Design of Torque Observer Based on Improved Elman Neural Network for Permanent Magnet Synchronous Motor

Abstract

PMSM generally has problems such as complex algorithm, difficult identification of motor parameters, and difficult accurate estimation of electromagnetic torque by mathematical model, which leads to the decrease of motor control precision and overall performance of drive system. In direct torque control, torque inaccuracy will affect motor performance. Therefore, for PMSM system, it is very important to improve the accuracy of torque estimation. In this paper, an improved Elman neural network electromagnetic torque observer is designed. The simulation results are verified and compared with the traditional Elman neural network. The results show that compared with the traditional Elman neural network estimation method, the improved torque observer has high precision torque output performance, higher control precision and accuracy.