an intelligent ice protection system for next …an intelligent ice protection system for next...

1

Abstract

This paper describes the development of anIntelligent Ice Protection System (IIPS) forintegration into an aircraft trajectoryoptimisation framework developed as part of theClean Sky programme. The IIPS is developed inMATLAB incorporating features of moreelectric systems and future air navigationenvironment. A typical flight from LondonAirport Heathrow (EGLL/LHR) airport toAmsterdam Airport Schiphol (EHAM/AMS) wasused as a case study. Initial results show thatfurther savings on fuel burn and flight timecould be achieved if icing phenomenon isconsidered in aircraft trajectory optimizationscheme.

Nomenclature

݉ : Aircraft massܸ : Aerodynamic speedܶ : Thrust magnitudeℎ : Altitudeܮ : Lift magnitudeܦ : Drag magnitude݃ : Gravity accelerationߛ : Flight path angle߯ : Heading angleܿ : Specific fuel consumptionφ : Geodetic latitudeߣ : Geodetic longitudeܴா : Earth radiusߤ : Bank or roll angleܲܪ : High Pressureܲܫ : Intermediate Pressureܲܫܫ ܵ : Intelligent Ice Protection Systemܦ&ܴ : Research and Development

1 Introduction

The growing trend of air travel has madeaviation the fastest growing source of globalwarming and climate change [1]. Based on thecurrent annual projection of about 5%, theannual passenger total is expected to increasefrom 3.1 billion in 2013 to 6.4 billion by 2030[2]. CO2 emission is by far the largest amongpollutants from air transport. In Europe alone, itis estimated that more than 300,000 tonnes ofCO2 is generated from aircraft operations perday [3]. As a result, EU initiated three streamcomprehensive projects/measures to mitigatethe impacts of aviation on the environment andfuel resources. These are R&D for greenertechnology, modernised air traffic managementsystems and market based measures.

The Clean Sky Joint Technology Initiative (JTI)is the flagship of the R&D projects for thegreening of air transport in EU. The Clean Skyis a private/public research partnership atEuropean level in the field of aviation todevelop the technologies necessary for a clean,innovative and competitive system of airtransport. This would be done through theachievement of ACARE targets of reducing theemissions of CO2, NOx and unburnthydrocarbons by 50%, 80% and 50%respectively by 2020 referenced to 2000standard [4]. One of the Clean Sky activities isthe Management of Trajectory and Mission(MTM) work package which is under Systems

AN INTELLIGENT ICE PROTECTION SYSTEM FORNEXT GENERATION AIRCRAFT TRAJECTORY

OPTIMISATION

Ahmed Shinkafi, Craig Lawson, Ravinka Seresinhe, Daniele Quaglia andIrfan Madani

Cranfield University, Aerospace Department, MK43 0AL, [email protected]

Keywords: Aircraft ice protection, next-generation, trajectory optimisation

A. SHINKAFI, C. LAWSON, R. SERESINHE, D. QUAGLIA AND I MADANI

2

for Green Operations (SGO) - IntegratedTechnology Demonstrators (ITD).

This research was carried out under the SGOITD and proposes a controllable ice protectionsystem for aircraft fitted with future airnavigation systems. The research aimed atinvestigating ways to operate aircraft at lowpower levels through changes in aircraftoperation strategy that takes into accountoptimised flight routings due to icingconditions. Conventional approaches totrajectory optimisation do not take the airframesystems penalty into account in contrast to realaircraft operation. This research has developed atool capable of sizing and simulating aircraftanti-icing performance for trajectoryoptimisation which would enable developmentof decision making processes dependent onweather within the flight management system;thus transforming the conventional IceProtection System (IPS) to a more intelligentsystem.

1.1 Objectives of the ResearchWhile continuing to have safety as a primaryobjective, this work was about utilising futureaircraft's Performance-Based Navigation (PBN)concept to investigate possible ways ofminimising IPS operational demand power thatwould lead to greater efficiency and capacity.The primary objectives of this work, therefore,are to:

develop a consistent and cohesivestrategy of managing in-flight icing in afuture ATM environment that enablesefficient flight planning,

investigate the response of today'scutting edge aircraft icing technologiesin the future 4D navigation environment,

use optimised aircraft trajectories toinvestigates ways to operate aircraft atlow power levels in icing conditions and

Conceptualise controllable ice protectionsystem for intelligent aircraft operation.

The secondary objectives are to build IPS modelfor the total system burden, and algorithm forintelligent operation so that potential savingsassociated with routing and other operational

issues may be explored. This would includeassessment of the cost, benefit and riskassociated with using routing for aircraft iceprotection.

1.2 Performance Based Navigation SystemGround-based systems have served the aviationcommunity well since inception; however asdemand for air transportation services increases,they do not permit the flexibility of point-to-point operations required for the future ATMenvironment [5]. Hence, ICAO has adopted thePerformance Based Navigation (PBN) system toaddress these challenges. PBN definesperformance requirements for aircraftnavigating on an Air Traffic Service (ATS)route, terminal procedure or in a designatedairspace. Through the application of AreaNavigation (RNAV) and Required NavigationPerformance (RNP) specifications, PBNprovides the means for flexible routes andterminal procedures [1]. The International AirTransport Association (IATA) estimated thatshorter PBN routes globally could cut CO2emissions by 13 million tonnes per year [6].Flying PBN routes eliminates 3.19 kg of CO2emissions for every kg of fuel savings [6]. Oneof the major advantages of PBN system is that itpermits optimal trajectory based operations byproviding very precise lateral and vertical flightpaths. From 2030 and beyond, aircraft areexpected to fly optimal trajectories that aredefined in the form of three dimensionalwaypoints plus associated required times (4D)of overfly [3].

