An improved sensorless FS-PTC of induction motors using estimated stator currents

Abstract

Conventionally a finite state predictive torque control (FS-PTC) strategy uses measured stator currents and estimated stator and rotor flux to predict stator flux and torque of induction motor (IM). In FS-PTC, the accuracy of the prediction model is directly dependent on the stator currents and the rotor speed. Direct application of measured stator currents into the prediction model degrades the control performance in terms of current total harmonic distortion (THD) and speed error, especially at lower speeds. This is because injection of noise into the prediction model leads to undesired switching actuation for the inverter. To avoid this problem, this paper proposes an improved prediction model for speed sensorless FS-PTC of IM drives. The estimated stator currents instead of measured currents are fed back to the controller and thus small stator current THD is confirmed. Extended Kalman filter (EKF), a promising state observer for sensorless control system, has been employed with FS-PTC to estimate rotor speed, rotor flux and stator currents accurately. The proposed control strategy has been verified experimentally, and improved torque and flux responses are achieved.