Research on multiobjective optimization method of permanent magnet synchronous motor based on PSO-BP neural network

In order to improve the drive performance of permanent magnet synchronous motors, a hierarchical multi-objective optimization method based on particle swarm optimization algorithm (PSO) and BP neural network combined with non-dominated sorting genetic algorithm (NSGA-II) is proposed in this paper. The results show that the particle swarm optimization-based BP neural network model significantly improves the prediction accuracy of the output parameters of the permanent magnet synchronous motor, with the coefficients of determination of 0.98605 (cogging torque amplitude), 0.99782 (average torque), and 0.9948 (torque pulsation), which are 2.54%, 3.31%, and 1.61% higher than those of the conventional BP neural network, respectively. In addition, the hierarchical optimization strategy is effective, in which the average torque is improved by 4.14%, the cogging torque is reduced by 84.93%, and the torque pulsation is reduced by 58.81%. The effectiveness and superiority of the proposed method in enhancing the performance of permanent magnet synchronous motors are verified.