As a rapidly growing new type of load, electric vehicles (EVs) can give full play to their bidirectional response capability through orderly regulation of EVs, which in turn can assist in grid peaking. It becomes a feasible measure to alleviate the grid peaking pressure by regulating electric vehicle clusters (EVC), but at the same time, the disorderly charging and discharging of EVs will bring about the anti-peaking phenomenon and the problem of decision variable explosion. To address the above issues, firstly, with the objective of minimizing the sum of peaking cost, generation cost, wind energy curtailment cost and load fluctuation cost, the model of EVC participation in power system peaking is established to determine the charging and discharging tasks of EVC; and then, the model of individual task allocation is established to formulate the optimal charging and discharging strategies for individual EVs under the scheduling tasks; Finally, it is solved by the multi-objective aggregation peaking method. The results show that the multi-objective aggregation peaking method can ensure the economy of the grid and users under the premise of satisfying the peaking demand.