annex xii final report effects of loads on asset ... · 3. critical aspect iii: transferability of...
TRANSCRIPT
![Page 1: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/1.jpg)
International Energy Agency Technology Collaboration Programme on
District Heating and Cooling including Combined Heat and Power
Annex XII final report
Effects of Loads on Asset Management of the 4th
Generation District Heating Networks
Date of publication: 31.03.2020
Authors: Ingo Weidlich, Gersena Banushi, Nazdaneh Yarahmadi, Ignacy
Jakubowicz, Jan Henrik Sällström, Alberto Vega, Jooyong Kim, Yeon Soo Kim,
Øyvind Nilsen, Thomas Grage, Georg Schuchardt, Fang Yang
![Page 2: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/2.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
2
This project has been independently funded by
the International Energy Agency Technology Collaboration Programme on
District Heating and Cooling including Combined Heat and Power
(IEA DHC).
Any views expressed in this publication are not necessarily those of IEA DHC.
IEA DHC can take no responsibility for the use of the information within this publication,
nor for any errors or omissions it may contain.
![Page 3: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/3.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
3
Executive Summary
This research project investigates the influence of future mechanical and thermal load
spectra on the service life of pre-insulated bonded single pipes, representing the
majority of currently operating DH pipelines, based on the 3rd generation technology
(3GDH). To minimize heat losses, these pipes have a composite cross-section with
three different material layers, including the steel pipe for the water supply, the
insulation foam of polyurethane (PUR), and an outer coating of High Density
Polyethylene (HDPE), interacting with the surrounding soil. The stiffness of the PUR
foam and its constant adhesion to the steel pipe are essential to properly transmit at the
HDPE coating the friction stresses from the surrounding soil. 3GDH pipes undergo
large temperature variations, associated with significant cyclic loading at the soil-pipe
interface, as well as within the pre-insulated DH pipe system, leading to accumulated
material damage and ageing.
Conversely, 4th generation DH networks (4GDH) operate at lower temperatures, also
integrating renewable energy sources, that are more volatile than the traditional ones,
based on 3GDH technology. The lower levels of operating temperature and the
increased amount of cyclic loading influence ageing and the service life of 4GDH
networks, requiring proper analysis of the system performance. This is fundamentally
important, in order to guarantee an efficient operation of DH networks, optimizing the
durability of the thermal insulation and the mechanical strength of the piping system.
Moreover, the elevated levels of operating temperature in 3GDH had led to the
development of pipe materials with worse thermal properties, able to sustain the high
temperature of the accelerated ageing tests, required by the European standards.
These do not address either the testing nor the design of 4GDH, highlighting the need
to further investigate the combined effect of increased volatility, and decreased thermal
loading associated with the integration of renewable energy sources in the DH network.
This research project aims to investigate the lifetime of 4GDH pre-insulated bonded
single pipes, developing a new approach as a function of the increased cyclic
mechanical and thermal loads, and the decreased thermal ageing.
![Page 4: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/4.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
4
The present report is structured in different chapters, describing the research carried
out within each workpackage of the research project.
Chapter 1 reports the fatigue analysis of the steel service pipe according to the
Palmgren-Miner rule implemented in EN 13941, considering temperature history data
collected from traditional and 4GDH pipe networks in Germany, Sweden, Norway, and
South Korea.
Chapter 2 describes the state-of-the-art on fatigue and damage accumulation theories
for the steel pipe material, identifying critical aspects in using linear and non-linear
theories. Moreover, the axial shear strength from naturally aged pipes is analyzed,
identifying the main parameters influencing the shear strength degradation of pre-
insulated piping systems.
Chapter 3 reports the experimental investigation of the adhesion strength of straight
pipes under the combined effect of thermal ageing and cyclic mechanical loading. The
adhesion strength of the aged pipes was evaluated using the plug method introduced
by RISE, while the degradation of the PUR material of some samples was analysed
further, using the Fourier transform infrared (FTIR) technique.
Chapter 4 reports the results of the tests performed on naturally aged pipes specimens,
gathered from four DH branches of KDHC in Korea, for measuring the adhesion
strength of the PUR foam.
Chapter 5 examines the consequences of the research project outcome for improving
the current network design, and asset management of 4GDH systems.
The analysis results indicate that the lifetime of 4GDH pipelines is expected to increase
due to the lower operating temperature, and the low impact of thermal loading volatility
in the network, compared to conventional DH. To accurately estimate the fatigue
damage, the measuring time interval in the temperature data should be sufficiently
small, highlighting the importance of data logging process. The performed accelerated
ageing tests demonstrated that the combined effect of mechanical loading and thermal
ageing accelerates the rate of chemical degradation of the PUR foam, leading to a
faster deterioration of the mechanical adhesion strength. The analysis of the naturally
aged DH pipelines showed that, in addition to the ageing time, the shear strength of DH
pipes depends on the temperature history, decreasing with the level of operating
temperature and amount of fluctuation. Finally, it is recommended to document the
![Page 5: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/5.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
5
operating temperature history and the most important properties of the pipe system
before installation, as well as during operation, contributing to better predictive
maintenance, and subsequent reduction of economic risks for replacement and repair.
In conclusion, the obtained results give a better understanding of the performance of
traditional and 4GDH pipelines in operation, that need to be suitably considered in the
engineering design standards of DH networks, contributing to a more sustainable and
energy efficient infrastructure.
![Page 6: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/6.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
6
Table of Contents
Executive Summary ....................................................................................................... 3
1 Identification of load characteristics and estimations of future conditions................ 9
1.1 Fatigue analysis of the steel service pipe .................................................... 9
1.1.1 The Palmgren-Miner rule ........................................................................... 13
1.2 Cycle counting methods ............................................................................ 15
1.2.1 Rainflow counting algorithm....................................................................... 16
1.3 Fatigue analysis of the collected sample data ........................................... 17
1.4 Results and discussion.............................................................................. 24
2 Investigation of fatigue theories and shear strength experience ............................ 28
2.1 Comparison of fatigue theories for steel .................................................... 28
2.1.1 State of the art – linear fatigue theories for ductile materials ..................... 29
2.1.2 Non-linear fatigue theories for ductile materials......................................... 31
2.1.3 Summary and evaluation of different theories of fatigue accumulation...... 34
2.2 Estimating fatigue resistances for future load conditions........................... 35
2.2.1 Definition of illustrative mechanical load spectrum .................................... 36
2.2.2 Quantification of fatigue resistances.......................................................... 43
2.2.3 Summary on the quantification of fatigue and interpretation of results ...... 44
2.2.4 Interpretation of results and evaluation of usability – critical aspects ........ 48
2.3 Compilation of shear strength data from naturally aged pipes................... 49
2.3.1 Naturally aged pipes.................................................................................. 50
![Page 7: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/7.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
7
2.3.2 Artificially aged pipes................................................................................. 55
2.3.3 Summary ................................................................................................... 57
3 Investigation of thermal ageing in combination with cyclic mechanical loads ........ 58
3.1 Background ............................................................................................... 58
3.2 Choice of objects and conditions ............................................................... 59
3.3 Selection of mechanical loading ................................................................ 59
3.4 Experiments............................................................................................... 61
3.4.1 Mechanical testing..................................................................................... 63
3.4.2 Fourier Transform Infrared spectroscopy................................................... 64
3.5 Results....................................................................................................... 65
3.5.1 Mechanical adhesive strength ................................................................... 66
3.5.2 Fourier-transform infrared spectroscopy.................................................... 69
3.6 Interpretation of results and evaluation of usability.................................... 73
4 Field tests .............................................................................................................. 74
4.1 Field selection and data acquisition........................................................... 74
4.1.1 Selecting and test preparation of naturally aged DH pipes........................ 74
4.1.2 Measuring the shear strength .................................................................... 79
4.2 Results....................................................................................................... 80
4.2.1 Shear strength of naturally aged DH pipes................................................ 80
4.2.2 FTIR analysis of naturally aged DH pipes.................................................. 95
5 Consequences for network design and asset management strategy..................... 97
5.1 Analyze consequences and impacts of future loads on the service pipe... 97
![Page 8: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/8.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
8
5.2 Consequences and impacts of future loads on the adhesion between
service pipe and insulation........................................................................................ 97
5.3 Recommendations concerning network design and asset management of
4th generation DH systems....................................................................................... 98
Concluding remarks ..................................................................................................... 99
Acknowledgments ...................................................................................................... 100
Bibliography................................................................................................................ 101
Appendix A: fatigue analysis results of the temperature data..................................... 105
![Page 9: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/9.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
9
1 Identification of load characteristics and estima tions of
future conditions
First, this section introduces the fatigue analysis of the steel service pipe, using the
failure criterion given by the Palmer-Miner rule, implemented in the European Standard
EN 13941, and the different cycle counting methods, including the rainflow counting
algorithm. Second, the fatigue analysis is performed considering the collected
temperature history data of 3GDH networks from different countries, provided by the
project partners. These results are then compared with those obtained from the
collected temperature history data on 4GDH pipes, in order to estimate the trend for
future load spectra, characterizing the energy transition in the heat sector.
1.1 Fatigue analysis of the steel service pipe
Fatigue is defined as material failure, caused by crack initiation and progressive growth,
due to repeated cyclic loading. At present, there are three major approaches for fatigue
design and analysis, like the traditional stress-based approach, strain-based approach,
and fracture mechanics approach (Dowling, 2013). The design standard of DH pipes
(EN 13941) uses the stress-based approach for the fatigue analysis of the steel service
pipe. Herein, the lifetime of a test specimen or an engineering component, subjected to
fully reversible cycle of stress range S, is measured in terms of the corresponding
number of loading cycles to failure N. Thus, the material fatigue performance is
characterized by the SN curve, also known as a Wöhler curve, representing the
magnitude of a cyclic stress, in terms of stress range (S) versus the number of cycles to
failure (N).
The lifetime where High-Cycle Fatigue (HCF) starts varies with material, typically in the
range 102 to 104 cycles, characterized by small elastic stress amplitudes. Conversely,
Low Cycle Fatigue (LCF) is caused by a relatively small number of cycles, on the order
of 102, associated with significant amounts of plastic deformation. Repeated heating
and cooling can cause a cyclic stress due to differential thermal expansion and
contraction, resulting in thermal fatigue (Dowling, 2013; Frederiksen and Werner, 2013).
In contrast to classical high or low cycle fatigue (Figure 1.1), steel structures subjected
extreme loading conditions, like earthquakes, may experience Ultra Low Cycle Fatigue
(ULCF), involving fewer than ten cycles with large strain amplitudes, on the order of ten
or more times the yield strain (Kanvinde et al., 2013; Fernandes et al., 2018).
![Page 10: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/10.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
10
Figure 1.1: Definition of the material failure mechanism as a function of the number of cycles (Bleck et al., 2009).
Fatigue strength is expressed in terms of series of SN curves (Figure 1.2), representing
the relationship between the stress range (S) and the number of cycles to failure (N).
Each SN curve refers to a particular construction detail, considering the effect of local
peak stresses (EN 13941, 2019).
Figure 1.2: SN curve for for steel service pipe components with butt weld and fillet weld (EN 13941, 2019).
![Page 11: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/11.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
11
The SN curves have been derived from fatigue test data obtained from appropriate
laboratory specimens tested under stress control or, for applied strains exceeding yield
(low cycle fatigue), under strain control. Continuity from low to high cycle regime is
achieved by expressing low cycle fatigue data in terms of the pseudo-elastic stress
range (i.e. strain range multiplied by elastic modulus, if necessary corrected for
plasticity).
Generally, SN curves vary with the material and its pre-processing. They are also
affected by mean stress, member geometry, especially the presence of notches,
surface finish, frequency of cycling, residual stress, thermal and chemical environment
(Dowling, 2013). The latter can cause stress corrosion cracking (SCC), initiating fatigue
earlier than expected, if the steel undergoes variable thermal loading. The presence of
water within a fatigue crack, as well as low-frequency temperature variations, typical of
district heating, adversely impact fatigue crack growth rates, by increasing the time for
environmental interactions per stress cycle (Christensen et al., 1999).
Most piping standards (EN 13941; EN 13480-3; ASME 31.1; ASME 31.3) are based on
the full-scale fatigue tests performed by Markl in the 1950's on a limited number of pipe
components, including straight pipes, bends and tees (Markl, 1952; 1955). The results
were a set of SN curves referring to an equivalent straight pipe, and the flexibility and
stress concentration factors for different pipe components. These factors, since their
introduction in the piping standards, have remained unchanged up to date. However,
full-scale fatigue tests are very expensive, requiring a considerable number of test
specimens to determine a statistically reliable SN curve. On the other hand, the stress
concentration factors can be evaluated by analytical and numerical methods, as hot-
spot values, and further compared to SN curves obtained from small scale material test
specimens. The latter must be further corrected for practical use with a number of
coefficients, taking into account the effects of rougher surface finish, temperature,
mean stress level, plastic strains, and electro-chemical environment (Randløv et al.,
1996; Christensen et al., 1999; Penderos, 1996).
The current design standard of DH systems (EN 13941, 2019), proposes the following
expression to evaluate the SN curve for the steel service pipe, based on the
aforementioned Markl's experimental work (Markl, 1952; 1955):
![Page 12: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/12.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
12
mNkS /1−⋅= (1.1) m
S
kN
= (1.2)
For low alloyed carbon steel (σy ≤ 360 N/mm2) normally used for preinsulated pipes the
factors k = 5000 N/mm2 and m = 4 can be used giving:
4/15000 −⋅= NS N/mm2 (1.3) 4
5000
=S
N (1.4)
Clearly, the SN curve should be used in combination with stress intensification factors
calculated or measured as hot-spot values. The curve includes the effect of a butt weld,
as well as reductions for rolled skin, temperature, and plastic yield (Figure 1.2). The
effect of the electro-chemical environment is not included, although a combination of
plastic strains and high pH-values of the heat carrier, characterizing district heating,
may severely reduce the fatigue life of DH pipelines (Christensen et al., 1999).
The curve presupposes that the stress range is calculated assuming a linear elastic
material behavior for the steel, also above yield.
The safety factor is applied by dividing the calculated number of cycles with γfat (Table
1.1), as a function of the project class, depending on the risk and consequences of
damage to the society or environment.
Table 1.1: Partial safety factor for action cycles
Project class A Project class B Project class C
γfat 5 6.67 10
The failure criterion is expressed by the Palmgren-Miner hypothesis:
fatfi
i
N
N
γ1≤∑ (1.5)
where:
Ni is the number of cycles with stress range Si during the required design life;
Nfi is the number of cycles of stress range Si to cause failure.
![Page 13: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/13.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
13
1.1.1 The Palmgren-Miner rule
Considering a situation of variable amplitude loading, (Figure 1.3), where a certain
stress amplitude σai is applied for a number of cycles Ni, and the number of cycles to
failure from the SN curve for σai is Nfi, then the fraction of life used is Ni/Nfi. The
Palmgren–Miner rule states that fatigue failure is expected when the life fractions
(Ni/Nfi), corresponding to all levels of stress amplitudes σai, sum to unity:
13
3
2
2
1
1 ==+++ ∑fi
i
fff N
N
N
N
N
N
N
NL (1.6)
Figure 1.3: Use of the Palmgren–Miner rule for life prediction for variable amplitude loading which is completely reversed (Dowling, 2013).
This simple rule was used by A. Palmgren in Sweden in the 1920s for predicting the life
of ball bearings. However, it was not widely known until its appearance in 1945 in a
paper by M. A. Miner (Dowling, 2013), subsequently finding large application in the
fatigue analysis and design of railways, bridges, aircraft and offshore structures.
In the design of district heating pipelines, the temperature history is used to calculate
the stress history. For simplicity, it is assumed that the stress variation Si is proportional
to the temperature variation ∆Ti:
ii TcS ∆⋅= (1.7) m
ifi Tc
kN
∆⋅= (1.8)
where, k and m are the constants defining the SN curve (Equation (1.1)), while c is the
proportionality constant between stress range (Si) and temperature variation (∆Ti):
![Page 14: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/14.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
14
( )i
T
T
i
i
T
dTE
T
Sc
i
∆
⋅=
∆=
∫0
α (1.9)
where the Young's modulus E and the linear thermal expansion coefficient α vary
linearly with the operating temperature, according to EN 13941 (2019). Consequently,
also c increases with temperature, but for simplicity in the design calculations, it can be
assumed as constant, equal to c = 2.505 N/mm2/K, considering an installation and
operation temperature equal to T0 = 10°C and Ti = 120°C, respectively.
The Palmgren-Miner rule can be expressed as:
∑∑ ∆⋅
= mii
m
fi
i TNk
c
N
N (1.10)
Moreover, the temperature history is simplified by calculating the number of full
temperature cycles N0, for the reference temperature ∆Tref, giving the same
accumulated damage as the temperature history presumed for the system:
mref
mm
ii
m
fi
i TNk
cTN
k
c
N
N ∆⋅
=∆⋅
= ∑∑ 0 (1.10)
∑ ∆⋅∆
= miim
ref
TNT
N1
0 (1.11)
therefore, the number of full temperature cycles N0 only depends on the coefficient m,
the reference temperature ∆Tref and the temperature history (Randløv et al., 1999;
EN 13941, 2019).
The design standard EN 13941 (2019) assumes as a reference temperature
∆Tref = 110°C, corresponding to a maximum number of equivalent load cycles N0,max, for
a straight pipe, given by:
fatfatfat
m
mreffi
i
m
mref c
k
TN
N
c
k
TN
γγγ1084121
2.505
5000
110
11114
4max,0 =⋅
=⋅
∆=⋅
∆= ∑ (1.12)
![Page 15: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/15.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
15
Clearly, the design standards of DH systems, must take into account their typical
mechanical behavior, that differs from other types of pipelines, such as the elevated
axial stresses due to the soil resistance, counteracting pipe thermal expansion.
According to EN 13941 (2019), the number of full action cycles, chosen in the
calculation for pipelines in normal operation, should not exceed the lowest number of
equivalent full action cycles indicated in Table 1.2 for the relevant years in operation.
Table 1.2: Equivalent full action cycles for m = 4 and ∆Tref = 110°C (EN 13941, 2019).
Number of full action cycles N 0 Pipeline characteristics 30 years operation 50 years operation
Transmission pipelines 100 to 250 170 to 420 Distribution pipelines 250 to 500 420 to 840 House connections 1000 to 2500 1700 to 4200
1.2 Cycle counting methods
Various counting methods have been developed to reduce cyclic time histories to some
simple form of cycle count for analysis and testing purposes. Two parameters counting
methods, such as range-pair-range, rainflow, and racetrack methods, have been
developed, reflecting some aspects of loading sequence. Rainflow counting is perhaps
the most widely accepted method for the identification of fatigue critical events and is
useful when pursuing a basic understanding of material behavior (Rice, 1997).
Most practical rainflow counting algorithms are based either on the 'availability matrix'
or the 'vector' mathematical concepts (Dowing and Socie, 1982). The 'availability
matrix' algorithm, developed by Wetzel (1971), requires the input signal to be divided
into a finite number of bands, used to define the numerical value of the range and mean
of each reversal. Corresponding to each band is an element in the availability matrix.
'Vector' based rainflow counting algorithms use a one dimensional array to keep track
of those peaks and valleys which have not formed a closed loop. In other words, once a
closed loop has been determined, the peak and valley associated with it can be
eliminated from the vector. First proposed by Matsuishi and Endo (1968), this approach
is generally regarded as the method leading to the best estimation of fatigue life. Due to
its importance, many other procedures, essentially equivalent, have been proposed in
literature (Downing et al., 1976; Okamura et al., 1979).
![Page 16: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/16.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
16
Finally, the ASTM standard E1049-85 reports a compilation of acceptable procedures
for cycle-counting methods employed in fatigue analysis (ASTM, 2011). The latter is
also referenced in the ISO 12110-2 (ISO 12110-2, 2013), presenting cycle counting
techniques and data reduction methods used in variable amplitude fatigue testing.
