Sliding Mode Fault-Tolerant Control Based on Fast Convergence Law for Faults of PMSM

Abstract

It is difficult for the traditional PI controller to meet high-performance control requirements under the demagnetization fault of a permanent magnet synchronous motor (PMSM). To address this problem, this study proposes a novel sliding mode fault-tolerant control method for PMSM demagnetization faults. First, the mathematical model of PMSM under demagnetization fault state is established, and the reasons for poor fault tolerance of the conventional PI controller are analyzed. A new convergence law is used to design the speed loop sliding mode feedback controller, and its stability is demonstrated. Meanwhile, an adaptive forgetting factor recursive least squares (AFRLS) flux linkage observer is designed to keep the controller parameter values in line with the actual motor parameter values, to reduce the impact of demagnetization faults on motor control performance and achieve fault-tolerant control of demagnetization faults. Finally, simulation and experimental comparison with conventional PI control demonstrate that the proposed method is more robust and resistant to interference.