Matches in SemOpenAlex for { <https://semopenalex.org/work/W2046793233> ?p ?o ?g. }
Showing items 1 to 72 of
72
with 100 items per page.
- W2046793233 endingPage "21" @default.
- W2046793233 startingPage "7" @default.
- W2046793233 abstract "This paper addresses the design of genetic algorithms in developing a hybrid neural network model for aluminium alloy flow stress prediction. The hybrid neural network model consists of a parallel grey-box model structure, with the resulting predictions combining the outputs from the constitutive equations and a neural network. Previous work shows that the hybrid neural network model can deliver better model performance than a neural network model or the constitutive equations. However, the level of performance improvement of the hybrid model depends on the quality of the constitutive model used. This motivates the search for a better constitutive model, with genetic algorithms being employed to optimize its parameters. The advantage of genetic algorithms is that they do not require any gradient information nor continuity assumption in searching for the best parameters. A number of genetic optimization schemes, with different coding schemes (such as binary coding and real-value chromosomes) and different genetic operators for selection, crossover and mutation, have been investigated. The real-value coded genetic algorithms converge much more rapidly and are more efficient since there is no need for chromosome encoding and decoding. Compared with previous work, the resulting hybrid model performance has been improved, mainly in the generalization capability and with a simpler neural network structure. Also, the model response surfaces are much smoother and more metallurgically convincing." @default.
- W2046793233 created "2016-06-24" @default.
- W2046793233 creator A5045735258 @default.
- W2046793233 creator A5069916821 @default.
- W2046793233 creator A5051050779 @default.
- W2046793233 date "2003-02-01" @default.
- W2046793233 modified "2023-09-23" @default.
- W2046793233 title "Genetic algorithms and hybrid neural network modelling for aluminium stress—strain prediction" @default.
- W2046793233 cites W2013520651 @default.
- W2046793233 cites W2022635608 @default.
- W2046793233 cites W2023104721 @default.
- W2046793233 cites W2040135606 @default.
- W2046793233 cites W2091759466 @default.
- W2046793233 cites W2142183404 @default.
- W2046793233 cites W2527319573 @default.
- W2046793233 doi "https://doi.org/10.1177/095965180321700103" @default.
- W2046793233 hasPublicationYear "2003" @default.
- W2046793233 type Work @default.
- W2046793233 sameAs 2046793233 @default.
- W2046793233 citedByCount "7" @default.
- W2046793233 countsByYear W20467932332012 @default.
- W2046793233 countsByYear W20467932332013 @default.
- W2046793233 countsByYear W20467932332014 @default.
- W2046793233 crossrefType "journal-article" @default.
- W2046793233 hasAuthorship W2046793233A5045735258 @default.
- W2046793233 hasAuthorship W2046793233A5051050779 @default.
- W2046793233 hasAuthorship W2046793233A5069916821 @default.
- W2046793233 hasConcept C105795698 @default.
- W2046793233 hasConcept C11413529 @default.
- W2046793233 hasConcept C119857082 @default.
- W2046793233 hasConcept C122507166 @default.
- W2046793233 hasConcept C154945302 @default.
- W2046793233 hasConcept C179518139 @default.
- W2046793233 hasConcept C2779990667 @default.
- W2046793233 hasConcept C33923547 @default.
- W2046793233 hasConcept C41008148 @default.
- W2046793233 hasConcept C50644808 @default.
- W2046793233 hasConcept C57273362 @default.
- W2046793233 hasConcept C8880873 @default.
- W2046793233 hasConceptScore W2046793233C105795698 @default.
- W2046793233 hasConceptScore W2046793233C11413529 @default.
- W2046793233 hasConceptScore W2046793233C119857082 @default.
- W2046793233 hasConceptScore W2046793233C122507166 @default.
- W2046793233 hasConceptScore W2046793233C154945302 @default.
- W2046793233 hasConceptScore W2046793233C179518139 @default.
- W2046793233 hasConceptScore W2046793233C2779990667 @default.
- W2046793233 hasConceptScore W2046793233C33923547 @default.
- W2046793233 hasConceptScore W2046793233C41008148 @default.
- W2046793233 hasConceptScore W2046793233C50644808 @default.
- W2046793233 hasConceptScore W2046793233C57273362 @default.
- W2046793233 hasConceptScore W2046793233C8880873 @default.
- W2046793233 hasIssue "1" @default.
- W2046793233 hasLocation W20467932331 @default.
- W2046793233 hasOpenAccess W2046793233 @default.
- W2046793233 hasPrimaryLocation W20467932331 @default.
- W2046793233 hasRelatedWork W1992342254 @default.
- W2046793233 hasRelatedWork W1996785991 @default.
- W2046793233 hasRelatedWork W1998415800 @default.
- W2046793233 hasRelatedWork W2024187501 @default.
- W2046793233 hasRelatedWork W2086367526 @default.
- W2046793233 hasRelatedWork W2371782252 @default.
- W2046793233 hasRelatedWork W2375901233 @default.
- W2046793233 hasRelatedWork W2383319832 @default.
- W2046793233 hasRelatedWork W3195272954 @default.
- W2046793233 hasRelatedWork W4312438320 @default.
- W2046793233 hasVolume "217" @default.
- W2046793233 isParatext "false" @default.
- W2046793233 isRetracted "false" @default.
- W2046793233 magId "2046793233" @default.
- W2046793233 workType "article" @default.