www.eera-avatar.eu

This project has received funding from the European Union’s Seventh Programme for research, technological development and demonstration under grand agreement No FP7-ENERGY-2013-1/n° 608396 .

AVATAR projectAdvanced Aerodynamic Tools for lArge Rotors

Gerard Schepers

2016 Wind Turbine Blade WorkshopSandia National Laboratories, USA

August 31st, 2016

FP7-ENERGY-2013-1/ n° 608396 2

List of Contents

q Introductionintotheprojectq DesignofAVATARReferenceWindTurbine1)q Aerodynamics athighReynoldsnumbers

• Airfoil measurements inpressurized DNW-HDGtunnel2)q Aero-elasticity oflargeturbinesatyaw

• BEMversusfreevortexwakeaerodynamic modelling3)

19-09-16

1) Acknowledgement G.Sieros,M.Stettner2) Acknowledgement O.Ceyhan,O.Pires3) AcknowledgementK.Boorsma,S.Voutsinas

Andallotherprojectpartners!

FP7-ENERGY-2013-1/ n° 608396 3

EU FP7 Project initiated by EERA

1. Energy Research Centre of the Netherlands, ECN (Coordinator)

2. Delft University of Technology, TUDelft

3. Technical University of Denmark, DTU

4. Fraunhofer IWES

5. University of Oldenburg, Forwind

6. University of Stuttgart, USTUTT

7. National Renewable Energy Centre, CENER

8. University of Liverpool/University of Glasgow, ULIV/UoG

9. Centre for Renewable Energy Sources and Saving, CRES

10.National Technical University of Greece, NTUA

11.Politecnico di Milano, Polimi

12.GE Global Research, Zweigniederlassung der General Electric Deutschland Holding GmbH, GE

13.LM Wind Power, LM

19-09-16

FP7-ENERGY-2013-1/ n° 608396 4

Period

• Project period: November 1st 2013- November 1st 2017

FP7-ENERGY-2013-1/ n° 608396 5

Main motivation for AVATAR:Aerodynamics of large wind turbines (10-20MW)

• We simply don’t know if present aerodynamic models are good enough to design 10MW+ turbines• 10MW+ rotors violate assumptions in current aerodynamic tools,

e.g.:– Reynoldsnumbereffects,– Compressibility effects– Thick(er)airfoils– Flowtransitionandseparation,– (More)flexibleblades– Flowdevices• Hence 10MW+ designs fall outside the validated range of

current state of the art tools.

19-09-16

FP7-ENERGY-2013-1/ n° 608396 6

Avatar: Main objective

To bring the aerodynamic and fluid-structure models to a

next level and calibrate them for all relevant aspects of

large (10MW+) wind turbines

19-09-16

FP7-ENERGY-2013-1/ n° 608396 7

Avatar: Work procedure– Problem:No10MWturbinesareonthemarketyetforvalidation– Hence:Validatesubmodels againstexperiments

• PressurizedHDGtunnelofGermanDutchWindTunnelfacilities(DNW)• AirfoilmeasurementsatReynoldsnumbersuptoRE=15M andlowMach(<0.2)

• LM:Windtunnelairfoilmeasurements• Forwind:Windtunnelairfoilmeasurementsatrepresentativeturbulence• TUDelft:Windtunnelexperimentsonairfoilswithvortexgenerators,flaps• NTUA:Windtunnelexperiments onairfoils with/withoutvortex generators• DTU:Danaero:Aerodynamicfieldexperimentsona2.3MWturbineandsupporting2Dwindtunnelmeasurements

19-09-16

FP7-ENERGY-2013-1/ n° 608396 8

Avatar: Work procedure

Usethedifferentmodelsfrompartnersintheprojecto Itisacooperation project!o Intheprojectwehavemanymodelswhichrangefromcomputational

efficient‘engineering’toolstohighfidelitybutcomputationallyexpensivetools

o Engineeringtoolsareneeded inindustrial designcodes1)o Highfidelitymodels(andintermediatemodels)feedinformationtowards

engineeringmodels

19-09-16

1)J.G.Schepers‘Engineeringmodelsinwindenergyaerodynamics,’(2012).http://repository.tudelft.nl/islandora/object/uuid:92123c07-cc12-4945-973f-103bd744ec87/?collection=research

