wave excitation; and moments caused by other modes of ship mo-tion. Among them, the roll-damping moment has been considered to be the most important contributor that needs to be correctly pre-dicted. The roll damping moment of a ship needs to be taken into account at the initial stage of ship design to secure the safety of a ship, and also to obtain a better understanding of ship motions in waves.

In order to better understand the roll-

damping effects for roll motions in irregular seas, a state-of-the-art review was conducted. This review covered both the validation of nu-merical results of roll damping, and numerical modelling of hydrodynamics for time-domain computer codes for large-amplitude roll mo-tions.

9.1. Validation of Predictions for Roll

Damping

Validation is important for numerical cal-culations, and the selection of adequate valida-tion data is important for accurate stability estimations. The following methods are commonly used to obtain validation data of roll damping:

Free-decay test. (A) Forced-roll test with sinusoidal-roll excita-

tion. (B) Forced-roll test around a fixed axis. (C)

Roll-motion data is also used to validate

roll-motion simulations. In the following sections, some validation data for numerical results are introduced.

388388

Figure 9-1 Experiments to obtain validation data of roll damping.

9.1.1. Damping Coefficients from Forced-

roll Tests One of the purposes of using roll damping

derived from forced-roll experiments is to calculate the roll amplitude in regular waves. In this case, the frequency-domain roll- damping coefficients are used in equations of motion.

For this purpose, the coefficient of roll

damping denotes the equivalent linear damp-ing coefficient. Although the value of the coefficient depends nonlinearly on the roll amplitude and angular velocity for a certain frequency and forward speed, it is assumed that the coefficient is constant during a specific motion for a given roll amplitude.

The equivalent linear damping coefficient

is obtained from a forced-roll test. There are two ways to perform a forced-roll test. In one test the model is forced to roll but with

small amplitudes, and is constrained in all other degrees of freedom. For largeamplitude forced rolls the model must be allowed to heave and pitch. In both tests the forcing-roll moment and the roll motion are measured starting after four swings from rest in order to remove transient effects (Ikeda, et al., 1988 and Katayama, et al., 2011). The equivalent linear damping can be obtained by frequency analysis of the measured roll moment based on the measured roll at the fundamental frequency component in phase with the roll angular velocity.

In the case of statistical analysis of irregu-

lar roll motions, there is another approach to the linearization of the roll- damping expres-sion that can be used. In this linearization, the linear and quadratic damping coefficients from a roll decay or forced roll experiment are added with the quadratic term weighted by the standard deviation of the roll angular velocity in random seas (ITTC, 2011b, Sect. 3.2).

Jaouen, et al. (2011) verified and validated

MARIN’s Unsteady Reynolds Averaged Navier-Stokes (URANS) code ReFRESCO for roll damping of two-dimensional hull sections by comparing the damping coefficients meas-ured by Ikeda, et al. (1978), Figure 9-2. Ikeda, et al. (1978) showed the measured roll- damping coefficient of Series 60, SR98, SR158, SR108, and also showed the effects of forward speed on the damping coefficient. Ikeda, et al. (1978) provided other useful measured data. Ikeda, et al. (1976, 1977b, 1979) provided detailed validation data for measured flows around a bilge keel using forced-roll tests. Ikeda, et al. (1977b) also showed a number of types of vortices on hulls (Figure 9-3) as well as measured flows around the bilge of a naked hull (Ikeda, et al., 1977a). Figure 9-4 shows the pressure distribution on a two-dimensional model with bilge keels.

A

[s]

[deg]

[s]

[deg]

BExcitation moment generating device

[s]

[deg]

CTowing Carriage

[Nm]

Forced rolling device

389389

Figure 9-2 Coefficients of added mass and damping (Jaouen, et al., 2011).

Bonfiglio, et al. (2011), using FLUENT

and CFD-code base on the open source librar-ies of OpenFOAM, and Henning (2011), evaluated the hydrodynamic damping and added mass coefficients of two-dimensional ship-like hull sections in the case of forced oscillations. The results from Bonfiglio, et al. were compared with measured results carried out by the Delft Hydrodynamics Laboratory (Vugts, 1968, 1970). Vugts carried out a forced-roll test for two-dimensional ship sec-tions (Figure 9-5) and showed the measured

added mass, damping and coupling coeffi-cients among roll, heave and pitch. He also carried out a forced-roll test with forward speed for a three-dimensional segmented model and showed the sectional added mass, damping and coupling coefficients among roll, heave and yaw.

Figure 9-3 Eddies near a hull (Ikeda, et al., 1977b).

Figure 9-4 Measured pressure distribution on a hull with bilge keels under forced

rolling (Ikeda, et al., 1977a). Paap (2005) investigated verification of

CFD calculations with forced-roll test results for a circular cylinder with various types of bilge keels and a free surface. The measured data included not only coefficients but also velocity vectors obtained by a particle image velocimetry (PIV) technique (Figure 9-6), i.e.,

390390

time histories of bilge-keel force and heighs of the radiated wave.

Bangun, et al. (2010) calculated the

hydrodynamic damping and added mass coefficients of two-dimensional rectangular sections with bilge keels and compared the predictions with measured results by Yago, et al. (2008). Yago, et al. (2008) showed the measured added mass, the equivalent linear total roll damping, and the wave component of roll damping.

Figure 9-5 Cross sections of cylinders (Vugts, 1970).

9.1.2. Free-decay Test Data Roll damping results obtained from a roll-

decay test are not the same as the results ob-tained from a forced-roll test (Figure 9-7). The difference between the two sets of results occurs particularly during the first few oscilla-tions because roll-decay motion is a transient motion. ITTC (2011, pp 19–20,) shows how to obtain a roll-damping coefficient from a

free-decay test. Some notes on how to carry out a free-decay test are indicated in IMO (2006, pp. 11).

To estimate the onset of large-amplitude

roll motions at the roll natural frequency, Sadat-Hosseini, et al. (2010) use roll-damping coefficients obtained from a roll-decay test in the equations of motion for a time-domain simulation.

Figure 9-6 Visualized vortex and velocity vectors (Paap, 2005).

391391

Figure 9-7 Comparison of measured results by free-decay tests and forced-roll tests by

sinusoidal harmonic-roll excitation (Handschel, et al., 2012).

Yang, et al. (2012) calculated roll damping

of DTMB Model 5512 at different initial roll angles by using the roll-decay simulation in CFD (Figure 9-8). Roll-damping coefficient results were compared with measured results by Irvine (2004). Wilson, et al. (2006) used an unsteady Reynolds Averaged Navier-Stokes (RANS) method (CFDShip- Iowa) to compute the motions of DTMB Model 5512, and the resulting flow and wave fields around the models; the calculated results were compared with the measured results obtained by Irvine, et al. (2004). Sadat- Hosseini, et al. (2010) calculated roll motion with forward speed of the Office of Naval Research (ONR) Tumblehome hull form by CFDShip-Iowa and compared it with the measured roll motion of DTMB Model 5415 (Irvine, 2004). Gao & Vassalos (2011) applied a RANS-based CFD solver to study the roll decay of an intact DTMB Model 5415. The computed roll-decay history and velocity contours were compared with the measured results by Irvine, et al. (2004) Figure 9-9. Irvine also provided measured data (roll motion, velocity field, and wave pattern around the hull) for DTMB Model 5512 (http:// www.iihr.uiowa.edu/shiphydro/efd-data5512- roll-decay/).

Figure 9-8 Measured roll motion and curve of extinction (Yang, et al., (2012).

9.1.3. Roll damping in time-domain

simulations of large-amplitude

motions For time-domain simulations of irregular

motion, roll damping must include the effects of transient motion (Ikeda, et al., 1988; Katayama, et al., 2010, 2013). This means that the validation data must include time histories of the force of moment, the motions and flow around the hull measured under transient and irregular motion conditions. Moreover, if the roll amplitude is large, the validation data must include the effects of nonlinearity caused by the large amplitudes of motion (e.g., Tanaka, et al., 1981; Bassler, 2013). For free motions in extreme waves, the waves which impact the model and the resulting motions are required for validation of the simulation.

392392

Figure 9-9 Contours of velocity (Gao & Vassalos., 2011).

Therefore, the following experimental data

may be required for validation of simulations for large-amplitude, irregular-roll motions:

Irregular- and large-amplitude forced- roll test data

Irregular- and large-amplitude roll mo-tion data in extreme, irregular waves

However, no studies exist which provide

measured data for all the required conditions. In this section, some studies relating to large- amplitude rolls or irregular roll motions are introduced.

The importance of flow-memory effects on

roll damping was discussed by Ikeda, et al. (1988). Using the Morison Equation instan-

taneous velocity and acceleration are used, for which flow memory is not accounted The Logarithmic Decrement Method, where an equation without memory is fitted to the mo-tion decay data, has the same deficiency as the Morison Equation Method. Ikeda, et al. (1988) showed through experiments that the drag coefficient on plates increased in the first few oscillations when the plate is started rest (Figure 9-10). It takes 3 to 4 oscillations be-fore a steady-state flow field is established and the drag becomes constant. For all Keulegan-Carpenter (KC) numbers investigated (defined as KC = UT/2h, where U is the maximum flow velocity in the oscillation period, T, and h is span of the bilge keel), the drag coefficient in the first oscillation is about half the value for that in a steady condition. This effect is caused by the interaction between previous and present vortices. Only after a few oscil-lations does a steady, disturbed-flow field exist around the object. An additional valuable observation reported from the experiments by Ikeda, et al. (1988) is that the memory effects remain important in irregular motion. When an oscillation has a larger amplitude than the oscillation after it, then the drag coefficient is larger than at a steady oscillation amplitude. When an oscillation is smaller in amplitude than the oscillation after it, the drag coefficient is similar to the drag found in the first oscilla-tion starting from rest.

Katayama, et al. (2011) investigated the ef-

fects of transient motion on the drag force of a flat plate. In the region of KC < 250, the drag coefficient for acceleration in one direction is larger than the drag coefficient for acceleration in a uniform flow and smaller than that in a steady oscillatory flow (Figure 9-11). More-over, in a transient condition under forced oscillation, the drag coefficients from the first to the third oscillation are smaller than that in a steady oscillatory flow. These facts may indicate that the characteristics of transient and

393393

non-periodic rolling affect the drag coefficient. Katayama, et al. (2013) proposed an empirical formula for the bilge-keel component of roll damping based on the results of Katayama, et al. (2010) that indicate that transient effects on bilge keel drag can affect the onset and ampli-tude of parametric rolling in a time-domain simulation.

Figure 9-10 Drag and added-mass coefficients of a sinusoidally oscillating flat

plate normal to the motion from rest (Ikeda, et al. 1988).

Tanaka, et al. (1981) discussed the effects

of shallow draft on roll damping for hulls with bilge keels. Under certain conditions bilge keels add no increase to roll damping for some shallow-draft ships (Figure 9-12). From ex-perimental and theoretical studies, it was shown that the wave-damping component was reduced by the interaction between the waves made by both the hull and bilge keel; the eddy- damping component of bilge-keel damping is

also reduced by deformation of the water sur-face. These same effects may occur for large-amplitude roll motions for normal-draft ships with bilge keels.

Figure 9-11 Comparison of drag coefficients of a flat plate from a forced-sway test and a unidirectional accelerating test. Equation (11) is a curve fit to measured data from a

unidirectional accelerating test with various accelerations (Katayama, et al., 2011).

Bassler (2013) analysed the hydrodynam-

ics of large-amplitude ship-roll motions as components of added inertia and damping based on the results of forced-roll tests and CFD. It was shown that the effects of hull geometry, bilge-keel geometry, deck edge, and the free surface all affect the hydrodynamic components during large- amplitude roll mo-tions. Results from the experiments included measurements, observations, and identification of the discrete processes that result in several physical phenomena relevant to large-amplitude roll motions, including bilge-keel interaction with the free surface (emergence and re-entry), vortex shedding, and the effect of vortex shedding on the forces and moments of both hull and the bilge keel. Figure 9-12 shows measured bilge-keel force at various roll amplitudes.

0 10 20 300

10

20CD

Kc

result in equation (11) ( )

result in steady flow ( )

measured by forced sway test ( )CDperi

CDacc

CD0

394394

9.2. Modelling of large-amplitude roll

motions. Understanding roll motion and its associ-

ated damping is essential for the safety of a ship since roll motion, coupled with other mo-tions, may lead to capsizing. Apart from en-vironmental uncertainties, the damping coeffi-cients in equations of motion cannot be de-rived accurately by theoretical means alone, so experimental studies (e.g.,, experimental forced-roll and roll-decrement tests) or nu-merical studies are necessary. Once the de-caying curve or forced-moment curve is ob-tained either from simulations or from model tests, damping coefficients can be obtained by several appropriate techniques.

Figure 9-12 Effects of draft and roll amplitude on non-dimensional roll damping

of a two-dimensional model. (Tanaka, et al., 1981).

Figure 9-13 Filtered roll-motion measurements and bilge-keel-force

measurements for DTMB Model #5699, at various roll amplitudes, φ = 15° (purple), 25° (black), 30° (red), 35° (green) and 40°

(blue) deg, ω = 2.5 rad/s, with distinct physical phenomena identified at various stages in the roll cycle. (Bassler, 2013). Since the pioneering work of Froude, con-

siderable attention has been paid by various researchers to roll damping. Even now roll damping continues to be studied because fluid viscosity and vessel-forward speed create many difficulties in making predictions of ship-roll motions due to roll damping..

9.2.1. Current engineering prediction

methods. Current ship-motion-prediction methods

rely primarily on potential-flow-based hydro-dynamic methods such as:

Strip theory methods Frequency-domain free-surface Green-

function panel methods, Frequency-domain Rankine-panel

methods, Time-domain free-surface Green-

function panel methods, Time-domain Rankine-panel methods.

395395

Except for Strip Theory the panel methods above may include fully three-dimensional effects of flow and free-surface boundary forward-speed effects, which are taken into account in strip theory methods. Local pres-sures, especially for shorter waves, were much better predicted by panel methods than by strip theory methods.

Frequency-domain methods deal with

linear-ship-motion problems based on steady flow. However, these methods may not be applicable to large-amplitude motions for ships with strong flare such as container ships. Therefore, a time-domain-panel method in-cluding nonlinearity of hull and free-surface boundary conditions for ship-motion predic-tion was developed. Generally, time- domain code requires considerable computational time for obtaining a solution. Although hydrody-namic forces, ship motions, and wave pressure are much better predicted using frequency- domain panel methods than by strip theory methods, the calculated accuracy of hydro-dynamic forces on lateral motions using frequency-domain panel methods is not satis-factory. This is due to the fact that vis-cous-flow effects are not accounted for in potential-flow methods and in some situations are introduced into the calculations through empirical corrections.

In fact, predicting roll effects analytically

has always been problematic because of sig-nificant viscous effects (i.e., the nonlinear na-ture of roll motions and the strong dependence of roll damping on forward speed). An ideal fluid theory cannot resolve such roll effects. In fact it has a tendency to under predict re-storing moments due to the cancellation of the unsteady pressures over the sides and bottom of a ship. Such predictions become progres-sively worse for round-bottom hulls. Conse-quently, potential flow methods must be sup-plemented with empirical information. A

great deal of effort has been directed at devel-oping coefficient-based approaches for roll prediction. The most important contribution to developing such coefficient-based methods was developed by researchers such as Tanaka, Himeno, Ikeda, and Blok. According to them, viscous-damping coefficients can be di-vided into components related to four effects; friction; lift associated with forward speed; bilge-keel local effects; and vortex-shedding.

Even though numerous sources exist for systematic empirical data, problems remain with limitations in specific ranges of geometry and operating parameters. The standard empirical approach, for example, involves subdividing damping into bare hull, appendage components, etc. These approaches have been used successfully when applied to hull forms for which they were developed. How-ever, these methods require new data when applied to new hull forms. 9.2.2. Requirements for large-amplitude

Roll motion prediction Typical wave-induced ship-motion solution

techniques are based on the assumption of using small-amplitude motions and potential flow so that the general 6-DOF nonlinear equations of motions are reduced to two sepa-rate sets of linear equations (i.e., vertical plane motions and lateral plane motions) and are solved in the frequency domain. Using those assumptions, predictions show good agree-ment for vertical-plane motions. For lateral- plane motions, potential-flow-method-based codes simulate viscous effects by incorporat-ing empirically derived roll-damping data. Predictions with these methods are limited to the range of geometry, frequency, and operat-ing parameters for which empirical data are valid. However, these methods are also lim-ited by scale effects.

396396

Even modest damping can significantly affect roll motions. However, our ability to predict roll motions continues to lag behind that of predicting pitch and heave. Unlike other degrees of freedom (that are dictated by potential flow-induced forces), roll is domi-nated by the turbulent, vortex-driven flows near the bilge of a ship. The hydrodynamics of roll motion of a ship is largely influenced by viscous effects which include drag on the hull form as it rolls and on flow separation from the bilge and keel where subsequent vortex formations account for a large amount of roll damping. Bilge keels significantly increase the damping of roll motions, and at forward speeds, bilge keels generate a lift force which also contributes to damping. At high speeds the lift on a hull can be a significant contributor to roll damping.

As ship roll is typically a lightly damped

motion with large wave-driven excitations, significant accelerations can occur near reso-nance. This raises a practical concern for owners who wish to maximize a ship’s range of operability in often marginal conditions. The ability to accurately predict rolling near resonance is therefore a crucial topic, and dic-tates the need for a better understanding of the viscous and vortical flows that drive damping. The level of understanding required is further reinforced by the fact that viscous flow often exhibits nonlinear (e.g., amplitude-dependent) behavior, and may, therefore affect already extreme motions in a nonlinear manner.

For large-amplitude roll motion, the ge-

ometry of a wetted-ship surface may have ab-rupt geometry changes and bilge keels may become less effective due to emergence and interaction with the free surface. As existing coefficient-based damping models were devel-oped for small- to moderate-roll motions, the amount of energy dissipation for large- amplitude roll motion may be over estimated,

resulting in under-predicted roll motion, e.g., Belenky, et al. (2009b).

9.2.3. CFD-based Prediction of Roll

Damping Since roll damping is dominated by vorti-

city, truly robust modeling of the problem re-quires a technique capable of predicting the creation of vorticity in the boundary layer, the shedding of vorticity upon boundary-layer separation, and the effects of turbulence on pressure in the shed-vortex cores. Thus, there is a critical need for development of methods for predicting both viscous flows and large- amplitude motions for surface ships with appendages. The most common numeri-cal technique for predicting roll damping in-volves the embedded vortex approach. This approach usually uses a vortex distribution over the body, shed-point vortices in the flow, and a separation model for the flow near the bilge corners. Unfortunately, separation models require some prior knowledge of the boundary-layer separation point, and are there-fore difficult to apply for round bilge-hull forms without bilge keels. The techniques are also generally limited to two dimensions (Yeung, et al., 2013).

Steady Reynolds Averaged Navier-Stokes

(RANS) methods to calculate resistance and propulsion are the most advanced methods to use when predicting ship resistance. For ship resistance and powering, CFD has become in-creasingly important and is now an indispensa-ble part of the design process. In compari-son, application of unsteady RANS methods to ship motions in waves is less developed due to obstacles from unsteady flows [i.e., ship mo-tions, and complex environments (e.g., inci-dent waves, wave breaking, and bubble flow)]. These obstacles increase required computer resources.

397397

Unsteady RANS methods have the poten-tial to produce superior roll motion predictions compared to other methods since the effects due to viscosity, creation of vorticity in the boundary layer, vortex shedding, and turbu-lence are naturally included in the calculations. In an effort to develop a physics-based ap-proach to the prediction of ship motions, most studies have focused on two-dimensional os-cillating bodies. Yeung & Ananthakrishnan (1992) were perhaps the first to attempt to cap-ture the flow attributes through the application of RANS techniques, and their efforts have set the direction for further studies in this area. RANS-equation methods have been used to study the flow around two-dimensional os-cillating cylinders (Korpus & Falzarano, 1997); Yeung, et al., 1998; Sarkar & Vassalos, 2000).

Accurate predictions of forces and mo-

ments on a three-dimensional, submerged cylinder fitted with bilge keels and with a prescribed roll motion was demonstrated in Miller, et al. (2002). Wilson, et al. (2006) demonstrated three-dimensional RANS results for ship-hull forms undergoing roll, but were limited to small-roll amplitudes. Numerical uncertainties for RANS simulations are esti-mated by using verification and validation (V&V) procedures.

The damping behaviour of a ship model

depends on the local effects on the hull and appendages. It also depends on the vortex effects on the pressure distribution on a ship bottom. Wanderley, et al. (2007) were able to show the influence of vortex shedding on the roll-damping contribution acting on the bottom pressure of a ship hull. The influence of vortex shedding in roll dynamics depends on the amplitude of motion. For small angles, the vortex quickly vanishes from the hull. For larger angles, however, the vortex in-creases until linked to the hull bottom, modify-

ing the pressure distribution in that region. The physics of damping behaviour at large an-gles is still an open question. It is clear that vortex shedding is the dominant aspect of the dynamics of systems with large damping. Oliveira & Fernandes (2009) have proposed a new approach to fit a nonlinear model with a set of data obtained from several roll-decay tests.

Bangun, et al. (2010) simulated forced- roll

motion from small- to moderate- angular amplitudes for a barge with various bilge-keel orientations. The vorticity contour and roll hydrodynamic coefficients of a rolling barge are calculated from velocity and pressure fields, respectively. In contrast to an inviscid fluid where damping is found to be small at high-wave frequencies, numerical results ob-tained from a viscous solver show that damping is large even when the wave fre-quency is high (i.e., when the convective flux dominates the flow over the diffusive flux). It is shown that larger roll-amplitude excitation will cause the vortices generated to interact very near the free surface. It remains a chal-lenge to solve a pressure-correction equation under such a condition.

Yang, et al. (2012) used CFD to simulate

DTMB Model 5512 roll-damping motions at different initial roll angles and the results showed good agreement with tank test data. It showed that the roll-damping coefficient is unrelated to the initial roll angle and varies linearly rather than nonlinearly if the roll angle is less than 20 degrees.

Stern, et al. (2013) summarised the CFD

progress on ship hydrodynamics. They showed that CFD studies mainly focus on heave- and pitch-motion simulations by RANS compared with roll simulations. Validation for local flow has not been conducted yet due to the complexity in local flow measurements

398398

for free-running models. For more computer- intensive applications such as seakeeping and route modelling, an extremely long simulation time and a range of operating conditions need to be covered. For these applications, the speed of current CFD solutions is still the lim-iting factor. Thus, using a faster method such as a system-based method should be consid-ered. However, the mathematical models for these methods could be improved using high-fidelity CFD solutions along with system identification techniques. In addition, inno-vative numerical methods for easier and faster CFD solutions are required. Finally, taking advantage of faster computers such as the next generation massively parallel multicore ma-chines should be considered. 10. CONCLUSIONS AND RECOMMEN-

DATIONS

10.1. Technical conclusions

A comprehensive state-of-the-art review on predicting ship stability in waves has been undertaken and has been particularly concerned with the definition of loss and survival of a ship, and modelling the internal geometry of a damaged ship. This review concludes with a discussion of modelling of extreme-wave con-ditions. 1. This review of modelling of damaged ships has reinforced the importance that methodolo-gies used to model damage must reflect the mechanisms involved with the physics of dam-aged-ship motions leading to loss of a vessel (i.e., sensitivity to scaling in model tests, nonlinear effects of progressive flooding, and floodwater effects on damping of roll on other degrees of freedom). Leak and collapse pres-

sures of watertight doors and bulkheads is an-other key area that must be covered for damaged-ship modelling. This review also considers the importance of taking air pressure into account during damage experiments and simulations. 2. A state-of-the-art review has been carried out concerning the definition of loss and sur-vival of a ship. It has been concluded that the two terms, loss and survival, under specific conditions, express complementary concepts. It is possible to identify many analogies but also differences while investigating the con-cepts of loss and survival for an intact ship versus a damaged one. Nevertheless, the prevalent trend in defining loss and survival is to focus on the capsizing event, but due to in-herent practical difficulties in dealing with this phenomenon, attention is often shifted to focus on the definitive representative roll-angle value. However, the critical roll-angle value is a particular characteristic of a specific ship under investigation. From a performance- based assessment perspective, it is recom-mended that attention also be paid to the loss of functional capabilities, which in some cases the functional capabilities (e.g., ship power production and delivery) are beyond the specific focus of the ITTC. 3. An investigation into uncertainty analysis for use in intact- and damaged-model tests to complement current procedures has been re-viewed and an outline guide has been provided. This investigation has focused on the uncertainty involved in making measurements during experiments such as roll, pitch, water height, etc. This investigation concluded that the impact of errors occurring while setting up the model can have a significant impact on the experimental results. However, understand-ing the source of the errors allows the effects of the errors to be minimised.

399399

4. An investigation on wave modelling spectra in the determination of dynamic in-stability of intact vessels has looked at nonlinear wave kinematics, statistical distri-bution of crest and trough height and nonlinear wave propagation. Progress is being made regarding methodologies for stability assess-ment of both intact and damaged ships, A number of modelling methods are presented to achieve realistic environmental conditions. 5. In order to better understand the uncertain-ties associated with results from experiments and simulations of extreme motions of intact vessels in realistic irregular seaways, a number of quantitative techniques which reflect the nature and magnitude of the phenomena of extreme motions have been reviewed. These techniques address the statistical reliability of both “linear” and “nonlinear” signals and events. Furthermore, these techniques were reviewed to determine extreme values and confidence intervals for nonlinear signals. 6. A state-of-the-art review has been carried out concerning the definition of vulnerability criteria (including long-term probability of loss of a ship) for intact and damaged ships. An outline of current developments is pre-sented and includes a vision of an harmonized approach for intact and damaged ships, high-lighting the different priorities that can be identified in the two states. Common ap-proaches are recommended to identify and dis-cuss the relevance and treatment of the envi-ronmental context, ship loading conditions, and time of exposure. These considerations must also be coupled with the current develop-ments of simulation tools for the prediction of nonlinear dynamic ship behaviour. Looking specifically at the case of damaged ship, the stochastic nature of flooding, especially in the transient progressive process, should be ad-dressed in conjunction with the proper stochas-tic treatment of the entire damage scenario.

A significant difference between an intact- and a damaged-ship situation with respect to safety assessments, is the issue of time-to-loss. For an intact ship the time-to-loss interval is so long that the estimation of the rare-event oc-currence implies the need to further develop methodologies for statistical extrapolation. For a damaged ship, the critical point to deter-mine is if the time-to-loss is sufficient to per-form emergency procedures or to evacuate the ship. The outcome from these investigations is extremely diverse which suggests a review is required for the identification of an efficient, final assessment index. 7. An investigation of model tests on damage stability in waves has examined air compres-sibility, scale effects on air pressure, and cur-rent test procedures. This investigation cov-ered the scale effects on air pressure on flooding-model tests under atmospheric condi-tions and how to deal with the inertia due to floodwater mass. The investigation concluded that the scale effects on air pressure are not significant in most cases, except for the case of trapped air and for a large-damage opening with a small-vent area. In line with these investigations, Procedure 7.5-02-07-04.2 has been updated. The inertia due to floodwater mass was investigated with regard to computational modelling. This included the momentum change description of floodwater, potential criteria for determining the amount of floodwater and a review of research related to floodwater dynamic properties; this has re-sulted in a revision to Procedure 7.5-02-0704.4. 8. In order to better understand the roll- damping effects for large-amplitude roll mo-tions in irregular seas, a state-of-the-art review was conducted. This covered both validation data for numerical results of time-domain computer codes of roll damping and numerical modelling of hydrodynamics for time-domain computer codes of roll damping. The review

400400

of validation data focused not only on large- amplitude irregular motion but also on small- amplitude regular motion. Some existing and useful model-scale experimental data has been identified for validation. These data are pre-sented separately as a total hydrodynamic moment, and roll damping with its compo-nents. 9. The committee has:

a. Updated Procedure 7.5-02-07-04.2 for Model Tests on Damage Stability in Waves.

b. Updated Procedure 7.5-02-07-04.4 for Simulation of Capsize Behaviour of Damaged Ships in Irregular Beam Seas.

10.2. Recommendations to the Conference Adopt the revised Procedure 7.5-02-07-

04.2, Model Tests on Damage Stability in Waves.

Adopt the revised Procedure 7.5-02-07- 04.4, Numerical Simulation of Capsize Behaviour of Damaged Ships in Irregular Beam Seas.

11. REFERENCES AND NOMENCLA-

TURE

11.1. References

Bačkalov, I. (2012) “A probabilistic analysis of stability regulations for river-sea ships.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Ball, R. E. & C. N. Calvano (1994) “Estab-lishing the fundamentals of a surface ship Survivability Design Discipline.” Naval Engineers J., 106(1):71–74.

Bangun, E. P., C. M. Wang & T. Utsunomiya (2010) “Hydrodynamic forces on a rolling barge with bilge keels.” Applied Ocean Research, 32:219–232.

Bassler, C. C. (2013) “Analysis and modeling of hydrodynamic components for ship roll motion in heavy weather.” Ph.D. Thesis, Virginia Polytechnic Institute and State University, Blacksburg, VA, xxiv+253 p.

Bassler, C. C., V. Belenky, G. Bulian, A. Francescutto, K. Spyrou & N. Umeda, (2009) “A review of available methods for application to second level vulnerability criteria.” Proc. 10th Int’l. Conf. Stability of Ships and Ocean Vehicles, St. Peters-burg, Russia.

Beaupuy, B., N. Stachelhausen, J.-Y. Billard, E. Mogicato, P. Vonier & J.-F. Leguen (2012) “Operability of French naval ships over 50 years.” Proc. 13th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 575–582.

Beck, R. F., W. E. Cummins, J. F. Dalzell, P. Mandel & W. C. Webster (1989) “Mo-tions in waves.” Principles of Naval Architecture, Vol. III, ed. by E. V. Lewis, Jersey City, NJ, Society of Naval Arch-itects & Marine Engineers, pp. 1–190.

Belenky, V., C. C. Bassler, K. Spyrou (2009a) “Dynamic stability assessment in early-stage ship design.” Proc. 10th Int’l. Conf. Stability of Ships and Ocean Vehicles, St. Petersburg, Russia.

401401

Belenky, V. & B. Campbell (2012) “Statistical extrapolation for direct stability assess-ment.” Proc. 13th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 243–256.

Belenky, V., A. M. Reed & K. M. Weems (2009b) “Probability of capsizing in beam seas with piecewise linear stochastic GZ curve.” Proc. 10th Int’l. Conf. Stability of Ships and Ocean Vehicles, St. Petersburg, Russia, pp. 637–650.

Belenky, V., K. M. Weems, C. C. Bassler, M. J. Dipper, B. Campbell & K. J. Spyrou (2012a) “Approaches to rare events in stochastic dynamics of ships.” Proba-bilistic Engineering Mechanics, 28:30–38.

Belenky, V., K. J. Spyrou & K. M. Weems (2012b) “Evaluation of the probability of surf-riding in irregular waves with the time-split method.” Proc. 11th Int’l Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Belenky, V., K. M. Weems & W-M. Lin (2007) “A probabilistic procedure for evaluating the dynamic stability and cap-sizing of naval vessels, phase 1: technology demonstration.” SAIC Report ASTD 08-017, 211 p.

Belenky, V., K. M. Weems, & V. Pipiras (2013) “Split-time method for calculation of probability of capsizing due to pure loss of stability.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France, pp. 70– 78.

Bendat, J. S. & A. G. Piersol (2010) Random Data, Analysis and Measurement Proced-ures. 4th edition, John Wily & Sons. Inc., New Jersey, USA.

Bonfiglio, L., S. Brizzolara & C. Chryssos-tomidis (2011) “Added mass and damping of oscillating bodies: a fully viscous nu-merical approach.” Recent Advances in Fluid Mechanics, Heat & Mass Transfer and Biology, pp. 210–215

Briggs, M. J. (1982) “Multichannel maximum entropy method of spectral analysis ap-plied to offshore platforms.” Proc. 14th Offshore Tech. Conf. OTC, Houston, TX.

Brouwer, J., J. Tukker & M. van Rijsbergen (2013) “Uncertainty analysis of finite length measurement signals.” Proc. 3rd Int’l. Conf. Advanced Model Measurement Technology for the EU Maritime Industry, Gdansk, Poland.

Buchner B., G. Forristall, K. Ewans, M. Christou & J. Henning (2011) “New insights in extreme crest height distribu-tions (A Summary of the Crest JIP).” Proc. ASME 30th Int’l. Conf. Offshore Mech. & Arctic Engin., Rotterdam, the Netherlands.

Clauss, G. F. (2008) “The taming of the shrew: tailoring freak wave sequences for sea-keeping tests,” J. Ship Research, 52(3):194–226.

Dankowski, H. (2012) “An explicit progres-sive flooding simulation method.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

402402

Dommermuth, D. G. & D. K. P. Yue (1987) “A high-order spectral method for the study of nonlinear gravity waves.” J. Fluid Mechanics, 184:267–288. (1988) Corrigendum, J. Fluid Mech., 189:589.

Gao, Q. & D. Vassalos (2011) “Numerical study of the roll decay of intact and damaged ships.” Proc. 12th Int’l. Ship Stability Workshop, Washington, D.C. pp. 277–282.

Goertzel, G. (1960) “Fourier analysis.” Mathematical Methods for Digital Com-puters, Ed. by A. Ralston & H. S. Wilf, John Wiley & Sons, pp. 258–262.

Green, A. E. & P. M. Naghdi (1986) “A nonlinear theory of water waves for finite and infinite depths.” Philos. Trans. Royal Soc. London, Series. A, 320(1552):37–70.

Green, A. E. & P. M. Naghdi (1987) “Further developments in a nonlinear theory of water waves for finite and infinite Depths.” Philos. Trans. Royal Soc. London, Series A, 324(1577):47–72.

Grim, O. (1961) “Beitrag zu dem Problem der Sicherheit des Schiffes im Seegang.” Schiff und Hafen, 6:490–497.

Gu, M., J. Lu & T. Wang (2012) “An investi-gation on stability under dead ship condition of a tumblehome hull.” Proc. 13th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 593– 598.

Gudmestad, O. T. (1993) “Measured and predicted deep water wave kinematics in regular and irregular seas.” Marine Structures, 6:1–73.

