Sensorless Control of Surface-Mount Permanent-Magnet Synchronous Motors Based on an Adaptive Super-Twisting Sliding Mode Observer

Abstract

The Sliding Mode Observer (SMO) is widely used for the sensorless control of Permanent-Magnet Synchronous Motors (PMSMs) due to its simple structure and strong parameter robustness. However, traditional SMOs have a limited speed range and suffer from chattering issues, which affect the accuracy of rotor position estimation. To address these problems, this paper proposes an Adaptive Super-Twisting SMO (AST-SMO) method. First, a fast super-twisting function is designed to resolve the step problem that occurs at the zero-crossing of the traditional sign function. Next, an adaptive-tracking high-order Sliding Mode Observer is constructed to extend the speed range of the SMO. The stability of the system is proven using the Lyapunov theorem. Finally, a sensorless control system for PMSMs is implemented and validated in MATLAB/SIMULINK. The results indicate that, compared to the traditional SMO, the AST-SMO reduces the back EMF THD from 20.03% to 14.2%. Additionally, the rotor estimation error across all speed ranges is less than 0.01. Therefore, AST-SMO offers a higher tracking accuracy, a wider speed range, and effectively suppresses sliding mode chattering and harmonic interference.