Assessment and improvement of mapping algorithms for non-matching meshes and geometries in computational FSI

Tianyang Wang, Roland Wüchner, Stefan Sicklinger, Kai Uwe Bletzinger

Research output: Contribution to journalArticlepeer-review

25 Scopus citations

Abstract

This paper investigates data mapping between non-matching meshes and geometries in fluid-structure interaction. Mapping algorithms for surface meshes including nearest element interpolation, the standard mortar method and the dual mortar method are studied and comparatively assessed. The inconsistency problem of mortar methods at curved edges of fluid-structure-interfaces is solved by a newly developed enforcing consistency approach, which is robust enough to handle even the case that fluid boundary facets are totally not in contact with structure boundary elements due to high fluid refinement. Besides, tests with representative geometries show that the mortar methods are suitable for conservative mapping but it is better to use the nearest element interpolation in a direct way, and moreover, the dual mortar method can give slight oscillations. This work also develops a co-rotating mapping algorithm for 1D beam elements. Its novelty lies in the ability of handling large displacements and rotations.

Original languageEnglish
Pages (from-to)793-816
Number of pages24
JournalComputational Mechanics
Volume57
Issue number5
DOIs
StatePublished - 1 May 2016

Keywords

  • Dual Lagrange multipliers
  • Fluid-structure interaction
  • Mortar method
  • Non-matching meshes
  • Nonlinear beam FSI

Fingerprint

Dive into the research topics of 'Assessment and improvement of mapping algorithms for non-matching meshes and geometries in computational FSI'. Together they form a unique fingerprint.

Cite this