Fault-tolerant control strategy of six-phase permanent magnet synchronous motor based on deadbeat current prediction

Abstract

The fault-tolerant control after phase loss is crucial in the studies of the six-phase permanent magnet synchronous motor (PMSM), and the one phase loss is the most frequent phase loss fault. To improve the system instability caused by nonlinear and time-varying perturbations of inductance parameters in a double-Y phase shifted 30° six-phase PMSM, an improved deadbeat predictive current fault-tolerant control (DPC-FTC) method is proposed in this study. The transformation matrix after single-phase open-phase is first reduced and reconstructed, and the reduced-dimensional voltage equation is derived. Based on this equation, the deadbeat current predictive control is then used to predict the expected voltage using the current feedback value and the reference value, so as to shorten the response time and improve the overall control effect. The voltage equation after parameter perturbation is rewritten, and the current discrete transfer function under constant expected voltage before and after parameter perturbation is calculated. Afterwards, to further improve the low stability of fault-tolerant control after phase loss, which is caused by the inductance parameter perturbation of the control system, the weight coefficient is introduced in order to enhance the deadbeat predictive current control so that it splits and optimizes the direct-quadrature axis current. The stability of the system is then analyzed. By changing the weight coefficient, the fault-tolerant control system has a wider stable working range. Finally, the simulation model and experimental platform are completed. The results show that the improved DPC-FTC method improves the permissible inductor parameter uptake range by a factor of 1/β, reduces the current static difference by 32.05% and 46.02% when the inductor parameter is mismatched by a factor of 2, reduces the current oscillation and effectively reduces the sensitivity of system stability to inductor parameter uptake.