1.2.1 Rainflow counting algorithm
In performing a rainflow cycle counting, a cycle is identified and if it meets the criterion
shown in Figure 1.4. A peak-valley-peak or valley-peak-valley combination X-Y-Z in the
loading history is considered to contain a cycle if the second range, ∆σYZ, is greater
than or equal to the first range, ∆σXY. If the second range is indeed larger or equal, then
a cycle equal to the first range (∆σXY) is counted. The mean value for this cycle,
specifically the average of σX and σY, is also an important parameter.
The complete procedure is described as follows, using the example of Figure 1.5:
Rainflow counting starts at the beginning of the recorded temperature variations, using
the criterion illustrated in Figure 1.4. If a cycle is counted, this information is recorded,
and its peak and valley are assumed not to exist for purposes of further cycle counting,
as illustrated for cycle E-F in (c). If no cycle can be counted at the current location, it is
proceeded until a count can be made.
Counting is complete when all of the history is exhausted. For this example, the cycles
counted are E-F, A-B, H-C, and D-G, with ranges and means as tabulated at the
bottom of Figure 1.5. For lengthy histories, it is convenient to present the results of
rainflow cycle counting as a matrix giving the numbers of cycles occurring at various
combinations of range and mean.
Figure 1.4: Condition for counting a cycle with the rainflow method (Dowling, 2013).
![Page 17: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/17.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
17
Figure 1.5: Example of rainflow cycle counting (Randløv et al., 1996; Dowling, 2013).
Commonly, the rain flow cycle counting for DH pipes is calculated considering directly
the temperature history, instead of the stress history, due to the assumed linear
relationship between the stress and temperature variations (Randløv et al., 1996).
1.3 Fatigue analysis of the collected sample data
First, this section presents 19 pairs of temperature history data from conventional DH
networks (in supply and return), gathered in different countries from the project partners.
Specifically, the Korea District Heating Corporation (KDHC) collected data from the two
locations of Goyang and Daegu Branch in South Korea, the Research Institutes of
Sweden AB (RISE) from the Gothenburg Energy company in Sweden, the Fortum Oslo
Varme AS (Fortum) from the Vika heat plant as well as different house connections in
Norway, and the German District Heating Research Institute (FFI) from the DH heating
network in Germany. Table 1.3 summarizes the characteristics of the pipe samples,
where the collected temperature data have been recorded, numbering the measured
points from 1 to 19, and indicating the supply and return pipe with "S" and "R" before
![Page 18: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/18.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
18
the number. Table 1.4 reports the information on the measured time period, including
the recording time interval ∆t, corresponding to each measured point.
For these temperature data, the measuring time ranges from 1 year to 11 years, except
for the single pair of data (S18/R18) collected in Germany, at the HKW network in
Hannover in 2011, over a period of 4 months. Moreover, the temperature data gathered
in South Korea and Sweden are recorded every hour (∆t = 60 min), while those in
Norway and Germany every 5 min and 15 min, respectively (Table 1.4).
Second, the section presents temperature data measured at different locations of a
4GDH network in Germany, heated with solar energy, as indicated in Table 1.5 and
Table 1.6. These temperature data have been collected within the recent German
research project "Solar district heating for the Brühl district in Chemnitz –
accompanying research (SolFW)" in 2018. The measured points regard different
locations the system including the main (S20/R20) and distribution (S21/R21) pipelines,
the two-zone thermal energy storage, the two solar collector fields (S22/R22; S23/R23),
and a house connection (S24/R24), illustrated in Figure 1.6 (Shrestha et al., 2018). The
measuring time for these data is one year, and the corresponding recording frequency
is one minute (∆t = 1 min), but for the house connection (∆t = 5 min).
This solar 4GDH system, is regulated so that the supply temperature in the distribution
network ranges between 75°C and 90°C, depending on the outdoor temperature. When
the solar heat is not sufficient to satisfy the consumers demand, the required heat
generation is complemented with a combined heat and power (CHP) plant, feeding the
heat directly into the distribution DH network.
The gathered temperature history data have been processed in Matlab (Mathworks,
2019), using the implemented rainflow cycle counting algorithm, for the fatigue analysis.
Then, the resulting rainflow cycle matrix, is used to evaluate the number of equivalent
full temperature cycles N0, using the Palmer-Miner rule. The latter is calculated for two
different values of the reference temperature ∆Tref, ∆Tref = 110°C, and ∆Tref = Tmax -
10°C (maximum temperature at the measuring point - 10°C), as also assumed in
Randløv et al. (1996).
Appendix A reports in detail the evaluated rainflow cycle matrix, as well as the number
of equivalent full temperature cycles N0 corresponding to each temperature history data.
![Page 19: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/19.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
19
Table 1.3: Characteristics of the measured points for the collected temperature history data regarding the steel pipe in conventional DH pipelines (3GDH).
Temperature history data
DH pipe Category N. mes. point *
Direction DN [mm]
OD [mm]
ts [mm] H [m]
S1 supply 65 140 4.5 1.6 Daegu branch House connection
R1 return 65 140 4.5 1.6
S2 supply 80 160 4.5 2.1
KDHC (South Korea)
Goyang branch House connection R2 return 80 160 4.5 2.1
Main S3 supply 200 315 4.5 1.7 Danska vägen
Main R3 return 200 315 4.5 1.7
Main S4 supply 400 630 6.3 2.5 Hisingsbron
Main R4 return 400 630 6.3 2.5
Main S5 supply 500 710 6.3 2.0-2.5 Marieholm
Main R5 return 500 710 6.3 2.0-2.5
Main S6 supply 300 500 5.6 1.5-2.0
RISE (Sweden)
Falutorget Main R6 return 300 500 5.6 1.5-2.0
S7 supply 50-500 125-710 0.6-1.5 Vika 2013 Main, Distribution
House connection R7 return 50-500 125-710 0.6-1.5
S8 supply 50-500 125-710 0.6-1.5 Vika 2016 Main, Distribution
House connection R8 return 50-500 125-710 0.6-1.5
S9 supply 65 76.1 2.9 0.6 Soerengkaia 153 Substation
R9 return 65 76.1 2.9 0.6
S10 supply 50 60.3 2.9 0.6 Brobekkveien 80 House connection
R10 return 50 60.3 2.9 0.6
S11 supply 100 114.3 3.6
Fortum (Norway)
Skøyen Terrasse 4 House connection
R11 return 100 114.3 3.6
Main S12 supply 300 323.9 5.6 0.6-0.8 COCA 2010
Main R12 return 300 323.9 5.6 0.6-0.8
Main S13 supply 900 914.0 10.0 - GKH 2010
Main R13 return 900 914.0 10.0 -
Main S14 supply 400-600 406.4-610 6.3-7.1 in conduit HKW 2010
Main R14 return 400-600 406.4-610 6.3-7.1 in conduit
Main S15 supply 200-700 219.1-711 4.5-8.0 0.6-0.8 KWH 2010
Main R15 return 200-700 219.1-711 4.5-8.0 0.6-0.8
Main S16 supply 300 323.9 5.6 0.6-0.8 COCA 2011
Main R16 return 300 323.9 5.6 0.6-0.8
Main S17 supply 900 914.0 10.0 - GKH 2011
Main R17 return 900 914.0 10.0 -
Main S18 supply 400-600 406.4-610 6.3-7.1 in conduit HKW 2011
Main R18 return 400-600 406.4-610 6.3-7.1 in conduit
Main S19 supply 200-700 219.1-711 4.5-8.0 0.6-0.8
FFI (Germany)
KWH 2011 Main R19 return 200-700 219.1-711 4.5-8.0 0.6-0.8
* N. mes. point = number of the measured point for the temperature data
![Page 20: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/20.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
20
Table 1.4: Characteristics of the collected temperature history data regarding the steel pipe in conventional DH pipelines (3GDH). N. mes. point* Start date Finish date Number of days
start-finish Number of
days recorded Number of
days missing Measuring
interval ∆t [min]
S1 01/01/2015 01/01/2018 1096 1096 0 60
R1 01/01/2015 01/01/2018 1096 1096 0 60
S2 01/01/2015 01/01/2018 1096 1096 0 60
R2 01/01/2015 01/01/2018 1096 1077 19 60
S3 27/03/2007 26/06/2017 3744 3744 0 60
R3 01/04/2009 26/06/2017 3008 3008 0 60
S4 01/12/2006 23/03/2018 4131 4131 0 60
R4 01/12/2006 23/03/2018 4131 4131 0 60
S5 11/01/2008 18/07/2017 3476 3476 0 60
R5 01/07/2007 18/07/2017 3671 3671 0 60
S6 01/12/2006 23/03/2018 4131 4131 0 60
R6 01/12/2006 23/03/2018 4131 4131 0 60
S7 01/01/2013 31/12/2013 365 365 0 5
R7 01/01/2013 31/12/2013 365 365 0 5
S8 01/01/2016 31/12/2016 366 366 0 5
R8 01/01/2016 31/12/2016 366 366 0 5
S9 20/08/2018 18/08/2019 364 364 0 5
R9 20/08/2018 18/08/2019 364 364 0 5
S10 30/08/2018 30/08/2019 365 365 0 5
R10 30/08/2018 30/08/2019 365 365 0 5
S11 30/08/2018 30/08/2019 365 365 0 5
R11 30/08/2018 30/08/2019 365 365 0 5
S12 01/01/2010 01/01/2011 365 365 0 15
R12 01/01/2010 01/01/2011 365 365 0 15
S13 01/01/2010 01/01/2011 365 365 0 15
R13 01/01/2010 01/01/2011 365 365 0 15
S14 01/01/2010 01/01/2011 365 365 0 15
R14 01/01/2010 01/01/2011 365 365 0 15
S15 01/01/2010 01/01/2011 365 365 0 15
R15 01/01/2010 01/01/2011 365 365 0 15
S16 01/01/2011 01/01/2012 365 365 0 15
R16 01/01/2011 01/01/2012 365 365 0 15
S17 01/01/2011 01/01/2012 365 365 0 15
R17 01/01/2011 01/01/2012 365 365 0 15
S18 01/01/2011 09/05/2011 128 128 0 15
R18 01/01/2011 09/05/2011 128 128 0 15
S19 01/01/2011 01/01/2012 365 365 0 15
R19 01/01/2011 01/01/2012 365 365 0 15 * N. mes. point = number of the measured point for the temperature data
![Page 21: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/21.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
21
Table 1.5: Characteristics of the measured points for the collected temperature history data regarding the steel pipe in the solar DH network in Chemnitz, Germany (4GDH).
Temperature history data Category N. mes.
point* Direction DN [mm]
OD [mm] ts [mm] H [m]
S20 supply 200 219.1 4.5 0.8-1.2 Main pipe
R20 return 200 219.1 4.5 0.8-1.2
S21 supply 250 273.0 5.0 0.8-1.2 Distribution pipe
R21 return 250 273.0 5.0 0.8-1.2
SL1 level 1
SL2 level 2
SL3 level 3 Heat storage
SL4 level 4
S22 supply 80 88.9 3.2 0.8-1.2 Solar field 1
R22 return 80 88.9 3.2 0.8-1.2
S23 supply 80 88.9 3.2 0.8-1.2 Solar field 2
R23 return 80 88.9 3.2 0.8-1.2
S24 supply 25 33.7 2.3 0.8-1.2
Solar DH (SolFW Project)
House connection R24 return 25 33.7 2.3 0.8-1.2
* N. mes. point = number of the measured point for the temperature data
Table 1.6: Characteristics of the collected temperature history data regarding the steel pipe in the solar DH network in Chemnitz, Germany (4GDH).
N. mes. point* Start date Finish date Number of days
start-finish Number of
days recorded Number of
days missing
Measuring interval ∆t
[min]
S20 01/01/2018 31/12/2018 365 365 0 1
R20 01/01/2018 31/12/2018 365 365 0 1
S21 01/01/2018 31/12/2018 365 365 0 1
R21 01/01/2018 31/12/2018 365 365 0 1
SL1 01/01/2018 31/12/2018 365 365 0 1
SL2 01/01/2018 31/12/2018 365 365 0 1
SL3 01/01/2018 31/12/2018 365 365 0 1
SL4 01/01/2018 31/12/2018 365 365 0 1
S22 01/01/2018 31/12/2018 365 365 0 1
R22 01/01/2018 31/12/2018 365 365 0 1
S23 01/01/2018 31/12/2018 365 365 0 1
R23 01/01/2018 31/12/2018 365 365 0 1
S24 01/01/2018 31/12/2018 365 365 0 5
R24 01/01/2018 31/12/2018 365 365 0 5 * N. mes. point = number of the measured point for the temperature data
![Page 22: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/22.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
22
Figure 1.6: Schematic representation of the supply station at the solar DH system in Chemnitz (Shrestha et al., 2018).
Table 1.7 and Table 1.8 summarize the evaluated fatigue damage for the collected
temperature history data, from conventional and 4th generation DH networks,
respectively, by linearly extrapolating the results for a lifetime of 30 years and 50 years.
For comparison purposes, the tables indicate also the allowable number of full
equivalent cycles N0, recommended in EN13941, depending on the pipe category.
![Page 23: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/23.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
23
Table 1.7: Results of the evaluated equivalent full temperature cycles N0, for all temperature history data from conventional DH pipes, considering two different values of the reference temperature ∆Tref, for a lifetime of 30 and 50 years, by linear extrapolation.
* N. mes. point = number of the measured point for the temperature data
N0(∆Tref = Tmax - 10°C) N0(∆Tref = 110°C) N0 (criteria EN 13941) N. mes. point*
Tmax (°C)
Tmean (°C) 30 years 50 years 30 years 50 years 30 years 50 years
S1 119.1 95.3 6.73 11.21 6.51 10.84 1000-2500 1700-4200
R1 84.4 39.7 6.42 10.70 1.35 2.25 1000-2500 1700-4200
S2 117.9 97.9 2.05 3.42 1.90 3.17 1000-2500 1700-4200
R2 71.3 45.2 4.64 7.74 0.45 0.75 1000-2500 1700-4200
S3 115.8 61.1 47.08 78.47 40.27 67.12 100-250 170-420
R3 99.4 45.7 22.03 36.72 9.59 15.99 100-250 170-420
S4 114.3 86.8 35.05 58.42 28.37 47.28 100-250 170-420
R4 127.4 33.1 11.28 18.80 14.63 24.38 100-250 170-420
S5 110.4 88.7 68.81 114.68 47.75 79.58 100-250 170-420
R5 75.6 24.3 19.63 32.71 2.49 4.14 100-250 170-420
S6 115.8 61.9 45.13 75.22 38.60 64.34 100-250 170-420
R6 127.4 33.1 11.28 18.80 14.63 24.38 100-250 170-420
S7 125.7 101.6 84.90 141.50 103.92 173.20 100-250 170-420
R7 105.4 60.6 60.55 100.92 34.26 57.09 100-250 170-420
S8 122.0 101.6 12.24 20.40 13.15 21.92 100-250 170-420
R8 105.6 57.2 86.03 143.38 49.08 81.80 100-250 170-420
S9 115.9 98.3 12.58 20.96 10.82 18.03 1000-2500 1700-4200
R9 55.1 38.3 1223.22 2038.70 34.56 57.61 1000-2500 1700-4200
S10 118.5 92.4 4.66 7.77 4.41 7.35 1000-2500 1700-4200
R10 62.8 45.8 297.93 496.56 15.82 26.36 1000-2500 1700-4200
S11 112.6 90.7 32.41 54.02 24.53 40.88 1000-2500 1700-4200
R11 60.2 38.7 1521.94 2536.56 66.01 110.02 1000-2500 1700-4200
S12 120.8 97.1 4.18 6.97 4.30 7.17 100-250 170-420
R12 80.0 61.3 2.26 3.76 0.37 0.62 100-250 170-420
S13 122.1 96.5 1.33 2.22 1.44 2.39 100-250 170-420
R13 73.5 59.1 0.34 0.56 0.04 0.06 100-250 170-420
S14 119.4 92.9 19.31 32.19 18.89 31.49 100-250 170-420
R14 83.1 60.5 14.31 23.85 2.79 4.65 100-250 170-420
S15 122.1 96.6 1.26 2.10 1.36 2.26 100-250 170-420
R15 68.5 58.5 0.50 0.84 0.04 0.07 100-250 170-420
S16 111.4 95.7 6.15 10.25 4.44 7.40 100-250 170-420
R16 80.0 60.1 3.15 5.24 0.52 0.86 100-250 170-420
S17 120.3 94.6 4.41 7.35 4.46 7.43 100-250 170-420
R17 65.9 58.4 1.94 3.24 0.13 0.22 100-250 170-420
S18 111.4 93.6 27.24 45.39 19.67 32.78 100-250 170-420
R18 83.0 64.2 45.96 76.60 8.91 14.86 100-250 170-420
S19 120.6 94.7 3.49 5.82 3.57 5.95 100-250 170-420
R19 68.2 57.7 0.62 1.04 0.05 0.08 100-250 170-420
![Page 24: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/24.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
24
Table 1.8: Results of the evaluated equivalent full temperature cycles N0, for all temperature history data from 4GDH pipes, considering two different values of the reference temperature ∆Tref, for a lifetime of 30 and 50 years, by linear extrapolation.
* N. mes. point = number of the measured point for the temperature data
1.4 Results and discussion
The number of full temperature cycles N0 for the temperature data collected in Norway
at the Vika heat plant in supply, in 2013 (S7), exceed the lower limit recommended by
EN 13941 for transmission pipelines, referring to a service life of 30 years and 50 years
(Table 1.7). For the rest of collected temperature data on conventional DH networks,
the calculated number of full temperature cycles N0 is below the standard limits.
On the other hand, for the temperature data gathered from the supply pipes in the two
solar fields of the 4GDH network in Chemnitz (S22, S23), the number of full
temperature cycles N0 exceeds the standard limits for the transmission and distribution
pipelines (Table 1.8). This elevated thermal loading fluctuation, occurring in a limited
portion of the network, is associated with the typical day-night solar thermal cycles.
Clearly, such volatility is drastically reduced below the standard limit, in the main
heating circuit, because of the controlling strategy adopted by the DH company.
Among 4GDH pipelines, the house connection return pipe (R24) presented the largest
number of fluctuations, due to the consumers operation, however below the standard
limits. Moreover, the number of full temperature cycles in the distribution pipes
N0(∆Tref = Tmax - 10°C) N0(∆Tref = 110°C) N0 (criteria EN 13941) N. mes. point*
Tmax (°C)
Tmean (°C) 30 years 50 years 30 years 50 years 30 years 50 years
S20 146.9 134.3 9.91 16.51 23.76 39.61 100-250 170-420
R20 104.7 59.091 10.97 18.29 6.03 10.05 100-250 170-420
S21 88.46 72.043 5.99 9.98 1.55 2.58 250-500 420-840
R21 63.7 55.374 2.47 4.12 0.14 0.23 250-500 420-840
SL1 65.68 44.166 394.83 658.05 25.92 43.20 - -
SL2 62.74 45.827 184.25 307.09 9.74 16.23 - -
SL3 86.34 51.367 102.01 170.02 23.66 39.44 - -
SL4 93.52 56.45 120.09 200.16 39.91 66.52 -
S22 127.54 41.362 643.01 1071.69 838.30 1397.16 - -
R22 76 38.488 852.80 1421.33 110.52 184.20 - -
S23 94.26 39.811 1923.84 3206.40 662.34 1103.91 - -
R23 64.46 37.344 1786.71 2977.84 107.35 178.91 - -
S24 82.4 70 340.69 567.82 63.93 106.56 1000-2500 1700-4200
R24 68.7 41.17 1426.31 2377.18 115.66 192.77 1000-2500 1700-4200
![Page 25: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/25.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
25
(S21/R21) is significantly lower than in the rest of the 4GDH system. Evidently, the
thermal fluctuations in the 4GDH transmission and house connection pipes are greater
in supply (S20) and return (R24), respectively (Table 1.8).
