This paper presents a real-time optimized dispatching model for large-scale cluster electric vehicles(EVs)to achieve the charging demand and safe operation of distribution network. For each new optimization scenario

the accessed EVs can cluster according to their desired completion time. The entire dispatching process for charging/discharging strategy can be divided into two steps

i. e.

the upper dispatching based on grey wolf optimization(GWO)algorithm for each cluster

and the bottom layer based on energy buffer consensus for each EV in the corresponding cluster. Simulation results demonstrate that the proposed model can significantly facilitate the real-time large-scale optimal dispatching of EVs

while GWO algorithm and energy buffer consensus are suitable to solve large-scale optimal dispatching with superior performance on availability and convergence rate. This work is supported by National Basic Research Program of China(973 Program)(No. 2013CB228205)and National Natural Science Foundation of China(No. 51477055).