Sensorless torque-controlled Induction Motor drive for EV applications

This paper presents a robust sensorless control technique for an Induction Motor (IM) used for Electric Vehicle (EV) applications. In order to estimate the rotor speed, a new back-electromotive-force (back-EMF) based-model reference adaptive system (MRAS) estimator is proposed. The new back-EMF MRAS is independent from the stator resistance and stator inductance parameters. Hence the proposed scheme is immune to these parameters variations. It also is robust against inverter nonlinearity and shows consistency at starting from standstill. The proposed scheme has been experimentally implemented and tested on a 19-kW induction motor (IM), of a type used for the automotive industry for EV applications. A comparison is carried out between the performance of the conventional and the proposed back-EMF MRAS. Tests include sensitivity to stator resistance variation and capability to start from standstill. Experimental results show the improved performance achieved using the proposed sensorless control scheme and its suitability for EV applications.