A New Electric Power Forecasting Model for Wind Farms

I.J. Ramírez-Rosado and L.A. Fernández-Jiménez (Spain)

Keywords

Short-term forecasting, wind electric energy prediction,neural networks, subtractive clustering.

Abstract

The integration of wind energy conversion systems in the electric power networks has become an important problem in their daily operation. The intermittent nature of wind makes difficult to guarantee the electric power generated in wind farms for the next hours. In this paper a new model based on clustering and data regression is used for the short-term forecast of the electric power generated in a wind farm. Only past values of electric power production are used. The results obtained, with real life data, from this model are better than those obtained from a relevant set of forecasting models using the same input data.

Important Links:



Go Back