1.3 Concept of Next generation AircraftThe next generation aircraft is considered to beextremely efficient by design, light weight(composites structures) with lower maintenanceand overall operating costs, and reduced impacton the environment. There is a lot of progresstowards the development of new and moreefficient de-icing systems that are compatiblewith next generation composite airframestructures. This includes the heater mattechnology, smart IPS and icephobic coatingsamong others. Projects such as Clean Sky,NextGen/SESAR, ON-WINGS, ERAST,

3

An Intelligent Ice Protection System for Next Generation Aircraft Trajectory Optimisation

Boeing CFD and EADS VoltAir are all projectspursued towards more efficient aircraft systemsenergy management and cleaner environment.

The next generation aircraft is conceptualisedutilising the All-Electric Aircraft (AEA)technology. Though the transition to AEAconfiguration is still far away, there is a greatachievement in a More Electric Aircraft (MEA)technology. In the MEA configuration, it isassumed that the majority of the airframesystems will be powered electrically retainingpresent form of hydraulic, mechanical orpneumatic power [7] whereas, the AEA conceptassumes that the all aircraft systems will bepowered electrically.

1.4 Recent Progress in Next Generation IPSThe classical pneumatic anti-icing systemtypically has a very simple on/off controlimplementation. It taps bleed air power from theengine which reduces the overall engineperformance. This causes high fuel consumptionand, CO2 and NOx emissions which affects airtransport costs and the environment impactalong the route. On the other hand, electricalanti/de-icing power could be provided bygenerators on-board. Goodrich Corp testedelectro-thermal technology on a Cessna 303Twing leading edge during the 2003/4 winter, andreported that between 20-50% energy was savedcompared to conventional anti-icing system [8].Boeing 787 is the first large fixed wing aircraftto use electro-thermal de-icing system onaircraft wings. The Boeing 787 powerconsumption was reduced to 45-75 kW usingthis technology compared to 150-200 kWneeded with classical technology [9]. Anotherrecent technology is the pulse electro-thermalde-icing system developed by Prof. VictorPetrenko [10]. According to Petrenko [11],when a pulse de-icer is fully optimised, only 1%of the energy requirement of a conventionalthermal de-icer would be demanded.

Conventional methods of protecting aircraftagainst inflight icing involve using Active IceProtection Systems (AIPS). However, AIPS arecharacterised by complexity and high fuel

consumption. Hence, there are on-goingresearch efforts aimed at developing Passive IceProtection Systems (PIPS) that are easy andrequire less energy. One prominent feature ofPIPS is the use of icephobic coatings on aircraftparts prone to inflight icing to reduce iceadherence to the surfaces.

Previous experiments have shown thaticephobic coatings have the potential tosignificantly reduce power consumption [12].However, these coatings must have certainchemical and physical properties to withstandaircraft harsh operating environment. Thus, theirdurability and expected service life have not yetbeen established so also their overall cost ascompared to AIPS. Erosion, corrosion andreaction with atmospheric substances vis-à-visits effects to the environment are a greatchallenge to understand at the moment.Response to lightening, electrostatic properties,interference with electromagnetic signals andavionic components are all issues that are notyet resolved [13].

2 Research Approach and Method

There are basically three methods of managingaircraft in-flight icing. These are throughdevelopment of efficient anti/de-icing systems,effective weather avoidance strategy, and or icetolerant aircraft designs. In the past, the effortsto reduce air transport cost and its negativeimpacts on the environment were mainlyfocused on aircraft and engine designs. Whileachieving safety, fuel consumption vis-à-viscosts could be minimised through climatecompatible operations by better exploitingfuture real time weather information.

Today, aircraft may deviate from its flight planonly for reasons of safety and weatherperturbations. However, due the growingimpacts of commercial aviation on theenvironment and oil resources, smart routingsare required to operate aircraft at low powerlevels to save fuel consumption. Recentadvances in intelligent ATM, digital control andcomputer based automation have provided


4

opportunities to investigate use of optimal flightroutes around disturbances which have thepotential to decrease safety and increase theenvironmental impact on aviation.

This approach, which is applicable to allairframe systems, is applied to the aircraft icingproblem in this work. In real sense, to achieveautonomous operation in icing conditions, acontrollable intelligent ice protection systemwhich enables the development of a decisionmaking process dependent on weather must beintegrated into the on-board FMS; thustransforming the conventional IPS into a moreintelligent system. The idea is to combine theeffect of technology and smart routing inaircraft anti-icing problem as illustrated inError! Reference source not found. Smartrouting is an avoidance strategy involvingweather-dependent optimized trajectories forefficient and low power levels operations.

Fig. 1 A Systematic Approach to ControllableAircraft Anti-icing in Future ATM Systems

In this approach, the IIPS takes only that muchpower required based on the icing inputs fromthe sensors or remote weather source. Thisapproach is an improvement on theconventional approach of representing only theaircraft dynamics and engines system;neglecting aircraft systems impacts.

