Speed sensorless vector control of induction motor based on reduced order extended Kalman filter

A vector control of induction motor by the estimated rotor speed and rotor flux using a new reduced-order extended Kalman filter is proposed. With this method, only rotor flux components are selected as the state variables. The rotor speed is regarded as a parameter, and the augmented states consist of rotor flux components and the rotor speed. The reduced-order extended Kalman filter is employed to estimate the rotor speed and rotor flux of an induction motor on line by using only stator voltage references and the measured stator currents. The estimated rotor flux is used for field orientation and flux control, and the estimated speed is used for overall speed control. Since the current control is performed at a synchronous rotating reference frame, the estimated rotor flux information is also used for the reference frame transformation of the current controller. The algorithm order reduction decreases the computational complexity and makes the proposed estimator technologically feasible to be implemented in real time. Computer simulations of the speed control have been carried out to test the usefulness of the estimation algorithm. The simulation results show good performance of the rotor flux and speed estimations.