The State of the Art in Short-term Prediction of Wind Power...DTU Wind Energy, Technical University of Denmark. Time and space scale of atmospheric motion. Long waves. Mid latitudes

The State of the Art in Short-term Prediction of Wind Power

Dr. Gregor GiebelDTU Wind EnergyAnd numerous co-authors, not all knowingly…

SANEDI, Sandton, South Africa, 19 Sept 2017

DTU Wind Energy, Technical University of Denmark

Outline

•Why predictions? and for whom?

•Predictions: The general information flow

•Some actual implementations

•Some knowledge from research

•Best Practice


State-of-the-Art for Wind Power


State-of-the-Art in Short-term Prediction


State-of-the-Art in Short-term Prediction

111 pages>380 references>700 citationsDOI: 10.11581/DTU:00000017

DTU Wind Energy, Technical University of Denmark54 18 September

2017

Sou

rce:

Win

dEur

ope

Dai

lyW

ind

new

slet

ter,

val

id for

13

Sep

t20

17.


Record in DK: 140% wind power!Low demand and high wind production in a summer night.Here given for all of Denmark.

Source: www.emd.dk/el/

59 18 September

2017


Smoothing


All Europe is connected

The

imag

e is

of 1

997,

but

it’s

not m

uch

chan

ged

since

then

However, local or cross-border transmission can be a bottle-neck


Data from 60 meteorological stations

•One year of three-hourly wind speed data from (mostly) 10 m

•Also many inland sites

•Then spatially averaged over all time series at every time step

•=> Total generation profile is less variable!


Cross-correlation versus distance

• Calculate cross-correlation

coefficient between every

pair of stations

• Result: Cross-correlation

decreases with distance

• Exponential fit has shape

parameter of ca. 700 km

0 1000 2000 3000 4000 5000-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

Cor

rela

tion

coef

ficie

nt (l

ag=0

hou

rs)

Distance [km]


Users


Users of forecasts

Who needs forecasts:

• Transmission companies in areas with high wind penetration (egEnerginet.dk, Tennet, 50Hertz, Red Electrica de España, CaISO, AEMO, ESKOM…)

• Wind power owners/operators with own market access (Iberdrola, DONG, Vattenfall, RWE Innogy, Vindkraft, DTEK, …)

• Electrical utilities (eg DONG Energy, Vattenfall, Acciona, Iberdrola, E.On, NUON, RWE, EnBW, ESKOM…)

• Everyone trading on markets with sizeable shares of wind power


Thomas Ackermann about TSOs

Source: 1st Workshop on Short-term Forecasting, Uni NSW, Sydney (AUS), December 2005


Level of users 2016

AU.SA

Slide source: Waldl: Operational wind & solar power forecasting - The Perfect Wind Power Prediction. Talk on the 1st Workshop on Large-scale Grid Integration of Renewables in India, Dehli, 6-8 Sept 2017


Timescales for wind forecasts

Source: The Future of Wind. White paper on Wind Power Monthly Expert reports, MetOffice, 2013

Min/sec

http://www.wpmexpertreports.com/Whitepaper/The-Future-of-Wind


Cost functionsDifferent users have different cost functionsEven within their organisationsMight not even be obvious for themUses:• trading wind power• definition of reserve requirements• unit commitment and economic dispatch• operation of combined wind-hydro• operation of wind associated with storage• design of optimal trading strategies• electricity market design• etc.


Strategic bidding vs TSO responsibilityREE showed first that aggregating the forecasts coming from the marketparticipants was worse than their in-house state-of-the-art tool

• Due to better information available at the TSO• Forecasts had variable quality (usually coming from the lowest bidder,

not necessarily the best)• Potential for strategic bidding of the market participants:

– E.g. allowing for a +- 20% corridor means that bidding the fullvolume increases the risk of error – safe bid is 80% max.

Requires online SCADA data at TSO, including active power, curtailment, maintenance information, …Often mandated in grid codes.

