Improvement on asynchronous motor system identification based on interactive MRAS

Abstract

Aiming at the inaccurate estimated rotate speed resulting from the parametric variation of the asynchronous motor, this paper designs an estimation method of the Model Reference Adaptive System (MRAS) that can identify the stator resistance and rotate speed simultaneously, and replaces PI link with sliding-mode control based on the approaching rate aiming at the complex PI gain coefficient adjustment existing in the common MRAS estimation method. This method can estimate the stator resistance and rotate speed value of the asynchronous motor during the operation only by detecting the stator side voltage and current, so as to make the whole system gain strong robustness on the change of the stator resistance, and improve the control accuracy of the whole system. The simulation experiment through the vector control on the asynchronous motor verifies the accuracy of the mentioned methods.