ITERATIVE LEARNING CONTROL DISTURBANCE OBSERVER-BASED ROBUST CONTROL FOR PMSM SPEED REGULATION FOR EV APPLICATION

Abstract

Achieving precise speed regulation of permanent magnet synchronous motors (PMSMs) in electric vehicles (EVs) remains a significant challenge, particularly under conditions of torque ripple at low speeds, varying load and mismatched external disturbances. This work proposes a novel control strategy that combines Iterative Learning Control (ILC) with a high gain Disturbance Observer (DOB) to enhance robustness and suppress torque ripples. A mathematical model of the PMSM is developed, incorporating control dynamics and mismatched disturbances to accurately represent the system’s behavior. The proposed ILC-DOB framework estimates periodic disturbances, including torque ripple, and utilizes state feedback and feedforward compensation to achieve superior disturbance rejection and robustness. Additionally, this approach integrates advanced control techniques such as Second-Order Sliding Mode Control (SOSMC), Model Predictive Control (MPC), and Reinforcement learning (RL) to further enhance performance. The composite ILC-DOB strategy not only mitigates torque ripples effectively but also adapts to dynamic environmental conditions, ensuring smooth EV operation. The framework’s applicability is demonstrated in high-speed, high-precision scenarios using digital microcontrollers and validated against standard EV reference velocity profiles from the Worldwide Harmonized Light Vehicles Test Procedure (WLTP). This work provides a robust solution for PMSM speed regulation, paving the way for improved performance in EV applications.