Wind power plant prediction by using neural networks

This paper introduces a method of short term wind power prediction for a wind power plant by training neural networks based on historical data of wind speed and wind direction. There are two steps in the process of wind power prediction. In the first step, raw data collected by plant information system is filtered by probabilistic neural network. This step prepares valid data to be used for building a prediction model. In the second step, a complex-valued recurrent neural network is applied to build a model to predict wind power. The test results of the prediction model are presented and analyzed at the end of the paper. The model proposed is shown to achieve a high accuracy with respect to the measured data.