To investigate the influence of the length of the data measuring interval ∆t, the number
of full temperature cycles N0 has been calculated for a service life of 30 years,
assuming a smaller measuring interval (∆t), when possible: 2 min, 4 min, 5 min, 15 min,
30 min, 45 min and 60 min. Figure 1.7 presents the variation of the number of full
temperature cycles N0, corresponding to a reference temperature ∆Tref = 110°C, and a
lifetime of 30 years, as a function of the sampling interval ∆t, for the temperature data
from conventional DH pipes in Norway and Germany. Similarly, Figure 1.8 shows the
variation of N0 with ∆t, for the 4GDH temperature data collected in Germany.
Generally, the number of full temperature cycles N0 decreases for greater values of the
measuring interval ∆t. This behavior depends on the measured data, and is stronger for
the temperature measurements S7 and S18, where the number of temperature cycles
N0 drops more than 70%, if measured every 60 min (Figure 1.7). Clearly, the return
house connection pipes exhibited a greater variation of the number of full temperature
cycles N0 with the sampling frequency ∆t, compared to the corresponding supply pipes
(Figure 1.7(b); Figure 1.8), due to the consumers operation.
Therefore, a large measuring time interval (∆t = 60min), can lead to an underestimation
of the number of full temperature cycles N0, and associated fatigue damage, depending
on the location in the network and pipe operating conditions. This is consistent with the
observations of Randløv et al. (1996), suggesting that a more frequent sampling is
necessary for low cycle fatigue analysis.
In conclusion, 4GDH pipelines are susceptible to greater thermal fluctuation in a limited
portion of the network, in the solar thermal circuit, associated with the typical day-night
solar thermal cycles. This elevated volatility drops drastically before entering the main
heating circuit, due to controlling measures adopted by DH companies. Moreover, other
4GDH systems with more plannable heat sources, like waste, biomass, geothermal, are
expected to have a lower temperature fluctuation, due to the absence of any localized
volatility as in the solar thermal circuit, and the controlling strategy of the DH operator.
Therefore, the lifetime of 4GDH pipelines is expected to increase due to the lower
operating temperature, and the low impact of thermal loading volatility in the network,
compared to conventional DH.
![Page 26: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/26.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
26
a)
0
20
40
60
80
100
120
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S7
R7
S8
R8
b)
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t N
umbe
r of f
ull t
empe
ratu
re c
ycle
s N
0
S9
R9S10R10S11R11
c)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S12
R12
S16
R16
d)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S13
R13
S17
R17
e)
0
2
4
6
8
10
12
14
16
18
20
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S14
R14
S18
R18
f)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S15
R15
S19
R19
Figure 1.7: Number of equivalent full temperature cycles N0, for a reference temperature ∆Tref = 110°C, and a lifetime of 30 years, as a function of the measuring interval ∆t, for the collected temperature data from conventional DH pipelines in Norway and Germany: a) S7, R7, S8, R8; b) S9, R9, S10, R10, S11, R11; c) S12, R12, S16, R16; d) S13, R13, S17, R17; e) S14, R14, S18, R18; f) S15, R15, S19, R19.
![Page 27: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/27.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
27
0
20
40
60
80
100
120
0 5 10 15 20 25 30 35 40 45 50 55 60
Measuring interval ∆t
Num
ber o
f ful
l tem
pera
ture
cyc
les
N0
S20
R20
S21
R21
S24
R24
Figure 1.8: Number of equivalent full temperature cycles N0, for a reference temperature ∆Tref = 110°C, and a lifetime of 30 years, as a function of the measuring interval ∆t, for the collected temperature data from 4GDH pipelines: a) S20, R20, S21, R21, S22, R22, S24, R24.
![Page 28: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/28.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
28
2 Investigation of fatigue theories and shear stren gth
experience
This chapter focuses on the influences of future mechanical load spectra on the
operational lifetime of steel service pipes. First, an overview on fatigue and damage
accumulation theories for steel is reported, followed by the different evaluation criteria.
Then, preliminary quantifications on fatigue resistances are performed, laying the
foundation for further analysis within this work package. Finally, impact parameters on
the shear strength of naturally aged pipes are described in order to identify the main
parameters for the compilation of shear strength data.
2.1 Comparison of fatigue theories for steel
Fatigue damage accumulation theories predict the effects of cyclic mechanical loads on
material components. Herein, the considered mechanical loads are quasi-static,
induced due to slowly changing temperatures in the DH system (thermally induced
mechanical load cycles). On the other hand, static loads (due to overburden heights)
and dynamic loads (mainly due to changing operational pressures) are not considered.
The application of fatigue theories allows predicting the technical lifetime of the
structural components. Based on past mechanical load cycles, the remaining lifetime
might be approximated as well.
Fatigue theories (linear or non-linear) assume that any mechanical load cycle causes a
certain damage in the structural components (Gudehus and Zenner, 1995; Reinhardt,
2010). The “Woehler-curves” (Wöhler-curves) relate the applied mechanical loads
(stresses) σimax to the number of mechanical load cycles Nimax to failure:
( )maxmaxmax iii Nσσ ∆=∆ (2.1)
Regarding the Woehler-curve, the correlation of Δσimax and Nimax may be described as
follows:
1. Short term tensile strength, involving a low number of load cycles Nimax and
highest individual loads bearable Δσmax(Nmin),
2. Fatigue strength: increasing number of load cycles Nimax, while individual loads
bearable Δσimax = Δσimax(Nimax) decrease (slope of decrease: k),
![Page 29: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/29.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
29
3. Endurance strength: Infinite number of load cycles Nimax; no damage caused by
individual loads occurring below the endurance strength Δσend(Nend), (Figure 2.1)
The damage D caused by each individual mechanical load Δσi(t) depends on the
fatigue theory itself. Thus, damage D occurring might be calculated by relating
individual number of mechanical load cycles ni(Δσi(t)) and the maximum number
mechanical load cycles Nimax(Δσi(t)):
( )maxii NnDD = , with ( )( )tnn iii σ∆= , and ( )( )tNN iii σ∆= maxmax (2.2)
This section describes different fatigue (damage accumulation) theories focussing on
ductile materials (such as steel). This is particularly important, because the application
of these theories to DH has not been updated since 1999 and most service pipes of DH
systems, as well as many relevant components of DH systems are made of steel.
Linear methods and theories (most commonly used for approximating the mechanical
ageing of DH media pipes) are compared to more complex (non-linear) theories
available today. Finally, criteria for the assessment of these theories are reported,
considering their impact in asset management and network design.
Figure 2.1: Woehler-curve of a ductile component, e. g. steel; a) Palmgren-Miner modified; b) Palmgren-Miner elementary.
2.1.1 State of the art – linear fatigue theories fo r ductile materials
Linear fatigue theories for ductile materials represent the simplest approach for the
calculation of fatigue. Linear theories directly relate the damage D to the ratio of
![Page 30: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/30.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
30
individual number of mechanical load cycles ni and maximum number mechanical load
cycles Nimax of an individual load (causing mechanical stress) σi(t):
maxii NnD Σ= (2.3)
The original linear fatigue theory for technical components bases on Palmgren
(Palmgren, 1924). Three decades later, Miner republished the concept (Miner, 1945).
Linear fatigue theories base on several assumptions:
1. Loads occurring are sinusoidal,
2. Failure occurs due to the total amount of work absorbed within the component/
specimen,
3. Work hardening is neglected (the energy absorbed during each load cycle stays
constant),
4. Failure at D = 1 is defined by crack initiation.
However, the original Miner rule does not consider any load occurring below the
endurance strength Δσend. Therefore, modifications of linear fatigue theories were
developed, considering the impact of load cycles below Δσend on the fatigue of the
component:
1. Palmgren-Miner elementary: neglecting the endurance strength Δσend of the
component, the section of fatigue strength is prolonged with the same slope k.
Thus, the influence of load cycles below Δσend on the fatigue of the component
might be considered.
2. Palmgren-Miner modified: considering the mechanisms behind crack initiation,
the section of fatigue strength is prolonged with a minor slope k* = 2k-1
(Haibach, 2006).
Summarizing, linear fatigue theories (Palmgren, 1924; Miner, 1945) directly relate
individual numbers of mechanical load cycles ni(Δσi(t)) and maximum numbers
mechanical load cycles Nimax(σi(t)). This is a straightforward approach for the
mathematical description of ageing due to stresses σi(t). However, linear fatigue
theories are state of the art in predicting damages of DH piping systems in Germany
(AGFW, 2007). This is mainly due to the simple structure of the theory and calculations,
as well as the easy handling of sequences for Δσi(t). On the other hand, influences of
![Page 31: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/31.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
31
sequential effects (high loads with initial damages of the material) are neglected,
whereas uncertainties of linear fatigue theories are quite relevant (Pöting, 2003). This is
mainly due to missing data on the variability of Nimax at the same stress Δσi(Nimax).
2.1.2 Non-linear fatigue theories for ductile mater ials
Fatigue damage is a highly complex and non-linear process. Thus, a linear description
seems to be inadequate for its description and non-linear descriptions were developed
for highly stressed components of ductile materials (e. g. steel). Thus, the development
of fatigue and damage D depends on additional material parameters. Herein, several
theories describe fatigue of ductile materials, based on:
1. several individual material parameters, that have to be measured within specific
test series, (Kujawaski and Ellyin, 1984; Leonatris, 1995; Linn and Scholz, 2013;
Miller and Zachariah, 1977; Nikbin, 2013; Schütz et al., 1983; Bernard-Connolly
et al., 1983),
2. material parameter χi that might be defined on the basis of the Woehler-curve,
(Haibach, 2006; Haibach, 1970; Subramanyan, 1976; Hashin and Laird, 1980;
Manson and Halford, 1981).
The latter theories (Bernard-Connolly et al., 1983; Haibach, 2006; Haibach, 1970;
Subramanyan, 1976; Hashin and Laird, 1980) are the main focus of this report, as no
test specimen are available for specific test series, which are not planned within this
research project.
Regarding the latter theories, fatigue and damage D occurring might be calculated by
correlating individual number of mechanical load cycles ni and maximum number
mechanical load cycles Nimax as well as the material parameter χi, whereas the material
parameter χi is defined by two-level endurance tests. Thus, fatigue may be calculated
for multi-level mechanical loads (Kujawaski and Ellyin, 1984):
1maxmax2
2
max2
2
max1
1
1
3
2
2
1
=
+
++
+
=
−
m
m
N
n
N
n
N
n
N
nD
m
m
χχ
χχ
χχ
K (2.4)
![Page 32: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/32.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
32
Thus, the simplest formulation of this equation for two-level mechanical loads may be
derived as well:
+
=
+
=
=
max2
2
max1
1
max2
2
max1
1
max1
12
11
N
n
N
n
N
n
N
n
N
nD
χχχχ
(2.5)
Based on the multi-level formulation, the definition of the material parameter χi differs
from theory to theory. Regarding the theories in focus of this research project, the
material parameter χi is defined by individual stresses occurring due to mechanical load
cycles σi(t) (Subramanyan, 1976) or by individual maximum number of mechanical load
cycles Nimax (Hashin and Laird, 1980; Manson and Halford, 1981).
2.1.2.1 Theories based on stress parameters
According to Subramanyan (1976), the material parameter χi may be defined by the
individual stress occurring at any mechanical load cycle σi(t) and the bearable stress of
the endurance strength of the material component σend. Hence, there is no
experimental effort for the definition of the material parameter χi:
endii σσ
χ∆−∆
= 1 (2.6)
and the formulation for two-level mechanical loads becomes:
+
=
∆−∆∆−∆
max2
2
max1
11
2
N
n
N
nD
end
end
σσσσ
(2.7)
Summarizing, Subramanyan (1976) directly relates individual numbers of mechanical
load cycles ni(Δσi(t)) and maximum numbers mechanical load cycles Nimax(Δσi(t)). In
addition, stress levels of previous load cycles and the stress Δσend of the endurance
strength are considered. Thus, the mathematical description of fatigue is directly related
to stresses occurring. This is a straightforward approach for the mathematical
description of ageing due to stresses Δσi(t). Non-linear fatigue theories based on
stresses are beyond state of the art in predicting damages of DH piping systems. This
is mainly due to the complex structure of non-linear theories and calculations.
![Page 33: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/33.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
33
2.1.2.2 Theories based on the number of mechanical load cycles
According to Hashin and Laird (1980), Manson and Halford (1981), the material
parameter χi may be defined by the maximum number of mechanical load cycles Nimax.
However, Hashin and Laird (1980) additionally considers the bearable number of
mechanical load cycles to cause failure at endurance strength Nend = Nmax(Δσend),
whereas the approach of Manson and Halford (1981) depends on a constant empirical
material parameter p. In both cases, there is no experimental effort for the definition of
the material parameter χi and p.
2.1.2.2.1 Theories based on endurance strength
According to Hashin and Laird (1980), the material parameter χi depends on the
number of mechanical load cycles to cause failure at endurance strength Nend, and the
maximum number of mechanical load cycles Nimax:
=
end
i
i
N
Nlog
1χ (2.8)
and the formulation for two-level mechanical loads becomes:
+
=
2
2log
log
1
11
2
N
n
N
nD end
end
N
N
N
N
(2.9)
Summarizing, Hashin and Laird (1980) directly relates individual numbers of
mechanical load cycles ni(Δσi(t)) and maximum numbers mechanical load cycles
Ni(Δσi(t)). In addition, the maximum number of previous mechanical load cycles Ni, and
the number of load cycles causing failure at endurance strength Nend(Δσend) are
considered. As thermally induced stresses are relevant for fatigue of components, this
approach seems to be not as transparent as previous approaches based on the relation
of developed stresses.
2.1.2.2.2 Theories based on empirical material para meters
According to Manson and Halford (1981), the material parameter χi depends on the
maximum number of mechanical load cycles Ni, and an additional empirical material
parameter p:
![Page 34: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/34.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
34
( )pii N=χ (2.10)
Based on empirical determinations, the additional material parameter p becomes 0.4.
Thus, the formulation for two-level mechanical loads becomes:
12
2
1
1
4.0
2
1
=
+
=
N
n
N
nD
N
n
(2.11)
Summarizing, Manson and Halford (1981) directly relates individual numbers of
mechanical load cycles ni(Δσi(t)) and maximum numbers mechanical load cycles
Ni(Δσi(t)). In addition, maximum number of previous mechanical load cycles Ni and an
empirical (and constant) material parameter (p) are considered. Again, this approach is
not as transparent as previous approaches basing on the relation of stresses occurring,
as thermally induced stresses are relevant for fatigue of components. Furthermore the
variability of the empirical material parameter p is not clarified. However, the individual
number of mechanical load cycles may be determined more easily from operational
data of DH systems than thermally induced stresses within DH systems.
Furthermore, this non-linear fatigue theory is beyond the state of the art in predicting
damages of DH piping systems. This is mainly due to the complex structure of this
theory. On the other hand, the influence of sequential effects (high loads with initial
damages of the material) is considered, whereas the handling of sequence effects is
rather complex, as each change in stress amplitude must be considered within an
individual term in the formulation. Finally, the uncertainties of this non-linear fatigue
theory are potentially minimal (Pöting, 2003). However, the latter must be clarified
within the first calculations on fatigue resistance, as shown in the following paragraph.
2.1.3 Summary and evaluation of different theories of fatigue accumulation
A summary on different theories for fatigue and damage accumulation is given in
Table 2.1. Proposed analytic evaluation criteria for the different theories are:
(i) transparency of mathematical/ physical approach, (ii) level of sophistication (in
comparison to state of the art in DH practice), (iii) usability, (iv) consideration of
sequential effects, and (v) relative accuracy in comparison to other theories (according
to literature survey). However, these criteria must be complemented according to the
results of first calculations and will finally be evaluated in the following paragraphs.
![Page 35: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/35.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
35
Table 2.1: Summary on analytic evaluation criteria for the different theories on fatigue and damage accumulation.
Theory
Basis: -Transparency of approach-
Level of Sophist.
Usability Sequential effects
Relative Accuracy,
(Pöting, 2003)
Linear Miner (1945)
Basis: no. of load cycles ni
-HIGH-
State of the art -LOW-
Easy structure
-HIGH-
Not Considered
-NO-
-LOW-
Non-linear A Subramanyan
(1976)
Basis: additional mat. Parameter χi(Δσi; Δσend)
-HIGH-
Beyond State of the art -HIGH-
complex structure
-LOW-
Considered
-YES-
-MEDIUM-
Non-linear B Hashin and
Laird (1980)
Basis: additional mat. Parameter χi(ni(Δσi); Nend)
-MEDIUM-
Beyond State of the art -HIGH-
complex structure
-LOW-
Considered
-YES-
-MEDIUM-
Non-linear C Manson (1981)
Basis: additional mat. Parameters
(empirical) χi(Ni(Δσi) p) -MEDIUM-
Beyond State of the art -HIGH-
Relatively complex structure
-MEDIUM-
Considered
-YES-
-BEST-
2.2 Estimating fatigue resistances for future load conditions
Fatigue and damage resistance according to linear and non-linear fatigue theories is
quantified illustratively within this chapter. For this purpose, a mechanical load
spectrum is utilized and modified, in order to evaluate the results obtained from
different theories on damage accumulation. Thus, fatigues occurring due to linear and
non-linear theories can be calculated, regarding:
1. The detailed load spectrum: no agglomeration of identical load cycles,
2. Sequential effects: relevant irregularities in the distribution of load cycles leading
to major differences for the damages D calculated regarding different load
scenarios:
![Page 36: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/36.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
36
a. precise mechanical load sequence representing in-situ data,
b. best-case scenario (load sequence: low-high),
c. worst-case scenario (load sequence: high-low).
3. Quantity of mechanical loads: loads occurring should be realistic, e. g. according
to AGFW (2007),
4. Typical mechanical load cycles of DH networks.
Finally, due to the special mathematical structure of non-linear theories, another major
issue must be considered regarding the material parameter χi = χi (Δσi; Δσend) or
χi = χi (Ni(Δσi)).
5. Relative characteristic of mechanical load cycle ΔΣi and ΔNi: the difference
between two consecutive mechanical load cycles Δσi - Δσi+1 and the previous
load cycle Δσi is of major importance for the extent of χi. Big gaps between two
consecutive mechanical loads will lead to an overestimation of χi (due to χi << 1),
and unrealistically high damages. Therefore, the absolute value of χi (according
to Subramanyan, 1976; Hashin and Laird, 1980; Manson and Halford, 1981)
must be restricted.
The impact of this last aspect might be regarded comparing the results on the damages
calculated for the ill- and well-conditioned load spectra. The minimum of χi is well below
0.2 for the ill-conditioned mechanical load spectrum, while being well above 0.7 for the
well-conditioned mechanical load spectrum (worst-case scenario).
2.2.1 Definition of illustrative mechanical load sp ectrum
In order to quantify fatigue due to a specific mechanical load spectrum, a Woehler-
curve has to be defined within a first step. Besides the definition of a Woehler-curve on
the basis of an experimental test series (which is not part of this research project),
these may be defined synthetically (Gudehus and Zenner, 1995; Hück et al., 1983), as
well as according to given standards and directives.
In accordance to the state of the art and practice in Germany (AGFW, 2007), a typical
Woehler-curve for the fatigue strength Δσimax = Δσimax(Nimax) of a media pipe is defined.
However, AGFW (2007) does not define any boundaries for Δσimax, such as short term
tensile strength Δσmax or endurance strength Δσend. However, these definitions are
![Page 37: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/37.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
37
necessary for the use of non-linear theories on damage accumulation. Figure 2.2
illustrates the Woehler-curve for a welded distribution pipeline made of P235GH
(1.0345), according to AGFW (2007). This Woehler-curve is supplemented by
considerations and calculations according to Hück et al. (1983).