Handschel, S., N. Köllisch, J. P. Soproni, M. Abdel-Maksoud (2012) “A numerical method for estimation of ship roll damp-ing for large amplitudes.” Proc. 29th Symp. Naval Hydrodynamics, Gothen- burg, Sweden, pp. 1–13.

Harmsen, E. (2006) “Development of proba-bility based quasi-static damage stability criteria for naval ships.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

Harvey, A. H., A. P. Peskin & S. A. Klein (2008) “NIST/ASME Steam Properties Version 2.22: Users’ Guide.” National Institute for Standards and Technology, Gaithersburg, MD.

Haver, S. &, O. J. Andersen (2000) “ Freak Waves: Rare Realizations of a Typical Population or Typical Realizations of a Rare Population?”, Proc. 10th Int’l. Off-shore and Polar Engineering Conf., Seattle, WA, pp. 123–132.

Henning, H. L. (2011) “Investigation of the heave, sway and roll motions of typical ship like hull sections using RANS numerical methods.” Masters Thesis, University of Stellenbosch, South Africa.

IAPWS (2008) “Supplementary release on properties of liquid water at 0.1 Mpa.” The Int’l Assn. for the Properties of Water & Steam, Berlin, Germany.

Ikeda, Y., Y. Himeno, N. Tanaka (1976) “On

roll damping force of ship: effects of friction of hull and normal force of bilge keels.” J. Soc. Naval Arch. Japan, 161:41–49. (in Japanese).

403403

Ikeda, Y., Y. Himeno & N. Tanaka (1977a) “One eddy making component of roll damping force on naked hull.” J. Soc. Naval Arch. Japan, 142:54–64.

Ikeda, Y., Y. Himeno & M. Tanaka (1978)

“Components of roll damping of ship at forward speed.” J. Soc. Naval Arch. Japan, 143:121–133.

Ikeda, Y., K. Komatsu, Y. Himeno & N. Tanaka (1977b) “On roll damping force of ship effects of hull surface pressure created by bilge keels.” J. Soc. Naval Arch. Japan, 165:31–40. (in Japanese).

Ikeda, Y., K. Komatsu, Y. Himeno & N.

Tanaka (1979) “On roll damping force of ship—effects of hull surface pressure created by bilge keels.” Report of Dept. of Naval Arch., Univ. Osaka Prefecture, No. 00402.

Ikeda, Y., K. Osa & N. Tatnaka (1988) “Vis-cous forces acting on irregular oscillating circular cylinders and flat plates.” J. Offshore Mech. and Arctic Engin., ASME, 110(2):140–147.

IMO (2005) MSC 80/24/Add. 1 “Report of the Maritime Safety Committee on its Eighti-eth Session.” London, UK.

IMO (2006) MSC.1/Circ 1200 Annex. “In-terim guidelines for alternative assessment of the weather criterion.” London, UK, 17 p.

IMO (2007a) MSC 82/24/Add.1 “Adoption of

Amendments to the Int’l. Convention for the Safety of Life at Sea, 1974, as Amended.” London, UK.

IMO (2007b) SLF 50/4/4 “Framework for the development of new generation criteria for intact stability.” submitted by Japan, the Netherlands, and the United States. London, UK, 6 p.

IMO (2008) SLF 51 WP2 “Revision of the intact stability code.” Report of the Working Group (Part 1), London, UK, 21 p.

IMO (2009) SLF 52/INF.2 “Information Collected by the Intercessional Correspon-dence Group on Intact Stability.” Submit-ted by Japan. London, UK, 138 p.

IMO (2010) SLF 52/WP1 “Development of new generation intact stability criteria.” Report of the Working Group (Part 1), London, UK, 15 p.

IMO (2011) SLF 54/3/1, Annex 1 “Framework for the second generation intact stability criteria, in development of second genera-tion intact stability criteria.” Report of the Working Group at SLF 53 (Part 2), Submitted by the Chairman of the Working Group, London, UK, 5 p.

IMO (2012) SLF 54/WP.3 “Development of second generation intact stability criteria, any other business.” Report of the Working Group (Part 1). London, UK, 24 p.

IMO (2013a) SLF 55/3/11 “Development of second generation intact stability criteria, comparison study of draft level 2 vulnerability criteria for stability under dead ship condition.” Submitted by Italy and Japan. London, UK, 4 p.

404404

IMO (2013b) SLF 55/WP.3 “Development of second generation intact stability criteria; development of amendments to Part B of the 2008 IS Code on Towing, and Anchor Handling Operations; Development of amendments to the Criterion For Maxi-mum Angle of Heel in Turns of the 2008 IS Code; Development of a mandatory code for ships operating in polar waters.” Report of the Working Group (Part 1). London, UK, 540 p.

IMO (2013c) SDC 1/INF.8 “Development of second generation intact stability criteria.” Information collected by the Correspon-dence Group on Intact Stability regarding second generation intact stability criteria development, Submitted by Japan. Lon-don, UK, 132 p.

IMO (2014) SDC 1/WP5 “Development of second generation intact stability criteria.” Report of the Working Group. London, UK, 5 p.

IOC, SCOR & IAPSO (2010) “The interna-tional thermodynamic equations of sea-water—2010: calculation and use of ther-modynamic properties.” Intergovernmen-tal Oceanographic Commission, Manuals and Guides No. 56, UNESCO, 196 pp.

Irvine, M., Jr. (2004) “Towing tank tests for surface combatant for coupled pitch and heave and free roll decay motions.” Doctorial thesis, University of Iowa, Iowa City, IA.

Irvine M., Jr., J. Longo & F. Stern (2004) “Towing tank tests for surface combatant for free roll decay and coupled pitch and Heave motions.” Proc. 25th Symp. Naval Hydrodynamics.

ISSW (2011) Proceedings of the 12th Interna-tional Ship Stability Workshop. V. Belenky. Ed., Washington, DC, iv+420 p.

ISSW (2013) Proceedings of the 13th Interna-tional Ship Stability Workshop. Brest, France.

ITTC (2008) “Guide in the expression of uncertainty in experimental hydrody-namics.” ITTC Recommended Procedures and Guidelines, 7.5-02-01-01, Revision 01, 25th Int’l. Towing Tank Conf., 17 p.

ITTC (2011a) “The Specialist Committee on Stability in Waves: Final Report and Recommendations to the 26th ITTC.” Proc. 26th ITTC, Rio de Janeiro, Brazil, 35 p..

ITTC (2011b) “Numerical estimation of roll damping.” ITTC Recommended Proce-dures 7.5-02-07-04.5, 26th Int’l. Towing Tank Conf., 33 p.

Jalonen, R., P. Ruponen, A. Jasionowski, P. Maurier, M. Kajosaari & A. Papanikolaou (2012) “FLOODSTAND overview of achievements.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles (STAB'12), Athens, Greece, pp. 819–830.

405405

Jaouen, F., A. Koop & G. Vaz (2011) “Predic-ting roll added mass and damping of a ship hull section using CFD.” Proc. ASME 30th Int’.l Conf. Offshore Mech. & Arctic Engin., Rotterdam, Netherlands, 11 p.

Jasionowski, A., D. Vassalos & L. Guarin (2004) “Theoretical developments on survival time post-damage.” Proc. 7th Int’l. Ship Stability Workshop, Shanghai, China.

Jasionowski, A., D. Vassalos & L. Guarin (2010) “Time-based survival criteria for passenger Ro-Ro vessels.” Proc. 11th Int’l. Ship Stability Workshop, Wageningen, Netherlands.

JCGM (2008) “Evaluation of measurement data—guide to the expression of uncer-tainty in measurement.” JCGM 100:2008 GUM 1995 with minor corrections, Joint Committee for Guides in Metrology, Bureau Int’l. des Poids Mesures (BIPM), Sèvres, France.

Karlberg, D., R. Jalonen & P. Ruponen (2011) “Sensitivity analysis for the input data in flooding simulation.” FLOODSTAND Report D2.6

Katayama, T., J. Umeda, H. Hashimoto & B. Yildiz (2013) “A study on roll damping estimation for non periodic motion.” Proc. 13rd Int’l. Ship Stability Workshop, Brest, France, pp. 44–49.

Katayama T., Y. Yoshioka, T. Kakinoki & Y. Ikeda (2010) “Some Topics for Estimation of Bilge-keel Component of Roll Damping.” Proc. 11th Int’l. Ship Stability Workshop, Wageningen, the Netherlands, pp. 225–230.

Katayama, T., Y. Yoshioka, T. Kakinoki & S. Miyamoto (2011) “A study on estimation method of bilge-keel component of roll damping for time domain simulation.” Proc. 12th Int’l. Ship Stability Workshop, Washington, DC, pp. 309–316.

Kobyliński, L. (2006) “Alternative stability requirements based on system and risk approach.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brazil.

Kobyliński, L. (2007) “Goal-based stability standards.” Proc. 9th Int’l. Ship Stability Workshop, Hamburg, Germany.

Kobyliński, L. (2012) “Stability criteria—50 years of experience and future prospects.” Proc. 13th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 91–100.

Korpus, R. & J. Falzarano (1997) “Prediction of viscous roll damping by unsteady Navier–Stokes techniques.” J. Offshore Mech. and Arctic Engin., 119:108–113.

406406

Krueger, S. & F. Kluwe (2010) “Development of threshold values for a minimum sta-bility criterion based on full scale acci-dents.” Technische Universitat Hamburg- Harburg, Institut fur Entwerfen von Schiffen und Schiffssicherheit, Final Report BMWi Project LaSSe—Loads on Ships at Sea, 03 SX 128 D, 167 p. (English Translation from the German original.)

Kubo, T., N. Umeda, S. Izawa & A. Matsuda, (2012) “Total stability failure probability of a ship in irregular beam wind and waves: model experiment and numerical simulation.” Proc. 11th Int’l. Conf. Stabi-lity of Ships and Ocean Vehicles, Athens, Greece.

Kwon, S., Q. Chen, G. Mermiris, & D. Vassalos, (2012) “Coupling of progres- sive structural failure and loss of stability in the safe return to port framework.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Lee, G. J. (2014) “Moment of Inertia of Liquid in a Tank.” Int’l. J. Naval Architecture and Ocean Engineering, 6(1):132–150.

Longuet-Higgens, M. S. (1957) “The statistical analysis of a random moving surface.” Philos. Trans. Royal Soc. London, Series A, 249:321–387.

Luke, Y. L. (1976) Mathematical Functions and their Approximations. New York, NY, Academic Press, Inc. 584 pp.

Miller, R., J. Gorski & D. Fry (2002) “Viscous roll predictions of a circular cylinder with bilge keels.” Proc. 24th Symposium on Naval Hydrodynamics, Fukuoka, Japan.

Montewka, J., F. Goerlandt, S. Ehlers, T. Hinz & P. Kujala (2013) “A risk framework for maritime transportation systems.” Proc. 13th Int’.l Ship Stability Workshop, Brest, France, pp. 58–69.

Naar, H. & S. Vaher (2010) “Numerical study on the critical pressure heads.” FLOOD-STAND Report D2.2a

Naess, A. & T. Moan (2012) Stochastic Dynamics of Marine Structures. Cam-bridge University Press, UK.

Newman, J. N. (1987) “Evaluation of the wave-resistance Green function: Part 1— The double integral.” J. Ship Research, 31(2):79–90.

Ochi, M. K. (1973) “On Prediction of extreme values.” J. Ship Research, 17(1):29–37.

Ochi, M. K. (1990) Applied Probability and Stochastic Processes. John Wiley and Sons.

Oliveira, A. C. & A. C. Fernandes (2009) “An alternative approach for roll damping estimation.” Proc. 4th Int’l. Workshop on Applied Offshore Hydrodynamics, Rio de Janeiro, Brazil.

407407

Paap, M. (2005) Verification of CFD cal-culations with experiments on a rolling circular cylinder with bilge keels in a free surface.” Masters Thesis, Delft University of Technology & Bluewater Energy Services, Delft, Netherlands.

Papanikolaou, A., C. Mains, S. Rusaas, R. Szalek, N. Tsakalakis, D. Vassalos & G. Zaraphonitis (2010) “GOALDS – Goal based damage stability.” Proc. 11th Int’l. Ship Stability Workshop, Wageningen, Netherlands.

Park, J. C., Y. Uno, H. Matsuo, T. Sato & H. Miyata (2001) “Reproduction of fully- nonlinear multi-directional waves by a 3D viscous numerical wave tank.” ISOPE 2001, Stavanger, Norway.

Pawłowski, M. (2008) “Closure on survival time.” Proc. 10th Int’l. Ship Stability Workshop, Daejeon, Korea.

Perrault, D. (2013) “Examination of the probability results for extreme roll of naval vessels.” Proc. 13th Int’l. Ship Stab-ility Workshop, Brest, France, pp. 178– 184.

Peters, A. J. & P. Harrison (2006) “A potential framework for a performance based damage stability standard for naval vessels.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

Peters, A. J. & D. Wing (2009), “Stability Criteria Evaluation and Performance Based Criteria Development for Damaged Naval Vessels.” Proc. 10th Int’l Conf. Stability of Ships and Ocean Vehicles, St. Petersburg, Russia.

Peters, W. S., V. Belenky & A. M. Reed, (2012) “On regulatory framework of direct stability assessment.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Peters, W., V. Belenky & K. J. Spyrou (2013) “On alignment of the second generation IMO intact stability criteria with the goal based standards: Evaluation of the safety level.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France, pp. 243–250.

Pierce, R. D. (1992) “Run length and statistical error estimation for seakeeping tests and trials.” Proc. Twenty-third American Tow-ing Tank Conf., pp. 73–82.

Press, W. H., B. P. Flannery, S. A. Teukolsky & W. T. Vetterling (1986) Numerical Recipes: The Art of Scientific Computing. Cambridge University Press, Cambridge, UK.

Provosto, M. & G. Z. Forristall, (2002) “Statistics of wave crests from models vs. measurements.” Proc. ASME 22th Int’l. Conf. Offshore Mech. & Arctic Engin., Oslo, Norway.

Ran, H., I. Rask & C. E. Janson (2012) “Damaged Ro-Pax vessel time to capsize.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Reed, A. M. (2009). “A naval perspective on ship stability.” Proc. 10th Int’l. Conf. Stability of Ships and Ocean Vehicles, St. Petersburg, Russia, pp. 21–44.

408408

Roddier, D. (2000) “Hydrodynamics of rolling cylinders with and without bilge keels.” PhD Dissertation, University of California at Berkeley.

Rosén A., L. Björnsson, M. Palmquist & E. Ovegård (2013) “On the influence of sea state idealizations and wave directionality in dynamic stability assessments.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France.

Ruponen, P. (2007) “Progressive flooding of a damaged passenger ship.” Doctoral Dis-sertation, Helsinki University of Tech-nology, Department of Mechanical Engin-eering, Helsinki, Finland.

Ruponen, P., M. Larmela, & P. Pennanen (2012) “Flooding prediction onboard a damaged ship.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Sadat-Hosseini, H., F. Stern, A. Olivieri, E. F. Campana, H. Hashimoto, N. Umeda, G. Bulian & A. Francescutto (2010) “Head-wave parametric rolling of a surface combatant.” Ocean Engineering, 37(10): 859–878.

Sarkar, T. & D. Vassalos (2000) “A RANS-based technique for simulation of the flow near a rolling cylinder at the free surface.” Mar. Sci. Technol., 5:66–77

Sclavounos, P. D. (1992) “On the quadratic effect of random gravity waves on floating bodies.” J. Fluid Mechanics, 242:475– 489.

Sheinberg, R., K. Stambaugh, A. Eisele & R. Leadbetter (2006) “Estimating probability of capsize for operator guidance.” Proc. 9th Int’.l Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

Shigunov, V., N. Themelis & K. J. Spyrou (2012) “Critical wave groups vs. direct Monte-Carlo simulations for typical stability failure modes of a container ship.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Skjong, R., E. Vanem & S. Rusås (2006) “Holistic and risk based approach to collision damage stability of passenger ships.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

Smith, T. & B. Campbell (2013) “On the validation of statistical extrapolation for stability failure rate.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France, pp. 79–87.

Spanos, D. & A. Papanikolaou (2007) “On the time to capsize of a damaged RoRo/ Passenger ship in waves.” Proc. 9th Int’l. Ship Stability Workshop, Hamburg, Ger-many.

Spanos, D. & A. Papanikolaou (2010) “On the time dependent survivability of ROPAX ships.” Proc. 11th Int’.l Ship Stability Workshop, Wageningen, Netherlands, pp. 143–147.

409409

Spanos, D. & A. Papanikolaou (2012) “Time dependent survivability against flooding of passenger ships in collision damages.” Proc. 11th Int’l Conf. Stability of Ships and Ocean Vehicles, Athens, Greece, pp. 401–410.

Spanos, D. & A. Papanikolaou (2014) “On the time for the abandonment of flooded passenger ships due to collision damages.” J. Mar. Sci. & Technol.

Spyrou, K. J. & I. Roupas (2006) “Damaged- ship survivability: A step beyond Wendel.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

STAB (2012) Proceedings of the 11th Interna-tional Conference on the Stability of Ships and Ocean Vehicles. K. J. Spyrou, N. Themelis & A. D. Papanikolaou, Eds., Athens, Greece, 927 p.

Stansberg, C. T., O. T. Gudmestad & S. K. Haver (2006) “Kinematics under extreme waves.” Proc. ASME 25th Int’l. Conf. Offshore Mech. & Arctic Engin., Hamburg, Germany, OMAE2006-92491.

Stern, F., J. Yang, Z. Wang, H. Sadat- Hosseini, M. Mousaviraad, S. Bhushan & T. Xing (2013) “Computational ship hydrodynamics: Nowadays and way for-ward.” Int’l. Shipbuilding Progress, 60:3– 105.

Tagg, R. & C. Tuzcu, (2002) “A performance- based assessment of the survival of dam-aged ships—Final outcome of the EU research project HARDER.” Proc. 6th Int’l. Ship Stability Workshop, New York, USA.

Tanaka, N., Y. Himeno, Y. Ikeda & K. Isomura (1981) “Experimental study on bilge-keel effect for shallow-draft ship.” J. Soc. Naval Arch. Japan, 180:69–75.

Tellkamp, J. & H. Cramer, (2002) “A methodology for design evaluation of damage stability.” Proc. 6th Int’l. Ship Stability Workshop, New York, USA.

Themelis, N. & K. J. Spyrou (2006) “Proba- bilistic assessment of resonant instability.” Proc. 9th Int’l. Conf. Stability of Ships and Ocean Vehicles, Rio de Janeiro, Brasil.

Thompson, A. & B. N. Taylor (2008) “Guide for the use of the International System of Units (SI).” NIST Special Publication 811, National Institute of Standards and Technology, Gaithersburg, MD.

Tsakalakis, N., J. Cichowicz & D. Vassalos (2010) “The capsize band concept re- visited.” Proc. 11th Int’l. Ship Stability Workshop, Wageningen, Netherlands.

Umeda, N. (2013) “Current status of second generation intact stability criteria development and some recent efforts.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France, pp. 138–157.

410410

van‘t Veer, R., J. de Kat & P. Cojeen (2002) “Large passenger ship safety: Time to sink.” Proc. 6th Int’.l Ship Stability Work-shop, New York, USA.

van‘t Veer R., W. Peters, A. L. Rimpela & J. de Kat (2004) “Exploring the influence of different arrangements of semi-watertight spaces on survivability of a damaged large passenger ship.” Proc. 7th Int’l. Ship Stability Workshop, Shanghai, China.

Vassalos, D. (2012) “Damage stability of passenger ships—notions and truths.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles (STAB'12), Athens, Greece, pp. 775–790.

Vassalos D. & A. Jasionowski (2013) “Emer-gency response in ship flooding casual-ties.” Proc. 13th Int’l. Ship Stability Work-shop, Brest, France.

Vassalos, D., A. Jasionowski & L. Guarin (2005) “Passenger ship safety—Science paving the way.” Proc. 8th Int’l. Ship Stability Workshop, Istanbul, Turkey.

Vugts, J. H. (1968) “The hydrodynamic coeffi-cients for swaying, heaving and rolling cylinders in a free surface." International Shipbuilding Progress, 15(167):251–276.

Vugts, J. H. (1970) “The hydrodynamic forces and ship motions in waves.” Doctoral Thesis, Delft University of Technology, Delft, Netherlands, 115 p.

Waals, O. J., A. B. Aalbers & J. A. Pinkster (2002) “Maximum likelihood method as a means to estimate the directional wave

spectrum and the mean wave drift force on a dynamically positioned vessel.” Proc. ASME 21th Int’l. Conf. Offshore Mech. & Arctic Engin., Oslo, Norway.

Wanderley, J. B. V., A. Ramiro, T. Reis, A. C. Fernandes & C. Levi (2007) “Numerical simulation of roll damping of a FPSO.” Proc. ASME 26th Int’l. Conf. Offshore Mech. & Arctic Engin., San Diego, USA, OMAE2007-29017.

Wang, L. & T. Moan (2004) “Probabilistic analysis of nonlinear wave loads on ships using Weibull, Generalized Gamma, and Pareto Distributions.” J. Ship Research, 48(3):202–217.

Webster, W. C. (2009) “Evolution and kine-matics of steep, random seas: A compar-ison with usual engineering estimates.” Proc. 4th Int’l. Workshop on Applied Offshore Hydrodynamics, Rio de Janeiro, Brazil.

Wilson, R. V., P. M. Carrica & F. Stern (2006) “Unsteady RANS method for ship motions with application to roll for a surface combatant.” Computers & Fluids, 3:501– 524.

Yago, K., Y. Ohkawa, T. Chuji & T. Utsu-nomiya (2008) “Experimental study on viscous damping force of box-shaped body with fin.” J. Soc. Naval Arch. Japan.

Yang, B., Z.-C. Wang & M. Wu (2012) “Numerical simulation of naval ships’ roll damping based on CFD.” Procedia Engineering, 37:14–18.

411411

Yeung, R. & P. Ananthakrishnan (1992) “Oscillation of a floating body in a viscous fluid.” J. Engin. Math., 26:211– 230

Yeung, R, S.-W. Liao & D. Roddier (1998) “On roll hydrodynamics of rectangular cylinders.” Proc. 8th Int’l. Offshore & Polar Engin. Conf., Montreal, Canada, Vol. 3. pp. 445–53.

Yeung, R, R. K. M. Seah &, J. T. Imamura (2013) “Lateral force and yaw moment on a slender body in forward motion at a yaw angle.” J. Offshore Mech. & Arctic Engin., 135(3): 031101– 031109.

Ypma, E. & E. Harmsen (2012) “Development of a new methodology to predict the capsize risk of ships.” Proc. 11th Int’l. Conf. Stability of Ships and Ocean Vehicles, Athens, Greece.

Zaraphonitis, G., A. Papanikolaou, C. Roussou & A. Kanelopoulou (2013) “Comparative study of damage stability regulations and their impact on the design and safety of modern ROPAX ships.” Proc. 13th Int’l. Ship Stability Workshop, Brest, France, pp. 235–242.

11.2. Nomenclature ASME American Society of Mechanical

Engineers BIPM Bureau Internat’l. Poids Mesures CFD Computational Fluid Dynamics CG Centre of Gravity CPU Central Processing Unit

CresT JIP Cooperative Research on Extreme Seas and their impacT Joint Industry Project

CTO Centrum Techniki Oretowej (Poland)

DoF Degree of Freedom DTMB David Taylor Model Basin EMLM Extended Maximum Likelihood

Method EPOT Envelope Peaks Over Threshold EU European Union FEM Finite Element Method FLOODSTAND Integrated Flooding Control

& Standards for Stability & Crisis Management (EU Project)

FP Forward Perpendicular FPSO Floating Production Storage &

Offloading GM Metacentric Height GOALDS Goal Based Damage Stability GZ Righting Moment Arm ICS Int’l. Council for Science IAPSO Int’l. Assoc. for Physical Sciences

of the Ocean IAPWS Int’l. Assoc. for Properties of

Water & Steam IMO Int’l. Maritime Organisation IOC Intergovernmental Oceanographic

Commission IS Intact Stability ISEI Insufficient Stability Event Index ISCG Intercessional Correspondence

Group ISO/ GUM International Standards

Group/Guide to the Expression of Uncertainty in Measurements

ISSW Int’l. Ship Stability Workshop ITTC Int’l. Towing Tank Conference JCGM Joint Committee for Guides in

Methodology JONSWAP Joint North Sea Wave Observation

Project KC Keulegan-Carpenter number (KC =

UT/2h)

412412

KRISO Korean Research Institute of Ships & Ocean Engineering

LaSSe Loads on Ships at Sea LCG Longitudinal Centre of Gravity LOAS Loss-of-Stability Accident MARIN Maritime Research Institute

Netherlands MEM Maximum Entropy Method MOI Moment of Inertia MSC Maritime Safety Committee NGS Nat’l. Geologic Survey NIST Nat’l. Institute of Standards &

Technology (USA) NOAA Nat’l. Oceanic & Atmospheric

Administration (USA) NSWCCD Carderock Division, Naval Surface

Warfare Centre OMAE Int’l. Conf. on Ocean, Offshore

and Arctic Engineering ONR Office of Naval Research PF Probability of Capsizing PIV Particle Image Velocimetry POT Peak Over Threshold RANS Reynolds Averaged Navier-Stokes RAO Response Amplitude Operator RDLI Relative Damage Loss Index RINA Royal Institution of Naval

Architects RO-RO Roll On-Roll Off ROPAX Roll On Passenger SAIC Science Applications Int’l. Corp. SCOR Scientific Committee on Ocean

Research SI Int’l. System of Units SiW Specialist Committee on Stability

in Waves SLF Sub-committee on Stability, Load

Lines & Fishing Vessels Safety SOLAS Int’l. Convention for the Safety of

Life at Sea STAB Int’l Conf. Stability of Ships &

Ocean Vehicles UNESCO United Nations Educational,

Scientific & Cultural Organization

URANS Unsteady Reynolds Averaged Navier-Stokes

V&V Verification & Validation VCG Vertical Centre of Gravity

413413

Quality Systems Group

Final Report and Recommendations to the 27th ITTC

1. GENERAL

1.1. Membership and Meetings Benedetti L., CNR-INSEAN (Secretary) Derradji A., NRC Ferrando M., University of Genova, (Chair) Johnson B., US Naval Academy (senior) Kobayashi E., Kobe Univ. Morabito M. G., US Naval Academy Park J. T. NSWC Carderock Div. Pérez Rojas Luis, ETSIN Sena Sales Jr J., LabOceano van Rijsbergen M., MARIN Woodward M. D., Newcastle Univ

From March 4th 2013 Morabito M. G., US Naval Academy replaced Johnson B., US Na-val Academy. Professor Jonson will remain a corresponding member of the group.

From September 2013 Park J. T. Naval Sur-

face Warfare Center Carderock Div. replaced Derradji A., NRC

The Group held four meetings as follows:

Rio de Janeiro, September 3rd 2011 Madrid, June 25th to 27th 2012 Annapolis, July 1st to 3rd 2013 Genoa, January 27th to 29th 2014.

From here on, in order to save space in the report, the Quality Systems Group will be ad-dressed as QSG.

1.2. Terms of Reference given by the 26th ITTC to the QSG.

1) Include a definition of the terms Verifi-

cation and Validation in the ITTC Symbols and Terminology List (to be done within first three months as a basis for the work of other committees).

2) Maintain the Manual of ITTC Recom-mended Procedures and Guidelines. Co-ordinate the modification and re-editing of the existing procedures according to the comments made by ITTC member organizations at the Conference and by the Technical Committees.

3) Support the Technical Committees in their work on Recommended Proce-dures. Supply the chairmen of the new committees at the beginning of the pe-riod with the MS Word versions of the relevant procedures and the template for the production of new procedures.

4) Observe the development or revision of ISO Standards regarding Quality Con-trol.

5) Update the ITTC Symbols and Termi-nology List.

6) Update the ITTC Dictionary of Hydro-mechanics.

414414

7) Cross-check the ITTC Symbols List and the Dictionary with other standards e.g. ISO.

8) Revise and update the existing ITTC Recommended Procedures according to the comments of Advisory Council, Technical Committees and the Confer-ence.

9) Before the third AC Meeting, review and edit new ITTC Recommended Pro-cedures with regard to formal Quality System requirements including format and compliance of the symbols with the ITTC Symbols and Terminology List.

10) Follow the implementation of the Benchmark data repository.

11) Support the technical committees with guidance on development, revision and update of uncertainty analysis proce-dures.

12) Observe ISO standards for uncertainty analysis, in particular the uncertainty analysis terminology.

13) Maintain Wiki for the 27th ITTC as a trial period and create link to it from the ITTC website.

2. PERFORMED TASKS 2.1. Include a definition of the terms Veri-

fication and Validation in the ITTC documents

The QSG has agreed on the following defi-

nitions: "Verification, Validation, and Accreditation

are three interrelated but distinct processes that gather and evaluate evidence to determine, based on the simulation's intended use, the simulation's capabilities, limitations, and per-

formance relative to the real-world objects it simulates."

• Verification is the process of determin-

ing that a model or simulation implemen-tation accurately represents the devel-oper's conceptual description and specifi-cation. (i.e., does the code accurately im-plement the theory that is proposed to model the problem at hand?)

• Validation is the process of determining the degree to which a model or simula-tion is an accurate representation of the real world from the perspective of the in-tended uses of the model or simulation. (i.e., does the theory and the code that implements the theory accurately model the relevant physical problem of inter-est?)

• Accreditation is the official determina-tion that a model or simulation, is accept-able for use for a specific purpose. (i.e., is the theory and the code that implements it adequate for modeling the physics rele-vant to a specific platform? In other words, are the theory and code relevant to the type of vessel for which it is being accredited?)

A letter was sent to all of the Chairmen

with the definitions proposed by the QSG and agreed upon by the AC Chairman.

The definitions have been entered into the

ITTC Dictionary of Hydromechanics.

2.2. Maintain the Manual of ITTC Rec-ommended Procedures and Guidelines

The revision of the Manual of ITTC Rec-

ommended Procedures and Guidelines con-cerned 54 documents:

415415

• 2 existing procedures were deleted • 17 new Procedures/Guidelines have

been evaluated, 16 have been approved and one postponed

• 35 existing procedures have been re-

viewed or updated, the revision of three of which have been postponed.

The revision outcome is illustrated in Table

1. In the process of revising the procedures

and guidelines submitted by the committees, some apparent inconsistencies have been found between the categorization (Procedure or Guideline) and the contents/titles of the docu-ments.

Specifically, 7.5-01-03-03, 7.5-02-02-02

and 7.5-02-07-03.7 were categorized as Proce-dures whereas the title and contents refer to them as being Guidelines.

Similarly, 7.5-02-07-04.4 was categorized as a Guideline whereas the contents refer to it as being a Procedure.

To rectify this situation, the Advisory

Council agreed to the change of categorization of the mentioned documents. Accordingly, documents 7.5-01-03-03, 7.5-02-02-02 and 7.5-02-07-03.7 will be marked as Guidelines and document 7.5-02-07-04.4 will be labelled as Procedure.

Table 1: Outcome of the Manual of ITTC Recommended Procedures and Guidelines Maintenance

New/ Revi-sed

Number Pr. /Gl Title

AC de-

cision

R 1.0-01 Description and Rules of the ITTC A R 1.0-02 Committee Structure of ITTC A N 1.0-04 P Decision Making Between Conferences A

R 4.2.3-01-02 G Guidelines for Preparation of Technical Committee and Group Reports A

R 4.2.3-01-03 P Work Instruction for Formatting ITTC Recommended Procedures A

R 7.5-01-03-01 P Uncertainty Analysis, Instrument Calibration A

R 7.5-01-03-03 P Guideline on the Uncertainty Analysis for Particle Im-age Velocimetry A

N 7.5-01-03-04 G Benchmark for PIV(2C) and SPIV(3C) setups A

R 7.5-02-01-01 P Guide to the Expression of Uncertainty in Experimental Hydrodynamics A

416416

N 7.5-02-01-04 G Guideline on Best Practices for the Applications of

PIV/SPIV in Towing Tanks and Cavitation Tunnels A

N 7.5-02-01-05 P Model scale noise measurements A

R 7.5-02-02-02 P General Guidelines for Uncertainty Analysis in Re-sistance Tests A

N 7.5-02-02-02.1 G Example for Uncertainty Analysis of Resistance tests in Towing Tank A

N 7.5-02-02-02.2 G Practical Guide for Uncertainty Analysis of Resistance Measurement in Routine Tests A

R 7.5-02-03-01.2 P Uncertainty Analysis Example for Propulsion Test PP R 7.5-02-03-01.4 P 1978 ITTC Performance Prediction Method A

N 7.5-02-03-01.6 G Hybrid Contra-Rotating Shaft Pod Propulsor Model Test A

R 7.5-02-03-02.1 P Open Water Test A

R 7.5-02-03-02.3 P Nominal Wake Measurements by LDV, Model Scale Experiments A

R 7.5-02-03-03.2 P Description of Cavitation Appearances A

R 7.5-02-03-03.3 P Cavitation Induced Pressure Fluctuations Model Scale Experiments A

R 7.5-02-03-03.4 P Cavitation Induced Pressure Fluctuations Numerical Prediction Methods A

R 7.5-02-04-01 P General Guidelines PP R 7.5-02-04-02 P Test Methods for Model Ice Properties R R 7.5-02-04-02.1 P Resistance Test in Level Ice PP R 7.5-02-05-04 P Seakeeping Tests A R 7.5-02-05-05 P Evaluation and Documentation of HSMV A R 7.5-02-06-01 P Free Running Model Tests A R 7.5-02-06-02 P Captive Model Test Procedure A R 7.5-02-06-03 P Validation of Manoeuvring Simulation Models A

R 7.5-02-06-04 P Uncertainty Analysis for manoeuvring predictions based on captive manoeuvring tests A

N 7.5-02-06-05 G Uncertainty Analysis for free running model tests A R 7.5-02-07-02.1 P Seakeeping Experiments A

R 7.5-02-07-02.2 P Predicting of Power Increase in Irregular Waves from Model Test A

R 7.5-02-07-02.3 P Experiments on Rarely Occurring Events A

R 7.5-02-07-02.4 P Validation of Seakeeping Computer Codes in the Fre-quency Domain D

R 7.5-02-07-03.3 P Model Tests on Tanker-Turret Systems D R 7.5-02-07-03.7 P Wave Energy Converter, Model Test Experiments A

417417

N 7.5-02-07-03.8 P Model Tests for Offshore Wind Turbines A N 7.5-02-07-03.9 P Model Tests for Current Turbines A N 7.5-02-07-03.10 G Guideline for VIV Testing A N 7.5-02-07-03.11 G Guidelines for VIM Testing PP R 7.5-02-07-04.2 P Model Tests on Damage Stability in Waves A

R 7.5-02-07-04.4 G Numerical Simulation of Capsize Behaviour of Dam-aged Ships in Irregular Beam Seas A

R 7.5-03-02-03 G Practical Guidelines for Ship CFD Applications A N 7.5-03-02-04 G Practical Guidelines for Ship Resistance CFD A N 7.5-03-03-01 G Practical Guidelines for Ship Self-Propulsion CFD A

N 7.5-03-03-02 G Practical Guidelines for RANS Calculation of Nominal Wakes A

N 7.5-03-04-02 G V&V of RANS Solutions in the Prediction of Manoeu-vring Capabilities A

R 7.5-04-01-01.1 P Preparation and Conduct of Speed/Power Trials A R 7.5-04-01-01.2 P Analysis of Speed/Power Trial Data A N 7.5-04-04-01 P Underwater Noise from Ships, Full Scale Measurements A

Legend

A = Accepted D = Deleted PP = Proposing Postponed

2.3. Support technical committees in their

work on Recommended Procedures

MS Word files containing the procedures to be updated, together with the template to be used for drafting new procedures has been sent to the Chairmen of the ITTC Committees.

