kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Time-resolved Adaptive Direct FEM Simulation of High-lift Aircraft Configurations: Chapter in "Numerical Simulation of the Aerodynamics of High-Lift Configurations'", Springer
KTH, School of Computer Science and Communication (CSC), Computational Science and Technology (CST).ORCID iD: 0000-0002-1695-8809
(BCAM - Basque Center for Applied Mathematics)
KTH, School of Computer Science and Communication (CSC), Computational Science and Technology (CST). (BCAM - Basque Center for Applied Mathematics)ORCID iD: 0000-0002-0191-7895
KTH, School of Computer Science and Communication (CSC), Computational Science and Technology (CST).ORCID iD: 0000-0002-5020-1631
Show others and affiliations
2018 (English)In: Numerical Simulation of the Aerodynamics of High-Lift Configurations / [ed] Omar Darío López Mejia andJaime A. Escobar Gomez, Springer, 2018, p. 67-92Chapter in book (Refereed)
Abstract [en]

We present an adaptive finite element method for time-resolved simulation of aerodynamics without any turbulence-model parameters, which is applied to a benchmark problem from the HiLiftPW-3workshop to compute the flowpast a JAXA Standard Model (JSM) aircraft model at realistic Reynolds numbers. The mesh is automatically constructed by the method as part of an adaptive algorithm based on a posteriori error estimation using adjoint techniques. No explicit turbulence model is used, and the effect of unresolved turbulent boundary layers is modeled by a simple parametrization of the wall shear stress in terms of a skin friction. In the case of very high Reynolds numbers, we approximate the small skin friction by zero skin friction, corresponding to a free-slip boundary condition, which results in a computational model without any model parameter to be tuned, and without the need for costly boundary-layer resolution. We introduce a numerical tripping-noise term to act as a seed for growth of perturbations; the results support that this triggers the correct physical separation at stall and has no significant pre-stall effect. We show that the methodology quantitavely and qualitatively captures the main features of the JSM experiment-aerodynamic forces and the stall mechanism-with a much coarser mesh resolution and lower computational cost than the state-of-the-art methods in the field, with convergence under mesh refinement by the adaptive method. Thus, the simulation methodology appears to be a possible answer to the challenge of reliably predicting turbulent-separated flows for a complete air vehicle.

Place, publisher, year, edition, pages
Springer, 2018. p. 67-92
National Category
Computational Mathematics
Identifiers
URN: urn:nbn:se:kth:diva-211705DOI: 10.1007/978-3-319-62136-4_5Scopus ID: 2-s2.0-85053970698OAI: oai:DiVA.org:kth-211705DiVA, id: diva2:1130499
Note

QC 20241108

Part of ISBN 978-3-319-62136-4, 978-3-319-62135-7

Available from: 2017-08-09 Created: 2017-08-09 Last updated: 2024-11-08Bibliographically approved
In thesis
1. Finite Element simulations: computations and applications to aerodynamics and biomedicine
Open this publication in new window or tab >>Finite Element simulations: computations and applications to aerodynamics and biomedicine
2020 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

  Partial Differential Equations describe a large number of phenomena of practical interest and their solution usually requires running huge simulations on supercomputing clusters.  Especially when dealing with turbulent flows, the cost of such simulations, if approached naively, makes them unfeasible, requiring modelling intervention.  This work is concerned with two main aspects in the field of Computational Sciences.  On the one hand we explore new directions in turbulence modelling and simulation of turbulent flows; we use an adaptive Finite Element Method and an \emph{infinite Reynolds number} model to reduce the computational cost of otherwise intractable simulations, showing that we are able to perform time-dependent computations of turbulent flows at very high Reynolds numbers, considered the main challenge in modern aerodynamics.  The other focus of this work is on biomedical applications.  We develop a computational model for (Cardiac) Radiofrequency Ablation, a popular clinical procedure administered to treat a variety of conditions, including arrhythmia.  Our model improves on the state of the art in several ways, most notably addressing the critical issue of accurately approximating the geometry of the configuration, which proves indispensable to correctly reproduce the physics of the phenomenon.

