Abstract: To alleviate the problems of high operating cost of PV-integrated EV charging station and large load fluctuation on the grid side,an optimization algorithm combining NSGA-Ⅲ （nondominated sorting genetic algorithm Ⅲ）and fuzzy clustering is proposed for optimization operation of energy storage. Firstly,on the basis of analyzing the structure of the charging station,a multi-objective operation model of the energy storage is established to minimize the load variance of grid side,the operation and maintenance cost of the energy storage,and the purchase cost from the grid. Then,NSGA-Ⅲ algorithm is applied to solve multi-objective model. For the Pareto sets containing a lot of information,it is difficult for the operators to make a decision. A method based on fuzzy clustering is proposed to select the optimal solution from Pareto sets.Finally,extensive experimental analyses demonstrate the efficiency of the proposed method,and show that,compared with particle swarm optimization,the economy of the system and the grid-side load level are improved on the basis of the load demand,so that the overall performance of the charging station is optimal comprehensively.