An Improved Continuous Control Set Model Predictive Current Control Algorithm for Permanent Magnet Synchronous Motor

Area of Science:

• Electrical Engineering
• Control Systems
• Power Electronics

Background:

• Continuous Control Set Model Predictive Current Control (CCS-MPCC) algorithms often suffer from poor robustness.
• Parameter uncertainties in motor models (resistance, flux linkage, inductance) degrade predictive control accuracy.
• Existing methods may not adequately compensate for real-time system disturbances.

Purpose of the Study:

• To design a robust CCS-MPCC algorithm by integrating a linear-nonlinear switching extended state observer.
• To improve the prediction accuracy and robustness of CCS-MPCC against parameter variations.
• To validate the proposed control strategy on a permanent magnet synchronous motor (PMSM) experimental platform.

Main Methods:

• Developed a model predictive controller utilizing linear-nonlinear switching extended state observers.
• Demonstrated theoretical equivalence between the observer-based deadbeat MPC and Active Disturbance Rejection Control (ADRC).
• Implemented real-time disturbance detection and feedback using the observer to correct for parameter deviations.

Main Results:

• The proposed algorithm exhibits superior speed response compared to traditional methods under parameter deviations.
• Observed significant reductions in current harmonic content: up to 1.03%, 1.25%, and 1.12% across three operating conditions.
• The linear-nonlinear switching extended state observer effectively compensates for system disturbances.

Conclusions:

• The integration of linear-nonlinear switching extended state observers significantly enhances the robustness of CCS-MPCC.
• The developed controller provides accurate current prediction and improved dynamic performance for PMSM drives.
• This approach offers a practical solution for mitigating the impact of parameter variations in predictive control systems.