86 18 September

2017


PredictionsHowTo


Short-Term Prediction Overview

Numerical Weather Prediction Prediction model

Online data

Orography

Roughness

Wind farm layout

End user

GRID

TRADING

Image sources: DWD, WAsP, Joensen/Nielsen/Madsen EWEC’97, Pittsburgh Post-Gazette, Red Electrica de España.


Short-Term Prediction Overview

End user

Image sources: DWD, WAsP, Joensen/Nielsen/Madsen EWEC’97, Pittsburgh Post-Gazette, Red Electrica de España.

GRID

TRADING


Online data

Orography

Roughness

Wind farm layout


End user

Image sources: DWD, WAsP, Joensen/Nielsen/Madsen EWEC’97, Red Electrica de España.


Stakeholders


Statistical power curve estimation• Establish best connection between NWP wind speed and measured power• Often non-parametric and not a function• Often recursively adapted with new online data


Data with turbine availability and curtailment

Slide source: Waldl: Operational wind & solar power forecasting - The Perfect Wind Power Prediction. Talk on the 1st Workshop on Large-scale Grid Integration of Renewables in India, Dehli, 6-8 Sept 2017

15

20

25

30

35

40

45

50

4:30

5:30

6:30

7:30

8:30

9:30

10:30

11:30

12:30

13:30

14:30

15:30

16:30

17:30

18:30

19:30

20:30

21:30

Unconstrained wind farm production

Limited turbine

availability

Grid curtailment

Power [%]

Time [hours]


Performance

DTU Wind Energy, Technical University of DenmarkSource: Advanced Short Range Wind Energy Forecasting Technologies-- Challenges, Solutions, and ValidationKristin Larson and Tillman Gneiting, Global WINDPOWER 2004, March 31, 2004


Time and space scale of atmospheric motion

Long waves

Mid latitudes Hs & Lsfronts

TyphonesTropical Storms

days to a week or more

Land-sea breeze

Mountain-valley breeze

Thunder-storms tornadoes water-spouts

small turbulent eddies

hours to days

mins to hours

secs to mins

micro-scale

2 m

meso-scale 20 km

synoptic-scale 2000 km

global-scale 2000 km

Typical life span

Typical

sizes

Source: Jesper Nissen


Synoptic scale meteorology

(http://www.metoffice.gov.uk/weather/uk/surface_pressure.html#view)

High pressure system

Low pressure system

Cold front

Warm front



Mesoscale Meteorology

H

HL

L

SEA LAND

Sea breeze circulation

Coastal low level jets

Thunderstorm

Picture from http://www.news2.dk/pdf/20060224X008.pdf



There are always unresolved processes that cannot be represenby a numerical model.

These features are approximated through Parametrization!

Micro scales

Turbulent strees seen on the sea surface scales

1-500 m

Dust devils scale 25-50 m



NWP: DMI-HIRLAM


Level vs. Phase errors

• There are two error categories from the NWP: level errors and phase errors

– Level error means not to predict the intensity of a storm

– Phase errors are misjudging the timing of the storm

– Phase errors are more frequent these days (but how to report them?)

• Ca 80% of the error comes from the NWP!


Phase and Level errors• Errors can be phase (timing) or level errors

0

100

200

300

400

500

600

700

800

900

1000

09/0

1 00

:00

09/0

1 04

:48

09/0

1 09

:36

09/0

1 14

:24

09/0

1 19

:12

10/0

1 00

:00

10/0

1 04

:48

10/0

1 09

:36

10/0

1 14

:24

10/0

1 19

:12

11/0

1 00

:00

Time

MW

Actual Production Short Term Forecast


Common evaluation criteria

Since none were available, the ANEMOS project codified common criteria for performance measurements of short-term forecasting systems:

• Mean Error• Mean Absolute Error• Root Mean Square Error• R2 (coefficient of determination)• Histogram of errors

• Also, use separate training and validation datasets• Present the errors normalised with the installed capacity

• Madsen, H., P. Pinson, G. Kariniotakis, H.Aa. Nielsen, T.S. Nielsen: Standardizing the Performance Evaluation of Short-term Wind Power Prediction Models. Wind Engineering 29(6), pp. 475-489, 2005

http://www.ingentaconnect.com/content/mscp/wind/2005/00000029/00000006/art00002


Typical results (1996 – now more like 10%)

