synthetic turbulence fluctuation in LES Modeling

SEB12129 · October 9, 2015, 09:55

Hi,

I am doing a validation test to model a fully turbulent channel flow to see if the bottom of the channel which is a riblet surface has some drag reduction performance compared to the top which is a flat plate. The validation test results are compared to two papers "Large eddy simulation of turbulent drag reduction by V-shaped riblets" by (Mohsen Jahanmiri, 2015) and "Towards large eddy simulations of turbulence drag reduction using sinusoidal riblets" by (Yulia Peet, 2007). In both of these papers, they refer to initialize computation by a certain model (k-omega or laminar), and then switch to LES model by adding synthetic turbulence fluctuations to the solutions. I think it is kind of stange to switch a RANS or laminar model into LES model since I have never met similar methods before, so I was wondering what might be the reason of that, how to do this in Star-ccm+ and how to determine the value of synthetic turbulence fluctuations to be added in the solution.

I will really appreciate if anyone could give some advice.

Yansheng

SEB12129 · October 9, 2015, 12:18

In addition, the physical condition is 3d, liquid, implicit unsteady, segregated flow, turbulent, constant density, Large Eddy Simullation, WALE subgrid scale, All y+ wall treatment. And the number of grid is 30*128*128 in streamwise, spanwise and vertical direction.

October 9, 2015, 09:55	synthetic turbulence fluctuation in LES Modeling	#1
SEB12129 New Member Yansheng Zhang Join Date: Sep 2015 Posts: 3 Rep Power: 10	Hi, I am doing a validation test to model a fully turbulent channel flow to see if the bottom of the channel which is a riblet surface has some drag reduction performance compared to the top which is a flat plate. The validation test results are compared to two papers "Large eddy simulation of turbulent drag reduction by V-shaped riblets" by (Mohsen Jahanmiri, 2015) and "Towards large eddy simulations of turbulence drag reduction using sinusoidal riblets" by (Yulia Peet, 2007). In both of these papers, they refer to initialize computation by a certain model (k-omega or laminar), and then switch to LES model by adding synthetic turbulence fluctuations to the solutions. I think it is kind of stange to switch a RANS or laminar model into LES model since I have never met similar methods before, so I was wondering what might be the reason of that, how to do this in Star-ccm+ and how to determine the value of synthetic turbulence fluctuations to be added in the solution. I will really appreciate if anyone could give some advice. Yansheng Last edited by SEB12129; October 9, 2015 at 12:19. Reason: mesh

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
LES turbulence model: how to write delta	Sylv	OpenFOAM Programming & Development	3	January 14, 2018 17:19
LES about BC cyclic and turbulence	tzqfly	OpenFOAM Running, Solving & CFD	8	October 11, 2014 08:20
LES turbulence decaying in channel flow	cfdIsMad	Main CFD Forum	6	August 21, 2009 12:17
Naive question about turbulence and LES	Renato	Main CFD Forum	1	October 4, 2005 23:01
turbulence modeling questions	llowen	Main CFD Forum	3	September 11, 1998 04:24

October 9, 2015, 12:18		#2
SEB12129 New Member Yansheng Zhang Join Date: Sep 2015 Posts: 3 Rep Power: 10	In addition, the physical condition is 3d, liquid, implicit unsteady, segregated flow, turbulent, constant density, Large Eddy Simullation, WALE subgrid scale, All y+ wall treatment. And the number of grid is 30128128 in streamwise, spanwise and vertical direction.