Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Framework for adaptive fluid-structure interaction with industrial applications
KTH, School of Computer Science and Communication (CSC), High Performance Computing and Visualization (HPCViz). Basque Center for Applied Mathematics (BCAM), Bizkaia Basque-Country, Spain .ORCID iD: 0000-0002-1695-8809
KTH, School of Computer Science and Communication (CSC), High Performance Computing and Visualization (HPCViz).
KTH, School of Computer Science and Communication (CSC), High Performance Computing and Visualization (HPCViz).ORCID iD: 0000-0003-4256-0463
2013 (English)In: International Journal of Materials Engineering Innovation, ISSN 1757-2754, Vol. 4, no 2, p. 166-186Article in journal (Refereed) Published
Abstract [en]

We present developments in the Unicorn-HPC framework for unified continuum mechanics, enabling adaptive finite element computation of fluid-structure interaction, and an overview of the larger FEniCS-HPC framework for automated solution of partial diffential equations of which Unicorn-HPC is a part. We formulate the basic model and finite element discretisation method and adaptive algorithms. We test the framework on a 2D model problem consisting of a flexible beam in channel flow, and to illustrate the capabilities of the computational framework, we show two application examples from industry and medicine. We simulate a flexible mixer plate in turbulent flow in an exhaust system where the target output is aeroacoustic quantities. The second example is a self-oscillating vocal fold configuration, where the ultimate goal is to predict how the voice is affected by physiological changes from aerodynamics. Here we give the displacement signal of a point on the folds.

Place, publisher, year, edition, pages
2013. Vol. 4, no 2, p. 166-186
Keywords [en]
Adaptive mesh refinement, Fluid structure interaction, High performance computing
National Category
Computational Mathematics
Identifiers
URN: urn:nbn:se:kth:diva-134185DOI: 10.1504/IJMATEI.2013.054394Scopus ID: 2-s2.0-84879009510OAI: oai:DiVA.org:kth-134185DiVA, id: diva2:665099
Note

QC 20131119

Available from: 2013-11-19 Created: 2013-11-19 Last updated: 2018-05-08Bibliographically approved
In thesis
1. Adaptive Finite Element Methods for Fluid Structure Interaction Problems with Applications to Human Phonation
Open this publication in new window or tab >>Adaptive Finite Element Methods for Fluid Structure Interaction Problems with Applications to Human Phonation
2018 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This work presents a unified framework for numerical solution of Fluid Structure Interaction (FSI) and acoustics problems with focus on human phonation. The Finite Element Method is employed for numerical investigation of partial differential equations that model conservation of momentum and mass. Since the resulting system of equations is very large, an efficient open source high performance implementation is constructed and provided. In order to gain accuracy for the numerical solutions, an adaptive mesh refinement strategy is employed which reduces the computational cost in comparison to a uniform refinement. Adaptive refinement of the mesh relies on computable error indicators which appear as a combination of a computable residual and the solution of a so-called dual problem acting as weights on computed residuals. The first main achievement of this thesis is to apply this strategy to numerical simulations of a benchmark problem for FSI. This FSI model is further extended for contact handling and applied to a realistic vocal folds geometry where the glottic wave formation was captured in the numerical simulations. This is the second achievement in the presented work. The FSI model is further coupled to an acoustics model through an acoustic analogy, for vocal folds with flow induced oscillations for a domain constructed to create the vowel /i/. The comparisons of the obtained pressure signal at specified points with respect to results from literature for the same vowel is reported, which is the final main result presented.

Abstract [sv]

Detta arbete presenterar en enhetlig ram för numerisk lösning av fluid-strukturinteraktion (FSI) och akustikproblem med fokus på det mänskligatalet. En finita elementmetod används för numerisk lösning av de partiella differentialekvationer som beskriver konserveringslagar för moment och massa.Eftersom det resulterande systemet av ekvationer är mycket stort, konstruerasen öppen källkod med hög prestanda. För att få hög noggrannhet i de numeriska lösningarna används en adaptiv nätförfiningsstrategi vilken minskar beräkningskostnaden jämfört med en uniform förfining.Adaptiv förfining av nätet bygger på beräknade felindikatorer som bygger på en kombination av en beräkningsbar residual och lösningen av ett såkallat dualt problem. Den första huvudresultatet av denna avhandling är attutveckla en och validera denna strategi för en FSI-modell i ett benchmarkproblem.Denna FSI-modell utvidgas vidare för att hantera kontaktmekanik, ochanvänds sedan för en realistisk modell av stämbandsstrukturerna där denglottiska vågformationen fångas i de numeriska simuleringarna. Detta är detandra huvudresultatet i det presenterade arbetet.FSI-modellen kopplas också till en akustikmodell genom en akustisk analogi,för modell konstruerad för att skapa vokalen / i /. Den erhållna trycksignaleni ett antal punkter jämförs med resultat från litteraturen, vilket är det slutligahuvudresultatet som presenteras.

Place, publisher, year, edition, pages
Stockholm, Sweden: KTH Royal Institute of Technology, 2018. p. 48
Series
TRITA-EECS-AVL ; 2018:38
Keywords
Fluid Structure Interaction, Finite Element Method, Contact Modeling, Acoustic Coupling, High Performance Computing
National Category
Computational Mathematics
Research subject
Applied and Computational Mathematics
Identifiers
urn:nbn:se:kth:diva-227252 (URN)978-91-7729-764-2 (ISBN)
Public defence
2018-05-23, F3, Sing-Sing, floor 2, KTH Kungliga Tekniska högskolan, Lindstedtsvägen 26, Stockholm, Stockholm, 10:30 (English)
Opponent
Supervisors
Funder
EU, FP7, Seventh Framework Programme, 308874Swedish Foundation for Strategic Research Swedish Research CouncilEU, European Research Council
Note

QC 20180509

Available from: 2018-05-09 Created: 2018-05-08 Last updated: 2018-05-09Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records BETA

Hoffman, Johan

Search in DiVA

By author/editor
Jansson, JohanDegirmenci, Niyazi CemHoffman, Johan
By organisation
High Performance Computing and Visualization (HPCViz)
Computational Mathematics

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 222 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf