[Hanzo]

Hi everybody,

I found the following issues in OpenFOAM concerning some turbulence analysis:

- the utility createTurbulenceFields only works for RAS computations
- there is no standard utility to average fields after computation has been done (to get U-mean, U-rms, R-Mean, Reff-Mean ...)

So I worked on these utilities and after mentioning it in some posts I got replies and private messages asking me if I could publish them. So here they are:

createTurbulenceFieldsLES

Is a tools which writes out the fields k,epsilon, R, Reff after a RAS or LES simulation has been performed.

postAverageTurbulenceFields

Inspired by eelcovv, who wrote the tool postAverage, I extended his tools to be also able to post average the fields R and Reff. Again RAS and LES are supported.
The post can be found here http://www.cfd-online.com/Forums/ope...rocessing.html

How to use these utilities

Please find attached the modified tutorials

- pitz_daily_les
- cavity_ras

Including Allrun scripts which show how to use these tools.

How to install

- copy the contents from the archives "postAverageTurbulenceFields.zip" and "createTurbulenceFieldsLES.zip" into the folder username-2.1.1/application/utilities

- go into each of the folders postAverageTurbulenceFields and createTurbulenceFieldsLES and type wmake

It also worked under OpenFOAM 1.6.

Known issues

- for some reason, which I did not figure out, averaging the Reynolds stress tensor for RAS computations does not work (but works for Reff)

- if the output setting for the averaging is set to certain values, nothing is written. However, it works for output intervals 1 and 2 (just delete the unnecessary files after averaging)

I am happy about any comments how to improve these tools or if there is a better way to post own code.

Best,
Hanzo

[atmcfd]

Hanzo,

Great work and thanks for all the efforts you have put in. I will give these tools a try and let you know.


Best,

A.T.M

[ghindi]

Hi there!

I'm relatively new to LES simulation and I'm struggling now to post-process energy spectrum to verify the -5/3 law for my model.

Does anyone know how to extract/create that field in OF?

Best regards,
Gustavo.

[rakendd]

Hi all
Sorry for opening up this old thread.
I want to calculate UPrime2Mean in LES post-processing.
createTurbulenceFieldsLES is calculating R and Reff but these values are different from UPrime2Mean
postAverageTurbulenceFields giving

Create time

Create mesh for time = 0

Adding fields for time 0
Reading field p

Reading field k

Reading field U

Reading/calculating face flux field phi

Selecting incompressible transport model Newtonian
Selecting turbulence model type LESModel
Selecting LES turbulence model oneEqEddy
--> FOAM Warning :
From function cubeRootVolDelta::calcDelta()
in file cubeRootVolDelta/cubeRootVolDelta.C at line 52
Case is 2D, LES is not strictly applicable

oneEqEddyCoeffs
{
ce 1.048;
ck 0.094;
}

Reading/calculating field UMean

Reading/calculating field pMean

Reading/calculating field UPrime2Mean

Reading/calculating field pPrime2Mean

fieldAverage: starting averaging at time 0

Calculating averages

but nothing is being calculated.
Can someone help me in this...
Thank you.

[jiez]

Nice job but there seems to be a bug:
For some reason, if do not specify time range, the createTurbulenceFieldsLES utility calculates only based on the velocity solution in the first read, resulting in identical k, epsilon, R, Reff for all times.

[jiez]

The bug I reported above can be easily fixed by a minor revision on the origin code. Attached please find an updated version.

[Bohus]

It is very good work guys, tahnks
But:
1. It is necessary to read mesh in each timestep for me, because I use the dynamic mesh.
2. I am not sure, but I think that Epsilon is calculated only as subgrid one, and the reading of nuSgs is also necessary. Then it works for me someway.

Best regards,
Bohus.

[Z.Q. Niu]

Hi ghindi,
Have you found the method of calculating the energy spectrum?

[anishtain4]

Usually you don't need the spectrum of the whole domain. You can have spectrum in time or a homogeneous direction, the easiest way to do it is to probe a location in time or sample a line in that direction then take FFT of it in Matlab or Python.

Remember, for time spectrum all of the times need to be loaded in memory which for big cases is not practical. For a directional spectrum always sampling is needed as the grid is unstructured and doing this for the whole domain introduces a huge computational cost.

[pedramtx]

Hello,
I've been trying to calculate rms of Lambda2. I added this quantity in controlDict/fieldAverage1, but got this warning:
--> FOAM Warning :
From function void Foam::fieldAverage::initialize()
in file fieldAverage/fieldAverage/fieldAverage.C at line 92
Field Lambda2 not found in database for averaging

Any suggestion on resolving the issue is appreciated.

Pedram

[pedramtx]

I resolved the issue! I modulated the solver that I use.
Thanks.