Optimal Tuning of PI Controller for Full-Order Flux Observer of Induction Motor Using the Immune Genetic Algorithm

Abstract

This paper presents a new method to tune the parameters of the adaptation PI controller of full-order flux observer. The method employs an Immune Genetic Algorithm (IGA) based optimization routine that can be implemented off-line. A novel fitness function is designed to assess both the estimation accuracy and the noise sensitivity of the rotor speed estimation system when each antibodys parameters are employed. The diversity of population is guaranteed by the evaluating of the antibody similarities function. The Roulette-wheel selection is used to choose the parents and large mutation probability is adopted to prevent the evolution from prematurity. The simulation results verify that the IGA has better performance in convergence speed and computation efficiency compared to the traditional GA.