Titre du document / Document title

Training and using neural networks to represent heuristic design knowledge

Auteur(s) / Author(s)

BIEDERMANN J. D. ⁽¹⁾ ; GRIERSON D. E. ⁽²⁾ ;

Affiliation(s) du ou des auteurs / Author(s) Affiliation(s)

⁽¹⁾ School of Engineering, University of Guelph, Ontario, CANADA
⁽²⁾ Department of Civil Engineering, University of Waterloo, Ontario, CANADA

Résumé / Abstract

The detailed stage of structural design has benefited considerably from computer automation of the numerically intensive tasks of structural analysis, optimization and conformance checking using the procedural programming approach. Such an approach, however, does not allow for the representation and utilization of heuristic knowledge implicit in previous design solutions, which is often difficult or impossible to represent algorithmically. This paper describes how artificial neural networks can be trained to learn heuristic knowledge from previous design solutions and how this knowledge can then be applied to produce a solution to a similar design problem.

Revue / Journal Title

Advances in engineering software   ISSN 0965-9978 

Source / Source

Congrès
International Conference on Computational Structures Technology, Athens , GRECE (30/08/1994)
1996, vol. 27, n^o 1-2 (179 p.)  (21 ref.), [Notes: Selection of reviewed papers], pp. 117-128

Langue / Language

Anglais

Editeur / Publisher

Elsevier, Oxford, ROYAUME-UNI  (1992) (Revue)

Mots-clés anglais / English Keywords

Knowledge representation ; Neural network ; Structural design ; Structural analysis ; Optimization ; Design ; Heuristic method ; Artificial intelligence ;

Mots-clés français / French Keywords

Représentation connaissances ; Réseau neuronal ; Calcul construction ; Analyse structurale ; Optimisation ; Conception ; Méthode heuristique ; Intelligence artificielle ;

Mots-clés espagnols / Spanish Keywords

Representación conocimientos ; Red neuronal ; Cálculo construcción ; Análisis estructural ; Optimización ; Diseño ; Método heurístico ; Inteligencia artificial ;

Localisation / Location

INIST-CNRS, Cote INIST : 18222, 35400006425889.0120