Bus Load Prediction Based on Subtraction Clustering and XGboost

Bus load in the power system has the characteristics of large number and high load fluctuation. The traditional bus load forecasting method fails to fully consider the influence of weather, week and holidays, etc. which have already cannot fit the demand of prediction accuracy. In order to solve the above problems, this paper proposes a bus load prediction method based on subtraction clustering (SCM algorithm) and extreme gradient enhancement (XGBoost). Firstly, subtractive clustering is used to cluster historical similar daily loads. Based on the clustering results, a decision tree is established to obtain the type of bus load on the day to be measured, and the daily load of bus on the day to be measured is predicted. Experimental results show that the proposed method has better prediction accuracy.