Composite Generalized Dynamic Predictive Control With Self-Tuning Horizon for Wide-Range Speed Regulation of PMSM Drives
Zhongkun Cao, Jianliang Mao, Xin Dong, Rafał Madoński, Chuanlin Zhang, Jun Yang
Abstract
This article proposes and investigates a novel composite generalized dynamic predictive control (GDPC) approach for wide-range speed regulation of permanent magnet synchronous motor (PMSM) drives under a non-cascade configuration. Firstly, to eliminate negative effects arising from both matched and mismatched disturbances within the receding-horizon optimization of the baseline generalized predictive control (GPC) design, two high-order sliding mode observers (HOSMOs) are constructed for disturbance estimation, which allows for offset-free tracking of the reference speed. Furthermore, a novel self-tuning horizon mechanism is introduced for wide-range speed regulation scenarios of PMSM drives. This feature enables autonomous and dynamic adjustment of the prediction horizon across diverse speed regulation ranges, resulting in significantly improved performance optimization compared to the conventional GPC method. Finally, the proposed GDPC methodology is implemented on a digital signal processor (DSP) hardware system. The simulations and experiments demonstrate the effectiveness of the proposed control approach.