Sensorless Control of PMSM Based on an Improved Super-Twisting Sliding Mode Observer

This paper presents an enhanced fast super-twisting algorithm sliding mode observer for sensorless control of permanent magnet synchronous motors. The proposed method improves upon conventional super-twisting algorithms by incorporating a linear correction term, which simultaneously boosts convergence speed and reduces chattering effects. These enhancements lead to superior dynamic response and steady-state estimation accuracy. Stability is guaranteed through rigorous Lyapunov analysis, with explicit convergence conditions mathematically derived. For practical implementation, rotor position and speed are accurately extracted from the estimated extended back-EMF using an optimized signal processing chain combining low-pass filtering and quadrature phase-locked loop techniques. Comprehensive simulation studies validate the proposed observer's performance advantages over traditional super-twisting approaches, demonstrating significant improvements in both position and speed estimation accuracy. The results confirm the method's effectiveness and practical viability for high-performance sensorless PMSM drives.