MPTC for Dual Three-Phase PMSM Based on Novel Virtual Voltage Vector

Model predictive torque control (MPTC), grounded in the utilization of virtual voltage vector (VVV), adeptly mitigates the prevailing current harmonics of dual three-phase permanent magnet synchronous motor (PMSM). Nevertheless, employing VVVs characterized by significant step increments and fixed amplitudes is liable to induce substantial torque ripple. Focusing on the negative impact of the amplitude and phase of the applied voltage vector on torque control, this paper introduces a MPTC method using novel VVV to effectively mitigate torque ripple and suppress current harmonics. Firstly, the amplitude and phase information of the expected reference voltage vector (RVV) is obtained by deadbeat direct torque and flux control (DB-DTFC) methods. Subsequently, a novel VVV set is constructed to reduce the effect of the phase difference between the expected RVV and the applied voltage vector. Then, based on the impact stemming from the amplitude deviation between three candidate voltage vectors and the expected RVV, the action time of these candidate voltage vectors are optimized in the form of duty cycle. The optimal vector is then determined through the cost function, thereby further reducing the torque ripple. Finally, experimental results are offered to verify the effectiveness of the proposed MPTC method.