Positivity Preservation and Adaptive Solution for the k-? Model of Turbulence

Ilinca, F.; Pelletier, Dominique

doi:10.2514/2.350

Cited by 98 publications

(53 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence the eddy viscosity T and the eddy conductivity T will always remain positive. Moreover, solutions from logarithms are more accurate because the fields of the logarithmic variables present smoother variations than those of k and [19]. …”

Section: Turbulence Modellingmentioning

confidence: 99%

See 1 more Smart Citation

Numerical simulation of the galvanizing process during GA to GI transition

Ilinca

Ajersch

Baril³

et al. 2006

Numerical Methods in Fluids

View full text Add to dashboard Cite

This paper presents the application of a three‐dimensional finite element solution algorithm for the prediction of the velocity, temperature and species concentration fields in an industrial continuous galvanizing bath. Simulations were carried out using a parallel CFD software developed at IMI‐NRC. The turbulent flow, heat and mass transfer has been solved using a high Reynolds number k–ε model. Simulations were carried out for the case when the density of the molten metal depends only on the temperature and also for the case when both temperature and Al concentration affect the density. When considering the buoyancy effect of the Al concentration, differences are especially apparent during the melting of ingots with high Al content. Otherwise, thermal effects are dominant. The continuous monitoring of the temperature and the Al and Fe content in an industrial bath was used to validate the flow, temperature and compositional variations. A period of three hours, corresponding to three different ingot additions, was simulated successfully, resulting in a good agreement of the temperature and compositional variations. Copyright © 2006 Crown in the right of Canada. Published by John Wiley & Sons, Ltd.

show abstract

Section: Turbulence Modellingmentioning

confidence: 99%

“…Turbulence equations are solved for the logarithms of turbulence variables [19,20]. This change of dependent variables guarantees that k and will remain positive throughout the computations.…”

Section: Turbulence Modellingmentioning

confidence: 99%

Numerical simulation of the galvanizing process during GA to GI transition

Ilinca

Ajersch

Baril³

et al. 2006

Numerical Methods in Fluids

View full text Add to dashboard Cite

show abstract

“…This ensures, for a Reynolds number of one million, that the non-dimensional wall distance in viscous unit, y þ , (see Eq. (11)) lies in [30,300] on all the wall while being as close as possible of the lower limit.…”

Section: Definition Of the Manufactured Solutionmentioning

confidence: 99%

“…To preserve positivity of the dependent variables, we work with the logarithmic form of these equations [30]. This can be viewed as using the following change of dependent variables:…”

Section: Governing Equationsmentioning

confidence: 99%

Verified predictions of shape sensitivities in wall-bounded turbulent flows by an adaptive finite-element method

Hay

Pelletier

Caro

2009

Journal of Computational Physics

View full text Add to dashboard Cite

“…The constants s k , s e , C el , C e2 and C m are as follow [15] s k = 1.0, s e = 1.3, C el = 1.44, C e2 = 1.92 and C m = 0.09 Increased robustness of the solution algorithm is obtained by solving the turbulence equations for the logarithms of turbulence variables [16]. In such a way, the positivity of turbulence variables is enforced resulting in increased accuracy and a faster solution approach.…”

Section: Process Simulationmentioning

confidence: 99%

Numerical Simulation of Flow, Temperature and Composition Variations in a Galvanizing Bath

Ajersch¹,

Ilinca²,

Hétu³

et al. 2005

Canadian Metallurgical Quarterly

View full text Add to dashboard Cite

-The modern hot dip galvanizing operation is a complex process subject to a number of configurational, physical, chemical and kinetic parameters. Small decreases in temperature can precipitate intermetallic dross particles, which can be entrained in the flow towards the strip leading to surface imperfections. The numerical simulations carried out in this study clearly define the spatial distribution of velocity, temperature and compositional variation in the bath. The modeling of the transient effects during ingot melting and non-melting periods have also identified the critical periods and zones within the bath where dross formation can occur within an operating cycle.Résumé -L'opération moderne de galvanisation à chaud est un procédé complexe exposé à plusieurs paramètres configurationnels, physiques, chimiques et cinétiques. De petites diminutions de température peuvent précipiter des particules intermétalliques de scories, lesquelles peuvent être entraînées dans l'écoulement vers la bande, conduisant à des imperfections de surface. Les simulations numériques effectuées dans cette étude définissent clairement la distribution spatiale de la vitesse, de température et de la variation de composition dans le bain. La modélisation des effets transitoires lors des périodes de fonte et d'absence de fonte du lingot a également identifié les périodes critiques ainsi que les zones dans le bain où la formation de scorie peut se produire à l'intérieur d'un cycle d'opération.

show abstract

Positivity Preservation and Adaptive Solution for the k-? Model of Turbulence

Cited by 98 publications

References 17 publications

Numerical simulation of the galvanizing process during GA to GI transition

Numerical simulation of the galvanizing process during GA to GI transition

Verified predictions of shape sensitivities in wall-bounded turbulent flows by an adaptive finite-element method

Numerical Simulation of Flow, Temperature and Composition Variations in a Galvanizing Bath

Contact Info

Product

Resources

About