The optimization of EKF algorithm based on current errors for speed-sensorless control of induction motors

Abstract

In this study, an optimized extended Kalman filter (EKF) is designed in order to achieve high performance speed-sensorless control of induction motors (IMs). The EKF algorithm estimates the stationary axis components of rotor fluxes, φ_rα and φ_rβ, the rotor angular velocity, ω_m, the rotor resistance, R'_r, besides the stationary axis components of stator currents, i_sα and i_sβ, which are measured as output as well. Also, an optimization process is performed using errors between measured and estimated current components to obtain satisfying estimation performance with EKF. The experimental results prove the high performance control of the proposed system.