Comparative study of sensorless control of Permanent magnet synchronous motor realised by sliding‐mode observer. Issue 6 (1st May 2020)

• Record Type: Journal Article
• Authors: He, Wenyun; Wu, Xiang; Chen, Junlei; Wang, Yu
• Is Part Of: IET power electronics. Volume 13:Issue 6(2020)
• Page Start: 1191
• Page End: 1199
• Publication Date: 2020-05-01
• Subjects: permanent magnet motors -- synchronous motors -- observers -- electric potential -- position control -- variable structure systems -- sensorless machine control -- stability; permanent‐magnet synchronous motor sensorless control systems -- sliding‐mode observer -- γδ coordinates -- sliding‐mode stability -- boundary layer width -- sliding‐mode gain -- backforward method -- estimation error -- speed error -- equivalent position error -- axis estimation system -- low‐speed range -- wider‐speed range -- estimation accuracy -- axis system -- stationary αβ coordinates -- system chattering reduction -- extended electromotive force estimation extraction -- filter cut‐off frequency -- γδ‐axis estimation system -- phase‐locked loop -- sufficient condition; Power electronics -- Periodicals; 621.31705
• DOI: 10.1049/iet-pel.2019.1153
• Languages: English
• ISSNs: 1755-4535

Abstract

Abstract : In this study, the permanent‐magnet synchronous motor sensorless control systems realised by sliding‐mode observers in stationary αβ and estimated γδ coordinates are systematically studied and compared. First, based on the sufficient condition of sliding‐mode stability, a more reasonable selection method of boundary layer width and sliding‐mode gain is derived, effectively reducing the system chattering. Furthermore, in extracting estimated extended electromotive force, the advantages and disadvantages of the backforward method in the αβ ‐axis and the influence of filter cut‐off frequency on estimation error in the γδ ‐axis are analysed. Finally, the estimation error, speed range, stability, and the effect of speed error of these two sensorless control systems are analysed and compared. The results indicate that these two systems have equivalent position error above a certain speed, and the γδ ‐axis estimation system has a larger position error in low‐speed range, but still operating well. Moreover, the γδ ‐axis estimation algorithm has better stability and wider‐speed range. Owing to the fast track of phase‐locked loop, the effect of speed error on estimation accuracy of the γδ ‐axis system can be neglected. Simulation and experiment verified the theoretical analysis and comparison.