Abstract [sv]

    Partiella differentialekvationer kan användas för att beskriva ett stort antal fenomen av praktiskt intresse.    Vanligtvis krävs enorma simuleringar på superdatorkluster för att hitta deras lösningar.    I synnerhet vid arbete med turbulent flöde.    Dessa simuleringar är så resurskrävande att utan specialbehandling så är de ohanterbara och kräver manuella modelleringsingrepp.    Denna avhandling består av två huvuddelar.    Först utforskar vi nya riktningar i turbulensmodellering och simulering av turbulent flöde.    Vi använder oss av en adaptiv finit elementmetod och en modell med  oändliga \emph{Reynoldstal} för att reducera beräkningskostnaden för annars ohanterbara simuleringar.    Avhandlingen visar att vi lyckats utföra tidsberoende beräkningar av turbulent flöde vid väldigt höga Reynoldstal, vilket är en av de stora utmaningarna i modern aerodynamik.    Den andra delen i denna avhandlingen fokuserar på biomedicinska tillämpningar.    Vi har utvecklat en modell för radiofrekvensablation, ett populärt medicinskt ingrepp som är del i behandlingen av ett flertal sjukdomar, inklusive arytmi.    Vår modell överträffar befintliga modeller på flera punkter.    Mest markant genom att noggrant approximera  konfigurationens geometri, vilket är väsentligt för att korrekt kunna reproducera fenomenets fysik.

Abstract [es]

    Las ecuaciones en derivadas parciales describen muchos fenómenos de interés práctico y sus soluciones suelen necesitar correr simulaciones muy costosas en clústers de cálculo.    En el ámbito de los flujos turbulentos, en particular, el coste de las simulaciones es demasiado grande si se utilizan métodos básicos, por eso es necesario modelizar el sistema.    Esta tesis doctoral trata principalmente de dos temas en Cálculo Científico.    Por un lado, estudiamos nuevos desarrollos en la modelización y simulación de flujos turbulentos; utilizamos un Método de Elementos Finitos adaptativo y un modelo de \emph{número de Reynolds infinito} para reducir el coste computacional de simulaciones que, sin estas modificaciones, serían demasiado costosas.    De esta manera conseguimos lograr simulaciones evolutivas de flujos turbulentos con número de Reynolds muy grande, lo cual se considera uno de los mayores retos en aerodinámica.    El otro pilar de esta tesis es una aplicación biomédica.    Desarrollamos un modelo computacional de Ablación (Cardiaca) por Radiofrecuencia, una terapia común para tratar varias enfermedades, por ejemplo algunas arritmias.    Nuestro modelo mejora los modelos existentes en varias maneras, y en particular en tratar de obtener una aproximación fiel de la geometría del sistema, lo cual se descubre ser crítico para simular correctamente la física del fenómeno.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2020. p. 68
Series
TRITA-EECS-AVL ; 2020:66
Keywords
radiofrequency ablation, finite elements, numerical simulations, partial differential equations, biomedical applications, HPC
National Category
Computational Mathematics
Research subject
Computer Science
Identifiers
urn:nbn:se:kth:diva-285936 (URN)978-91-7873-710-9 (ISBN)
Public defence
2020-12-11, Kollegiesalen, Brinellvägen 8, Stockholm, 14:00 (English)
Opponent
Supervisors
Note

QC 20201118

Available from: 2020-11-16 Created: 2020-11-13 Last updated: 2022-06-25Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopushttps://link.springer.com/chapter/10.1007/978-3-319-62136-4_5

Authority records

Jansson, JohanLeoni, MassimilianoJansson, NiclasHoffman, Johan

Search in DiVA

By author/editor
Jansson, JohanLeoni, MassimilianoJansson, NiclasHoffman, Johan
By organisation
Computational Science and Technology (CST)
Computational Mathematics

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 938 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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