Application of Machine Learning for Speed and Torque Prediction of PMS Motor in Electric Vehicles

Abstract

Permanent Magnet Synchronous (PMS) motor has huge applications in Electric Vehicles. Therefore, a correct prediction of both speed and torque is required for satisfactory result. A dataset is considered having real time data of ambient temperature, coolant temperature, direct axis and quadrature axis voltage and current, yoke temperature, rotor temperature and stator temperature for prediction of motor speed and torque. This dataset is collected from the test bench of University of Paderbon laboratory. Various machine learning models have been applied on the dataset. The result shows that Fine Tree is the best model for prediction of both speed and torque of the permanent magnet synchronous motor having lowest RMSE of 0.029224 and 0.052538 for prediction of speed and torque respectively. © 2020 IEEE.