Robust Model Predictive Current Control for Dual-Star Induction Machine

Abstract

Predictive current control (PCC) is considered as an effective and efficient strategy for controlling multiphase drives, offering superior flexibility, fast dynamic response, and reduced computational complexity compared to conventional control methods. This paper presents a robust PCC approach for a dual-star induction machine (DSIM), integrated with direct field-oriented control (DFOC), using proportional-integral (PI) controllers to control mechanical speed and flux to analyze the efficiency of the drive system's behavior in complex challenging scenarios caused by motor's external perturbations and parameters uncertainties. The proposed PCC algorithm incorporates a two-step-ahead prediction horizon to evaluate a cost function that minimizes the deviation between reference and predicted stator currents. The control signal is selected from a finite set of voltage vectors (VVs) provided by a two-level voltage source inverter (2L-VSI), and the optimal switching states combination is selected to ensure precise control and improved performance. The proposed framework is validated through comprehensive simulations conducted in the Simulink/MATLAB environment. The findings achieved superior disturbances rejection capabilities, and minimized steady-state error. Furthermore, the system highlights an effective performance and robustness against simultaneous extreme parameters variations, especially under full load for very low speed scenario.