Abstract
Aimed at the problems of system chattering and large observation errors in the sensorless control of a permanent magnet synchronous motor (PMSM) based on a traditional sliding mode observer (SMO), a combined reaching law algorithm based on the exponential reaching law and arcsine saturation function reaching law is proposed to improve the sliding mode observer. An appropriate positive real number is taken and compared with the product of controller gain and stator current observation error to judge the system position in sliding mode motion. In the early stage of sliding mode motion, the exponential reaching law is utilized, and then, in the latter and stable stages of sliding mode motion, the arcsine saturation function reaching law is used. The stability of the observer is proved by Lyapunov theory. The simulation and experimental data show that the speed error of the sliding mode observer based on the combined reaching law is reduced by 80% compared with the traditional sliding mode observer, and the chattering problem is also improved.