Robust Position Estimator for Sensorless Control of Traction Motor in Electric Vehicles

Although the sensorless control of traction motor can improve the mileage, cost, and reliability of electric vehicles (EVs), its practical use remains challenging due to the vulnerability of conventional position estimator for sensorless control to severe load torque. However, this limitation can be overcome by utilizing the kinematic and dynamic sensors embedded in the latest EVs. Therefore, this study proposes a new position estimator that utilizing the total torque and speed error as the inputs by reprocessing the vehicle acceleration information to improve the robustness against the load torque. In addition, gains of proposed position estimator are determined to nullify errors that may occur during the reprocessing of information. As the results, the proposed position estimator can achieve the high estimation bandwidth, while improving the overall estimation performance. Furthermore, the practical effectiveness of proposed position estimator is verified by both the experimental tests and hardware-in-the-loop (HIL) tests under various speed and load torque conditions.