Flow past flap with prescribed velocity

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • maryrose
    Member
    • May 2019
    • 32

    Flow past flap with prescribed velocity

    Hi,

    I'm trying to run the 'flow past a flap' FSI test case with a rigid solid. Is it possible to apply a prescribed velocity instead of a pressure gradient between the inlet and the outlet?
    I would like to apply an initial velocity of 1 in the x-direction to the inlet, outlet, and to the fluid. I've tried to apply it as a boundary condition to the inlet and outlet but my results show a near zero velocity in the fluid domain, aside from an area of higher velocity at the base of the flap.

    Many thanks,
    Maryrose
  • jshim777
    Junior Member
    • Mar 2018
    • 11

    #2
    Hi Maryrose

    Yes you can prescribe a "fluid normal velocity" surface load on the inlet. Note that the velocity is with respect to the normal to the surface so if you want the fluid to go inside as is usually done you have to set the velocity to -1 instead of 1. Typically for the outlet you can prescribe 0 pressure which you do by applying a zero fluid dilatation BC. Because of the formulation of the governing equations you may also have to prescribed the dilatation to 0 on the inlet corners. A tutorial that describe prescribing BCs on a FSI problem can be found in the Preview manual, which may be useful.

    Also for the flap it might be better to use a very stiff neo-Hookean or Mooney-Rivlin rather than a rigid material because rigid materials are dealt with differently compared to other solids and you have to add additional rigid constraints and boundary conditions.

    Jay

    Comment

    • maryrose
      Member
      • May 2019
      • 32

      #3
      Hi Jay,

      Thanks for your help. I ran a test with a normal inlet velocity but the fluid pressure results I'm getting are in the order of 10^8. How should I prescribe the pressure in the .feb file?

      Many thanks,
      Maryrose

      Comment

      • jshim777
        Junior Member
        • Mar 2018
        • 11

        #4
        Hi Maryrose

        Did you fix the inlet corners so that the dilatation is 0? Because dilatation is indeterminate there, if you dont do this step the pressure will be higher at those locations. However, elsewhere the pressures should not be blowing up. Some other potential sources of error: Did you also make sure to prescribe the FSI traction at the interface between the fluid and solid? Did you also fix the velocity at the interface (no-slip condition)? Did you give the fluid reasonable material parameters (such as for viscosity, bulk modulus, etc)? The solid as well? Did you prescribe 0 dilatation at the outlet? If you still get large pressures, I can take a look if you send it to me (jjs2215 (at) columbia (dot) edu)

        thanks
        Jay

        Comment

        • maryrose
          Member
          • May 2019
          • 32

          #5
          Hi Jay,

          I set up the simulation as you described but perhaps I missed something. There is an initial pressure wave which moves from the inlet to the outlet but after that, the pressure appears to be uniform and the magnitude continues to increase throughout the simulation.
          I've sent you on the input file.

          Many thanks,
          Maryrose

          Comment

          • jshim777
            Junior Member
            • Mar 2018
            • 11

            #6
            Hi Maryrose

            I took a look and there were a few issues. For the dilatation boundary conditions you have to check the dilatation box like you do for the velocity boundary conditions or else it does not appear in the .feb file. I also changed some of the displacement boundary conditions and velocity boundary conditions. The model itself was also very small, so I scaled it up using the PreView transform function, or else the solid would move too much even at that high of a stiffness. I also changed the time step size to be bigger (but not too big) and made it so that auto time stepper can be used so that the solver can restart if it happens to run into any negative jacobians (which can happen to the mesh with FSI). Also the backflow and tangential stabilization loads should only be prescribed at the outlet because that is basically prescribing a fluid traction that conflicts with velocity boundary conditions (but not with dilatation/pressure BCs).

            Note that when I ran this problem with these changes, it runs but not particularly well. I think the high stiffness of the solid and the fact that the width of the flow suddenly changes drastically and expands again make it difficult to solve. The flow eventually starts to shed vorticies and the pressure looks reasonable (high before the flap, lower after). You can try to play with some of the options in the step or refine the mesh to try to make it run better. I sent you these files through email.

            Thanks
            Jay

            Comment

            • maryrose
              Member
              • May 2019
              • 32

              #7
              Thanks Jay, the results are much better now.

              Maryrose

              Comment

              Working...
              X
              😀
              😂
              🥰
              😘
              🤢
              😎
              😞
              😡
              👍
              👎