Wind farm wind power prediction method based on CEEMDAN and DE optimized DNN neural network

Forecasting the generation of renewable energy power plants is increasingly becoming one of the basic technologies to ensure the safe and stable operation of power grids. In this paper, a new wind farm wind power prediction method based on CEEMDAN and DE optimized DNN neural network is proposed. Firstly, CEEMDAN is used to decompose the preliminary processed wind power historical data, and the LASSO method is used to eliminate the noise signal and re-fit. Then, the DE optimization algorithm is used to optimize the performance of the DNN neural network. Finally, the optimized DNN neural network is used to predict the short-term wind power of the wind farm. The CE-DE-RBF, CE-DE-BP, and CE-DE-LSSVM models were used as comparison models. Predictive experiments were performed using real data from a wind power plant in northern China. The test results fully demonstrate that the proposed model has higher prediction accuracy in terms of three performance indicators than other comparison models.