Learning Algorithm And Its Application Based On Improved Fuzzy Neural Network Parameters Of The Ga

Fuzzy theory provides a method for dealing with uncertain information. Neural network is usually used to approach nonlinear mapping. Fuzzy neural network combines both strongpoints, it is a powerful tool to deal with nonlinear problems. The learning algorithm of fuzzy neural network is vital for the research of its theory and applications, simultaneously it is very worth paying more attention on parameters learning, since the structure learning can be transformed into parameters learning.In this paper, at first, analyzing the development and current situation of fuzzy neural network, genetic algorithm, modeling of nonlinear system and adaptive filter.Introducing the theory of fuzzy system, neural network and the structure and learning algorithm of fuzzy neural network.Expatiating the flow and basic principle of genetic algorithm, an improved strategy of genetic algorithm is put forward after concluding and summarizing the shortcomings of genetic algorithm, that is to say, mixed coding, side-by-side managing of crossover and mutation, adaptive crossover and dynamic mutation, inserting emigration and back propagation operators, and so on.Adopting this improved means to optimize and adjust the parameters of fuzzy neural network and with which applied to modeling of nonlinear system and adaptive filter, it is verifed by simulation and compare with other means.