The Committees were also supplied with

the “Guidelines for Preparation of Technical Committee and Working Group Reports”.

2.4. Observe the development or revision of ISO Standards regarding Quality Con-trol.

The present version of the ISO Standards

for Quality Management System is the ISO 9001:2008. This version basically re-narrates ISO 9001:2000. The 2008 version only intro-

duced clarifications to the existing require-ments of ISO 9001:2000 and some changes intended to improve consistency with ISO 14001:2004. No new requirements were added.

Nevertheless, a new version of the Standard

will be published in December 2015 by the ISO, pending favourable vote by the members in March 2015.

The process involves a number of draft re-

leases and interested parties have been invited to comment at various stages of the Standard’s production. The first draft, called the ‘Commit-tee Draft’, of ISO 9001:2015 was published in May 2013 and was available for consultation among members of ISO/TC 176/SC 2 (the ISO committee that is leading the revision process) until August 2013.

418418

The impact of this revision will be similar

to, if not greater than the 2000 edition, which was a major change for accreditation bodies, certification bodies, training organizations, implementing organizations, procurement or-ganizations, consultants and customers. The transition period for ISO 9001:2000 was three years and the expectation is that for the 2015 revision it will be the same, so activity is being planned up to 2018.

The general changes can be summarized as

follows:

• Adoption of a high-level structure and terminology of Annex SL, a unified guideline used for the development of all new ISO standards (ISO/IEC, 2013).

• New redaction to increase clarity and ac-cessibility, reducing room for interpreta-tion.

• Introduction of two new clauses relating

to the context of the organization: under-standing the organization and its context and understanding the needs and expecta-tions of interested parties.

• Renders the adoption of a process ap-

proach in the implementation of a quality management system more explicit, by in-cluding a clause, which specifies the re-quirements for the adoption of a process approach.

• Replaces the term ‘products’ by ‘goods

and services’, in order to remove the ex-isting bias towards organizations dealing with physical products. As a result, the new standard will be applicable for or-ganizations of any kind.

• Does not contain a clause with specific

requirements for preventive action. ISO

motivates this decision by arguing that prevention is the task of the quality man-agement system in its entirety, as op-posed to a specific subsection of it.

However, this updated edition makes refer-

ence only to the Quality Management System and the Technical Procedures are not affected. The Technical Procedures are normally based on ITTC Procedures.

QSG feels obliged to submit to the Confer-

ence another quality standard that could be appropriate for our experimental activity: the ISO 17025 (ISO 2013), which sets general re-quirements for the competence of testing and calibration laboratories, being the global qual-ity standard for testing and calibration laborato-ries. It is the basis for accreditation from an accreditation body, but an accreditation body for towing tanks does not exist. The current release was published in 2005.

Two main clauses are included in ISO/IEC

17025 – Management Requirements and Tech-nical Requirements. Management requirements are related to the operation and effectiveness of the quality management system within the laboratory, and this clause has similar require-ments to ISO 9001. Technical requirements address the competence of staff; testing meth-odology; equipment and quality; and reporting of test and calibration results.

Implementing ISO/IEC 17025 has benefits

for laboratories, but the work and costs in-volved should be considered before proceed-ing.

419419

2.5. Update the ITTC Symbols and Termi-

nology List. After the last revision, the List is found to

be up-to-date and does not require a major check.

Some minor maintenance has been per-

formed as follows:

• The symbol CAPP has been added; • The symbol LPP has been added to the

alphabetical list where it was missing; • The symbol c0.7 has been added as a con-

sequence of the cross check with the ISO Standards;

• The symbol Re0.7 has been added.

2.6. Update the ITTC Dictionary of Hydromechanics.

Revised or new entries:

• Rake angle; • Skew; • Pod.

Added figures:

• Co-ordinate planes; • Rake; • Set back; • Blade section.

2.7. Cross-check the ITTC Symbols List and the Dictionary with other stan-dards e.g. ISO.

QSG considered the following standards,

which were provided by the AC and were not cross checked during the preceding period be-cause of time shortage:

ISO 3715-1:2002; Ships and marine tech-nology – Propulsion plants for ships -- Part 1: Vocabulary for geometry of propellers,

ISO 3715-2:2001; Ships and marine tech-

nology -- Propulsion plants for ships -- Part 2: Vocabulary for controllable-pitch propeller plants,

ISO 7255:1985; Shipbuilding -- Active con-

trol units of ships – Vocabulary, ISO 7462:1985; Shipbuilding -- Principal

ship dimensions -- Terminology and definitions for computer applications,

ISO 8384:2000; Ships and marine technol-

ogy -- Dredgers – Vocabulary, ISO/TR 13298:1998; Ships and marine

technology -- Vocabulary of general terms, ISO 19018:2004; Ships and marine tech-

nology -- Terms, abbreviations, graphical sym-bols and concepts on navigation.

As regards ISO/TR 13298, ISO 19018 and

ISO 8384, no modification to the dictionary is required.

The following gives a list of the addi-

tions/modifications required by the relevant standard.

ISO 3715-1:

• Leading edge, blade; • Leading edge, foil section; • Pitch; • Pitch angle; • Pitch, at a certain radius; • Pitch, blade mean; • Pitch, propeller mean; • Propeller reference system, cylindrical; • Propeller reference system, rectangular;

420420

• Trailing edge, blade; • Trailing edge, foil section.

ISO 3715-2:

• Propeller; • Pitch angle, range of; • Pitch, design propeller; • Pitch, maximum ahead; • Pitch, maximum astern; • Pitch, nominal; • Blade position; • Blade position, angle of; • Propeller Windmilling.

ISO 7462:

• Axis co-ordinate; • Baseline; • Section; • all occurrences of Beam replaced with

Breadth; • Displacement Volume; • Moulded.

ISO 7255: added 6 new definitions:

• Lateral thruster; • Retractable lateral thruster; • Rudder-propeller; • Swivelling rudder-propeller; • Retractable Rudder-propeller; • Active rudder.

The Symbol and Terminology List has been

accordingly updated where required. Some definitions given in the ISO Stan-

dards conflict with the relevant ITTC defini-tions:

• the definition of Body Axes given in ISO

3715-1;

• the definition of Skew given in ISO 3715-1 ;

• the definition of Propeller reference sys-

tem, cylindrical given in ISO 3715-1; • the definition of Baseline given in ISO

7462.

For the time being, pending decisions of the AC or the Conference, a statement was added to the questioned definitions warning that ‘the definition is not in line with ISO’.

2.8. Revise and update the existing ITTC Recommended Procedures.

The QSG updated the following

procedures:

4.2.3-01-03 Work Instruction for format-ting ITTC Recommended Procedures and Guidelines

7.5-01-03-01 Uncertainty Analysis, In-strument Calibration

7.5-02-01-01 Guide to the Expression of Uncertainty in Experimental Hydrodynamics

Procedure 4.2.3-01-03 has been updated in

order to ameliorate the format of the Recom-mended Procedures.

Minor corrections have been made to the

ITTC version of the GUM, Procedure 7.5-02-02-02 ITTC (2014a). Equations (20) and (21) have been corrected where a factor of 4 was omitted from the coefficient for the shaft rota-tional rate, un, for the thrust and torque coeffi-cients. The explanation on the tolerance for weights has been restated in Section 14.1, and equation (31c) has been corrected. The refer-ences to ISO have been replaced with JCGM.

421421

Errors were also discovered in the Instru-

ment Calibration Procedure, 7.5-01-03-01 ITTC (2014b). A sign error has been corrected in equation (19). The tolerance of weights in section 5 has been re-stated, and equation (24) corrected. The reference list has been updated.

Section 7 has been added on Direct Digital

Calibration. Frequency signals from shaft rota-tional rate and carriage speed from a wheel have been processed with frequency to voltage (f-v) convertors. F-v converters are subject to drift. Current data acquisition cards have a counter port and built-in timing so that fre-quency can be measured directly without the need for an f-v converter.

During the revision process the QSG ob-

served the non-compliance of procedures on sea-trials 7.5-04-01-01.2.1 and 7.5-04-01-01.2.2 with UA concepts and JCGM GUM standards. Furthermore QSG noted some in-

consistencies in the use of symbols and the use of symbols not in the SaT List (such as RAW Rwave, rho and rho0).

Since these procedures are somehow con-

nected with IMO and cannot be promptly cor-rected, Postponing the further updating of these documents to the next ITTC period is recom-mended.

In the framework of above-mentioned pro-

cedures QSG suggests the develop of a new procedure on full-scale torque measurements.

2.9. Review and edit new ITTC Recom-mended Procedures with regard to formal Quality System requirements

The QSG review process regarded 35 exist-

ing and 17 new procedures adding to a total of 52 documents, as illustrated in Table 2.

Table 2: List of the reviewed procedures

Committee Procedure No. Procedure title

Advisory Council / Executive Committee/Secretariat

1.0-01 Description and Rules of the ITTC 1.0-02 Committee Structure of ITTC

1.0-04 Decision Making between Confer-ences

4.2.3-01-02 Guidelines for Preparation of Techni-cal Committee and Group Reports

Resistance

7.5-02-02-02.1 Example for Uncertainty Analysis of Resistance Tests in Towing Tank

7.5-02-02-02.2 Practical Guide for Uncertainty Analysis of Resistance Measurement in Routine Tests

7.5-02-02-02 General Guideline for Uncertainty Analysis in Resistance Tests

Propulsion

7.5-02-03-01.4 1978 ITTC Performance Prediction Method

7.5-02-03-01.6 Hybrid Contra-Rotating Shaft Pod Propulsors Model Test

7.5-02-03-02.3 Nominal Wake Measurement by LDV Model Scale Experiments

422422

7.5-02-03-03.2 Description of Cavitation Appearances

7.5-02-03-03.3 Cavitation Induced Pressure Fluctua-tions Model Scale Experiments

7.5-02-03-03.4 Cavitation-Induced Pressure Fluctua-tions: Numerical Prediction Methods

Manoeuvring

7.5-02-05-05 Evaluation and Documentation of HSMV

7.5-02-06-01 Free Running Model Tests 7.5-02-06-02 Captive Model Test Procedure

7.5-02-06-03 Validation of Manoeuvring Simulation Models

7.5-02-06-04 Uncertainty Analysis for manoeuvring predictions based on captive manoeu-vring tests

7.5-02-06-05 Uncertainty Analysis for free running model tests

7.5-03-04-02 Validation and Verification of RANS Solutions in the Prediction of Ma-noeuvring Capabilities

Seakeeping

7.5-02-05-04 Seakeeping Tests 7.5-02-07-02.1 Seakeeping Experiments

7.5-02-07-02.2 Seakeeping prediction of Power In-crease in Irregular Waves from Model Tests

7.5-02-07-02.3 Experiments on Rarely Occurring Events

Ocean Engineering 7.5-02-07-03.10 Guideline for VIV Testing 7.5-02-07-03.11 Guideline for VIM Testing

Stability in Waves

7.5-02-07-04.2 Model Tests on Damage Stability in Waves

7.5-02-07-04.4 Simulation of Capsize Behaviour of Damaged Ships in Irregular Beam Seas

CFD in Ship Hydrodynamics

7.5-03-02-03 Practical Guidelines for Ship CFD Applications

7.5-03-02-04 Practical Guidelines for Ship Resis-tance CFD

7.5-03-03-01 Practical Guidelines for Ship Self-Propulsion CFD

7.5-03-03-02 Practical Guidelines for RANS Calcu-lation of Nominal Wakes

Detailed Flow Measurements 7.5-01-03-03 Guideline on the Uncertainty Analysis for Particle Image Velocimetry

7.5-01-03-04 Benchmark for PIV(2C) and

423423

SPIV(3C) setups

7.5-02-01-04 Guideline on Best Practices for the Applications of PIV/SPIV in Towing Tanks and Cavitation Tunnels

Performance of Ships in Service 7.5-04-01-01.1 Preparation and Conduct of Speed/Power Trials

7.5-04-01-01.2 Analysis of Speed/Power Trial Data

Hydrodynamic Noise 7.5-02-01-05 Model scale noise measurements

7.5-04-04-01 Underwater Noise from Ships, Full Scale Measurements

Testing of Marine Renewable De-vices

7.5-02-07-03.7 Wave Energy Converter Model Test Experiments

7.5-02-07-03.8 Model Tests for Offshore Wind Tur-bines

7.5-02-07-03.9 Model Tests for Current Turbines

Ice 7.5-02-04-01 General Guidelines for Ice Model

Testing 7.5-02-04-02 Test Methods for Model Ice Properties 7.5-02-04-02.1 Resistance Test in Ice

Quality Systems Group

4.2.3-01-03 Work Instruction for formatting ITTC Recommended Procedures and Guide-lines

7.5-01-03-01 Uncertainty Analysis, Instrument Cali-bration

7.5-02-01-01 Guide to the Expression of Uncer-tainty in Experimental Hydrodynamics

The great majority of the procedures re-

quired an enormous amount of editing with respect to format. This is probably due to the fact that procedure 4.2.3-01-03 was not suffi-ciently clear about the use of Styles. The new version of the document will help to obtain documents in line with the ITTC agreed for-mat.

The document 0.0 Register has been up-

dated accordingly. A template in word format has been pre-

pared to write new procedures in the next ITTC period. To write a new procedure, an author

will open the new file with the following tem-plate:

ProcTemplate_2017.dotx

The file is included in the CD.

2.10. Follow the implementation of the Benchmark data repository.

The Benchmark Data Repository structure

has been decided by the 26th ITTC.

424424

The QSG has tried to locate the benchmark

data, in order to supply them to the web-site administrator for publication. The task proved to be extremely difficult since nobody seems to know who actually has the required data.

To this effect, a request was forwarded to

the AC through the ITTC Secretary aimed go obtain information about the benchmark data location, but no news has been obtained.

The QSG proposes to the Conference to in-

sert into the ToR of each of the Committees an item regarding the location of the performed benchmark data about relevant topics, and to provide the information to the next QSG.

2.11. Support the technical committees with guidance on development, revision and update of uncertainty analysis proce-dures.

During the first meeting in Rio de Janeiro

the Group decided to support the technical committees by appointing a QSG member to follow their work and provide guidance and assistance if required.

Liaison Committee

Rojas Resistance and Propulsion

Woodward Manoeuvering - Stability in

Waves

Sales Seakeeping - Ocean Engineer-ing

Rijsbergen CFD - Detailed Flow Measure-ment and Noise

Benedetti

Perfomance of Ships in Service

Derradij Marine Renewable Devices – Ice

On December 2011 letters have been sent to

all Chairmen asking to appoint a person to liaise with QSG.

Following the invitation letter, a request for support was received from the Manoeuvring Committee. The Manoeuvring Committee (MC) agreed at their first meeting to invite the QSG representative on uncertainty analysis to subsequent meetings. Attending the 2nd meet-ing of the MC (19-21 Nov 2012 – Nantes, France) the QSG representative presented an overview of the changes needed to come in line with ISO and assisted in a workshop to re-view/develop one procedure:

• 7.5-02-06-04 – “Force and Moment Un-

certainty Analysis, Example for Planar Motion Mechanism Test”, Effective date 2008, Rev. 00.

Attending the 3rd meeting of the MC (5-7

June 2013 – Antwerp, Belgium) the QSG rep-resentative assisted with the development of two procedures:

425425

• 7.5-06-05 – “Uncertainty Analysis for

Free-running Model Tests”, Effective date 2014, Rev. 00.

• 7.5-02-06-04 – “Uncertainty Analysis for Manoeuvring Prediction based on Cap-tive Manoeuvring Tests”, Effective date 2014, Rev. 01.

Following the 3rd meeting, additional mate-

rials were worked on in collaboration to assist with the development of the procedures.

Furthermore, the adoption of uncertainty

analysis in hydrodynamic metrology is a neces-sary and on-going task. However, expertise, in this ever-developing field, cannot reasonably be expected to be present in every technical committee.

The format of maintaining a core group of

uncertainty specialists within the QSG, pro-vides a critical mass for the cross-fertilisation of ideas while at the same time providing con-sistency in the support to the ITTC community.

Maintain a core group of uncertainty spe-

cialists within QSG is recommended, which is sufficient in size to achieve succession plan-ning and knowledge transmission.

2.12. Observe ISO standards for uncertainty analysis.

Since the publication of the uncertainty

procedures from the 25th ITTC, which were based upon the ISO Guide to the Uncertainty in Measurement, the responsibility of the GUM and the International Vocabulary for Metrology (VIM) has been transferred to the Bureau In-ternational des Poids Measures (BIPM) and the Joint Committee for Guides in Metrology (JCGM). The latest information may be found on the BIPM web page:

http://www.bipm.org/en/committees/jc/jcgm/ The GUM has been re-released as JCGM

(2008a) and the VIM as JCGM (2008b). A total of seven documents are being developed in support of the GUM. Four have been com-pleted: JCGM (2008c), JCGM (2009), JCGM (2011), and JCGM (2012). Three more are in preparation: JCGM (20xxa, b, c).

JCGM (2009) is an introduction to the

GUM. The body of this report is 15 pages and serves as a good introduction to the GUM. The other six supplemental documents apply to more advanced users of uncertainty analysis.

2.13. Maintain Wiki for the 27th ITTC as a trial period and create link to it from the ITTC website.

The link to Wiki Dictionary has been added

into the ITTC website and it is operative (www.ittc.info). The Wiki is also accessible directly from:

http://www.ittcwiki.org/doku.php

The ITTC Wiki online tool has been main-

tained operative as instructed by 26th ITTC. The online version of the Wiki Dictionary has been updated to reflect the changes approved at the 26th ITTC.

The way of operating and updating the

Wiki has followed so far the policy to imple-ment changes on line only after the ITTC has approved the Dictionary itself (i.e. updates happen every 3-year period). In the spirit of Wiki as a tool, inter-session updates should be allowed under the disclaimer that the online version is not an adopted version of the ITTC Dictionary. This proposal is included in the recommendations of Section 5.

426426

A report on the analytics of the usage of the

website is contained in Annex A to this report.

3. OTHER MATTERS 3.1. Survey of Uncertainty Analysis Proce-

dure Usage

Anecdotal evidence exists in the ITTC community in difficulty in the application of the ITTC uncertainty analysis procedure, ITTC (2014a). For the 28th ITTC, a survey is pro-posed on the application of ITTC (2014a). From the survey results, a simplified step-by-step procedure will be developed on uncer-tainty analysis for novices. Perhaps a second workshop on uncertainty analysis should be conducted by the 28th ITTC.

Further, a surveying of the extent and

breadth of uptake of uncertainty analysis tech-niques and procedures by the hydrodynamic testing community is recommended. This should evaluate the extent to which the four key stages of uncertainty analysis are imple-mented; viz.:

1. Type-B evaluation of zero-order repli-

cation level uncertainties including cal-ibration with traceability to a national standard.

2. Type-A evaluations of the above men-tioned uncertainty sources instead.

3. Evaluation of random uncertainties as-

sessed by time series analysis, repeat measurements or reproduction meas-urements.

4. Evaluation of systematic modeling un-certainties due to model size and inter-facility bias.

3.2. New Procedure on Torsionmeters

Development of a new procedure on tor-sionmeters is proposed for ship trials. Modifi-cations to ships have been proposed for fuel savings such as Cusanelli and Karafiath (2012). The claim is made that 1 % fuel savings from improved ship performance of DDG 51 class ships would result in an annual fuel savings of 100,000 USD per ship. However, if the fuel consumption is known with an uncertainty of 2 %, the 1 % savings is meaningless. No esti-mate exists on the uncertainty in fuel consump-tion for the U. S. Navy. This dilemma is what Kline (1985) has called the “hopeless” experi-ment.

An accurate measure of fuel consumption

by the propulsion system may not be possible since fuel consumption may go to other sources. Consequently, the only direct meas-urement of the ship propulsion performance is the computation of power from the shaft speed and torque. Insel (2008) has described some of the challenges associated with ship powering measurements. Insel (1985) concludes that the uncertainty in ship powering is between 3 and 5 % of full-power. Environmental conditions will increase the uncertainty; consequently, ship trials should be conducted at a low sea state (low waves and wind speed) and low cur-rent. With current estimates in powering, the claimed reduction in power is likely smaller than the estimated uncertainty. Again the ship modification and the ship trial would be a “hopeless” experiment.

If a ship trial is performed at favourable en-

vironmental conditions, the uncertainty in powering must reduced to as low a value as practical. The primary contributor to the uncer-tainty in power is from the torque measure-ment. At present, the best device is a calibrated torsionmeter such as the one in Figure 1.

427427

Torque is measured as the relative rota-

tional displacement of the two rings in the fig-ure. The displacement is measured by strain gages mounted in the connecting bar. The dis-placement and voltages are measured in a cali-bration fixture with instruments traceable to a National Metrology Institute (NMI). Data are transmitted by a wireless device.

Figure 1. Drawing of torsionmeter installation.

The torque is then computed from the fol-

lowing: )/(RLGJQ δ= (1a)

where δ is the measured deflection, G the modulus of rigidity or shear modulus of elastic-ity, R the radial distance to the sensor, L the length of the span between the two rings, and J is the polar moment of inertia. For a hollow shaft,

)/()32/( 4i

4o DDJ −= π (1b)

and Do is the outside diameter of the shaft and Di the inside diameter.

The only item that cannot be measured di-rectly is the value of G. From ITTC (2002), the expanded relative uncertainty in G is ±2.3 %.

From an uncertainty analysis, G is the domi-nant term. Thus, the uncertainty in torque and power is 2.3 % of the full-scale calibration value. In an example calculation, ITTC (2002) estimated the expanded uncertainty in power from all sources as 2.8 % by comparison with Insel (2008) of 3 to 5 %. A reduction in the uncertainty requires a measured value of G with an uncertainty estimate.

However, an ultrasonic gage can measure

the shear-wave velocity, Vs of the shaft mate-rial and G computed by the following:

2sVG ρ= (2)

where ρ is the density of the material. Al-though density may not have been measured for a particular shaft, the density probably has a relatively low uncertainty. In principle, density can be computed from the shaft weight and geometry, but such a calculation would require NMI traceable measurements of the weight and shaft dimensions.

The outside diameter, Do, of the shaft may

be measured directly with a micrometer caliper or indirectly from a circumference measure-ment with a tape, and the inside diameter, Di, from the wall thickness, t, as follows:

tDD 2oi −= (3)

or

tCD 2/i −= π (4) where C is the shaft circumference as measured with a tape measure. The wall thickness is measured with an ultrasonic gage. For calibra-tion of the ultrasonic gage, gage blocks should be manufactured from the same-class material as the shaft with documented measurements of

428428

G, the dimensions, density, and their uncertain-ties.

The uncertainty in torque may be computed from Equations (1) from the law of propagation of uncertainty either by analytical methods or central finite differencing from JGCG (2008a) and ITTC (2014a). The result for the relative standard uncertainty is

22222

+

+

+

+

=

Lu

Ru

Ju

Guu

Qu LRJGQ

δδ

(3) The uncertainty in the polar moment of inertia for the shaft is

23ii

23oo )()()8/( DuDuu DDJ += π (4)

If the outside diameter is measured by a mi-crometer, the uncertainty of the inside diameter is

22oi 4 tDD uuu += (5)

If the outside diameter is computed from a measurement of the circumference, the uncer-tainty in the inside diameter is

22i 4)/( tCD uuu += π (6)

The standard uncertainty in G from the shear wave velocity measurement from Equa-tion (2) is

242)2( ρρ uVuVu sVssG += (7)

For conventional steel, the values for the shear modulus of elasticity are as follows:

• G = 8.0 x 106 Pa • ρ = 7800 kg/m3 • Vs = 3200 m/s

The estimated uncertainties for the various elements are listed in Table 3. A complete es-timate requires the dimensions of the torsion-meter. As a preliminary estimate, the uncer-tainty G is computed from Equation (7), where the uncertainties in the shear wave velocity and density are assumed as 1 % and 2 %, respec-tively. The uncertainties in G from the two elements are respectively, ±120 x 106 and ±164 x 106 Pa. The combined expanded uncertainty in G is then ±200 x 106 Pa or ±0.25 %. An ac-curate assessment of density uncertainty is then necessary. If G is the dominant term in the un-certainty estimate, the uncertainty in torque and power is approximately ±0.25 %. With the in-clusion of the other terms, ±0.50 % expanded uncertainty in power appears to be reasonable. Table 3. Uncertainty estimates for elements of

torque calculation.

Symbol Units Value

C mm 1.0

Di mm 0.025

Do mm 0.025

L mm 0.10

R mm 0.10

t mm 0.025

Vs m/s 2.4

ρ kg/m3 16 For a better assessment of the uncertainty

the following procedures should be followed:

• Calibration of the displacement sensor per the ITTC calibration procedure, ITTC (2014b).

429429

• Calibration of the ultrasonic device for

the shear velocity and thickness with gage blocks.

• Multiple measurements of the shaft out-side diameter, wall thickness and shear wave velocity. Measurements are rec-ommended at eight (8) equal increments or 45° increments around the shaft both forward and aft of the torsionmeter for a total of 16 measurements. If the outside diameter is computed from the circum-ference, the circumference should be measured forward and aft of the torsion-meter.

• Measurements of R and L should be pro-vided by the manufacturer with uncer-tainty estimates.

• Measurement of shaft speed by direct digital methods per ITTC (2014b).

4. CONCLUSIONS

The revision of the ITTC Quality Manual concerned 54 documents. Two existing proce-dures were deleted, 17 new proce-dures/Guidelines have been added, 35 existing procedures have been reviewed or updated, of which 3 have been postponed.

The cross checking of the Dictionary and

the Symbols and Terminology List with ISO standards has been completed and produced a number of new entries in the ITTC documents. A decision is still required by the Conference about the discrepancy in the definition of Skew between ISO and the ITTC definitions.

The Dictionary and the Symbol and Termi-

nology List have been updated and some mis-takes have been rectified.

The development of three new documents is

proposed to the Conference for the next ITTC period:

• a new procedure on full scale torque measurements,

• a guideline with number 7.5-02-01-02 and working title: “Guideline to Practical Implementation of Uncertainty Analysis”

• a procedure on the determination of a type A uncertainty estimate of a mean value from signal analysis

5. RECOMMENDATIONS TO THE CONFERENCE

The QSG recommends to the Full Conference to:

adopt the revised procedure 4.2.3-01-03 Work Instruction for formatting ITTC Recom-mended Procedures and Guidelines;

adopt the revised procedure 7.5-01-03-01

Uncertainty Analysis, Instrument Calibration; adopt the revised procedure 7.5-02-01-01

Guide to the Expression of Uncertainty in Ex-perimental Hydrodynamics;adopt the revised Symbols and Terminology List;

adopt the name of “ITTC Dictionary of

Hydromechanics” in place of “Dictionary of Ship Hydrodynamics”;

adopt the revised ITTC Dictionary of

Hydromechanics Version 2014; enhance the liaison with ISO with a view to

reconcile the differences in definitions between ISO standards and ITTC definitions as laid down in the abovementioned procedures

allow the Wiki tool to implement updates to

the Dictionary also between conferences.

430430

6. RECOMMENDATIONS FOR FU-

TURE WORK The following future work is recom-

mended: support the Technical Committees in their

work on Recommended Procedures. Supply the chairmen of the new committees at the begin-ning of the period with the MS Word versions of the relevant procedures and the template for the production of new procedures,

maintain the Manual of ITTC Recom-

mended Procedures and Guidelines. Co-ordinate the modification and re-editing of the existing procedures according to the comments made by ITTC member organizations at the Conference and by the Technical Committees,

observe the development or revision of ISO

Standards regarding Quality Control, update the ITTC Symbols and Terminology

List, update the ITTC Dictionary of Hydrome-

chanics, revise and update the existing ITTC Rec-

ommended Procedures according to the com-ments of Advisory Council, Technical Com-mittees and the Conference,

before the third AC Meeting, review and

edit new ITTC Recommended Procedures with regard to formal Quality System requirements including format and compliance of the sym-bols with the ITTC Symbols and Terminology List,

follow the implementation of the Bench-

mark data repository,

support the Technical Committees with guidance on development, revision and update of uncertainty analysis procedures,

observe ISO standards for uncertainty

analysis, in particular the uncertainty analysis terminology,

review developments in metrology theory and uncertainty analysis and issue appropriate Procedures,

continue to maintain the online Wiki tool

keeping it up to date and in line with the adopted documents of the ITTC,

include a new section of the Dictionary

dedicated to Offshore Engineering, as prepara-tion for an extension of ITTC procedures to this fast developing field,

include into the Dictionary a section dealing

with planning craft and a section on pods, include into the Dictionary a section on

pods, develop a guideline with number 7.5-02-01-

02 and working title: “Guideline to Practical Implementation of Uncertainty Analysis”. This guideline should assist committee members (primarily beginners but also experienced in the field of UA) in making an adequate uncer-tainty analysis in both pre-test and post-test situations. It should provide an overview of all the steps to be taken in an uncertainty analysis and refers to existing procedures such as 7.5-02-01-01 on basic techniques and 7.5-01-03-01 on calibration,

develop a procedure on the determination of

a type A uncertainty estimate of a mean value from signal analysis, based on Brouwer et al. (2013). This analysis provides an uncertainty estimate in cases where instead of multiple

431431

repeat or reproduction measurements, only a single time series is available,

surveying the extent and breadth of uptake

of uncertainty analysis techniques and proce-dures by the hydrodynamic testing community,

develop a new procedure on torsionmeters

for ship trials.

7. REFERENCES Brouwer, J., Tukker, J., van Rijsbergen, M.,

2013, “Uncertainty Analysis of Finite Length Measurement Signals”, 3rd Interna-tional Conference on Advanced Model Measurement Technologies for the Mari-time Industry, Gdansk, Poland.

Cusanelli, D. S. and Karafiath, G., 2012, “Hy-

drodynamic Energy Saving Enhancements for DDT 51-Class Ships”, Naval Engineers Journal, American Society of Naval Engi-neers, No. 124-2, pp. 123-138.

Insel, M., 2008, “Uncertainty in the analysis of

speed and powering trials”, Ocean Engi-neering, Elsevier, Vol. 35, pp. 1183-1193.

ISO/IEC 2013 ISO/IEC, Directives, Part 1.

“Consolidated ISO Supplement Procedures specific to ISO, Fourth Edition, 2013. An-nex SL (normative). Proposals for man-agement system standards.

ITTC, 2002, “The Specialist Committee on

Speed and Powering Trials, Final Report and Recommendations to the 23rd ITTC”, Proceedings of the 23rd International Tow-ing Tank Conference, Vol. II, p. 355.

ITTC, 2014a, “Guide to the Expression of Un-

certain in Experimental Hydrodynamics”, ITTC Procedure 7.5-02-02-02, Revision 02,

2th International Towing Tank Conference, draft.

ITTC, 2014b, “Uncertainty Analysis Instru-

ment Calibration”, ITTC Procedure 7.5-01-03-01, Revision 01, 27th International Tow-ing Tank Conference, draft.

JCGM, 2008a, “Evaluation of measurement data – Guide to the expression of uncer-tainty in measurement,” JCGM 100:2008 GUM 1995 with minor corrections, Joint Committee for Guides in Metrology, Bu-reau International des Poids Mesures (BIPM), Sèvres, France.

JCGM, 2008b, “International vocabulary of

metrology – Basic and general concepts and associated terms (VIM)” JCGM 200:2008 VIM, Joint Committee for Guides in Me-trology, Bureau International des Poids Me-sures (BIPM), Sèvres, France.

JCGM, 2008c, “Evaluation of measurement

data — Supplement 1 to the ‘Guide to the expression of uncertainty in measurement’ — Propagation of distributions using a Monte Carlo method”, JCGM 101:2008, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France.

JCGM, 2009, “Evaluation of measurement data

— An introduction to the ‘Guide to the ex-pression of uncertainty in measurement’ and related documents”, JCGM 104:2009, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France.

432432

JCGM, 2011, “Evaluation of measurement data

– Supplement 2 to the ‘Guide to the expres-sion of uncertainty in measurement’ — Ex-tension to any number of output quantities”, JCGM 102:2009, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France.

JCGM, 2012, “Evaluation of measurement data

— The role of measurement uncertainty in conformity assessment”, JCGM 106:2012, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France.

JCGM, 20xxa, “Evaluation of measurement

data — Supplement 3 to the ‘Guide to the expression of uncertainty in measurement’ — Modeling”, JCGM 20xx:103. Joint Committee for Guides in Metrology, Bu-reau International des Poids Mesures (BIPM), Sèvres, France, in preparation.

