Fluids

Fluids

Porous Zone and High Continuity Residual

    • PurdueME
      Subscriber

      I have a simulation where I want to model a porous zone in steady state. When I set the zone to non-porous I get a solution which converges quickly. When I set the zone to a porous zone the continuity residual does not converge within a few thousand iterations. In my model I have 3 zones, two of which are non-porous. When I set the middle zone to be porous the issue arises. 


       


      I have a velocity inlet condition and a pressure outlet condition. 


       


      When I set the boundary condition at the inlet to be a pressure boundary condition the solution converges rapidly. 


       


      Any advice would be greatly appreciated. 

    • RK
      Ansys Employee

      Hello, 


      You have mentioned that with pressure inlet you get convergence. Can you please be more specific on your question? 

    • PurdueME
      Subscriber

      I am simulating a phenomena where I know the inlet volumetric flow, which in this case I am converting to velocity.


       


      I am wondering why there would be a difference in the convergence behavior, or how to address the lack of convergence for the continuity residual. 


       


      I am using the viscous flow model with default values for the solvers. 

    • Karthik R
      Administrator

      What are your solver settings? Are you using a Coupled Pseudo-Transient to solve this model? Also, which viscous model is this? Do you have turbulence here? 


      Can you share what your inlet pressure value is? Also, what velocity are you using?


      Thanks.


      Karthik

    • PurdueME
      Subscriber

      I think I found the answer digging through the manual. For anyone else following the thread I found the answer at the url:


       


      https://www.afs.enea.it/project/neptunius/docs/fluent/html/ug/node233.htm#sec-porous-limitations under the section "Solution Strategies for Porous Media". 


       


      I didn't read it to the end initially Grinning Face with Sweat on Samsung One UI 1.5


       

    • PurdueME
      Subscriber

      But to answer your question I am using the following:


       



      • Solver Settings

        • .   Pressure Based. Absolute Velocity, Steady, Planar (2D)



      • Coupled Pseudo-Transient

        •    Yes



      • Viscous Model

        •    Laminar



      • Turbulence

        •    No



    • PurdueME
      Subscriber

      It seems that, following the advice of the manual, if I initialize the region with a pressure I achieve rapid convergence. 

    • Karthik R
      Administrator

      Perfect! Could you please mark your answer as 'Is Solution' to benefit others on the community?


      Thank you.


      Karthik

Viewing 7 reply threads
  • You must be logged in to reply to this topic.