Good morning folks:
At our company we are performing an extended use of Code_saturne since a few time ago with come modifications on the code according to our needs and generally concerning Lagrangian multiphase simulations.
In the other hand, we are performing some "experiments" of chemical reactions in Eulerian simulation through the use of the transported species based on some additional subroutines but recently I have been asking myself about the posibility of perform non-reactive and perfectly dispersed multiphase simulations (for example water/oil or air/water simulations) by using adequate-configured species.
So, I have some questions about this:
- Has somebody done this approach before?
- Are the transport equations prepared/alterable to perform this simulations via subroutines?
- What's the scope of the multiphase CS capability (http://code-saturne.org/cms/it-possible ... simulation)
Kind regards
Eulerian multiphase simulation
Forum rules
Please read the forum usage recommendations before posting.
Please read the forum usage recommendations before posting.
Re: Eulerian multiphase simulation
Hi,
No one replies to this topic. I'll try.
I have not done this before, but I think:
For two-fluid model, use two CS instances defined in the same flow domain --- one for each phase. Use CS' volume-coupling for the interface between the two phases. There also needs an iterative scheme to make sure the volume fractions sum up to 1. And, the dispersion phase needs to handle particle-particle interactions (kinetic theory), slip BC, and particle rebound from the wall. For these, studying the implementation in MFix code (https://mfix.netl.doe.gov/mfix/) may be helpful.
- ffan
No one replies to this topic. I'll try.
I have not done this before, but I think:
For two-fluid model, use two CS instances defined in the same flow domain --- one for each phase. Use CS' volume-coupling for the interface between the two phases. There also needs an iterative scheme to make sure the volume fractions sum up to 1. And, the dispersion phase needs to handle particle-particle interactions (kinetic theory), slip BC, and particle rebound from the wall. For these, studying the implementation in MFix code (https://mfix.netl.doe.gov/mfix/) may be helpful.
- ffan