A Direct Model Predictive Control Strategy with an Implicit Modulator for Six-Phase PMSMs

Abstract

This paper proposes a direct model predictive control (MPC) scheme for asymmetric six-phase permanent magnet synchronous machines (PMSMs), which combines control and modulation in one computation stage. By emulating the switching pattern of space vector modulation (SVM), the MPC problem is formulated as a four-dimensional current control problem where the switching sequences and instants are computed and directly applied to the inverters. This implicit modulation addresses the issue of a variable switching frequency and spread harmonic spectra of conventional direct MPC methods. Moreover, the effect of the modulation constraints and controller bandwidth on the system performance is investigated as well. To verify the effectiveness of the proposed control strategy, experiments are carried out with an asymmetric six-phase PMSM driven by two three-phase two-level inverters.