Load Prediction of Charging Station Based on Regression Tree of Cart Model

Load forecasting is an important service for the reliable operation and effective management of the smart grid. To achieve this goal, we refine the load characteristics of the electric vehicle charging station according to the actual charging power and weather forecast data of electric vehicle charging stations, and propose a load forecasting method of charging stations based on classification and regression tree, which can enhance the scalability and robustness compared with classical regression models. We evaluate our model with several simulation examples from real data. Experimental results indicate that the proposed model can obtain higher performances than existing models for load forecasting.