LES Open bounded jet - Smagorinsky or one equation KE?

diogof · December 28, 2016, 11:47

Hey everyone,

I'm currently working on a FLUENT assignment which involves the use of LES in order to simulate a turbulent pulsating jet. The physical problem consists of a jet with a rectangular base 0.1x0.02m, at the base of a 2x4x0.02m domain, where each of the 0.02m sides is meant to be modelled as a symmetry boundary, with two grid points along the 0.02m side. The pulse is sine wave which is defined as a function of time in an UDF. The side walls and top wall are open, whereas the rest of the bottom base is a wall.

I am currently having some difficulty trying to choose which subgrid-scale model to use.

I think I understand that the default WALE model, as well as the WMLES models are probably not the best models to use in this case because they are tuned for wall-bounded flows. I think the best two options are either the dynamic Smagorinsky-Lilly (eddy viscosity) model or the Kinetic Energy Transport model.

When looking at papers online however, I can't find any reasons to choose one model over the other, and so I am stumped when trying to justify which model I have used and why. Do you have any recommendations?

Thank you in advance!

LuckyTran · January 3, 2017, 18:11

I agree, ignore the WALE and WMLES stuff since they are not appropriate.

The original Smagorinsky-Lily and dynamic Smagorinsky SGS models are algebraic. The kinetic energy transport solves a transport equation for the subgrid scale kinetic energy. In this regard, I think the kinetic energy transport model is vastly superior (in theory). It's similar to using Prandtl's mixing length as a turbulence model versus the one-equation Spalart-Allmaras model. How well this theory is implemented and how accurately the numerical results compare with real world is another story.

Furthermore, I wouldn't expect much difference between original Smagorinsky versus Dynamic Smagorinsky. They should be similar.

diogof · January 13, 2017, 14:36

Thanks for your advice! For most of the results I've seen on papers, when comparing Smagorinsky and kinetic energy transport model to real-world data they both seem to be pretty close. I think I can justify using either one.

December 28, 2016, 11:47	LES Open bounded jet - Smagorinsky or one equation KE?	#1
diogof New Member Join Date: Dec 2016 Posts: 3 Rep Power: 9	Hey everyone, I'm currently working on a FLUENT assignment which involves the use of LES in order to simulate a turbulent pulsating jet. The physical problem consists of a jet with a rectangular base 0.1x0.02m, at the base of a 2x4x0.02m domain, where each of the 0.02m sides is meant to be modelled as a symmetry boundary, with two grid points along the 0.02m side. The pulse is sine wave which is defined as a function of time in an UDF. The side walls and top wall are open, whereas the rest of the bottom base is a wall. I am currently having some difficulty trying to choose which subgrid-scale model to use. I think I understand that the default WALE model, as well as the WMLES models are probably not the best models to use in this case because they are tuned for wall-bounded flows. I think the best two options are either the dynamic Smagorinsky-Lilly (eddy viscosity) model or the Kinetic Energy Transport model. When looking at papers online however, I can't find any reasons to choose one model over the other, and so I am stumped when trying to justify which model I have used and why. Do you have any recommendations? Thank you in advance!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Custom Thermophysical Properties	wsmith02	OpenFOAM	4	June 1, 2023 14:30
Problem installing on Ubuntu 9.10 -> 'Cannot open : No such file or directory'	mfiandor	OpenFOAM Installation	2	January 25, 2010 09:50
Version 15 on Mac OS X	gschaider	OpenFOAM Installation	113	December 2, 2009 10:23
DxFoam reader update	hjasak	OpenFOAM Post-Processing	69	April 24, 2008 01:24
OpenFOAM with IBM AIX	matthias	OpenFOAM Installation	20	March 25, 2008 02:36

January 3, 2017, 18:11		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,673 Rep Power: 65	I agree, ignore the WALE and WMLES stuff since they are not appropriate. The original Smagorinsky-Lily and dynamic Smagorinsky SGS models are algebraic. The kinetic energy transport solves a transport equation for the subgrid scale kinetic energy. In this regard, I think the kinetic energy transport model is vastly superior (in theory). It's similar to using Prandtl's mixing length as a turbulence model versus the one-equation Spalart-Allmaras model. How well this theory is implemented and how accurately the numerical results compare with real world is another story. Furthermore, I wouldn't expect much difference between original Smagorinsky versus Dynamic Smagorinsky. They should be similar.

January 13, 2017, 14:36		#3
diogof New Member Join Date: Dec 2016 Posts: 3 Rep Power: 9	Thanks for your advice! For most of the results I've seen on papers, when comparing Smagorinsky and kinetic energy transport model to real-world data they both seem to be pretty close. I think I can justify using either one.