Iterative Fast Super-Twisting Flux Sliding Mode Observer for SPMSM with Tangent Quadrature Phase-Locked Loop

Traditional low-order flux sliding mode observer (FSMO) and quadrature phase-locked loop (QPLL) structures generally encounter issues such as estimated signal chattering and inadequate dynamic performance. To overcome these challenges, this paper proposes an iterative fast super-twisting flux sliding mode observer (IFST-FSMO) and a tangent quadrature phase-locked loop (TQPLL) for sensorless control of surface-mounted permanent magnet synchronous motors (SPMSMs). Building on the traditional super-twisting algorithm (STA), the IFST-FSMO is proposed to accelerate convergence and enhance chattering suppression, which incorporates a linear term and utilizes the hyperbolic tangent function to replace the intrinsic sign function. Notably, the feedback matrix is redesigned to ensure the algorithm’s stability during speed reversal. Furthermore, an iterative calculation strategy is implemented under low-speed and light-load conditions, improving steady-state accuracy of estimated flux while avoiding increased computational burden at medium and high speeds. Regarding position estimation, a novel TQPLL with correction factor is proposed, utilizing the tangent function of the electrical angle error to achieve normalization and bandwidth adaptation. Ultimately, the proposed method is implemented on a motor test platform. Comparative experimental results demonstrate that the IFST-FSMO combined with TQPLL exhibits superior dynamic response and steady-state accuracy, while achieving efficient speed reversal.