Improved Deadbeat Predictive Current Control of Permanent Magnet Synchronous Motor Using a Novel Stator Current and Disturbance Observer

Thanks to the merits of superior dynamic response capability and current tracking performance, the deadbeat predictive current control (DPCC) has become a research hotspot for the permanent magnet synchronous motor (PMSM) drive system. However, DPCC is a model parameter sensitive control method. If there is a motor parameter mismatch, the performance of the DPCC drive system in terms of expected voltage vector, current harmonics, and torque ripple would be influenced. In this paper, firstly, a novel power sliding mode reaching law is proposed, which shortens the convergence time of the system state no matter what the initial state is. Then, an improved non-homogeneous disturbance observer (NHDO) with the proposed power sliding mode reaching law is established, which guarantees d-q axis current errors converge to zero when the PMSM drive system suffers uncertain disturbances, such as motor parameter mismatch. Finally, an improved DPCC using the novel stator current and disturbance observer, which includes the proposed power sliding mode reaching law and NHDO, is established. Hence the accuracy of the predicted current increases significantly, and voltage vectors can be immediately compensated once disturbances occur. Both simulation and platform experiments verify that the improved DPCC can maintain the current tracking performance with lower current ripples than the traditional DPCC when the major motor parameters mismatch. The proposed novel stator current and disturbance observer may also enhance the PMSM's drive performance under other control strategies.