Skip to content
Conference proceedings

A comparison of metamodeling techniques for crashworthiness optimization

Stander N, Roux W, Giger M, Redhe M, Fedorova N, Haarhoff J...(+6 more)

Collection of Technical Papers - 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, vol. 4 (2004) pp. 2116-2127

  • 6

    Readers

    Mendeley users who have this article in their library.
  • 13

    Citations

    Citations of this article.
  • N/A

    Views

    ScienceDirect users who have downloaded this article.
Sign in to save reference

Abstract

This crashworthiness optimization study compares the use of three metamodeling techniques while using a sequential random search method as a control procedure. The three methods currently applied are (i) the Successive Linear Response Surface Method, (ii) the Updated Neural Network method and (iii) the Kriging method. Three crashworthiness examples, including a full vehicle multidisciplinary analysis, are investigated. It is shown that, although NN and Kriging seem to require a larger number of initial points, the three metamodeling methods have comparable efficiency when attempting to achieve a converged result. The Neural Network and Kriging methods have the advantage that they can be updated to construct a reasonable global approximation with higher accuracy at the optimum. Copyright © 2004 by Livermore Software Technology Corporation.

Author-supplied keywords

  • Approximation theory
  • Computer simulation
  • Convergence of numerical methods
  • Neural networks
  • Optimization
  • Problem solving
  • Random access storage
  • Reliability

Find this document

Get full text

Cite this document

Choose a citation style from the tabs below