turbulent kinetic energy budget terms with openfoam

Thangam · August 1, 2014, 07:03

Hi,

Is there a way to write out all the components(production, convection, dissipation and turbulent diffusion) of the tke budget separately for a LES case? Any pointers??

Thanks.

Thangam.

wyldckat · August 13, 2014, 16:06

Greetings Thangam,

Can you provide more information about each variable you are looking to export?
I ask this because even though I'm familiar with many of OpenFOAM's way of working with things, on the other hand I'm not familiar with the terminology in LES. Therefore I need more details on what exactly you're looking for.

Best regards,
Bruno

PS: Re-posting does not make it easier for people to answer your questions. What's more important is that more details are provided with each new post you provide, because if people did not answer your question, it's likely that not enough information was provided in the first place. And many of the people that might be able to answer your question, don't have the time to still have to ask first: What on Earth are you talking about?

Thangam · August 14, 2014, 01:55

My bad I thought these were quite obvious! Anyway,

The complete form of the turbulent kinetic energy can be written as,

$P_k - C_k - \epsilon_k + D_k = 0$ ,

where,

is the production term of

given by,

$P_k = - \overline{u\prime_i u\prime_m} \frac{\partial U_i}{\partial x_m}$

is the mean convection term of

given by,

$C_k = U_m \frac{\partial k}{\partial x_m}$

$\epsilon_k$ is the dissipation term of

given by,

$\epsilon_k = \nu \overline{ \frac{\partial u\prime_i}{\partial x_m}\frac{\partial u\prime_i}{\partial x_m}}$

and

is the diffusion transport term of

given by,

$D_k = \frac{1}{2} \frac{\partial}{\partial x_m}( T_{iim}^{(\nu)} - T_{iim}^{(p\prime)} - T_{iim}^{(u\prime)} )$ and

contains

the viscous diffusion,

$T_{iim}^{(\nu)} = - \nu \overline{\frac{\partial u_i\prime u_i\prime}{\partial x_m}}$

pressure diffusion,

$T_{iim}^{(p \prime)} = \frac{2}{\rho}\overline{u_i \prime p\prime \delta_{im}}$

and the turbulent diffusion given by,

$T_{iim}^{(u \prime)} = \overline{u_i u_i u_m}$

The problem is I am not able to identify where these terms are embedded in the code and how to write them out by creating a small utility.

Thanks.

dkxls · August 14, 2014, 05:56

Quote:

Originally Posted by Thangam

The problem is I am not able to identify where these terms are embedded in the code and how to write them out by creating a small utility.

You are familiar with the source code of the one equation model?

Code:

src/turbulenceModels/compressible/LES/oneEqEddy/oneEqEddy.H
src/turbulenceModels/compressible/LES/oneEqEddy/oneEqEddy.C

wyldckat · August 17, 2014, 15:00

Greetings to all!

To complement Armin's answer: the class "oneEqEddy" is derived from the class "GenEddyVisc", which in turn derives from "LESModel". This last class is the base class for accessing turbulence data that is readily available on the solver's side.
For example, have a look at the solver "applications/solvers/heatTransfer/thermoFoam" in OpenFOAM 2.3, namely in the file "setAlphaEff.H". There you can see how it explicitly loads the turbulence of type "LESModel".

Correlating the information given in #3 with the information here on the wiki, it looks like the terms you've looking for are these: http://www.cfd-online.com/Wiki/Kinet...id-scale_model
Checking Armin's answer in post #4 does reveal the same equations in this method:

Code:

void oneEqEddy::correct(const tmp<volTensorField>& tgradU)

All of the terms of the equation are all solved directly in a single equation and are not available for outside use.

Which means that you will have to either create a new derived class from "oneEqEddy" or modify directly OpenFOAM's class "oneEqEddy", to write out each term of the equation.
Or you can modify the solver and do a dynamically cast to the instance of "LESModel" class to "oneEqEddy" and then access directly the "k()" method and other methods that give the components used in the equation that is solved in "oneEqEddy::correct".

Although I don't know how practical that is, because each term of the equation is in the form of "fvScalarMatrix" and not of in the form of a standard field.

Best regards,
Bruno

syavash · September 5, 2015, 18:58

Quote:

Originally Posted by wyldckat

Greetings to all!

To complement Armin's answer: the class "oneEqEddy" is derived from the class "GenEddyVisc", which in turn derives from "LESModel". This last class is the base class for accessing turbulence data that is readily available on the solver's side.
For example, have a look at the solver "applications/solvers/heatTransfer/thermoFoam" in OpenFOAM 2.3, namely in the file "setAlphaEff.H". There you can see how it explicitly loads the turbulence of type "LESModel".

Correlating the information given in #3 with the information here on the wiki, it looks like the terms you've looking for are these: http://www.cfd-online.com/Wiki/Kinet...id-scale_model
Checking Armin's answer in post #4 does reveal the same equations in this method:

Code:

void oneEqEddy::correct(const tmp<volTensorField>& tgradU)

All of the terms of the equation are all solved directly in a single equation and are not available for outside use.

Which means that you will have to either create a new derived class from "oneEqEddy" or modify directly OpenFOAM's class "oneEqEddy", to write out each term of the equation.
Or you can modify the solver and do a dynamically cast to the instance of "LESModel" class to "oneEqEddy" and then access directly the "k()" method and other methods that give the components used in the equation that is solved in "oneEqEddy::correct".

Although I don't know how practical that is, because each term of the equation is in the form of "fvScalarMatrix" and not of in the form of a standard field.

Best regards,
Bruno

I had a similar experience with Tecplot in extracting turbulent production terms from Reynolds stress tensor and mean velocity gradients.
I could effectively use Tecplot "Calculate Variable" option to construct each term of the production and dissipation.
I think paraFoam can perform such calculations as well.

