Abstract: A multi-objective optimization electric vehicle charging model is proposed to address the issues of grid load stability and user charging costs caused by disorderly charging of electric vehicles. Considering the different load pressures of the power grid and the urgency of user charging needs at different time periods in public and residential areas, a multi-objective optimization model for electric vehicles is proposed to reduce user charging costs and peak valley load differences while reducing grid load fluctuations. The improved multi-objective particle swarm optimization algorithm is adopted for solving, and the weight values are adjusted by optimizing the learning factor and introducing dynamic inertia. Levy flight disturbance particle swarm optimization improves the multi-objective particle swarm optimization algorithm. Combining the time-of-use electricity price for example analysis, the results show that the charging model based on the improved multi-objective particle swarm optimization algorithm has fast convergence speed, can jump out of local optima, better multi-objective optimization, and achieve the goal of reducing the peak valley difference of power grid load and charging costs.