FP7-ENERGY-2013-1/ n° 608396 9

Avatar: Work procedure

• Demonstratethevalueoftheimprovedtoolson10MWreference rotors withandwithoutflowcontroldevices1. INNWIND.EUreferencerotor(moreorlessconventional

designphilosophy)2. AVATARreferencerotorwhichshouldbemore

challengingfromanaerodynamicpointofview(e.g.lowerinduction,longer,moreslenderblades,thickerairfoils,highertipspeed).

• Compareresultsfrom‘old’andimprovedmodelsattheendoftheproject

FP7-ENERGY-2013-1/ n° 608396 10

List of Contents

q Introductionintotheprojectq DesignofAVATARReferenceWindTurbine1)q Aerodynamics athighReynoldsnumbers

• Airfoil measurements inpressurized DNW-HDGtunnelq Aero-elasticity oflargeturbinesatyaw

• BEMversusfreevortexwakeaerodynamic modelling

19-09-161) Acknowledgement G.Sieros,M.Stettner

AVATAR RWT

Power: 10MW 10MW

Rotordiameter: 178.3m 205.8m

WTPD: 400W/m2 300W/m2

Axialinduction: 0.3 0.24

RPMàTip speed 9.8rpmà 90m/s 9.8à103.4m/s

Hubheight: 119m 132.7m

FP7-ENERGY-2013-1/ n° 608396

Classical Approach versus Low Induction

• PowerCoefficient flataroundBetzmaximum(a=1/3)

• Aerodynamic loadcoefficientstronglydependantona

• Increasediameterà maintainaerodynamicloadsà increasepower

2

3)1(4

21 aaAU

PCP ==

0

0.2

0.4

0.6

0.8

1

1.2

0 0.1 0.2 0.3 0.4 0.5 0.6

Cdax

Cp

)1(4

21

.2

. aaAU

axDC axD ==

a[-]

FP7-ENERGY-2013-1/ n° 608396 13

Design of AVATAR RWT

• 5% Increase in energy production due to larger diameter• Key rotor load levels are maintained• Non-rotor loads exceededà Redesign of AVATAR rotor at end of project• Note: LCOE of AVATAR turbine assessed

in 1) taking into account additionaladvantage of lower wake effects

1)R.Quinn, B.Bulder,J.G.SchepersAparametricinvestigation intotheeffectoflowinductionrotor(LIR)windturbinesontheLCoE ofa1GWoffshorewindfarminaNorthSeawindclimate,EERA-Deepwind 2016

FP7-ENERGY-2013-1/ n° 608396

Design of AVATAR RWT

• The operational conditions

Section Thickness Re (rated) Ma (rated) Re (Min) Ma (Min)60.0% 7.0×106 0.05 4.4×106 0.03

40.1% 11.0×106 0.07 7.0×106 0.05

35.0% 14.0×106 0.09 9.0×106 0.06

30.0% 17.0×106 0.12 10.0×106 0.07

24.0% 20.0×106 0.16 12.0×106 0.10

24.0% 16.0×106 0.25 11.0×106 0.15

24.0% 13.0×106 0.30 8.0×106 0.18

21.0% 20.0×106 0.16 12.0×106 0.10

21.0% 16.0×106 0.25 11.0×106 0.15

21.0% 13.0×106 0.30 8.0×106 0.18

Section Thickness Re (rated) Ma (rated) Re (Min) Ma (Min)60.0% 7.0×106 0.05 4.4×106 0.03