The short term tensile strength, based on the tensile strength Δσmax of a typical media
pipe, made of P235GH (1.0345) Δσmax, may be defined by:
2max 400mm
N=∆σ (2.12)
where Δσmax defined above is validated by the definition of a Wohler-curve according to
Hück et al. (1983).
The fatigue strength Δσimax(Nmin), based on AGFW (2007), may be defined for a welded
distribution pipeline, made of P235GH (1.0345), by:
( ) 25,0
2max 67,65000 −⋅⋅=∆ ii Nmm
Nσ (2.13)
which leads to:
( ) ( ) 664,3664,367,65000400 maxmin25,0
22max =∆⇒⋅⋅==∆ − σσ Nmm
N
mm
N (2.14)
The endurance strength, based on Hück et al. (1983), the endurance strength
Δσend(Nend) may be defined for a welded distribution pipeline made of P235GH (1.0345)
by:
( ) ( ) 000,000,76067,65000 max2
25,0
2=∆⇒≈⋅⋅=∆ − σσ endendend N
mm
NN
mm
N (2.15)
Summarizing, relevant amplitudes for stresses occurring Δσi are in the range of
60 N/mm² < Δσi < 400 N/mm². Based on this Woehler-curve, mechanical load spectra
may be defined. For illustrative purposes, these load spectra must consider easy
applicability, as well as the impact of sequential effects (asymmetric mechanical load
sequence). Therefore, three different load spectra are regarded:
1. Precise mechanical load sequence representing in-situ data,
2. Best-case scenario, represented by descending mechanical loads (low-high),
![Page 38: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/38.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
38
3. Worst-case scenario, represented by ascending mechanical loads (high-low).
Figure 2.2: Woehler-curve of a welded distribution line; P235GH (1.0345) (AGFW, 2007; Hück et al., 1983).
2.2.1.1 Ill-conditioned load spectrum
The ill-conditioned mechanical load spectrum, disregarding homogeneity of loads, is
given in Figure 2.3. The load spectrum reflects typical amplitudes for stresses Δσi
within media pipes of DH systems. Based on this mechanical load spectrum, a
mechanical load sequence is given, as indicated in Table 2.2 (minimum amplitudes for
stress, 100 N/mm² = Δσmin > Δσend). Thus, fatigue and damage occurring may be
approximated and evaluated for different linear and non-linear theories. The
unfavourable relative characteristic of the mechanical load cycles is given in Table 2.2,
as well as with χi > 0.16 (Manson and Halford, 1981).
![Page 39: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/39.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
39
Figure 2.3: Illustrative example of an ill-conditioned mechanical load sequence of distribution pipeline, showing precise mechanical load sequence (above), as well as resulting best- and worst-case scenarios.
![Page 40: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/40.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
40
Table 2.2: Mechanical load sequences of ill-conditioned mechanical load amplitude and material parameters χi for different load scenarios.
Precise mechanical sequence
Δσi [N/mm²]
ni [-] Occurrence
Ni [-]
χi [-] (Subramanya
n, 1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
300 1 1h < t < 3h Ca. 11,560
0.165 0.314 0.172
100 1 4h < t < 6h Ca. 950,000
1.00 1.00 1.00
100 1 6h < t < 8h Ca. 950,000
7.33 3.49 7.42
350 1 1h < t < 9h Ca. 6,300
n.a.n. n.a.n. n.a.n.
Best-case scenario (low-high)
Δσi [N/mm²]
ni [-] Occurrence
Ni [-]
χi [-] (Subramany
an, 1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
100 2 s. above Ca. 950,000
6.06 3.19 5.80
300 1 s. above Ca. 11,560
1.21 1.10 1.28
350 1 s. above Ca. 6,300
n.a.n. n.a.n. n.a.n.
Worst-case scenario (high-low)
Δσi [N/mm²]
ni [-] Occurrence
Ni [-]
χi [-] (Subramany
an, 1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
350 1 s. above Ca. 6,300
0.827 0.912 0.781
300 1 s. above Ca. 11,560
0.165 0.314 0.172
100 2 s. above Ca. 950,000
n.a.n. n.a.n. n.a.n.
![Page 41: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/41.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
41
2.2.1.2 Well-conditioned load spectrum
The well-conditioned mechanical load spectrum regarding homogeneity of loads is
given in Figure 2.4. The load spectrum reflects typical amplitudes for stresses Δσi
within media pipes of DH systems. Based on this mechanical load spectrum, a
mechanical load sequence is given, as indicated in Table 2.3 (minimum amplitudes for
stress, 100N/mm² = Δσmin > Δσend). Thus, fatigue and accumulated damage may be
approximated and evaluated for different linear and non-linear theories. The favourable
relative characteristic of mechanical load cycles is also given in Table 2.3, with χi > 0.74
(Manson and Halford, 1981).
Figure 2.4: Illustrative example of an well-conditioned mechanical load sequence of distribution line precise mechanical load sequence (above), as well as resulting best- and worst-case scenarios.
![Page 42: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/42.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
42
Table 2.3: Mechanical load sequences of well-conditioned mechanical load amplitude and material parameters χi, for different load scenarios.
Precise mechanical sequence
Δσi [N/mm²]
ni [-] Occurrence
Ni
[-]
χi [-] (Subramanyan,
1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
300 1 1h < t < 3h Ca. 11,560
0.791 0.866 0.747
250 1 4h < t < 6h Ca. 23,650
1.00 1.00 1.00
250 1 6h < t < 8h Ca. 23,650
1.53 1.237 1.713
350 1 1h < t < 9h Ca. 6,300
n.a.n. n.a.n. n.a.n.
Best-case scenario (low-high)
Δσi [N/mm²]
ni [-] Occurrence Ni [-]
χi [-] (Subramanyan,
1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
250 2 s. above Ca. 23,650
1.26 1.13 1.34
300 1 s. above Ca. 11,560
1.21 1.10 1.28
350 1 s. above Ca. 6,300
n.a.n. n.a.n. n.a.n.
Worst-case scenario (high-low)
Δσi [N/mm²]
ni [-] Occurrence
Ni [-]
χi [-] (Subramanyan,
1976)
χi [-] (Hashin,
Laird, 1980)
χi [-] (Manson,
Halford, 1981)
350 1 s. above Ca. 6,300
0.827 0.912 0.781
300 1 s. above Ca. 11,560
0.791 0.886 0.747
250 2 s. above Ca.
23,650 n.a.n. n.a.n. n.a.n.
![Page 43: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/43.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
43
2.2.2 Quantification of fatigue resistances
This section quantifies the impact of different fatigue theories on the consumed lifetime,
as well as the deviation between these theories, while considering the impact of
sequential effects. For this purpose, fatigue occurring due the different load scenarios
are compared (precise mechanical load sequence, best-case scenario, and worst-case
scenario, as shown in Figure 2.3 and Figure 2.4).
2.2.2.1 Results for ill-conditioned load spectrum
Table 2.4 reports the damage D occurring, based on an ill-conditioned load spectrum,
for different linear (Palmgren, 1924; Miner, 1945) and non-linear theories
(Subramanyan, 1976; Hashin and Laird, 1980; Manson and Halford, 1981).
Table 2.4: Damages occurring due to the ill-conditioned mechanical load sequence in Figure 2.3, for different theories of damage accumulation.
Theory Precise sequence Best-case scenario Worst-case scenario
D = D(ni; Ni(σi)) Linear Miner (1945) D = 0.165%
D = D(ni; Ni(σi);χi(σi; σend)) Non-linear A Subramanyan
(1976) D = 0.119% D = 0.119 % D = 40.32 %
D = D(ni(σi); Nend; χi(ni(σi); Nend)) Non-linear B Hashin; Laird
(1980) D = 0.135 % D = 0.135 % D = 15.30 %
D = D(ni; Ni(σi); χi(Ni(σi); p)) Non-linear C Manson (1981) D = 0.114 % D = 0.114 % D = 40.57 %
2.2.2.2 Results for well-conditioned load spectrum
Table 2.5 reports the damages D occurring, based on an well-conditioned load
spectrum, for different linear (Palmgren, 1924; Miner, 1945) and non-linear theories
(Subramanyan, 1976; Hashin and Laird, 1980; Manson and Halford, 1981).
![Page 44: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/44.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
44
Table 2.5: Damages occurring due to the well-conditioned mechanical load sequence in Figure 4 for different theories for damage accumulation.
Theory Precise sequence Best-case scenario Worst-case scenario
D = D(ni; Ni(σi)) Linear Miner (1945) D = 0.221 %
D = D(ni; Ni(σi);χi(σi; σend)) Non-linear A Subramanyan
(1976) D = 0.123 % D = 0.121 % D = 1.34 %
D = D(ni(σi); Nend; χi(ni(σi); Nend)) Non-linear B Hashin; Laird
(1980) D = 0.150 % D = 0.146 % D = 0.554 %
D = D(ni; Ni(σi); χi(Ni(σi); p)) Non-linear C Manson (1981) D = 0.116 % D = 0.115 % D = 2.06 %
2.2.3 Summary on the quantification of fatigue and interpretation of results
Results for the ill- and well-conditioned load spectrum are discussed for linear and non-
linear theories. Herein, the dependency of the results on the highest load cycles within
the load spectrum is analyzed. For this purpose, the damage calculation considers the
impact of the highest mechanical load cycles on the overall damage D due to:
1. The singular load cycle of 350N/mm² (D350),
2. The singular load cycle of 300N/mm² (D300),
3. Both highest load cycles (D350&300), and the remaining damage that occurs due to
the lowest load cycles D250 = D - D350&300 (D from Table 2.5).
![Page 45: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/45.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
45
2.2.3.1 Overall results
Based on the results of Table 2.4 and Table 2.5 for the ill- and well-conditioned load
spectrum, the damages D calculated depend on the load scenario, and load spectrum:
1. Precise mechanical load scenario:
Damages D calculated are in a narrow range for all theories and spectra (ill-,
well-conditioned) considered. Comparing different theories, deviations for D are
higher for the well-conditioned load spectrum. D is in a realistic range for all
theories.
2. Best-case scenario:
Damages D calculated are in a narrow range for all theories and spectra (ill-,
well-conditioned) considered. Comparing different theories, deviations for D are
higher for the well-conditioned load spectrum. D is in a realistic range for all
theories. Regarding all load spectra, sequential effects have a minor impact on
D. The latter is slightly below the damage values for the precise mechanical load
scenario.
3. Worst-case scenario:
Damages D calculated are significantly higher for all non-linear theories, and in
comparison to other load scenarios. Comparing different theories, deviations for
D are in the same range for all load spectra. D is realistic for the well-conditioned
load spectrum, and out of any realistic range for the ill-conditioned load
spectrum. Regarding the ill-conditioned load spectrum, D strongly depends on
sequential effects by a factor superior to 100, whereas this effect is diminished
for the well-conditioned load scenario (by factor inferior to 15). D is higher for the
ill-conditioned load spectrum than for the well-conditioned load spectrum.
2.2.3.2 Impact of highest load cycles on overall damage D
Following considerations are limited to the well-conditioned load spectrum, as results of
the ill-conditioned load spectrum are questionable (due to the relative characteristic of
load occurring):
1. Despite of the lower loads, damages D calculated for the ill-conditioned
mechanical load spectrum are higher than for the well-conditioned load
spectrum.
![Page 46: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/46.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
46
2. Damages D calculated for the ill-conditioned load spectrum are out of any
realistic range.
3. Damages D calculated for the ill-conditioned load spectrum strongly depend on
the lowest loads due to sequential effects, which is in direct contradiction to any
physical considerations.
The impact of the highest mechanical load cycles (350N/mm² and 300N/mm², as well
as both loads in combination) on the overall damage D calculated is given in Table 2.6.
Table 2.6: Damages occurring due to singular load cycles (D350 & D300) and both
highest load cycles (D350&D300) for the well-conditioned load spectrum.
Theory Precise sequence
Best-case scenario
Worst-case scenario
D = D(ni; Ni(σi))
D350 = 0.107 %
D300 = 0.058 %
Linear Miner
(Palmgren, 1924; Miner, 1945)
D350&300 = 0.163 % → D250 = 0.058
D = D(ni; Ni(σi);χi(σi; σend))
D350 ≈ 0.107 % D350 = 0.107 %%
D300 ≈ 0.058 % D300 = 0.058 %
Non-linear A Subramanyan
(Subramanyan, 1976)
D350&300 ≈ 0.119 % → D250 = 0.004 %
D350&300 = 0.456 % → D250 = 0.884 %
D = D(ni(σi); Nend; χi(ni(σi); Nend))
D350 ≈ 0.107 % D350 = 0. 107 %
D300 ≈ 0.058 % D300 = 0.058 %
Non-linear B Hashin; Laird (Hashin and Laird, 1980)
D350&300 ≈ 0.135 % → D250 = 0.015 %
D350&300 = 0.252 % → D250 = 0.302 %
D = D(ni; Ni(σi); χi(Ni(σi); p))
D350 ≈ 0.107 % D350 = 0.107 %
D300 ≈ 0.058 % D300 = 0.058 %
Non-linear C Manson
(Manson and Halford, 1981)
D350&300 ≈ 0.114 % → D250 = 0.002 %
D350&300 = 0.534 % → D250 = 1.526 %
![Page 47: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/47.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
47
Based on these results for the well-conditioned load spectrum, the damage D
calculated depends on the load scenario and theory considered:
1. Precise mechanical load scenario:
(i) Damages D350 and D300 are identical for all theories,
(ii) D350&300 is highest for the linear theory,
(iii) Considering all theories, D350&300 are in a quite narrow range for all theories
(deviation < 45% between (Palmgren, 1924; Miner, 1945) and (Manson and Halford,
1981)),
(iv) Whereas non-linear effects result in diminished damages D350&300 (compare D350&300
of (Palmgren, 1924; Miner, 1945) to D350&300 of (Subramanyan, 1976; Hashin and Laird,
1980; Manson and Halford, 1981)).
2. Best-case scenario:
(i) Damages D350 and D300 are identical for all theories,
(ii) D350&300 is highest for the linear theory. Considering all theories, D350&300 are in a
quite narrow range for all theories (deviation < 45% between (Palmgren, 1924; Miner,
1945) and (Manson and Halford, 1981)),
(iii) Whereas non-linear effects result in diminished damages D350&300 (compare
D350&300 of (Palmgren, 1924; Miner, 1945) to D350&300 of (Subramanyan, 1976; Hashin
and Laird, 1980; Manson and Halford, 1981)),
(iv) Considering non-linear theories, D350&300 deviates insignificantly between the best-
case scenario, and the precise mechanical load scenario (deviations < 2% for
(Subramanyan, 1976; Hashin and Laird, 1980; Manson and Halford, 1981)).
3. Worst-case scenario:
(i) Damages D350 and D300 are identical for all theories,
(ii) D350&300 is higher for non-linear theories and highest for the linear theory according
to (Manson and Halford, 1981),
(iii) Considering non-linear theories, D350&300 is in a broader range (deviations up to
about 120% between (Subramanyan, 1976) and (Manson and Halford, 1981)),
(iv) Non-linear effects result in enhanced damages D350&300 (compare D350&300 of
(Palmgren, 1924; Miner, 1945) to D350&300 of (Subramanyan, 1976; Hashin and Laird,
1980; Manson and Halford, 1981), while D350 + D300 < D350&300, for all non-linear
theories),
![Page 48: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/48.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
48
(v) Considering non-linear theories, D350&300 deviate significantly between the worst-
case scenario and the precise mechanical load scenario (deviations up to ca. 370% for
(Manson and Halford, 1981)).
2.2.4 Interpretation of results and evaluation of u sability – critical aspects
The damage D occurring due to different load sequences (ill- and well- conditioned),
and scenarios (precise mechanical load sequence, as well as best- and worst-case
scenario for the same load sequence) has been examined.
1. Regarding best-case scenarios, the impact of sequential effects on the damage
D calculated is of minor significance (minor impact of sequential effects). This
has been confirmed by previous examinations on more complex load spectra at
FFI.
2. Regarding best-case scenarios, and precise mechanical load sequences, the
impact of high loads on damages D calculated, is disproportionately high. Thus,
the damage D calculated within the illustrative example depends on high loads
by more than 90%. This impact is of the same magnitude as within linear
theories.
3. Regarding the worst-case scenarios, the impact of sequential effects on the
damage D calculated is of major significance (major impact of sequential
effects). This also has been confirmed by previous examinations on more
complex load spectra at FFI.
4. Regarding the worst-case scenarios, the impact of low loads on damages D
calculated, is disproportionately high. Thus, the damage D calculated within the
illustrative example depends on low loads by more than 50% (up to 75%). This
impact seems to be realistic, comparing it to the impact of low loads within linear
theories.
Summarizing, non-linear theories seem to be generally suitable for the description of
mechanical ageing of media pipes within DH systems. Thus, these theories may deliver
more realistic results for damages occurring due to mechanical loads. On the other
hand, some critical aspects concerning the use of non-linear theories, the impact of
sequential effects, and the transferability of non-linear theories to DH systems have
been identified:
![Page 49: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/49.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
49
1. Critical aspect I: implementation of operational data for non-linear theories.
Linear and non-linear theories on damage accumulation deliver realistic results for
the damage prediction. However, special attention must be given to the conditioning
(esp. relative characteristic) of the loads occurring. Bigger gaps between following
load cycles result in unrealistically high damage D, as shown in the worst-case
scenario, for the ill-conditioned load spectrum. However, as load cycles in-situ can
not be conditioned according to this mathematical stipulation, this might be
problematic in implementing operational data for non-linear theories, applying to
DH-networks.
2. Critical aspect II: Unrealistic impact of sequential effects.
The impact of favorable and unfavorable sequential effects is considered by all non-
linear theories. Due to these effects, the damage D, calculated due to the lowest
mechanical loads, increases significantly for every load scenario. Thus, the damage
occurring due to low-load cycles is 20 to 750 higher (Subramanyan, 1976; Hashin
and Laird, 1980), comparing the results for D250 in Table 2.6.
3. Critical aspect III: Transferability of non-linear theories.
The transferability of non-linear theories on media pipes of DH-systems must be
discussed, as non-linear theories have been developed for highly-stressed
components, e. g. Bernard-Connolly et al. (1983), typically based on two-level
endurance tests, whereas media pipes in DH systems undergo limited multi-level
stresses.
Finally, regarding literature and obtained results, the presumed better relative accuracy
of Manson and Halford (1981), according to Pöting (2003), cannot be verified (as
evident in critical aspect II). Therefore, in future steps, the results from all non-linear
theories (Subramanyan, 1976; Hashin and Laird, 1980; Manson and Halford, 1981)
should be considered, and compared to those of linear theories (Palmgren, 1924; Miner,
1945) in order to compare, and validate the obtained results.
2.3 Compilation of shear strength data from natural ly aged pipes
Varying mechanical and thermal loads have a negative effect on the lifetime and
usability of any component within technical systems. Regarding DH systems,
mechanical and thermal loads influence the ageing of media pipes, thermal insulations
and casing. This section focuses on the shear strength of pre-insulated pipe systems,
![Page 50: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/50.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
50
as this parameter is of major importance for pipe stability, and reliability of supply on a
long-term perspective. First, the axial shear strength data from naturally aged pipes
examined at FFI are compiled and analysed. Herein, the parameters influencing the
axial shear strength of naturally aged pre-insulated piping systems are identified and
described. Secondly, referring to the test results of the artificially aged pipes at FFI, the
impact parameters on the evolution of the axial shear strength of the thermal insulation,
are determined.
2.3.1 Naturally aged pipes
The FFI has conducted studies on the composite properties of naturally aged
preinsulated bonded DH pipe. The 16 examined pipes were exposed after different
operating time from six different places.
