A combined model based on secondary decomposition technique and grey wolf optimizer for short-term wind power forecasting

Short-term wind power forecasting plays an important role in generation systems. In order to improve the accuracy of forecasting, many researchers have proposed a large number models. However, traditional models ignore data preprocessing and limitations single model, resulting low accuracy. Aiming at shortcomings existing models, combined model based on secondary decomposition technique grey wolf optimizer (GWO) is proposed. process firstly, complete ensemble empirical mode adaptive noise (CEEMDAN) wavelet transform (WT) are used preprocess data. Then, least squares support vector machine (LSSVM), extreme learning (ELM) back propagation neural network (BPNN) established forecast decomposed components respectively. performance, parameters LSSVM, ELM, BPNN tuned by GWO. Finally, GWO determine weight coefficient each weighted combination obtain final result. The simulation results show that has better performance than other