Due to the rapid development of electric vehicles,there is much potential for distributed charging piles to participate in power grid management. Typical evaluation indicators of load operation are compared, load fluctuation rate is selected as the objective of optimization, multiple target optimization model for load optimization with distributed charging piles is established, and the key characteristics of load fluctuation rate are extracted to establish the simple target function describing load fluctuation rate. The inflexibility of electricity pricing is solved by introduction of incentive points, and optimization target function of end user cost is built. Typical summer load conditions of certain distribution system feeder line located in Nanjing is taken for instance. MATLAB is used to simulate the load optimization featuring peak shifting and valley filling participated by 2 000 EVs and 500 EVs. It turns out that the fluctuation rate of grid load reduces obviously and the end users save cost greatly.