A new approach for overcoming failure of current sensors in states estimation of an Induction motor

Failures and uncertainties of current sensors are key issues in states and parameters estimation of Induction motors. In this paper, an EKF equipped with Output Augmented Fusion method (OAF) is used to design a new observer for dominance over these challenges. The proposed observer has the ability of speed estimation in a wide range with less sensitivity to parameters' variations and also failures and uncertainties of current sensors. An EKF is employed to estimate speed as a dynamic state variable over wide range of variation, using extension of states vector with equation of motion and other major parameters consisted of stator/rotor resistances and load torque. Moreover, using OAF algorithm, output matrices are extended to overcome failure probability and uncertainties of current sensors. The effectiveness of this approach is validated by simulation.