An EKF-Based Estimator for the Speed Sensorless Vector Control of Induction Motors

Abstract

This article offers a solution to the performance deteriorating effect of uncertainties in the sensorless control of induction motors (IMs). The major contribution of the study is the development and implementation of an extended Kalman filter (EKF) algorithm that takes electrical and mechanical uncertainties into account. In this regard, this is the first known study to estimate the mechanical uncertainties together with the rotor resistance, R′ _r , without injecting high frequency signals. The EKF algorithm also estimates the rotor flux, angular velocity and stator currents with no apriori knowledge on the states and initial values taken as zero. Experiments performed under unknown load torque and with rotor resistance variations up to twice the rated value demonstrate the good performance and robustness of the estimation method. The algorithm also estimates the mechanical uncertainties as a constant state to capture the unknown viscous and Coulomb friction in steady-state; therefore, it could be used to improve the performance of the velocity or position control of IMs, if utilized in combination with a compensation scheme.