Abstract: In order to improve the fuel economy of extended range electric vehicle (EREV), an energy management strategy for EREV based on Multi-Agent System (MAS) networking technology is proposed. The EREV vehicle dynamics model was built in Simulink environment, and the Java Development Framework (JADE) platform was used to construct each power component agent to realize the networking. The thermostats energy management strategy was improved by MAS networking technology. The multi-agent networking system of power components was connected with Simulink vehicle dynamics model by MACSimJX component, and the simulation verification was carried out under WLTC condition. The simulation results show that compared with the dynamic programming control strategy, the fuel saving rate of the vehicle is 3.3%, the power consumption is 1.7% higher, and the total energy consumption rate is 1.4% lower. Compared with the traditional electric auxiliary control strategy, the fuel saving rate of the vehicle is 18.2%, the power consumption is reduced by 12.4%, and the total energy consumption rate is reduced by 16.6%. MAS networking technology can effectively improve the energy control effect of the vehicle, and provide theoretical support and a new research method for hybrid electric vehicle energy management strategy.