40.1% 11.0×106 0.07 7.0×106 0.05

35.0% 14.0×106 0.09 9.0×106 0.06

30.0% 17.0×106 0.12 10.0×106 0.07

24.0% 20.0×106 0.16 12.0×106 0.10

24.0% 16.0×106 0.25 11.0×106 0.15

24.0% 13.0×106 0.30 8.0×106 0.18

21.0% 20.0×106 0.16 12.0×106 0.10

21.0% 16.0×106 0.25 11.0×106 0.15

21.0% 13.0×106 0.30 8.0×106 0.18

FP7-ENERGY-2013-1/ n° 608396 15

List of Contents

q Introductionintotheprojectq DesignofAVATARReferenceWindTurbineq Aerodynamics athighReynolds numbers

• Airfoil measurements inpressurized DNW-HDGtunnel1)q Aero-elasticity oflargeturbinesatyaw

• BEMversusfreevortexwakeaerodynamic modelling

19-09-16

1) Acknowledgement O.Ceyhan,O.Pires

FP7-ENERGY-2013-1/ n° 608396

Measurements in DNW-HDG pressurized tunnel

• Measurements up to Re = 15M (and low M)• DU00-W-212 selected as common airfoil– Alsomeasuredby:– LMuptoRE=6M– Forwind atcontrolledturbulentconditionsuptoRe=1M

• Results are brought into a ‘blind test’ – Measurementscomparedwithcalculations– Blindtestincludedparticipantsoutsideproject

DNW-HDGmodel,c=15cm

FP7-ENERGY-2013-1/ n° 608396

DNW-HDG Wind Tunnel

General Tunnel Characteristics

Test section : 60cmx60cmFan Rpm : 200 – 820 Fan blade angle

fixedTunnel Pres. : 1 – 100×105 PaTunnel Temp.: ambient to 45°C

FP7-ENERGY-2013-1/ n° 608396

Test Section

Probeholderforhotwire

90PTsathalfspan

Wakerake118pitottubes6pressuretaps

Kulites4pressureside1suctionside

3-componentBalance

Sensicam &UVLED’sFlowvisualization

FP7-ENERGY-2013-1/ n° 608396

Participants/Codes

Test1.Re=3mil

Test2.Re=6mil-1

Test3.Re=6mil-2

Test4.Re=9mil-1

Test5.Re=9mil-2

Test6. Test7.Re=12mil Re=15mil

Pt [bars] 12 34 67 34 67 67 60Vtunnel [m/s] 25.6 19 10 28.6 15 20 28.4

FullCFD

DTU/EllipSys FullyturbulentTransition Yes Yes Yes Yes Yes Yes Yes

KIEL/TAU FullyturbulentTransition Yes Yes Yes Yes Yes Yes Yes

NTUA/Mapflow FullyturbulentTransition Yes Yes Yes Yes Yes Yes Yes

UoG/HMB Fullyturbulent Yes YesTransition Yes Yes

Forwind-IWES/OpenFOAM Fullyturbulent Yes Yes Yes Yes Yes Yes YesTransition

Pane

lMetho

ds USTUTT/XFOILvUSTUTTFullyturbulent

Transition Yes Yes Yes Yes Yes Yes Yes

ORECatapult/XFOILv6.96Fullyturbulent

Transition Yes Yes Yes Yes Yes Yes Yes

FP7-ENERGY-2013-1/ n° 608396

DNW-HDG Full CFD calculations vs measurementsEffect in Blade Design parameter: Cl/Cd

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpDTU-EllipsysNTUA-MaPFlowKiel-TAUF-I-OpenFOAMUoG-HMB

Test1 Rey=3đ106

Ti3 ( 0.09% )

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

AlphaC

l/Cd

ExpDTU-EllipsysNTUA-MaPFlowKiel-TAU-F-I-OpenFOAMUoG-HMB

Test3 Rey=6đ106

Ti3 ( 0.2% )

