Prediction of power network planning demand coefficient using eXtreme Gradient Boosting algorithm

In order to solve the problems of inaccurate prediction of demand coefficient in traditional load forecasting based on empirical value, which is easy to cause investment waste or equipment operation safety risks, a refined prediction method of demand coefficient of power network planning based on XGBoost algorithm is proposed. Firstly, the information of power application system is obtained, and the multi-dimensional data such as the capacity and load of power users are obtained from the database tables. Then the refined prediction model of demand coefficient is constructed based on XGBoost. Finally, the demand coefficient is predicted by the newly reported user information. The application results show that the proposed prediction method of demand coefficient based on XGBoost can effectively improve the prediction accuracy, and has good research and practical value.