0 3 6 9 12 15 18 21 24 27 30 33 36-5

0

5

10

15

20

25

30

35

40

-251.5

0.0

251.5

503.0

754.5

1006.0

1257.5

1509.0

1760.5

2012.0

Peak capacity=5030 kWNøjsomhedsodde wind farm

Persistence HWP MeanRM

S Er

ror [

% o

f nam

epla

te c

apac

ity]

Forecast length [h]

RM

S Error [kW]



0 3 6 9 12 15 18 21 24 27 30 33 36

0

5

10

15

20

25

30

35

40

0.0

251.5

503.0

754.5

1006.0

1257.5

1509.0

1760.5

2012.0


Persistence HWP NewRef MeanRM

S Er

ror [

% o

f nam

epla

te c

apac

ity]

Forecast length [h]

RM

S Error [kW]



0 3 6 9 12 15 18 21 24 27 30 33 36

0

5

10

15

20

25

30

35

40

0.0

251.5

503.0

754.5

1006.0

1257.5

1509.0

1760.5

2012.0


Persistence HWP NewRef HWP / MOSRM

S Er

ror [

% o

f nam

epla

te c

apac

ity]

Forecast length [h]

RM

S Error [kW]

Typical forecast accuracy

of a modern system


Wind Speed Dependency• Wind speed errors from the NWP are quite similar across the whole range

of wind speeds• Folded through the power curve gives large errors for rising parts, small

errors for flat part

Source: M. Lange, D. Heinemann: Accuracy of Short Term Wind Power Predictions Depending on Meteorological Conditions. Proceedings of the Global Wind Power Conference, Paris 2002Central plot from Lange, M., and U. Focken: Physical Approach to Short-Term Wind Power Prediction. Berlin: Springer-Verlag, 2005

http://www.springer.com/engineering/power+engineering/book/978-3-540-25662-5


Forecast accuracy, historical (eg ISET)

• Forecasting got better during the last years

• Some of it piggybacks on improvements in meteorology

• Some is due to better interface to meteorological models (e.g., using 100m wind speed)

• Some is using multi-model approach

Graph shows error in E.Oncontrol zone over the years, with references from the paper

B. Lange et.al.: Wind Power Prediction in Germany - Recent advances and future challenges. Paper on the EWEC 2006 in Athens.


Smoothing of forecast errors

• Focken et al looked into the spatial smoothing of forecast errors – left is actual, right is derived model

• Therefore, predictions for a region always are better than predictions for a single wind farm

• Source: Lange, M., and U. Focken: Physical Approach to Short-Term Wind Power Prediction. Berlin: Springer-Verlag, 2005

http://www.springer.com/engineering/power+engineering/book/978-3-540-25662-5


History


Ed McCarthy 1985-87

• Predicted for the large wind farms in California (Altamont, San Gorgognio etc)

• Was run in the summers of 1985-87• On a HP 41CX programmable calculator

• Using meteorological observations and local upper air observations• The program was built around a climatological study of the site and

had a forecast horizon of 24 hours. • It forecast daily average wind speeds with better skill than either

persistence or climatology alone.


Prediktor

• Applied in Eastern Denmark between 1993 and 1999

• Similar: Previento


Previento

ω*

ω

Power

Prediction

Similar to Prediktor, but uses more stringent physical downscaling (incl stability) and specialised upscaling

Operational at EWE, E.On, RWE, Vattenfall, EnBWUniversity of Oldenburg / energy & meteo systems GmbH


Wind Power Prediction Tool• Developed at IMM/DTU• Operational in Western DK

1994• Operational for all of DK

1999• Statistical non-parametric

adaptive models for prediction of representative farms

• Upscaling statistically to installed capacity

• Employs data cleaning

• Similar: Sipreólico, WPMS, MORE-CARE


Fraunhofer IWES WPMS

• Wind Power Management System = Nowcasting + Forecasting

• In use at E.On Netz since 2001, RWE since 6/2003, Vattenfall Europe 2004

• E.On case: 50 representative wind farms (soon more) from WMEP -> ANN upscaling = Nowcast

• DWD Lokalmodell and others provide for forecast

• Accuracy: after 7 hours purely NWP dominated (5% RMS for E.On Netz total area)


New for WE: 100m winds

•100m wind forecasts and analysis (from ECMWF) publically available from August 2010...

