Liu, Jun (2024)
Unstructured finite volume methods for two-phase flows with high density ratios.
Technische Universität Darmstadt
doi: 10.26083/tuprints-00028266
Dissertation, Erstveröffentlichung, Verlagsversion
Kurzbeschreibung (Abstract)
The dissertation presents a comprehensive study of advanced computational methods for simulating incompressible two-phase flows, particularly addressing challenges associated with high-density ratios, mass and momentum conservation, and non-orthogonality errors in unstructured Finite Volume methods. The thesis extends the unstructured Level Set / Front Tracking (LENT) method, introducing the \rhoLENT approach to ensure numerical consistency between mass and momentum conservation in the collocated Finite Volume discretization of the single-field two-phase Navier-Stokes equations. This method demonstrates exact numerical stability for two-phase momentum advection across a wide range of density and viscosity ratios, effectively handling challenging fluid pairings such as mercury/air and water/air, and scenarios involving strong interactions between phases.
Further, the study applies the consistency conditions derived for the \rhoLENT method to geometric flux-based Volume-of-Fluid (VOF) methods. It reveals that standard computations of mass fluxes in these methods can disrupt the equivalence between scaled volume fraction equations and mass conservation equations, depending on temporal and convective term discretization schemes. The thesis proposes a dual solution approach: a consistent combination of temporal discretization and interpolation schemes, and an auxiliary mass conservation equation with a geometric calculation of face-centered densities. This approach is validated for extensive density and viscosity ratios, demonstrating its robustness and effectiveness.
Additionally, the dissertation tackles non-orthogonality errors in unstructured Finite Volume methods, which can compromise force-balanced discretization in simulating incompressible two-phase flows. A novel, deterministic residual-based control of non-orthogonality correction is introduced, removing the number of non-orthogonality corrections as a global parameter from the simulation process. This method ensures force balance, particularly for surface tension and gravity forces, and is verified on polyhedral unstructured meshes with different non-orthogonality levels.
Overall, this dissertation provides contributions by developing, verifying and validating advanced methodologies to accurately and efficiently simulating incompressible two-phase flows under complex conditions. These developments have improved applications in industrial multiphase microfluidics, where precise computational fluid dynamics is crucial.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2024 | ||||
Autor(en): | Liu, Jun | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Unstructured finite volume methods for two-phase flows with high density ratios | ||||
Sprache: | Englisch | ||||
Referenten: | Marić, Dr.-Ing. Tomislav ; Bothe, Prof. Dr. Dieter | ||||
Publikationsjahr: | 29 Oktober 2024 | ||||
Ort: | Darmstadt | ||||
Kollation: | xxxi, 207 Seiten | ||||
Datum der mündlichen Prüfung: | 30 September 2024 | ||||
DOI: | 10.26083/tuprints-00028266 | ||||
URL / URN: | https://tuprints.ulb.tu-darmstadt.de/28266 | ||||
Kurzbeschreibung (Abstract): | The dissertation presents a comprehensive study of advanced computational methods for simulating incompressible two-phase flows, particularly addressing challenges associated with high-density ratios, mass and momentum conservation, and non-orthogonality errors in unstructured Finite Volume methods. The thesis extends the unstructured Level Set / Front Tracking (LENT) method, introducing the \rhoLENT approach to ensure numerical consistency between mass and momentum conservation in the collocated Finite Volume discretization of the single-field two-phase Navier-Stokes equations. This method demonstrates exact numerical stability for two-phase momentum advection across a wide range of density and viscosity ratios, effectively handling challenging fluid pairings such as mercury/air and water/air, and scenarios involving strong interactions between phases. Further, the study applies the consistency conditions derived for the \rhoLENT method to geometric flux-based Volume-of-Fluid (VOF) methods. It reveals that standard computations of mass fluxes in these methods can disrupt the equivalence between scaled volume fraction equations and mass conservation equations, depending on temporal and convective term discretization schemes. The thesis proposes a dual solution approach: a consistent combination of temporal discretization and interpolation schemes, and an auxiliary mass conservation equation with a geometric calculation of face-centered densities. This approach is validated for extensive density and viscosity ratios, demonstrating its robustness and effectiveness. Additionally, the dissertation tackles non-orthogonality errors in unstructured Finite Volume methods, which can compromise force-balanced discretization in simulating incompressible two-phase flows. A novel, deterministic residual-based control of non-orthogonality correction is introduced, removing the number of non-orthogonality corrections as a global parameter from the simulation process. This method ensures force balance, particularly for surface tension and gravity forces, and is verified on polyhedral unstructured meshes with different non-orthogonality levels. Overall, this dissertation provides contributions by developing, verifying and validating advanced methodologies to accurately and efficiently simulating incompressible two-phase flows under complex conditions. These developments have improved applications in industrial multiphase microfluidics, where precise computational fluid dynamics is crucial. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
Status: | Verlagsversion | ||||
URN: | urn:nbn:de:tuda-tuprints-282663 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 500 Naturwissenschaften und Mathematik > 510 Mathematik 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau |
||||
Fachbereich(e)/-gebiet(e): | 04 Fachbereich Mathematik 04 Fachbereich Mathematik > Analysis 04 Fachbereich Mathematik > Analysis > Mathematische Modellierung und Analysis 04 Fachbereich Mathematik > Mathematische Modellierung und Analysis (MMA) |
||||
Hinterlegungsdatum: | 29 Okt 2024 13:08 | ||||
Letzte Änderung: | 05 Nov 2024 13:35 | ||||
PPN: | |||||
Referenten: | Marić, Dr.-Ing. Tomislav ; Bothe, Prof. Dr. Dieter | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 30 September 2024 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |