
[Sponsors] 
June 2, 2016, 22:12 
Compressible vs incompressible SGS model!

#1 
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
Hi everyone,
I want to simulate combustion in Internal Combustion Engine with LES. Now I am trying to implement some SGS models to my code. But I found many SGS models are suited for incompressible problems. Fortunately, I found several compressible SGS models: Code:
compressible SGS models: compressible Smagorinsky: [1] Yoshizawa A. Statistical theory for compressible turbulent shear flows, with the application to subgrid modeling[J]. Physics of Fluids(19581988), 1986, 29(7):21522164. dynamic(Germano)compressible Smagorinsky: [2] Moin P, Squires K, Cabot W, et al. A dynamic subgridscale model for compressible turbulence and scalar transport[J]. Physics of Fluids A: Fluid Dynamics, 1991, 3(11):2746. DOI: 10.1063/1.858164. dynamic(Lilly)compressible Smagorinsky: [3] Martνn M P, Piomelli U, Candler G V. SubgridScale Models for Compressible LargeEddy Simulations[J]. Theoretical & Computational Fluid Dynamics, 2000, 13(5):361376. compressible problems. Unfortunately, most of SGS models are incompressible version: Code:
incompressible SGS models: Smagorinsky: [1] Smagorinsky J. General circulation experiments with the primitive equations: I. the basic experiment*[J]. Monthly weather review, 1963, 91(3):99164. Keqn: [2] Yoshizawa A, Horiuti K. A statisticallyderived subgridscale kinetic energy model for the largeeddy simulation of turbulent flows[J]. Journal of the Physical Society of Japan, 1985, 54(54):28342839. dynamic(Germano) Smagorinsky: [3] Germano M, Piomelli U, Moin P, et al. A dynamic subgridscale eddy viscosity model[J]. Physics of Fluids A: Fluid Dynamics (19891993), 1991, 3(7):17601765. dynamic(Lilly) Smagorinsky: [4] Lilly D K. A proposed modification of the Germano subgrid‐scale closure method[J]. Physics of Fluids A Fluid Dynamics, 1992, 4(4):633. dynamicKeqn: [5] Kim WW, Menon S. A new dynamic oneequation subgridscale model for large eddy simulations[J]. AIAA, Aerospace Sciences Meeting and Exhibit, 33 rd, Reno, NV, 1995. dynamicLagrangian: [6] Meneveau C, Lund T S, Cabot W H. A Lagrangian dynamic subgridscale model of turbulence[J]. Journal of Fluid Mechanics, 1996, 319:353385. I don't know: [7] Menon S, Yeung P K, Kim W W. Effect of subgrid models on the computed interscale energy transfer in isotropic turbulence[J]. Computers & Fluids, 1996, 25(2):165180. WALE: [8] Nicoud F, Ducros F. Subgridscale stress modelling based on the square of the velocity gradient tensor[J]. Flow, turbulence and Combustion, 1999, 62(3):183200. dynamic structure1: [9] Pomraning E, Rutland C J. Dynamic oneequation nonviscosity largeeddy simulation model[J]. AIAA journal, 2002, 40(4):689701. dynamic structure2: [10] Chumakov S G, Rutl C J. Dynamic structure subgrid‐scale models for large eddy simulation[J]. International Journal for Numerical Methods in Fluids, 2005, 47(8‐9):911923. I don't know whether these incompressible models can be used to simulate compressible problems. 

June 3, 2016, 04:43 

#2 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,551
Rep Power: 58 
One of the main difference is in the isotropic term related to div v, for compressible flows you need to supply a model for both isotropic and deviatoric terms.
Have a look here http://www.springer.com/us/book/9789048128181 

June 4, 2016, 03:42 

#3  
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
Quote:
I am a firstyear graduate student， and I am just start my research. After reading some papers, I found many people simulate combustion in the internal combustion engine with an incompressible SGS model. I don't know why. I think a compressible SGS model is necessary in such simulation. 

June 4, 2016, 04:08 

#4 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,551
Rep Power: 58 
it depends on the choice of a lowMach model


June 4, 2016, 04:17 

#5 
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
But I never heard Mach number in the cylinder of engine, except for air intake port of engine.
I think the compressibility in the engine is related to the reaction. Reaction causes the big variation of density. As far as I know, we didn't mention Mach number in the cylinder combustion simulation. So, which kind of SGS models should I choose? Maybe incompressilbe model is OK. But compressible model is better. Do I understand this correctly? 

June 4, 2016, 04:31 

#6  
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,551
Rep Power: 58 
Quote:
well, no...for the flow in the cylinder you have a fully compressible case... if you search in google you will find a lot of papers. Try with "LES" and "incylinder" words 

June 4, 2016, 04:43 

#7  
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
Quote:
Such as: Code:
[1] Jhavar R, Rutland C J. Using Large Eddy Simulations to Study Mixing Effects in Early Injection Diesel Engine Combustion[J]. 2006. DOI: 10.4271/2006010871. [2] Som S, Senecal P K, Laboratory A N, et al. Comparison of RANS and LES Turbulence Models against Constant Volume Diesel Experiments[J]. 2012. [3] Som S, Longman D E, Luo Z, et al. Simulating Flame LiftOff Characteristics of Diesel and Biodiesel Fuels Using Detailed ChemicalKinetic Mechanisms and Large Eddy Simulation Turbulence Model[J]. Journal of Energy Resources Technology, 2012, 134(3):189206. Now I'm quite confused! 

June 4, 2016, 04:51 

#8 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,551
Rep Power: 58 
Some studies are related to the valve flowing, without any compression/expansion phase. In such case you can adopt the incompressible model. But if you consider the movement of the piston the compressible effects are relevant.


June 4, 2016, 05:17 

#9  
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
Quote:
And, Am I correct now ? 

June 4, 2016, 05:27 

#10 
Senior Member
Filippo Maria Denaro
Join Date: Jul 2010
Posts: 5,551
Rep Power: 58 
If you have combustion I suppose that the SGS model enters into the other equations... for lowMach model the density variation term can be coupled.
I never worked personally on LES of combustion I suggest following the literature (have a look to the book of Sagaut for reference), your flow problem is often found in the SAE literature 

June 4, 2016, 05:39 

#11  
Senior Member
Yan Zhang
Join Date: May 2014
Posts: 116
Rep Power: 8 
Quote:
I have bought that book recently. And I will read it carefully. 

Tags 
compressible, incompressible, large eddy simulation, les, sgs 
Thread Tools  Search this Thread 
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Turbulence kinetic energy not defined for incompressible SA model  FernandoSoares  OpenFOAM Running, Solving & CFD  7  October 24, 2019 12:07 
Superlinear speedup in OpenFOAM 13  msrinath80  OpenFOAM Running, Solving & CFD  18  March 3, 2015 06:36 
Compressible turbulence model issues  351Cleveland  OpenFOAM  5  October 24, 2013 16:41 
Reynolds Stress Model incompressible  sven  OpenFOAM  3  August 29, 2012 14:13 
strange behaviour of the oneequation SGS model  cfdmarkus  OpenFOAM Running, Solving & CFD  18  August 14, 2009 06:33 