JCGM, 20xxb, “Evaluation of measurement

data — Concepts and basic principles”, JCGM 20xx:105, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France, in preparation.

JCGM, 20xxc, Evaluation of measurement data

— Applications of the least-squares meth-od, JCGM 20xx:107, Joint Committee for Guides in Metrology, Bureau International des Poids Mesures (BIPM), Sèvres, France, in preparation.

Kline, S. J., 1985, “The Purpose of Uncertainty

Analysis”, Journal of Fluids Engineering, American Society of Mechanical Engineers, Vol. 107, No. 2, pp. 153-160.

433433

8. ANNEX A – REPORT ON THE AC-

TIVITY ON THE WIKI WEBSITE.

As agreed at 26th ITTC in 2011, QSG main-tained and further developed the ITTC Wiki page dedicated to the Dictionary of Hydrome-chanics at the following web address:

http://www.ittcwiki.org/doku.php/start hosted at CNR-INSEAN, Roma, Italy.

ITTC Wiki Home page

The structure of the ITTC Wiki dictionary is:

General Vessel Geometry and Stability Resistance Propeller (including propeller geometry) Cavitation Seakeeping Manoeuvrability Performance (in the context of speed and power) Alphabetic dictionary and fully reflects the structure of the ITTC Dic-tionary as agreed at 26th ITTC contained in the pdf version.

As for the pdf version it is also possible to browse the Alphabetic version.

Table A1 lists some of the modifications

implemented on the online tool after the 26th ITTC to match the changes approved by the Conference.

As can be seen looking at the table, quite relevant effort has been dedicated to figures and schemes to be included in the web pages. This is an area that still needs attention and further work in order to achieve a uniform ac-ceptable level in terms of quality of the images, drawings, table, sketches that are present in the Dictionary. Several schemes have been re-drawn from scratch.

An example is shown below:

If the proposal of the QSG on the Diction-

ary is accepted by the 27th ITTC, the web pages

434434

will be aligned again with the new proposed structure. Then, the occasional further im-provement can be implemented.

A link to the ITTC Wiki pages has been

added on the ITTC web site to increase the visibility.

An analysis of the visitors and their behav-

iour when visiting the ITTC Wiki has been carried out and in the following most relevant data collected between September 2011 (26th ITTC) and April 2014 are showed below.

SESSIONS 27.141 USERS 20.269

PAGE VIEWS 64.129

BOUNCE RATE 61.85%

More than 27.000 Sessions and more than

20.000 users have been registered.

Trend of the Sessions between 26th ITTC and April 2014

Trend of the users Analytics between 26th ITTC and April 2014

The Trends above show slow constant in-

crease of the number of visitors.

New Visitors vs. Returning Visitors

Looking into how visits are geographically

distributed is also interesting.

Visits by Country

435435

Visits by Continent

Visits by Citiy

Visits:First 10 Countries

Visits by Continent – figures

As can be seen by the previous analysis

Europe has registered more than 12.000 visitors followed by Asia and the Americas. This result is not coming as a surprise given the large number of ITTC members in Europe and Asia. For the Americas the two countries that have shown significant interest are US and Brazil.

Singapore with more than 700 visits is

ranked 9th. This suggests that not only re-searchers and technicians belonging to ITTC members’ organization have been visiting the ITTC Wiki pages but also other type of profes-sionals with a maritime interest. A big part of those are (with an educated guess) students and maritime universities in general. This fact is an extremely positive signal that should not be underestimated given the constant shortage of qualified human capital experienced by several ITTC organizations.

For improved visibility, further dissemina-

tion actions could be imagined such as ITTC members to add a link to the ITTC Wiki to the web pages of their organizations.

436436

Table A1: Modifications approved by the 26th ITTC Conference implemented on the Wiki tool

2014/03/14 11:34 structured_dictionary:propeller – [Blade position] ubuwiki 2014/02/26 18:42 structured_dictionary:shipgeometry – [Beam] ubuwiki 2014/02/26 18:35 structured_dictionary:general – [Axes co-ordinate] ubuwiki 2014/01/10 13:24 propeller:structureddictionary2008_img_25.jpg – created ubuwiki 2014/01/10 10:45 structured_dictionary:seakeeping – [Added mass [M]] ubuwiki 2014/01/10 10:05 manoeuvrability:manoeuvrability7_1_ok.png – created ubuwiki 2014/01/10 10:00 structured_dictionary:performance – [Admiralty coefficient] ubuwiki 2014/01/10 09:54 start – [ITTC Dictionary] ubuwiki 2013/10/09 15:18 document – [reference] ubuwiki 2013/10/09 15:05 structured_symbols_list2011.pdf – created ubuwiki 2013/10/09 15:05 ittc_alphabet_dictionary_2011.pdf – created ubuwiki 2013/10/09 15:05 ittc_structured_dictionary_2011.pdf – created ubuwiki 2013/10/09 15:05 alphabetic_ittc_symbols_list2011.pdf – created ubuwiki 2013/10/09 15:04 dw-backup-20120621-185027.tar.bz2 – removed ubuwiki 2013/10/09 15:04 dictionary2011.pdf – removed ubuwiki 2013/10/09 15:04 alphabetdictionary2011.pdf – removed ubuwiki 2013/10/09 15:04 dw-backup-20130724-100631.tar.bz2 – removed ubuwiki 2013/10/09 15:03 dw-backup-20120621-185044.tar.bz2 – removed ubuwiki 2013/10/09 15:03 dw-backup-20121016-141033.tar.bz2 – removed ubuwiki 2013/10/09 15:03 dw-backup-20121016-145543.tar.bz2 – removed ubuwiki 2013/10/09 15:03 dw-backup-20120621-181458.tar.bz2 – removed ubuwiki 2013/10/09 15:03 dw-backup-20120611-083203.tar.bz2 – removed ubuwiki 2013/04/08 14:34 sidebar – ubuwiki 2013/02/04 12:42 trysyntax:trysyntax – [Graphic notes] ubuwiki 2012/09/03 15:52 proposal:dictionary:planing_resistance_and_trim_01.png – created marco.ferrando-prof 2012/05/06 13:06 proposal:dictionary:speed_en.png – created marco.ferrando-prof 2012/05/05 20:59 propeller:set_back.png – created marco.ferrando-prof 2012/05/05 20:44 structured_dictionary:set_back.png – created marco.ferrando-prof 2012/05/05 19:24 propeller:wing.png – created marco.ferrando-prof 2012/05/05 18:07 propeller:skew_3_3d.png – created marco.ferrando-prof 2012/05/05 18:07 propeller:propeller_lines.png – created marco.ferrando-prof 2012/05/05 17:43 propeller:rake_02.png – created marco.ferrando-prof 2012/05/05 17:43 structured_dictionary:rake_02.png – removed marco.ferrando-prof 2012/05/05 10:00 propeller:pitch_00.png – created marco.ferrando-prof 2012/05/05 00:16 propeller:sezioni_2.png – created marco.ferrando-prof 2012/04/19 14:09 flag_uk.jpg – removed ubuwiki 2012/03/21 18:12 flag-italian.jpg – created ubuwiki 2012/03/13 12:14 ittcrepository:pr-00_2.doc – dario 2012/03/13 12:08 ittcrepository:vort2.avi – dario 2012/03/13 11:16 ittcrepository:3dbis.flv – dario 2012/03/13 11:16 ittcrepository:fs_sezu.avi – dario 2012/03/13 10:31 ittcrepository:little_x_manual_4_5.pdf – dario

437437

2011/11/17 15:03 100_1835.jpg – ubuwiki 2011/05/30 12:10 efcampana.png – ubuwiki 2011/04/19 12:39 propeller:figure30.jpg – dario 2011/04/19 12:23 propeller:figure30.png – dario 2011/04/19 12:15 propeller:figure36.png – dario 2011/04/19 12:14 propeller:figura4-3.png – dario 2011/04/19 12:10 propeller:figure32.png – dario 2011/04/19 12:09 propeller:figure31.png – dario 2011/04/19 12:09 propeller:figure31.gif – dario 2011/04/19 12:04 propeller:figura31.png – dario 2011/04/19 11:52 structured_dictionary:figure34.png – dario 2011/04/19 11:52 structured_dictionary:figure30.png – dario 2011/04/19 11:52 propeller:figure34.png – dario 2011/04/19 10:49 propeller:figure13.png – dario 2011/04/13 09:26 structured_dictionary:figure29.png – ubuwiki 2011/04/13 09:26 propeller:figure26.png – ubuwiki 2011/04/13 09:10 propeller:figure27.png – ubuwiki 2011/04/13 09:08 propeller:figure23.png – ubuwiki 2011/04/13 08:53 propeller:figure22.png – ubuwiki 2011/04/05 23:29 propeller:hub_01a_2d.png – marco.ferrando-prof 2011/04/05 23:29 propeller:hub_01_2d.png – marco.ferrando-prof

438438

Appendix 1

Committees of the 27th ITTC EXECUTIVE COMMITTEE Stig Sand, FORCE Technology, Denmark, Chairman (until 2013) Peter K. Sørensen, FORCE Technology, Denmark, Chairman (from 2013) Susanne Abrahamsson, SSPA, Sweden (Northern Europe Representative) Jürgen Friesch, HSVA, Germany (Central Europe Representative) Daniele Ranocchia, INSEAN, Italy (Southern Europe Representative) F. Mary Williams, NRC, Canada (Americas Representative until 2013) Antonio Fernandes, LabOceano, Brazil (Americas representative from 2013) Masashi Kashiwagi, Osaka University, Japan (Pacific Islands Representative) Suak Ho Van, KRISO, Korea (East Asia Representative) Gerhard Strasser, Vienna Model Basin, Austria, AC Chairman, ex-officio Aage Damsgaard, FORCE Technology, Denmark, ITTC and EC Secretary, ex-officio ADVISORY COUNCIL Neil Bose, AMC, Australia Gerhard Strasser, Vienna Model Basin, Austria Marcelo Neves, LabOceano, Brazil Kostadin Yossifov, BSHC, Bulgaria (until 2013) Rumen Kishev, BSHC, Bulgaria (from 2013) F. Mary Williams, NRC, Canada (until 2013) James Millan, NRC, Canada (from 2013) Baoshan Wu, CSSRC, China Jianming Yang, SJTU, China Marta Pedisic Buca, Brodarski Institute, Croatia Stig Sand, FORCE Technology, Denmark (until 2013) Christian Schack, FORCE Technology, Denmark (from 2013) Seppo Kivimaa, VTT, Finland Guillaume de Garidel, DGA Hydrodynamics, France (until 2014) Roland Joannic, DGA Hydrodynamics, Franse (from 2014) Pierre Ferrant, Ecole Centrale de Nantes, France Jürgen Friesch, HSVA, Germany Manfred Mehmel, SVA Potsdam, Germany (until 2014) Christian Ernst-Georg Masilge, SVA Potsdam, Germany (from 2014) Cornel Thill, DST, Germany (until 2014) Bettar Ould el Moctar, DST, Germany (from 2014)

439439

Giovanni Caprino, CETENA, Italy Daniele Ranocchia, INSEAN, Italy Takuya Omori, JMUC, Japan Kazuyuki Yamakita, Meguro Model Basin, Japan Chiharu Kawakita, MHI, Japan Noriyoki Sasaki, NMRI, Japan (until 2013) Shotari Uto, NMRI, Japan (from 2013) Hajime Yamaguchi, University of Tokyo, Japan Hyun Soo Shin, HHI, Korea (until 2014) Young Sik Jang, HHI, Korea (from 2014) Suak Ho Van, MOERI, Korea Ho-Hwan Chun, Pusan National University, Korea Bas Buchner, MARIN, The Netherlands Kourosh Koushan, MARINTEK, Norway Leszek Wilczynski, CTO, Poland Alexander Pustoshny, Krylov, Russia Luis Palao Lechuga, CEHIPAR, Spain (until 2014) Emilio Fajardo, CEHIPAR, Spain (from 2014) Susanne Abrahamsson, SSPA, Sweden Paul Crossland, QinetiQ, UK Mehmet Atlar, Newcastle University, UK Jon Etxegoien, NSWCCD, USA Robert F. Beck, University of Michigan, USA TECHNICAL COMMITTEES Resistance Committee Prof. Stephen Turnock, University of Southampton (Chair) Dr. Hisao Tanaka, Universal Shipbuilding Dr. Jin Kim, MOERI Prof. Baoshan Wu, CSSRC Thomas Fu, NSWCCD Bertrand Alessandrini, ECN Takinaci Ali Can, ITU T. Mikkola, Aalto Uni

440440

Propulsion Committee Didier Fréchou, DGA (Chair) Dr. Takuya Ohmori, JMUC Prof. Moon Chan Kim, Pusan National Uni. Dr. Chenjun Yang, SJTU Steve Ceccio, Univ. of Michigan Emin Korkut, ITU Rainer Grabert, SVA Potsdam Tom Dinham-Peren, BMT V. Borusevich, Krylov Manoeuvring Committee Frans Quadvlieg, MARIN (Chair) Prof. Yoshitaka Furukawa, Kyushu University Dr. Jonathan Duffy, AMC Dr. Sun Young Kim, MOERI Prof. Xiaofei Mao, Wuhan UST Eduardo Tannuri, U Sao Paulo Pierre Emanuel Guillerm, ECN Dr. G. Delefortrie, Uni Ghent & Flanders C. Simonsen, FORCE Seakeeping Committee Prof. Young Hwan Kim, Seoul National University (Chair) Dr. Katsuji Tanizawa, NMRI Dr. Giles Thomas, AMC Prof. Quanming Miao, CSSRC (resigned 2012) Greg Hermanski, NRC David Hayden, NSWCCD Pepijn de Jong, Delft Uni. Dr. Dominic Hudson, Uni Southampton D. Fathi, Marintek Dr. Chengsheng WU, CSSRC (from 2012) Ocean Engineering Committee Wei Qiu, Memorial University (Chair) Dr. Takashi Mikami, Mitsui Akishima Prof. Xuefeng Wang, SJTU Dr. Dong Yeon Lee, Samsung HI Sergio Sphaier, LabOceano Jean-Marc Rousset, ECN Prof. Longbin Tao, Newcastle Uni. H. Lie, Marintek V. Magarovski, Krylov

441441

Stability in Waves Committee Arthur Reed, NSWCCD (Chair) Prof. Toru Katayama, Osaka Prefecture Uni. Dr. Gyung Jung Lee, MOERI Prof. Wenyang Duan, Harbin Eng. Uni. Paola Gualeni, Unige Frans van Walree, MARIN Andy Peters, Qinetiq Specialist Committee on CFD in Marine Hydrodynamics Dr. Takanori Hino, Yokohama National Uni. (Chair) Prof. Shin Hyung Rhee, Seoul National Uni. Dr. Decheng Wan, SJTU Sung-Eun Kim, NSWCCD Pablo M. Carrica, Univ. of Iowa Riccardo Broglia, INSEAN Peter Bull, Qinetiq Dr. Ignazio Maria Viola, Newcastle Uni. D-Q. Li, SSPA I. Saisto, VTT Specialist Committee on Detailed Flow Measurement Techniques Paisan Atsavapranee, NSWCCD (Chair) Dr. Shigeki Nagaya, IHI Dr. Feng Zhao, CSSRC Prof. In Won Lee, Pusan National Uni. Mario Felli, INSEAN C. Muthanna, Marintek Specialist Committee on Performance of Ships in Service A. Minchev, FORCE (Chair) Dr. Masaru Tsujimoto, NMRI Mr. Michio Takai, Sumitomo HIMEC Dr. Jinbao Wang, MARIC Mr. Heungwon Seo, Hyundai HI Angelo Olivieri, INSEAN G. Grigoropoulos, NTUA Dr. Uwe Hollenbach, HSVA Henk van der Boom, MARIN S. Werner, SSPA W. Gorski, CTO

442442

Specialist Committee on Hydrodynamic Noise Elena Ciappi, INSEAN (Chair) Mr. Chiharu Kawakita, Mitsubishi Heavy Ind. Prof. Denghai Tang, CSSRC Dr. Gil Hwan Choi, Hyundai HI Dr. Theodore Farabee, NSWCCD Herbert Bretschneider, HSVA Johan Bosschers, MARIN Specialist Committee on Hydrodynamic Modelling of Renewable Energy Devices Dr. Sandy Day, Uni Glasgow/Strathclyde (Chair) Prof. Motohiko Murai, Yokohama National Uni Dr. Irene Penesis, AMC Prof. Hyunkyung Shin, Uni. of Ulsan Prof. Yanping He, SJTU Arnold Fontaine, Penn State Univ. Aurélien Babarit, ECN Francesco Salvatore, INSEAN M. Kraskowski, CTO Specialist Committee on Ice Peter Jochmann, HSVA (Chair) Prof. Akihisa Konno, Kogakuin University Prof. Qianjing Yue, Dalian UT Michael Lau, NRC Dr. Rod Sampson, Newcastle Uni. (until 2013) K. Sazonov, Krylov T. Leiviskä, Aker Arctic J. Römeling, FORCE (until 2013) R. von Bock und Polach, Aalto Uni J. Huffmeier, SSPA (until 2013) V. Westerberg, SSPA (from 2013) Quality Systems Group Marco Ferrando, UNIGE (Chair) Prof. Eiichi Kobayashi, Kobe Univ. Bruce Johnson, US Naval Academy Ahmed Derradji, NRC (until 2013) Joel Sena Sales Jr., LabOceano Lanfranco Benedetti, INSEAN Luis Pérez Rojas, ETSIN M. van Rijsbergen, MARIN Michael Woodward, Newcastle Univ. Michael G. Morabito, US Naval Academy (from 2012) Joel Park, NSWCCD (from 2013)

Updated 2014-05-25

443443

Appendix 2

Tasks and Structure of the 27th ITTC Technical Committees and Groups

1. STRUCTURE OF TECHNICAL

COMMITTEES The structure of the technical committees

has changed slightly from the 26th ITTC and now includes six General Committees, six Specialist Committees and one Group.

2. TERMS OF REFERENCE FOR THE GENERAL AND SPECIALIST TECHNICAL COMMITTEES AND GROUPS

2.1. General Committees

Each General Committee will be

responsible for a general subject area. It will review the state-of-the-art, identify the need for research and development, and carry out longer terms studies with broad impact.

An important part of the work of the General Committees will be to establish Procedures and Guidelines to help the ITTC Member Organizations maintain their institutional credibility with regard to quality assurance of products and services such as predictions and evaluations, and quality assurance of designs. The General Committees will develop detailed plans in accordance with Conference Recommendations and their work should be directed towards the techniques and understanding of physical and numerical

modelling as a means of predicting full-scale behaviour. While maintaining an awareness of progress, fundamental theoretical studies and fundamental aspects of numerical fluid computation should be covered by other fora. Procedures and Guidelines shall contain only techniques which are applicable in commercial practice.

Each General Committee will submit a report on the results of its work to the Full Conference. The conclusions and the recommendations of the General Committee should be structured as follows:

1. General technical conclusions 2. Recommendations to the Full Conference,

which require actions such as, e.g., adopting ITTC procedures.

3. Proposals for future work of the General Committee and identification of tasks, which may be appropriate for Specialist Committees. These proposals shall be submitted to the Advisory Council which will compile these proposals and present them to the Full Conference.

2.2. Specialist Committees The ITTC Advisory Council will propose

Specialist Committees. Each Specialist Committee will be responsible for studying a specific technical problem. The Specialist

444444

Committees will be appointed for a limited duration. It is expected that they will complete their tasks within maximum two ITTC periods (6 years). They shall interact closely with the appropriate General Committees. The tasks of a Specialist Committee can include establishing Procedures and/or Guidelines. Procedures and Guidelines shall contain only techniques which are applicable in commercial practice.

Each Specialist Committee will present a

final report on the results of its work to the Full Conference and interim reports on progress if the duration of the committee spans more than one Conference. The conclusions and the recommendations of the Specialist Committee should be structured as follows:

1. General technical conclusions 2. Recommendations to the Full Conference,

which require actions such as, e.g., adopting ITTC procedures.

3. Proposals for future work of and identification of tasks, which may be appropriate for Specialist Committees. These proposals shall be submitted to the Advisory Council which will compile these proposals and present them to the Full Conference.

2.3. Groups

Groups may be established from time to

time by the Executive Committee to carry out specific tasks for the Conference, which are not technical issues. Membership of a Group should not exceed three consecutive terms of three years, but the Executive Committee may make exceptions. Also, normally, Groups shall have fewer members than the Technical Committees. Such Groups shall be dissolved upon completion of their respective tasks.

3. MECHANISM FOR IDENTIFYING NEW SPECIALIST TECHNICAL COMMITTEES

As part of their Terms of Reference, the

General Committees shall consider the need for new tasks and include appropriate proposals in their technical reports. If the Advisory Council identifies a need for a new Specialist Committee when it reviews the draft recommendations of the General Committees, the Council will prepare and agree a statement of the technical aims and objectives for the work of the Specialist Committee.

Independently of the proposals of the

General Committees, the Advisory Council will keep under continuous review the requirement for Specialist Committees.

When the Advisory Council has agreed the

need for a new Specialist Committee, the draft statement of technical aims and objectives will be presented to the Executive Committee for endorsement. If the Executive Committee approves the formation of a new Specialist Committee, it will present the proposal to the Full Conference for approval.

4. PROPOSED STRUCTURE OF THE TECHNICAL COMMITTEES AND GROUP FOR 27THITTC

4.1. General Committees

• Resistance • Propulsion • Manoeuvring • Seakeeping • Ocean Engineering • Stability in Waves

445445

4.2. Specialist Committees

• CFD in Marine Hydrodynamics • Detailed Flow Measurement

Techniques • Performance of Ships in Service • Hydrodynamic Noise • Hydrodynamic Modelling of

Marine Renewable Devices • Ice

4.3. Groups

• Quality Systems Group

5. TASKS OF THE TECHNICAL COMMITTEES AND GROUPS OF THE 27TH ITTC

5.1. General Terms of Reference

1. All committees shall observe the Terms of Reference and general obligations. The committees are expected to perform all the tasks defined in this document. However, should a committee be unable to do this, it shall consult the Advisory Committee with regard to reduction of the work.

2. Each technical committee shall consider any unfinished items from previous committees and report to the Advisory Council by 1st December 2011 in order to clarify whether these items should be included in the Terms of Reference.

3. All committees shall identify areas of mutual interest with other committees and the concerned committees shall establish active co-operation in these areas.

4. In their work, the committees shall

follow the guidelines given in ITTC Recommended Procedure 1.0-03, General Guideline for the Activities of Technical Committees, Liaison with the Executive Committee and Advisory Council.

5. Procedures and Guidelines must be in the format defined in the ITTC Recommended Procedure 4.2.3-01-03, Work Instruction for Formatting ITTC Recommended Procedures, and they will be included in the ITTC Quality Manual. Symbols and terminology must be in accordance with those used in the current version of the ITTC Symbols and Terminology List. If necessary, new symbols should be proposed in collaboration with the Quality Systems Group.

6. All new procedures for uncertainty analysis in experiments shall follow the ISO (1995) ‘Guide to the Expression of Uncertainty in Measurements’ (also known as ISO-GUM). It is not required to update existing procedures on uncertainty analysis to follow this standard. If a procedure for uncertainty analysis is for other reasons updated, it shall follow the ISO standard.

7. Committees that have a task to review ITTC Recommended Procedures shall identify and report any changes proposed in their first annual report to the Advisory Council. The changes approved by the Advisory Council should be implemented in the second year and the draft revised procedure submitted to the Advisory Council for comment.

446446

8. Committees that have a task to write new procedures or guidelines shall submit an outline of these with their first annual report to the Advisory Council. The outline shall be reviewed by the Advisory Council and comments made to the committees. The draft new procedure or guideline shall be prepared during the second year and submitted to the Advisory Council for review.

9. All new and revised procedures shall, as far as feasible, include procedure for uncertainty analysis.

10. New and revised draft procedures shall subsequently be updated, incorporating the comments made by the Advisory Council, and in February of the third year be submitted to the Quality Systems Group for formal check and to the Advisory Council for final review and approval.

11. Committee reports to the Conference should be structured in line with the terms of reference of the committee and in accordance with Recommended Procedure 4.2.3-01-02, Guidelines for Preparation of Technical Committee and Working Group Reports

5.2. Terms of Reference for the General Committees

Resistance Committee

1. Update the state-of-the-art for

predicting the resistance of different ship concepts emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. The potential impact of new technological developments on the ITTC,

b. New experimental techniques and extrapolation methods,

c. New benchmark data, d. The practical applications of

computational methods to resistance predictions and scaling,

e. The need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

2. Review ITTC Recommended

Procedures relevant to resistance and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

c. Implement updated uncertainty analysis spreadsheet for resistance test.

3. Continue the analysis of the ITTC

worldwide series for identifying facility biases.

4. Review definitions of surface roughness and develop a guideline for its measurement.

5. Review results from tests that correlate skin friction with surface roughness.

6. Review trends and new developments in experimental techniques on unsteady flows and dynamic free surface phenomena.

7. Review new developments on model

manufacturing devices and methods.

447447

8. Review the development and evaluate improvements in design methods and the capabilities of numerical optimization applications, such as Simulation Based Design environments, with special emphasis on design of new ship concepts, geometry manipulation and parameterization, surrogate models and variable fidelity applications. (The fundamental assumption that an optimal hull shape is one that minimizes the calm water resistance may no longer be appropriate given the developments in CFD that give the designer the ability to make assessment of both wave and viscous effects for added resistance in waves as well as the interaction between hull-propulsor and appendages.)

Propulsion Committee

1. Update the state-of-the-art for predicting for propulsion systems emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. The potential impact of new technological developments on the ITTC including new types of propulsors, azimuthing thrusters and propulsors with flexible blades,

b. New experimental techniques and extrapolation methods,

c. New benchmark data, d. The practical applications of

computational methods to the propulsion systems predictions and scaling,

e. New developments of experimental and CFD methods applicable to the prediction of cavitation,

f. The need for R&D for improving methods of model experiments,

numerical modelling and full-scale measurements.

g. Monitoring the developments regarding high-speed marine vehicles

2. Review ITTC Recommended

Procedures relevant to propulsion and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

3. Liaise with the Specialist Committee on

Performance of Ships in Service, especially regarding power prediction and consequences of EEDI.

4. Assess where CFD results can be introduced to support experimental model testing by monitoring status of CFD to perform full scale powering, resistance, cavitation and wake simulations and their correlation with full scale data. Identify the needs for hybrid procedures combining experimental and numerical methods.

5. Prepare a state-of-the-art review of modelling and scaling unconventional propulsion and wake improving devices.

6. Examine methods of target wake simulation, e.g. “smart” dummy approach.

7. Examine wake fraction scaling for twin screw ships and show the consequences on existing procedures.

8. Examine the possibilities of CFD-methods regarding scaling of

448448

conventional and unconventional propeller open water data. Initiate a comparative CFD-calculation project.

9. Develop guidelines for hybrid propulsor testing.

10. Continue with the monitoring of existing full scale data for podded propulsion. If there is available data, refine the existing Procedure.

Manoeuvring Committee

1. Update the state-of-the-art for predicting the manoeuvring behaviour of ships emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. the potential impact of new technological developments on the ITTC

b. developments in manoeuvring and course keeping in waves.

c. new experiment techniques and extrapolation methods,

d. new benchmark data e. the practical applications of

computational methods to manoeuvring predictions and scaling.

f. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

g. the effects of free surface, roll, sinkage, and trim in numerical simulation of manoeuvring.

2. Review ITTC Recommended Procedures relevant to manoeuvring and

a. Identify any requirements for changes in the light of current

practice and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

3. Complete the work on the Procedure

7.5-02-06-04, Uncertainty Analysis; Forces and Moment, Example for Planar Motion Mechanism Test, based on ISO approach. The present procedure 7.5-02-06-04 and the subsection on uncertainty analysis in the Procedure 7.5-02-06-02, Captive Model Test Procedure, prepared by the 23rd ITTC are based on the ASME approach. In view of the work already carried out for the procedure 7.5-02-06-04, consider to keep the elaborated ASME example as one of the Appendices to the to-be-renewed 7.5-02-06-04.

4. Based on results of the SIMMAN

workshop held in 2008 and its next edition, continue the already initiated work to generate a guideline on Verification and Validation of RANS tools in the prediction of manoeuvring capabilities. Liaise with the QSG with respect to definitions of Verification and Validation.

5. Restricted waters:

a. Produce a guideline for experimental methods.

b. Complete the initiated one for numerical methods which may serve as a basis for recommended procedures for manoeuvring in restricted waters.

6. Free running model tests:

a. Update the procedure 7.5-02-06-01, Free Running Model Test

449449

Procedure, in particular to include objective statements on the initial conditions of free manoeuvring model tests.

b. Elaborate the already initiated procedure on uncertainty analysis for free running manoeuvring model tests, including an example.

7. Scale effects in manoeuvring:

a. Report on knowledge and collect, analyse and summarize data on scale effects for manoeuvring predictions.

8. Review developments in methods and

draft a validation procedure of combined manoeuvring and seakeeping with respect to simulation. Liaise with the Seakeeping Committee and the Stability in Waves Committee.

9. Support the organisation of a second

SIMMAN workshop.

10. Manoeuvring criteria and relations to IMO:

a. Report on manoeuvring criteria for ships not directly covered by IMO like POD and waterjet driven vessels, naval ships, inland ships, HSMV, etc.

b. Study possible criteria for manoeuvring at low speed and in shallow waters and if warranted communicate findings to IMO.

Seakeeping Committee

Note: The Seakeeping Committee is primarily concerned with the behaviour of ships underway in waves. The Ocean Engineering Committee covers moored and dynamically positioned ships. The modelling and simulation

of waves, wind and current is the primary responsibility of the Ocean Engineering Committee, with the cooperation of the Seakeeping Committee and the Stability in Waves Committee.

1. Update the state-of-the-art for

predicting the behaviour of ships in waves emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. the potential impact of new technological developments on the ITTC

b. new experiment techniques and extrapolation methods,

c. new benchmark data d. the practical applications of

computational methods to sea-keeping predictions and scaling.

e. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

2. Review ITTC Recommended

Procedures relevant to seakeeping and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

c. Introduce a definition of slamming.

3. Liaise with ISSC, the Ocean

Engineering Committee, The Stability in Waves Committee and the Specialist Committee on Performance of Ships in Service.

4. Update existing ITTC Recommended Procedure 7.5-02-07-02.5, Verification

450450

and Validation of Linear and Weakly Non-Linear Seakeeping Codes, to reflect the outcomes of the Verification and Validation workshop held in 2010.

5. Investigate methodology for

Verification and Validation of fully non-linear seakeeping viscous flow codes.

6. Develop a guideline for the verification

and outline further developments required for validation of hydroelastic seakeeping codes.

7. Jointly organize and participate in the

joint ISSC/ITTC workshop on uncertainty in measurement and prediction of wave loads and responses.

8. Establish a numerical and experimental

process for estimating fw, in the EEDI calculation. Liaise with the Specialist Committee on Performance In Service.

9. Develop a unified method for sloshing

experiments drawing on the methods developed by the classification societies. Identify benchmark data for sloshing in LNG carriers.

10. Review and update the Procedure 7.5-02-05-04, Seakeeping Tests, for High Speed Marine Vehicles.

Ocean Engineering Committee

Note: The Ocean Engineering committee covers moored and dynamically positioned ships and floating structures. The modelling and simulation of waves, wind and current is the primary responsibility of the Ocean Engineering Committee, with the cooperation of the Seakeeping Committee and the Stability in Waves Committee.

1. Update the state-of-the-art for predicting the behaviour of bottom founded or stationary floating structures including moored and dynamically positioned ships emphasising developments since the 2011 ITTC Conference. The committee report should include sections on:

a. the potential impact of new technological developments on the ITTC.

b. new experimental techniques, extrapolation methods,

c. new benchmark data, d. the practical applications of

computational methods to prediction and scaling.

e. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

2. Review ITTC Recommended

Procedures relevant to ocean engineering and

a. Identify any requirements for changes in the light of current practice, and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

3. Complete the VIV and VIM guideline

and benchmark study initiated by the Specialist Committee in Vortex Induced Vibrations of the 26th ITTC. The report on the benchmark test shall include clear definition of all test parameters.

4. Complete and report on the wave run-up benchmark study for a single cylinder.

451451

5. Carry out a wave run-up benchmark study for cases of four columns using the experimental data from Marintek.

6. Investigate and report on thruster-thruster interaction, ventilation and their scaling for DP systems.

7. Investigate and report on physical and numerical modeling of vessels in side-by-side operations with an emphasis on wave elevation in the gap.

8. Investigate and report on motions of large ships and floating structures in shallow water.

9. Jointly organize and participate in the joint ISSC/ITTC workshop on uncertainty in measurement and prediction of wave loads and responses.

Stability in Waves Committee

Note: The Stability in Waves Committee covers the stability of intact and damaged ships in waves. The modelling and simulation of waves, wind and current is the primary responsibility of the Ocean Engineering Committee, with the cooperation of the Seakeeping Committee and the Stability in Waves Committee.

1. Update the state-of-the-art for

predicting the stability in waves, emphasizing developments since the 2011 ITTC conference. The committee report should include sections on:

a. Definition of loss and survival of the ship;

b. The amount of detail required for modeling the internal geometry of the ship;

c. Leak and collapse pressures for water tight doors and bulkheads; and

d. Modeling of extreme wave conditions.

2. Review ITTC Recommended Procedures relevant to stability and

a. Identify any requirements for changes in the light of current practice, and, if approved by the Advisory Council, update them.

b. Identify the need for new procedures and outline the purpose and content of these.

3. Investigate uncertainty analysis for intact and damaged model tests to complement current procedures.

4. Investigate the criteria for modeling wave spectra in the determination of dynamic instability of intact vessels, i.e. wave steepness, non-linearity, frequency contents of the spectrum, statistical distribution of wave and crest height and spatial behaviour of the waves.

5. Develop better understanding of uncertainties associated with the results from experiments and simulations of extreme motions of intact vessels in realistic irregular seaways and develop quantitative techniques which reflect the nature and magnitude of the phenomena.

6. Review vulnerability criteria (including long term probability of loss of the ship) for intact and damaged ships, and outline further developments that are required.

