[lakeat]

Hi,

As you might know, in a wind tunnel testing, it is generated by means of grid. Precisely speaking, this is not isotropic, however my question is:

Does anyone know how to generate this kind of turbulence in CFD simulation.

Thanks,

[jola]

I assume that you want to generate this to use as an inlet condition in an LES simulation or something. My old university dept has been doing some work on this. I found one paper online here: http://www.tfd.chalmers.se/~lada/postscript_files/paper_ceas_davidson.pdf

I do not know if this is the best way, but the guys behind it usually produce good stuff.

[lakeat]

Originally Posted by jola

I assume that you want to generate this to use as an inlet condition in an LES simulation or something. My old university dept has been doing some work on this. I found one paper online here: http://www.tfd.chalmers.se/~lada/postscript_files/paper_ceas_davidson.pdf

I do not know if this is the best way, but the guys behind it usually produce good stuff.

Actually, I read this paper for a few times, for your group is quite active in CFD, so one may easily find the reference to Davidson's paper.

The point is:
1. This is a synthetic method, I wonder is there is any non-synthetic method, like recircle as in OpenFOAM, or precursur methods, etc. And any successful works?
2. I am quite interested in the difference between these methods.

For a flat plate turbulence, or boundary layer turbulence, Spalart's work and Lund's may give fairly good results. But in wind tunnel testing, isotropic turbulence is still needed.

[lakeat]

Another thing I am worrying about the synthetic method is:

Synthetic method will not always produce a correct turbulence coherent structures, and if so, how could we evaluate the results, partly successful or what?

I am saying that, we might get some good results that compares well with the EXPERIMENTS, but that does not mean it is very correct in everyway.

I will give an example in our lab: our need is not just to generate the turbulence, but also its integral length scale, (for civil engineering structures, you can imagine), and see will the integral length scale affect the vibration or not?

So,
Is my concern valid?