To improve the accuracy of short-term wind power prediction

a short-term wind power integrated prediction model based on error correction is proposed. A wind power prediction model is established by using the XGBoost model based on improved particle swarm optimization firstly. Based on the relationship between wind speed and power

the prediction error of the XGBoost model is divided into low wind speed power error

middle-speed power error

and high wind speed power error. The random forest is trained under each type of error

then the corresponding power error prediction model is obtained. The short-term wind power prediction results based on error correction can be obtained by the addition of XGBoost model prediction results and power error prediction results. The research results show that the proposed model improves the short-term wind power prediction accuracy through integrated learning and residual learning methods

therefore

the proposed model can promote the wind power consumption capacity and improve the economy of the power system.