Regards,
Syavash

syavash · July 15, 2016, 10:21

For anyone interested in this subject, I have created a new thread.

http://www.cfd-online.com/Forums/ope...e-budgets.html

Regards,
Syavash

August 1, 2014, 07:03	turbulent kinetic energy budget terms with openfoam	#1
Thangam Member Thangam Natarajan Join Date: Dec 2010 Location: Perth Posts: 70 Rep Power: 16	Hi, Is there a way to write out all the components(production, convection, dissipation and turbulent diffusion) of the tke budget separately for a LES case? Any pointers?? Thanks. Thangam.

August 13, 2014, 16:06		#2
wyldckat Retired Super Moderator Bruno Santos Join Date: Mar 2009 Location: Lisbon, Portugal Posts: 10,982 Blog Entries: 45 Rep Power: 128	Greetings Thangam, Can you provide more information about each variable you are looking to export? I ask this because even though I'm familiar with many of OpenFOAM's way of working with things, on the other hand I'm not familiar with the terminology in LES. Therefore I need more details on what exactly you're looking for. Best regards, Bruno PS: Re-posting does not make it easier for people to answer your questions. What's more important is that more details are provided with each new post you provide, because if people did not answer your question, it's likely that not enough information was provided in the first place. And many of the people that might be able to answer your question, don't have the time to still have to ask first: What on Earth are you talking about? __________________ OpenFOAM: FAQ \| Getting started Forum: How to get help, to post code/output and forum guide Read this before sending me PM Last edited by wyldckat; August 13, 2014 at 16:10. Reason: see "PS:"

August 17, 2014, 15:00		#5
wyldckat Retired Super Moderator Bruno Santos Join Date: Mar 2009 Location: Lisbon, Portugal Posts: 10,982 Blog Entries: 45 Rep Power: 128	Greetings to all! To complement Armin's answer: the class "oneEqEddy" is derived from the class "GenEddyVisc", which in turn derives from "LESModel". This last class is the base class for accessing turbulence data that is readily available on the solver's side. For example, have a look at the solver "applications/solvers/heatTransfer/thermoFoam" in OpenFOAM 2.3, namely in the file "setAlphaEff.H". There you can see how it explicitly loads the turbulence of type "LESModel". Correlating the information given in #3 with the information here on the wiki, it looks like the terms you've looking for are these: http://www.cfd-online.com/Wiki/Kinet...id-scale_model Checking Armin's answer in post #4 does reveal the same equations in this method: Code: void oneEqEddy::correct(const tmp<volTensorField>& tgradU) All of the terms of the equation are all solved directly in a single equation and are not available for outside use. Which means that you will have to either create a new derived class from "oneEqEddy" or modify directly OpenFOAM's class "oneEqEddy", to write out each term of the equation. Or you can modify the solver and do a dynamically cast to the instance of "LESModel" class to "oneEqEddy" and then access directly the "k()" method and other methods that give the components used in the equation that is solved in "oneEqEddy::correct". Although I don't know how practical that is, because each term of the equation is in the form of "fvScalarMatrix" and not of in the form of a standard field. Best regards, Bruno amolrajan likes this. __________________ OpenFOAM: FAQ \| Getting started Forum: How to get help, to post code/output and forum guide Read this before sending me PM

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Suggestion for a new sub-forum at OpenFOAM's Forum	wyldckat	Site Help, Feedback & Discussions	20	October 28, 2014 10:04
Kinetic Turbulent Energy...	mohammad_mec87	Main CFD Forum	4	July 2, 2014 05:35
Significance of turbulent kinetic energy	oj.bulmer	CFX	1	February 25, 2014 18:30
Turbulent kinetic energy at wall?	alanlove	FLUENT	1	November 23, 2012 12:24
turbulent kinetic energy and dissipation	panneerselvam.R	CFX	1	July 22, 2005 02:55

August 14, 2014, 01:55		#3
Thangam Member Thangam Natarajan Join Date: Dec 2010 Location: Perth Posts: 70 Rep Power: 16	My bad I thought these were quite obvious! Anyway, The complete form of the turbulent kinetic energy can be written as, $P_k - C_k - \epsilon_k + D_k = 0$ , where, is the production term of given by, $P_k = - \overline{u\prime_i u\prime_m} \frac{\partial U_i}{\partial x_m}$ is the mean convection term of given by, $C_k = U_m \frac{\partial k}{\partial x_m}$ $\epsilon_k$ is the dissipation term of given by, $\epsilon_k = \nu \overline{ \frac{\partial u\prime_i}{\partial x_m}\frac{\partial u\prime_i}{\partial x_m}}$ and is the diffusion transport term of given by, $D_k = \frac{1}{2} \frac{\partial}{\partial x_m}( T_{iim}^{(\nu)} - T_{iim}^{(p\prime)} - T_{iim}^{(u\prime)} )$ and contains the viscous diffusion, $T_{iim}^{(\nu)} = - \nu \overline{\frac{\partial u_i\prime u_i\prime}{\partial x_m}}$ pressure diffusion, $T_{iim}^{(p \prime)} = \frac{2}{\rho}\overline{u_i \prime p\prime \delta_{im}}$ and the turbulent diffusion given by, $T_{iim}^{(u \prime)} = \overline{u_i u_i u_m}$ The problem is I am not able to identify where these terms are embedded in the code and how to write them out by creating a small utility. Thanks. amolrajan, manuc and jadidi.cfd like this.

July 15, 2016, 10:21		#7
syavash Senior Member Ehsan Asgari Join Date: Apr 2010 Posts: 473 Rep Power: 18	For anyone interested in this subject, I have created a new thread. http://www.cfd-online.com/Forums/ope...e-budgets.html Regards, Syavash