Advanced Tools for Wind Power Integration into Electrical Networks

R. Blonbou, S. Monjoly, and R. Calif (France)


Wind power prediction; Bayesian neural networks; Classification; Dirichlet mixture.


In the short term, wind energy is the most promising endogenous sources of energy for the French overseas territories. However, to anticipate for wind power variability is crucial for island power system management. Their small size and the fact that they are not connected to large utility network impose supplementary constraints to wind power integration. Wind power fluctuations limit the penetration rate of Grid connected wind energy. This paper describes two approaches developed for the sake of wind energy forecasting and wind power fluctuations characterization respectively. The first approach directly addresses the wind energy forecasting problem; we present an adaptive wind energy prediction scheme that uses an artificial neural network as predictor. The second approach aims to identify and classify the different meteorological regime encountered in order to anticipate for the rapid wind variations.

Important Links:

Go Back