Short-Term Forecasting of Wind Power Generation Based on the Similar Day and Elman Neural Network

Wind power forecasting is significant to reduce the impact of wind power generation integration on the power grid. According to the characteristics of power generation of wind power system and the factors impacting wind power output, a selecting method of the similar days is proposed. By the historical data similar to the features of forecasted day are selected and considered as the training sets. Elman Neural Network is used to calculate wind power output. The method is validated by wind power system data, and the forecast error is calculated and analyzed. The results show the method has high accuracy, which provides reference to short-term forecasting of wind power generation.