First, the axial shear strength of the pipes in naturally aged condition was determined
according to DIN EN 253 5.4 (DIN EN 253, 2009. The axial length of the test specimen
is equal to 2.5 times the polyurethane insulation thickness, but at least 200 mm. The
tests were carried out on a system with a 100 kN load cell (accuracy class 0.1). The
force was applied centrally, in the vertical axis direction, to the service pipe of the test
specimen, at a speed of 5 mm/min, while the jacket casing likewise rested centrically
on the abutment. The axial displacement of the service pipe during the force application
was monitored with an inductive position sensor (accuracy class 0.2). The test
temperature was 23±2°C.
The recording of the measured results during the test took place by means of electronic
data processing. The amount of the axial shear strength was calculated according to
the equation below, considering the dead weight of the service pipe in the axial force
Fax:
πτ
⋅⋅=
s
axax DL
F (2.16)
where
τax axial shear strength, in MPa; Fax axial force, in N; L length of specimen, in mm; Ds outside diameter of the service pipe, in mm
![Page 51: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/51.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
51
All available important data of the 16 pipes, as well as the measured axial shear
strength, and the foam density in naturally aged condition are summarized in Table 2.7.
On this basis, the following analysis was carried out to determine the relationship
between the axial shear strength and the various and possible influencing parameters.
Table 2.7: Overview of the 16 pipe samples.
1-1 1-2 2-1 2-2 3-1 3-2 4-1 4-2
excavation date 06.2006 11.2007 12.2007 10.2007
installation date 1992 12.1984 1986 1997
operating time (year) 15,5 23 22 11
DN steel medium pipe (-) 65 150 150 250
sample length (mm) 140 250 250 400
overlap height (m) 0,95 0,95 1,1 1,0
design temperatur (°C) 130 130 130 130
design pressure (bar) 16 25 25 16
supply pipe (x) X X X X
return pipe (x) X X X X
axial shear strength at
room temperature (MPa) 0,172 0,253 0,225 0,274 0,213 0,184 0,110 0,204
foam density (kg/m³) 63,1 77,5 99,9 98,5 59,1 62,5 75,7 64,8
5-1 5-2 5-3 5-4 6-1 6-2 6-3 6-4
excavation date 12.2009 06.2012
installation date 1985 2003
operating time (year) 25 10
DN steel medium pipe (-) 150 150
sample length (mm) 250 250
overlap height (m) 0,5 1,2
design temperatur (°C) 120 130
design pressure (bar) 25 25
supply pipe (x) X X X X
return pipe (x) X X X X
axial shear strength at
room temperature (MPa) 0,120 0,147 0,336 0,297 0,185 0,232 0,299 0,255
foam density (kg/m³) 81,2 103,3 70,6 87,9 78,7 64,3 76,8 65,6
S1 S2 S3 S4
S5 S6
Figure 2.5 shows the measured axial shear strength in naturally aged condition,
depending on the type of pipe (supply or return). With one exception, the supply line
has lower values than the return line. On average, the axial shear strength of the supply
line is 0.176 MPa, while that of the return line is 0.263 MPa. This difference shows that
thermal loading is an important factor influencing the service life of the pipeline. For
![Page 52: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/52.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
52
comparison, the minimum requirement for axial shear strength according to DIN
EN 253 for the pipeline in unaged condition (0.12 MPa at test temperature 23±2°C) was
also indicated. Of the 16 pipes, two supply pipes (S4-1 after 11 operating years, S5-1
after 25 operating years) are below or equal to the limit value.
Figure 2.5: The axial shear strength versus the pipe type.
Figure 2.6: The axial shear strength versus the pipe size.
Figure 2.6 represents the axial shear strength as a function of the pipe size. Clearly, 12
pipes of the 16 test samples have a nominal diameter of 150 mm, which is a typical
nominal size of most built-in district heating pipelines. The measured values are
relatively widely scattered between 0.12 MPa and 0.34 MPa. Despite the lack of
sample size for smaller and larger dimensions, the thin pipe has the greater residual
![Page 53: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/53.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
53
axial shear strength values than the thick pipe (Figure 2.6). Similar results were also
reported in (Meigen and Schuricht, 2004; Schuricht, 2004).
Figure 2.7 represents the measured axial shear strengths as a function of the operating
time. Evidently, the supply pipe exhibits a lower residual axial shear strength than the
return pipe. Considering a potential line for the supply and return data, the two lines are
almost horizontal and parallel. Thus, in this case, there seems to be no correlation
between the residual axial shear strength and the operating time.
Figure 2.7: The axial shear strength versus the operating time.
In addition to the axial shear strength, the foam density was determined according to
DIN EN 253. Figure 2.8 shows the measured axial shear strength versus the measured
foam density. The foam density of the investigated pipes was very different, but all
meet the requirements of DIN EN 253 (the minimum requirement is 55 kg/m³). A
relationship between the residual axial shear strength and the foam density could not
be directly deduced.
![Page 54: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/54.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
54
Figure 2.8: The axial shear strength versus the foam density.
All specimens were tested until the foam failed and cracked. Table 2.8 summarizes the
positions of the fracture in the axial shear strength test. All of the samples are produced
with the discontinuous processing method, while most of the fracture occurs in the first
cell layers near the steel-service-pipe. Only in the case of sample 6, the break occurred
both in proximity of the service pipe, and in the polyurethane insulation foam (Table
2.8).
![Page 55: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/55.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
55
Table 2.8: Manufacturing process, and the break-position of the 16 pipe samples.
Pipe Manufacturing process Break
S 1-1 Disconti PUR/St
S 1-2 Disconti PUR/St
S 2-1 Disconti PUR/St
S 2-2 Disconti PUR/St
S 3-1 Disconti PUR/St
S 3-2 Disconti PUR/St
S 4-1 Disconti PUR/St
S 4-2 Disconti PUR/St
S 5-1 Disconti PUR/St
S 5-2 Disconti PUR/St
S 5-3 Disconti PUR/St
S 5-4 Disconti PUR/St
S 6-1 Disconti PUR/St & PUR/PUR
S 6-2 Disconti PUR/St & PUR/PUR
S 6-3 Disconti PUR/St & PUR/PUR
S 6-4 Disconti PUR/St & PUR/PUR
PUR/St: Break in the first cell layers near the steel-medium-pipe
PUR/PUR: Break in the polyurthane foam insulation
2.3.2 Artificially aged pipes
To estimate the remaining service life of the exposed pipes, the 16 pipes at the FFI
were subjected to the accelerated thermal aging, according to DIN EN 253, 5.4.3.
The service pipe was heated during the artificial aging at a temperature of 160°C,
during which the casing was maintained at a room temperature. After 600 h (25 days),
1200 h (50days), 1800 h (75 days) aging, one test specimen was cut from the pipeline,
and the axial shear strength was determined again at room temperature (23 ± 2°C).
This allowed to determine the evolution of axial shear strength as a function of the
aging time, until reaching the critical shear strength.
The measured axial shear strength after artificial aging, for the pipe specimens are
plotted in Figure 2.9 (for each pipe), and in Figure 2.10 (for all 16 pipes). The evolution
of the shear strength is very individual. However, the average bond strength was
greatly reduced after aging for 600 hours, while achieving the standard limit (DIN EN
253) after 1200 hours of aging.
![Page 56: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/56.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
56
Figure 2.9: The measured axial shear strength of the artifically aged pipes versus the ageing time.
![Page 57: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/57.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
57
Figure 2.10: The measured axial shear strength of all artifically aged pipes versus the ageing time.
2.3.3 Summary
Despite the lack of data of the operating temperature history, and the axial shear
strength of the pipe in unaged condition, the enclosed investigation showed a
significant influence of the thermal load on the axial shear strength, as also noted in
(AGFW, 2011; Meigen and Schuricht, 2004; Schuricht, 2004).
In addition to the thermal load, other parameters, including age, pipe size, and foam
density were analyzed. Despite the small number of samples, the results showed a
large scattering. Therefore, it is very difficult to derive a direct relationship between the
residual shear strength and the operating time, as well as the foam density.
From the test result, it is strongly recommended to keep a documentation of the new
pipe well before built-in. Moreover, as much data as possible should be collected during
operation (Herbst, 2015). On the basis of this data, it would possible to determine the
specifications for a thermal aging test procedure in the future, and derive the long term
temperature resistance of preinsulated bonded pipe. This would lead to more accurate
forecasts of the status change in district heating networks, and produce optimized plans
for maintenance and refurbishment measures.
![Page 58: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/58.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
58
3 Investigation of thermal ageing in combination wi th cyclic
mechanical loads
3.1 Background
Mechano-chemical degradation of polymers is well-known, but poorly understood. It
comprises of all aspects of stress-induced scissions of chain molecules (Kausch, 1986;
Terselius et al., 1986). In the presence of oxygen, the radicals formed react
immediately with oxygen, followed by rearrangement reaction which make the break
permanent. All polymers are strongly anisotropic at the molecular level (1-10 nm),
where the material properties vary with the different crystallographic orientations.
Generally, two failure mechanisms can occur: chain slippage and chain scission. Many
parameters, beside the stress level and loading time, affect these mechanisms, such as
degree of chain orientation, presence of entanglements and / or cross-links (Andrews,
1969). The kinetic theory of polymer fraction, the effect of motion, and physical
properties of molecules on the macroscopic behaviour have been studied during five
decades. The effect of deformation and rupture of molecular chains, crystals and
morphological structures on the strength of the polymeric materials in certain
applications have been investigated (Kausch, 1986).
Thermal degradation, and additional mechanical stress have a synergy effect on
molecular rearrangements, which may cause disentanglements and slips of chain
segments to local reorientations and formations of voids. Three principal mechanisms
may contribute to fatigue failure: thermal softening, excessive creep or flow, and the
initiation and propagation of fatigue cracks (Kausch, 1986). Moreover, fatigue cracks
also depend on stress levels, strain amplitudes, deformations, frequency, ambient and
internal temperatures (Andrews, 1969).
The consequence of chain scissions, reduction of molecular weight, and radical
formations in polymeric material results in accelerated degradation by e.g. oxidative
degradation, and in some cases by cross-linking (Casale et al., 1975; Lauer, 1975).
These parameters intensify by the ageing of the polymeric materials during their service
life. Time and temperature are crucial parameters in environmental degradation for
polyurethane in an application such as DH pipes.
The estimation of the lifetime of DH pipes is commonly based on accelerated thermal
ageing, regardless of other influencing factors. The objective of this chapter is to
![Page 59: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/59.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
59
investigate the ageing of DH pipes subjected to both cyclic mechanical loads and
elevated temperatures, by focusing on polyurethane degradation.
3.2 Choice of objects and conditions
The analysis was performed with the following choices and conditions.
Object: the investigations were carried out using straight DH-pipes (DN 50/160 mm),
which are usually employed for 3GDH, and composed of a steel pipe as a service pipe,
rigid polyurethane foam as insulation, and a high density polyethylene pipe as a casing.
The DH pipes used for the experiments were produced by Powerpipe Systems AB
using a discontinuous processing method. These new samples were used in order to
have the same history for the tested pipe with and without load, as well as for the pipes
aged at different temperatures.
Mechanical load: a varying axial load was applied to simulate the shear occurring in the
expansion zone of a pipe in the ground. The maximum shear stress occurring at the
interface of the service pipe and the foam was 31 kPa, and the time period of the cycle
was 1 hour.
Conditions: The ambient temperature was 23 ± 2°C, and internal temperatures in the
service pipe were 130, and 140 ± 2 ⁰C, respectively.
3.3 Selection of mechanical loading
In the sliding zones, the pipe assemblies are exposed to high temperature and
mechanical loading, due to varying operating temperature and thermal expansion. In
these zones, the axial forces vary along the pipe due to friction and sliding. The soil
friction causes axial shear stresses in the pipe assembly, reaching a maximum in the
polyurethane (PUR) insulation, at the interface with the service pipe. Here, the PUR
insulation also degrades faster, given the higher temperature. The failure mode
investigated is the loss of adhesion between the PUR insulation and the steel service
pipe. This could cause thermal expansion of the steel service pipe, leading to fatigue
failure at critical locations like joints or weld defects. Hence, the end of life of a pipe
segment is due to that failure mode. The shear stresses only depend on the friction
force per unit length of pipe, which in turn depends on the friction coefficient, and the
soil pressure on the pipe. It is assumed that the pipe is buried above the ground water
level, so that the ground water pressure does not act on the casing.
![Page 60: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/60.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
60
In the experiments, pipes of dimension DN 50/160 mm were chosen, where DN 50
refers to the service pipe, and 160 mm to the casing. The reasons for this choice are
that it has been used previously in two research projects (Yarahmadi et al., 2017a;
2017b), it is commonly used, and it is manageable in the laboratory. The chosen length
of the pipes was 3.4 m. Table 3.1 presents the input parameters for calculating a
reasonable magnitude of the axial load in the experiments. The calculations follow the
equations given in the standard EN 13941 (2019). Different soil covers are used in the
participating countries Korea, Germany and Sweden. The maximum applied soil cover
in Korean and Germany is about 1.5 m and in Sweden the soil cover is usually 0.6 m.
In the calculations the soil cover 1.2 m was chosen. The friction coefficient is usually
about 0.4, but the maximum according to standard is 0.6, which was chosen in the
present analysis. These choices will give a relatively high value on the friction force.
Table 3.1: Input data for calculation of axial force in experiments.
Parameter Symbol Value Unit Friction coefficient µ 0.6
Soil pressure coefficient K0 0,5
Weight density of soil γ 20 kN/m3
Density of water ρw 1000 kg/m
3
Density of steel ρs 7850 kg/m
3
Density of polyethylene ρpe
960 kg/m3
Density of polyurethane ρpur
60 kg/m3
Cover HC 1.2 m
Length of pipe in experiment 3.4 m Axial force F 19 kN
The pipe with the chosen dimension may be used for a house connection. In
EN 13941 (2019), the estimated number of full thermal cycles with the range 110°C is
1000 during a service life of 30 years. The house connection is subjected to a spectrum
of different thermal cycle ranges. The temperature range influences the maximum pipe
axial force and the length of the expansion zone, i.e., the length of deformed pipe
during the cycle. Here, a simplified case is treated, and a representative temperature
range is assumed to be half the range of the full cycle. The fatigue curve of the welded
service pipe can be written:
![Page 61: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/61.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
61
b
S
kN
= (3.1)
where, N is the number of cycles until failure at the stress range S. The shape of the
fatigue curve is given by the parameters k and b. According to EN 13941 (2019), the
exponent is equal to b = 4. Hence, the corresponding number of cycles with half the
range during the service life of 30 years is:
( ) 16000551101000 40 ==N (3.2)
For the experiments, the chosen axial load range is 20 kN, while the time period of the
load cycle is one hour. This means that the desired number of load cycles of the
experiments can be reached during less than two years.
3.4 Experiments
Thermal ageing at relative high temperatures is a common method to estimate the
technical lifetime of DH pipes. According to EN 253, temperatures around 160 and
170°C are suitable for accelerated thermal ageing. However, in a previous project
(Yarahmadi et al., 2017a; 2017b), it was shown that it is important to choose the correct
ageing temperature. Too high temperature activates other degradation mechanisms
which are not relevant for the district heating application.
In the previous project (Yarahmadi and Sällström, 2015), DH pipes were aged at 130,
150, and 170°C, while the mechanical adhesion between the steel pipe and the PUR
foam was studied by testing and measuring the shear strength. The results suggested
a different behaviour for ageing temperatures equal to 150°C and higher, compared to
130°C. Figure 3.1 shows that these results can not be used to determine the activation
energy Ea in the Arrhenius relationship:
−⋅=RT
EAk aexp (3.3)
The reason is the different behaviour of the shear strength of the pipe aged at 130°C,
compared to the pipes aged at higher temperatures.
![Page 62: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/62.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
62
In the current project, five DH pipes with dimension DN50/160 were prepared for both
studying thermal ageing at elevated temperatures, and thermal ageing in combination
with mechanical axial loads. One pipe was kept at room temperature as a reference.
Four pipes were connected to the electrical installation for thermal ageing, and two of
them were also connected to the axial loading equipment during thermal ageing, as
shown in Figure 3.2. The piston acts on the service pipe, and the cylinder is connected
with bars to several axial positions to the casing. The selected ageing temperatures are
130 and 140°C, based on the results from the previous project (Yarahmadi et al., 2017a;
2017b).
Two electrical cylinders were programmed to apply a cyclic axial load of range 20 kN,
as shown in Figure 3.3.
All pipes were tested for mechanical adhesive strength, and some of the obtained PUR
samples were analysed using Fourier transform infrared (FTIR) technique.
Figure 3.1: Mechanical adhesive strength results expressed in percentage of the initial break torque (Yarahmadi et al., 2017b).
![Page 63: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/63.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
63
Figure 3.2: Schematic electrical and mechanical installation for testing of DH pipes.
Figure 3.3: Schematic axial load cycle.
3.4.1 Mechanical testing
The adhesive strength was measured as shear strength that was determined by
applying a torque to an uncovered polyurethane cylindrical plug (drill core), attached to
the service pipe, as shown in Figure 3.4.
The fracture occurs between the service pipe and the polyurethane foam (adhesion), or
in the foam (cohesion) close to the service pipe, when a cylindrical specimen is twisted
off, as the torque is measured.
The shear strength is defined as the ultimate shear stress τu occurring at the
circumference of the plug. It is calculated by use of:
3
16d
Mu π
τ = (3.4)
![Page 64: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/64.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
64
where M is the applied maximum torque, and d is the diameter of the plug.
Figure 3.4: Sketch of mechanical testing.
3.4.2 Fourier Transform Infrared spectroscopy
Changes in the PUR foam’s chemical structure because of ageing can be identified by
FTIR spectroscopy analyses. Some reactions between polyol, NCO, and the additives
are not completed during the PUR manufacturing process. Thus, the reactions can take
place during the early phase of application, leading to property changes of the PUR
foam. An important characteristic of PUR foam is the so-called NCO index, which is a
measure of the amount of NCO used, compared to the stoichiometric amount required.
Foams with a very high NCO index (large excess of NCO) have a large proportion of
the trimetric cyclic isocyanurate structure, which increases the thermal stability of the
polymer.
This section describes the effect of thermomechanical exposure on chemical structure
of PUR in pre-insulated DH pipe, analyzed using the FTIR spectroscopic method.
Some PUR plugs from the mechanical tested samples were chosen for analysis of their
chemical structure. In all PURs, the repeating unit is the urethane linkage produced
from the reaction of an isocyanate (–N=C=O) with an alcohol (–OH). The urethane
group (–NH–CO–O–) is the weak link in PUR material, regarding degradation process.
In this investigation, degradation of the urethane group was used as an indicator for
changes in the chemical structure of PUR. Thermo-oxidative stability of PUR depends
on its structure, especially the chemical structure of the polyol. The intensity of the
![Page 65: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/65.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
65
methylene diphenyl diisocyanate (MDI) aromatic ring deformation (aromatic ν(C=C)
vibration) at 1595 cm–1 was used as the internal reference during usage of Fourier
transform infrared spectroscopy (FTIR) in attenuated total reflection (ATR) mode.
3.5 Results
The DH pipes have been aged according to the description in the previous chapter, and
the mechanical adhesive strength was evaluated using the RISE plug method, after
different thermal and mechanical ageing intervals. Figure 3.5 shows the installation of
the DH pipes in the test chamber, and Table 3.2 summarizes the test information.
Figure 3.5: Electrical and mechanical test installation.
Table 3.2: Test information.
Pipe name Mechanical ageing by Cyclic load (kN)
Thermal ageing ( °C)
Reference 0 23 DHP130 0 130
DHP130 load 0-20 130 DHP140 0 140
DHP140 load 0-20 140
![Page 66: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/66.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
66
The electrical cylinders were intended to follow the top curve (blue curve in Figure 3.6).
However, no compensation for the relaxation of the polyurethane was considered,
resulting in a decrease of the load (middle and bottom curves for the pipes exposed to
130 and 140ºC, respectively, in Figure 3.6) during the interval of 28 min. Instead of
controlling the force after the desired maximum was reached, the displacement was
kept constant by applying a brake to the piston of the loading cylinder. The reason for
this was to minimize the wear of the loading cylinder.
