A novel model predictive flux control for permanent magnet synchronous motor considering torque constraint

Abstract

In order to enhance the torque response speed of permanent magnet synchronous motor (PMSM) while minimizing torque pulsation, a novel model predictive flux control (NMPFC) method is proposed in this article. First, the bisection method is employed to determine the optimal flux linkage trajectory through operating point control (OPC) and reference prerotation (RPR). In the dynamic process, the output flux linkage gradually converges to its optimal trajectory, which decreases the torque dynamic response time and improves the torque dynamic performance. Second, within the principle of continuous control set model predictive flux control (CCS-MPFC), a cost function incorporating both L1 and L2 norms is constructed, and the cost function incorporates both torque reference tracking and flux linkage reference tracking terms, which can effectively suppress torque pulsation, simplify the selection of weight factor, and improve the torque steady-state performance. Finally, a closed-loop flux observer is designed based on the PMSM current and voltage equations to estimate the flux linkage, so that the estimated flux linkage converges rapidly approaching the actual value, providing accurate flux linkage parameters for RPR and CCS-MPFC. Compared with other methods, this article further reduces torque pulsation and accelerates torque convergence by directly introducing the L1 norm of the torque error into the cost function. The advantages of the L1 norm over the L2 norm are discussed, and a reference range for the weight factor is provided. Experimental results show that this method improves the torque dynamic performance and steady-state performance of the system.