Improved super-twisting-observer-based finite-control-set model-predictive fault-tolerant current control of PMSM considering demagnetization fault

Temperature and other factors will cause the permanent-magnet flux to demagnetize. The demagnetization fault inevitably results in the prediction error since the finite-control-set model-predictive current control (FCS-MPCC) method is sensitive to the parameter variation of permanent magnet synchronous motor (PMSM). This paper presents a improved super-twisting-observer-based FCS model-predictive fault-tolerant current control (ISTO-based FCS-MPFTCC) method to compensate for the prediction error caused by the demagnetization fault. The unified PMSM model of healthy and demagnetized is first derived. Next, a improved super-twisting-observer (ISTO) is designed to detect permanent-magnet flux. The estimated permanent-magnet flux updated the prediction model. Thus, the presented method performs fault-tolerant current-control of PMSM on the demagnetization fault. It eliminates the influence caused by the demagnetization fault. The experiment results validate the effectiveness of the presented ISTO-based FCS-FCS-MPFTCC method. •An improved super-twisting-observer-based FCS model-predictive fault-tolerant current control (ISTO-based FCS-MPFTCC) method is presented.•An improved super-twisting-observer (ISTO) is designed to estimate permanent-magnet flux.•The presented ISTO-based FCS-MPFTCC method performs demagnetization fault-tolerant control for a PMSM drive system.•The method eliminates the parameter mismatch caused by the demagnetization fault.