Wind Power forecasting using fuzzy neural networks enhanced with on-line prediction risk assessment

Pierre Pinson (1) , Georges Kariniotakis (1)

1 CEP - Centre Énergétique et Procédés

The paper presents an advanced wind forecasting system that uses on-line SCAnA measurements, as well as numerical weather predictions (NWP) as input, to predict the power production of wind park8 48 hours ahead. The prediction system integrates models based on adaptive fuzzy-neural networks configured either for short-term (1-10 hours) or longterm (1-48 hours) forecasting. The paper presents detailed oneyear evaluation results ofthe models on the case study oflreland, where the output of several wind farms is predicted using HIRLAM meteorological forecasts as input A method for the online estimation of confidence intervals of the forecasts is developed together with an appropriate index for assessing online the risk due to the inaccuracy of the numerical weather predictions.

Mots clés: Wind power; short-term forecasting; numerical weather predictions; on-line software; adaptive fuzzy-neural networks; confidence intervals; prediction risk

Domaines: Sciences de l'ingénieur [physics]; Energie électrique

Pierre Pinson, Georges Kariniotakis. Wind Power forecasting using fuzzy neural networks enhanced with on-line prediction risk assessment. 2003 IEEE Bologna Powzer Tech conference, Jun 2003, Bologna, Italy. 8 p. - ISBN 0-7803-7968-3, ⟨10.1109/PTC.2003.1304289⟩. ⟨hal-00530561⟩