3 Trajectory Optimisation Method

3.1 GATAC Simulation FrameworkGreener Aircraft Trajectories under ATMConstraints (GATAC) is a multi-objectiveoptimization framework for planningenvironmentally efficient trajectories. Thesoftware is co-developed between University ofMalta and Cranfield University. The Universityof Malta developed the infrastructure codewhile Cranfield University developed theoptimizer and models.

Fig. 2 GATAC Integration Framework Architecture[14]

The current GATAC v3 release allows the userto use four different optimisers to solvetrajectory optimization problems. The overalloptimisation block diagram of a genericproblem using GATAC is shown in Fig. 2.Further reading on GATAC can be obtained in[15].

3.2 Optimiser ChoiceThe following optimisers are available inGATAC v3:

Non-dominated Sorting GeneticAlgorithm Multi-Objective (NSGAMO)

Multi-Objective Tabu Search (MOTS) Hybrid Optimiser (HYOP)

The NSGAMO optimiser is capable ofperforming multi-objective optimisation underconstraints and is based on Genetic Algorithm(GA). The properties of GA algorithms tooptimise problems with local minimumsperfectly fit to the needs of this work. A bi-objective optimisation scheme which results in acreation of Pareto front was used to optimise thefuel and time. The theoretical optimal Paretofront solutions are ଵݔ ∈ ݔ,[0,1] = 0 where

5


i=2,…,n. The development processes andcapabilities of these models and the GATACsimulation framework are discussed below.

4 Models

To enable the simulation of aircraft operation inicing condition, an aircraft dynamics model andmajor airframe systems models were developedand integrated in the GATAC framework asillustrated in Fig. 3. The framework wasdeveloped and validated by using Cranfieldmodels.

Fig. 3 Models Interface Architecture

The model suite generates a representation ofairline operating costs in order to be able toprovide a realistic indication of the overall costsassociated with the flight time, fuel burn, noiseand emissions of the flight mission beingoptimized.

4.1 Aircraft ModelA representative model similar to Airbus A320based on EUROCONTROL BADA datasetparameters was used to model aircraft motion.The A320 aircraft was chosen because it is oneof the Clean Sky baseline aircraft for technologydemonstration. In addition, the A320 aircraftoffers advanced navigation technology such asRequired Navigation Performance (RNP)capability and Future Air Navigation System(FANS) which are part of the requirements forthe use of this methodology. The RNP reducesapproach distances for landing while reducingfuel consumption and CO2 emissions, whileFANS optimizes flight path and reduces aircraftspacing. Airbus A320 is an aircraft of the futureand therefore suitable for this research.

4.2 Ice Protection System PerformanceModellingThe essence of modelling the IPS operation wasto estimate the system power requirement undervarious icing conditions. The IPS is modelledbased on the Messinger [16] method utilizesconvection, sensible heating,evaporation/sublimation, kinetic energy andviscosity terms in the conservation energyequation to find the equilibrium temperature ofan unheated icing surface. Based on thismethod, the anti/de-icing energy estimated isequal to the energy resulting from the heatbalance which includes sensible heating,(ሶ௦௦ݍ) convective cooling (ሶ௩ݍ) andevaporative cooling .(ሶ௩ݍ) However, heatgains due to kinetic energy of the impingingdroplets and air must be accounted by thekinetic heating (ሶݍ) and aerodynamic heating(ሶݍ) terms, respectively. The anti-icingenergy equation is thus given as:

=ሶ௧ݍ ሶ௦௦ݍ + ሶ௩ݍ + ሶ௩ݍ − −ሶݍ

ሶݍ (1)ሶ௦௦ݍ = ݉ሶ.ܥೌ( ௦ܶ − ஶܶ ) (2)

ሶ௩ݍ = 0.7ℎܮቂோೞೠିಮ

ಮቃ (3)

ሶݍ = ݉ሶ௩ಮమ

ଶ(4)

ݍ = ܴℎ௩ಮమ

ଶುೌ൨ (5)

Where the local heat transfer coefficient, ℎ iscalculated from the following empirical relation:

ℎ = .ݑܰబ

௫(6)

The Nusselt (Nu), Prandtl (Pr) and Reynolds(Re) numbers are dimensionless quantities thatare calculated from the following relationships:

ݑܰ = 0.0296.ܴ ௫݁.଼.ܲݎ.ସ (7)

=ݎܲ .ఓ

బ(8)

ܴ݁=ఘಾ ೄಽ..

ఓ(9)

A.

The A320 anti-ice system uses both hot air andelectrical heating to protect the critical areas ofthe aircraft. During flight, hot air from theengine IP and HP bleed ports is used to heat theengine nacelle and, slats 3, 4 and 5energy is used instead for de-icing windscreen,probes and waste water drain mast.the parameters used are shown in Table 2model however, is reconfigurablemedium to large fixed wing aircraftanti/de-icing modelling processes were coveredin [17]. The electrical power requirement isgiven in shaft kW and the bleed requirement isgiven in bleed kg/s; the two of which areconnected to the engine offtake model. Thedifference between this model and the baselineaicraft AI is in the operation. In this model, theIPS algorithm penalises the engine based onicing inputs from sensors/weather data andaircraft mission parameters such as air speedand altitude. In this way, the system demandsonly the amount of power required for thecurrent operation in contrast to the baseline AIsystem whereby a fixed value of kg/s and kWare provided for ice protection.

