Global function approximations using wavelet neural networks

Kwang Ho Shin, Jongsoo Lee

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Feed-forward neural networks have been widely used as function approximation tools in the context of global approximate optimization. In the present study, a wavelet neural network (WNN) which is based on wavelet transform theory is suggested as an alternative to a traditional back-propagation neural network (BPN). The basic theory of wavelet neural network is briefly described, and approximation performance is tested using a nonlinear multimodal function and a composite rotor blade analysis problem. Laplacian of Gaussian function, Mexican function, and Morlet function are considered during the construction of WNN architectures. In addition, approximation results from WNN are compared with those from BPN.

Original languageEnglish
Pages (from-to)753-759
Number of pages7
JournalTransactions of the Korean Society of Mechanical Engineers, A
Volume33
Issue number8
DOIs
Publication statusPublished - 2009 Aug 1

Fingerprint

Neural networks
Backpropagation
Feedforward neural networks
Network architecture
Wavelet transforms
Turbomachine blades
Rotors
Composite materials

All Science Journal Classification (ASJC) codes

  • Mechanical Engineering

Cite this

@article{b08f4e5577184ea093763dcb187a2475,
title = "Global function approximations using wavelet neural networks",
abstract = "Feed-forward neural networks have been widely used as function approximation tools in the context of global approximate optimization. In the present study, a wavelet neural network (WNN) which is based on wavelet transform theory is suggested as an alternative to a traditional back-propagation neural network (BPN). The basic theory of wavelet neural network is briefly described, and approximation performance is tested using a nonlinear multimodal function and a composite rotor blade analysis problem. Laplacian of Gaussian function, Mexican function, and Morlet function are considered during the construction of WNN architectures. In addition, approximation results from WNN are compared with those from BPN.",
author = "Shin, {Kwang Ho} and Jongsoo Lee",
year = "2009",
month = "8",
day = "1",
doi = "10.3795/KSME-A.2009.33.8.753",
language = "English",
volume = "33",
pages = "753--759",
journal = "Transactions of the Korean Society of Mechanical Engineers, A",
issn = "1226-4873",
publisher = "Korean Society of Mechanical Engineers",
number = "8",

}

Global function approximations using wavelet neural networks. / Shin, Kwang Ho; Lee, Jongsoo.

In: Transactions of the Korean Society of Mechanical Engineers, A, Vol. 33, No. 8, 01.08.2009, p. 753-759.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Global function approximations using wavelet neural networks

AU - Shin, Kwang Ho

AU - Lee, Jongsoo

PY - 2009/8/1

Y1 - 2009/8/1

N2 - Feed-forward neural networks have been widely used as function approximation tools in the context of global approximate optimization. In the present study, a wavelet neural network (WNN) which is based on wavelet transform theory is suggested as an alternative to a traditional back-propagation neural network (BPN). The basic theory of wavelet neural network is briefly described, and approximation performance is tested using a nonlinear multimodal function and a composite rotor blade analysis problem. Laplacian of Gaussian function, Mexican function, and Morlet function are considered during the construction of WNN architectures. In addition, approximation results from WNN are compared with those from BPN.

AB - Feed-forward neural networks have been widely used as function approximation tools in the context of global approximate optimization. In the present study, a wavelet neural network (WNN) which is based on wavelet transform theory is suggested as an alternative to a traditional back-propagation neural network (BPN). The basic theory of wavelet neural network is briefly described, and approximation performance is tested using a nonlinear multimodal function and a composite rotor blade analysis problem. Laplacian of Gaussian function, Mexican function, and Morlet function are considered during the construction of WNN architectures. In addition, approximation results from WNN are compared with those from BPN.

UR - http://www.scopus.com/inward/record.url?scp=69149103797&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69149103797&partnerID=8YFLogxK

U2 - 10.3795/KSME-A.2009.33.8.753

DO - 10.3795/KSME-A.2009.33.8.753

M3 - Article

AN - SCOPUS:69149103797

VL - 33

SP - 753

EP - 759

JO - Transactions of the Korean Society of Mechanical Engineers, A

JF - Transactions of the Korean Society of Mechanical Engineers, A

SN - 1226-4873

IS - 8

ER -