EERA Design Tool for Offshore

wind farm Cluster (DTOC) PETER HAUGE MADSEN. Director Charlotte Hasager. Senior scientist

DTU Wind Energy

Support by

Verification of wind farm wake

ClusterDesign & EERA DTOC workshop

Amsterdam, 24 September 2014

Contributions from

WP leader

P-E. Réthoré

SCADA data analysis and wake model intercomparison

K.S. Hansen

Wake modelling

G. Sieros, J. Prospathopoulos, J.M.L.M. Palma, V.C. Gomes, G. Schepers, P. Stuart, T. Young, J.S. Rodrigo, G.C. Larsen, T.J. Larsen, S. Ott, O. Rathmann, A. Pena, M. Gaumond, R.J. Barthelmie, S.C. Pryor, P. Volker, E. Cantero, J. Badger, A. Palomares

Satellite data analysis

A.Mouche, R.Husson, M. Badger, A. Dibella

Wind farm wake comparison

• Horns Rev 1 SCADA data comparing nine wake models

• Lillgrund SCADA data comparing ten wake models

• Rødsand-2 SCADA data comparing several wake models

• Alpha ventus SCADA data comparing long-range scanning

wind lidar, ship-based wind lidar wind, FINO-1 met mast

several wake models

• Satellite SAR from more than 10 large offfshore wind

farms comparison qualitative to wake models

EERA DTOC portfolio of models

Name Partner Status Programs Input/

output

Script/

GUI

Database

interface IPR Com

CFDWake CENER Fluent, C++,

OpenFOAM ASCII script Yes

CorWind Risoe DTU Ope DOS exe

Delphi CSV files no no + +

CRES-farm CRES Ope Linux/ Fortran77 ASCII no no +

CRES--flowNS CRES Ope Linux/ Fortran77 ASCII no no

DWM Risoe DTU Ope Fortran, pc, pc-

cluster ASCII script +

ECNS ECN Beta Linux/ Fortran90 ASCII No No +

EeFarm ECN Alpha Matlab Matlab

scripts

Script/

GUI yes + +

Farm-farm

interaction ECN Ope Fortran ASCII No no +

FarmFlow ECN Ope Delphi ASCII/

binary GUI Yes + +

FlowARSM CRES Alpha Linux/ Fortran77 ASCII no no

FUGA Risoe DTU Ope Fortran, C, Delphi,

pc ASCII Script/ GUI No +

NET-OP SINTEF Proto type Matlab ASCII script No +

Skiron/WAM CENER Ope Unix/ Fortran GRIB script yes

TOPFARM Risoe DTU Beta Matlab/C/ Fortran ASCII script +

UAEP Risoe DTU Matlab, pc ASCII/ binary no yes

VENTOS UPorto Beta Unix/ Fortran ASCII no yes + +

WAsP Risoe DTU Ope Windows pc ASCII Script/ GUI No + +

WCMS Fraunhofer Ope Matlab/JAVA OracleDB yes +

WRF Risoe DTU Ope Unix, Linux,

Fortran90 netCDF Shell script yes

WRF/ROMS CIEMAT Ope Linux/ Fortran netCDF script yes + 4

EERA DTOC wake models

Name Partner Status Programs Input/

output

Script/

GUI

Database

interface IPR Com

CFDWake CENER Fluent, C++,

OpenFOAM ASCII script Yes

CorWind Risoe DTU Ope DOS exe

Delphi CSV files no no + +

CRES-farm CRES Ope Linux/ Fortran77 ASCII no no +

CRES--flowNS CRES Ope Linux/ Fortran77 ASCII no no

DWM Risoe DTU Ope Fortran, pc, pc-

cluster ASCII script +

ECNS ECN Beta Linux/ Fortran90 ASCII No No +

EeFarm ECN Alpha Matlab Matlab

scripts

Script/

GUI yes + +

Farm-farm

interaction ECN Ope Fortran ASCII No no +

FarmFlow ECN Ope Delphi ASCII/

binary GUI Yes + +

FlowARSM CRES Alpha Linux/ Fortran77 ASCII no no

FUGA Risoe DTU Ope Fortran, C, Delphi,

pc ASCII Script/ GUI No +

NET-OP SINTEF Proto type Matlab ASCII script No +

Skiron/WAM CENER Ope Unix/ Fortran GRIB script yes

TOPFARM Risoe DTU Beta Matlab/C/ Fortran ASCII script +

UAEP Risoe DTU Matlab, pc ASCII/ binary no yes

VENTOS UPorto Beta Unix/ Fortran ASCII no yes + +

WAsP Risoe DTU Ope Windows pc ASCII Script/ GUI No + +

WCMS Fraunhofer Ope Matlab/JAVA OracleDB yes +

WRF Risoe DTU Ope Unix, Linux,

Fortran90 netCDF Shell script yes

WRF/ROMS CIEMAT Ope Linux/ Fortran netCDF script yes + 5

Horns Rev 1

We acknowledge DONG energy and

Vattenfall AB for granting access to the

SCADA data from the Horns Rev 1 offshore

wind farm.

Wind farm wake comparison

Horns Rev 1 - conclusion

Benchmarking of wind farm scale wake models in the EERA-DTOC Project, P.-E. Réthoré, K.S. Hansen,

R.J. Barthelmie, S.C. Pryor, G. Sieros, J. Prospathopoulos, J.M.L.M. Palma, V.C. Gomes, G. Schepers, P.