Figure 3.6: Axial load cycle between 0 and 20 kN with a total cycle time of 58 min.
3.5.1 Mechanical adhesive strength
The testing periods were chosen to represent significant events observed in the
previous project (Yarahmadi et al., 2017a; 2017b) at similar temperatures, and are
presented in Table 3.3. The number of cycles, and the ageing time differ, since the
thermal ageing has been running more or less continuously, but there have been
interruptions of the mechanical loading.
![Page 67: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/67.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
67
Table 3.3: Test time for both temperatures.
DHP130 DHP140 Ageing time (h) # Load cycles Testing date Ageing time (h) # Load cycles Testing date
720 637 2018-07-27 529 565 2018-07-19
1320 1106 2018-08-21 1008 992 2018-08-09
3600 2550 2018-12-10 2136 2042 2018-09-25
5064 4000 2019-02-11 2976 2850 2018-11-01
7176 5536 2019-05-17 4800 4350 2019-01-17
8112 6536 2019-06-28 6072 5614 2019-03-18
8088 7114 2019-06-12
The results obtained from the mechanical tests are presented in Figure 3.7, Figure 3.8,
Figure 3.9, and Figure 3.10.
Figure 3.7: Mechanical adhesive strength from all four DH pipes aged at 130 and 140°C with and without loads.
![Page 68: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/68.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
68
Figure 3.8: Mechanical adhesive strength from DH pipes aged at 130 and 140°C with load.
Figure 3.9: Mechanical adhesive strength from DH pipes aged at 140°C with/without load.
![Page 69: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/69.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
69
Figure 3.10: Mechanical adhesive strength from DH pipes aged at 130°C with/without load.
The experimental results confirmed the mechano-chemical effect, where a faster
deterioration of the mechanical adhesive strength was obtained by combining thermal
and mechanical ageing, using axial loading and elevated temperature simultaneously.
3.5.2 Fourier-transform infrared spectroscopy
Some PUR plugs from the mechanical tests were chosen for analysis of their chemical
structure. The purpose was to study the effect of synergistic degradation mechanisms
by thermal and mechanical ageing on the chemical structure of the PUR material. Two
slices from each selected PUR plug were studied. One slice was taken near to the
contact surface between PUR and the steel service pipe. Another slice was taken 20
mm above this contact surface, as shown in Figure 3.11.
Figure 3.12 shows a FTIR spectrum from a PUR sample. The analysis focuses on the
following peaks:
• C-H in methyl at 2975 cm-1
• Unreacted NCO at 2277 cm-1
• C=O in the urethane group at 1712 cm-1 and N – H at 1512 cm-1
• Aromatic rings C=C at 1595 cm-1
![Page 70: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/70.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
70
• Isocyanurate rings at 1411 cm-1
(a) (b)
Figure 3.11: PUR plugs from mechanical tests (a), and PUR slices for FTIR test (b).
Figure 3.12: FTIR spectrum of a PUR sample.
For a correct comparison, the absorbance values of the relevant peaks were
normalized using the peak at 1595 cm-1 corresponding to the aromatic ring, regarded
as the internal reference. The normalized results are presented in Figure 3.13 and
Figure 3.14. Differences between the carbonyl peaks at 1712 cm-1 from samples close
to the steel pipe (2 mm from the steel pipe), aged at 130°C, loaded, after 150 and 211
![Page 71: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/71.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
71
days (Figure 13a), and corresponding pipe aged at 130°C, unloaded, (Figure 13c) are
obvious. Similar differences are measured using pipes after ageing at 140°C (Figures
14a and 14c). The PUR degradation is also observed in the changes in peak intensity
at 1512 cm-1 (Figures 3.13 and 3.14). The changes increased with increasing ageing
time. The FTIR analyses strongly indicate that the combined effect of cyclic mechanical
loading, and thermal ageing accelerates the rate of chemical degradation of PUR foam,
reducing its shear strength.
(a) (b)
(c) (d)
Figure 3.13: FTIR analysis of the DH pipes aged at 130 °C with and without load.
![Page 72: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/72.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
72
(a) (b)
(c) (d)
Figure 3.14: FTIR analysis of the DH pipes aged at 140°C, with and without load.
The results from the shear strength measurements confirmed the results from previous
investigations with thermal ageing at 130°C, showing after the initial part of ageing, that
the adhesive strength remains unchanged for a very long period (plateau phase). This
investigation has also confirmed that thermal ageing at 140°C exhibits the same type of
ageing behaviour, but at a higher rate of deterioration. In the pipes exposed to a
combination of thermal ageing and axial loads, the first part of the ageing process
observed in thermal ageing is missing. The adhesive strength decreases until a plateau
is reached at the level of about 44% of the original adhesive strength, which is
significantly lower compared to thermal ageing. The plateau phase is also reached
roughly after half the time of the corresponding thermal ageing, due to mechano-
chemical processes. The FTIR analyses revealed that this effect originates from the
chemical degradation of PUR, and is not only a result of fatigue.
![Page 73: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/73.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
73
3.6 Interpretation of results and evaluation of usa bility
The aim of this chapter was, to understand the effect of simultaneity of load and
thermal ageing on the adhesion strength between steel pipe and PUR insulation in a
DH-pipe. The measurements have been done using the RISE plug method, and the
FTIR spectroscopy. Both measurements confirm an increase of the degradation
process with combination of load and thermal ageing, for both pipes aged at 130 and
140°C. Clearly, the lower the applied temperature, the slower the process of material
ageing.
To perform such study, some assumption and calculation were needed. The estimation
of the relevant load acting on service pipe, as a result of the axial movements caused
by changes in temperature, and the estimated number of load cycles occurring in the
entire service life of a DH pipe, was calculated before set up of the experiment. The
RISE plug method, and the FTIR are good complementary methods for evaluating the
synergic effect of mechanical load, and thermal ageing on the deterioration of
properties for loaded pipe, in comparison to unloaded pipes.
![Page 74: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/74.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
74
4 Field tests
The purpose of this workpackage is to perform the field study of naturally aged DH
pipes, in order to compare the mechanical and chemical effects between high and low
temperature. Therefore, a number of test specimens of naturally aged pipes were
acquired from the DH sites in South Korea to analyze the operating temperature, and
evaluate the shear strength.
On-site supply and return pipes were tested to analyze the effects of high and low
temperatures. The temperature level of supply and return pipes in South Korea is about
100~120°C, and 40~60°C, respectively. The temperature variation in return pipes can
be assumed analogous to that in 4GDH, with relatively more thermal fluctuation,
compared to the current supply temperature.
4.1 Field selection and data acquisition
4.1.1 Selecting and test preparation of naturally a ged DH pipes
The naturally aged DH pipes were gathered from four DH branches of KDHC in South
Korea for shear strength tests. The minimum length of test pipes is 1000 mm, to obtain
at least three specimens (SPMs) for axial shear test, and one SPM for spare and/or gas
analysis, etc. Figure 4.1 shows an example of sampling location, while the method to
cut the test SPMs from each sampling pipe is shown in Figure 4.2. Figure 4.3 shows a
view of acquiring the pipe samples in the field.
The collected pipe samples were sealed on the field as shown in Figure 4.4, and
transported immediately to the test site. In order to perform the shear strength test, the
pipe samples were cut to 200 mm length, as shown in Figure 2.5. The cut specimens
were wrapped with aluminum foil to minimize the property changes before the test, as
shown in Figure 2.6.
Table 4.1 summarizes the characteristics of the sampling pipes, including the sampling
branch and location, dimension, age, installation time, as well as the operating
temperature in supply and return.
![Page 75: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/75.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
75
Figure 4.1: Example of sampling location for field test.
Figure 4.2: Test specimen cutting.
![Page 76: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/76.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
76
Figure 4.3: Pipe sample acquisition.
Figure 4.4: Acquired samples: (left) Daegu branch, (right) Goyang branch.
![Page 77: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/77.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
77
Figure 4.5: Sample cutting.
Figure 4.6: Prepared test samples.
![Page 78: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/78.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
78
Table 4.1: Dimension, ageing time and sample number of naturally aged pipes.
Dimension Branch and
location Carrier [mm]
Casing [mm]
Installation year
Sampling and test year Aged year
Type (Supply / Return)
SPM No
Daegu 65 140 1997 2018 21 R #01
S #02
S #03
R #04 Goyang 80 160 1995 2018 23
R #05
S #06
S #07
R #08 Suwon-1 125 225 2001 2019 18
R #09
S #10
S #11
S #12
R #13
Suwon-2 125 225 1998 2019 21
R #14
S #15
S #16
S #17
R #18
R #19
Suwon-3 100 200 1998 2019 21
R #20
S #21
S #22
R #23 Jungang-1 100 200 1987 2019 32
R #24
S #25
S #26
S #27
R #28
R #29
Jungang-2 80 160 1987 2019 32
R #30
S #31
S #32
S #33
R #34
R #35
Jungang-3 80 160 1987 2019 32
R #36
![Page 79: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/79.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
79
4.1.2 Measuring the shear strength
Figure 4.7 shows the pipe specimen before the test, and Figure 4.8 shows the test
device, and set-up for the axial shear strength test. The test was carried out in
accordance with EN253.
Figure 4.7: Test specimen before the test.
Figure 4.8: Shear strength test set-up.
![Page 80: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/80.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
80
4.2 Results
4.2.1 Shear strength of naturally aged DH pipes
Figure 4.9 shows the tested pipe specimens. The speed of stroke was 5 mm/min and
the axial force was recorded. Figure 4.10 to 4.17 show the load-stroke curve obtained
by the shear strength test at each sampling location.
SPM #01 SPM #02 SPM #03 SPM #04
SPM #05 SPM #06 SPM #07 SPM #08
SPM #09 SPM #10 SPM #11 SPM #12
Figure 4.9: Test specimens after the test.
![Page 81: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/81.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
81
SPM #13 SPM #14 SPM #15 SPM #16
SPM #17 SPM #18 SPM #19 SPM #20
SPM #21 SPM #22 SPM #23 SPM #24
Figure 4.9: Test specimens after the test (contd.).
![Page 82: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/82.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
82
SPM #25 SPM #26 SPM #27 SPM #28
SPM #29 SPM #30 SPM #31 SPM #32
SPM #33 SPM #34 SPM #35 SPM #36
Figure 4.9: Test specimens after the test (contd.).
![Page 83: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/83.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
83
Figure 4.10: Shear strength curves / Daegu.
Figure 4.11: Shear strength curves / Goyang.
![Page 84: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/84.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
84
Figure 4.12: Shear strength curves / Suwon-1.
Figure 4.13: Shear strength curves / Suwon-2.
![Page 85: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/85.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
85
Figure 4.14: Shear strength curves / Suwon-3.
Figure 4.15: Shear strength curves / Jungang-1.
![Page 86: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/86.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
86
Figure 4.16: Shear strength curves / Jungang-2.
Figure 4.17: Shear strength curves / Jungang-3.
![Page 87: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/87.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
87
According to the standard EN 253, the shear strength is calculated from the following
formula:
πτ
⋅⋅=
s
axax DL
F (4.1)
where,
τax axial shear strength, in MPa;
Fax axial force, in N;
L length of specimen, in mm;
Ds outside diameter of the service pipe, in mm
The calculated shear strength of the SPMs are summarized in Table 4.2.
Figure 4.18 shows the shear strength of 36 naturally aged DH pipes by ageing year.
The shear strength of the supply and return pipes is widely distributed in the range of
0.13 to 0.46MPa. At first glance, it does not seem to follow any special trend,
depending on the ageing year. However, the linear trend line, assuming the initial shear
strength as 0.4MPa, shows that the shear strength drop rate of the supply pipes is
faster than the return pipes. In this case, the expected lifetime for reaching the
minimum required shear strength of 0.12MPa, according to EN 253, was 45.9 years for
the supply pipes, and 140 years for the return pipes.
Moreover, despite the same ageing time, the decrease in shear strength is dependent
on the temperature history.
![Page 88: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/88.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
88
Table 4.2: Calculation of shear strength.
SPM No.
Type (Supply / Return)
Max. axial load [kgf]
Max. axial load [N]
SPM length [mm]
Outside diameter of service pipe
[mm]
Axial shear strength
[kPa]
#01 R 1239 12152 200 76.3 253
#02 S 2090 20503 200 89.1 366
#03 S 2361 23162 200 89.1 414
#04 R 2674 26232 200 89.1 469
#05 R 2671 26203 200 89.1 468
#06 S 2222 21798 200 139.8 248
#07 S 2098 20581 200 139.8 234
#08 R 2241 21979 200 139.8 250
#09 R 2467 24201 200 139.8 276
#10 S 1164 11419 200 139.8 130
#11 S 1200 11767 200 139.8 134
#12 S 1376 13494 200 139.8 154
#13 R 1403 13759 200 139.8 157
#14 R 1443 14151 200 139.8 161
#15 S 1269 12449 200 114.3 173
#16 S 1504 14749 200 114.3 205
#17 S 1544 15147 200 114.3 211
#18 R 2805 27517 200 114.3 383
#19 R 2569 25197 200 114.3 351
#20 R 2864 28096 200 114.3 391
#21 S 900 8829 200 114.3 123
#22 S 858 8412 200 114.3 117
#23 R 1125 11031 200 114.3 154
#24 R 1098 10766 200 114.3 150
#25 S 1619 15877 200 89.1 284
#26 S 1856 18207 200 89.1 325
#27 S 2434 23873 200 89.1 426
#28 R 2512 24638 200 89.1 440
#29 R 2471 24236 200 89.1 433
#30 R 2598 25486 200 89.1 455
#31 S 1107 10860 200 89.1 194
#32 S 1197 11738 200 89.1 210
#33 S 1330 13047 200 89.1 233
#34 R 2523 24746 200 89.1 442
#35 R 2336 22916 200 89.1 409
#36 R 2649 25987 200 89.1 464
![Page 89: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/89.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
89
Figure 4.18: Axial shear strength - All.
Figure 4.19 shows only the shear strength of the supply pipes. SPMs #10 to 12 and
#15 to 17 have the same ageing years of 21, but the measured shear strength varies
between 0.13 to 0.15MPa, and 0.17 to 0.21MPa, respectively. This trend can be more
clearly distinguished in #21 to 22 and #25 to 27, with the same 32 aged years. The
return pipes have the same tendency, as shown in Figure 4.20.
![Page 90: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/90.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
90
Figure 4.19: Axial shear strength – supply pipes.
Figure 4.20: Axial shear strength – return pipes.
The operating temperature of tested SPMs was analyzed to explain the cause of these
trend. Figure 4.21 to 4.27 show the temperature histories of tested SPMs over the last
![Page 91: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/91.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
91
3 years. The temperature histories of Daegu (Figure 4.21) and Goyang (Figure 4.22)
branches were also used in the fatigue analysis in workpackage 1 (chapter 1).
Figure 4.24 and Figure 4.25 represent the temperature history of SPM #10 to 14, and
SPM #15 to 20, respectively. In the case of “Suwon-2 (Figure 4.24)”, the heating site is
a government office building, showing very large daily temperature changes. High
demand heating, and hot water supply is concentrated only during daytime, and supply
is cut off at night and during the holidays. It is typical of intermittent (on and off)
operation, due to the nature of government offices. On the other hand, “Suwon-3
(Figure 4.25)” shows a typical pattern of heat supply in apartment.
The same trend can be found in the temperature history “Jungang-1 (Figure 4.26)” of
the SPMs #21 to 24 and “Jungang-2 (Figure 4.27)” of the SPMs #25 to 30. The return
temperature of “Jungang-1” occurs at a greater fluctuation than “Jungang-2”, which
may be associated with the decrease in shear strength.
In conclusion, the analysis of naturally aged pipes from DH sites in South Korea
showed that the shear strength of the DH pipes is related not only to operating
temperature itself, but also to the extent of temperature fluctuation.
The results demonstrate the validity of the new test method performed in workpackage
3 (chapter 3), considering the effect of temperature and axial load at the same time. It
also presents important implications to consider when determining the lifetime of DH
pipes in 4GDH operating condition, associated with large changes in supply
temperature.
Figure 4.21: Operating temperature – Daegu.
![Page 92: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/92.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
92
Figure 4.22: Operating temperature – Goyang.
Figure 4.23: Operating temperature – Suwon-1.
![Page 93: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/93.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
93
Figure 4.24: Operating temperature – Suwon-2
Figure 4.25: Operating temperature – Suwon-3.
![Page 94: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/94.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
94
Figure 4.26: Operating temperature – Jungang-1.
Figure 4.27: Operating temperature – Jungang-2 and Jungang-3.
![Page 95: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/95.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
95
4.2.2 FTIR analysis of naturally aged DH pipes
Two parts of naturally aged pipe, provided by KDHC, have been evaluated at RISE,
using the Piopsy Method (a combination of RISE plug method and FTIR analysis).
These two pieces for FTIR test were obtained from same pipe gathered from Suwon-2
location (SPMs #10 to #14) for shear strength test (Table 4.1). Herein one piece was
from a supply pipe, and the other one from a return pipe, as shown in Figure 4.28.
Figure 4.28: Two naturally aged pipe pieces: one from a supply pipe (FTIR SPL-S) and another from a return pipe (FTIR SPL-R).
Three plug tests were carried out on each piece of pipe. The obtained results are
summarized in Table 4.3. The values obtained by the plug method showed results
almost similar to the axial shear strength test. The relative shear strength between the
supply and return pipes is equal to σsup /σret = 0.78.
![Page 96: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/96.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
96
Table 4.3: Results from the plug tests.
Return Supply
Test #1 1.77 1.5
Test #2 1.97 1.45
Test #3 1.71 1.32
Average 1.82 1.42
St. dev. 0.14 0.09
The chemical structure of the obtained plugs was analysed using the FTIR-ATR
method. Two slices of approximately 2 mm were cut from each plug. One slice from the
contact surface between PUR and steel service pipe, and the other slice 20 mm above
this point. Three measurements were performed on each slice. The absorption values
were normalized using the peak of aromatic ring as an internal reference. The
absorption indexes of some relevant peaks are shown in Figure 4.29. Peaks 1712 and
1512 shows higher degradation of the contact surface at the supply pipe.
Figure 4.29. Normalized absorption indexes of some relevant peaks (S= supply and R= Return).
The result from FTIR indicate significant changes for the supply pipe near the steel pipe,
in comparison to the return pipe. Moreover, the results of FTIR analysis of naturally
aged DH pipes are similar to those of artificial ageing at 130~140°C with cyclic loads,
performed in workpackage 3 (chapter 3). This proves that the new accelerated ageing
techniques suggested in this study represent well the natural ageing conditions.
![Page 97: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/97.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
97
5 Consequences for network design and asset managem ent
strategy
The purpose of this workpackage is to examine the consequences of the results
obtained within the present research project for improving the current network design.
5.1 Analyze consequences and impacts of future load s on the service pipe
Considering the lower impact of operating temperature, and thermal loading fluctuation
in the network, the lifetime of pre-insulated bonded single pipes used in 4GDH is
expected to increase, in comparison to 3GDH pipelines, with the same design
configuration. The latter are typically designed for 30 years, although many pipelines in
operation are even older (Weidlich and Schuchardt, 2017). Consequently, 4GDH
pipelines would manifest a slower material degradation, and potentially lead to the
implementation of new pipe materials with enhanced thermo-mechanical properties,
economically and environmentally more beneficial.
This conclusion can be transferred to other DH networks. However, the findings in this
study are based on a limited number of data sets, while several other network set-ups,
and operational modes exist. Presumably, DH networks with power to heat devices,
fast unloading big storages, and deviating customer behaviour could manifest different
load characteristics.
One important lesson learned is related to the data logging process. The sensitivity
analysis performed in this study showed that the measuring time interval in the
temperature data should be sufficiently small, to accurately estimate the number of full
temperature cycles, and associated fatigue damage, depending on the particular
operating conditions of the DH pipes.