Table 1. Modelling and Simulation Parameters

InputsParameter ValuesAltitude (h) User defineAmbient temperature ( ஶܶ ) User defineSurface heat transfer area (S0) User defineFlight speed (VTAS) User defineClouds liquid water content (LWC) User define

Internal constantsMean aerodynamic chord (LMAC) 2.2Slat length (ySLAT) 3.14Leading edge sweep (߮ா) 27.5Skin temperature ( ௦ܶ) 5MVD (dmed) 20Pressure (P) f(h)Saturation pressure (e) f(T)Relative humidity (Rh) 100Specific heat of air (ೌܥ) 1005

Specific heat of water ೢܥ) ೌ) @ 0°C 1859

Specific density of water (ρwater) 1000Latent heat for water evaporation (Le) 2257Latent heat of fusion (Lf) of ice 332.5Air density (ρ) f(h)Absolute viscosity of air (µ) 1.5636x10Thermal conductivity of air (k0) 0.0228

OutputsHeat flux (௧̇ݍ) ResultBleed mass flow rate (݉̇ ௗ) Result

A. SHINKAFI, C. LAWSON, R. SERESINHE, D. QUAGLIA AND I M

6

ice system uses both hot air andelectrical heating to protect the critical areas of

During flight, hot air from theand HP bleed ports is used to heat the

engine nacelle and, slats 3, 4 and 5. Electricalicing windscreen,

probes and waste water drain mast. Summary ofshown in Table 2. The

rable for anymedium to large fixed wing aircraft. The

ling processes were coveredThe electrical power requirement is

given in shaft kW and the bleed requirement isthe two of which are

connected to the engine offtake model. Thedifference between this model and the baselineaicraft AI is in the operation. In this model, the

penalises the engine based onicing inputs from sensors/weather data and

ft mission parameters such as air speedand altitude. In this way, the system demandsonly the amount of power required for thecurrent operation in contrast to the baseline AIsystem whereby a fixed value of kg/s and kW

Modelling and Simulation Parameters

Values UnitsUser define ftUser define °CUser define m2

User define ktUser define g/m3

mm°°CµmhPahPa%J/kg.K

J/kg.K

Kg/m3

kJ/kgkJ/kgkg/m3

1.5636x10-5 kg/s.mW/m.K

kWkg/s

4.2.1 IPS Model ValidationThe model performance was evaluated based onan icing experimental test conducted by AlKhalil et al. [18] on the engine intake of aturbine aircraft. The same test caseKhalil experiment was run with the modeldeveloped in this work and the results comparedvery well with the experimental result as shownin Fig. 4.

Fig. 4 IPS Model Validation with Experimental Data

The percentage deviation of the model resultfrom the experimental resultpower plotted in Fig. 5discrepancies in all the six cases.

Fig. 5 Percentage Deviation from Experimental Data

Although, there is value in investigating morecases, with the satisfactoryand the fact that the range here covers theAppendix C temperature range (0 tothe most severe CM icing condition makes theresult of the model valid.

4.3 Aircraft Dynamics ModelThe Aircraft Dynamics Model (ADM) isintegrated model in GATAC which is

R. SERESINHE, D. QUAGLIA AND I MADANI

alidationhe model performance was evaluated based on

an icing experimental test conducted by Al-on the engine intake of a

The same test case as the Al-was run with the model

developed in this work and the results comparedvery well with the experimental result as shown

IPS Model Validation with Experimental Data

percentage deviation of the model resultfrom the experimental result in terms of total

5 shows less than 20%in all the six cases.

Percentage Deviation from Experimental Data

Although, there is value in investigating morecases, with the satisfactory sensitivity resultsand the fact that the range here covers theAppendix C temperature range (0 to -30°C) forthe most severe CM icing condition makes theresult of the model valid.

Dynamics ModelThe Aircraft Dynamics Model (ADM) is anintegrated model in GATAC which is in charge

7


of the aircraft trajectory generation of a genericaircraft between two pre-defined positions in a3D space [19]. The generic aircraft is modelledusing the rigid body idealisation with varyingmass under aerodynamic, propulsive andgravitational forces with assumption ofsymmetrical aircraft with thrust force parallel tothe motion. In addition the assumptions ofspherical, non-rotating Earth and no windatmosphere are also introduced to simply theproblem. The aircraft motion is described byusing point mass with three degrees of freedomand the resulting differential algebraic equationsare listed below:

݉ௗ

ௗ௧= ܶ− ܦ − ݉݃ sin(ߛ) (10)

ܸ݉ (ߛ)ݏܿௗஎ

ௗ௧= ݅ݏܮ (ߤ݊) (11)

ܸ݉ௗఊ

ௗ௧= −(ߤ)cosܮ ݉݃ cos (ߛ) (12)

ௗ

ௗ௧= −ܿܶ (13)

(ܴா + ℎ)ௗఝ

ௗ௧= ܸ (ߛ)ݏܿ cos(χ) (14)

(ܴா + ℎ)ௗఒ

ௗ௧= ܸ sin(ߛ)ݏܿ ( )߯ (15)

ௗ

ௗ௧= ݅ݏܸ (ߛ݊) (16)

The aerodynamic forces are modelled by dragpolar characteristic provided by BADA dataset[20] and the gravitational forces are modelledusing the International Standard Atmosphere(ISA) datum value (9.81 ms-2). The ADMgenerated 3D trajectories based on variablesprovided by the optimiser regarding waypointpositions and altitude and airspeed informationalong the trajectory. Several input parameterssuch as initial and final positions and speed andaircraft initial mass are required to support theoptimal variable to generate the trajectory andevaluate the overall fuel consumption and flighttime, and emission indexes. The optimisationprocess will evaluate many possible trajectoriesby varying the trajectory variables previouslyintroduced and refine the search by minimizingthe imposed objectives.

