Deadbeat Predictive Control Method for PMSM Based on Parallel Anti-Disturbance Structure
摘要
Permanent magnet synchronous motor (PMSM) drive systems suffer from reduced speed and current tracking control accuracy due to parameter variations and external disturbances. To enhance the system tracking performance and control bandwidth, while reducing the inherent complexity of the traditional speed-current cascade control architecture, a deadbeat predictive control with Parallel Anti-Disturbance Structure (PAS-DPC) is proposed. Firstly, the discrete mathematical model of the PMSM is established, and based on this, the system deadbeat model is developed. Secondly, a disturbance rejection loop is constructed based on the error model caused by parameter variations, effectively suppressing disturbances generated by system parameter perturbation. Then, the controller structure of the disturbance rejection loop is modified to improve the suppression performance against external disturbances. Finally, a real-time observer is utilized to estimate the external load torque. Through performance analysis of the PAS-DPC controller under reference signals, it is confirmed that the system maintains excellent transient response and steady-state tracking performance. A PMSM drive system was built to verify the effectiveness of the proposed method through comparison.