Robust Predictive Current Control Using Luenberger Observer for IMs in the Stationary Reference Frame

Abstract

Induction motors are widely utilized in numerous applications due to their robustness, straightforward design, and affordability. Nevertheless, the precision of motor control can be affected by parameter mismatches and unmodeled system dynamics. Advanced control techniques, such as finite control set-model predictive control (FCS-MPC), have been designed to mitigate these issues. The FCS-MPC uses the system’s mathematical model to predict future behavior of the system and select the most suitable control signal from a discrete set of options. Although FCS-MPC proves to be effective, traditional implementations can still face challenges from uncertainties, such as parameter changes or external disturbances. This article presents a novel robust predictive current control for induction motors in the stationary reference frame. It incorporates a discrete Luenberger observer to address the machine’s parameter mismatches. A discrete-time voltage model is derived from the motor’s dynamics, enabling the integration of a discrete Luenberger observer. This integration enhances controller performance, especially in the presence of parameter mismatches. In contrast to conventional FCS-MPC approaches, this method integrates stator voltage into the cost function instead of current, torque, or magnetic flux. Experimental results on an induction motor, tested under both steady-state and dynamic scenarios, confirm that this approach outperforms traditional predictive current control regarding robustness. The new method’s enhanced control capability to better track the reference, even under significant variations in inductance and resistance, underscores its potential as an effective solution.