Modulated Model Predictive Speed Controller for PMSM Drives Employing Voltage-Based Cost Function

Abstract

Various electrical drive systems have widely implemented the classical cascaded field-oriented control (FOC) topology, including speed loop, current loop, and modulation. On the other hand, modulated model predictive control (M2PC) has been employed recently for different applications for faster dynamic response and better power quality. The FOC topology's speed and current control loops can be merged to simplify the control system structure and improve the system dynamics. Therefore, a noncascaded speed loop controller employing M2PC for permanent magnet synchronous motors is introduced. The required simulation work has been developed to analyze the algorithm performance compared to proportional integral (PI), noncascaded model predictive control, and M2PC controllers. In addition, it has been applied practically through a dedicated testing rig, and results are investigated showing its merits including harmonic content, dynamic behavior, and robustness against parameter mismatch.