Comparative Analysis of Full-Order SMO and STM-MRAS in SPMSM Sensorless Drive System

The permanent magnet synchronous motor (PMSM) drive system based on sliding mode control (SMC) strategy is widely used in sensorless control due to its good characteristics of fast dynamic response and strong anti-interference ability. However, traditional SMC can cause uncontrollable chattering problem, which greatly affects the control performance of the motor drive system. In order to suppress the chattering phenomenon, this article proposes two different improvement strategies to suppress the chattering problem. One is a full-order sliding mode observer (SMO) algorithm based on the EMF and alpha- beta axis current as the observation, which achieves the purpose of suppressing chattering by designing an observer with filter characteristics. Another approach is to combine super-twisting. SMC with model reference adaptive system to propose a super-twisting sliding mode control (STM)-model reference adaptive algorithm (MRAS) algorithm, which effectively improves the control performance of the traditional proportional-integral (PI)-MARS algorithm. Finally, a motor algorithm testing platform based on MATLAB automatic code generation was designed, and the experimental results showed that the two suppression control strategies would effectively reduce chattering, with a reduction of 69.3% in amplitude and good control performance.