Considering the large amount of calculation and difficult control instruction acceptance for electric vehicle(EV)consumers by using traditional centralized control strategies

a vehicle-to-grid(V2G)management framework of hierarchical architecture is adopted based on distributed control

which is applicable to large-scale EV optimization scheduling. In the framework

EV agents submit their charging schemes to local operator agents(LAs)that is in the control of distribution networks center agents(DNAs). Through building the charging and discharging load model of a single EV agent

the V2G optimization model of EV agent for maximizing its profit is built. LA aims at minimizing the peak-valley difference and DNA is committed to ensure the safety of distribution networks. In the case of IEEE 33-bus system with 3 LAs

the behaviors and strategies of each layer are simulated to verify the effectiveness of proposed hierarchical architecture. This work is supported by National Natural Science Foundation of China(No. 51577115)and National Key Technologies R&D Program(No.2013BAA01B04).