4.4 Engine ModelThe engine model is based on performance dataobtained by Cranfield’s in-house developed gasturbine performance code called Turbomatch.

The Turbomatch model is similar to the CFMInternational CFM56-5B4 turbofan engine. Themodel was validated using EASA [21] andICAO [22] data. The engine was simulated for avast envelope of off-design conditions and aperformance database was created inMatlab/Simulink. Fig. 6 shows the performancedata obtained from the engine model.

Fig. 6 SFC vs net thrust at SL, ISA

Fig. 7 shows the Simulink model of the engine.

Fig. 7 Turbofan engine Simulink model – Inputs &Outputs

4.5 Off-take ModelIn current large commercial aircraft, thesecondary power system which includes the IPSis powered from energy extracted from theaircraft engines. As per the nature of the sub-systems the extractions can be divided intobleed air power and shaft power. Typically thebleed air provides power for ECS and IPS. Theshaft power extractions from the engines areused to operate the hydraulic pumps whichenable the hydraulic system to operate theconventional actuators for the flight controls.Moreover the shaft power extraction from theengines also provides power for the electrical


8

generators mounted on the engines which runthe electrics of the aircraft.

The Off-takes model provides the interfacebetween the aircraft systems and the engine.The model is based on the methodology in [23].The off-takes model takes the fuel flow for anoperating condition of the aircraft as an inputand then corrects it to represent the penalty dueto the operation of aircraft systems. In this casethe IPS is operated with bleed air extracted fromthe HPC of the engine. The model wasdeveloped in Matlab/Simulink and shown inFig. 8.

Fig. 8 Off-takes model – Inputs & Outputs

4.6 Emissions ModelThe emissions model was based on the P3T3methodology in SAE International, 2009. It is acorrection based method where emissionsmeasurements taken in accordance of ICAO

Annex 16 certification engine testing arecorrected to required altitude, using combustoroperating parameters at both ground level andthe required altitude. The baseline emissionsindices were extracted from. The model also hasthe capability to correct the emissions as per thehumidity, but for this study the relativehumidity was set at 60%. The model wasdeveloped in Matlab/Simulink and theschematic with the inputs and outputs as shownin Fig. 9.

Fig. 9 Emissions model –Inputs & Outputs

Further reading on the emissions model can befound in [24].

5 Discussions and Analysis of Results

5.1 Case OverviewThe selected test case is a flight from LondonAirport Heathrow (EGLL/LHR) to AmsterdamAirport Schiphol (EHAM/AMS). Fig. 10 showsa graphical projection of a typical real flighttrajectory obtained from FlightAware, a privateflight tracking company [25].

Fig. 10

The British Airways and KLM Royal Dutchairlines are the major airlines operating thisroute. The BA uses A319 and A320 aircraftwhereas KLM uses F70 and E190 on this route.This analysis considers an A320this route in the presence of icinThe degree of protection provided was basedCS25 Appendix C icing envelopeTable 2.

Table 2. Appendix C Icing Envelope

Case T0(°C)

LWC(g/m3)

MVD(µm)

1 0 0.63 202 -10 0.43 203 -20 0.22 204 -30 0.15 20

Typically, aircraft encounters icing during climbto cruise altitude, and descent and hold. This isdue the fact that the clouds that contain supercooled water droplets exist at lower flightlevels, and during descent and climb, aircraftreduces speed which lessens the effects ofkinetic heating on the airframe. Between 7,000ft and 13,500 ft during departure and arrival, thealgorithm simulates icing condition whichautomatically puts on the IPS and penaliseengine accordingly.

5.2 Fuel vs TimeCase 3 which represents the most probable icingcondition was used for optimising aircrafttrajectory with a view to minimisfuel consumption and total flight time

9

10 London – Amsterdam Flight Route [25]

The British Airways and KLM Royal Dutchairlines are the major airlines operating thisroute. The BA uses A319 and A320 aircraftwhereas KLM uses F70 and E190 on this route.

A320 aircraft flyingthis route in the presence of icing conditions.The degree of protection provided was based on

icing envelope shown in

Altituderange (ft)0-120000-170000-220000-22000

Typically, aircraft encounters icing during climbto cruise altitude, and descent and hold. This isdue the fact that the clouds that contain super-cooled water droplets exist at lower flightlevels, and during descent and climb, aircraft

ch lessens the effects ofBetween 7,000

ft and 13,500 ft during departure and arrival, thealgorithm simulates icing condition whichautomatically puts on the IPS and penalises the

which represents the most probable icingor optimising aircraft

a view to minimising the totaltime.

