Identification of induction machine parameters using a new adaptive genetic algorithm

Interests in seeking accurate and reliable parameters for identification methods of induction machines constitute major modeling concerns for performance prediction and assessment. This article presents an optimization, technique-based, parameters identification for the machine steady-state operation. This method is based on a genetic algorithm incorporating a new adaptive scheme for a computing time reduction. It aims to accurately identify the parameters by solving a nonlinear curve fitting problem. Finally, the obtained machine performances of the adaptive genetic algorithm method are compared with both reference and near-least-square-error estimator using experimental variable load measurements.