With the large-scale charging of increasing electric vehicles

which is of great randomness and autonomy

the far-reaching impacts on power grids can hardly be handled using centralized control. In order to solve this problem

the customer satisfaction degree is modeled as an index of vehicle user charging preferences

based on which a coordinated pricing strategy for charging stations is proposed in the multi-agent framework to implement the price-based decentralized control of electric vehicles. Furthermore

the proposed strategy is modeled on a Java multi-agent development framework (JADE) platform

which is capable of realizing the behavior and communication among electric vehicles

charging stations and smart grid agents. In the end

the grid-friendly charging behavior of electric vehicles using the proposed strategy is proved by simulation results

while the impact of vehicle user preferences and grid constraints is analyzed in detail. This work is supported by National Natural Science Foundation of China (No. 51577115) and National Key Technologies R&D Program (No. 2013BAA1B04).