Model Predictive Control of Permanent Magnet Synchronous Motor Based on Optimal Switching Frequency

With the advancement of electrified drive systems, model predictive control for permanent magnet synchronous motors has become a focus of research. However, existing algorithms have issues such as motor operational losses due to frequent power device switching. To tackle these challenges, this study introduces the MPC algorithm for PMSM, incorporating optimal switching frequency principles and integrating a penalty function into the MPC framework. Leveraging upon this foundation, through the manipulation of the penalty factor, the PMSM's dq-axis current performance can be tracked even at low switching frequency. Remarkably, employing an OSF-MPC switching frequency of 864.8hz leads to a 38.3% reduction in current ripple (
$^{\Delta}\mathrm{i}_{\text{RMS}}$
) and a substantial 59.8% decrease in current harmonic (
$\mathrm{I}_{\text{THD}}$
). Additional observations from PMSM operation in an alternative working scenario show that as the motor speed drops to 10rpm and then stabilizes, compared to the MPC algorithm, the OSF-MPC algorithm succeeds in reducing speed fluctuations by 66.8%. The above findings highlight the significant improvement provided by the OSF-MPC algorithm regarding PMSM operational efficiency and control system robustness. The OSF-MPC algorithm has promising application potential in the progressively expanding landscape of electrification.