Research on prediction model of electric vehicle thermal management system based on particle swarm optimization- Back propagation neural network

The variation of temperature in various components of the Electric Vehicle Thermal Management (EVTM) system is the main factor affecting energy consumption. A good temperature prediction method is a prerequisite for optimizing energy strategies. The temperature variation of each component in the thermal management system is influenced by various factors, such as ambient temperature, vehicle speed, and air conditioning compressor speed. Therefore, this paper proposes a temperature prediction method based on Particle Swarm Optimization (PSO) and Back propagation Neural Network (BP). The PSO-BP method is used to update the weights and thresholds of the neural network. Real-time data collected from road experiments are used to ensure the accuracy of the influencing factors. Simulation results show that compared to the BP neural network, the proposed PSO-BP prediction method reduces the prediction errors by 66%, 75%, and 25% respectively.