Model‐Free Adaptive Super‐Twisting Sliding Mode Speed Control Based on RBFNN Estimator for PMLSM Drive Systems

ABSTRACT Considering the various unknown and uncertain parameters as well as load disturbances of permanent magnet linear synchronous motor (PMLSM) drive systems, this paper proposes a novel model‐free adaptive super‐twisting (MFAST) speed control strategy based on radial basis function neural network (RBFNN) estimator to ensure the satisfactory performance and strong robustness of the speed control. First, by considering all possible unknown and uncertain parameters, the ultralocal model of PMLSM is constructed. Next, the RBFNN estimator is designed to estimate the unknown parameters of the above‐mentioned ultralocal model. Finally, the RBFNN‐based MFAST control law is proposed to guarantee PMLSM drive systems' robustness against various internal and external disturbances. StarSim HIL experiment results demonstrate that the synthesised RBFNN‐based MFAST control strategy can enable PMLSM drive systems to possess high accuracy, remarkable rapidity and strong robustness.