Stuart, T. Young, J.S. Rodrigo, G.C. Larsen, T.J. Larsen, S. Ott, O. Rathmann, A. Pena, M. Gaumond and

C. B. Hasager, International Conference on Aerodynamics of Offshore Wind Energy Systems and Wakes

(ICOWES 2013), 17-19 June 2013, Copenhagen, Denmark.

EERA DTOC publications

EERA DTOC publications (continued)

Lillgrund

We acknowledge Vattenfall AB for granting access to

the SCADA data from the Lillgrund offshore wind farm.

Lillgrund available measurements

• 65 m mast (wind speed, turbulence, wind direction, air temperature),

before wind farm installation with high quality, period: 2003 – 2006

• 65 m mast (wind speed, turbulence, wind direction, air temperature)

with medium quality, period: 2008 – 2010

• SCADA data as 10 minute statistics (mean values and stdev from each

wind turbine). Signals: power, pitch, rpm, nacelle wind speed and

position. Period 2008 – 2012

1. SCADA is the processed wind farm data to be compared with the wind farm

wake models;

2. FUGA is a linearized actuator disc eddy-viscosity CFD model for offshore

wind farm wake developed by DTU;

3. CRESflowNS is an elliptic k-ε actuator disc CFD model tailored for offshore

wake simulation developed by CRES;

4. FarmFlow is a parabolized k-ε actuator disc CFD model tailored for offshore

wake simulation developed by ECN;

5. GCL is the G.C. Larsen eddy-viscosity wake model v2009 developed by DTU;

6. NOJ is the original N.O Jensen model;

7. AD/Ainslie is an eddy-viscosity wake model developed by RES-LTD.

Wind Farm wake models used in the benchmark

63 simulation results have been provided from the

10 participants.

Benchmark matrix

Park

Row:3-120deg Row:B-222deg Row:5-120deg Row:D-222deg TI-3.3D TI-4.3D Efficiency

DTU FUGA 1 1 1 1 1 1 1

CRES CRESflowNS 1 1 1 1

ECN FarmFlow 1 1 1 1 1 1 1

DTU GCJ-BinAve 1 1 1 1 1 1 1

DTU GCJ-GauUnc 1 1 1 1 1 1 1

DTU NOJ-BinAve 1 1 1 1 1

DTU NOJ-GauUnc 1 1 1 1 1

DTU NOJ(Penã) 1 1 1 1 1 1 1

RES-LTD AD/Ainslie 1 1 1 1 1 1 1

CENER GCJ-GauUnc 1 1 1 1 1 1 1

sum 10 10 10 10 7 7 9

Missing turbine(s)Complete rows Turbulence

Institution/model

EERA-DTOC

Flow case complete rows

Power deficit along a row of turbines - 3.3D & 4.3 D spacing

at 9 m/s;

Flow case complete rows 3.3 D spacing

Power deficit along a row of turbines - 3.3D & 4.3 D spacing –

with ”missing” wind turbines at 9 m/s;

Flow case missing turbines

Flow case 3.3 D spacing with missing turbines

Decreased deficit -

due to speed recovery

1. Maximum power deficit – as function of turbulence

intensity for a pair of turbines with 3.3D spacing

Flow case turbulence dependence

3.3D spacing

Flow case turbulence dependence

4.3D spacing

Park efficiency for 0 – 360º inflow at 9 m/s & Δ=3º.

Inflow conditions:

-Wind direction (derived)

-Wind speed (derived)

Flow case 360 degrees

Flow case 360 degrees

Flow case 360 degrees

• Good agreement between wake model results and

measurements

• All models were able to predict the increased deficit

between closely spaced turbines

• The speed recovery was well reproduced

• Linear relation between deficit and turbulence was well

reproduced

• Park power deficit for 0 - 360º inflow was well reproduced

within 4-5% at 9 m/s

Conclusion from Lillgrund

Rødsand-2

We acknowledge E. On for granting access to the SCADA

data from the Rødsand-2 offshore wind farm.

Courtesy: E.On

Rødsand 2

SCADA data from all turbines

10 minute statistical data from meteorological mast

(No data from neighbouring Nysted farm)

Rødsand 2 preliminary results

Rødsand 2 preliminary results

Rødsand 2 preliminary results

Rødsand 2 preliminary results

Rødsand 2 preliminary results

Rødsand 2 preliminary results

Conclusion Rødsand-2 preliminary!

• Very interesting case with bending rows of turbines

• Is modelled both by microscale and mesoscale

• Cluster effect from neighbouring wind farm is clear

• Intercomparison of several wake models is in progress

SAR satellite images

Satellite case using Radarsat-2 SAR

WRF wake modelling

WRF wake modelling

WRF wake modelling

WRF wake modelling

Conclusion on satellite wake

• 10 m above sea level

• Varying wind direction

• No SCADA data

• Work is not yet completed

Alpha Ventus

We acknowledge RAVE for granting access to the

SCADA data from the Alpha Ventus offshore wind farm.

Courtesy: Alpha Ventus

42 24-9-2014

Lidar measurements (ForWind & Fraunhofer IWES)

• Long range wind scanner measurements from fixed

positions

• Ship based LIDAR measurements

• FINO-1 met mast

Conclusion on Alpha Ventus

• Work is on-going

• No preliminary conclusion yet

Support by

EERA DTOC project

FP7-ENERGY-2011-1/ n°282797