7. Update ITTC Recommended Procedure 7.5-02-07-04.2, Model Tests on Damage Stability in Waves, paying specific attention to:

452452

a. Investigate the significance of scale effects in air pressure on flooding model tests under atmospheric conditions. Comment on the need to perform flooding model tests under scaled air pressure conditions.

b. Investigate how to deal with the inertia due to the flood water mass.

8. Cooperate with IMO SLF subcommittee.

9. Investigate the roll damping for large amplitude roll motions in irregular seas. Review suitable data for future benchmarking of time domain computer codes.

5.3. Terms of Reference for Specialist Committees

Specialist Committee on CFD in Marine Hydrodynamics

Computational capabilities are making

progress in the design and evaluation processes for many vehicles of interest including marine vehicles. Although inviscid methods are still often used, RANS codes, DES, LES and DNS are starting to play a larger role in the study of flow fields generated by marine vehicles. It is inevitable that these methods will have an even larger role in the future as computer power increases and the application of such codes matures even further. However, it will still take considerable effort to have the confidence in these methods that currently exist with the same level as in model tests, since grid resolution, turbulence modeling and other sources of uncertainties are still major

factors which affect the accuracy of solutions. In ITTC, as the range of application of CFD has been extended, the issues have been discussed in several committees, (Resistance, Manoeuvring, Propulsion, Seakeeping and Ocean Engineering Committees, for example). The purpose of this specialist committee is to comprehensively review the past work on the areas treated separately by those committees. General conclusions on the status of practical applications of CFD and suggestions for future CFD applications will be beneficial to all members of ITTC.

1. Review from an interdisciplinary

perspective, the current status of CFD in areas of importance to the ITTC. Include resistance, propulsion, propulsors, manoeuvring, steep and breaking wave simulation, seakeeping, ocean engineering and steady and unsteady flow field prediction at model and full scale.

2. Review the developments and identify the need for research in steady and unsteady computational fluid dynamics at full scale, including the implementation of real-time CFD analyses for the use in manoeuvring simulators.

3. Define which benchmark data are needed for CFD validation. Include the requirement for experimental data. Create a list of benchmark experimental data for validation of different aspects of CFD for hydrodynamics of ships and offshore structures, including the output needed from such experiments and the level of experimental uncertainty required.

453453

4. Check the need for formal procedures and guidelines on CFD verification and validation in specific areas.

5. Update the guideline 7.5-03-02-03, Practical Guidelines for Ship CFD Applications.

6. Review use and validation of CFD methods for wake scaling and determination of nominal full-scale wake.

7. Develop procedure for RANS simulation of model scale and full scale nominal wakes.

8. Review recent developments in techniques for direct numerical simulation of wakes (LES, DNS, SPH, ect).

Specialist Committee on Detailed Flow Measurement Techniques

1. Survey and report on the existing

detailed flow visualization, measurement techniques and data analysis methods.

2. Develop best-practice guidelines for the applications of PIV/SPIV in tow tanks and cavitation tunnels.

3. Develop experimental benchmarks for the verification of PIV/SPIV setup.

4. Perform an uncertainty analysis to assess PIV error sources beyond those considered in existing ITTC Recommended Procedure 7.5-01-03-03. These include peak locking errors, error due to improper light sheet overlap, effects of velocity gradients in the interrogation region, etc.

5. Develop a Guideline for SPIV uncertainty analysis.

6. Collaborate with the Specialist Committee on CFD to develop methods for the validation of CFD codes using detailed flow measurements.

Specialist Committee on Hydro-dynamic Modeling of Renewable Energy Devices

1. Review and update the guideline on

wave energy converters.

2. Develop guidelines for the physical modeling of wind and current/tidal renewable energy systems, both floating and bottom fixed structures.

3. Produce a guideline for large scale tests in open environment.

4. Investigate and report on techniques for the modeling of power take-off (PTO) systems.

5. Review and report on techniques for the numerical modeling of renewable energy systems.

6. Investigate and suggest improvements for wind load modeling on wind turbine devices during physical model testing.

7. Identify the parameters that cause the largest uncertainties in the results of physical model experiments and the extrapolation to full scale.

8. Investigate and report on the correct modeling for renewable energy system arrays (farms).

454454

Specialist Committee on Hydro-dynamic Noise

1. Create an overview of the

characteristics of hydrodynamic noise sources (including machinery and equipment, e.g. sonars) and their influence on the marine environment

2. Create an overview of existing national and international regulations regarding hydrodynamic noise

3. Check the existing methods and develop relevant guidelines how to perform both model and full scale noise measurements

4. Identify scale effects in prediction of hydrodynamically generated noises ( flow noise, cavitation noise,..)

5. Examine the possibilities to predict full scale values (correlation and operational requirements).

Specialist Committee on Performance on Ships in Service

The purpose of the Committee is to improve the performance predictions (especially for large ships) for service conditions covering the whole life cycle of the ship, keeping in mind the EEDI and EEOI development within IMO.

1. Cooperate directly with the AC and ITTC’s representative in IMO with regard to EEDI.

2. Liaise with the Resistance, Propulsion and Seakeeping Committees as relevant, specifically with regard to estimating fw, in the EEDI calculation.

3. Monitor and review the state of the art for EEDI and EEOI prediction and determination methods, including CFD based ones.

4. Review the existing procedures for the ship model testing with regard to the requirements arising from the EEDI prediction process, including ITTC Recommended Procedure 7.5-02-07-02.2, Prediction of Power Increase in Irregular Waves from Model Tests, and liaise with the Seakeeping Committee to decide whether an update of the procedure is required.

5. Identify and describe the practical aspects of the EEDI prediction process involving ship model testing, and develop a guideline for EEDI prediction.

6. Take into account minimum power requirements for safe and effective manoeuvring with respect to the EEDI formula (sea margin).

7. Describe the type of data (and the quality of that data) that should be recorded during full scale monitoring trials, including the issues of surface roughness.

8. Review the existing ITTC trial test procedures in this context. Review the existing speed correction methods for Full Scale Trial Measurements including ISO 15016, and come up with recommendation if the problems are identified, taking into account the MARIN report as contained in document MEPC 62/5/5.

9. Review the technologies (hydrodynamic issues) for enhancement of the powering performance, such as

455455

speed reduction, energy saving devices, hull form and propeller design etc.

10. Investigate the experimental and numerical possibilities to estimate the effect of manoeuvering and wind to the added resistance.

11. Look for full scale data that will allow to improve powering estimation taking into account the surface roughness (hull, appendages and propeller).

12. Examine the possibilities for numerical methods in the prediction of the influence of surface roughness on the power prediction.

Specialist Committee on Ice

1. Ice properties modeling (full scale and model scale) considering various conditions, ridges, pressurized ice, etc. for both offshore structures and ships.

a. Review and update the state of the art, regarding new relevant ice conditions such as brash ice channels (related to Ice Class powering requirements) both in frozen channel and fresh channel

b. Examine methods to model and measure various ice properties

c. Gather information of scatter in model ice properties within one ice sheet (statistical distribution)

2. Define which existing ice related

procedures need to be checked and if new ones need to be developed

3. Look into operational conditions in

freezing seas (in view of the climate change) in terms of relevant modelling. Conditions needed to be modeled are for example:

a. Brash ice channels b. Icing c. Ice and waves, wind, current;

ice dynamics

4. Review the existing numerical methods (offshore structures and ships) concerning:

a. Model ice failure b. Ice resistance, propulsion,

manoeuvring, ice loads c. Operational simulation incl.

positioning in ice 5.4. Terms of Reference for the Groups

Quality Systems Group

1. Include a definition of the terms Verification and Validation in the ITTC Symbols and Terminology List (to be done within first three months as a basis for the work of other committees).

2. Maintain the Manual of ITTC Recommended Procedures and Guidelines. Co-ordinate the modification and re-editing of the existing procedures according to the comments made by ITTC member organizations at the Conference and by the Technical Committees.

3. Support the Technical Committees in their work on Recommended Procedures. Supply the chairmen of the new committees at the beginning of the period with the MS Word versions of the relevant procedures and the template for the production of new procedures.

4. Observe the development or revision of ISO Standards regarding Quality Control.

456456

5. Update the ITTC Symbols and Terminology List.

6. Update the ITTC Dictionary of Hydromechanics.

7. Cross-check the ITTC Symbols List and the Dictionary with other standards e.g. ISO.

8. Revise and update the existing ITTC Recommended Procedures according to the comments of Advisory Council, Technical Committees and the Conference.

9. Before the third AC Meeting, review and edit new ITTC Recommended Procedures with regard to formal Quality System requirements including format and compliance of the symbols with the ITTC Symbols and Terminology List.

10. Follow the implementation of the Benchmark data repository.

11. Support the technical committees with guidance on development, revision and update of uncertainty analysis procedures.

12. Observe ISO standards for uncertainty analysis, in particular the uncertainty analysis terminology.

13. Maintain Wiki for the 27th ITTC as a trial period and create link to it from the ITTC website.

457457

Appendix 3

Tasks and structure of the 28th ITTC technical committees and groups

1. STRUCTURE OF TECHNICAL

COMMITTEES

The structure of the technical committees includes six General Committees, six Specialist Committees and one Group. 2. TERMS OF REFERENCE FOR

THE GENERAL AND SPECIALIST TECHNICAL COMMITTEES AND GROUPS

2.1 General Committees

Each General Committee will be responsi-ble for a general subject area. It will review the state-of-the-art, identify the need for research and development, and carry out longer terms studies with broad impact.

An important part of the work of the Gen-

eral Committees will be to establish Procedures and Guidelines to help the ITTC Member Or-ganizations maintain their institutional credibil-ity with regard to quality assurance of products and services such as predictions and evalua-tions, and quality assurance of designs. The General Committees will develop detailed plans in accordance with Conference Recom-mendations and their work should be directed towards the techniques and understanding of physical and numerical modelling as a means of predicting full-scale behaviour. While main-

taining an awareness of progress, fundamental theoretical studies and fundamental aspects of numerical fluid computation should be covered by other fora. Procedures and Guidelines shall contain only techniques which are applicable in commercial practice.

Each General Committee will submit a re-

port on the results of its work to the Full Con-ference. The conclusions and the recommenda-tions of the General Committee report should be structured as follows:

1. General technical conclusions 2. Recommendations to the Full Con-

ference, which require actions such as, e.g., adopting ITTC procedures.

In addition, each General Committee shall

submit proposals for future work of the General Committee and identification of tasks, which may be appropriate for Specialist Committees. These proposals shall be submitted to the Advi-sory Council which will compile the proposals and present them to the Full Conference. 2.2 Specialist Committees

The ITTC Advisory Council will propose Specialist Committees. Each Specialist Com-mittee will be responsible for studying a spe-cific technical problem. The Specialist Com-mittees will be appointed for a limited duration. It is expected that they will complete their tasks

458458

within maximum two ITTC periods (6 years). They shall interact closely with the appropriate General Committees. The tasks of a Specialist Committee can include establishing Procedures and/or Guidelines. Procedures and Guidelines shall contain only techniques which are appli-cable in commercial practice.

Each Specialist Committee will present a

final report on the results of its work to the Full Conference and interim reports on progress if the duration of the committee spans more than one Conference. The conclusions and the rec-ommendations of the Specialist Committee report should be structured as follows:

1. General technical conclusions 2. Recommendations to the Full Con-

ference, which require actions such as, e.g., adopting ITTC procedures.

In addition, each Specialist Committee shall

submit proposals for future work of and identi-fication of tasks, which may be appropriate for Specialist Committees. These proposals shall be submitted to the Advisory Council which will compile the proposals and present them to the Full Conference. 2.3 Groups

Groups may be established from time to time by the Executive Committee to carry out specific tasks for the Conference, which are not technical issues.

Each Group will present a final report on the results of its work to the Full Conference. The conclusions and the recommendations of the Group report should be structured as fol-lows:

1. General technical conclusions

2. Recommendations to the Full Con-ference, which require actions such as, e.g., adopting ITTC procedures.

In addition, each Group shall submit pro-

posals for future work of and identification of tasks, which may be appropriate for General and Specialist Committees. These proposals shall be submitted to the Advisory Council which will compile the proposals and present them to the Full Conference. 3. MECHANISM FOR IDENTIFYING

NEW SPECIALIST TECHNICAL COMMITTEES

As part of their Terms of Reference, the

General Committees shall consider the need for new tasks and include appropriate proposals in their technical reports. If the Advisory Council identifies a need for a new Specialist Commit-tee when it reviews the draft recommendations of the General Committees, the Council will prepare and agree a statement of the technical aims and objectives for the work of the Special-ist Committee.

Independently of the proposals of the Gen-

eral Committees, the Advisory Council will keep under continuous review the requirement for Specialist Committees.

When the Advisory Council has agreed the

need for a new Specialist Committee, the draft statement of technical aims and objectives will be presented to the Executive Committee for endorsement. If the Executive Committee ap-proves the formation of a new Specialist Com-mittee, it will present the proposal to the Full Conference for approval.

459459

4. PROPOSED STRUCTURE OF THE

TECHNICAL COMMITTEES AND GROUPS FOR 28TH ITTC

4.1 General Committees

• Resistance • Propulsion • Manoeuvring • Seakeeping • Ocean Engineering • Stability in Waves

4.2 Specialist Committees

• Performance of Ships in Service • Hydrodynamic Noise • Hydrodynamic Modelling of Marine

Renewable Energy Devices • Ice • Energy Saving Methods • Modelling of Environmental Condi-

tions 4.3 Groups

• Quality Systems Group 5. TASKS OF THE TECHNICAL

COMMITTEES AND GROUPS OF THE 28TH ITTC

5.1 General Terms of Reference

1. All committees shall observe the Terms of Reference and general ob-ligations. The committees are ex-pected to perform all the tasks de-fined in this document. However,

should a committee be unable to do this, it shall consult the Advisory Committee with regard to reduction of the work.

2. Each technical committee shall con-

sider any unfinished items from previous committees and report to the Advisory Council by 1st De-cember 2014 in order to clarify whether these items should be in-cluded in the Terms of Reference.

3. All committees shall identify areas

of mutual interest with other com-mittees and the concerned commit-tees shall establish active co-operation in these areas.

4. In their work, the committees shall

follow the guidelines given in ITTC Recommended Procedure 1.0-03, General Guideline for the Activities of Technical Committees, Liaison with the Executive Committee and Advisory Council.

5. Procedures and guidelines must be

in the format defined in the ITTC Recommended Procedure 4.2.3-01-03, Work Instruction for Formatting ITTC Recommended Procedures, and they will be included in the ITTC Quality Manual. Symbols and terminology must be in accordance with those used in the current ver-sion of the ITTC Symbols and Ter-minology List. If necessary, new symbols should be proposed in col-laboration with the Quality Systems Group. Recommended Procedure 4.2.3-01-03 contains a template, which shall be used for new proce-dures and guidelines.

460460

6. All new procedures for uncertainty

analysis in experiments shall follow the ISO (1995) ‘Guide to the Ex-pression of Uncertainty in Meas-urements’ (also known as ISO-GUM). It is not required to update existing procedures on uncertainty analysis to follow this standard. If a procedure for uncertainty analysis is for other reasons updated, it shall follow the ISO standard.

7. Committees that have a task to re-

view ITTC Recommended Proce-dures shall identify and report any changes proposed in their first annu-al report to the Advisory Council. The changes approved by the Advi-sory Council should be implemented in the second year and the draft re-vised procedure submitted to the Advisory Council for comment.

8. All general committees shall survey

and review new techniques within CFD in their area and shall include the results thereof in their report.

9. All general committees shall moni-

tor advances in the application of detailed flow measurements in the ITTC community to assess the need for detailed evaluation and imple-mentation of best-practice, uncer-tainty analysis, and benchmark guidelines.

10. Committees that have a task to write

new procedures or guidelines shall submit an outline of these with their first annual report to the Advisory Council. The outline shall be re-viewed by the Advisory Council and comments made to the committees.

The draft new procedure or guide-line shall be prepared during the se-cond year and submitted to the Ad-visory Council for review.

11. All new and revised procedures

shall, as far as feasible, include a procedure for uncertainty analysis.

12. New and revised draft procedures

shall subsequently be updated, in-corporating the comments made by the Advisory Council, and in Febru-ary of the third year be submitted to the Advisory Council for final re-view and approval. Thereafter, the Quality Systems Group shall per-form a formal check of the proce-dures.

13. Committee reports to the Confer-

ence should be structured in line with the terms of reference of the committee and in accordance with Recommended Procedure 4.2.3-01-02, Guidelines for Preparation of Committee and Group Reports.

5.2 Terms of Reference for the General

Committees Resistance Committee

1. Update the state-of-the-art for pre-dicting the resistance of different ship concepts emphasizing devel-opments since the 2014 ITTC Full Conference. The committee report should include sections on: a. The potential impact of new

technological developments on the ITTC

461461

b. New experimental tech-

niques and extrapolation methods

c. New benchmark data d. The practical applications of

computational methods to re-sistance predictions and scal-ing

e. The need for R&D for im-proving methods of model experiments, numerical modelling and full-scale measurements.

2. During the first year, review ITTC

Recommended Procedures relevant to resistance and resistance specific CFD procedures, and a. Identify any requirements for

changes in the light of cur-rent practice, and, if ap-proved by the Advisory Council, update them

b. Identify the need for new procedures and outline the purpose and contents of the-se

3. Review definitions of ship surface

roughness and develop a guideline for its measurement, hereunder re-solve differences between ISO 4287 and the widely used BMT roughness measurement system. Include effect of coatings and through-life chang-es.

4. Review trends and new develop-ments on understanding the phe-nomenon of unsteady free surface flows, including their influence on added resistance and experimental techniques.

5. Develop a new procedure for meas-urement of wave pattern generated by the hull model , and wave re-sistance analysis, including its influ-ence through uncertainty analysis for extrapolation.

6. Review roughness of models and

appendages produced by rapid pro-totyping. Assess effects of this roughness on resistance.

7. Propose guidance for ITTC mem-

bers to reduce/manage their uncer-tainty as a result of the worldwide resistance benchmark tests of previ-ous ITTCs.

8. Review turbulence stimulation

methods and devices from the point of view of their physics and update the relevant procedure 7.5-01-01-01 Ship models. Check occurrence of turbulence stimulation methods in other procedures and update as needed.

9. Develop a procedure for verification

and validation of the detailed flow field data

10. An ITTC benchmark study shall be

initiated according to 7.5-01-03-04 Benchmark for PIV (2C) and SPIV (3C) setups. The benchmark study would involve PIV measurements performed on a flow of interest, with fully detailed uncertainty analysis. The results can then be compared with similar measurements done in different facilities or with high-quality CFD computations from var-ious organizations.

462462

Propulsion Committee

1. Update the state-of-the-art for pre-dicting the propulsive performance of ships, emphasizing developments since the 2014 ITTC Conference. The committee report should in-clude sections on: a. The potential impact of new

technological developments on the ITTC including new types of propulsors (e.g. hybrid propulsors), azimuthing thrust-ers and propulsors with flexible blades

b. New experimental techniques and extrapolation methods

c. New benchmark data d. The practical applications of

computational methods to the propulsion systems predictions and scaling

e. New developments of experi-mental and computational meth-ods applicable to the prediction of cavitation

f. The need for R&D for improv-ing methods of model experi-ments, numerical modelling and full-scale measurements

g. Monitoring the developments regarding high-speed marine vehicles.

2. During the first year, review ITTC

Recommended Procedures relevant to propulsion and cavitation, includ-ing CFD procedures, and a. Identify any requirements for

changes in the light of current practice and, if approved by the AC, update them,

b. Identify the need for new proce-dures and outline the purpose and contents of these.

3. Liaise with the Specialist Commit-

tee on Energy Saving Methods on subjects of common interest.

4. Liaise with the Specialist Commit-tee on Performance of Ships in Ser-vice regarding consequences of EEDI, especially with respect to ITTC Recommended Procedure 7.5-02-03-01.4, 1978 ITTC Perfor-mance Prediction Method, with spe-cial emphasis on the proposed value of the propeller roughness (to high), ΔCF and CA, also for different draft conditions. Harmonize the formulae in ITTC Recommended Procedures 7.5-02-03-01.4 and 7.5-02-03-01.2.

5. Develop new roughness correction methods for both hull and propeller.

6. Continue with the monitoring of ex-

isting full scale data for podded pro-pulsion. If there is available data, re-fine the existing procedure.

7. Review and update guideline 7.5-

02-03-01.6, Hybrid Contra-Rotating Shaft Propulsors Model Test.

8. Review and update, if required,

Recommended Procedure 7.5-02-05-03.2 Waterjet System Perfor-mance.

9. Develop an extension of the existing

procedure 7.5-02-03-01.4, 1978 ITTC Performance Prediction Method for triple shaft vessels.

463463

10. Examine methods of target wake

simulation with the support of CFD (smart dummy).

11. Continue the given task 8 from the

former period (Examine the possi-bilities of CFD methods regarding scaling of unconventional propeller open water data. Initiate a compara-tive CFD calculation project).

12. Monitor the use of and, if possible,

develop guidelines for quasi-steady open water propeller and propulsion model tests.

Manoeuvring Committee

1. Update the state-of-the-art for pre-dicting the manoeuvring behaviour of ships, emphasizing developments since the 2014 ITTC Conference. The committee report should in-clude sections on: a. the potential impact of new

technological developments on the ITTC

b. developments in manoeuvring and especially course keeping in waves

c. new experiment techniques and extrapolation methods

d. the practical applications of computational methods to manoeuvring predictions and scaling, including CFD methods

e. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements

f. the effects of free surface, roll, sinkage, heel and trim in numer-ical simulation of manoeuvring.

2. During the first year, review ITTC Recommended Procedures relevant to manoeuvring, including CFD procedures, and a. Identify any requirements for

changes in the light of current practice and, if approved by the Advisory Council, update them

b. Identify the need for new proce-dures and outline the purpose and contents of these.

3. Liaise with the Specialist Commit-

tee on Ice with regard to the possible updating of ITTC Recommended Procedure 7.5-02-04-02.3, Manoeu-vring in Ice.

4. Update ITTC Recommended Proce-dure 7.5-04-02-01, Full Scale Manoeuvring Trials Procedure. In-clude consideration of full scale to model scale correlation. In particu-lar, examine the model scale to full scale correlation of steering control in manoeuvring.

5. Update ITTC Recommended Proce-

dure 7.5-02-06-02, Captive Model Test Procedure, with particular at-tention to the use of PMM and hex-apod.

6. Continue work in order to have a

full set of benchmark data for each of the benchmark hulls (KVLCC2, KCS, 5415, HTC, SUBOFF and S175 – manoeuvring in waves).

7. Extend the uncertainty analysis for captive model tests from measure-ments towards a new procedure that provides the uncertainty analysis for the use of captive model tests in the

464464

predictions of manoeuvring. Elabo-rate with an example.

8. Develop a procedure for verification and validation of ship manoeuvring simulation methods, including CFD.

9. Review testing methods for ships in ports and harbours, including ship mooring loads, safe speed limits for moving and its impact on moored vessels, manoeuvring concerns such as squat and bank effects.

10. Conduct a concise review and report on the specific aspects of the manoeuvring of vessels in restricted waters such as ports and harbours.

11. Organize a joint workshop on manoeuvring in waves with the Seakeeping and the Stability in Waves Committees and the Special-ist Committee on Performance of Ships in Service, for example on the subject of minimum power require-ments for safe manoeuvring in ad-verse sea conditions and model test-ing methods to investigate this.

Seakeeping Committee

Note: The Seakeeping Committee is primar-ily concerned with the behaviour of ships un-derway in waves. The Ocean Engineering Committee covers moored and dynamically positioned ships. For the 28th ITTC, the model-ling and simulation of waves, wind and current is the primary responsibility of the Specialist Committee on Modelling of Environmental Conditions, with the cooperation of the Ocean Engineering, the Seakeeping and the Stability in Waves Committees.

1. Update the state-of-the-art for pre-dicting the behaviour of ships in waves, emphasizing developments since the 2014 ITTC Conference. The committee report should in-clude sections on: a. the potential impact of new

technological developments on the ITTC

b. new experiment techniques and extrapolation methods

c. new benchmark data d. the practical applications of

computational methods to seakeeping predictions and scal-ing, including CFD methods

e. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

2. Review ITTC Recommended Pro-

cedures relevant to seakeeping, in-cluding CFD procedures, and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them

b. Identify the need for new proce-dures and outline the purpose and contents of these.

3. Update ITTC Recommended Proce-

dure 7.5-02-07-02.5, Verification and Validation of Linear and Weak-ly Non-linear Seakeeping Computer Codes to include the verification and validation of ship hydro-elasticity codes. It is recommended that the developed section/procedure is re-viewed by ISSC Loads and Re-sponse Committee.

465465

4. Update ITTC Recommended Proce-

dure 7.5-02-07-02.1, Seakeeping Experiments, to include tests specif-ic to active stabilization systems, with particular attention to modeling the control system and prediction of full scale behaviour. If possible, up-date the corresponding procedure for high speed craft.

5. Review ITTC Recommended Pro-

cedures 7.5-02-05-06, Structural Loads, and 7.5-02-05-07, Dynamic Instability Tests, and propose up-dates, if any.

6. Develop a new procedure for the de-termination of speed reduction coef-ficient fw. Liaise with the Specialist Committee on Performance of Ships in Service.

7. Develop a new procedure for model scale sloshing experiments.

8. Review the research considering the impact of seakeeping on propulsion and manoeuvring performance. Li-aise with the Manoeuvring Commit-tee.

9. Survey and/or collect benchmark data for seakeeping problems, such as motion, loads, sloshing, slam-ming, added resistance, full-scale measurements.

10. Continue the collaboration with ISSC committees, including Loads and Responses and Environment Committees.

11. Support a joint workshop on manoeuvring in waves with the

Manoeuvring and the Stability in Waves Committees and the Special-ist Committee on Performance of Ships in Service, for example on the subject of minimum power require-ments for safe manoeuvring in ad-verse sea conditions and model test-ing methods to investigate this.

Ocean Engineering Committee

Note: The Ocean Engineering Committee covers moored and dynamically positioned ships and floating structures. For the 28th ITTC, the modelling and simulation of waves, wind and current is the primary responsibility of the Specialist Committee on Modelling of Environmental Conditions, with the coopera-tion of the Ocean Engineering, the Seakeeping and the Stability in Waves Committees.

1. Update the state-of-the-art for pre-dicting the behaviour of bottom founded or stationary floating struc-tures, including moored and dynam-ically positioned ships, emphasizing developments since the 2014 ITTC Conference. The committee report should include sections on: a. the potential impact of new

technological developments on the ITTC

b. new experimental techniques and extrapolation methods

c. new benchmark data d. the practical applications of

computational methods to pre-diction and scaling

e. the need for R&D for improving methods of model experiments, numerical modelling and full-scale measurements.

466466

2. Review ITTC Recommended Pro-

cedures relevant to ocean engineer-ing, including CFD procedures, and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them

b. Identify the need for new proce-dures and outline the purpose and contents of these.

3. Complete the VIM guideline 7.5-02-

07-03.11, which was initiated by the Ocean Engineering Committee of the 27th ITTC.

4. Continue work to further quantify the uncertainty sources in ocean en-gineering model tests. Prepare a list of uncertainty parameters with quantified values and conduct com-prehensive uncertainty analyses for selected tests. Collaborate with the Seakeeping Committee and the Spe-cialist Committee on Modelling of Environmental Conditions.

5. Investigate the modelling of wind

for Ocean Engineering tests with at-tention to inflow conditions in co-operation with the Specialist Com-mittee on Modelling of Environ-mental Conditions.

6. Develop a guideline for model tests

of multi-bodies operating in close proximity.

7. Develop a procedure for the analysis

of model tests in irregular waves.

8. Continue investigating thruster-thruster interaction, ventilation and their scaling for DP systems.

9. Conduct a concise review and report on specific aspects of moored struc-tures and vessels in port such as the effects of passing ships and harbor resonance.

10. Continue wave run-up benchmark

studies for cases of four columns (circular and square cross-sections) using the experimental data from Marintek.

11. Review the state-of-the-art in off-

shore mining, including model test-ing methods.

Stability in Waves Committee

Note: The Stability in Waves Committee covers the stability of intact and damaged ships in waves. For the 28th ITTC, the modelling and simulation of waves, wind and current is the primary responsibility of the Specialist Com-mittee on Modelling of Environmental Condi-tions, with the cooperation of the Ocean Engi-neering, the Seakeeping and the Stability in Waves Committees.

1. Update the state-of-the-art for pre-dicting the stability of ships in waves, emphasizing developments since the 2014 ITTC conference. The committee report should in-clude sections on: a. the potential impact of new

technological developments on the ITTC

b. new experimental techniques and extrapolation methods

c. new benchmark data d. the practical applications of

computational methods to pre-diction and scaling

467467

e. the need for R&D for improving

methods of model experiments, numerical modelling and full-scale measurements.

2. Review ITTC Recommended Pro-

cedures relevant to stability, includ-ing CFD procedures, and a. Identify any requirements for

changes in the light of current practice, and, if approved by the Advisory Council, update them

b. Identify the need for new proce-dures and outline the purpose and contents of these.

3. Update ITTC Recommended Proce-

dure 7.5-02-07-04.1, Model Tests on Intact Stability, to include uncer-tainty analysis of measurements, in-cluding an example.

4. Consider whether the guideline for

the prediction of the occurrence and magnitude of parametric rolling should be updated to a procedure.

5. Continue to investigate the criteria

for modelling wave spectra in the determination of dynamic instability of intact vessels, i.e. wave steepness, non-linearity, frequency contents of the spectrum, statistical distribution of wave and crest height and spatial behaviour of the waves. Liaise with the Specialist Committee on Model-ling of Environmental Conditions.

6. Include in the report an example of

uncertainty analysis in measure-ments for damage stability model tests.

7. Review surface tracking methods for measuring flow between com-partments of damaged ships.

8. Review the current approaches for

quantifying time to loss and report in the impact for model testing.

9. Examine the model scale to full

scale correlation of steering control and active motion control in extreme waves.

10. Support a joint workshop on

manoeuvring in waves with the Manoeuvring and the Seakeeping Committees and the Specialist Committee on Performance of Ships in Service, for example on the sub-ject of minimum power require-ments for safe manoeuvring in ad-verse sea conditions and model test-ing methods to investigate this.

11. Continue cooperation with IMO SDC (formerly SLF) subcommittee.

5.3 Terms of Reference for Specialist

Committees Specialist Committee on Hydrodynamic Modelling of Marine Renewable Energy De-vices

Work relating to wave energy converters (WEC):

1. Develop guidelines for uncertainty prediction for WECs.

2. Monitor and report on developments in power take-off (PTO) modelling both for physical and numerical pre-diction of power capture.

468468

3. Review and report on the progress

made on the modelling of WEC ar-rays.

4. Review and report on challenges as-

sociated with the performance of WECs in irregular wave spectra, particularly as it relates to physical modelling.

5. Check willingness of participants

for the “round-robin” test campaign before starting work.

6. Review and report on integrated

WEC simulation tools based on multi-body solvers which are in de-velopment.

Work relating to current turbines:

1. Develop specific uncertainty analy-

sis guidelines / example for horizon-tal axis turbines

2. Report on development in physical and numerical techniques for predic-tion of performance of current tur-bines, with particular emphasis on unsteady flows, off-axis conditions, and other phenomena which offer particular challenges to current de-vices.

3. Report on the progress made on the

modelling of arrays.

4. Report on progress in testing at full-scale and moderate scale in-sea test sites.

Work relating to offshore wind:

1. Wind Field Modelling including Froude/Reynolds scaling challenges for the turbine in cooperation with the Specialist Committee on Model-ling of Environmental Conditions.

2. The impact of control strategies and

other features on full-scale devices on global response to allow im-proved understanding of the impact of simplifications adopted in model tests.

3. Report on integrated tools for simu-

lation of floating wind turbine in-cluding platform, mooring, turbine and control system.

4. Report on developments in full-

scale demonstrators of floating wind turbines.

Specialist Committee on Hydrodynamic Noise

1. Continue development of the guide-lines produced during the 27th ITTC and monitor how these guidelines are being implemented by the tow-ing tank community.

2. Identify scale effects in prediction of hydrodynamically generated noises (flow noise, cavitation noise, etc.).

3. Examine the possibilities to predict

full scale values (correlation and op-erational requirements) from model scale noise measurements.

4. Review uncertainties associated

with model scale noise measure-

469469

ments and full scale noise measure-ments, including variability between sister ships and influence of opera-tional conditions during sea trials, such as manoeuvring and sea state.

5. Check the existing methodologies

regarding full scale noise measure-ments in shallow and restricted wa-ter and provide, if possible, guide-lines. Establish communication with ISO working groups active on this topic.

6. Update the overview of national and

international regulations and stand-ards regarding hydrodynamic noise.

7. Review the developments of pre-

dicting methods (theoretical and numerical) for underwater noise sources characterisation and for far field propagation.

8. Define and, if possible, conduct

benchmarking tests of model test noise measurements, preferably for a ship for which full scale noise measurements are available.

Specialist Committee on Performance on Ships in Service

The purpose of the committee is to improve the performance predictions (especially for large ships) for service conditions covering the whole life-cycle of the ship, keeping in mind the EEDI and EEOI development within IMO.

1. Liaise with the Resistance, Propul-sion and Seakeeping Committees as relevant, specifically with regard to estimating fw, in the EEDI calcula-tion and CA guideline.

2. Monitor and review the state of the

art for EEDI and EEOI prediction and determination methods, includ-ing CFD based methods.

3. Monitor and review the state of the

art with regard to minimum power requirements for safe and effective manoeuvring and requirements aris-ing from the EEDI formula (sea margin).

4. Investigate the following aspects of

the analysis of speed/power sea trial results: a. Temperature and density correc-

tion to take into account temper-ature/density gradient

b. ISO proposed ‘iterative method' as an alternative for load varia-tion method and current elimina-tion.

c. Statistical properties for the re-sults of load variation tests

d. New shallow water correction method to replace Lackenby

e. Influence of headings and wind on sea trials

f. Wave limits for the wave cor-rection methods

g. Application of CFD methods for wind loads

h. Expansion of the wind coeffi-cient database for more ship types

i. More extensive validation of the wave correction methods (STA I , STA2, NMRI)

j. Feedback of speed/power data for correlation purpose especial-ly for the design and EEDI draft.

