Temperature compensation strategy of output torque for permanent magnet synchronous motor based on BP neural network

Due to the advantages of high power factor and high efficacy, the permanent magnet synchronous motor (PMSM) is widely applied to the traction motor of electrical vehicles. However, in the operation of PMSM, the temperature will change with an influence on the output torque through the fluctuation of flux and resistance, etc. This leads the PMSM system to the fluctuation of output torque. Based on the analysis of the temperature influence on PMSM, this paper builds a Simulink electrical motor model with temperature as a variable, and constructs the BP neural network compensator. In this process, the neural network is optimized by the Mind Evolution Algorithm (MEA), compensating the impact of temperature on the output torque. According to the simulation result, it is found that the BP neural network temperature compensator proposed in this paper has an outstanding compensation effect of smooth output torque. Therefore the proposed strategy of temperature compensation here is quite efficient.