5.2 Consequences and impacts of future loads on the adhesion
between service pipe and insulation
The analysis of the shear strength data from naturally aged pipes demonstrated that
thermal loading has a significant influence on the pipe axial shear strength. Moreover,
the observed axial shear strength for a thin pipe was greater than for a thick pipe.
However, detailed information on the operating temperature history and initial pipe
![Page 98: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/98.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
98
material data are necessary, for obtaining a direct relationship between the residual
shear strength and the operating time, as well as the foam density.
The implemented laboratory test set-up allowed to investigate accelerated ageing of
DH pipes, considering the combined effect of thermo-oxidative degradation and cyclic
mechanical stresses. While changes of the adhesive strength in pipes exposed only to
thermal ageing follow the same pattern as in previous investigations, changes in
adhesive strength in pipes, exposed to the combined mechanical and thermal ageing,
exhibit partly a different behaviour. The adhesive strength decreases until a plateau is
reached at a level of about 44% of the original adhesive strength, which is significantly
lower compared with thermal ageing. The plateau is reached after about 200 days at
130ºC, and after about 100 days at 140ºC, which roughly corresponds to half the
thermal ageing time without mechanical stresses.
Clearly, the accelerated ageing tests demonstrated that combined effect of mechanical
loading and thermal ageing does not only cause mechanical fatigue, but it also
accelerates the rate of chemical degradation of the PUR foam.
Furthermore, in the naturally aged pipes, the rate of the shear strength drop in the
supply pipes is faster than the return pipes, due to the higher operating temperature. In
addition to the ageing time, the shear strength of the naturally aged DH pipes depends
on the temperature history, requiring proper data logging.
5.3 Recommendations concerning network design and a sset management of 4th generation DH systems
Asset management for 4GDH is not common today. This kind of infrastructure is young
compared to conventional DH, or to other infrastructure (as for instance sewers, where
predictive maintenance is daily business). Therefore, the first important
recommendation for utilities with 4GDH networks is to start data gathering, based on
asset management, as soon as possible. Temperature history and failures should be
documented, with highest accuracy possible. Specifically, the frequency of the
measured temperature data should be sufficiently small (e.g. 5 min), to accurately
estimate the number of full temperature cycles, and associated fatigue damage of the
steel service pipe, depending on the particular operating conditions of the DH pipes.
These data should be stored in a reliable way. In addition, it is recommended to
document the starting point of the most important properties of the pipe before
![Page 99: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/99.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
99
installation. At least, the shear strength, the compression strength, and the thermal
conductivity should be determined according to EN 253. Although this standard is valid
only for 3GDH pipes, the adopted methods are transferable to different types of DH
pipes. For low and very low temperatures some deviations have to be taken into
account, considering the latest standards in this field (for instance prEN 17414 and
prEN 17415 for district cooling).
The lower operating temperature in 4GDH would increase the lifetime of preinsulated
district heating pipes, designed according to EN 253. This would create new
opportunities for new pipe material, and deviations from EN 253, which is expected to
realize economic potentials for 4GDH. This development perspective challenges
existing ageing models, and the current asset management strategies. Therefore, it is
recommended to create a material and load database for each district heating network,
taking into account the pipe materials with initial properties, trench conditions, and load
history. This would lay the foundation for a good predictive maintenance, and a
reduction of economic risks for replacement and repair.
Concluding remarks
The aim of this research project is to determine the lifetime of 4th generation DH
heating networks, developing a new approach, that takes into account the increased
cyclic mechanical and thermal loads, as well the decreased thermal ageing.
The fatigue analysis of the steel pipe, performed considering the collected temperature
data from conventional and 4th generation DH pipelines, showed that the latter are
subjected to greater thermal fluctuation in the limited region of the solar thermal circuit.
The fatigue demand drops drastically before entering the main heating circuit, due to
controlling measures adopted by DH companies. Therefore, the lifetime of 4GDH
pipelines is expected to increase due to the lower operating temperature, and the low
impact of thermal loading volatility in the network, compared to conventional DH.
Moreover, a large measuring time interval in the temperature data, can lead to an
underestimation of the number of full temperature cycles, and associated fatigue
damage, depending on the particular operating conditions of the DH pipes. This
indicates that a more frequent sampling is necessary, while performing low cycle
fatigue analysis, requiring proper data logging.
![Page 100: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/100.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
100
The experimental investigation of the adhesion strength of DH pipes showed a faster
deterioration of the mechanical adhesion strength, when combining thermal ageing and
axial loading simultaneously. Clearly, the most severe conditions are at the interface
between the steel pipe and the foam material, undergoing larger strains and greater
chemical changes, triggered by the mechanical load. Interestingly, the FTIR analyses
indicated that the observed higher rate of deterioration originates from the chemical
degradation, and is not only a result of fatigue.
The analysis of the naturally aged DH pipelines showed that, in addition to the ageing
time, the shear strength of DH pipes depends on the temperature history, decreasing
with the level of operating temperature and amount of fluctuation.
The obtained results permit to gain a better understanding of the performance of
traditional and 4GDH pipelines in operation, that need to be suitably considered in the
engineering design standards of DH networks, contributing to a more sustainable and
energy efficient infrastructure.
Acknowledgments
The authors would like to acknowledge Prof. Thorsten Urbaneck from the Department
of Mechanical Engineering at the Chemnitz University of Technology, and Mr. Thomas
Göschel from inetz GmbH, for providing helpful guidance and detailed information on
the collected temperature data within the recent German research project "Solar district
heating for the Brühl district in Chemnitz – accompanying research (SolFW)". Thanks is
extended to Mr. Stefan Hay from AGFW GmbH, for the helpful support in gathering
4GDH temperature data. Thanks are extended to Enercity Netz GmbH for providing the
DH network data, and the test samples of DH pipes, as well as the colleagues at FFI for
all the support in performing the tests. Thanks are extended to Power Pipe AB for the
valuable discussion and manufacturing of DH pipes, as well as to the colleagues at
Pipe Centre at RISE for all the support and help to set up the experimental part.
Moreover, the authors would like to acknowledge Mr. Eun Sick Kang from the
Pipeteckorea Co. Ltd. for the cooperation in the axial shear strength test of naturally
aged pipes in Korea. Finally, thanks are extended to the members of the IEA DHC
Expert Group for their useful input and guidance during this research project, and in
reviewing the final report.
![Page 101: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/101.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
101
Bibliography
AGFW-Arbeitsblatt FW 401 – Teil 10 (2007). Verlegung und Statik von
Kunststoffmantelrohren (KMR) für Fernwärmenetze – statische Auslegung; Grundlagen
der Spannungsermittlung. Hrsg. v. AGFW Arbeitsgemeinschaft für Wärme und
Heizkraftwirtschaft e.V. beim VDEW e.V., Frankfurt a. M.
Andrews, E.H. (1969) Fatigue in polymers, Testing of Polymers, Vol. IV, W. Brown, ed,
New York: Interscience, 237.
ASME B31.1 (2016) Power piping, American society of mechanical engineers.
ASME B31.3 (2016) Process piping, American society of mechanical engineers.
ASTM Standard E 1049, 1985 (2011). Standard Practices for Cycle Counting in Fatigue
Analysis. West Conshohocken, PA: ASTM International, 2011.
Bernard-Connolly, M., Bui-Quoc, T., Biron, A. (1983). Multilevel strain controlled fatigue
on a type 304 stainless steel. Journal of Engineering Materials and Technology, 105(3),
188-194.
Bleck, W., Dahl, W., Nonn, A., Amlung, L., Feldmann, M., Schäfer, D., Eichler, B.
(2009). Numerical and experimental analyses of damage behaviour of steel moment
connection. Engineering Fracture Mechanics, 76(10), 1531-1547.
Casale, A., Porter, R.S., Johnson, J.F. (1975) Rubber Chem. and Techn. 44/2, 534-577.
Christensen, R., Hansen K.E., Neergaard L.B., Randlov P., Olsson N. (1999). Fatigue
Analysis of District Heating Systems, IEA District Heating and Cooling, Programme of
Research, Development and Demonstration on District Heating and Cooling.
Dowling, N. E. (2013). Mechanical Behaviour of Materials. Engineering Methods for
Deformation, Fracture, and Fatigue. Fourth Edition. Pearson.
Downing, S., Galliart, D., Berenyi, T. (1976). A Neuber's rule fatigue analysis procedure
for use with a mobile computer. SAE Transactions, 1252-1261.
Downing, S.D., Socie, D.F. (1982). Simple rainflow counting algorithm. International
Journal of Fatigue. 1. pp. 31-40.
EN 13480-3 (2017) Metallic industrial piping – Part 3: design and calculation, European
committee for standardization (CEN).
EN 13941 (2019) Design and installation of preinsulated bonded pipe systems for
district heating.
![Page 102: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/102.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
102
EN 253 (2019) District heating pipes - Bonded single pipe systems for directly buried
hot water networks - Factory made pipe assembly of steel service pipe, polyurethane
thermal insulation and a casing of polyethylene.
prEN 17414 (2019) District cooling pipes - Factory made flexible pipe systems - Part 1:
Classification, general requirements and test methods; Part 2: Bonded system with
plastic service pipes; requirements and test methods; Part 3: Non bonded system with
plastic service pipes; requirements and test methods.
prEN 17415 (2019) District cooling pipes - Bonded single pipe systems for directly
buried cold water networks - Part 1: Factory made pipe assembly of steel or plastic
service pipe, polyurethane thermal insulation and a casing of polyethylene.
Fernandes, A. A., de Jesus, A. M., Jorge, R. N. (2018). Monotonic and Ultra-Low-Cycle
Fatigue Behaviour of Pipeline Steels: Experimental and Numerical Approaches.
Springer.
Frederiksen S., Werner S. (2013). District heating and cooling. Sweden:
Studentlitteratur, 697.
Gudehus, H.,Zenner, H. (1995). Leitfaden für eine Betriebsfestigkeitsrechnung. Verlag
Stahleisen mbH.
Haibach, E. (2006). Betriebsfestigkeit: Verfahren und Daten zur Bauteilberechnung.
Springer-Verlag.
Haibach, E. (1970). Modifizierte lineare Schadensakkumulations-hypothese zur
Berücksichtigung des Dauerfestigkeitsabfalls mit fortschreitender Schädigung.
Laboratorium für Betriebsfestigkeit.
Hashin, Z., Laird, C. (1979). Cumulative damage under two level cycling: some
theoretical predictions and test data. Fatigue & Fracture of Engineering Materials &
Structures, 2(4), 345-350.
Hück, M., Thrainer, L., Schütz, W. (1983). Berechnung von Wöhlerlinien für Bauteile
aus Stahl, Stahlguß und Grauguß: synthetische Wöhlerlinien. Stahleisen.
ISO 12110-2 (2013). Metallic materials - Fatigue testing - Variable amplitude fatigue
testing - Part 2: Cycle counting and related data reduction methods.
Kanvinde, A. M., Fell, B. V., Deierlein, G. G. (2007). Physics-based continuum models
by Abaqus to simulate fracture and ultra low cycle fatigue in steel structures. Structural
Engineering Research Frontiers, 249(33), 14.
![Page 103: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/103.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
103
Kausch, H.H. (1986) Polymer fracture, 2nd ed., Springer, Heidelberg. New York.
Lauer, W. (1975) Kautschuk + Gummi, Kunstoffe 28/9, 536-613.
Leonatris, G. (1995). Ein Konzept zur Lebensdauervorhersage, Dissertation,
Universität-Gesamthochschule Kassel.
Linn, S., Scholz, A. (2013). Creep-fatigue lifetime assessment with phenomenological
and constitutive material laws. Procedia Engineering, 55, 607-611.
Manson, S. S., Halford, G. R. (1981). Practical implementation of the double linear
damage rule and damage curve approach for treating cumulative fatigue damage.
International journal of fracture, 17(2), 169-192.
Markl, A. R. C. (1955). Piping-flexibility analysis. Trans. ASME, 77(2), 12743.
Markl, A.R.C. (1952) Fatigue testing of piping components, Trans. ASME, 1952, pp.
287-303
Mathworks. (2018). Matlab Documentation.
https://ch.mathworks.com/help/signal/ref/rainflow.html#d120e139322
Matsuishi, M., Endo, T. (1968). Fatigue of metals subjected to varying stress, Japan
society of Mechanical Engineers, Fukuoka, Japan.
Miner, M. A. (1945). Cumulative fatigue damage. Journal of applied mechanics, 12(3),
A159-A164.
Miller, K. J., Zachariah, K. P. (1977). Cumulative damage laws for fatigue crack
initiation and stage I propagation. The Journal of Strain Analysis for Engineering Design,
12(4), 262-270.
Nikbin, K. (2013). Creep/fatigue crack growth testing, modelling and component life
assessment of welds. Procedia Engineering, 55, 380-393.
Okamura, H., Sakai, S., Susuki, I. (1979). Cumulative fatigue damage under random
loads. Fatigue and Fracture of Engineering Materials and Structures, 1(4), 409-419.
Palmgren, A. (1924). Die Lebensdauer von Kugellagern, VDI-Zeitschrift, 58, 339-341.
Penderos, M. (1996). Thermal Cycling of Directly Buried District Heating Networks:
Application of Cumulative Damage Theory to Some Field Cases.
Pöting, S. (2003). Lebensdauerabschätzung im High-cycle-fatigue-Bereich.
Papierflieger.
![Page 104: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/104.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
104
Randløv, R., Hansen, K. E., Penderos, M. (1996). Temperature Variations in
Preinsulated DH Pipes Low Cycle Fatigue. IEA District Heating and Cooling,
Programme of Research, Development and Demonstration on District Heating and
Cooling.
Reinhardt, H. W., Reinhardt, H. W. (2010). Ingenieurbaustoffe. Berlin: Ernst & Sohn.
Rice R. C. (1997). SAE fatigue design handbook, 3rd Edition.
Shrestha, N. L., Urbaneck, T., Oppelt, T., Platzer, B., Göschel, T., Uhlig, U., & Frey, H.
(2017). Implementation of large solar thermal system into district heating network in
Chemnitz (Germany). In International Solar Energy Society (ed.): ISES Solar World
Conference (pp. 322-332).
Subramanyan, S. (1976). A cumulative damage rule based on the knee point of the SN
curve. Journal of Engineering Materials and Technology, 98(4), 316-321.
Terselius, B., Gedde, U. W., Jansson, J.F. (1986) Mechano-chemical Phenomena in
Polymers- Failure of Plastics edited by Brostow/Corneliussen, Hanser Publisher.
Weidlich, I., Schuchardt, G. K. (2017). New Approach for Asset Management in District
Heating (DH) Networks. Energy Procedia, 113, 22-27.
Wetzel, R. M. (1971). A method of fatigue damage analysis. PhD Thesis, Department
of Civil Engineering, University of Waterloo, Ontario, Canada.
Yarahmadi, N., Sällström, J.H. (2015) Improved maintenance strategies for district
heating pipe-lines, IEA DHC/CHP Annex10.
Yarahmadi, N., Vega, A., Jakubowicz, I. (2017). Life time for district heating pipe:
Phase 1. Stockholm: Energiforsk AB.
Yarahmadi, N., Vega, A., Jakubowicz, I. (2017). Livslängd för fjärrvärmerör: Phase 2.
Life time for district heating pipe: Energiforsk AB.
![Page 105: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/105.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
105
Appendix A: fatigue analysis results of the tempera ture data
This section reports in detail the rainflow cycle matrix, as well as the number of
equivalent full temperature cycles N0 corresponding to each temperature history data
analyzed in workpackage 1 (chapter 1).
Figure A-1. Histogram of the cycle counts as a function of temperature mean and range, for the Daegu branch supply pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Table A-1. Results of the equivalent full temperature cycles N0 for the Daegu branch supply pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Daegu supply pipe (S1) N0 N0 (30 years) N 0 (50 years) Tmax Tmean ∆Tref = Tmax - 10°C 0.67 6.73 11.21
∆Tref = 110°C 0.65 6.50 10.84 119.1°C 95.3°C
![Page 106: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/106.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
106
Figure A-2. Histogram of the cycle counts as a function of temperature range, for the Daegu branch return pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Table A-2. Results of the evaluated full temperature cycles N0 for the Daegu branch return pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Daegu return pipe (R1) N 0 N0 (30 years) N 0 (50 years) Tmax Tmean ∆Tref = Tmax - 10°C 0.64 6.42 10.70
∆Tref = 110°C 0.13 1.35 2.25 84.4°C 39.7°C
![Page 107: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/107.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
107
Figure A-3. Histogram of the cycle counts as a function of temperature range, for the Goyang branch supply pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Table A-3. Results of the evaluated full temperature cycles N0 for the Goyang branch supply pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Goyang supply pipe (S2) N 0 N0 (30 years) N 0 (50 years) Tmax Tmean ∆Tref = Tmax - 10°C 0.21 2.05 3.42
∆Tref = 110°C 0.19 1.90 3.17 117.9°C 97.9°C
![Page 108: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/108.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
108
Figure A-4. Histogram of the cycle counts as a function of temperature range, for the Goyang branch return pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Table A-4. Results of the evaluated full temperature cycles N0 for the Goyang branch return pipe (measuring period: 3 years, from 01/01/2015 to 01/01/2018).
Goyang return pipe (R2) N0 N0 (30 years) N 0 (50 years) Tmax Tmean ∆Tref = Tmax - 10°C 0.46 4.64 7.74
∆Tref = 110°C 0.04 0.45 0.75 71.3°C 45.2°C
![Page 109: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/109.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
109
Figure A-5. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Danska supply pipe (measuring period: 10 years, 2 months, 30 days (3744 days), from 27/03/2007 to 26/06/2017).
Table A-5. Results of the evaluated full temperature cycles N0 for the Göteborg-Danska supply pipe (measuring period: 10 years, 2 months, 30 days (3744 days)).
Göteborg-Danska supply pipe (S3) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 16.09 47.08 78.47 ∆Tref = 110°C 13.76 40.27 67.12
115.8°C 61.1°C
![Page 110: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/110.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
110
Figure A-6. Histogram of the cycle counts as a function of temperature range, for the Göteborg Danska return pipe (measuring period: 8 years, 2 months, 25 days (3008 days), from 01/04/2009 to 26/06/2017).
Table A-6. Results of the evaluated full temperature cycles N0 for the Göteborg-Danska return pipe (measuring period: 8 years, 2 months, 25 days (3008 days)).
Göteborg-Danska return pipe (R3) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 6.05 22.03 36.72 ∆Tref = 110°C 2.63 9.59 15.99
99.4°C 45.7°C
![Page 111: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/111.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
111
Figure A-7. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Hisingsbron supply pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Table A-7. Results of the evaluated full temperature cycles N0 for the Göteborg-Hisingsbron supply pipe (measuring period: 11 years, 3 months, 23 days, (4131 days)).
Göteborg- Hisingsbron supply pipe (S4) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 13.21 35.05 58.42 ∆Tref = 110°C 10.69 28.37 47.28
114.3°C 86.8°C
![Page 112: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/112.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
112
Figure A-8. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Hisingsbron return pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Table A-8. Results of the evaluated full temperature cycles N0 for the Göteborg-Hisingsbron return pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Göteborg- Hisingsbron return pipe (R4) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 4.25 11.28 18.80 ∆Tref = 110°C 5.51 14.63 24.38
127.4°C 33.1°C
![Page 113: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/113.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
113
Figure A-9. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Marieholm supply pipe (measuring period: 9 years, 6 months, 7 days (3476 days)).
Table A-9. Results of the evaluated full temperature cycles N0 for the Göteborg-Marieholm supply pipe (measuring period: 9 years, 6 months, 7 days, (3476 days)).