5.2.1 DepartureFig. 11 shows the Pareto fronts for theconventional method and the enhanced methodafter optimisation of the departure.the loop optimisation approachfuel savings over usingmethod. The two resultsrespect to minimum timeexpected in the case oflike this one.

Fig. 11 Difference in Pareto Front betweenConventional and Icing in the Loop Approaches

5.2.2 London - Amsterdam RouteThe Pareto front for Coptimisation is shown in

shows the Pareto fronts for theconventional method and the enhanced methodafter optimisation of the departure. The icing inthe loop optimisation approach gave a 2.1 %fuel savings over using the conventional

results are quite close withrespect to minimum time trajectory. This is

n the case of a short flight segment

Difference in Pareto Front betweenConventional and Icing in the Loop Approaches

Amsterdam RouteCase 3 for full routeFig. 12.

A.

Fig. 12 Pareto Front for London-Amsterdam RouteOptimisation

The Pareto shows a difference of 252between the minimum time and minimum fueltrajectories within 3 minutesarrival time. A further simulation was carriedout using Case 1 which represents a condition ofhighest liquid water concentration andhighest icing. The two Pareto fronts arecompared in Fig. 13. A difference of 66 kg fuelburn was realised between the two cases.

Fig. 13 Pareto Front of Fuel vs Time Simulation

The minimum fuel and minimum timetrajectories for Case 3 were compared with atypical trajectory flown by commercial aircraftfrom EGLL/LHR to Amsterdam AirportSchiphol EHAM/AMS as shown inminimum time trajectory was very close to thetypical trajectory of commercial aircraft. One ofthe things learnt from this result is that thetypical aircraft finishes climb earlier than thesimulated baseline aircraft. This could beassociated with ATM operational requirement atLHR.

A. SHINKAFI, C. LAWSON, R. SERESINHE, D. QUAGLIA AND I M

10

Amsterdam Route

The Pareto shows a difference of 252 kgimum time and minimum fuel

s difference ofmulation was carried

represents a condition ofhighest liquid water concentration and the

The two Pareto fronts areA difference of 66 kg fuel

realised between the two cases.

Pareto Front of Fuel vs Time Simulation

The minimum fuel and minimum timewere compared with a

typical trajectory flown by commercial aircraftfrom EGLL/LHR to Amsterdam AirportSchiphol EHAM/AMS as shown in Fig. 14. The

time trajectory was very close to theercial aircraft. One of

things learnt from this result is that thetypical aircraft finishes climb earlier than thesimulated baseline aircraft. This could beassociated with ATM operational requirement at

Fig. 14 Trajectory for Min Time and Min Fuel

Fig 17 shows the speed variation with distance.As usual, the minimum fuel trajectory has lowerspeed profile than minimum time and typicaltrajectories. Because of this difference in speedprofile, the fuel flow requirementinvestigated.

Fig. 15 TAS Comparison

Theoretically, IPS power demand is a functionof air speed and altitude. Hence it can be seenthat 200 km from LHR, the IPS wasautonomously operated with rest to theminimum time trajectory.flow comparison.

R. SERESINHE, D. QUAGLIA AND I MADANI

ctory for Min Time and Min Fuel

Fig 17 shows the speed variation with distance.As usual, the minimum fuel trajectory has lowerspeed profile than minimum time and typicaltrajectories. Because of this difference in speedprofile, the fuel flow requirement was

TAS Comparison

IPS power demand is a functionof air speed and altitude. Hence it can be seenthat 200 km from LHR, the IPS wasautonomously operated with rest to theminimum time trajectory. Fig. 16 shows bleed


Fig. 16 Bleed Flow due to IPS

Fig. 17 shows the response of the fuel flow tothe engine based on the trajectory choice.Although, the initial mass of the aircraft in thetypical trajectory is not known, same conditionhas been applied to it as the study case for easeof comparison.

Fig. 17 Fuel Flow vs Distance

The result is as expected with minimum fueltrajectory having lower fuel flow requirement asthe minimum time or the low altitude typicaltrajectory.

Fig. 18 Fuel Consumption Comparison

In can be noted in Fig. 18 that min fueltrajectory saves more fuel (10.3%) than thetypical and min time trajectory. This is because

11


Bleed Flow due to IPS

shows the response of the fuel flow toed on the trajectory choice.

Although, the initial mass of the aircraft in thetypical trajectory is not known, same condition

dy case for ease

low vs Distance

result is as expected with minimum fueltrajectory having lower fuel flow requirement asthe minimum time or the low altitude typical

Fuel Consumption Comparison

that min fueltrajectory saves more fuel (10.3%) than the

. This is because

the min time trajectory is achieved throughhigher TAS as can be seen inthe typical trajectory is at similar TAS with theminimum fuel trajectory, it is at a lower altitudewhere air drag is highest.

6 Summary and Conclusions

This work enabled the developand tools contributing to the evaluation ofaircraft fuel burn under various environmentalconditions. These models and tools enabled adependable assessment of fuelpollutant emissions reduction thrprofile optimization.quantitative estimates of the energymajor airframe systemsdevelopment process for the GATACAircraft Trajectories under ATM Constraintssoftware.

