The random charging of electric vehicles(EVs)will cause many adverse effects on the security and economical operation of a power system. A multi-objective optimization model is proposed for real-time charging control aimed at maximization of the EV owners' benefit and minimization of the active power loss. By using the alternate direction method of multiplier(ADMM)

the centralized optimization model of charging can be converted into individual sub-problems in the decentralized optimization model with the device as the unit. For each iteration of ADMM

only a bit of information is exchanged between the device and the adjacent interactive information points

which is conducive to protecting user information security. Meanwhile

some disadvantages due to centralized optimization can be overcome

such as high communication requirements and high computational overhead. Simulations for IEEE 33-bus and actual 119-bus network systems show that the results of the decentralized optimization model and the centralized model are conformable. Moreover

the proposed algorithm shows high computing efficiency

low communication cost and good applicability to the rolling scheduling schema in real-time. This work is supported by National Basic Research Program of China(973 Program)(No. 2013CB228205)and National Natural Science Foundation of China(No. 51367004).