Sensorless low/zero speed control of induction motors with EKF estimation

This study aims an improved estimation performance for the speed sensorless control of induction motors (IMs) over a very low/zero speed range. For this purpose, an Extended Kalman Filter (EKF) observer is designed to overcome the difficulties of flux/speed estimation at and around zero speed. The estimation problems related to low speed operation are further emphasized due to stator resistance, R_s uncertainties. Thus, the observer is designed to estimate the stator resistance, load torque and velocity, which is taken into account via the equation of motion. With this approach, the rotor flux information is reflected to the stator, thus improving the estimation accuracy. The simulation results obtained for the EKF approach in combination with the Direct Vector Control (DVC) of an IM, indicate improved performance.