•... Featured in the Weather Eye column of The Times (19th of August 2010)

Evaluation of 100m deterministic and ensemble forecastsFocus on sites with available observations (e.g. Fino)Potential benefit from getting more model-level wind forecasts


Evolution of the state of the art

1990

Deterministic approachesHybrid approaches

Probabilistic approaches1st BenchmarkingTowards standardisationCombined forecasts (multi model/NWPs)

2002 Anemos 2013

176

Next generation of toolsFocus on extremesDiversify predicted informationSpatio-tempRamp forecastingCut-off forecastingAlarming, risk indicesLink to meteorology etc

SafeWind2008

ANEMOS.plusdemos

End-users :• TSOs• DSOs• Island system

operators• Utilities• Traders

Functions considered:

Power system

scheduling

Reservesestimation

Congestion management

Wind/storage

coordination

Probabilisticforecasting

Optimal trading

38

ANEMOS./plus

Link forecasts to the applicationPower system management/tradingStochastic optimisationDemonstration


Research results


Doubling the number of NWP• Used DMI and DWD for six test cases in Denmark• Result: the combination of inputs is better

SyltholmFjaldene

MiddelgrundenKlim

Hagesholm

Tunø Knob

G. Giebel, A. Boone: A Comparison of DMI-Hirlam and DWD-Lokalmodell for Short-Term Forecasting. Poster on the EWEC, London, Nov 2004


Benefit of multiple NWPs• Combining two NWPs improves results• Alaiz data, Hirlam and MM5 (CENER) as NWP

H.Aa. Nielsen: Slides on project meeting, PSO project Intelligent Prognosis Systems, January 2006 at Risø


Spatio-temporal improvement of forecasts

Work done by DTU-IMM (now DTU-Compute) and ENFOR in the SafeWind project.

J. Tastu, P. Pinson, E. Kotwa, H.Aa. Nielsen, H. Madsen (2011). Spatio-temporal analysis and modeling of wind power forecast errors. Wind Energy 14(1), pp. 43-60


Anemos


Overview• 3 EU sponsored projects (20 M€ total, 15 M€ from EU)• Started 2002• Most important European institutes, many important end users, some

meteorological providers• Common shell for models developed… • … and installed at utilities / TSOs• Comparison of existing models• Advancing the state-of-the-art in statistical, physical and offshore

predictions

• Also commercial installations: AEMO, SONI

• Many slides in this section come from Georges Kariniotakis, Anemos project leader at Ecole des Mines / Armines.


The Consortium of Anemos

IASA


• Accuracy!• Uncertainty!

• 1990ies to 2005

Meteorology

Wind power forecasting technology

Operational decision making

Evolution

Graphics: George Kariniotakis


• Feedback to meteorology, dedicated wind powermeteorological forecasts

• 2000-

Meteorology



Evolution



Meteorology



• Specialised forecast products for power system issues:• Trading• Ramps / Variability• Medium-range forecasts• Congestion / Storage management• …

Evolution



Meteorology



Evolution


• Specialised forecast products for power system issues:• Trading• Ramps / Variability• Medium-range forecasts• Congestion / Storage management• …


Meteorology



Evolution


• Increased interaction between wind powerforecasting and meteorological community


The forecasting system

Generic architecture

An operational prediction platform is developed covering a wide range of end-user requirements: single wind farm forecasting up to regional/national forecasting.


Portfolio of models• Short-term (0-6 hours) statistical • Medium-term (0-48/72 hours) statistical & physical from the leading

model providers (choose some or all)• Combined approaches• Regional / National prediction • On-line uncertainty estimation • Probabilistic forecasts• Risk assessment• Numerical Weather Predictions by alternative models as input


Operational experience

Real-time operation & demonstration in 9 countries by 10 end-users.

