[FabianF]

Dear Foamers,

I am trying to solve a fluid domain, which should include radiation in just one part of the domain. Below is a simplified case given, which just includes the relevant properties needed to understand the problem.

The red part in the attached picture represents the part of the domain in which radiation does not play a role at all. The fluid then gets into the "geometry" where it is surrounded by walls. In this part of the domain I want to include the calculation of the radiation. Both parts can be seperated by more or less one surface along faces. In the following I will call this surface "internal surface".

I want to impose a boundary condition (BC) for the radiation on the "green" side of the internal surface, but I don't want to have any BC for the fluid dynamics here.

Has somebody of you done something similar, and/or can somebody give me a hint in which direction I should search further.

So far I have considered to try the following ideas:

1) define the internal surface as boundary, and then set this boundary to something like "empty/interior/not of interest" for all values that do not have to take care about this boundary (e.g. p,U,k, .....)
-> Is there such kind of condition in OF?

2) Use some interface condition? I have heard of AMI cyclic which works for the fluid dynamics. But in a first test with AMI cyclic, I could not impose the radiation BC on the AMI cyclic patch.
-> Should I investigate further in this direction? Does something better than AMI cyclic exist?

3) Use a multiregion approach. At this point I do not know how multiregion works, and it would be Fluid-to-identicalFluid regions.
I am not sure if it is really necessary to seperate a domain of the same fluid into two regions.
-> Is it still a good idea to try this? Actually the fluid does not interact with the radiation itself (no absorbtion or emission), only with hot and cold walls.

4) Do something else, but what? Any Ideas?


I would be very grateful if any one could help me.

Greetings
Fabian



P.S. The real model is meshed in ICEM CFD. Then exported to fluent-mesh and imported to OF by fluent3dToFoam. So if I have to set up things already there, just tell me.

[Aunura]

Hello,Fabian. I have similar problems like you described recently. I want to know whether you have solved this problem.

[FabianF]

Originally Posted by Aunura

Hello,Fabian. I have similar problems like you described recently. I want to know whether you have solved this problem.

Hello Aunura,

thanks for digging up this 5 year old thread. To be honest I do not know any more how I solved this and if I really did or just worked around. If I find some time, I might try to look into the stuff I have done back then; but I am not confident that I can get back into into it. I have not been working with Openfoam since then.