Reynolds-averaged modeling of turbulence damping near a large-scale interface in two-phase flow

Frederix, E.M.A.; Mathur, Akshat; Dovizio, D.; Geurts, Bernardus J.; Komen, E.M.J.

doi:10.1016/j.nucengdes.2018.04.010

Cited by 34 publications

(36 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Even though the k-ω SST model was used for the turbulence modeling in the present study, the proposed length scale could be used for all other ω-based turbulence models as well. In addition, Frederix et al [12] developed a methodology to implement Egorov's model into -based turbulence models, which makes it straightforward to implement Egorov's model together with ∆n = y p in all the -based models.…”

Section: Notes On Y Pmentioning

confidence: 99%

“…As a result, many values, e.g., 10 [21], 20 [16], 100 [8], 500 [17], and 2500 [10], are used or recommended. Few attempts [12,23] have been made with the hope of finding a large enough value for B that is mesh size independent. The following discussion aims at showing that there is no large enough value for B.…”

Section: There Is No Large Enough Value For Bmentioning

confidence: 99%

“…For instance, according to both experimental studies [4,5] and direct numerical simulations (DNS) [6,7], the gas-liquid interface in stratified flows behaves similarly to a solid wall in single-phase flows. In order to reproduce this wall-like behavior in CFD simulations where turbulence models are used, researchers have made various attempts trying to damp the turbulence around the interface such that the interface could behave more like a wall [8][9][10][11][12]. Even though all these methods are essentially phenomenological models, they are being used and will still be used in the foreseen future.…”

Section: Introductionmentioning

confidence: 99%

“…On the one hand, a proper value should be selected for the model parameter, B, to match the numerical results with the experimental data. However, B is mesh size sensitive, and no universal guideline exits for the selection of B [8,12,16,17]. On the other hand, Egorov's model employs the same treatment for both gas and liquid phases, which is referred to as the symmetric treatment by Frederix et al [12].…”

Section: Introductionmentioning

confidence: 99%

“…However, B is mesh size sensitive, and no universal guideline exits for the selection of B [8,12,16,17]. On the other hand, Egorov's model employs the same treatment for both gas and liquid phases, which is referred to as the symmetric treatment by Frederix et al [12]. It has been pointed out by many researchers [10,12,18] that an asymmetric treatment should be introduced to model the different turbulence behaviors on the different sides of the interface.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Progress in Phenomenological Modeling of Turbulence Damping around a Two-Phase Interface

Fan

Anglart

2019

Fluids

View full text Add to dashboard Cite

The presence of a moving interface in two-phase flows challenges the accurate computational fluid dynamics (CFD) modeling, especially when the flow is turbulent. For such flows, single-phase-based turbulence models are usually used for the turbulence modeling together with certain modifications including the turbulence damping around the interface. Due to the insufficient understanding of the damping mechanism, the phenomenological modeling approach is always used. Egorov’s model is the most widely-used turbulence damping model due to its simple formulation and implementation. However, the original Egorov model suffers from the mesh size dependency issue and uses a questionable symmetric treatment for both liquid and gas phases. By introducing more physics, this paper introduces a new length scale for Egorov’s model, making it independent of mesh sizes in the tangential direction of the interface. An asymmetric treatment is also developed, which leads to more physical predictions for both the turbulent kinetic energy and the velocity field.

show abstract

Section: Notes On Y Pmentioning

confidence: 99%

Section: There Is No Large Enough Value For Bmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Progress in Phenomenological Modeling of Turbulence Damping around a Two-Phase Interface

Fan

Anglart

2019

Fluids

View full text Add to dashboard Cite

show abstract

Experimental and numerical study of stratified viscous oil–water flow

et al. 2021

View full text Add to dashboard Cite

Stratified two-phase flows of oil and water are important to the energy industry, and models capable of predicting this type of flow are primordial. Many studies focus on fluids with low viscosity, but a high viscosity oil in the mixture significantly changes its behavior. We gathered experimental data of pressure drop, volumetric fractions, and flow-pattern data of a stratified liquid-liquid flow with high viscosity ratio. In addition, a wire-mesh sensor provided tomographic views of the flow. The data were compared with computational fluid dynamics (CFD) models using OpenFOAM and a one-dimensional model. CFD simulations used an interface capturing method, and turbulence damping was introduced to avoid high eddy viscosity at the interface region. Reynolds Average Navier-Stokes and large eddy simulations were used to account for turbulence, and they showed significant differences. The comparisons showed good overall results for pressure drop, volumetric fractions, and phase distributions between CFD and experiments.

show abstract