Construction of fuzzy basis function networks using adaptive least squares method

Author

Lee, Cheol W. ; Shin, Yung C.

Author_Institution

Sch. of Mech. Eng., Purdue Univ., West Lafayette, IN, USA

fYear

2001

fDate

25-28 July 2001

Firstpage

2630

Abstract

A novel algorithm based on the least squares (LS) method and genetic algorithm (GA) is proposed for autonomous learning and construction of FBFN´s when training data are available. The proposed algorithms add significant fuzzy basis functions (FBF) at each iteration during training, based on error reduction measures. The adaptive least squares (ALS) algorithm based on the combined LS and GA, realizes hybrid structure-parameter learning without any human intervention. Simulation studies are performed with numerical examples for comparison with conventional algorithms. The ALS algorithm is applied to the construction of a fuzzy basis function network model for surface roughness in a grinding process using experimental data

Keywords

fuzzy neural nets; genetic algorithms; grinding; learning (artificial intelligence); least squares approximations; process control; radial basis function networks; ALS algorithm; FBF; FBFN; GA; adaptive least squares algorithm; adaptive least squares method; autonomous learning; combined LS/GA; error reduction measures; experimental data; fuzzy basis function network model; fuzzy basis function networks; fuzzy basis functions; genetic algorithm; grinding process; hybrid structure-parameter learning; surface roughness; training data; Adaptive systems; Artificial neural networks; Backpropagation algorithms; Feedforward neural networks; Fuzzy neural networks; Fuzzy systems; Humans; Inference algorithms; Mechanical engineering; Neural networks;

fLanguage

English

Publisher

ieee

Conference_Titel

IFSA World Congress and 20th NAFIPS International Conference, 2001. Joint 9th

Conference_Location

Vancouver, BC

Print_ISBN

0-7803-7078-3

Type

conf

DOI

10.1109/NAFIPS.2001.943638

Filename

943638