Although, operation depends solely on theroute, operator and aircraft type, it can be notedfrom the solution of the GATAC trajectoryoptimization that there are fuel savings andenvironmental benefits in flying optimizedroutes in comparison to a typical flight routebetween LHR to EGLL. Both BA anairlines fly the route between 22,000 ft and25,000 ft. Meanwhile, the most economic flightprofile obtained from the GATAC optimizationis a steady climb from LHR to 37,000 ft andcruising for about 5 minutes before gradualdescent to EGLL. The perslower altitudes by the regular operators of theroute could be explainedminimum time trajectory. ATM constraintalso likely to be a factor.

7 Future Work

Future work involves the integration of theweather model and impstrategy in aircraft icing problem.avoidance strategy is only applicable in aPerformance-Based Navigation (PBN)environment. It is therefore basednetwork-enabled informationdesigned for future ATM environment where


the min time trajectory is achieved throughhigher TAS as can be seen in Fig. 15. Althoughthe typical trajectory is at similar TAS with the

fuel trajectory, it is at a lower altitudewhere air drag is highest.

Conclusions

This work enabled the development of modelsand tools contributing to the evaluation of

under various environmental. These models and tools enabled a

dependable assessment of fuel consumption andpollutant emissions reduction through a mission

The models providequantitative estimates of the energy used bymajor airframe systems and were partdevelopment process for the GATAC (GreenerAircraft Trajectories under ATM Constraints)

Although, operation depends solely on thete, operator and aircraft type, it can be noted

from the solution of the GATAC trajectoryoptimization that there are fuel savings andenvironmental benefits in flying optimizedroutes in comparison to a typical flight routebetween LHR to EGLL. Both BA and KLMairlines fly the route between 22,000 ft and25,000 ft. Meanwhile, the most economic flightprofile obtained from the GATAC optimizationis a steady climb from LHR to 37,000 ft andcruising for about 5 minutes before gradualdescent to EGLL. The persistence of usinglower altitudes by the regular operators of theroute could be explained considering theminimum time trajectory. ATM constraints arealso likely to be a factor.

work involves the integration of theweather model and implementing avoidance

aircraft icing problem. Theavoidance strategy is only applicable in a

Based Navigation (PBN)is therefore based on shared

enabled information capability. It isdesigned for future ATM environment where


12

flight operators must have access to current andplanned strategies to deal with congestion andother airspace constraints. These constraintsmight include scheduled times of use for specialactivity airspace for military, security or spaceoperations. They could also include changes toprocedure due to current or forecastweather/congestion as well as the status of areaproperties and facilities, such as out-of-servicenavigational aids. In summary, flight operatorsmust continuously obtain up-to-date regular aswell as potential ATM limitations, from groundoperations to the intended flight trajectory.

In the future work, smart routing techniques willbe used to optimize aircraft trajectory in thepresence of icing condition. The majority ofcommercial aircraft nowadays carry an AirborneWeather Radar (AWR) system that is most oftenbuilt into the aircraft nose. These on-boardweather radars provide the pilots with a localweather picture ahead of the aircraft whichallows them to identify and avoid specific,undesirable weather formations.

Ice is detected inflight using devices that candetect the presence of ice and send advisorysignal to pilot in some cases trigger a de-icingaction. Ice detectors can be used in a primary oradvisory role. Currently, there are no remote oron-board sensors that can reliably and routinelyquantify liquid water content or drop size. Atthe moment, the AWR system on-boardcommercial aircraft is used for the detection ofthunderstorms, areas of strong precipitation andturbulence only. Some AWR systems have amaximum range of up to 180 nm with 360-degree coverage. However, since precipitation isassociated with clouds formation, and cloudsand moisture are the main factors for aircrafticing, it can be used in combination withtemperature, LWC and droplet size models forpredicting and avoiding icing areas.

To achieve autonomous capability, the IIPSmust be connected with the FMS whose primaryfunction is the management of the flyingaircraft. The FMS using both the GPS and INSto determine the aircraft position can guide the

aircraft along a new path safely andconveniently. When icing hazard is detected,two methods are involved in solving theproblem. First is to activate the de-icingsequence through electrically powered thermalheating of the critical surfaces; second, to avoidthe icing condition by modifying the trajectoryto fly in a less severe icing condition. Thecontrol law governing the decision process overthe efficient operation between the nominal andoptimised trajectories is given by:

ݑ = [ܶℎ, ∅, ]݊ (17)

Where ܶℎ = thrust, ∅ = bank angle and n =vertical load factor; all of which are function oftime. The thrust component affects thelongitudinal motion mainly, whereas ∅ affectsthe lateral motion and n affects the verticalmotion of the aircraft.

Copyright Statement

The authors confirm that they, and/or theircompany or organization, hold copyright on allof the original material included in this paper.The authors also confirm that they haveobtained permission, from the copyright holderof any third party material included in thispaper, to publish it as part of their paper. Theauthors confirm that they give permission, orhave obtained permission from the copyrightholder of this paper, for the publication anddistribution of this paper as part of theICAS2014 proceedings or as individual off-prints from the proceedings.

References

[1] ICAO (2014), ICAO Performance BasedNavigation Programme, available at:http://www.icao.int/safety/pbn/Pages/default.aspx (accessed 02/25).

[2] ICAO (2014), 2013 ICAO Air TransportResults Confirm Ruburst PassengerDemand, Sluggish Cargo Market, availableat:http://www.icao.int/Newsroom/News%20D

13


oc%202013/COM.43.13.ECON-RESULTS.Final-2.en.pdf (accessedJan/15).

