Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Conferences
ISSN 2753-3239 CCC: 2
PROCEEDINGS OF THE ELEVENTH INTERNATIONAL CONFERENCE ON ENGINEERING COMPUTATIONAL TECHNOLOGY Edited by: B.H.V. Topping and P. Iványi
Paper 8.4
Across-Partition Contact Analysis with Domain Decomposition Q. Fang and B.A. Izzuddin
Department of Civil and Environmental Engineering, Imperial College London, United Kingdom Q. Fang, B. A. Izzuddin, "Across-Partition Contact Analysis with Domain
Decomposition", in B.H.V. Topping, P. Iványi, (Editors), "Proceedings of the Eleventh International Conference on Engineering Computational Technology", Civil-Comp Press, Edinburgh, UK,
Online volume: CCC 2, Paper 8.4, 2022, doi:10.4203/ccc.2.8.4
Keywords: finite element, contact analysis, node-to-segment contact, parallel
analysis, domain decomposition method, dual Lagrange multipliers.
Abstract
The integration of the node-to-surface contact element into parallel analysis with
across-partition contact boundaries can offer significant reduction in computational
demands for large-scale contact simulations. In the context of contact analyses with
parallel computation , dual-interface-based domain decomposition methods are
adopted in the current work for their considerable flexibility without degradation of
accuracy and convergence behaviour. However, nonlinearity in contact problems still
poses great difficulties for dual-interface-based domain decomposition methods. In
addition, the a priori unpredictability of the contact regions and contact states
complicate the across-partition contact enforcement. The current work purposed a
dual-interface-based domain decomposition method emphasising across-partition
contact coupling. A pair of fully decomposed node-to-surface contact element are
proposed to discretise the across-partition contact boundaries. The assumption of
small incremental displacements is adopted, which a) avoids the excessive coupling
between the decomposed master and slave, b) significantly reduces the
communication overhead, and c) facilitates a flexible across-partition adaptive
analysis.
download the full-text of this paper (PDF, 6 pages, 421 Kb)
go to the previous paper |
|