Optimization of Adaptation Gains of Full-order Flux Observer in Sensorless Induction Motor Drives Using Genetic Algorithm

摘要

This study presents a new optimization method of the adaptation PI gains of the full-order flux observer in the sensorless induction motor drives. The new method employs a Genetic Algorithm (GA) based optimization routine that can be implemented off-line. A suitable fitness function is defined to assess the tracking performance, the noise sensitivity and the stability of the rotor speed estimation system when each individual?s parameters are employed. The tournament selection is used to choose the parent individuals and a large mutation probability is used to prevent the evolution from the prematurity. The PI gains calculated according to the design guidelines are put in the initial population to quicken the optimization procedure. With the help of the proposed method, the desirable PI gains can be obtained and the optimization procedure is fast and efficient. Simulation results show that the estimated speed tracks the practical speed well when the obtained PI gains are employed. Simulation results validate the proposed method in the study. Since, the efficient optimization ability, the Genetic Algorithm (GA) is pretty suitable for the optimization of the adaptation PI gains of the full-order flux observer in the sensorless induction motor drives.