France (utility)Germany (utility)Greece (utility)Ireland (TSO)Spain (TSO, developer)Portugal (TSO)UK (Northern Ireland TSO)

Denmark (utility)

Australia (TSO, commercial installation)


IEA Wind Task 36


IEA Wind Task 36 ForecastingInternational Energy Agency (IEA) has several Energy Technology Networks, e.g. Wind Power.Forecasting Task (36) runs 1/2016-12/2018 (and probably new phase afterthat). Some 200 people from weather services, operational forecasters, academiaand end users (TSOs, operators, traders, …).

Participation is open for IEA Wind member countries, but news aredistributed via a mailing list (send mail to [email protected]).

See details at www.ieawindforecasting.dk.

266 18 September

2017

mailto:[email protected]


Technical Results

Mainly: published 5 lists, useful for peers• Tall masts for NWP verification, and how to

access their data• Field experiments in wind power meteorology• Openly available benchmarks for power

forecasts• Research projects in the field• Future research issues


Use of probabilistic forecastingOpen Access journal paper48 pages on the use of uncertaintyforecasts in the power industry

Definition – Methods –Communication of Uncertainty – End User Cases – Pitfalls -Recommendations

Source: http://www.mdpi.com/1996-1073/10/9/1402/

272 18 September

2017

http://www.mdpi.com/1996-1073/10/9/1402/


Best Practice


Best Practice

• Get a model (24/7)• Get another model (NWP and / or short-term forecasting model)• Use online power data for first hours and power curve calibration• Work together with service provider / academia / weather

service to continuously improve model accuracy• Reduce error by predicting for a larger area (smoothing)• Balance all errors together, not just wind• Use the uncertainty / pdf

• Do forecasting on TSO level, not necessarily on wind farm / developer level

• Use intraday trading

• Use longer forecasts for maintenance planning• Meteorological training for the operators• Meteorological hotline for special cases

• Also in report on powwow.risoe.dk (Giebel and Kariniotakis: Best Practice in Short-term Forecasting. A User’s Guide. Project report for the POW’WOW project, 6 pages, 2009)

http://powwow.risoe.dk/publ/GGiebelGKariniotakis-STPBestPractice_POW'WOW.pdf

The State of the Art in Short-term Prediction of Wind PowerOutlineState-of-the-Art for Wind PowerState-of-the-Art in Short-term PredictionState-of-the-Art in Short-term PredictionSlide Number 54Record in DK: 140% wind power!SmoothingAll Europe is connectedData from 60 meteorological stationsCross-correlation versus distanceUsersUsers of forecastsThomas Ackermann about TSOsLevel of users 2016Timescales for wind forecastsCost functionsStrategic bidding vs TSO responsibilityPredictions�HowToShort-Term Prediction OverviewShort-Term Prediction OverviewStakeholdersStatistical power curve estimationData with turbine availability and curtailmentPerformanceSlide Number 110Slide Number 111Synoptic scale meteorologyMesoscale MeteorologySlide Number 114NWP: DMI-HIRLAMLevel vs. Phase errorsPhase and Level errorsCommon evaluation criteriaTypical results (1996 – now more like 10%)Typical results (1996 – now more like 10%)Typical results (1996 – now more like 10%)Wind Speed DependencyForecast accuracy, historical (eg ISET)Smoothing of forecast errorsHistoryEd McCarthy 1985-87Prediktor PrevientoWind Power Prediction ToolFraunhofer IWES WPMSNew for WE: 100m windsEvolution of the state of the artResearch resultsDoubling the number of NWPBenefit of multiple NWPsSpatio-temporal improvement of forecastsAnemosOverviewThe Consortium of AnemosEvolutionEvolutionEvolutionEvolutionEvolutionEvolutionSlide Number 258Portfolio of modelsSlide Number 264IEA Wind Task 36IEA Wind Task 36 ForecastingSlide Number 270Use of probabilistic forecastingBest PracticeBest Practice

Documents

The State of the Art in Short-term Prediction of Wind Power...DTU Wind Energy, Technical University of Denmark. Time and space scale of atmospheric motion. Long waves. Mid latitudes