Permanent Magnet Synchronous Motor Drive Using Deep-Neural-Network-Based Vector Control for Electric Vehicle Applications

This paper presents a novel artificial intelligence-based approach for the permanent magnet synchronous machine (PMSM) current and speed control using deep neural networks (DNN). Motor parameters change due to the factors such as temperature, magnetic saturation and armature reaction. The need for such a controller arises since conventional proportional integral (PI) controllers do not perform optimally during such variations. Moreover, today's enhanced controller hardware enables the implementation of such control techniques. This paper evaluates the performance and robustness of the proposed DNN based controllers when the motor parameters and load vary. The simulation results compare the performance of both conventional PI controllers and the proposed DNN based controllers. The simulation results show that the proposed DNN based controllers can outperform conventional PI controllers in terms of settling time, dynamic response, and robustness to parameter variations.