Super-twisting sliding-mode observer-based model reference adaptive speed control for PMSM drives

This paper develops a super-twisting sliding-mode observer-based model reference adaptive speed controller (STSMO-MRA-SC) for the permanent-magnet synchronous motor-based variable speed drive (PMSM-VSD) system. A stable first-order linear model is selected as the reference model to describe the required speed trajectory. To make the actual speed of the PMSM-VSD system follow this trajectory, the proposed STSMO-MRA-SC comprises three terms: (1) the stabilization term dependent on known parameters of the motion dynamics and the selected reference model for stabilizing the speed tracking error dynamics asymptotically; (2) the disturbance compensation term based on the STSMO for compensating the lumped disturbance in the speed tracking error dynamics; and (3) the error compensation term updated online by the adaptive law for confronting the estimation error of the STSMO in practice. Comparative experimental tests among the classic MRA-SC, the radial basis function neural network-based MRA-SC and the proposed STSMO-MRA-SC are performed. Experimental results have verified the effectiveness and the superiority of the proposed STSMO-MRA-SC.