An intelligent method for wind power forecasting based on integrated power slope events prediction and wind speed forecasting

In this paper, we study an intelligent wind power prediction method by taking the prediction time horizons and prediction accuracy into account. The wind power slope events are defined, and multiple support vector machines are applied to the classification of slope down/up events for multistep-ahead scenarios. The wind speed series are decomposed by using the maximum overlap discrete wavelet transform (MODWT), and each decomposed signal is forecast using an adaptive wavelet neural network (AWNN) individually. The network is trained for wind speed prediction up to 24h ahead. Based on slope events forecasting and wind speed forecasting, an improved radial basis function neural network (RBFNN) is proposed to predict wind power up to 24h ahead. The proposed model is tested by using wind power data collected from a real wind farm. The analysis results validate that both the prediction time horizons and the prediction accuracy are guaranteed, and the proposed method can be applied to the optimal scheduling of wind farms 1day in advance. (c) 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.