This paper proposes a novel adaptive super-twisting (AST) speed control strategy based on generalized proportional integral observer (GPIO) for permanent magnet linear synchronous motor (PMLSM) system. First, considering the uncertainty of nonlinear friction, thrust ripple and load variation, a GPIO is adopted to estimate the lumped disturbance of system. Second, for conventional super-twisting control (STC), its high switch gain results in deteriorative dynamics. To overcome its drawback, a new AST strategy for PMLSM system is put forward, which can adaptively adjust gain and thus effectively weaken sliding-mode chattering. Finally, taking into account compensation to the lumped disturbance estimated by GPIO, a novel AST-based speed compound controller (SCC) is synthesized so that the speed of PMLSM system is capable of tracking its reference value rapidly and accurately, and the anti-disturbance ability of system can be enhanced. The experimental results confirm that the SCC can enable PMLSM system to achieve high tracking precision and possess strong robustness against disturbance.

PMLSM， AST， speed control， generalized proportional integral observer， time-varying disturbance