Model predictive torque control of PMSM for EV drives: A comparative study of finite control set and predictive dead-beat control schemes

Abstract

In this paper, two schemes of a model predictive control (MPC), namely finite control set with modulation control (FCS-MPC) and dead-beat model predictive control (DB-MPC) as a continuous control set (CCS-MPC) are applied to permanent magnet synchronous motor (PMSM) for torque control. The FCS-MPC strategy chooses the optimal output voltage vector with its optimal modulation interval. In CCS-MPC strategy, a current/torque control structure that utilizes the advantage of the dead-beat controller and combines it with a reference current prediction is presented. Then, we compare the performance of the two approaches. The comparison is mainly handled from two sides: (1) ripples reduction of torque and current waves which is assessed at steady-state, (2) fast response of torque at transient times. For a neutral comparison, both techniques are implemented to the same PMSM at the same operational circumstances. The performances of both MPC structures are found out by simulation results. As a result, torque and current ripples are reduced by CCS; whereas the FCS presents somewhat faster torque response.