470470

5. On the basis of the results of these

investigations update the speed/power sea trial procedures 7.5-04-01-01.1 and -01.2.

6. Explore “Ship in Service” issues a. Investigate the monitoring and

analysis of speed/power perfor-mance of ships in service

b. Investigate feedback of speed power data to seakeeping com-mittee to get reliable data for fw from full-scale measurements

c. Investigate feedback of speed/power data for fw

d. Investigate the influence of ship hull surface degradation due to fouling and aging on the speed/power performance and consider the related EEOI issues originating from IMO require-ments

e. Establish a guideline for the use of CA for different draft condi-tions.

Specialist Committee on Ice

1. Finalize the updating of the guide-lines commenced during the 27th ITTC, including: 7.5-02-04-01and 7.5-02-04-02.1.

2. Revise the following procedures: • 7.5-02-04-02.3 Maneuvering

Tests In Ice (liaise with the Manoeuvring Committee)

• 7.5-02-04-02.2 Propulsion Test in Ice (liaise with the Propulsion Committee)

• 7.5-04-03-01 Ship trials in ice • 7.5-02-04-02.5 Experimental

Uncertainty Analysis for Ship

Resistance in Ice Tank Test-ing

3. Review the following items with a

view to developing procedures and guidelines in the future:

• Station keeping in ice, for moored, DP and thruster-assisted moored structures and ships.

• Offshore structures – fixed and floating

• Scalability of ice with the goal of new guidelines on full-scale ice property testing

• Brash Ice characterization and modelling.

4. Evaluate numerical and semi-

empirical prediction for operation in ice.

Specialist Committee on Modelling of Envi-ronmental Conditions

This new committee will deal with the modelling of realistic environmental conditions in a reliable, reproducible way in a model ba-sin. During the first period, the focus of the committee shall be in order of priority, waves, wind and current.

1. Review the report of Ocean Engi-neering Committee of the 24th ITTC (2005)

2. Propose and develop guidelines where appropriate.

Suggested topics may include:

Waves: 1. Non-linear effects – analysis, con-

trol

471471

2. Interactions with current and wind

3. Distribution of extremes

4. Wave grouping (characterization and reproduction)

5. Short-crested wave modelling

6. Deterministic generation of extreme

waves

7. Confinement a. Wave frequency and low fre-

quency reflections b. Radiation and reflection from

model, beach, etc.

8. Measurement and analysis of long- and short-crested waves

9. Non-stationary power spectrum (time and space)

10. Wave breaking – influence on statis-

tics and kinematics

11. Geographical consistency of wave spectrum selection

Wind: 1. Interaction with waves

2. Gusting (including squalls)

3. Turbulence

4. Vertical profiles

5. Horizontal variation

6. Measurements

7. Geographical consistency of wind conditions

Current:

1. Interactions with waves

2. Turbulence

3. Vertical profiles (including current reversal)

4. Horizontal variation

5. Measurements

Specialist Committee on Energy Saving Methods

1. Conduct a systematic survey of en-ergy saving methods, devices, appli-cations and possible savings, includ-ing the influence on the EEDI for-mula.

2. Identify the physical mechanisms on energy saving on ships.

3. Conduct a survey on the frictional

drag reduction methods, including air lubrication and surface treatment.

4. Conduct a survey on energy savings

based on the use of wind energy.

5. Monitor the CFD methods, model tests and scaling procedures for en-ergy saving devices.

6. Conduct a survey on existing full

scale data on the effect of energy saving methods.

7. Identify the needs for new model

test procedures (resistance and pro-

472472

pulsion, extrapolation methods) to investigate the effect of energy sav-ing methods.

5.4 Terms of Reference for the Groups Quality Systems Group

1. Update all ITTC Recommended Procedures and Guidelines to con-form to the requirements of Rec-ommended Procedure 4.2.3-01-03, Work Instruction for Formatting ITTC Recommended Procedures and Guidelines.

2. Support the Technical Committees in their work on Recommended Procedures. Supply the chairmen of the new committees with the MS Word versions of the relevant pro-cedures.

3. Maintain the Manual of ITTC Rec-ommended Procedures and Guide-lines. Co-ordinate the modification and re-editing of the existing proce-dures according to the comments made by ITTC member organiza-tions at the Conference and by the Technical Committees.

4. Observe the development or revi-

sion of ISO Standards regarding Quality Control.

5. Update the ITTC Symbols and Ter-

minology List. Consolidate some inconsistencies in the use of sym-bols and add recognized symbols not contained in the SaT list to the list(such as RAW, Rwave, rho ad rho0).

6. To further develop a liaison with

ISO with a view to reconcile the dif-ferences in definitions between ISO standards 15016, EEDI definitions and ITTC definitions as laid down in the procedures 7.5-04-01-01.1 and -01.2

7. Update the ITTC Dictionary of Hy-

dromechanics.

8. Revise and update the existing ITTC Recommended Procedures accord-ing to the comments of Advisory Council, Technical Committees and the Conference.

9. After the third AC Meeting, review

and edit new ITTC Recommended Procedures with regard to formal Quality System requirements in-cluding format and compliance of the symbols with the ITTC Symbols and Terminology List.

10. Support the Technical Committees

with guidance on development, re-vision and update of uncertainty analysis procedures.

11. Observe ISO standards for uncer-

tainty analysis, in particular the un-certainty analysis terminology.

12. Review developments in metrology

theory and uncertainty analysis and issue appropriate Procedures.

13. Continue to maintain the online

Wiki tool keeping it up to date and in line with the adopted documents of the ITTC.

473473

14. To develop a procedure on the de-termination of a type A uncertainty estimate of a mean value from sig-nal analysis, based on Brouwer et al. (2013). This analysis provides an uncertainty estimate in cases where instead of multiple repeat or repro-duction measurements, only a single time series is available.

15. To develop a guideline with work-ing title: “Guideline to Practical Im-plementation of Uncertainty Analy-sis”.

16. At the beginning of the period, or-

ganize an electronic repository of in-formation and data on the bench-marks cases. ITTC member organi-zations should then be invited to participate in the adoption of the benchmark and contribute to the da-tabase.

17. Survey the extent and breadth of up-

take of uncertainty analysis tech-niques and procedures by the hy-drodynamic testing community.

18. Include new sections in the Diction-

ary dedicated to Offshore Engineer-ing, Planing Craft and Pods.

474474

Appendix 4

Description and Rules of the ITTC (International Towing Tank Conference)

Proposed for adoption by the 27th Full Conference

1. DESCRIPTION The International Towing Tank Conference

(ITTC) is a world-wide independent association of hydrodynamics research organizations that operate towing tanks or similar model test labo-ratories. ITTC members support the designers, builders and operators of ships and marine in-stallations by giving advice and information regarding the performance, safety and environ-mental impact of ships and marine installations using the results of physical model tests, nu-merical modelling and full-scale measurements.

2. AIMS

The aims of the ITTC are:

(a) To stimulate progress in solving the technical problems which are of importance to towing tank organizations and model test labo-ratories; (b) To stimulate research in areas in which a better knowledge is required in order to im-prove methods of predicting the full-scale hy-drodynamic performance of ships and marine installations;

(c) To stimulate the improvement of meth-ods of model experiments, numerical modelling and full-scale measurements;

(d) To recommend procedures for carrying out physical model experiments, numerical

modelling and full scale measurements of ships and marine installations;

(e) To validate the accuracy of full-scale predictions for quality assurance;

(f) To formulate collective policy on mat-ters of common interest;

(g) To provide an effective organization for the interchange of information.

3. ACTIVITIES

The aims of the ITTC shall be pursued by:

(a) Stimulating research into specific top-ics;

(b) Organizing and encouraging meetings to review progress in this research;

(c) Making such recommendations and decisions on joint action and policy as seem desirable to the members of the ITTC;

(d) Establishing procedures and guidelines to help the member organizations to maintain their institutional credibility with regard to quality assurance of products and services, such as, performance prediction and evaluation of designs by either experimental or computa-tional means;

475475

(e) Recording and publishing discussions taking place at ITTC meetings.

4. MEMBERSHIP Membership of the ITTC shall be open to

all towing tank organizations or model test laboratories that carry out hydrodynamic work in support of the designers, builders and opera-tors of ships and marine installations, and to other organizations that contribute to the aims of the ITTC.

Applications for membership shall be made

to the Executive Committee through the ITTC Secretary. Each such organization shall satisfy the Executive Committee that it is eligible for membership.

Each member organization shall be repre-

sented by its director or other senior officer having the authority to bind the member or-ganization in matters relating to the ITTC (the designated representative).

A membership fee shall be payable by all

member organizations. The Executive Commit-tee shall propose the fee for the next three years for approval by the Full Conference. The fee shall be payable by October 1st of the year in which the triennial conference (the Conference) is held.

A member organization, which has not paid

the fee by May 31st in the year following the Conference, shall no longer be a member or-ganization of ITTC and the name of the organi-zation shall be removed from the membership list. The Executi-ve Committee may extend this deadline if unusual financial or administrative circumstances delay the payment of the fee.

5. FULL CONFERENCE The Full Conference comprises the desig-

nated representatives of member organizations eligible to vote and present at general sessions that take place during the Conference.

5.1. Roles and responsibilities The Full Conference shall: (a) Determine the policies of the ITTC; (b) Approve changes to the rules of the ITTC;

(c) Appoint the Chairman of the Executive Committee and the ITTC Secretary;

(d) Appoint the Chairman and members of each technical committee or group;

(e) Approve financial reports and plans and the ITTC membership fee;

(f) Approve the host organization for the next Conference;

(g) Approve terms of reference for techni-cal committees and groups;

(h) Approve recommended procedures and guidelines.

Only member organizations are eligible to vote. The vote shall be exercised by the desig-nated representative of the organization and no organization shall be entitled to more than one vote. A designated representative who is unable to attend the meeting may choose to delegate the voting rights of the member organization to another employee of the organization. The des-ignated representative must inform the Chair-man of the Executive Committee of the name of the alternate before the start of the general

476476

session at which the vote will take place. Postal or email votes shall not be allowed.

Voting may be by secret ballot or a show of

hands as determined by the Executive Commit-tee. An affirmative vote of at least 2/3 of mem-bers present shall be required to carry a motion.

A record of the decisions of the Full Con-

ference shall be published in the proceedings of the Conference.

5.2. Decision making between Full Conferences

If for any reason a decision is required in

the time gap between Full Conferences with regard to the items listed in 5.1, the Executive Committee unanimously and supported by a majority of the Advisory Council is mandated to make such decision. If a unanimous decision cannot be made, a decision shall be made in accordance with a procedure approved by the Full Conference.

Any decisions made in accordance with 5.2

shall be reported to the Full Conference and recorded in the proceedings of the Conference.

6. EXECUTIVE COMMITTEE 6.1. Roles and responsibilities The Executive committee shall: (a) Implement the decisions of the Full Conference;

(b) Represent the ITTC between Confer-ences;

(c) Replace members of technical commit-tees or groups as necessary between Confer-ences;

(d) Accept new member organizations to the ITTC;

(e) Manage the income from the ITTC membership fees and any amounts transferred from the Advisory Council.

(f) Approve the arrangements and associ-ated costs and registration fees for the Confer-ence;

(g) Prepare a report on its activities for presentation at a general session of the Con-ference.

The Executive Committee shall propose the following for approval by the Full Conference:

(a) The Executive Committee Chairman, ITTC Secretary and members and Chairmen of technical committees and groups; (b) The terms of reference of technical committees and groups;

(c) Recommended procedures and guide-lines;

(d) The host organization for the next Con-ference;

(e) A financial plan and the ITTC member-ship fee.

In order to pursue the aims of the ITTC the Executive Committee may initiate formal inter-actions or collaborations between the ITTC and other organizations (for example the IMO or ISSC). The Executive Committee may require technical committees to carry out specific tasks in support of such interactions.

Votes by the Executive Committee may be

by a show of hands or secret ballot at the call of the Chairman. A simple majority shall carry a motion.

477477

The Executive Committee shall meet at

least three times between Conferences.

6.2. Membership

The Executive Committee shall normally consist of seven full-voting members including the Chairman.

(a) There shall be one representative from each of the six geographic areas listed in Annex A. (the area representative). Where at all pos-sible, the area representative shall represent a member organization of the Advisory Council. The Executive Committee may approve excep-tions to the area representative being from a member organization of the Advisory Council. Each area representative shall normally serve for two terms of three years each. The area rep-resentatives shall be appointed at least one-half year prior to the Conference by the member organizations of that area. Each region shall decide on its own procedure for selection (elec-tion) of its area representative.

(b) The Chairman of the next Executive Committee shall be appointed by the Full Con-ference at the end of the Conference and act as Chairman until the end of the next Conference. The Chairman of the Executive Committee is usually the designated representative of the member organization that will host the next Conference, but the Executive Committee may propose as its Chairman the designated repre-sentative of any member organization in the area where the next Conference will be held. The Vice Chairman of the Executive Com-mittee shall be elected by the Executive Com-mittee from its members. In the absence of the Chairman, meetings of the Executive Commit-tee shall be conducted by the Vice Chairman.

The following shall be ex-officio non-voting members of the Executive Committee:

(a) The Chairman of the Advisory Council (b) The ITTC Secretary

(c) The past Chairman of the Executive Committee. If the past Chairman is the repre-sentative of a geographic area then that person shall be a full voting member of the Executive Committee.

(d) The Conference Organizer, if that per-son is not a member of the Executive Commit-tee.

The Executive Committee Secretary shall

be proposed by Chairman of the Executive Committee for approval by the Executive Committee and shall normally serve for the term of one Conference.

The Executive Committee Secretary shall

work in support of the Executive Committee and carry out duties assigned by the Executive Committee. The duties may include work relat-ing to the organization of the next Conference such as making detail arrangements for the Conference, editing and publishing the pro-ceedings and communicating with member organizations concerning the Conference.

7. ADVISORY COUNCIL 7.1. Roles and responsibilities

The Advisory Council proposes to the Ex-ecutive Committee the topics that should be addressed by the ITTC, bearing in mind that the primary aim of the ITTC is to solve technical problems of importance to towing tank organi-zations. It proposes new specialist committees and recommends terms of reference for all technical committees based on input from tech-

478478

nical committees, ITTC members at large and the expertise and priorities of Advisory Council members.

The Advisory Council proposes recom-

mended procedures and guidelines to the Ex-ecutive Committee based on proposals by tech-nical committees and groups.

The Advisory Council may provide advice

or recommendations to the Executive Commit-tee on any other topics agreed by the Chairmen of the Executive Committee and Advisory Council.

The Advisory Council may set up mecha-

nisms to support and monitor the work of Technical Committees. The Advisory Council may communicate with technical committees through the ITTC Secretary.

Votes on matters other than the appointment

of the Chairman or Vice Chairman may be by a show of hands or secret ballot at the call of the Chairman. A simple majority shall carry a mo-tion.

The Advisory Council shall meet at least

three times between Conferences at times and places coordinated with meetings of the Execu-tive Committee.

7.2. Membership The Executive Committee appoints mem-

bers to the Advisory Council. Applications for membership shall be made to the Executive Committee through the area representative. Each such organization must satisfy the Execu-tive Committee that:

(a) The purpose of the organization is the prediction of performance of marine vehicles, marine structures and marine installations. The organization provides information, on a fee-for-service basis, to clients who are the designers,

builders, owners or operators of these assets. The work is directed and executed by full time professional staff. The organization may also conduct research, technology development, and education activities, provided the funding for these is secondary to its client revenue. (b) It has a long history of work in support of the ITTC as evidenced by membership of Committees and Groups, providing data in support of committee and group work, or mak-ing written contributions to committees and groups;

(c) It operates at least two model test facili-ties and has the capability of performing a vari-ety of experimental and numerical investiga-tions within the scope of the ITTC.

No limit shall be put on the total number of

members. However, the Executive Committee shall confirm the membership of each member of the Advisory Council once every six years. In order to remain a member of the Advisory Council members must demonstrate to the Ex-ecutive Committee that they meet the criteria (a), (b) and (c) and that in addition, they have had a record of regular attendance at meetings of the Advisory Council and the Full Con-ference and have made meaningful contri-butions to the Advisory Council. Half the Ad-visory Council member organizations shall be confirmed every three years. The Advisory Council shall recommend the process for con-firmation to the Executive Committee.

Member organizations appointed to the Ad-

visory Council shall be represented on the Ad-visory Council by their designated ITTC repre-sentative. In the event of the designated repre-sentative being unable to attend a meeting, the member organization may send an alternate who shall be a senior technical member of the management of the member organization, able to contribute to technical discussions on hydro-

479479

dynamic testing, numerical modelling and full-scale measurement.

Each member of the Executive Committee

shall be an ex-officio member of the Advisory Council if he/she is not already a member in his own right as a representative of a member or-ganization.

The Chairman and Vice Chairman of the

Advisory Council shall be elected by its mem-bers between one year and one-half year prior to the next Conference. The election shall be by secret ballot, the candidate with the maximum number of votes shall be elected. The Chairman shall take office immediately following the end of this Conference. In the absence of the Chairman, the meetings of the Advisory Coun-cil shall be conducted by the Vice Chairman.

Secretarial support to the Advisory Council

shall be provided by the ITTC Secretary.

7.3. Advisory Council fee

Advisory Council members shall pay a fee to provide sufficient money to cover the cost of the additional workload on the ITTC Secretary of performing secretarial duties directly for the Advisory Council. The fee shall be approved by the Advisory Council and paid at the same time and under the same conditions as the ITTC membership fee. The Advisory Council shall be responsible for managing the income from the Advisory Council fee.

8. TECHNICAL COMMITTEES 8.1. Roles and responsibilities

The technical committees carry out the technical work of the ITTC defined in their terms of reference. The results shall be docu-

mented in reports published in the proceedings of the Conference.

The technical committees shall consist of

two types. One type on "general subject areas" (general committees), such as: Resistance, Pro-pulsion and Manoeuvring, shall be continuing committees. The other type on "special subject areas" (specialist committees), such as Water-jets, where a specific technical problem needs to be addressed shall be of limited duration.

Technical committees shall develop detailed

plans in accordance with their terms of refer-ence. The work of all technical committees shall be directed towards the techniques and understanding of physical and numerical mod-elling as a means of predicting full-scale behav-iour. While maintaining an awareness of pro-gress, fundamental theoretical studies and fun-damental aspects of numerical fluid compu-tation shall be covered by other forums, such as the ONR Symposium on Naval Hydrodynamics or Conference on Numerical Ship Hydro-dynamics.

Technical committees may contact member

organizations to request assistance (for exam-ple, by completing a questionnaire, participat-ing in comparisons of the results of experi-ments or calculations or providing other infor-mation) or accept offers assistance from mem-ber organizations or individuals to help them carry out their work. Written contributions to the program of work of a technical committee may be submitted to its Chairman by any member organization or individual. The techni-cal committee may include a short abstract of any such contribution in its report, with an in-dication of the source from which the full document may be obtained. The conclusions and recommendations published in the commit-tee report are the sole responsibility of the committee.

480480

A technical committee may make informal contact with technical committees of other or-ganizations which may be working in areas of interest to the ITTC committee.

The report of a technical committee shall

reflect the opinion of the complete committee. If the committee is unable come to a consensus, the different opinions of committee members shall be published. The length, structure and format of the report shall be in accordance with guidelines set by the Executive Committee. The conclusions shall be structured into two separate parts:

(a) General technical conclusions; (b) Recommendations to the Full Confer-ence to adopt new or revised Recommended Procedures or Guidelines.

Technical committees may make proposals for future work in the subject area covered by the committee. Such proposals shall be com-municated to the Advisory Council through the ITTC Secretary.

Reporting schedules for the technical com-

mittees shall be set by the Executive Commit-tee and communicated by the ITTC Secretary.

Technical committees shall meet no more

than four times between Conferences.

8.2. Membership

Each technical committee shall normally consist of not more than eight members, includ-ing the Chairman. The Chairman and members shall in all cases be selected for their personal contributions to, interest in, and ability to con-tribute to the subject area of that technical committee. Formal qualifications and a bal-anced geographic representation shall also be considered in the selection process. The organi-

zation sponsoring the candidate must have agreed to support the candidate financially in carrying out his/her committee work and travel to committee meetings.

For general committees, each geographic

area shall be allowed to present to the Execu-tive Committee a "curriculum vitae" of only one candidate for each committee. Once the first six positions have been successfully filled, each geographic area that wishes to nominate an additional candidate may put a "curriculum vitae" of another candidate forward for the two remaining positions on that technical commit-tee. For specialist committees each geographic area may nominate any number of candidates.

The membership of each technical commit-

tee shall be reviewed by the Full Conference at intervals of not more than three years. A person shall not serve on technical committees for more than a total of four three-year terms, and shall not be a member on any one technical committee for more than three terms.

A member of a technical committee who is

unable to continue in committee work shall be replaced according to the following guidelines:

(a) Where possible a suitable candidate should be found from the geographical area of the member to be replaced; (b) The Area representative for that area shall, after consultation with the Chairman of the technical committee, recommend the name of the replacement member to the Executive Committee;

(c) If a suitable replacement cannot be found from the area of the member to be re-placed, the Executive Committee shall solicit proposals from other area representatives. The Executive Committee shall consult with the Chairman of the Technical Committee on the

481481

suitability of candidates for the work of the Committee;

(d) The Executive Committee shall appoint the new technical committee member.

9. GROUPS

The Executive Committee may establish groups to carry out specific non-technical tasks for the ITTC. Examples of groups are the Sym-bols and Terminology Group and the Quality Control Group. Groups may have fewer mem-bers than the technical committees. Member-ship on a group shall normally not exceed three consecutive terms of three years, but the Ex-ecutive Committee may make exceptions. Groups shall be disbanded upon completion of their tasks. Groups shall meet no more than four times between Conferences.

10. SERVING IN MORE THAN ONE CAPACITY

No person shall serve in more than one offi-

cial capacity, or on more than one technical committee, at the same time. The official ca-pacities are:

(a) Membership of the Executive Commit-tee; (b) Chairman of the Advisory Council;

(c) Chairman of a technical committee or group.

A member of the Executive Committee or

the Chairman of the Advisory Council shall not also be a member of a technical committee or group except for short periods of time at the expressed recommendation of the Executive Committee.

11. ITTC SECRETARY 11.1. Roles and responsibilities

The ITTC Secretary shall undertake all ad-ministrative and secretarial tasks in support of the operation of the ITTC except those specifi-cally assigned by the Executive Committee to the Executive Committee Secretary.

The duties of the ITTC Secretary may in-

clude maintaining lists of ITTC memberships, publishing the ITTC Newsletter and maintain-ing the ITTC website. The ITTC Secretary pro-vides secretarial support to the Executive Committee and the Advisory Council and is the primary point of contact for communications within the ITTC and between outside organiza-tions and the ITTC.

The ITTC Secretary shall be responsible for

the administration of ITTC funds. The ITTC Secretary shall:

(a) set up a bank account for ITTC funds; (b) collect ITTC membership fees and Ad-visory Council fees;

(c) make separate records of income and expenditure for the ITTC membership fees and AC fees;

(d) prepare proposed budgets and financial reports for the Executive Committee and Advi-sory Council;

(e) make authorized withdrawals from the account.

11.2. Selection of the ITTC Secretary

The ITTC Secretary shall be employed by

or in the case of a retiree, directly supported by a member organization of the Advisory Council

482482

which undertakes to provide necessary services such as office space, Fax, email etc. (The host organization for the ITTC Secretary). The Sec-retary shall have experience as a representative on the Advisory Council or as a member of a technical committee.

The Advisory Council shall give the name

of a qualified person willing to become ITTC Secretary for the next ITTC period to the Ex-ecutive Committee between one year and one-half year prior to the next Conference. The name shall be chosen by secret ballot and the candidate with the maximum number of votes shall be passed on to the Executive Committee. The Executive Committee shall propose the ITTC Secretary to the Full Conference for ap-pointment.

The ITTC Secretary shall normally serve

for two terms. The remuneration for the ITTC Secretary

shall be decided by the Executive Committee.

12. MANAGEMENT OF ITTC FUNDS

The Executive Committee shall be responsi-ble for the management of income from ITTC membership fees and funds transferred to it by the Advisory Council. Income from member-ship fees shall be used to cover the costs of the ITTC organization, including the remuneration of the ITTC Secretary, part of the cost of pro-ducing the proceedings of the Conference and other costs approved by the Executive Commit-tee.

The Advisory Council shall be responsible

for the management of income from Advisory Council membership fees. Income from the Advisory Council fee shall be transferred to the Executive Committee to cover the cost of the additional workload on the ITTC Secretary of performing secretarial duties directly for the

Advisory Council and the cost of any other activities approved by the Advisory Council for implementation by the Executive Committee.

Once each year the Executive Committee

and Advisory Council shall review and approve budgets for the money for which they have responsibility. The budgets shall show actual income and expenditures to date, including any balance or deficit remaining from previous ITTC periods, and income and expenditures planned for the remainder of the current ITTC period.

The ITTC Secretary shall set up a separate

bank account for ITTC funds. The ITTC mem-bership fees and Advisory Council fees may be kept in the same bank account (the ITTC Ac-count), but the ITTC Secretary must maintain separate records of the income and expendi-tures of money from both sources. Withdrawals from the account shall be made only by the ITTC Secretary with the written authority of the Chairman of the Executive Committee or the Chairman of the Advisory Council as ap-propriate. Cheques must be co-signed by the senior financial officer of the host organization for the ITTC Secretary.

A financial report shall be included in the

Executive Committee Report to the Con-ference. The Executive Committee shall also present an outline financial plan for the upcom-ing period including a proposal for the ITTC membership fee, for approval by the Full Con-ference. The financial reporting period for the ITTC is from October 1st in the year of the Conference to September 30th in the year of the next Conference.

13. THE CONFERENCE

The Conference shall be held at three-year intervals.

483483

Invitations from organizations to host the Conference of the next interval must be sent to the Executive Committee, through the area representative, at least one year before the Con-ference of the current interval.

The Executive Committee and the Full Con-

ference shall ensure a balanced rotation of the Conference venue among the six geographic areas. Each area shall decide on its own proce-dure for the rotation of venue among the coun-tries in the area.

The host organization for the Conference

may be either an ITTC member organization or an association whose mandate or aims are rele-vant to the aims of the ITTC, such as the American Towing Tank Conference, the Soci-ety of Naval Architects of Japan or the British Marine Hydrodynamics Panel.

The host organization shall have overall re-

sponsibility for the organization of the Confer-ence.

When the host organization is an ITTC

member, the Conference organizer shall be the designated representative of the host organiza-tion. When the host is a local association, the Conference organizer shall be the designated representative of an ITTC member organization chosen by the association.

The Conference organizer shall be respon-

sible for the detailed arrangements for the Con-ference including the preparation and publica-tion of the Conference proceedings.

The arrangements, associated costs and reg-

istration fees for the Conference must be pro-posed by the host organization for approval by the Executive Committee.

Participation in the Conference is by invita-

tion only. The host organization shall invite designated representatives of ITTC member

organizations and members of technical com-mittees and groups to the Conference. The host organization may also invite observers and sen-iors to attend. The names of observers shall be proposed by their area representative. Seniors are persons now retired who have had a long association with the ITTC and whose at-tendance is proposed by their area repre-sentative and endorsed by the Executive Com-mittee. The host organization shall offer re-duced registration fees to seniors.

13.1. Conference arrangements

The Conference shall include general and technical sessions

General Sessions shall include discussion of

the report of the Executive Committee and presentations of proposals from the Executive Committee for decisions by a vote of the Full Conference. The agenda and decision record of the general sessions shall be published in the proceedings of the Conference. General Ses-sions shall be chaired by the Chairman of the Executive Committee.

Technical sessions shall discuss the reports

and recommendations of the technical commit-tees. No discussion shall be permitted that is not directly related to the report and recom-mendations under consideration. The Confer-ence proceedings shall not be used as vehicles for disseminating technical papers. Technical sessions shall be chaired by members of the Executive Committee or Advisory Council.

The Conference may also include group

discussions, to provide opportunity for dis-cussion of topics of current interest to mem-bers. The Advisory Council shall propose top-ics for group discussions to the Executive Committee. The Executive Committee shall choose suitably qualified individuals to organ-ize and Chair the group discussions. A sum-

484484

mary of the discussion shall be published in the proceedings.

Designated representatives, members of

technical committees and groups, observers and seniors may participate in discussions at tech-nical sessions (including submitting written discussion) and in group discussions. Desig-nated representatives may submit written dis-cussion on behalf of colleagues from their or-ganization. Presentation of written discussion during the technical session shall be at the dis-cretion of the session chairman. Only desig-nated representatives of member organizations may participate in discussions at general ses-sions.

14. COMMUNICATIONS

The Executive Committee shall regularly communicate with member organizations on activities relating to the work of the Executive Committee, the Advisory Council and technical committees and groups and other any the mat-ters judged by the Executive Committee to be of concern to ITTC member organizations. The communications may be through the use of a web site, the publication of a newsletter or any other means chosen by the Executive Committee.

Member organizations may bring issues to

the attention of the ITTC through their area representative. Members of the Advisory Council may do so at a meeting of the Advisory Council.

14.1. ITTC Website

There shall be only one ITTC web site. The ITTC Secretary shall maintain the site.

The ITTC website shall provide access to:

(a) Membership information, rules, pro-cedures and guidelines, and the archive of Con-ference proceedings; (b) Information relating to the upcoming Conference, including location, hotels, travel, technical and social programs, and committee reports, and other documentation for discussion at the Conference.

14.2. ITTC Newsletter

A newsletter may be used to communicate with member organizations. The newsletter shall be published twice a year. It shall be ed-ited and produced by the ITTC Secretary. The newsletter may be published in paper or elec-tronic form.

485485

ANNEX A

Geographic areas

Area Countries Included Americas Argentina,

Brazil, Canada, Chile, Ecuador, Mexico, USA, Venezuela

Central Europe Austria, Belgium, Germany, The Netherlands, United Kingdom

East Asia China, Korea

Northern Europe Denmark, Finland, Norway, Poland, Russia, Sweden

Pacific Islands Australia, India, Indonesia, Japan, Malaysia

Southern Europe Bulgaria, Croatia, France, Greece, Iran, Israel, Italy, Portugal, Romania, Spain, Turkey

ANNEX B

Notes on the organization and operation of the ITTC

This Annex provides information to help

new members of ITTC or members joining committees for the first time, understand the workings of the ITTC organization. It includes background information, explains some rules in more detail than is appropriate in the formal rules document and includes brief descriptions of current practice.

The Annex is supplementary to the Rules,

and does not take the place of the Rules. In case of a perceived conflict between this Annex and the Rules, the Rules shall be followed.

Definitions In previous versions of the Rules and collo-

quially, the words ‘International Towing Tank Conference’, its initials, ITTC, and shortened form, ‘Conference’ have been used to mean different things depending on the context. The present Rules attempt to avoid this confusion by using these words with specific meanings:

The four letters ITTC means the associa-

tion of towing tank organizations which func-tions according to these rules.

The Conference means the tri-annual meet-

ing of ITTC member organizations. The Full Conference means the repre-

sentatives of member organizations with au-thority to vote.

486486

The name “International Towing Tank Conference” is not used in the Rules except as the title. In other documents it may be used to mean the organization (ITTC) or the tri-annual meeting (the Conference), depending on the context.

In addition the following three words are

used in the Rules:

Shall: Conveys commitment to doing some-thing. Of these three words, shall is used in the Rules in paragraphs which describe the opera-tion of the ITTC. Must: The action is mandatory; there are no al-ternatives; gives emphasis; stronger than shall. May: The action is optional; the choice is up to the person performing the action.

These definitions are also used in this An-nex.

Brief history of the ITTC In 1933 23 representatives of tanks from 10

countries including the superintendents of 9 tanks met in The Hague, to “confer in an open and confidential manner on their own methods and also on the manner of publication of tank results.” The program for the new Conference of Ship Tank Superintendents was focused on the everyday business of tanks. The conference appointed a committee to work out “in a more definite way the general conclusions.” This was the forerunner of the technical committees we have today. All decisions were made by all those present at the “the conference”.

This simple organization continued until

1948, when a Standing Committee of six re-gional members was formed to give continuity from one Conference to another. it later became the Executive Committee. Up until 1948 indi-

vidual Conference attendees made presenta-tions, but from 1948 discussions at the Confer-ence were based on reports of the technical committees. This continues to be the structure of the ITTC Conferences.

As the size and number of topics considered

by the ITTC increased, there was concern that the ITTC should not evolve into a diffuse or-ganization loosely concerned with ship hydro-dynamics. In addition, the member organiza-tions whose primary business was model test-ing for clients were worried that they would be outnumbered by the tanks operated by educa-tional and research institutions which did not share the same responsibilities to customers. There was a possibility that ITTC might adopt procedures and policies that would be harmful to the relationships between the more commer-cial facilities and their customers. The Advi-sory Council was formed at the 13th ITTC in 1972 in response to these concerns. The pur-pose of the Advisory Council was (and still is) to recommend the subjects to be considered “bearing in mind the primary aim of the con-ference is to solve technical problems of impor-tance to tank superintendents.” Organizations represented on the Advisory Council were se-lected from member organizations which met criteria chosen to show that their primary busi-ness was model testing for clients. In many ways the Advisory Council represents the community of Tank Superintendents which first met over 80 years ago.

Aims of the ITTC

The Aims of the ITTC written in the Rules

have changed very little from the aims of the ITTC expressed in the first meetings over 70 years ago. Over the years they have been re-vised to keep them up to date by including nu-merical modelling and full-scale trials and work done by ITTC members on marine instal-lations other than ships. The aims include stimulating relevant research in hydrodynam-

487487

ics, but the exchange of information concerning research in theoretical hydrodynamics and fun-damental aspects of numerical fluid computa-tions are not included. These are covered by other forums, such as the ONR Symposium on Naval Hydrodynamics or Conference on Nu-merical Ship Hydrodynamics. The ITTC estab-lishes the need for research, encourages re-search and provides for coordination of re-search carried out by its members, but does not, as an organization, fund or carry out research. The aims are written to ensure that the ITTC continues to focus on its unique role of meeting the needs of its members for giving advice and information on full-scale performance to the designers, builders and operators of ships and marine installations based on physical and nu-merical modelling.

