MRAS based Speed identification and online updating of rotor time constant for sensorless field oriented controlled induction motor

Abstract

To improve the performance of a sensorless indirect field oriented control of induction drive a speed estimation method insensitive to machines parameters using model reference adaptive system (MRAS) along with online identification of rotor time constant is proposed. Proposed speed identification method uses reactive power as tuning signal for estimating the speed. In conventional MRAS methods, reference and adaptive model suffer from the integral problem and are sensitive to temperature variation of machine parameters. Thus, reducing the accuracy of speed estimator especially at low speeds. The proposed method can produce fast speed estimation, easy to implement, free from integrator and is robust to variations in the parameter error giving improved performance for speed sensor less control of induction motor drive over wide speed range including low speeds. Also, indirect field oriented control involve feed forward adjustment of the slip frequency which is dependent upon varying rotor time constant and may distort the field orientation if not compensated. Therefore online identification of rotor flux MRAS based rotor time constant is proposed along with speed identification. The superiority of the proposed method is verified by MATLAB/SIMULATION over wide range of speed.