Sliding-Mode Predictive Speed Control for SPMSM Drives Based on Ultra-Local Model

Abstract

To improve the speed tracking precision and overall steady-state performance of the permanent magnet synchronous motor (PMSM) drives, a sliding-mode predictive speed control method is proposed. Firstly, an ultra-local model (ULM) is built to ensure the anti-disturbance ability. Then, a novel discrete sliding-mode surface (SMS) is designed to achieve finite-time convergence of the motor speed. Simultaneously, a quadratic cost function is developed based on the predicted values of SMS. And then, the implemented model predictive control (MPC) algorithm not only efficiently generates the optimal q-axis reference current by the ULM, but also autonomously optimizes the motion trajectory of the motor speed. Specifically, the overall motor performance ultimately depends on the designed SMS, which also emphasizes the crucial role of MPC. Additionally, by evaluating the key controller gain that affects the balance between speed dynamics and system steady-state performance using mathematical and transfer function methods, an adaptive gain based on the Gaussian function is designed to effectively address this issue. Finally, the experiments are performed, and the results validate the superior performance of the proposed method. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.