Abstract: An electric vehicle (EV) charging load prediction model was established to predict the time distribution characteristics of EV charging load by Monte Carlo simulation, a source-load matching optimization strategy was proposed considering the parameters of charging piles and charging/swapping modes, and a comprehensive evaluation system for wind and solar energy storage system was established. Using improved genetic algorithms for planning and solving to obtain the optimal capacity configuration. Taking a service area in Shandong Province as an example, the results show that the EV charging load prediction curve presents a "bimodal" pattern during the day. The number of charging piles and the maximum charging power in the service area are reasonably set, and the charging/swapping mode with battery swapping participation is selected to make the load energy consumption curve close to the photovoltaic power generation curve. Compared with the initial scheme, the improved genetic algorithm can improve the closeness of the optimal scheme, and can realize the charging of energy storage system and the connection of residual power.