diplomska naloga
Rok Hamzić (Author), Matjaž Četina (Mentor), Sebastjan Bratina (Thesis defence commission member), Jože Peternelj (Thesis defence commission member), Mario Krzyk (Co-mentor)

Abstract

Dopolnitev modela PCFLOW2D s k-e modelom turbulence za nestalni tok

Keywords

gradbeništvo;diplomska dela;UNI;matematični model;PCFLOW2D;k-e model turbulence;model Smagorinsky;metoda končnih volumnov;hibridna shema;nestalni tok;numerična difuzija;PGI Visual Fortran;

Data

Language: Slovenian
Year of publishing:
Source: Ljubljana
Typology: 2.11 - Undergraduate Thesis
Organization: UL FGG - Faculty of Civil and Geodetic Engineering
Publisher: [R. Hamzić]
UDC: 532.5:626/627(043.2)
COBISS: 6064993 Link will open in a new window
Views: 1954
Downloads: 559
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Upgrade of PCFLOW2D model with k-e turbulence model for unsteady flow
Secondary abstract: This thesis presents basic mathematical models used in calculation of turbulent flows worldwide. We upgraded the PCFLOW2D program with the Smagorinsky model and the k-�à turbulence model for unsteady flow. PCFLOW2D is derived from the TEACH program and is written in the Fortran 77 programming language. PCFLOW2D is a two-dimensional depth-averaged model which calculates turbulent flow in open channels. It uses finite volume method of Patankar and Spalding to calculate partial differential equations for unsteady flow. We utilised our upgrades on the mathematical model with hybrid numerical scheme to calculate unsteady flow in a laboratory channel with a side discharge and on a kayaking track on the river So.a near Solkan. Both study cases were calculated using the model of constant effective viscosity, the Smagorinsky model and the k-�à model and the results were compared. Analysis shows that while the new model works, it has some problems with numerical diffusion. It produces physically realistic results, with insufficient length of recirculation, due to increased numerical diffusion. We attempted to solve the problem by condensing the numerical grid, but were unable to entirely remove the numerical diffusion. We also tested an up-to-date Fortran compiler, PGI Visual Fortran, that compared to the current compiler, reduced computational times by 45% on average.
Secondary keywords: mathematical model;PCFLOW2D;k-ĺ turbulence model;Smagorinsky model;finite volume method;hybrid scheme;unsteady flow;numerical diffusion;PGI Visual Fortran
File type: application/pdf
Type (COBISS): Undergraduate thesis
Thesis comment: Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages: XV, 47 str.
Type (ePrints): thesis
Title (ePrints): Upgrade of PCFLOW2D model with k-e turbulence model for unsteady flow
Keywords (ePrints): matematični model;PCFLOW2D;model turbulence;model Smagorinsky;metoda končnih volumnov;hibridna shema;nestalni tok;numerična difuzija;PGI Visual Fortran
Keywords (ePrints, secondary language): mathematical model;PCFLOW2D;k-ĺ turbulence model;Smagorinsky model;finite volume method;hybrid scheme;unsteady flow;numerical diffusion;PGI Visual Fortran
Abstract (ePrints): V diplomski nalogi so predstavljeni osnovni matematični modeli in numerične sheme, ki se uporabljajo za izračun turbulentnih tokov pri nas in po svetu. Osredotočili smo se na model Smagorinsky in k-sigma model turbulence za nestalni tok, ki smo ju tudi vgradili v program PCFLOW2D. Program PCFLOW2D je napisan v programskem jeziku Fortran 77 in izhaja iz programa TEACH. PCFLOW2D je dvodimenzionalni matematični model globinsko povprečnega toka s prosto gladino. Za izračun parcialnih diferencialnih enačb nestalnega toka program uporablja metodo končnih volumnov Patankarja in Spaldinga. Posodobitve programa smo preizkusili na primeru laboratorijskega kanala z bo.nim vtokom in primeru tok v kajaka.ki progi na So.i pri Solkanu. Oba primera smo izra.unali z modelom konstantne efektivne viskoznosti, modelom Smagorinsky in k-�Ă modelom turbulence pri hibridni numeri.ni shemi in pri laboratorijskem kanalu primerjali tokovne slike z eksperimetom. Iz analize smo ugotovili, da vgrajeni model turbulence deluje in prika.e fizikalno realne rezultate. Vendar dobimo zaradi pove.ane numeri.ne difuzije v dolo.enih predelih toka premajhno dolžino recirkulacije. Problem smo poskusili rešiti z zgoščevanjem numerične mreže, vendar numerične difuzije nismo uspeli popolnoma odpraviti. Preiskusili smo tudi novejši Fortranov prevajalnik, PGI Visual Fortran in ugotovili, da skrajša računske čase v primerjavi z dosedanjim prevajalnikom v povprečju za 45%.
Abstract (ePrints, secondary language): This thesis presents basic mathematical models used in calculation of turbulent flows worldwide. We upgraded the PCFLOW2D program with the Smagorinsky model and the k-�à turbulence model for unsteady flow. PCFLOW2D is derived from the TEACH program and is written in the Fortran 77 programming language. PCFLOW2D is a two-dimensional depth-averaged model which calculates turbulent flow in open channels. It uses finite volume method of Patankar and Spalding to calculate partial differential equations for unsteady flow. We utilised our upgrades on the mathematical model with hybrid numerical scheme to calculate unsteady flow in a laboratory channel with a side discharge and on a kayaking track on the river So.a near Solkan. Both study cases were calculated using the model of constant effective viscosity, the Smagorinsky model and the k-�à model and the results were compared. Analysis shows that while the new model works, it has some problems with numerical diffusion. It produces physically realistic results, with insufficient length of recirculation, due to increased numerical diffusion. We attempted to solve the problem by condensing the numerical grid, but were unable to entirely remove the numerical diffusion. We also tested an up-to-date Fortran compiler, PGI Visual Fortran, that compared to the current compiler, reduced computational times by 45% on average.
Keywords (ePrints, secondary language): mathematical model;PCFLOW2D;k-ĺ turbulence model;Smagorinsky model;finite volume method;hybrid scheme;unsteady flow;numerical diffusion;PGI Visual Fortran
ID: 8312625