State Estimation for Induction Motor Speed-sensorless Control Based on Strong Tracking Filter

In order to solve the poor performance of state estimation and instability at low speeds, a strong tracking filter (STF) method for the joint estimation of speed, rotor flux and load torque is proposed. The torque equation and load torque are introduced into the state equation, and the filter gain matrix is adjusted online by introducing the time-varying fading factor into the covariance matrix of the predicted state. Compared with the state observer algorithm, simulation results show that the STF algorithm can effectively realize the joint state estimation of speed, rotor flux and load torque of induction motor. The proposed method can improve the estimation performance and stability at low speeds, and realize smooth transition of speed and stator current when switching between motoring mode and regenerating mode with superior dynamic performance, high estimation accuracy and strong robustness against load disturbance.