A new combination model for short-term wind power prediction

Abstract

Short-term wind power prediction is important to the dispatch and operation of power system. A prediction model based on the rough set, principal component analysis (PCA) and Elman neural network (ElmanNN) is constructed for short-term wind speed forecasting to improve the prediction accuracy of short-term wind power. The wind speed prediction model is established by using ElmanNN, and PCA is used to extract the feature of wind speed data, which optimizes the inputs of ElmanNN. Furthermore, excitation function and the structures of network are improved to search for the optimum solution to function of convergence rate and prediction accuracy. To solve large error and prediction accuracy fluctuations of the ElmanNN model at the peak value of wind speed, the rough set theory is proposed to compensate and correct the predicted values to further improve the forecasted results. Finally, the predictive value of the wind power is obtained by the power conversion. Experiment results show that the new combination model proposed in this paper has higher prediction accuracy compared to another model and has certain application value.