Prediction Method Based on Random Forest Regression for Sympathetic Intush Peak

To solve the problem of line protection action caused by sympathetic inrush, this paper proposes to predict its peak before closing, avoid closing switch under dangerous conditions and try to apply random forest regression to the prediction of the sympathetic inrush peak. This article first comprehensively analyzes influencing factors of sympathetic inrush, then input influencing factors are screened according to characteristic importance calculated by random forest, and obtain the optimal prediction model through the grid search and cross-validation. Finally, a substation under construction in the Lingang district is taken as an example for testing, and different regression models are applied for comparative tests. The experimental results show that the method proposed in this paper has better applicability and can accurately and rapidly predict the peak of sympathetic inrush.