General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from orbit.dtu.dk on: Dec 20, 2017

On the impact of the assimilation of nacelle winds and yaw angles with WRF-FDDA andWRF-DART for short-term wind energy predictions

Draxl, Caroline; delle Monache, Luca; Liu, Yubao; Cheng, William; Descombes, Gael; Vandenberghe,Francois; Romine, Glen; Knievel, Jason; Hahmann, Andrea N.; Giebel, Gregor

Publication date:2011

Document VersionPublisher's PDF, also known as Version of record

Link back to DTU Orbit

Citation (APA):Draxl, C., delle Monache, L., Liu, Y., Cheng, W., Descombes, G., Vandenberghe, F., ... Giebel, G. (2011). Onthe impact of the assimilation of nacelle winds and yaw angles with WRF-FDDA and WRF-DART for short-termwind energy predictions [Sound/Visual production (digital)]. 14th Conference on Mesoscale Processes/15thConference on Aviation, Range, and Aerospace Meteorology, Los Angeles (US), 1-4 Aug, 01/01/2011,http://ams.confex.com/ams/14Meso15ARAM/flvgateway.cgi/id/18490?recordingid=18490

Risø DTU NCAR

...a bit about everything...

Caroline Draxl

Caroline Draxl

Luca delle Monache, Yubao Liu, William Cheng, Gael Descombes, Luca delle Monache, Yubao Liu, William Cheng, Gael Descombes, FranFranççois Vandenberghe, Glen Romine, Jason Knievel (NCAR)ois Vandenberghe, Glen Romine, Jason Knievel (NCAR)Andrea N. Hahmann, Gregor Giebel (Risø DTU)Andrea N. Hahmann, Gregor Giebel (Risø DTU)

Assimilation of nacelle winds and yaw angles for Short-Term Wind Energy Predictions

Risø DTU National Laboratory for Sustainable Energy

Forecast errors are expensive

Trading of wind power on energy markets

Fluctuating wind and power is challenging for grid operators

Optimisation of power plant mix

Maintainance planning

Enhanced value and acceptance of wind power

Wind turbine measurements become increasingly available and constitute a new set of measurements in the PBL

How to treat them in data assimilation?

What is their benefit in the system??

Nacelle winds and yaw angles

Nacelle winds and yaw angles areused by the turbine control system for optimal turbine operation.

Nacelle winds:

Sonic anemometers measure wind on Horns Rev turbines @ 70 m.

Yaw Angles for wind directions(u and v in data assimilation):

The nacelle turns itself into the wind.

Issues with nacelle winds

Disturbance depends on

Design of turbine and nacelle

Pitch/stall regulation

Height of anemometer and position on nacelle

Operation/standstill

Position in wind farm: wake

Nacelle transfer function

Restricted data access

Wind Farm layout and model setup

30 km

10 km3.3 km

Horns Rev Wind Farm Measurement mast

Denmark10 km

Assumption of spatially independentobs errors in data assimilation violated

=> data thinningWRF FDDA V3.2.1

Find optimal case studies

→ 4 days in 4 different months in 2005:

54 hours with a 6h-forecast every hour 54 forecasts→

low pressure over northern Europe, pre or post cold front

WIND70 m tower280°

260°wake

6 km

Data thinning

Statistics compared for different turbine groups

Medians of

all turbines

WIND70 m tower280°

260°wake

Data thinning

Medians of

all turbines

upwind half farm

WIND70 m tower280°

260°wake

Data thinning

Medians of

all turbines

upwind half

first 2 upwind turbine rows

WIND70 m tower280°

260°wake

Data thinning

Medians of

all turbines

upwind half

first 2 upwind turbine rows

first upwind turbine row

Direction measurements from turbine yaws

WIND70 m tower280°

260°wake

Data thinning

Medians of

all turbines

upwind half

first 2 upwind turbine rows

first upwind turbine row

Direction measurements from turbine yaws

and from 68 m tower to the north of the farm

WIND70 m tower280°

260°wake

Data thinning

Medians of

all turbines

upwind half

first 2 upwind turbine rows

first upwind turbine row

Direction measurements from turbine yaws

and from 68 m tower to the north of the farm

+ METAR, SYNOP, ACARS, Radiosondes, Ships

WIND280°

260°70 m towerwake

1h 2h 3h 4h 5h 6h Lead time

BIA

S (

model

– o

bs)

m/s

0.7

5 0.8

0.8

5 0.9

0

.95 1

1.0

5 1.1

1

.15

Bias reduced in first 2 hours, noisy

no FDDA

Bias reduced in first 2 hours, noisy

1h 2h 3h 4h 5h 6h Lead time

BIA

S (

model

– o

bs)

m/s

0.7

5 0.8

0.8

5 0.9

0

.95 1

1

.05 1.1

1

.15

All + MADIShalf + MADIS

no FDDAMADIS only

Bias reduced in first 2 hours, noisy

1h 2h 3h 4h 5h 6h Lead time

BIA

S (

model

– o

bs)

m/s

0.7

5 0.8

0.8

5 0.9

0

.95

1 1

.05 1.1

1

.15

All + MADIShalf + MADISrow 1 + MADISRows 1&2 + MADISRows 1&2 + tower + MADISno FDDAMADIS only

Bias reduced in first 2 hours, noisy

1h 2h 3h 4h 5h 6h Lead time

1h 2h 3h 4h 5h 6h Lead time

BIA

S (

model

– o

bs)

m/s

0.7

5 0.8

0.8

5 0.9

0

.95

1 1

.05 1.1

1

.15

R

MSE m

/s1.4

1

.5 1.6

1.7

1

. 8 1

.9 2

2.1

2.2

2.3

All + MADIShalf + MADISrow 1 + MADISRows 1&2 + MADISRows 1&2 + tower + MADISno FDDAMADIS only

Bias reduced in first 2 hours, noisy

1h 2h 3h 4h 5h 6h Lead time

1h 2h 3h 4h 5h 6h Lead time

BIA

S (

model

– o

bs)

m/s

0.7

5 0.8

0.8

5 0.9

0

.95

1 1

.05 1.1

1

.15

R

MSE m

/s1.4

1

.5 1.6

1.7

1

. 8 1

.9 2

2.1

2.2

2.3

All + MADIShalf + MADISrow 1 + MADISRows 1&2 + MADISRows 1&2 + tower + MADISno FDDAMADIS only

Outliers Weight too high ?

WRF DART and plans

DART: Community Ensemble Kalman filter system maintained by NCAR

EnKF: flow-dependent background error covariances.

Observation error needs to be assigned to measurements – knowledgeof measurements is important.Observation errors are used by EnKF as a weight for the information provided by the measurements.

Source: Tom Hamill, http://www.esrl.noaa.gov/psd/people/tom.hamill

A new growing & promising data set

CAREFULwith nacelle winds !!!

Thinning strategies

FDDA: 2 hour benefit

Improve forecasts inland

WRF DARTMore promising ?

?

Assimilation in the PBL

A new growing & promising data set

THANK YOU !cadr@risoe.dtu.dk

CAREFULwith nacelle winds !!!

Thinning strategies

FDDA: 2 hour benefit

Improveforecasts inland

WRF DARTMore promising ?

?

Assimilation in the PBL