Estimated position correction algorithm of surface-mounted permanent-magnet synchronous motor based on variable gain steepest gradient descent method

Abstract

This paper addresses the position-estimation deviation issue of the sensorless drive method and a new algorithm for estimating position correction of surface-mounted permanent magnet synchronous motor (SPMSM) is proposed. To improve the estimation accuracy of sensorless methods, the estimated position correction algorithm based on variable gain steepest gradient descent position error observer is proposed in this paper. The position error is estimated by the flux observation, and the steepest gradient descent method is used to correct the integration process. In addition, the dynamic and accuracy performance of the position error observer are improved by variable gain cycle iterative optimization. Experimental evaluation validated the effectiveness of the proposed algorithm, and the results show that the proposed method has the characteristics of high position observation accuracy and strong robustness.