The ITTC Organization

Members

Members of the ITTC are organizations that

satisfy the Executive Committee that they meet the criteria for membership stated in the Rules. (The ITTC does not have individual member-ships; people participate in ITTC activities as representatives of member organizations).

Designated representatives

Designated representatives are directors or

senior officers of member organizations who have authority to bind the organization in mat-ters relating to ITTC. Each member organiza-tion has one designated representative.

Full Conference

Decision making authority for the ITTC

rests with its member organizations. The Full Conference is the collective name of the desig-nated representatives from member organiza-tions present at general sessions held during the Conference. Votes taken during general ses-

sions at the Conference are recorded as deci-sions of the Full Conference.

Executive Committee

The Executive Committee is in effect, the

‘governing body’ of the ITTC. The Chairman is usually the organizer of the next Conference and members are representatives from each of 6 geographic areas. The Executive Committee implements decisions of the Full Conference and may take actions between Conferences. The agenda of the Executive Committee in-cludes applications for membership of the ITTC, membership of technical committees, arrangements for the next Conference, financial matters and relationships with other organiza-tions.

Chairman of the Executive Committee

The Chairman of the Executive Committee

is the leader of the ITTC and Chairs general sessions at the Conference.

Advisory Council

The Advisory Council drives the technical

agenda of the ITTC. It is comprised of about 30 of the larger member organizations whose pri-mary business is model testing for clients and have had a long history of involvement with the ITTC. The Advisory Council identifies topics of importance to the ITTC, drafts terms of ref-erence for the technical committees and groups and provides ongoing support and monitoring of the technical committees as they carry out the work. It reviews proposed recommended procedures in detail, and ensures they are ap-propriate for practical application in work for clients. It reviews annual progress reports from technical committees.

488488

Advisory Council Working Groups To do its work effectively, the Advisory

Council has set up four working groups. Each working group has responsibility for a technical area of importance to the ITTC. Members of the working groups are members of the Advi-sory Council who have an expertise or particu-lar interest in the subjects covered by the group. The working groups take the lead in dealing with technical matters in their area of expertise and report at meetings of the Advisory Council. Technical Committees

Technical committees carry out the techni-

cal work of the ITTC. Members of the technical committees are chosen for their ability to carry out the work and to ensure a geographic distri-bution of membership. The Executive Commit-tee chooses the Chairmen. The scope of work is defined in the terms of reference for the com-mittee. There are two types of technical com-mittee; general technical committees are con-cerned with areas of continuing long-term im-portance to ITTC member organizations and specialist technical committees that address specific topics and are of limited duration. All the technical committees have equivalent re-sponsibilities. There is no hierarchy between technical committees. The reports of technical committees primarily contain reviews of re-search relevant to ITTC members and are not comparable in format or content with publica-tions in technical journals or at other confer-ences.

Groups

Groups are similar to technical committees

except that their work is primarily non-technical (for example symbols, quality con-trol).

The ITTC Secretary The ITTC Secretary is a central point of

contact for communications between ITTC members and to and from organizations outside the ITTC. The ITTC Secretary undertakes sec-retarial tasks in support of the operation of the ITTC except those undertaken by the Executive Committee Secretary. The duties of the Secre-tary include maintaining lists of memberships, the administration and collection of member-ship fees, publishing the ITTC newsletter, maintaining the ITTC website and preparing agenda and minutes of meetings of the Full Conference.

Executive Committee Secretary

The Executive Committee Secretary is pri-

marily concerned with the organization of the Conference. These duties are common to the organization of any conference and include making physical arrangements, setting up the registration process and arranging for publica-tion of the reports of technical committees and discussions. The Executive Committee Secre-tary also performs duties in support of the Ex-ecutive Committee, such as preparing agenda and minutes of meetings.

The Conference

The Conference is held once every three

years, usually in September. The Conference agenda is based on the presentation and discus-sion of reports of technical committees, not presentations of papers by individuals. The plenary or general sessions are the opportunity for representatives of ITTC member organi-zations to discuss and for the Full Conference to vote on recommendations from the Execu-tive Committee.

The Conference venue and host organiza-

tion are chosen to ensure a balanced rotation between geographic areas. The host organiza-

489489

tion has overall responsibility for ensuring the Conference meets the requirements of the ITTC as described in the Rules and communicated by the Executive Committee. The detailed ar-rangements for the Conference are the respon-sibility of the Conference Organizer who is the designated representative of the host organiza-tion. The Executive Committee must approve the arrangements and associated costs for the Conference. The Conference Organizer is usu-ally Chairman the Executive Committee. The rules are written to allow for the possibility that the Conference Organizer might have little ex-perience of ITTC and that a different person might chair the Executive Committee. Up to the 25th Conference this situation has never arisen.

Participation in the Conference is by invita-

tion only. Invitations are sent to all designated representatives and members of technical committees and groups. In addition area repre-sentatives may propose observers and seniors to attend. Employees of ITTC member organi-zations who are neither designated representa-tives nor members of technical committees or groups may attend the Conference as observers. Observers may also be persons with an interest in the work of the ITTC who are not affiliated with ITTC member organizations. Examples are representatives from ship designers and builders, classifications societies or other ma-rine research organizations. Representatives of commercial companies with an interest in mar-keting to ITTC members may attend the Con-ference as observers, but no provision is made at most venues for the distribution of advertis-ing material or product demonstrations.

Meetings

ITTC committees (including the Executive

Committee and Advisory Council) meet three or four times between Conferences. The cost of attending these meetings is a significant cost to committee members’ organizations and every

effort is taken to minimize them. Meetings are often scheduled to coincide with major confer-ences likely to be attended by several commit-tee members and the cost to the host is kept small by using in non-commercial facilities whenever possible. The high cost of long dis-tance air travel is distributed among members by holding meetings in different geographic areas.

ITTC Fees

Member organizations pay a membership

fee by which ITTC funds are raised. The ITTC funds are used to cover the cost of the ITTC organization, including paying for the ITTC Secretary and a proportion of the cost of pub-lishing the proceedings of the Conference. (Conference proceedings are distributed to all members, whether they attend the Conference or not). Registration fees paid by Conference attendees cover the cost of the Conference and the remainder of the cost of the Proceedings.

Decision making process

The Full Conference is the decision-making

authority for the ITTC. Decisions by technical committees, the Advisory Council and Execu-tive Committee (other than those concerning only the internal operation of these committees) are made as recommendations for adoption to the next level on the organization as follows:

1) Technical committees or groups 2) Advisory Council 3) Executive Committee 4) Vote by the Full Conference

For example, when a technical committee

recommends the adoption of a procedure, the procedure is first reviewed in detail by the Ad-visory Council. (In doing this the Advisory Council will make use of the expertise in the corresponding Advisory Council working group). If the Advisory Council supports the

490490

recommendation it will pass the procedure to the Executive Committee for consideration. If the Executive Committee also supports the rec-ommendation, (Rejection is unlikely because members of the Executive Committee are also members of the Advisory Council) it will rec-

ommend its adoption by the ITTC member organizations through a vote of the Full Con-ference.

491491

ITTC – Recommended Procedures and Guidelines

1.0-04 Page 18 of 4

Decision Making between Conferences Effective Date 2014

Revision 00

Updated / Edited by Approved

Executive Committee of the 27th ITTC

27th ITTC 2014

Date: 03/2014 Date: 09/ 2014

Table of Contents

1. purpose ................................................. 19

2. definitions ............................................. 19

3. unanimous agreement ......................... 20

4. disagreement ........................................ 20

5. RECORDING AND REPORTING OF DECISIONS ......................................... 20

5.1 Executive Committee meetings ...... 20 5.2 Full Conferences .............................. 21

492492

ITTC – Recommended Procedures and Guidelines

1.0-04 Page 19 of 4

Decision making between Conferences Effective Date 2014

Revision 00

Updated / Edited by Approved

Executive Committee of the 27th ITTC

27th ITTC 2014

Date: 03/2014 Date: 09/ 2014

Decision making between Conferences

1. PURPOSE

The Full Conference is the highest authority of the ITTC and responsible for all policies and technical matters. The Executive Committee represents ITTC between Conferences. The close cooperation with external bodies such as IMO and ISO has demonstrated the need for the Executive Committee being able to make decisions on policy and/or technical matters between Conferences. The purpose of the pre-sent document is to establish a formal proce-dure for making such decisions.

Two situations are envisaged, viz.

• Unanimous Agreement: The Execu-tive Committee members eligible to vote agree unanimously to the pro-posed decision and are supported by a majority of the Advisory Council members.

• Disagreement: No unanimous agreement is reached among the Ex-ecutive Committee members eligi-ble to vote or support is not obtained from a majority of the Advisory Council.

2. DEFINITIONS

ISSUE - means an issue that may arise at any level of the ITTC (TC, AC, EC) and has either of the following characteristics:

• A communication with a third party/organization on ITTC’s poli-cies, attitude, position or profes-sional opinion on matters which may have a bearing on ITTC’s im-age, credibility, independence or may have legal or financial implica-tions

• Has a short time frame (in the sense that it is shorter than the committee life span and the Rules based deci-sion processes)

• A situation which cannot be solved unanimously

Whenever in doubt, any member of a tech-nical committee, the Advisory Council or the Executive Committee can always address the Executive Committee for an evaluation whether a given subject shall be treated as an ISSUE.

CORRESPONDENCE - shall as regards this procedure always be by e-mail. It shall include:

493493

ITTC – Recommended Procedures and Guidelines

1.0-04 Page 20 of 4

Decision making between Conferences Effective Date 2014

Revision 00

• A clear indication of the type of re-sponse required

• A clear indication of the deadline for the response (a typical time frame shall be two weeks)

VOTES - can be given as YES, NO and ABSTAIN. Decision will be made by simple majority according to the ITTC Rules.

3. UNANIMOUS AGREEMENT

In case a unanimous agreement is reached among the Executive Committee members eli-gible to vote and the support is obtained by a majority of the Advisory Council members, the decision is executed.

4. DISAGREEMENT

In case a unanimous agreement is not reached, a task force shall be established, com-prising

• Advisory Council Chairman

• Chairman of the relevant Advisory Council Working Group (re. 1.0-03)

• Chairman of the relevant technical committee

The task force is chaired by the Advisory Council Chairman and shall, as its first action, produce an ISSUE note. This is a maximum two page description of the ISSUE with the following contents:

• Problem description

• Differences

• Task force views

• Recommendations

• Appendices (technical)

The ISSUE note shall be distributed to the Executive Committee and Advisory Council members.

Within a period of two weeks from receiv-ing the ISSUE note, the Advisory Council members shall send an OPINION note to their respective Area Representative in the Execu-tive Committee. This note shall be brief and clear and may in its simplest form just be a vote, Yes, No, or Abstain.

Based on the OPINION notes, the Area Representative produces a POSITION note, which he submits to the Chairman of the Ex-ecutive Committee within a further time frame of two weeks. The POSITION note shall con-tain the views of the Advisory Council mem-bers from that area and a recommendation. Regardless of the extent of responses from the Advisory Council members, the Area Repre-sentative shall formulate the Area position on the ISSUE.

The Executive Committee decides on the ISSUE by simple majority and the appropriate action is taken to execute the decision.

5. RECORDING AND REPORTING OF DECISIONS

5.1 Executive Committee meetings

All decisions made by this procedure, whether unanimously or by the procedure de-scribed in Section 4, shall be recorded in the Minutes of the Executive Committee meetings.

494494

ITTC – Recommended Procedures and Guidelines

1.0-04 Page 21 of 4

Decision making between Conferences Effective Date 2014

Revision 00

5.2 Full Conferences

All decisions made by this procedure, whether unanimously or by the procedure de-scribed in Section 4, shall be reported to the Full Conference in the Executive Committee report and shall be recorded in the Proceedings of the Conference.

495495

Appendix 5

Member organisations

Argentina Universidad de Buenos Aires Departamento de Ingeniería Naval Paseo Colón 850 Buenos Aires C1063ACV Argentina Att. Prof. M. A. Colpachi Tel.: +54 11 4343 0891/2775 Fax: +54 4345 726 Email: mcolpac@fi.uba.ar; dindir@fi.uba.ar URL: http://www.fi.uba.ar Australia Australian Maritime College National Centre for Maritime Engineering and Hydrodynamics, University of Tasmania Locked Bag 1395 Launceston, Tasmania 7250 Australia Att. Prof. Neil Bose Tel.: +61 3 6335 4403 Fax: +61 3 6335 4720 Email: n.bose@amc.edu.au URL: http://www.amc.edu.au

Austria Schiffbautechnische Versuchsantstalt in Wien Brigittenauerlände 256 A-1200 Wien Austria Att. Prof. Gerhard Strasser Tel.: +43 1 330 3732 Fax: +43 1 332 9385 Email: prof.dr.g.strasser@sva.at URL: http://www.sva.at Belgium University of Liege - ANAST Department ArGEnCo - Sector: TLU+C Instutut du Genie Civil Bat. B52/3 (Niv. +1) Chemin des Chevreuils 1 B-4000 Liege 1 Belgium Att. Ass. Prof. HAGE André Tel.: +32 4 366 9225 Mobile: +32 479 958 585 Fax: +32 4 366 9133 Email: ahage@ulg.ac.be URL: http://www.ulg.ac.be/anast

496496

Universiteit Gent & Flanders Hydralulic Research Towing Tank for Manoeuvres in Shallow Water Ugent - Maritime Technology Division Technologiepark - Zwijnaarde 904 B-9052 Gent Belgium Att. Prof. Marc Vantorre Tel.: +32 9 264 5555 Fax: +32 9 264 5843 Email: Marc.Vantorre@ugent.be URL: http://maritiem.ugent.be Brazil Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT Centro de Engenharia Naval e Oceânica Av. Prof. Almeida Prado, 532, Cidade Universitária Butantã, São Paulo, SP CEP: 01333-030 Brazil Att. Dr. Carlos Daher Padovezi, Director Tel.: +55 11 37674729 Fax: +55 11 37674051 Email: padnaval@ipt.br URL: http://www.ipt.br LabOceano - Brazilian Ocean Technology Laboratory Parque Tecnológico - Quadra 07 Ilha do Fundão - Cidade Universitária Caixa Postal: 68508 CEP: 21945 - 970 Rio de Janeiro Brazil Att. Prof. Antonio Carlos Fernandes Tel.: +55 21 38676771 Fax: +55 21 25628715 Email: acfernandes@peno.coppe.ufrj.br URL: http://www.laboceano.coppe.ufrj.br

LabOceano - Brazilian Ocean Technology Laboratory Parque Tecnológico - Quadra 07 Ilha do Fundão - Cidade Universitária Caixa Postal: 68508 CEP: 21945 - 970 Rio de Janeiro Brazil Att. Marcelo Neves Tel.: +55 21 38676771 Fax: +55 21 25628715 Email: masn@peno.coppe.ufrj.br URL: http://www.laboceano.coppe.ufrj.br Bulgaria Bulgarian Ship Hydrodynamics Centre William Froude Str. 1 Kv. Asparuhovo, P.O. Box 58 9003 Varna Bulgaria Att. Prof. Dr. Rumen Kishev Tel.: +359 52 370 500 Fax: +359 52 370 514 Email: r.kishev@bshc.bg URL: http://www.bshc.bg Canada Memorial University of Newfoundland Ocean Engineering Research Centre Faculty of Engineering and Applied Science St. John's, NF A1B 3X5 Canada Att. Dr. Wei Qiu Tel.: +1 709 737 8970 Fax: +1 709 737 2116 Email: qiuw@mun.ca URL: http://www.engr.mun.ca/research/ocean.php

497497

National Research Council of Canada Institute for Ocean Technology Arctic Avenue P.O. Box 12093 St. John's, Newfoundland A1B 3T5 Canada Att. Dr. James Millan Tel.: +1 709 772 2472 Fax: +1 709 772 2462 Email: james.millan@nrc-cnrc.gc.ca URL: http://www.nrc.ca China China Ship Scientific Research Centre (CSSRC) P.O. Box 116 Wuxi, Jiangsu 214082 China Att. Prof. Baoshan WU Tel.: +86 510 8555 5299 Fax: +86 510 8555 5725 Email: wbscssrc@163.com URL: http://www.cssrc.com.cn Dalian University of Technology School of Naval Architecture and Ocean Engineering 2 Ling-gong Road Dalian 116024 China Att. Prof. Zhi Zong Tel.: +86 411 8470 7694 Fax: +86 411 8470 7337 Email: zongzhi@dlut.edu.cn URL: http://www.dlut.edu.cn

Harbin Engineering University Department of Naval Architecture and Ocean Engineering Harbin, Heilongjiang 150001 China Att. Dr. Xiongliang Yao Tel.: +86 451 8251 9900 ext. 8296 Fax: +86 451 8251 8443 Email: heu-yxl@163.com URL: http://www.hrbeu.edu.cn Huazhong University of Science and Technology Department of Naval Architecture and Ocean Engineering Wuhan, Hubei 430074 China Att. Prof. Yao Zhao Tel.: +86 27 8754 3958 Fax: +86 27 8754 2946 Email: yzhaozzz@hust.edu.cn URL: http://www.hust.edu.cn Jiangsu University of Science and Technology School of Naval Architecture and Ocean Engineering No. 2 Mengxi Road Zhenjiang 212003 China Att. Prof. Renqing Zhu Tel.: +86 511 84401133 Fax: +86 511 84421823 Email: zjcyzrq@hotmail.com URL: http://202.195.195.151/index.asp

498498

Marine Design and Research Institute of China (MARIC) 1688 Xizangnan Road Shanghai 200011 China Att. Prof. Jinbao Wang Tel.: +86 21 6315 0560 ext. 802 Fax: +86 21 6315 1167 Email: wang:jb@maric.com.cn URL: http://www.maric.com.cn Shanghai Jiao Tong University (SJTU), School of Naval Architecture, Ocean and Civil Engineering 800 Dong Chuan Road Shanghai 200240 China Att. Prof. Jianming Yang Tel.: +86 21 6293 3082 Fax: +86 21 6293 3160 Email: jmyang@sjtu.edu.cn URL: http://www.sjtu.edu.cn Shanghai Ship and Shipping Research Institute (SSSRI) 600 Minsheng Road Shanghai 200135 China Att. Professor Xiaping Chen Tel.: +86 21 5885 6638 ext. 2585 Fax: +86 21 5821 2824 Email: srd@sssri.com; xpchen603@sina.com URL: http://www.sssri.com

Wuhan University of Technology School of Transportation Yujiatou, Wuchang District Wuhan, Hubei 430063 China Att. Dr. Xiaofei Mao Tel.: +86 27 8655 1193 Fax: +86 27 8786 3980 Email: mxfzh@163.com URL: http://www.whut.edu.cn Croatia Brodarski Institute, Ship Hydrodynamics and Physical Modelling Ave. V. Holjevca 20 HR.10020 Zagreb Croatia Att. Ms. Marta Pedisic Buca Tel.: +385 1650 4102 Fax: +385 1650 4230 Email: marta.pedisic@hrbi.hr URL: http://www.hrbi.hr Denmark FORCE Technology Division for Maritime Industry Hjortekaersvej 99 2800 Kgs. Lyngby Denmark Att. Mr. Peter Kr. Sorensen Tel.: +45 72 15 77 00 Fax: +45 72 15 77 01 Email: pks@force.dk URL: http://www.force.dk

499499

FORCE Technology Division for Maritime Industry Hjortekaersvej 99 2800 Kgs. Lyngby Denmark Att. Dr. Christian Schack Tel.: +45 72 15 7700 Fax: +45 72 15 77 01 Email: crs@force.dk URL: http://www.force.dk Finland Helsinki University of Technology, Ship Laboratory P.O.Box 5300, 02015 TKK FIN-02015 TKK Finland Att. Prof. Jerzy Matusiak Tel.: +358 9 451 3480 Fax: +358 9 451 3493 Email: Jerzy.Matusiak@tkk.fi URL: http://www.tkk.fi/Units/Ship Aker Arctic Technology Inc. Sjöfaragatan 6 FI-00980 Helsinki Finland Att. Reko-Antti Suojanen Tel.: +358 10 670 2540 Fax: +358 10 670 2527 Email: reko-antti.suojanen@akerarctic.fi URL: http://www.akerarctic.fi

VTT Vuorimiehentie 3, Espoo P.O. Box 1000 FIN-02044 VTT Finland Att. Dr. Seppo Kivimaa Tel.: +358 20 722 6223 Fax: +358 20 722 7053 Email: seppo.kivimaa@vtt.fi URL: http://www.vtt.fi France DGA Hydrodynamics Chaussée du Vexin BP 510 F-27105 Val de Reuil France Att. Dr. Roland Joannic Tel.: +33 2 3259 7701 Fax: +33 2 3259 7702 Email: roland.joannic@intradef.gouv.fr URL: http://www.bassin.fr École Centrale de Nantes Laboratoire de Mécanique des Fluides 1 Rue de la Noe, B.P. 92101 F-44321 Nantes Cedex 3 France Att. Dr. Pierre Ferrant Tel.: +33 2 4037 1631 Fax: +33 2 4037 2523 Email: Pierre.Ferrant@ec-nantes.fr URL: http://www.ec-nantes.fr

500500

Germany Hamburgische Schiffbau Versuchsanstalt GmbH (HSVA) Bramfelder Strasse 164 D-22305 Hamburg Germany Att. Dipl.-Ing. Juergen Friesch Tel.: +49 40 6920 3216 Fax: +49 40 6920 3345 Email: Friesch@hsva.de URL: http://www.hsva.de Schiffbau Versuchsanstalt Potsdam GmbH Marquardter Chaussee 100 D-14469 Potsdam Germany Att. Dr. Christian Masilge Tel.: +49 331 567 1244 Fax: +49 331 567 1249 Email: masilge@sva-potsdam.de URL: http://www.sva-potsdam.de Development Centre for Ship Technology and Transport Systems Oststrasse 77 D-47057 Duisburg Germany Att. Dr.-Ing. Cornel Thill Tel.: +49 203 993 6920 Fax: +49 203 361 373 Email: thill@dst-org.de URL: http://www.dst-org.de

Technische Universität Berlin Fachgebiet Dynamik Maritimer Systeme Institut für Land- und Seeverkehr - Bereich Schiffs- und Meerestechnik Sekr. SG 17 Salzufer 17-19 10587 Berlin Germany Att. Prof. Dr.-Ing. Andres Cura Hochbaum Tel.: +49 30 314 26010 Fax: +49 30 314 22885 Email: cura@tu-berlin.de URL: http://www.tu-berlin.de/ Greece National Technical University of Athens Department of Naval Architecture and Marine Engineering 9 Heroon Polytechniou Str., Zografou Athens, 157-73 Greece Att. Prof. George D. Tzabiras Tel.: +30 1 772 1107 Fax: +30 1 772 1036 Email: tzab@fluid.mech.ntua.gr; gregory@central.ntua.gr URL: http://www.naval.ntua.gr

501501

India Naval Science and Technological Laboratory Ministry of Defence R&D Organisation Vigyan Nagar Visakhapatnam 503 027, Andhra Pradesh India Att. Dr. P. K. Panigrahi Tel.: +91 891 2586 076 Fax: +91 891 2559 464 Email: panigrahi.pk@nstl.drdo.in URL: http://www.drdo.org/labs/nr&d/nstl/index.shtml Indonesia UPT BPPH Indonesian Hydrodynamic Laboratory Jl. Hidrodinamika Kompleks ITS, Sukolilo Surabaya 60002 Indonesia Att. Dr. Erwandi Tel.: +62 31 594 7849 Fax: +62 31 594 8066 Email: erwandi@webmail.bppt.go.id URL: www.indonesian-hydrolab.com Iran Isfahan University of Technology Subsea Research & Development Institute Isfahan 84156-83111 Iran Att. Dr. Ahmad Reza Zamani Tel.: +98 311 3912515 Fax: +98 311 3912518 Email: arzamani@cc.iut.ac.ir URL: http://subseard.iut.ac.ir

Sharif University of Technology P.O. Box 11365-9567 Azadi Ave. Tehran Iran Att. Dr. Mohammad Saeed Seif Tel.: +98 21 6600 5549 Fax: +98 21 6600 0021 Email: seif@sharif.edu URL: http://mech.sharif.edu/~mel/ Italy Centro Esperienze Idrodinamiche Marina Militare (CEIMM) Ministero Difesa Marina Via di Vallerano 149 I-00196 Roma Italy Att. Cdr. Domenico Guadalupi Tel.: +39 06 3680 6427 Fax: +39 06 3680 5773 Email: domenico.guadalupi@marina.difesa.it URL: http://www.marina.difesa.it/ceimm Centro per gli Studi di Tecnica Navale (CETENA) Via Ippolito d'Aste 5 I-16121 Genova Italy Att. Dr. Giovanni Caprino Tel.: +39 010 599 5471 Fax: +39 010 599 5790 Email: giovanni.caprino@cetena.it URL: http://www.cetena.it

502502

Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN) Via di Vallerano 139 I-00128 Roma Italy Att. Dr. Daniele Ranocchia Tel.: +39 06 5029 217 Fax: +39 06 507 0619 Email: daniele.ranocchia@cnr.it URL: http://www.insean.it Università de Genova Dipartimento di Ingegneria Navale e Tecnologie Marine (DINAV) Via Montallegro 1 I-16145 Genova Italy Att. Prof. Carlo Podenzana-Bonvino Tel.: +39 010 353 2426 Fax: +39 010 353 2127 Email: podenzana@dinav.unige.it URL: http://dinav.unige.it Università di Napoli Dipartimento di Ingegneria Navale Via Claudio 21 I-80125 Napoli Italy Att. Prof. Carlo Bertorello Tel.: +39 081 760 3700 Fax: +39 081 239 0380 Email: bertorel@unina.it URL: http://www.din.unina.it

Università di Trieste Dipartimento di Ingegneria Navale, del Mare e per l'Ambiente (DINMA) Via A. Valerio 10 I-34127 Trieste Italy Att. Prof. Alberto Francesutto Tel.: +39 040 676 3404 Fax: +39 040 676 3443 Email: francesc@univ.trieste.it URL: http://www.dinma.univ.trieste.it Japan Akashi Ship Model Basin Co. Ltd. 3-1, Kawasaki-cho Akashi-City, 673-0014 Japan Att. Director, Dr. Yasunroi Iwasaki Tel.: +81 78 922 1200 Fax: +39 78 922 1205 Email: iwasaki_y@asmb.co.jp URL: http://www.asmb.co.jp Akishima Laboratories (Mitsui Zosen) Inc. 1-1-50, Tsutsujigaoka Akishima City, Tokyo 196-0012 Japan Att. Dr. Hiroyuki Nakagawa Tel.: +81 42 545 3116 Fax: +81 42 545 3113 Email: nakagawa@ak.mes.co.jp URL: http://www.mes.co.jp/Akiken/index-j.html

503503

Hiroshima University Department of Social & Environmental Engineering 1-4-1, Kagamiyama Higashi-Hiroshima 739-8527 Japan Att. Prof. Hironori Yasukawa Tel.: +81 82 424 7777 Fax: +81 82 424 7194 Email: yasukawa@naoe.hiroshima-u.ac.jp URL: http://www.naoe.hiroshima-u.ac.jp Japan Marine United Corporation Technical Research Center, Hydrodynamics Research Group 1-3 Kumozukokan-cho, Tsu-City Mie-pref. 514-0398 Japan Att. Dr. Eng. Takuya Omori Tel.: +81 59 238 6405 Fax: +81 59 238 6442 Email: omori-takuya@jmuc.co.jp URL: http://www.jmuc.co.jp/en/ Kobe University Graduate School of Marine Sciences 5-1-1 Fukoeminami, Higashinadaku Kobe 658-0022 Japan Att. Prof. Eiichi Kobayashi Tel.: +81 78 431 4541 Fax: +81 78 431 6361 Email: kobayasi@maritime.kobe-u.ac.jp URL: http://www.maritime.kobe-u.ac.jp

Kyushu University Department of Naval Architecture and Marine Systems Engineering 744 Motooka, Nishishi-ku Fukuoka 819-0395 Japan Att. Prof. Jun Ando Tel.: +81 92 802 3449 Fax: +81 92 802 2268 Email: ando@nams.kyushu-u.ac.jp URL: http://www.nams.kyushu-u.ac.jp Kyushu University Research Institute for Applied Mechanics 6-1 Kasuga-Koen, Kasuga-Shi Fukuoka 816-8580 Japan Att. Prof. Masahiko Nakamura Tel.: +81 92 583 7752 Fax: +81 92 583 7754 Email: naka@riam.kyushu-u.ac.jp URL: http://www.riam.kyushu-u.ac.jp Meguro Model Basin Naval Sytstems Research Center Technical Research and Development Institute Ministry of Defense 2-2-1 Nakameguro, Meguro-Ku Tokyo 153-8630 Japan Att. Dr. Kazuyuki Yamakita Tel.: +81 3 5721 7005 ext. 6280 Fax: +81 3 3731 6144 Email: ymkt@cs.trdi.mod.go.jp URL: http://cs.trdi.mod.go.jp

504504

Mitsubishi Heavy Industries Ltd. Nagasaki Research and Development Center 5-717-1 Fukahori-Machi Nagasaki 851-0392 Japan Att. Mr. Chiharu Kawakita Tel.: +81 95 834 2600 Fax: +81 95 834 2385 Email: chiharu_kawakita@mhi.co.jp URL: http://www.mhi.co.jp/ngsrdc/english/senpaku/senpaku_top.html Nagasaki Institute of Applied Science 536 Amiba-cho Nagasaki 851-0193 Japan Att. Prof. Shigeru Hayashita Tel.: +81 95 838 3207 Fax: +81 95 837 0491 Email: hayashita_shigeru@nias.ac.jp URL: http://www.nias.ac.jp National Maritime Research Institute 6-38-1 Shinkawa, Mitaka-City Tokyo 181-0004 Japan Att. Dr. Shotari Uto Tel.: +81 42 241 3505 Fax: +81 42 241 3053 Email: uto@nmri.go.jp URL: http://www.nmri.go.jp

National Research Institute of Fisheries Engineering Fishing Research Agency 7620-7 Hasaki Kamisu Ibaraki Ibaraki 314-0408 Japan Att. Mr. Akihiko Matsuda Tel.: +81 47 944 5929 Fax: +81 47 944 1875 Email: amatsuda@fra.affrc.go.jp URL: http://www.nrife.affrc.go.jp Osaka Prefecture University Department of Marine System Engineering 1-1 Gakuen-cho, Sakai Osaka 599-8531 Japan Att. Prof. Yoshiho Ikeda Tel.: +81 72 254 9343 Fax: +81 72 254 9914 Email: ikeda@marine.osakafu-u.ac.jp URL: http://www.marine.osakafu-u.ac.jp Osaka University Department of Naval Architecture and Ocean Engineering 2-1 Yamadaoka, Suita Osaka 565-0871 Japan Att. Prof. Masashi Kashiwagi Tel.: +81 66 879 7738 Fax: +81 66 879 7594 Email: kashi@naoe.eng.osaka-u.ac.jp URL: http://www.naoe.eng.osaka-u.ac.jp

505505

Shipbuilding Research Centre of Japan KICHIJOJI SUBARU Building 1-6-1 Kichijoji minamimachi-cho, Musashino Tokyo 180-0003 Japan Att. Dr. Yushu Washio Tel.: +81 422 40 2826 Fax: +81 422 40 2829 Email: y_washio@srcj.or.jp URL: http://www.srcj.or.jp Sumitomo Heavy Industries Ltd. Marine&Engineering Co., Ltd. 19, Natsushima-cho, YOKOSUKA Kanagawa 237-8555 Japan Att. Dr. Michio Takai Tel.: +81 46 869 1616 Fax: +81 46 869 1705 Email: mic_takai@shi.co.jp URL: http://www.shi.co.jp Tokyo University of Marine Science and Technology 2-1-6 Etchujima,Koto-ku Tokyo 135-8533 Japan Att. Prof. Kiyokazu Minami Tel.: +81 3 5245 7397 Fax: +81 3 5245 7397 Email: minami@kaiyodai.ac.jp URL: http://www.kaiyodai.ac.jp

University of Tokyo Department of Ocean Technology, Policy and Environment Graduate School of Frontier Sciences Kibanto Bldg. 6E5 Kashiwanoha 5-1-5 Kashiwa-shi, Chiba 277-8561 Japan Att. Prof. Hajime Yamaguchi Tel.: +81 4 7136 4114 Fax: +81 3 3815 8364 Email: H-yama@k.u-tokyo.ac.jp URL: http://www.k.u-tokyo.ac.jp/pros-e/otpe-e/index-e.htm Yokohama National University Department of Naval Architecture and Ocean Engineering 79-5 Tokiwadai, Hodogaya-ku Yokohama 240-8501 Japan Att. Prof. Kazuo Suzuki Tel.: +81 45 339 4086 Fax: +81 45 339 4099 Email: kaz@ynu.ac.jp URL: http://www.ynu.ac.jp Korea Hyundai Heavy Industries Co. Ltd. Hyundai Maritime Research Institute 1 Cheonha-Dong, Dong-Ku Ulsan 682-792 Korea Att. Dr. Young Sik Jang Tel.: +82 52 202 2115 Fax: +82 52 202 3410 Email: ysjang@hhi.co.kr URL: http://www.hhi.co.kr

506506

Inha University Department of Naval Architecture and Ocean Engineering 253-Yonghyun-Dong, Nam-Ku Inchon 402-751 Korea Att. Prof. Young-Gill Lee Tel.: +82 32 860 7340 Fax: +82 32 864 5850 Email: younglee@inha.ac.kr URL: http://www.naoe.inha.ac.kr Korea Research Institute of Ships and Ocean Engineering (KRISO, formerly MOERI) 1312-32 Yuseongdaero, Yuseong-gu Daejeon 305-343 Korea Att. Dr. Suak-Ho Van Tel.: +82 42 868 7100 Fax: +82 42 868 7714 Email: shvan@kriso.re.kr URL: http://www.kriso.re.kr Pusan National University Department of Naval Architecture and Ocean Engineering San 30 Changjon-Dong, Kumjong-Ku Pusan 609-735 Korea Att. Prof. Ho-Hwan Chun Tel.: +82 51 510 2341 Fax: +82 51 512 8836 Email: chunahh@pusan.ac.kr URL: http://www.pusan.ac.kr

