An alternative to multiscale simulators based on model order reduction for nonlinear fracture mechanics

Lee Margetts, O Lagrouche (Editor), A El Kacimi (Editor), P Woodward (Editor), G Medero (Editor)

    Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

    Abstract

    We present an adaptive model order reduction for paramatrised nonlinear fracture mechanics. Efficient corrections of the intial reduced space, provided by the proper orthogonal decomposition, are performed on-the-fly during the solution process. The proposed strategy establishes a bridge between model order reduction and full-scale augmented Newton-Krylov solvers. We demonstrate its efficiency through numerical examples.
    Original languageEnglish
    Title of host publicationProceedings of the 19th UK Conference of the Association for Computational Mechanics in Engineering
    EditorsO Lagrouche, A El Kacimi, P Woodward, G Medero
    Place of PublicationEdinburgh, Scotland
    PublisherHeriot-Watt University
    Pages89-92
    Number of pages4
    ISBN (Print)ISBN 978-0-9565951-1-9
    Publication statusPublished - 5 Apr 2011
    Event19th UK Conference of the Association for Computational Mechanics in Engineering - Heriot-Watt University, Edinburgh
    Duration: 5 Apr 20116 Apr 2011
    http://web.sbe.hw.ac.uk/acme2011/index.htm

    Conference

    Conference19th UK Conference of the Association for Computational Mechanics in Engineering
    CityHeriot-Watt University, Edinburgh
    Period5/04/116/04/11
    Internet address

    Keywords

    • Nonlinear fracture
    • Model order reduction
    • Newton-Krylov
    • Augmented conjugate gradient

    Fingerprint

    Dive into the research topics of 'An alternative to multiscale simulators based on model order reduction for nonlinear fracture mechanics'. Together they form a unique fingerprint.

    Cite this