FP7-ENERGY-2013-1/ n° 608396

DNW-HDG Full CFD calculations vs measurementsEffect in Blade Design parameter: Cl/Cd

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpDTU-EllipsysNTUA-MaPFlowKiel-TAUF-I-OpenFOAM

Test5 Rey=9đ106

Ti3 ( 0.24% )

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpDTU-EllipsysNTUA-MaPFlowKiel-TAUF-I-OpenFOAM

Test7 Rey=15đ106

Ti3 ( 0.33% )

FP7-ENERGY-2013-1/ n° 608396

DNW-HDG Panel code calculations vs measurementsEffect in Blade Design parameter: Cl/Cd

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpORE-XFOILUoS-XFOIL*DTU-EllipSys

Test1 Rey=3đ106

Ti3 ( 0.09% )

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpORE-XFOILUoS-XFOIL*DTU-EllipSys

Test3 Rey=6đ106

Ti3 ( 0.2% )

FP7-ENERGY-2013-1/ n° 608396

DNW-HDG Panel code calculations vs measurementsEffect in Blade Design parameter: Cl/Cd

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpORE-XFOILUoS-XFOIL*DTU-EllipSys

Test5 Rey=9đ106

Ti3 ( 0.24% )

-5 0 5 10 150

20

40

60

80

100

120

140

160

180

Alpha

Cl/C

d

ExpORE-XFOILUoS-XFOIL*DTU-EllipSys

Test7 Rey=15đ106

Ti3 ( 0.33% )

FP7-ENERGY-2013-1/ n° 608396

Blind test, main conclusion

– cl/cd largely determines theoptimal aerodynamicdesign– importance ofboundary layer transition oncl/cdhasbeenconfirmed:– eN method performswellatall Reynoldsnumbers– Correlationbased transitionmethod (stateofthe artinmany CFDtools!)deficient athighReynoldsnumbers 1)

– Newversion ofcorrelation based transitionmodelisdevelopedwithbetter performance2)

1) Niels N. Sørensen et al Prediction of airfoil performance at high Reynolds numbers EFMC 2014, Copenhagen 17-20 Sept 2014 2) S. Colonia et al Calibration of the g equation transition model for High Reynolds flows at low Mach To be published at the Science

of Matking Torque, October 2016

FP7-ENERGY-2013-1/ n° 608396 25

List of Contents

q Introductionintotheprojectq DesignofAVATARReferenceWindTurbineq Aerodynamics athighReynoldsnumbers

• Airfoil measurements inpressurized DNW-HDGtunnelq Aero-elasticity oflargeturbinesatyaw

• BEMversusfreevortexwakeaerodynamic modelling1)

19-09-161) AcknowledgementK.Boorsma,S.Voutsinas

ECN Aero-module

26June17,2010

• ECNAeroModule:Onecodewithaero-modelsofdifferentdegreesoffidelity coupledtosamestructuralsolver(PHATAS/FOCUS)• BEM andAWSM (Freeandprescribedvortexwakemodel)• Straightforwardcomparisonofdifferentaerodynamicmodelswithsameinput• Resultsareshownofazimuthalvariationofinducedvelocityatyawed conditions:• AWSM:Physical modelling ofazimuthalvariation ofinduced velocity atyaw• 2BEMimplementations: Conventional Glauert model1)and ECNdeveloped rootvortexmodel2)and 3)

1) H.Glauert, AgeneraltheoryforAutogyroARC-R-M,1111, 1926

2) J.G.SchepersAnengineeringmodelforyawedconditions,developedonbasisofwindtunnelmeasurements,AIAA-99-003937thAerospaceSciencesMeetingandExhibit, AerospaceSciencesMeetings,1999

3) J.G.Schepers‘Engineeringmodelsinwindenergyaerodynamics,’TUDelft PhDthesisISBN:9789461915078, 2012

INNWIND.EU turbineResults at 30 degrees yaw (30% and 95% span)

27

§ Induced velocity variation at30%span:• 90degrees azimuth:

• Rootvortexeffects clearly apparentinvortexwakemethods (AWSM and AWSM-Prescribed)• Conventional Glauertmodelfor induced velocity variation deficient• Rootvortexeffects predicted wellby ECN-rv

§ Induced velocity variation at30%span:Effectofrootvortexatouter blade partinECN-rv will be reduced

FP7-ENERGY-2013-1/ n° 608396 28

Summary

• AVATAR is an EU FP7 projects which aims to validate, improve and calibrateaerodynamic models for 10MW+ turbines with and without flow devices and withand without aero-elastic effects

• Several (wind tunnel) measurements have been taken which have helped to validateand improve (sub) models relevant for 10MW+ turbines

• Models of different degrees of fidelity are evaluated on two 10 MW reference wind turbines:– AVATARlowinduction turbinewith specialaerodynamic challenges– INNWIND.EUconventional induction turbine

FP7-ENERGY-2013-1/ n° 608396 29

Summary

• The amount of results from the AVATAR project is far too much to present in 15-20 minutes• AVATAR led to several model improvements and lessons learned e.g:

– Transition modelling• Originalcorrelation based transitionmodels deficient athighReynoldsnumbers• Updated version ofcorrelation based transition models hasbecome available• Guidelineson theemploymentoftransition(andturbulence)modelsformulated

– Improved predictions for lowMachconditions incompressible codes– Improved models for thick airfoils– Improved modelsfor flowdevices (flaps, vortexgenerators,rootspoilers)– Firstpossibilities for rotormodelimprovements identified (many morewill follow…)– Improved calculational efficiency and reduced timeto set-upcalculation– Expansion/integration ofcodes(e.g.coupling ofaero-elastic models to freevortexwakemethods)– Bestpracticeguidelines ongridresolutions andtemporalresolution

• Needforvortexcorrectionsatsomegridtopologieswithlimiteddomainsize– CalibrationofengineeringmethodsfromCFDnotstraightforward

• TechniquesaredevelopedandtestedtoderiveliftinglinevariablesfromCFD– Manymoreimprovementsandlessons learnedwillfollow!

FP7-ENERGY-2013-1/ n° 608396 30

Dissemination of results

– All technical deliverablesarepublic:http://www.eera-avatar.eu/publications-results-and-links/

– Webbased validation platformisunder development(http://windbench.net/avatar)

– AVATARsessionatIRPWIND-EERAJointWindR&Dconference,September19,Amsterdam,theNetherlands– Introduction (GerardSchepers,ECN)– AdvancedAerodynamicModelling(NielsSørensen,DTU)– Flowdeviceaerodynamics(AlvaroGonzalez,CENER)– Roadmapforfurtherengineeringmodellingimprovements(GerardSchepers,ECN)

– AVATARsessionatScience ofMakingTorque,October 7,MunichGermany– 5Oralpresentations

– LatestresultsfromtheEUprojectAVATAR:Aerodynamicmodellingof10MWwindturbines(GerardSchepers(ECN) etal)– CFDcodecomparison for2Dairfoilflows (NielsSørensen (DTU)etal)– Experimentalbenchmarkandcodevalidationforairfoilsequippedwithpassivevortexgenerators (DanielBaldacchino (TUDelft)etal)– ComputingtheflowpastVortexGenerators:Comparison betweenRANSSimulationsandExperiments(M.Manolesos(NTUA) etal)– ResultsoftheAVATARprojectforthevalidationof2DaerodynamicmodelswithexperimentaldataoftheDU95W180airfoilwithunsteadyflap(CarlosFereira (TUDelft) etal)

– Introduction to 7posters

Consortium

FP7-ENERGY-2013-1/ n° 608396

Coordinator:

Partners in alphabetical order:

This project has received funding from the European Union’s Seventh Programme for research, technological development and demonstration under grand agreement No FP7-ENERGY-2013-1/n° 608396 .

Thank you for your attention