Sliding Mode Observer Based Sensorless Model Predictive Current Control for Induction Motor

Abstract

In this paper, a sensorless model predictive current control based on sliding mode observer for induction motor is proposed. The predictive method is based on examining the feasible voltage vector in a prescribed cost function. The voltage vector that minimizes the stator current error in the cost function is selected. An advanced model predictive current control is proposed to solve the problem of system noise amplification in the differential link caused by the current slope calculation in the back electromotive force model (EFM). The predictive model includes the sliding mode feedbacks which are employed to observe the real-time rotor speed and flux linkage. Besides, the feedback gains satisfy the Lyapunov's law of stability. In order to verify the fine dynamic and static performance of the proposed method, simulation and experimental results are presented.