A comparison of PSO and GA combined with LS and RLS in identification using fuzzy gaussian neural networks

Author

Shafiabady, Niusha ; Teshnehlab, M. ; Shooredeh, M. Aliyari

Author_Institution

Dept. of Mechatron. Eng., Azad Univ. Sci. & Res. Branch, Tehran, Iran

fYear

2009

fDate

5-8 July 2009

Firstpage

2081

Lastpage

2086

Abstract

In this article, a new method for training the parameters is discussed and we have compared the function of particle swarm optimization with genetic algorithm in training the standard deviation and centers in the antecedent part of fuzzy Gaussian neural network. We have applied least square and recursive least square in training the weights of this fuzzy neural network in the conclusion part. There are four sets of data used to examine the proposed learning strategy to achieve the proper learning mode.

Keywords

Gaussian processes; fuzzy neural nets; genetic algorithms; learning (artificial intelligence); least squares approximations; particle swarm optimisation; GA algorithm; LS method; PSO algorithm; RLS method; fuzzy Gaussian neural network; genetic algorithm; learning strategy; least square method; parameter training; particle swarm optimization; recursive least square method; standard deviation; Fuzzy neural networks; Neural networks; Resonance light scattering; Fuzzy Gaussian Neural Network; Genetic Algorithm; Gradient Descent; Identification; Least Square; Particle Swarm Optimization; Recursive Least Square;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics, 2009. ISIE 2009. IEEE International Symposium on

Conference_Location

Seoul

Print_ISBN

978-1-4244-4347-5

Electronic_ISBN

978-1-4244-4349-9

Type

conf

DOI

10.1109/ISIE.2009.5217923

Filename

5217923