Study on permanent magnet synchronous motor control with improved BP neural network

Abstract In this paper, an improved back-propagation (BP) neural network magnetic field-oriented control (FOC) control strategy is investigated and applied to the speed control of a permanent magnet synchronous motor (PMSM). To address the problem that the traditional PI-based FOC algorithm is difficult to adapt the parameters in the face of complex environmental changes, a method to optimize the FOC using the Bacterial Foraging Optimization Algorithm (BFOA) to optimize the PID parameters of the BP neural network is proposed. The proposed scheme is simulated in Matlab’s Simulink, and compared to the traditional PI- and BFOA_PID-based FOC, the improved BFOA_BP_PID-based FOC algorithm shows faster response speed, smaller speed error, lower torque fluctuation under no-load, a step change in speed, and a sudden change in load conditions, showing good stability and robustness.