A dynamic flame surface density model for large eddy simulation of turbulent premixed combustion

Knikker, Ronnie; Veynante, Denis; Meneveau, Charles

doi:10.1063/1.1780549

Cited by 80 publications

(82 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where η i is the inner cut-off scale and D f is the fractal dimension of the flame surface [29][30][31][32][33][34]. Due to the close relation between gen = (N c /D) 1/2 and SDR, the analogy of power-law closure for FSD has been extended in previous analyses [6,7] for possible modelling of SDRÑ c :…”

Section: Mathematical Backgroundmentioning

confidence: 99%

“…Dynamic evaluation of power-law exponents were successfully used for the generalised Flame Surface Density (FSD) (i.e. gen = |∇c|) closure in the past [29,31] and given the close relation between FSD and SDR (i.e. gen = (N c /D) 1/2 ) it is worthwhile to consider if the dynamic evaluation of power-law exponent α D could lead to a satisfactory prediction ofÑ c .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Closure of Scalar Dissipation Rate for Large Eddy Simulations of Turbulent Premixed Combustion: A Direct Numerical Simulations Analysis

Gao

Chakraborty

Swaminathan

2015

Flow Turbulence Combust

View full text Add to dashboard Cite

Dynamic algebraic closure of scalar dissipation rate (SDR) of reaction progress variable in the context of Large Eddy Simulations (LES) of turbulent premixed combustion has been addressed here using a power-law based expression and a model, which was originally proposed for Reynolds Averaged Navier Stokes (RANS) simulations, but has recently been extended for LES. The performances of these models have been assessed based on a-priori analysis of a Direct Numerical Simulations (DNS) database of statistically planar turbulent premixed flames with a range of different values of heat release parameter τ , turbulent Reynolds number Re t and global Lewis number Le. It has been found that the power-law model with a single constant exponent α D does not adequately capture the volume-averaged behaviour of density-weighted SDR and this problem is particularly severe especially for Le << 1 flames. The deficiency of the power-law model with a single power-law exponent arises due to multi-fractal nature of SDR. The dynamic evaluation of the model parameter for the algebraic model, which was originally proposed in the context of RANS and has been extended here for LES, has been shown to capture the local behaviour of SDR better than the power-law model. It has been demonstrated that the empirical parameterisation of a model parameter for the static version of the RANS-extended SDR model can be avoided using a dynamic formulation which captures the local behaviour of SDR either comparably or better than the static formulation for a range of different values of τ, Le and Re t , without sacrificing the prediction of the volume-averaged SDR. Keywords

show abstract

Section: Mathematical Backgroundmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dynamic Closure of Scalar Dissipation Rate for Large Eddy Simulations of Turbulent Premixed Combustion: A Direct Numerical Simulations Analysis

Gao

Chakraborty

Swaminathan

2015

Flow Turbulence Combust

View full text Add to dashboard Cite

show abstract

“…In LES, filter width, is generally considered as outer cut-off scale and for inner cut-off scale there are several expressions available in the literature related to Gibson scale or Kolmogorov scale or laminar flame thickness. In the present study, we considered inner cut-off scale as three times of the laminar flame thickness following Knikker et al [2004]. Hence, is calculated as:…”

Section: Modelling Using Wrinkling Flame Factormentioning

confidence: 99%

“…Defining the flame wrinkling factor Ξ in the equation (3), as a ratio of flame surface density to its projection in the normal direction of the flame propagation [Knikker et al 2004] and identifying flame surface as a fractal surface between inner and outer cut-off scales leads to:…”

Section: Modelling Using Wrinkling Flame Factormentioning

confidence: 99%

A Dynamic SGS Model for Les of Turbulent Premixed Flames

Gubba

Ibrahim

Malalasekera

2008

International Symposium on Advances in Computational Heat Transfer

View full text Add to dashboard Cite

Numerical simulations are carried out for transient turbulent premixed flames using large eddy simulations (LES) technique. The sub-grid scale (SGS) mean chemical reaction rate is calculated using a simple algebraic relation for flame surface density (FSD), with a dynamic formulation for the model coefficient. The dynamic model is derived based on fractal theory and a flame wrinkling factor and implemented in compressible LES code. The developed model is used to simulate turbulent premixed flames of stoichiometric propane/air mixture in a vented combustion chamber, propagating over built-in solid obstacles. The fractal dimension is dynamically calculated by viewing the flame front as a fractal surface, using the SGS velocity fluctuations and the strained laminar burning velocity. The model predictions are validated against experimental measurements where good agreements are obtained.

show abstract

“…These approaches can be broadly categorized into two classes, namely, flamelets and non-flamelets or geometrical and statistical (Gicquel et al, 2012). The geometrical category of flamelets includes thickened flame (Colin et al, 2000;De and Acharya, 2009), flame surface density or flame-wrinkling (see, for example, Boger et al 1998;Chakraborty and Cant 2007;Chakraborty and Klein 2008;Gubba et al, 2012;Hawkes and Cant 2001;Knikker et al, 2004;Wang et al, 2012), and level-set or G equation (Moureau et al, 2008;Pitsch, 2005). The statistical category of flamelets includes approaches, such as EBU (eddy-break-up model), algebraic closure involving scalar dissipation rate (Butz et al, 2015;Dunstan et al, 2013;Gao et al, 2014;Langella et al, 2015;Ma et al, 2014), and presumed probability density function (PDF) methodology with laminar flamelets.…”

Section: Introductionmentioning

confidence: 99%

Large Eddy Simulation of Premixed Combustion: Sensitivity to Subgrid Scale Velocity Modeling

Langella

Swaminathan

Gao

et al. 2016

Combustion Science and Technology

View full text Add to dashboard Cite

An algebraic reaction rate closure involving filtered scalar dissipation rate of reaction progress variable is studied. The filtered scalar dissipation rate closure requires a model for sub-grid scale velocity, u 0 Δ , which is estimated using four algebraic models and transported subgrid scale kinetic energy. A priori analyses using direct numerical simulation (DNS) data show that the filtered dissipation rate, and thus the reaction rate closure, has some sensitivity to the u 0 Δ model. The sensitivity of various statistics obtained from large eddy simulation (LES) of three piloted Bunsen flames of stoichiometric methaneair mixture to the modeling of u 0 Δ is observed to be weaker compared to that for the DNS analysis. Moreover, analysis using transported sub-grid scale kinetic energy does not indicate a necessity to include flame-generated turbulence in the modeling of u 0 Δ for the Bunsen flames in the thin reaction zones regime. The measured and computed flame brush structures are compared and studied and the algebraic closure for the filtered reaction rate is found to be quite good. ARTICLE HISTORY

show abstract

A dynamic flame surface density model for large eddy simulation of turbulent premixed combustion

Cited by 80 publications

References 15 publications

Dynamic Closure of Scalar Dissipation Rate for Large Eddy Simulations of Turbulent Premixed Combustion: A Direct Numerical Simulations Analysis

Dynamic Closure of Scalar Dissipation Rate for Large Eddy Simulations of Turbulent Premixed Combustion: A Direct Numerical Simulations Analysis

A Dynamic SGS Model for Les of Turbulent Premixed Flames

Large Eddy Simulation of Premixed Combustion: Sensitivity to Subgrid Scale Velocity Modeling

Contact Info

Product

Resources

About