Study On Energy Management Strategy Of Extended Range Electric Vehicle Based On Particle Swarm Optimization Algorithm

The Structure of Extended Range Electric Vehicle (EREV) power system is similar to that of traditional Series Hybrid Electric Vehicle. The battery of EREV is main power source, and its capacity is larger so that it can meet the drivers’ commuter time requirement in a city. Range Extender (RE), included engine and generator, is auxiliary power system to ensure drivers’long-distance travel. Compared with Electric Vehicle, EREV can also achieve a certain all-electric range. Due to engine, EREV can solve the problem of expensive battery cost and eliminate drivers’ anxiety for short range. In one word, EREV is a good model to solve resources shortage and environmental pollution.The thesis takes Energy Management Strategy (EMS) of EREV as the object, puts forward a logic threshold EMS. After EMS is optimized, simulation is conducted. Finally, an EMS based on Artificial Neural Network (ANN) is proposed. The main contents are as follows:Firstly, based on the structure characteristic of EREV, the selection and parameter matching of key components, including motor, battery and RE, are finished.Secondly, after analyzing the status of current EMS and control strategy of traditional hybrid electric vehicles, the logic threshold EMS is presented. Particle Swarm Optimization (PSO) algorithm is used to optimize thresholds and engine fixed working points to improve EMS.Thirdly, after establishment of the vehicle model, Energy Management Strategy is tested by off-line simulation.Fourthly, the vehicle real-time simulation platform is established based on D2P (from Development to Production) rapid prototype development. After harness connection, ECU controller and pedal operating mechanism are utilized to test the real-time performance and the control effect of EMS.Fifthly, taking the randomness of actual driving cycle into account and learning form optimization experiences, an Energy Management Strategy based on ANN is built by the training of respective data so that it can avoid specified driving cycle restriction and adapt to all kinds of driving condition to achieve better control.The research of thesis has proved EMS is reasonable and it has certain theoretical significance and practical value. In the meanwhile, the study of the paper can provide a theoretical basis and guidance for the deeper research and EMS design.