EXPERIMENTAL BASED INVESTIGATION OF INDUCTION MOTOR IDENTIFICATION AND CLASSIFICATION

Abstract

An invariant parameters based modeling and an offline identification of a single cage, double cage and deep bar Induction Motor (IM) are developed. Using steady state electric measurements (voltage, stator current and active power), the IM identification is developed by performing a locked rotor test for different frequencies. The linear Least Squares Technique (LST) and the Genetic Algorithm (GA) are used so as to classify the IM according to its rotor type (single cage, double cage or deep bar). The identification and classification algorithms are validated on four IMs. The accuracy and validity of the algorithms are verified as the NRMSE between measured and simulated speed during starting are less than 2,24%.