Knowledge-based global optimization of cold-formed steel columns

H. Liu, Takeru Igusa, B. W. Schafer

Research output: Contribution to journalArticle

Abstract

A knowledge-based global optimization scheme, well-suited for the cross-sections of cold-formed steel structures was developed. The scheme operated within a low-dimensional expert system and used a numerical implementation of the direct strength method (DSM) for the objective function. The expert knowledge for the optimization process was evaluated and added using the feature-based classification trees which indicated the need for tip lengths etc. An expert-based feature space and its associated classification tree effectively encapsulated the knowledge gained in the design optimization.

Original languageEnglish (US)
Pages (from-to)785-801
Number of pages17
JournalThin-Walled Structures
Volume42
Issue number6
DOIs
Publication statusPublished - Jun 1 2004

    Fingerprint

Keywords

  • Classifier model
  • Cold-formed steel
  • Global optimization
  • Knowledge

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanical Engineering

Cite this