Electric Vehicle Charging Demand Forecasting Based on City Grid Attribute Classification

In order to improve the accuracy of user travel behavior and traffic road condition description in EV charging demand prediction research, a method of EV charging demand prediction based on urban grid attribute division is proposed. Firstly, the study area is determined based on the distribution of net car trips, and then the study area is precisely divided into functional areas based on the urban point-of-interest data crawled by Python, and then the spatio-temporal characteristics of residents' trips are obtained by mining; finally, considering the charging characteristics of electric vehicles, a complete charging demand prediction model is established, and the travel behavior of electric vehicles under different spatio-temporal distributions in the Second Ring Road area of Chengdu is simulated by Monte Carlo sampling method and The simulation results show that the proposed charging demand prediction method can effectively predict the charging demand in different areas and different scenarios.