PERFORMANCE EVALUATION OF THREE-LEG VOLTAGE SOURCE INVERTER FED UNSYMMETRICAL TWO-PHASE INDUCTION MOTOR BASED ON GENETIC ALGORITHM FOR PARAMETER ESTIMATION

This paper presents an evaluation method of performance characteristics in terms of loss, currents and electromagnetic torque of an unsymmetrical two-phase induction motor driven by a three-leg Voltage Source Inverter (VSI) providing unbalanced two-phase voltages. The model parameters for consideration of loss and dynamic performance of the motor are estimated by using Genetic Algorithm (GA). Also, synthesized winding current waveforms based on a super-position method and frequency domain of known harmonic voltages are investigated. The methodology of the proposed GA applied to parameter estimation of the unsymmetrical two-phase induction motor is fully given. Carrier-based unbalanced Space Vector Pulse With Modulation (SVPWM) is employed and implemented on a low-cost microcontroller. In order to prove the validity of the model with the parameters obtained by GA, performance comparison with the experiment and the model with the parameters obtained by conventional test in laboratory has been made. The results of simulation with the proposed parameters and experiment are in good agreement.