PMSM Control Research Based on Particle Swarm Optimization BP Neural Network

Abstract

This paper presents a control method of the combination of particle swarm optimization algorithm (PSO) and BP neural network for the control of PMSM. PSO can easily and quickly find a optimal parameters of PI, which can be used to generate the study sample space of BP neural network .The BP neural network can be off-line learning, then the network after learning apply to PI controller to control PMSM. This method combines the advantages of PSO and BP neural network learning ability. Compare with the traditional PI control, this method shows a better control performance, can quickly learn ideal PI parameters for motor control.