CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Experimental data mapping. (http://www.cfd-online.com/Forums/openfoam/83768-experimental-data-mapping.html)

paka January 10, 2011 22:26

Experimental data mapping.
 
Good day to you all,

I have some experimental data available and I would like to map it on my OpenFOAM computational grid. I plan to use interFoam solver which relates to VoF problem. The initial condition is such as I know the initial water domain shape from experiments, which was extracted from video. I also know the water fluid velocity field, which say is constant through the whole water domain.

How can I map this kind of experimental data on top of my computational mesh?

The way I see it, I need to setFields (or something like this) on top of my computational grid, so I get irregularly shaped water domain. Second step involves velocity assignment to his irregularly shaped water domain.

Any tips on how to do it, possibly tutorials? Anyone has any experience in this matter?

Many very thanks and wish you all a great day,
K

nakul January 11, 2011 07:46

Hi

I haven't tried this before but hope this helps. There is a file in buoyantCavity tutorial in buoyantSimpleFoam (heat transfer) called createGraphs in validation folder. It has some code to plot simulation data over experimental data.

Have a look at it and see if that helps you.

Bernhard January 11, 2011 08:02

Another suggestion may be to use the mapFields utility. You can interpolate the data from one mesh to another, and then use the interpolated field as an initial condition.

paka January 11, 2011 14:52

Thank you very much for your replies, but this is not what I want.

In some sense, I want to do what Bernard wrote. However, I do not have any mesh data, I just have incoming water shape profile (from experiments) and I know the velocity of the water entering the domain. Now I would like to map/put those experimental results (from that particular time frame) on top of my computational mesh, so it will start from this prescribed point/conditions.

Many very thanks for you time,
K

Bernhard January 12, 2011 06:00

So I assume you now have data like this:
x y alpha u v w
Or something like that? In princple it is possible to construct a mesh out of your xy data (I know there is an option in Tecplot), then your only problem is to convert your data to a format that OpenFOAM can handle. I can give you no useful advise how to do this unfortunately. Maybe it is a better idea to just write your own utility, that can handle this data, that would be useful as a more general method. Good luck!

paka January 13, 2011 17:27

Yes, something like this. I less care on x-y position, say I can use arbitrary grid and I can apply my water fluid volume at any horizontal location. Said that,tThe thing I care the most is the that water domain shape which I know and I know it's velocity. Long story short, let's say, I want to map it onto my domain at different horizontal locations and see if there is any effect of this initial location on it's final flow shape/configuration.

Many very thanks for you reply. I was looking for a clear answer, which I got.

Krystian


All times are GMT -4. The time now is 05:24.