Real-Time Validation of Enhanced Permanent Magnet Synchronous Motor Drive Using Dense-Neural-Network-Based Control

High-performance current and speed control are required to obtain smooth output torque, current tracking, and speed tracking in permanent-magnet synchronous motor (PMSM) drives. The motor speed and stator current control rely on multiple nonlinear motor parameters, which play a crucial role in shaping the performance of PMSM. Moreover, tuning the speed and current controller parameters using the conventional control technique depends on these PMSM parameters, also variation of these parameters will have a decisive influence on the dynamic performance of PMSM. To enhance the robustness of vector control and tracking methodology against PMSM parameter uncertainties and load disturbances, a novel artificial intelligence (AI)-based advanced speed and current control technique for PMSM is proposed in this article. Subsequently, the methodology for designing and training the suggested Dense Neural Network (DNN) controllers are elicited. The proposed controllers can handle the inevitable fluctuation and non-linearity in motor parameters at different load points and drive conditions. The proposed DNN scheme is validated in terms of settling time, dynamic responsiveness, tolerance to parameter fluctuations, and overall robustness. A comparative analysis is conducted against adaptive proportional-integral (API) control applied to the same PMSM within the OPAL-RT real-time simulator (RTS). The viability of the proposed control scheme is substantiated through simulation, Software-In-the-Loop (SIL) and Hardware-In-the-Loop (HIL) testing with an RTS and an automotive-grade controller board across diverse conditions.