Research on Model Predictive Control Strategy for Permanent Magnet Synchronous Motors Incorporating Parameter Mismatch Considerations

To address the computational complexity and parameter sensitivity of conventional three-vector model predictive current control (TV-MPCC) in permanent magnet synchronous motor (PMSM) drives, this study proposes an enhanced MPC strategy with systematic parameter mismatch compensation. A deadbeat current control framework translates current tracking error minimization into voltage vector optimization, establishing a streamlined cost function for dynamic performance enhancement. Sector-based preselection of voltage vectors narrows candidate combinations by prioritizing the dominant sector, substantially reducing computational iterations per control cycle. A Luenberger state observer dynamically estimates and compensates for disturbances caused by inductance and flux linkage mismatches, ensuring robust model fidelity. Experimental validation on a surface-mounted PMSM (SPMSM) driven by a three-phase two-level voltage-source inverter (2L-VSI) confirms the strategy’s effectiveness under transient conditions including startup, load variation, and speed regulation. Comparative simulations demonstrate notable improvements in computational efficiency and stability while preserving steady-state precision and dynamic responsiveness.