A multilevel domain decomposition solver for monolithic fluid-structure interaction problems

Simone Bnà, Sandro Manservisi, Eugenio Aulisa

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

3 Scopus citations

Abstract

This paper deals with multilevel domain decomposition techniques inside a multigrid framework to solve fluidstructure interaction (FSI) problems via the Finite Element Method (FEM). The mathematical problem consists of the Navier-Stokes equations in the Arbitrary-Lagrangian-Eulerian (ALE) formulation coupled with a non-linear incompressible structure model (Neo-Hookean). The coupling between the structure and the fluid, due to the important added-mass effect, is enforced within a unique solver. This kind of problem can be tackled efficiently by domain decomposition techniques. The solution strategy presented in this work is to solve several small local subproblems over subdomain patches using Vanka smoothing inside a multigrid algorithm. The results of a benchmark are presented in order to show the potentiality of the proposed solution method in terms of accuracy and efficiency.

Original languageEnglish
Title of host publication11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013
Pages871-874
Number of pages4
DOIs
StatePublished - 2013
Event11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013 - Rhodes, Greece
Duration: Sep 21 2013Sep 27 2013

Publication series

NameAIP Conference Proceedings
Volume1558
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013
Country/TerritoryGreece
CityRhodes
Period09/21/1309/27/13

Keywords

  • domain decomposition
  • fluid-structure interaction
  • monolithic
  • multilevel

Fingerprint

Dive into the research topics of 'A multilevel domain decomposition solver for monolithic fluid-structure interaction problems'. Together they form a unique fingerprint.

Cite this