[3] SESAR (2012), The Roadmap forSustatinable Air Traffic Management:European ATM Master Plan, Edition 2,EU.

[4] Advisory Council for AeronauticsResearch in Europe (2008),Addendum to the Strategic ResearchAgenda, 2008. Brussels, ACARE.

[5] Schumann, U. (2012), "Volcanic, Weatherand Climate Effects on Air Transport", 28thInternational Congress of the AeronauticalSciences, 23-28 September, 2012, Brisbane,Optimage Ltd, Brisbane, Australia, .

[6] ICAO (2009), Performance BasedNavigation, Document 9613, ICAO.

[7] Laskaridis, P. (2004), PerformanceInvestigations and Systems Architecturesfor the More Electric Aircraft (PhD thesis),Cranfield University, Cranfield University.

[8] Botura, G. C., Sweet, D. and Flosdorf, D.(2005), Development and Demonstration ofLow Power Electrothermal De-icingSystem, AIAA 2005-1460, AIAA,Washington.

[9] GKN (2010), Composite Aircraft WingResearch gets underway at GKNAerospace, available at:http://www.gknaerospace.com/newsarticle.aspx?page=S633463542178688750&ArchiveID=5&CategoryID=33&ItemID=313&src= (accessed 05/15).

[10] Petrenko, V., ( 2006), Method forModifying Friction Between an Object andIce or Snow (Patent No. US 7,034,257),219/482 ed., H05B 3/02, Hanover/US.

[11] Petrenko, V. F., Sullivan, C. R.,Kozlyuk, V., Petrenko, F. V. and

Veerasamy, V. (2011), "PulseElectrothermal De-icer (PETD)", ColdRegions Science and Technology, , no. 65,pp. 70-78.

[12] Fortin, G. (2013), "Considerations on theUse of Hydrophobic, Superhydrophobic orIcephobic Coatings as a Part of the AircraftIce Protection System", SAE 2013 AerotechCongress and Exhibition, 24-26 September2013, Montereal, Canada, .

[13] Shinkafi, A. and Lawson, C. (2013),"Evaluating Inflight Ice Protection Methodsfor Applications on Next GenerationAircraft", Journal of AerospaceEngineering and Technology, , no. 2231-038X.

[14] Sammut, M., Xuereb, M., Chircop, K.,Camilleri, W., Dimech, E., Karumbaiah, D.and Pervier, H. (2013), System for GreenOperations (SGO) ITD: GATAC V3 UserManual, , Clean Sky.

[15] Chircop, K., Xuereb, M., Zammit-Mangion, D. and Cachia, E. (2010), "AGeneric Framework for Multi-parameterOptimization of Flight Trajectories",Internationla Congress of the AeronauticalSciences, vol. ICAS2010.

[16] Messinger, B. L. (1953), "EquilibriumTemperature of an Unheated Icing Surfaceas a Function of Air Speed", Journal of TheAeronautical Sciences, vol. 20, no. 1, pp.29-42.

[17] Shinkafi, A. and Lawson, C. (2014),"Enhanced Method of Conceptual Sizing ofAircraft Electro-Thermal De-icingSystem", ICAA 2014 : InternationalConference on Aeronautics andAstronautics, Vol. 9, 5-6 June 2014, WorldAcademy of Science, Engineering andTechnology, New York, .

[18] Al-Khalil, A., Hitzigrath, R., Philippi, O.and Bidwell, C. (2000), "Icing Analysis and


14

Test of a Business Jet Engine Inlet Duct",38th Aerospace Sciences Meeting &Exhibit, 10-13 January 2000, .

[19] Quaglia, D., Ramasamy, S. and Gardi,A. (2014), Systems for Green Operations(SGO) ITD Software Design Description:Aircraft Dynamics Model (ADM) for3D/4D Trajectories, SGO-WP 3.1-C-U-OUT-0327, Clean Sky, CranfieldUniversity, UK.

[20] EUROCONTROL (2009), Base ofAircraft Data (BADA) AircraftPerformance Modelling Report, 2009-009,EUROCONTROL, Bretigny-sur-Orge.

[21] EASA (2012), E.003 CFM InternationalS.A. - CFM56-5B and 5C Series Engines,TCDS E.003, EASA, Cologne, Germany.

[22] ICAO (2013), ICAO Engine ExhaustEmissions Data Bank Subsonic Engines, ,EASA, Cologne, Germany.

[23] Scholz, D., Seresinhe, R., Staack, I. andLawson, C. (2013), Fuel Comsumption dueto Shaft Power Off-take from the Engine, ,AST, Hamburg, Germany.

[24] Seresinhe, R., Shinkafi, A., Quaglia, D.,Lawson, C. and Madani, I. (2014),"Airframe Systems Power Off-takeModelling in More-Electric LargeCommercial Aircraft for use in TrajectoryOptimisation", 29th Conference of theInternational Council of the Aerospace, 7-12 Sep, St. Petersburg, IACAS, St.Petersburg, Russia, .

[25] FlightAware (2014), Live FlightTracking, available at:http://uk.flightaware.com/live/flight/BAW442/history/20140418/1700Z/EGLL/EHAM(accessed May/23).

an intelligent ice protection system for next …an intelligent ice protection system for next...

Documents