Abstract: In order to improve the stability and security of wind power integration, the accuracy of wind power prediction is of great significance. The method of wind power forecasting based on the wavelet packet and echo state network is proposed, and the specific application steps are given. Firstly, the historical data of meteorological factors such as wind speed, wind direction, temperature, humidity and wind power is decomposed by wavelet packet decomposition, which can accurately reflect the regularity of historical data. Secondly, the echo state network is applied to model and predict each decomposed signal, which improves the speed and accuracy of modeling. Finally, the prediction results of each decomposed signal are synthesized to obtain the predicted value of wind power.