[grayback87]

Simulation: Density current into a stratified basin with fresh and salty water.

Hello everybody. I will appreciate some clues or whatever that help me to find a solution to the problem which is driving my crazy for a long time.

I know that several of our mates in cfd-forum struggled with the same problem.¿How to model fresh-water with salt-water?. ( I think that Fluent should face this problem, making it more easier, because nowadays there are a lot of engineering investigation in the interaction and mixing of this fluids)


I have three different densities, but this 3 fluids are the same,
1.Fresh Water with RHO=998 kg/m3 T=295k
2.Salt Water with RHO=1010 kg/m3 T=295k
3.Salt Water (Density current) with RHO =1199 kg/m3 T=295 k

Reading the forum and some papers I thought that the best way to model this problem was with Multiphase-Problem. Many scientist used fluent with an unknown multiphase model (they didn't say what specific model was used). Although I tried to use some of them I am not very comfortable with them because I am not able to do what I want. I suppose that the model that best fits the problem is the Eulerian model, but I am not sure.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
I tried to face this problem from different point of views:

-(1)My first option was to create a new material with a artificial density function in order to create 2 different zones with a density step; initializing the solution with the temperature gave me the density of the fresh water, and then patching the zone where the salt water was supposed to be. Then the density current enter through the inlet BC with a constant Temperature which gave it the desired density. Everything was ok except that this model did not behave as I previously supposed.

-(2)My 2nd option was to create an Scalar (salinity) and try to use the UNESCO (which relate density-temperature-salinity) formula to define the density. But it wasn't possible because , as far as I know, Fluent only allows to define the density in terms of Temperature or an UDF (and I don't know how to insert in my UDF the value of the UDS).

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I only see 2 solutions to this problem.
1.- Be able to write a proper UDF with the UNESCO formula and the UDS
2.- Use the right multiphase problem. ¿Eulerian Model?.

Any suggestion? Thank you

References (they used fluent):
1-Cfd modelling of desalination plant brine discharge system. G. Zeil Q.Zhang Australian Water Association.
2-Numerical Simulation of Lab-Scale-Brine-Water Mixing Experiments. Sandia National Laboratories.
3-Experimental and Numerical Investigation of Desalination Plant Outfalls in Limited Disposal Areas.
4-Numerical Simulation of Flushing of Trapped Salt Water from a Bar-Blocked Estuary

[hwangpo]

I saw plenty of papers using VOF model to simulate selective withdrawal from multiple-layer flow. But actually most of them are two-layer flow.

[Zaktatir]

I will suggest to used the Muti-Fluid VOF approach with right materials. The stratification is density driven. If your fluids are always quite separated then you may use the One Fluid formulation. Otherwise and trying to resolve regions with sharp interface (not smeared) then the Zonal discretization may the best solution!.

The problem is not a big deal! Start the Stratification in horizontal simulation or with something like a bend!

[Noman422]

i have to analyze a model of salinity salt gradient solar pond in ansys fluent,but i don;t know to how to apply the boundary conditions,and three different salt layers in fluent.can any one share the salinity salt gradient solar model ansys file with me.i would be appreciate.

nauman