Fluids

Fluids

Fluid Flow in porous media

    • navidz
      Subscriber

      Hi Everyone,


      I want to model fluid flow inside a porous media using Fluent. This question might be very simple but I cannot figure it out and I would be appreciate for the comments and helps. So the problem is:


      Assume that I have two parts on top of each other. The bottom part is sand and the top part is water. And there is an object on top of the sand that is fixed and is not moving. However, it can pump fluid into the sand from its bottom as in Figure bellow


      Consiguration


      So the problems are:


      1- I know how to define porous media in Fluent by enabling the porous zone in "Cell zone condition", and I can give the permeability and parameters to define inertia in the zone and that is not the problem. But I feel it is more of multi-physics problem since the sands can also move. But lets just put this aside and for now considering that the sand is not moving. So how can we define other parameters like degree of saturation in the sand. I know that the permeability parameter that we can give, already embed porosity inside it but how about other parameters? I guess maybe it has to do with TBDATA, that I do not know how to use with Fluent.


      2- How to define a radial inlet flow as in above figure. In inlet options of Fluent, I can either say perpendicular to inlet surface and define the direction. But it will define all the inlet flow in one direction rather than in every direction. Should I define it through UDF? is there any other easier way?


      3- So one of my main goals is to see how fluid will come out of sand from the sides of the object as bellow   


      is there something specific I need to do or defining the "Mesh Interface" is sufficient? Will fluid (in red stream lines) merge and join the water on top?


      4- To come back to the first question, is it possible to model the movement of the sand too?


      5- Now imagine that the solid body on top of the sand that is fixed, can now move. How can I define an object that is not completely closed. That is, when I put one part of the body as a place for inlet that inject the fluid inside the sand then the square body is not close and the inlet part is open so I cannot model it with Fluent as bellow:



      What is the workaround for this problem?


      Thanks,

    • Raef.Kobeissi
      Subscriber

      Hello,


      You cannot model the motion of the sand. The sand domain will be considered a fluid with a high resistance which reflects its porosity.Check this document which explains the porous medium in detail:


      EDIT (rwoolhou) : Removed sharcnet link to ANSYS documentation.  Porous media is covered in the Fluent Users Guide under Cell Zones. 


       


      Also check these videos I created on Porous medium:



    • Rob
      Ansys Employee

      If you want to move the sand too the Eulerian Grannular model is the best starting point. 

    • navidz
      Subscriber

      Thank you so much Raef for your reply and very nice tutorial videos that you put online, I personally have used them in multiple occasions and have learned a lot from them.


      I have watched the both videos that you mentioned here. So I understand that the degree of saturation cannot be adjusted with cell zone parameters, am I correct. I am asking this because when I pump fluid I would like to see how the fluid disperse in the domain. And for sure it has some connections with degree of saturations. If there is no fluid in the sand, then fluid can easily penetrate into the sand versus the case that the fluid need to push the fluid inside the sand asiade. So I am guessing that in one case fluid will go deep in the sand and in the other one, it will come out from the sides.


      Also, as one of the questions that I mentioned, I have a problem with injecting fluid in all directions. What I am doing at this point is making a very small nubbin in the shape of a half sphere and make it my inlet and ask the inlet fluid to be normal to the surface. Is there a better way?

    • navidz
      Subscriber

      rwoolhou, thank you for your answer. Is there an specific tutorial or something that I can learn from for this type of problem?


       

    • DrAmine
      Ansys Employee

      To get started check Using the Eulerian Granular Multiphase Model with Heat Transfer standard tutorial. 

    • aims
      Subscriber
      Hello nPlease i am trying to do a simulation with ansys cfx of a near wellbore flow. But i keep having convergence error. nPlease i will b glad if someone could help me with tutorials of flow in porous media nI have use many but i still dnt get my problem solven
Viewing 6 reply threads
  • You must be logged in to reply to this topic.