Title: OPTIMIZING ENSEMBLE PREDICTION SYSTEMS FOR WIND POWER FORECASTS
1OPTIMIZING ENSEMBLE PREDICTION SYSTEMS FOR WIND
POWER FORECASTS
European Centre for Medium-Range Weather
Forecasts (ECMWF) Michael Denhard, Renate
Hagedorn ForWind - University of Oldenburg Jens
Tambke , Thomas Petroliagis, Miriam Wolff, Jinhua
Jiang
2ECMWF
is an independent international organization,
supported by 18 member states
12 co-operating states
Co-operating agreements
Czech Republic
Croatia
Estonia
Hungary
Iceland
Lithuania
Montenegro
Morocco
Romania
Serbia
Slovenia
Slovakia
3ForWind, the Center for Wind Energy Research of
the Universities of Oldenburg and Hannover,
combines scientific know-how with research geared
towards the industry. ForWind bundles the
competencies of both universities and is an adept
industry contact. ForWind conducts basic
research in the wind energy sector, cooperating
in industrial projects from an independent
scientific perspective.
4Numerical Weather forecasts
Numerical model to describe the processes
in the earth system
5Deterministic Forecasting
Is this forecast correct?
Temperature
Forecast time
Initial condition
Forecast
6The Lorenz Attractor
one flap of a sea-gulls wing may forever
change the future course of the weather
(Lorenz, 1963)
7Ensemble Forecasting
Temperature
Forecast time
Initial condition
Forecast
Complete description of weather prediction in
terms of a Probability Density Function (PDF)
8Flow dependence of forecast errors
26th June 1995
26th June 1994
9EPS appearing on TV
Netherlands
Germany
Courtesy of Robert Mureau, KNMI
10Safe Wind
Multi-scale data assimilation, advanced wind
modelling and forecasting with emphasis to
extreme weather situations for a safe large-scale
wind power integration.
EU-FP7 Project
11WP-5 Optimized Ensemble System
September 2008
August 2012
12Numerical forecast systems in Europe
TIGGE (10 global EPS) Thorpex Interactive Grand
Global Ensemble
global high resolution models
13WP-5 high resolution deterministic system
Sea Level Pressure und 10 m Winds 00 UTC, 12
December 2005 Anlysis and forecasts (a) T799L91
and (b) T511L60
ECMWF global high resolution deterministic model
14WP-5 Combined System
Weight of high resolution determinitic run
15WP5 Online ensemble
Update weights of ensemble forecasts with respect
to recent observations (every 3 hours)
16WP-5 Kombiniertes Vorhersagesystem
- Wind and temperature forecasts on model levels
- Weather pattern related Extreme Forecast Index
(EFI) - Gust factor
- Convert wind and temperature PDFs to windpower
PDF
17SafeWind WP5 Summary
- 6 Tasks, 11 Deliverables, 5 (direct) Partners
Task Description Partners Inp Out
5.1 Probabilistic verification tool (wind gust verif.) ECMWF (7), UNIOL (3) WP2 T6.5 T6.5
5.2 Evaluation of novel ensemble techniques ECMWF (12), UNIOL (4), MF(5) T5.1 T5.5
5.3 Use of Local Area Model EPS UNIOL (8), ECMWF (1) T5.5
5.4 Weather regime dep. Extreme Forecast Index UNIOL (4), ECMWF (1), MF(1) T4.5 T5.5
5.5 Combined meteorological Prediction Systems ECMWF (12), UNIOL (11) T5.1- T5.4 T5.5
5.6 CPS applied to wind power forecasts UNIOL (4), ARMINES (3), ECMWF (1), ENERGINET (1) T5.5 T1.1 T1.4 WP9