High performance variable-speed machines incorporate a model for the system in either the controller or state estimation stages. The accuracy and general robustness of the machine is dependant on this model. Therefore, it must accurately represent both the electrical and electromagnetic interactions within the machine and associated mechanical systems. Recently, some new technologies have been tested in the field of electromechanics like neural networks, fuzzy logic, simulated annealing and genetic algorithms. These methods are increasingly being utilized in solving electric machine problems.; In this thesis, a genetic algorithm (GA)—a form of artificial intelligence—is employed to identify the electric parameters of induction motors. The variables used to calculate the electric parameters are the measured stator currents, stator voltages and rotor speed. The variables are acquired by using Data Acquisition System and Lab VIEW Software. Free acceleration test is performed on 7.5 hp induction motor, using a constant frequency power supply. The performance of the identification scheme is demonstrated with simulated and measured data, and electric parameters obtained using this method are compared with parameters obtained from IEEE standard tests. Based on the results, the method proved to be worth considering in optimizing induction machines and can be applied to a variety of induction motor parameter estimation problems.
展开▼
