Speed Sensorless of an Induction Motor Using Self-tuning Fuzzy Identification

This paper presents a novel method for sensor-less speed estimation of the induction motor, based on the self-tuning fuzzy identifier incorporating a three-layered network. The back-propagation algorithm is used to adjust the parameter of membership function, as well as to minimize the difference between the actual flux output and the output as predicted by the self- tuning fuzzy identifier, so as to enable precise estimation of speed tracks of the actual motor speed. The experimental results demonstrated that the self- tuning fuzzy identifier possesses a fairly good tracking capability and can accurately and rapidly identify the actual speed of the induction.