Abstract:To improve the prediction accuracy of short-term power load after a large number of distributed energy resources such as photovoltaic and wind power are connected to the power grid, and to promote the improvement of the absorption capacity of the power grid, this paper adopts the wavelet-radial basis function(RBF) neural network for photovoltaic output and short-term power load, and uses BPNN, RBFNN, WNN and ELMANNN neural networks to predict the power data after the overall average empirical mode decomposition(EEMD) are used for wind power, and uses particle swarm algorithm and grey correlation degree to correct the wind power prediction data. Finally, the equivalent load concept is used to analyze the influence of photovoltaic and wind power integration on load forecasting, and the three models are effectively combined to obtain the equivalent load model of short-term load prediction of power system considering photovoltaic and wind power generation. Case analysis shows that compared with other methods, the proposed method has higher prediction accuracy in short-term load prediction of power system considering photovoltaic and wind power integration.