A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning

Detalhes bibliográficos
Autor(a) principal: Pontes, Fabrício José [UNESP]
Data de Publicação: 2010
Outros Autores: Silva, Messias Borges [UNESP], Ferreira, João Roberto, De Paiva, Anderson Paulo, Balestrassi, Pedro Paulo, Schönhorst, Gustavo Bonnard
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://dx.doi.org/10.1590/s1678-58782010000500010
http://hdl.handle.net/11449/226282
Resumo: The use of artificial neural networks for prediction in hard turning has received considerable attention in literature. An often quoted drawback of ANNs is the lack of a systematic way for the design of high performance networks. This study presents a DOE based approach for the design of ANNs of Radial Basis Function (RBF) architecture applied to surface roughness prediction in turning of AISI 52100 hardened steel. Experimental factors are the number of radial units on the hidden layer, the algorithm employed to calculate the spread factor of radial units and the algorithm employed to calculate radial function centers. DOE is employed to select levels of factors that benefit network prediction skills. Experiments with data sets of distinct sizes were conducted and network configurations leading to high performance were identified. ANN models obtained proved capable to predict roughness in accurate, precise and affordable way. Results pointed significant factors for network design and revealed that interaction effects between design parameters have significant influence on network performance for the task proposed. The work concludes that the DOE methodology constitutes a better approach to the design of RBF networks for roughness prediction than the most common trial and error approach. Copyright © 2010 by ABCM.
id UNSP_90a3886bce41f50945c740588079cdfc
oai_identifier_str oai:repositorio.unesp.br:11449/226282
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turningAISI 52100 hardened steelDesign of experimentsHard turningRadial basis function neural networksSurface roughnessThe use of artificial neural networks for prediction in hard turning has received considerable attention in literature. An often quoted drawback of ANNs is the lack of a systematic way for the design of high performance networks. This study presents a DOE based approach for the design of ANNs of Radial Basis Function (RBF) architecture applied to surface roughness prediction in turning of AISI 52100 hardened steel. Experimental factors are the number of radial units on the hidden layer, the algorithm employed to calculate the spread factor of radial units and the algorithm employed to calculate radial function centers. DOE is employed to select levels of factors that benefit network prediction skills. Experiments with data sets of distinct sizes were conducted and network configurations leading to high performance were identified. ANN models obtained proved capable to predict roughness in accurate, precise and affordable way. Results pointed significant factors for network design and revealed that interaction effects between design parameters have significant influence on network performance for the task proposed. The work concludes that the DOE methodology constitutes a better approach to the design of RBF networks for roughness prediction than the most common trial and error approach. Copyright © 2010 by ABCM.Departamento de Mecânica Faculdade de Engenharia de Guaratinguetá UNESP - Universidade Estadual Paulista, 12516-410 Guaratinguetá, São PauloInstituto de Eng. Produção e Gestão UNIFEI - Universidade Federal de Itajubá, Caixa Postal 50, 37500-903 Itajubá, Minas GeraisDepartamento de Mecânica Faculdade de Engenharia de Guaratinguetá UNESP - Universidade Estadual Paulista, 12516-410 Guaratinguetá, São PauloUniversidade Estadual Paulista (UNESP)UNIFEI - Universidade Federal de ItajubáPontes, Fabrício José [UNESP]Silva, Messias Borges [UNESP]Ferreira, João RobertoDe Paiva, Anderson PauloBalestrassi, Pedro PauloSchönhorst, Gustavo Bonnard2022-04-28T22:36:58Z2022-04-28T22:36:58Z2010-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article503-510http://dx.doi.org/10.1590/s1678-58782010000500010Journal of the Brazilian Society of Mechanical Sciences and Engineering, v. 32, n. 5 SPEC. ISSUE, p. 503-510, 2010.1806-36911678-5878http://hdl.handle.net/11449/22628210.1590/s1678-587820100005000102-s2.0-79952969703Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengJournal of the Brazilian Society of Mechanical Sciences and Engineeringinfo:eu-repo/semantics/openAccess2024-07-01T20:32:31Zoai:repositorio.unesp.br:11449/226282Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T22:41:36.191444Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
title A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
spellingShingle A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
Pontes, Fabrício José [UNESP]
AISI 52100 hardened steel
Design of experiments
Hard turning
Radial basis function neural networks
Surface roughness
title_short A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
title_full A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
title_fullStr A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
title_full_unstemmed A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
title_sort A DOE based approach for the design of RBF artificial neural networks applied to prediction of surface roughness in AISI 52100 hardened steel turning
author Pontes, Fabrício José [UNESP]
author_facet Pontes, Fabrício José [UNESP]
Silva, Messias Borges [UNESP]
Ferreira, João Roberto
De Paiva, Anderson Paulo
Balestrassi, Pedro Paulo
Schönhorst, Gustavo Bonnard
author_role author
author2 Silva, Messias Borges [UNESP]
Ferreira, João Roberto
De Paiva, Anderson Paulo
Balestrassi, Pedro Paulo
Schönhorst, Gustavo Bonnard
author2_role author
author
author
author
author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (UNESP)
UNIFEI - Universidade Federal de Itajubá
dc.contributor.author.fl_str_mv Pontes, Fabrício José [UNESP]
Silva, Messias Borges [UNESP]
Ferreira, João Roberto
De Paiva, Anderson Paulo
Balestrassi, Pedro Paulo
Schönhorst, Gustavo Bonnard
dc.subject.por.fl_str_mv AISI 52100 hardened steel
Design of experiments
Hard turning
Radial basis function neural networks
Surface roughness
topic AISI 52100 hardened steel
Design of experiments
Hard turning
Radial basis function neural networks
Surface roughness
description The use of artificial neural networks for prediction in hard turning has received considerable attention in literature. An often quoted drawback of ANNs is the lack of a systematic way for the design of high performance networks. This study presents a DOE based approach for the design of ANNs of Radial Basis Function (RBF) architecture applied to surface roughness prediction in turning of AISI 52100 hardened steel. Experimental factors are the number of radial units on the hidden layer, the algorithm employed to calculate the spread factor of radial units and the algorithm employed to calculate radial function centers. DOE is employed to select levels of factors that benefit network prediction skills. Experiments with data sets of distinct sizes were conducted and network configurations leading to high performance were identified. ANN models obtained proved capable to predict roughness in accurate, precise and affordable way. Results pointed significant factors for network design and revealed that interaction effects between design parameters have significant influence on network performance for the task proposed. The work concludes that the DOE methodology constitutes a better approach to the design of RBF networks for roughness prediction than the most common trial and error approach. Copyright © 2010 by ABCM.
publishDate 2010
dc.date.none.fl_str_mv 2010-01-01
2022-04-28T22:36:58Z
2022-04-28T22:36:58Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://dx.doi.org/10.1590/s1678-58782010000500010
Journal of the Brazilian Society of Mechanical Sciences and Engineering, v. 32, n. 5 SPEC. ISSUE, p. 503-510, 2010.
1806-3691
1678-5878
http://hdl.handle.net/11449/226282
10.1590/s1678-58782010000500010
2-s2.0-79952969703
url http://dx.doi.org/10.1590/s1678-58782010000500010
http://hdl.handle.net/11449/226282
identifier_str_mv Journal of the Brazilian Society of Mechanical Sciences and Engineering, v. 32, n. 5 SPEC. ISSUE, p. 503-510, 2010.
1806-3691
1678-5878
10.1590/s1678-58782010000500010
2-s2.0-79952969703
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Journal of the Brazilian Society of Mechanical Sciences and Engineering
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 503-510
dc.source.none.fl_str_mv Scopus
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1808129451530125312