On the usefulness of non-gradient approaches in topology optimization

    Research output: Contribution to journalJournal articleResearchpeer-review

    Abstract

    Topology optimization is a highly developed tool for structural design and is by now being extensively used in mechanical, automotive and aerospace industries throughout the world. Gradient-based topology optimization algorithms may efficiently solve fine-resolution problems with thousands and up to millions of design variables using a few hundred (finite element) function evaluations (and even less than 50 in some commercial codes). Nevertheless, non-gradient topology optimization approaches that require orders of magnitude more function evaluations for extremely low resolution examples keep appearing in the literature. This forum article discusses the practical and scientific relevance of publishing papers that use immense computational resources for solving simple problems for which there already exist efficient solution techniques.
    Original languageEnglish
    JournalStructural and Multidisciplinary Optimization
    Volume43
    Issue number5
    Pages (from-to)589-596
    ISSN1615-147X
    DOIs
    Publication statusPublished - 2011

    Keywords

    • Discrete optimization
    • Topology optimization
    • Stochastic optimization
    • Generic algorithms

    Fingerprint

    Dive into the research topics of 'On the usefulness of non-gradient approaches in topology optimization'. Together they form a unique fingerprint.

    Cite this