Prediction of Wind Power Generation based on Chaotic Phase Space Reconstruction Models

Abstract

The development of wind generation has rapidly progressed over the last decade, but it must be integrated into power grids and electric utility systems. However, it cannot be dispatched like conventional generators because the power generated by the wind changes rapidly because of the continuous fluctuation of wind speed and direction. So it is very important to predict the wind power generation. This paper discusses why the wind power generation can be predicted in short-term, and how to setup the construction of an ANN (artificial neural network) prediction model of wind power based on chaotic time series. The analysis of modeling with low dimensions nonlinear dynamics indicates that time series of wind power generation have chaotic characteristics, and wind power can be predicted in short-term. Phase space reconstruction method can be used for ANN model design. The data from the wind farm located in the Saihanba China are used for this study.