Samsung Ship Model Basin (SSMB) Samsung Heavy Industries Co. Ltd. 103-28, Munji-dong, Yuseong-gu Daejeon, 305-380 Korea Att. Dr. Booki Kim Tel.: +82 42 865 4700 Fax: +82 42 865 4736 Email: booki.kim@samsung.com URL: http://www.shi.samsung.co.kr Seoul National University Department of Naval Architecture and Ocean Engineering 1 Gwanak-ro, Gwanak-gu Seoul 151-744 Korea Att. Prof. Jung-Chun Suh Tel.: +82 2 880 7341 Fax: +82 2 888 9298 Email: jungsuh@snu.ac.kr URL: http://naoe3.snu.ac.kr University of Ulsan School of Transportation System Engineering San 29 Moogeo-Dong, Nam-Ku Ulsan 680-749 Korea Att. Prof. Hyun-Kyoung Shin Tel.: +82 52 259 2157 Fax: +82 52 259 2677 Email: hkshin@uou.ulsan.ac.kr URL: http://www.ulsan.ehome.ac.kr

507507

Malaysia Universiti Teknologi Malaysia Marine Technology Centre (MTC) Faculty of Mechanical Engineering 81300 Skudai, Johor Malaysia Att. Prof. Engr. Dr. Ab Saman Abd Kader, Director, Marine Technology Centre Tel.: +0601 9717 5860 (HP) Fax: Email: abdsaman@utm.my samasak1968@gmail.com URL: http://www.fkm.utm.my/marine The Netherlands Maritime Research Instutute Netherlands (MARIN) P.O. Box 28 NL-6700 AA Wageningen The Netherlands Att. Dr. Bas Buchner Tel.: +31 317 493 333 Fax: +31 317 493 245 Email: b.buchner@marin.nl URL: http://www.marin.nl Delft University of Technology Department of Marine Technology Mekelweg 2 NL-2628 CD Delft The Netherlands Att. Prof. Rene H. M. Huijsmans Tel.: +31 15 278 3598 Fax: +31 15 278 1836 Email: r.h.m.huijsmans@tudelft.nl URL: http://www.3me.tudelft.nl

Norway Norwegian Marine Technology Research Institute (MARINTEK) P.O. Box 4125, Valentinlyst N-7002 Trondheim Norway Att. Dr. Kourosh Koushan Tel.: +47 73 59 5500 Fax: +47 73 59 5776 Email: kourosh.koushan@marintek.sintef.no URL: http://www.marintek.sintef.no Poland Ship Design and Research Centre (CTO S.A.) Al. Rzeczypospolitej 8 PL-80-369 Gdansk Poland Att. Dr. Leszek Wilczynski Tel.: +48 58 307 4214 Fax: +48 58 307 4212 Email: research@cto.gda.pl URL: http://www.cto.gda.pl Technical University of Gdansk Ocean Engineering and Ship Technology Narutowicza Str. 11/12 PL-80-952 Gdansk Poland Att. Prof. Marek Dzida Tel.: +48 58 472 557 Fax: +48 58 414 712 Email: sekoce@pg.gda.pl URL: http://www.pg.gda.pl/~wwwoce/WOiOSite/HTMLdocs/English/Home.htm

508508

Russia Krylov Shipbuilding Research Institute 44 Moskovskoye Shosse 196158 St. Petersburg Russia Att. Dr. A.V. Pustoshny Tel.: +7 812 127 9596 (9647) (9348) Fax: +7 812 127 9595 (9632) (9594) Email: krylov@krylov.spb.ru URL: http://www.krylov.com.ru Spain Canal de Experiencias Hidroninámicas de El Pardo (CEHIPAR) Carretera de la Sierra s/n 24048 El Pardo-Madrid Spain Att. Emilio Fajardo Tel.: +34 91 376 2101 Fax: +34 91 376 0176 Email: efajardo@cehipar.es URL: http://www.cheipar.es Escuela Técnica Superior de Ingenieros Navales (ETSIN) Universidad Politécnica de Madrid Avda Arco de la Victoria s/n 28040 Madrid Spain Att. Prof. Luis Pérez-Rojas Tel.: +34 91 336 7154 Fax: +34 91 544 2149 Email: luis.perezrojas@upm.es URL: http://www.etsin.upm.es

Sweden Rolls-Royce Hydrodynamic Research Centre P.O. Box 1010 SE-68129 Kristinehamn Sweden Att. Mr. Michael Forslund Tel.: +46 550 84299 Fax: +46 550 84470 Email: michael.forslund@rolls-royce.com URL: http://www.rolls-royce.com SSPA Sweden AB P.O. Box 24001 SE-400 22 Göteborg Sweden Att. Ms Susanne Abrahamsson Tel.: +46 31 772 9000 Fax: +46 31 772 9124 Email: susanne.abrahamsson@sspa.se URL: http://www.sspa.se Turkey Istanbul Technical University Department of Naval Architecture and Marine Engineering, Maslak Sariyer 34469 Istanbul Turkey Att. Prof. S. Bal Tel.: +90 212 285 6485 Fax: +90 212 285 6454 Email: sbal@itu.edu.tr URL: http://nutkulab.itu.edu.tr

509509

United Kingdom BMT Defense Services Ltd. 12 Little Park Road Fareham Hampshire, PO15 5SU United Kingdom Att. Dr. Tom Dinham-Peren Tel.: +44 1489 553100 Fax: +44 1489 553101 Email: tperen@bmtdsl.co.uk URL: http://www.bmtseatech.co.uk QinetiQ Maritime Platforms Haslar Marine Technology Park Haslar Road, Gosport Hampsjhire, PO12 2AG United Kingdom Att. Dr. Paul Crossland Tel.: +44 23 9233 5506 Fax: +44 23 9233 5461 Email: pcrossland@QinetiQ.com URL: http://www.qinetiq.com Cranfield University, School of Engineering Whittle Building, Ground Floor, G113 Cranfield, Bedfordshire MK43 0AL United Kingdom Att. Dr. Maurizio Collu Tel.: +44 1234 754779 Fax: Email: maurizio.collu@cranfield.ac.uk URL: www.cranfield.ac.uk

Newcastle University School of Marine Science and Technology Armstrong Building Queen Victoria Road Newcastle Upon Tyne, NE1 7RU United Kingdom Att. Prof. Mehmet Atlar Tel.: +44 191 222 8977 (5067) Fax: +44 191 222 5491 (5067) Email: mehmet.atlar@ncl.ac.uk URL: http://www.marinetech.ncl.ac.uk University of Southampton Froude Building (28) School of Engineering Sciences Southampton, SO17 1BJ United Kingdom Att. Dr. Stephen R. Turnock Tel.: +44 23 8059 2488 Fax: +44 23 8059 3299 Email: srt@soton.ac.uk URL: http://www.ses.soton.ac.uk Universities of Glasgow and Strathclyde Department of Naval Architecture and Marine Engineering 100 Montrose Street Glasgow, G4 OLZ United Kingdom Att. Prof. Atilla Incecik Tel.: +44 141 548 4093 Fax: +44 141 552 2879 Email: atilla.incecik@na-me.ac.uk URL: http://www.na-me.ac.uk

510510

USA Applied Research Labratory Pennsylvania State University P.O. Box 3-0 State College, PA, 16804-0030 USA Att. Dr. Arnold A. Fontaine Tel.: +1 814 863 1765 Fax: +1 814 865 3287 Email: aaf1@arl.psu.edu URL: http://www.arl.psu.edu Iowa Institute of Hydraulic Research University of Iowa Iowa City, IA 52242-1585 USA Att. Prof. Frederick Stern Tel.: +1 319 335 5215 Fax: +1 319 335 5238 Email: frederick-stern@uiowa.edu URL: http://www.iihr.uiowa.edu Naval Surface Warfare Center, Carderock Division David Taylor Model Basin 9500 MacArthur Blvd. W. Bethesda, MD 20817-5700 USA Att. Dr. Jon Etxegoien Tel.: +1 301 227 1578 Fax: +1 301 227 2584 Email: jon.etxegoien@navy.mil URL: http://www.dt.navy.mil

Stevens Institute of Technology Davidson Laboratory 711 Hudson Street Hoboken, NJ 07030 USA Att. Dr. Raju Datla Tel.: +1 201 216 5568 Fax: +1 201 216 8214 Email: rdatla@stevens.edu URL: http://www.stevens.edu/engineering/cms United States Naval Academy Naval Architect & Ocean Engineering Department 590 Holloway Road, Stop 11 D Annapolis, MD 21402-5042 USA Att. Prof. G.J. White Tel.: +1 410 293 6423 Fax: +1 410 293 2219 Email: greg@usna.edu URL: http://usna.edu/Hydromechanics University of Michigan Department of Naval Architecture and Marine Engineering 2600 Draper Road Ann Arbor, MI 48109-2145 USA Att. Prof. Robert F. Beck Tel.: +1 734 764 0282 Fax: +1 734 936 8820 Email: rbeck@umich.edu URL: http://www.engin.umich.edu/dept/name

511511

University of New Orleans Department of Naval Architecture and Marine Engineering 911 Engineering Building New Orleans, LA 70148 USA Att. Prof. Robert G. Latorre Tel.: +1 504 280 7180 Fax: +1 504 280 5542 Email: rlatorre@uno.edu URL: http://www.uno.edu/~engr/towtank

512512

Appendix 6

Designated Representatives(R), Committee Members (M) and Observers (O) invited to attend the 27th ITTC Conference

Name Organisation Country Category

Prof. M. A. Colpachi Universidad de Buenos Aires Argentina (R) Prof. Neil Bose Australian Maritime College Australia (R) Dr. Jonathan Duffy Australian Maritime College Australia (M) Dr. Giles Thomas Australian Maritime College Australia (M) Dr. Irene Penesis Australian Maritime College Australia (M) Dr. Brendon Anderson DSTO Australia (O) Dr. Stuart Cannon DSTO Australia (O) Ms. Liz Lakey J.P.Kenny Australia (O) Dr. Gregor Macfarlane AMC-UTAS Australia (O) Dr. Hayden Marcollo AMOG Australia (O) Associate Professor Paul Brandner

AMC-UTAS Australia (O)

Dr. Martin Renilson Renilson Marine Australia (O) Prof. Gerhard Strasser Schiffbautechnische Versuchsantstalt in Wien Austria (R) Dr. Clemens Strasser Schiffbautechnische Versuchsantstalt in Wien Austria (O) Ass. Prof. HAGE André University of Liege - ANAST Belgium (R) Prof. Marc Vantorre Universiteit Gent & Flanders Hydralulic Research

Towing Tank for Manoeuvres in Shallow Water Belgium (R)

Dr. G. Delefortrie Universiteit Gent & Flanders Hydralulic Research Towing Tank for Manoeuvres in Shallow Water

Belgium (M)

Prof. Dr. Ir. Katrien Eloot Flanders Hydraulic Research Belgium (O) Dr. Adrian Constantinescu DN&T Belgium (O) Dr. Carlos Daher Padovezi, Director

Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT

Brazil (R)

Prof. Antonio Carlos Fernandes

LabOceano - Brazilian Ocean Technology Laboratory Brazil (R)

Eduardo Tannuri Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT

Brazil (M)

Sergio Sphaier LabOceano - Brazilian Ocean Technology Laboratory Brazil (M) Joel Sena Sales Jr. LabOceano - Brazilian Ocean Technology Laboratory Brazil (M) Marcelo Neves LabOceano - Brazilian Ocean Technology Laboratory Brazil (O) Dr. James Manoel Guimarães Weiss, Director

Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT

Brazil (O)

Dr. Toshi-ichi Tachibana Instituto de Pesquisas Tecnológicas do Estado de São Paulo - IPT

Brazil (O)

513513

Andre Fujarra University of São Paulo Brazil (O) Alexandre Simos University of São Paulo Brazil (O) Prof. Dr. Rumen Kishev Bulgarian Ship Hydrodynamics Centre Bulgaria (R) Dr. Kostadin Yossifov Bulgarian Ship Hydrodynamics Centre Bulgaria (O) Dr. Wei Qiu Memorial University of Newfoundland Canada (R) Dr. James Millan National Research Council of Canada Canada (R) Greg Hermanski National Research Council of Canada Canada (M) Michael Lau National Research Council of Canada Canada (M) Heather Peng, ph.D., P.Eng. Memorial University, Faculty of Engineering and

Applied Science Canada (O)

David Murdey Canada (O) Prof. Baoshan WU China Ship Scientific Research Centre (CSSRC) China (R) Prof. Zhi Zong Dalian University of Technology China (R) Dr. Xiongliang Yao Harbin Engineering University China (R) Prof. Yao Zhao Huazhong University of Science and Technology China (R) Prof. Renqing Zhu Jiangsu University of Science and Technology China (R) Prof. Jinbao Wang Marine Design and Research Institute of China

(MARIC) China (R)

Prof. Jianming Yang School of Naval Architecture, Ocean and Civil Engineering Shanghai Jiao Tong University

China (R)

Professor Xiaping Chen Shanghai Ship and Shipping Research Institute (SSSRI)

China (R)

Dr. Xiaofei Mao Wuhan University of Technology China (R) Dr. Chenjun Yang School of Naval Architecture, Ocean and Civil

Engineering Shanghai Jiao Tong University

China (M)

Dr. Chengsheng Wu China Ship Scientific Research Centre (CSSRC) China (M) Prof. Xuefeng Wang School of Naval Architecture, Ocean and Civil

Engineering Shanghai Jiao Tong University

China (M)

Prof. Wenyang Duan Harbin Engineering University China (M) Dr. Decheng Wan School of Naval Architecture, Ocean and Civil

Engineering Shanghai Jiao Tong University

China (M)

Dr. Feng Zhao China Ship Scientific Research Centre (CSSRC) China (M) Prof. Denghai Tang China Ship Scientific Research Centre (CSSRC) China (M) Prof. Yanping He School of Naval Architecture, Ocean and Civil

Engineering Shanghai Jiao Tong University

China (M)

Prof. Qianjing Yue Dalian University of Technology China (M) Zuyuan LIU, Prof. Wuhan Transportation University China (O) Tingqiu LI, Prof. Wuhan Transportation University China (O) Keqiang CHEN, Prof. Wuhan Transportation University China (O) Xiaoming HU, Engineer Wuhan Transportation University China (O) Zhiguo ZHANG, Professor Huazhong University of Science and Technology China (O) Yanzhuo XUE, Prof. Harbin Engineering University China (O) Debo HUANG, Prof. Harbin Engineering University China (O) Chunyu GUO, Prof. Harbin Engineering University China (O)

514514

Zhuang KANG, Prof. Harbin Engineering University China (O) Zili WANG, President Jiangsu University of Science and Technology China (O) Zhiyong JIANG, Head Jiangsu University of Science and Technology China (O) Rening ZHU, Professor Jiangsu University of Science and Technology China (O) Zhidong WANG, Professor Jiangsu University of Science and Technology China (O) Daming YANG, Associate Professor

Jiangsu University of Science and Technology China (O)

Gang XU, Phd Jiangsu University of Science and Technology China (O) Zhenping WENG, Director China Ship Scientific Research Centre (CSSRC) China (O) Kai YAN, Prof. China Ship Scientific Research Centre (CSSRC) China (O) Shitang DONG, Prof. China Ship Scientific Research Centre (CSSRC) China (O) Changyun Chen, Prof Shanghai Ship and Shipping Research Institute China (O) Weimin Chen, Associate Professor

Shanghai Ship and Shipping Research Institute China (O)

Lei Yan, Associate Professor Shanghai Ship and Shipping Research Institute China (O) Ms. Marta Pedisic Buca Brodarski Institute, Ship Hydrodynamics and

Physical Modelling Croatia (R)

Mrs. Gordana Semijalac Brodarski Institute Croatia (O) Mr. Stanislav Ruzic Brodarski Institute Croatia (O) Mr. Boris Bucan Brodarski Institute Croatia (O) Mr. Peter Kr. Sorensen FORCE Technology Denmark (R) Dr. Claus Simonsen FORCE Technology Denmark (M) Dr. Anton Minchev FORCE Technology Denmark (M) Dr. Christian Schack FORCE Technology Denmark (O) Mr. Aage Damsgaard FORCE Technology Denmark (O) Jørgen Juncher Jensen Danmarks Tekniske Universitet

Institut for Mekanisk Teknologi Denmark (O)

Poul Andersen Danmarks Tekniske Universitet Institut for Mekanisk Teknologi

Denmark (O)

Ulrik Dam Nielsen Danmarks Tekniske Universitet Institut for Mekanisk Teknologi

Denmark (O)

Arne Hasle Nielsen Denmark (O) Stig Sand Denmark (O) Brian Skov Josefsen Student Denmark (O) Lasse Normann de-Boer Student Denmark (O) Mads Martinsen Student Denmark (O) Jan Runge Student Denmark (O) Audun Støme Student Denmark (O) Christian Nielsen Student Denmark (O) Henrik Mikkelsen Student Denmark (O) Nicolai Slaatto Student Denmark (O) Pelle Bo Regener Student Denmark (O) Anders Boy Student Denmark (O) Kasper Fønns Bach Student Denmark (O) Prof. Jerzy Matusiak Helsinki University of Technology, Ship Laboratory Finland (R) Reko-Antti Suojanen Aker Arctic Technology Inc. Finland (R) Dr. Seppo Kivimaa VTT Finland (R) T. Mikkola Aalto University Finland (M)

515515

I. Saisto VTT Finland (M) T. Leiviskä Aker Arctic Technology Inc. Finland (M) R. von Bock und Polach Aalto University Finland (M) Dr. Harri Soininen Finland (O) Mr. Raimo Hämälainen STX Finland Finland (O) Mr. Tomi Veikonheimo ABB Oy Finland (O) Mr. Janne Niittymäki Foreship Finland (O) Dr. Pierre Ferrant École Centrale de Nantes

Laboratoire de Mécanique des Fluides France (R)

Roland Joannic DGA Hydrodynamics France (R) Bertrand Alessandrini École Centrale de Nantes

Laboratoire de Mécanique des Fluides France (M)

Didier Fréchou DGA Hydrodynamics France (M) Pierre Emmanuel Guillerm École Centrale de Nantes

Laboratoire de Mécanique des Fluides France (M)

Jean-Marc Rousset École Centrale de Nantes Laboratoire de Mécanique des Fluides

France (M)

Aurélien Babarit École Centrale de Nantes Laboratoire de Mécanique des Fluides

France (M)

Guillaume de Garidel DGA Hydrodynamics France (O) Dipl.-Ing. Juergen Friesch Hamburgische Schiffbau Versuchsanstalt GmbH

(HSVA) Germany (R)

Prof. Dr.-Ing. Andres Cura Hochbaum

Technische Universität Berlin Germany (R)

Dr. Christian Ernst-Georg Masilge

Schiffbau Versuchsanstalt Potsdam GmbH Germany (R)

Prof. Dr. Ing. Bettar Ould el Moctar

Development Centre for Ship Technology and Transport Systems

Germany (R)

Rainer Grabert Schiffbau Versuchsanstalt Potsdam GmbH Germany (M) Dr. Uwe Hollenbach Hamburgische Schiffbau Versuchsanstalt GmbH

(HSVA) Germany (M)

Herbert Bretschneider Hamburgische Schiffbau Versuchsanstalt GmbH (HSVA)

Germany (M)

Peter Jochmann Hamburgische Schiffbau Versuchsanstalt GmbH (HSVA)

Germany (M)

Dr. Manfred Mehmel Schiffbau Versuchsanstalt Potsdam GmbH Germany (O) Dr.-Ing. Cornel Thill Development Centre for Ship Technology and

Transport Systems Germany (O)

Mr. Carsten Meier Lavision GmbH Germany (O) Dr. Thomas Rüggeberg Bundesministerium für Wirtschaft und Technologie Germany (O) Prof. Stefan Krüger TU Hamburg-Harburg, Institut für Entwerfen von

Schiffen und Schiffsicherheit Germany (O)

Mrs. Elke Proos Forscungszentrum Jülich GmbH Germany (O) Prof. Günther F. Clauss Technische Universität Berlin Germany (O) Dipl.-Ing. Karsten Rieck Technische Universität Berlin Germany (O) Michael Schmiechen Germany (O) Prof. George D. Tzabiras National Technical University of Athens Greece (R) G. Grigoropoulos National Technical University of Athens Greece (M)

516516

Dr. P. K. Panigrahi Naval Science and Technological Laboratory India (R) Prof. V. Anantha Subramanian

IIT Madras India (O)

Dr. Erwandi UPT BPPH Indonesian Hydrodynamic Laboratory Indonesia (R) Dr. Ahmad Reza Zamani Isfahan University of Technology Iran (R) Dr. Mohammad Saeed Seif Sharif University of Technology Iran (R) Cdr. Domenico Guadalupi Centro Esperienze Idrodinamiche Marina Militare

(CEIMM) Italy (R)

Dr. Giovanni Caprino Centro per gli Studi di Tecnica Navale (CETENA) Italy (R) Dr. Daniele Ranocchia Instituto Nazionale per Studi ed Esperienze di

Architettura Navale (INSEAN) Italy (R)

Prof. Carlo Podenzana-Bonvino

Università de Genova Dipartimento di Ingegneria Navale e Tecnologie Marine (DINAV)

Italy (R)

Prof. Pasquale Casella Università di Napoli Dipartimento di Ingegneria Navale

Italy (R)

Prof. Alberto Francesutto Università di Trieste Dipartimento di Ingegneria Navale, del Mare e per l'Ambiente (DINMA)

Italy (R)

Paola Gualeni Università de Genova Dipartimento di Ingegneria Navale e Tecnologie Marine (DINAV)

Italy (M)

Riccardo Broglia Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Mario Felli Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Angelo Olivieri Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Elena Ciappi Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Francesco Salvatore Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Marco Ferrando Università de Genova Dipartimento di Ingegneria Navale e Tecnologie Marine (DINAV)

Italy (M)

Lanfranco Benedetti Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (M)

Prof. Georgio Contento University of Trieste, Dept. Of Engineering and Architecture

Italy (O)

Dr. Guido Lupieri University of Trieste, Dept. Of Engineering and Architecture

Italy (O)

Dr. Fabio di Felice Instituto Nazionale per Studi ed Esperienze di Architettura Navale (INSEAN)

Italy (O)

Director, Dr. Yasunroi Iwasaki

Akashi Ship Model Basin Co. Ltd. Japan (R)

Dr. Hiroyuki Nakagawa Akishima Laboratories (Mitsui Zosen) Inc. Japan (R) Prof. Hironori Yasukawa Hiroshima University Japan (R) Dr. Eng. Takuya Omori Japan Marine United Corporation Japan (R) Prof. Eiichi Kobayashi Kobe University Japan (R)

517517

Prof. Jun Ando Kyushu University Japan (R) Prof. Masahiko Nakamura Kyushu University Japan (R) Dr. Kazuyuki Yamakita Meguro Model Basin Japan (R) Mr. Chiharu Kawakita Mitsubishi Heavy Industries Ltd. Japan (R) Dr. Shotari Uto National Maritime Research Institute Japan (R) Mr. Akihiko Matsuda National Research Institute of Fisheries Engineering Japan (R) Prof. Yoshiho Ikeda Osaka Prefecture University Japan (R) Prof. Masashi Kashiwagi Osaka University Japan (R) Dr. Yushu Washio Shipbuilding Research Centre of Japan Japan (R) Dr. Michio Takai Sumitomo Heavy Industries Ltd.

Marine&Engineering Co., Ltd. Japan (R)

Prof. Hajime Yamaguchi University of Tokyo Japan (R) Prof. Kazuo Suzuki Yokohama National University Japan (R) Professor Kiyokazu MINAMI Tokyo University of Marine Science and Technology Japan (R) Prof. Shigeru Hayashita Nagasaki Institute of Applied Science Japan (R) Dr. Hisao Tanaka Japan Marine United Corporation Japan (M) Prof. Yoshitaka Furukawa Kyushu University Japan (M) Dr. Katsuji Tanizawa National Maritime Research Institute Japan (M) Dr. Takashi Mikami Mitsui Akishima Japan (M) Prof. Toru Katayama Osaka Prefecture University Japan (M) Dr. Takanori Hino Yokohama National University Japan (M) Dr. Shigeki Nagaya IHI Japan (M) Prof. Motohiko Murai Yokohama National University Japan (M) Prof. Akahisa Konno Kogakuin University Japan (M) Dr. Chaniku Shin Nagasaki Institute of Applied Science Japan (O) Prof. Toshio Iseki Tokyo University of Marine Science and Technology Japan (O) Professor Kunihiro IKEGAMI Department of Naval Architecture, Faculty of

Engineering, Nagasaki Institute of Applied Science

Japan (O)

Mister Kei SATO Mitsubishi Heavy Industries Japan (O) Doctor Kinya TAMURA Japan (O) Professor TODA, Yasuyuki Osaka University Japan (O) Professor Naoji Toki Ehime University Japan (O) Doctor Ryo Yakushiji R&D Assessment Division, Technical Research and

Development Institute, Ministry of Defense Japan (O)

Prof. Emeritus Kuniharu NAKATAKE

Kyushu University Japan (O)

Prof.Takeshi KINOSHITA Department of Oceanic Architecture and Engineering, Ost, Nihon University

Japan (O)

Dr. Noriyuki SASAKI Maritime Technology Group, Monohakobi Technology Institute

Japan (O)

Dr. Kinya TAMURA Japan (O) Prof. Hisaaki MAEDA Japan (O) Prof. Katsuro KIJIMA President, Nagasaki Institute of Applied Science Japan (O) Prof. Emer. Hiroharu KATO Japan (O) Dr. Young Sik Jang Hyundai Heavy Industries Co. Ltd. Korea (R) Prof. Young-Gill Lee Inha University Korea (R) Dr. Suak-Ho Van Maritime and Ocean Engineering Research Institute Korea (R)

518518

(MOERI)

Prof. Ho-Hwan Chun Pusan National University Korea (R) Dr. Booki Kim Samsung Ship Model Basin (SSMB) Korea (R) Prof. Hyun-Kyoung Shin University of Ulsan Korea (R) Prof. Jung-Chun Suh Seoul National University Korea (R) Dr. Jin Kim Maritime and Ocean Engineering Research Institute

(MOERI) Korea (M)

Prof. Moon Chan Kim Pusan National University Korea (M) Dr. Sun Young Kim Maritime and Ocean Engineering Research Institute

(MOERI) Korea (M)

Prof. Young Hwan Kim Seoul National University Korea (M) Dr. Dong Yeon Lee Samsung Ship Model Basin (SSMB) Korea (M) Dr. Gyung Jung Lee Maritime and Ocean Engineering Research Institute

(MOERI) Korea (M)

Prof. Shin Hyung Rhee Seoul National University Korea (M) Prof. In Won Lee Pusan National University Korea (M) Heungwon Seo Hyundai Heavy Industries Co. Ltd. Korea (M) Dr. Gil Hwan Choi Hyundai Heavy Industries Co. Ltd. Korea (M) Dr. Seung-Myun Hwangbo Samsung Ship Model Basin (SSMB) Korea (O) Prof. Key Pyo Rhee Seoul National University Korea (O) Dr. Bong-Jun Chang Hyundai Heavy Industries Co. Ltd Korea (O) Dr. Youngjae Sung Hyundai Heavy Industries Co. Ltd Korea (O) Dr. Seong Jae Jeong National Fisheries R&D Institute Korea (O) Prof. Seung-Hee Lee Inha University Korea (O) Prof. Sang-Hyun Kim Inha University Korea (O) Prof. Sungbu Suh Dong Eui University Korea (O) Prof. Kwang Hyo Jung Pusan National University Korea (O) Prof. Nakwan Kim Seoul National University Korea (O) Dr. Se-Eun Kim Daewoo Shipbuilding & Marine Engineering Korea (O) Dr. Seung Il Yang KRISO Korea (O) Prof. Ab Saman Abd Kader Universiti Teknologi Malaysia Malaysia (R) Dr. Montasir Osman Ahmed Univeriti Teknologi Petronas, Civil Engineering

Department Malaysia (O)

Dr. Mohd Shahir Liew Univeriti Teknologi Petronas, Civil Engineering Department

Malaysia (O)

Dr. Kourosh Koushan Norwegian Marine Technology Research Institute (MARINTEK)

Norway (R)

D. Fathi Norwegian Marine Technology Research Institute (MARINTEK)

Norway (M)

H. Lie Norwegian Marine Technology Research Institute (MARINTEK)

Norway (M)

C. Muthanna Norwegian Marine Technology Research Institute (MARINTEK)

Norway (M)

Dr. Andreas Krapp Jotun A/S Norway (O) Dr. Leszek Wilczynski Ship Design and Research Centre (CTO S.A.) Poland (R) Prof. Marek Dzida Technical University of Gdansk Poland (R) W. Gorski Ship Design and Research Centre (CTO S.A.) Poland (M) M. Kraskowski Ship Design and Research Centre (CTO S.A.) Poland (M) Dr. A.V. Pustoshny Krylov Shipbuilding Research Institute Russia (R)

519519

V. Borusevich Krylov Shipbuilding Research Institute Russia (M) V. Magarovski Krylov Shipbuilding Research Institute Russia (M) K. Sazonov Krylov Shipbuilding Research Institute Russia (M) Prof. Luis Pérez-Rojas Escuela Técnica Superior de Ingenieros Navales

(ETSIN) Spain (R)

Emilio Fajardo CEHIPAR Spain (R) Luis Palao Lechuga Canal de Experiencias Hidroninámicas de El Pardo

(CEHIPAR) Spain (O)

Mr. Michael Forslund Rolls-Royce Hydrodynamic Research Centre Sweden (R) Ms Susanne Abrahamsson SSPA Sweden AB Sweden (R) D-Q- Li SSPA Sweden AB Sweden (M) Sofia Werner SSPA Sweden AB Sweden (M) V. Westerberg SSPA Sweden AB Sweden (M) Willem van Berlekom Sweden (O) Charlotta Nordenfelt Qualisys AB Sweden (O) Mr. Lars T. Gustafsson SSPA Sweden AB Sweden (O) Prof. Lars Larsson Chalmers University of Technology Sweden (O) Prof. Göran Bark Chalmers University of Technology Sweden (O) Prof. Rickard Bensow Chalmers University of Technology Sweden (O) Prof. Jakob Kuttenkeuler KTH Royal Institute of Technology Sweden (O) Frederik Gerhardt SSPA Sweden AB Sweden (O) Dr. Bas Buchner Maritime Research Instutute Netherlands (MARIN) The

Netherlands (R)

Prof. Rene H. M. Huijsmans Delft University of Technology The Netherlands

(R)

Frans Quadvlieg Maritime Research Instutute Netherlands (MARIN) The Netherlands

(M)

Pepijn de Jong Delft University of Technology The Netherlands

(M)

Frans van Walree Maritime Research Instutute Netherlands (MARIN) The Netherlands

(M)

Henk van der Boom Maritime Research Instutute Netherlands (MARIN) The Netherlands

(M)

Johan Bosschers Maritime Research Instutute Netherlands (MARIN) The Netherlands

(M)

M. van Rijsbergen Maritime Research Instutute Netherlands (MARIN) The Netherlands

(M)

Dr. M.W.C. Oosterveld The Netherlands

(O)

Prof. S. Bal Istanbul Technical University Turkey (R) Ass. Prof. Emin Korkut Istanbul Technical University Turkey (M) Takinaci Ali Can Istanbul Technical University Turkey (M) Dr. Tom Dinham-Peren BMT Defense Services Ltd. UK (R) Dr. Paul Crossland QinetiQ UK (R) Dr. Maurizio Collu Cranfield University, School of Engineering UK (R) Prof. Mehmet Atlar Newcastle University UK (R) Dr. Stephen R. Turnock University of Southampton UK (R) Prof. Atilla Incecik Universities of Glasgow and Strathclyde UK (R)

520520

Dr. Dominic Hudson University of Southampton UK (M) Prof. Longbin Tao Newcastle University UK (M) Andy Peters QinetiQ UK (M) Peter Bull QinetiQ UK (M) Dr. Ignazio Maria Viola Edinburgh University UK (M) Dr. Sandy Day Universities of Glasgow and Strathclyde UK (M) Michael Woodward Newcastle University UK (M) Steve Curtis Cussons Technology Ltd. UK (O) Shaun Ross Cussons Technology Ltd. UK (O) Dr. Reddy Devalapalli Lloyd´s Register UK (O) Dr. Alex Phillips University of Southampton UK (O) Dr. Mahdi Khorasanchi University of Strathclyde UK (O) Mr. Edward Nixon University of Strathclyde UK (O) Jennifer Norris The European Marine Energy Centre (EMEC) Ltd. UK (O) Dr. Arnold A. Fontaine Applied Research Labratory USA (R) Prof. Frederick Stern Iowa Institute of Hydraulic Research USA (R) Dr. Jon Etxegoien Naval Surface Warfare Center, Carderock Division USA (R) Dr. Raju Datla Stevens Institute of Technology USA (R) Prof. G.J. White United States Naval Academy USA (R) Prof. Robert F. Beck University of Michigan USA (R) Prof. Robert G. Latorre University of New Orleans USA (R) Dr. Roger H. Compton Webb Institute USA (R) Thomas Fu Naval Surface Warfare Center, Carderock Division USA (M) Steve Ceccio University of Michigan USA (M) David Hayden Naval Surface Warfare Center, Carderock Division USA (M) Arthur Reed Naval Surface Warfare Center, Carderock Division USA (M) Sung.Eun Kim Naval Surface Warfare Center, Carderock Division USA (M) Pablo M. Carrica Iowa Institute of Hydraulic Research USA (M) Paisan Atsavapranee Naval Surface Warfare Center, Carderock Division USA (M) Dr. Theodore Farabee Naval Surface Warfare Center, Carderock Division USA (M) Bruce Johnson USA (M) Michael G. Morabito United States Naval Academy USA (M) Joel Park Naval Surface Warfare Center, Carderock Division USA (M) Dr. Joseph T. Arcano, Technical Director

Naval Surface Warfare Center, Carderock Division USA (O)

William B. Morgan USA (O)

521521

http://www.ittc.info