Abstract

A very simple sequential model predictive control (SMPC) is recently proposed to achieve high performance control of induction motor drives. By evaluating two separate cost functions for the torque and amplitude of the stator flux linkage in a cascaded way, the weighting factor in the conventional MPC is eliminated. However, it is shown in this paper that the conventional SMPC cannot achieve stable operation over the full speed range, if the cost function for the stator flux linkage is first evaluated. To solve this problem, this paper proposes a generalized SMPC (GSMPC), which is effective when any one of the two cost functions is first evaluated. Compared to the conventional SMPC, GSMPC not only eliminates the limitation on the execution order of two cost functions, but also presents less stator flux ripples and lower current total harmonic distortion with even lower average switching frequency. Furthermore, a simple field-weakening strategy is proposed and combined with GSMPC to widen the speed range by adjusting the torque and stator flux linkage reference online. Both simulation and experimental results are presented to confirm the effectiveness of the proposed method.