Göteborg-Marieholm supply pipe (S5) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 21.83 68.81 114.68 ∆Tref = 110°C 15.15 47.75 79.58
110.4°C 88.7°C
![Page 114: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/114.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
114
Figure A-10. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Marieholm return pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Table A-10. Results of the evaluated full temperature cycles N0 for the Göteborg-Hisingsbron return pipe (measuring period: 11 years, 3 months, 23 days, (4131 days)).
Göteborg-Marieholm return pipe (R5) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 6.57 19.63 32.71 ∆Tref = 110°C 0.83 2.49 4.14
75.6°C 24.3°C
![Page 115: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/115.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
115
Figure A-11. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Falutorget supply pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Table A-11. Results of the evaluated full temperature cycles N0 for the Göteborg-Falutorget supply pipe (measuring period: 11 years, 3 months, 23 days, (4131 days)).
Göteborg- Falutorget supply pipe (S6) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 17.01 45.13 75.22 ∆Tref = 110°C 14.55 38.60 64.34
115.8°C 61.9°C
![Page 116: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/116.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
116
Figure A-12. Histogram of the cycle counts as a function of temperature range, for the Göteborg-Falutorget return pipe (measuring period: 11 years, 3 months, 23 days (4131 days)).
Table A-12. Results of the evaluated full temperature cycles N0 for the Göteborg-Falutorget return pipe (measuring period: 11 years, 3 months, 23 days, (4131 days)).
Göteborg- Falutorget return pipe (R6) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 4.25 11.28 18.80 ∆Tref = 110°C 5.51 14.63 24.38
127.4°C 33.1°C
![Page 117: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/117.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
117
Figure A-13. Histogram of the cycle counts as a function of temperature range, for the Oslo Vika 2013 supply pipe (measuring period: 1 year (365 days), from 01/01/2013 to 31/12/2013).
Table A-13. Results of the evaluated full temperature cycles N0 for the Oslo Vika 2013 supply pipe (measuring period: 1 year (365 days)).
Oslo Vika 2013 supply pipe (S7) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 2.83 84.90 141.50 ∆Tref = 110°C 3.46 103.92 173.20
125.7°C 101.6°C
![Page 118: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/118.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
118
Figure A-14. Histogram of the cycle counts as a function of temperature range, for the Oslo Vika 2013 return pipe (measuring period: 1 year (365 days), from 01/01/2013 to 31/12/2013)).
Table A-14. Results of the evaluated full temperature cycles N0 for the Oslo Vika 2013 return pipe (measuring period: 1 year (360 days)).
Oslo Vika 2013 return pipe (R7) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 2.02 60.55 100.92 ∆Tref = 110°C 1.14 34.26 57.09
105.4°C 60.6°C
![Page 119: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/119.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
119
Figure A-15. Histogram of the cycle counts as a function of temperature range, for the Oslo Vika 2016 supply pipe (measuring period: 1 year (365 days), from 01/01/2013 to 31/12/2013)).
Table A-15. Results of the evaluated full temperature cycles N0 for the Oslo Vika 2016 supply pipe (measuring period: 1 year (360 days)).
Oslo Vika 2016 supply pipe (S8) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.41 12.24 20.40 ∆Tref = 110°C 0.44 13.15 21.92
122.0°C 101.6°C
![Page 120: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/120.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
120
Figure A-16. Histogram of the cycle counts as a function of temperature range, for the Oslo Vika 2016 return pipe (measuring period: 1 year (365 days), from 01/01/2016 to 31/12/2016)).
Table A-16. Results of the evaluated full temperature cycles N0 for the Oslo Vika 2016 return pipe (measuring period: 1 year (360 days)).
Oslo Vika 2016 return pipe (R8) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 2.87 86.03 143.38 ∆Tref = 110°C 1.64 49.08 81.80
105.6°C 57.2°C
![Page 121: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/121.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
121
Figure A-17. Histogram of the cycle counts as a function of temperature range, for the Soerengkaia 153 supply pipe (measuring period: 1 year (364 days), from 20/08/2018 to 18/08/2019)).
Table A-17. Results of the evaluated full temperature cycles N0 for the Soerengkaia 153 supply pipe (measuring period: 1 year (364 days)).
Soerengkaia 153 supply pipe (S9) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.42 12.58 20.96 ∆Tref = 110°C 0.36 10.82 18.03
115.9°C 98.3°C
![Page 122: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/122.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
122
Figure A-18. Histogram of the cycle counts as a function of temperature range, for the Soerengkaia 153 return pipe (measuring period: 1 year (365 days), from 20/08/2018 to 18/08/2019)).
Table A-18. Results of the evaluated full temperature cycles N0 for the Soerengkaia 153 return pipe (measuring period: 1 year (364 days)).
Soerengkaia 153 return pipe (R9) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 40.63 1223.22 2038.70 ∆Tref = 110°C 1.15 34.56 57.61
55.1°C 38.3°C
![Page 123: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/123.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
123
Figure A-19. Histogram of the cycle counts as a function of temperature range, for the Brobekkveiien 80 supply pipe (measuring period: 1 year (365 days), from 30/08/2018 to 30/08/2019)).
Table A-19. Results of the evaluated full temperature cycles N0 for the Brobekkveiien 80 supply pipe (measuring period: 1 year (365 days)).
Brobekkveiien 80 supply pipe (S10) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.16 4.66 7.77 ∆Tref = 110°C 0.15 4.41 7.35
118.5°C 92.4°C
![Page 124: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/124.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
124
Figure A-20. Histogram of the cycle counts as a function of temperature range, for the Brobekkveiien 80 return pipe (measuring period: 1 year (365 days), from 30/08/2018 to 30/08/2019)).
Table A-20. Results of the evaluated full temperature cycles N0 for the Brobekkveiien 80 return pipe (measuring period: 1 year (365 days)).
Brobekkveiien 80 return pipe (R10) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 9.92 297.93 496.56 ∆Tref = 110°C 0.53 15.82 26.36
62.8°C 45.8°C
![Page 125: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/125.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
125
Figure A-21. Histogram of the cycle counts as a function of temperature range, for the Skoeyen terasse 4 supply pipe (measuring period: 1 year (365 days), from 30/08/2018 to 30/08/2019)).
Table A-21. Results of the evaluated full temperature cycles N0 for the Skoeyen terasse 4 supply pipe (measuring period: 1 year (365 days)).
Skoeyen terasse 4 supply pipe (S11) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 1.08 32.41 54.02 ∆Tref = 110°C 0.82 24.53 40.88
112.6°C 90.7°C
![Page 126: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/126.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
126
Figure A-22. Histogram of the cycle counts as a function of temperature range, for the Skoeyen terasse 4 return pipe (measuring period: 1 year (365 days), from 30/08/2018 to 30/08/2019)).
Table A-22. Results of the evaluated full temperature cycles N0 for the Skoeyen terasse 4 return pipe (measuring period: 1 year (365 days)).
Skoeyen terasse 4 return pipe (R11) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 50.70 1521.94 2536.56 ∆Tref = 110°C 2.20 66.01 110.02
60.2°C 38.7°C
![Page 127: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/127.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
127
Figure A-17. Histogram of the cycle counts as a function of temperature range, for the Hannover CoCa 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-17. Results of the evaluated full temperature cycles N0 for the CoCa 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
CoCa 2010 supply pipe (S12) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.14 4.18 6.97
∆Tref = 110°C 0.14 4.30 7.17 120.8°C 97.1°C
![Page 128: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/128.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
128
Figure A-18. Histogram of the cycle counts as a function of temperature range, for the Hannover CoCa 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-18. Results of the evaluated full temperature cycles N0 for the CoCa 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
CoCa 2010 return pipe (R12) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.08 2.26 3.76
∆Tref = 110°C 0.01 0.37 0.62 80.0°C 61.3°C
![Page 129: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/129.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
129
Figure A-19. Histogram of the cycle counts as a function of temperature range, for the Hannover GKH 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-19. Results of the evaluated full temperature cycles N0 for the GKH 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
GKH 2010 supply pipe (S13) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.04 1.33 2.22
∆Tref = 110°C 0.05 1.44 2.39 122.1°C 96.5°C
![Page 130: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/130.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
130
Figure A-20. Histogram of the cycle counts as a function of temperature range, for the Hannover GKH 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-20. Results of the evaluated full temperature cycles N0 for the GKH 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
GKH 2010 return pipe (R13) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.01 0.34 0.56
∆Tref = 110°C 0.001 0.04 0.06 73.5°C 59.1°C
![Page 131: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/131.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
131
Figure A-21. Histogram of the cycle counts as a function of temperature range, for the Hannover HKW 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-21. Results of the evaluated full temperature cycles N0 for the HKW 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
HKW 2010 supply pipe (S14) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.64 19.31 32.19
∆Tref = 110°C 0.63 18.89 31.49 119.4°C 92.9°C
![Page 132: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/132.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
132
Figure A-22. Histogram of the cycle counts as a function of temperature range, for the Hannover HKW 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-22. Results of the evaluated full temperature cycles N0 for the HKW 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
HKW 2010 return pipe (R14) N0 N0 (30 years) N0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.48 14.31 23.85
∆Tref = 110°C 0.09 2.79 4.65 83.1°C 60.5°C
![Page 133: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/133.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
133
Figure A-23. Histogram of the cycle counts as a function of temperature range, for the Hannover KWH 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-23. Results of the evaluated full temperature cycles N0 for the KWH 2010 supply pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
KWH 2010 supply pipe (S15) N0 N0 (30 years) N0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.04 1.26 2.10
∆Tref = 110°C 0.05 1.36 2.26 122.1°C 96.6°C
![Page 134: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/134.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
134
Figure A-24. Histogram of the cycle counts as a function of temperature range, for the Hannover KWH 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
Table A-24. Results of the evaluated full temperature cycles N0 for the KWH 2010 return pipe (measuring period: 1 year (365 days), from 01/01/2010 to 01/01/2011).
KWH 2010 return pipe (R15) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.02 0.50 0.84
∆Tref = 110°C 0.001 0.04 0.07 68.5°C 58.5°C
![Page 135: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/135.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
135
Figure A-25. Histogram of the cycle counts as a function of temperature range, for the Hannover CoCa 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-25. Results of the evaluated full temperature cycles N0 for the CoCa 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
CoCa 2011 supply pipe (S16) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.20 6.15 10.25
∆Tref = 110°C 0.15 4.44 7.40 111.4°C 95.7°C
![Page 136: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/136.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
136
Figure A-26. Histogram of the cycle counts as a function of temperature range, for the Hannover CoCa 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-26. Results of the evaluated full temperature cycles N0 for the CoCa 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
CoCa 2011 return pipe (R16) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.10 3.15 5.24
∆Tref = 110°C 0.02 0.52 0.86 80.0°C 60.1°C
![Page 137: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/137.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
137
Figure A-27. Histogram of the cycle counts as a function of temperature range, for the Hannover GKH 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-27. Results of the evaluated full temperature cycles N0 for the GKH 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
GKH 2011 supply pipe (S17) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.15 4.41 7.35
∆Tref = 110°C 0.15 4.46 7.43 120.3°C 94.6°C
![Page 138: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/138.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
138
Figure A-28. Histogram of the cycle counts as a function of temperature range, for the Hannover GKH 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-28. Results of the evaluated full temperature cycles N0 for the GKH 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
GKH 2011 return pipe (R17) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.06 1.94 3.24
∆Tref = 110°C 0.004 0.13 0.22 65.9°C 58.4°C
![Page 139: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/139.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
139
Figure A-29. Histogram of the cycle counts as a function of temperature range, for the Hannover HKW 2011 supply pipe (measuring period: 1 year (128 days), from 01/01/2011 to 09/05/2011).
Table A-29. Results of the evaluated full temperature cycles N0 for the HKW 2011 supply pipe (measuring period: 1 year (128 days), from 01/01/2011 to 09/05/2011).
HKW 2011 supply pipe (S18) N 0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.32 27.24 45.39
∆Tref = 110°C 0.23 19.67 32.78 111.4°C 93.6°C
![Page 140: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/140.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
140
Figure A-30. Histogram of the cycle counts as a function of temperature range, for the Hannover HKW 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 09/05/2011).
Table A-30. Results of the evaluated full temperature cycles N0 for the HKW 2011 return pipe (measuring period: 1 year (128 days), from 01/01/2011 to 09/05/2011).
HKW 2011 return pipe (R18) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.54 45.96 76.60
∆Tref = 110°C 0.10 8.91 14.86 83.0°C 64.2°C
![Page 141: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/141.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
141
Figure A-31. Histogram of the cycle counts as a function of temperature range, for the Hannover KWH 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-31. Results of the evaluated full temperature cycles N0 for the KWH 2011 supply pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
KWH 2011 supply pipe (S19) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.12 3.49 5.82
∆Tref = 110°C 0.12 3.57 5.95 120.6°C 94.7°C
![Page 142: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/142.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
142
Figure A-32. Histogram of the cycle counts as a function of temperature range, for the Hannover KWH 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
Table A-32. Results of the evaluated full temperature cycles N0 for the KWH 2011 return pipe (measuring period: 1 year (365 days), from 01/01/2011 to 01/01/2012).
KWH 2011 return pipe (R19) N0 N0 (30 years) N 0 (50 years) T max Tmean ∆Tref = Tmax - 10°C 0.02 0.62 1.04
∆Tref = 110°C 0.002 0.05 0.08 68.2°C 57.7°C
![Page 143: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/143.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
143
Figure A-33. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH main network supply pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-33. Results of the evaluated full temperature cycles N0 for the solar thermal DH main network supply pipe (measuring period: 365 days ).
Main network supply pipe (S20) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.330 9.91 16.51 ∆Tref = 110°C 0.792 23.76 39.61
146.9°C 134.3°C
![Page 144: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/144.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
144
Figure A-34. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH main network return pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-34. Results of the evaluated full temperature cycles N0 for the solar thermal DH main network return pipe (measuring period: 365 days ).
Main network return pipe (R20) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.366 10.97 18.29 ∆Tref = 110°C 0.201 6.03 10.05
104.7°C 59.1°C
![Page 145: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/145.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
145
Figure A-35. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH secondary network supply pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-35. Results of the evaluated full temperature cycles N0 for the solar thermal DH secondary network supply pipe (measuring period: 365 days ).
Secondary network supply pipe (S21) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.199 5.99 9.98 ∆Tref = 110°C 0.052 1.55 2.58
88.5°C 72.0°C
![Page 146: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/146.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
146
Figure A-36. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH secondary network return pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-36. Results of the evaluated full temperature cycles N0 for the solar thermal DH secondary network return pipe (measuring period: 365 days ).
Secondary network return pipe (R21) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 0.082 2.47 4.12 ∆Tref = 110°C 0.005 0.14 0.23
63.7°C 55.4°C
![Page 147: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/147.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
147
Figure A-37. Histogram of the cycle counts as a function of temperature range, for the heat storage temperature level 1 (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-37. Results of the evaluated full temperature cycles N0 for the heat storage temperature level 1 (measuring period: 365 days ).
Heat storage temperature level 1 (SL1) N0 N0 (30 years) N 0 (50 years) T max Tmean
∆Tref = Tmax - 10°C 13.152 394.83 658.05 ∆Tref = 110°C 0.863 25.92 43.20
65.7°C 44.2°C
![Page 148: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/148.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
148
Figure A-38. Histogram of the cycle counts as a function of temperature range, for the heat storage temperature level 2 (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-38. Results of the evaluated full temperature cycles N0 for the heat storage temperature level 2 (measuring period: 365 days ).
Heat storage temperature level 2 (SL2) N0 N0 (30 years) N 0 (50 years) T max Tmean
∆Tref = Tmax - 10°C 6.138 184.25 307.09 ∆Tref = 110°C 0.324 9.74 16.23
62.7°C 45.8°C
![Page 149: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/149.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
149
Figure A-39. Histogram of the cycle counts as a function of temperature range, for the heat storage temperature level 3 (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-39. Results of the evaluated full temperature cycles N0 for the heat storage temperature level 3 (measuring period: 365 days).
Heat storage temperature level 3 (SL3) N0 N0 (30 years) N 0 (50 years) T max Tmean
∆Tref = Tmax - 10°C 3.398 102.01 170.02 ∆Tref = 110°C 0.788 23.66 39.44
86.3°C 51.4°C
![Page 150: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/150.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
150
Figure A-40. Histogram of the cycle counts as a function of temperature range, for the heat storage temperature level 4 (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-40. Results of the evaluated full temperature cycles N0 for the heat storage temperature level 4 (measuring period: 365 days).
Heat storage temperature level 4 (SL4) N0 N0 (30 years) N 0 (50 years) T max Tmean
∆Tref = Tmax - 10°C 4.000 120.09 200.16 ∆Tref = 110°C 1.330 39.91 66.52
93.5°C 56.5°C
![Page 151: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/151.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
151
Figure A-41. Histogram of the cycle counts as a function of temperature range, for the solar thermal field 1 supply pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-41. Results of the evaluated full temperature cycles N0 for the solar thermal field 1 supply pipe (measuring period: 365 days).
solar thermal field 1 supply pipe (S22) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 21.419 643.01 1071.69 ∆Tref = 110°C 27.924 838.30 1397.16
127.5°C 41.4°C
![Page 152: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/152.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
152
Figure A-42. Histogram of the cycle counts as a function of temperature range, for the solar thermal field 1 return pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-42. Results of the evaluated full temperature cycles N0 for the solar thermal field 1 return pipe (measuring period: 365 days).
solar thermal field 1 return pipe (R22) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 28.407 852.80 1421.33 ∆Tref = 110°C 3.682 110.52 184.20
76.0°C 38.5°C
![Page 153: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/153.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
153
Figure A-43. Histogram of the cycle counts as a function of temperature range, for the solar thermal field 2 supply pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-43. Results of the evaluated full temperature cycles N0 for the solar thermal field 2 supply pipe (measuring period: 365 days ).
solar thermal field 2 supply pipe (S23) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 64.084 1923.84 3206.40 ∆Tref = 110°C 22.063 662.34 1103.91
94.3°C 39.8°C
![Page 154: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/154.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
154
Figure A-44. Histogram of the cycle counts as a function of temperature range, for the solar thermal field 2 return pipe (measuring period: 365 days, from 1/01/2018 to 31/12/2018, ∆t = 1min).
Table A-44. Results of the evaluated full temperature cycles N0 for the solar thermal field 2 return pipe (measuring period: 365 days ).
solar thermal field 2 return pipe (R23) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 59.516 1786.71 2977.84 ∆Tref = 110°C 3.576 107.35 178.91
64.5°C 37.3°C
![Page 155: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/155.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
155
Figure A-45. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH house connection supply pipe (measuring period: 360 days, from 06/12/2018 to 02/12/2019, ∆t = 5min).
Table A-45. Results of the evaluated full temperature cycles N0 for the solar thermal DH house connection supply pipe (measuring period: 360 days ).
House connection supply pipe (S24) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 11.206 340.69 567.82 ∆Tref = 110°C 2.103 63.93 106.56
82.4°C 70.0°C
![Page 156: Annex XII final report Effects of Loads on Asset ... · 3. Critical aspect III: Transferability of non-linear theories. The transferability of non-linear theories on media pipes of](https://reader034.vdocuments.net/reader034/viewer/2022042211/5eb45128062c011da641370f/html5/thumbnails/156.jpg)
Annex XII final report
Effects of Loads on Asset Management of the
4th Generation District Heating Networks
156
Figure A-46. Histogram of the cycle counts as a function of temperature range, for the solar thermal DH house connection return pipe (measuring period: 360 days, from 06/12/2018 to 02/12/2019, ∆t = 5min).
Table A-46. Results of the evaluated full temperature cycles N0 for the solar thermal DH house connection return pipe (measuring period: 360 days ).
House connection return pipe (R24) N0 N0 (30 years) N 0 (50 years) Tmax Tmean
∆Tref = Tmax - 10°C 46.914 1426.31 2377.18 ∆Tref = 110°C 3.804 115.66 192.77
68.7°C 41.2°C