I am trying to simulate the motion of a beam using the overset mesh method. Let me explain the purpose of the simulation:
There is a beam inside a pipe, and air is introduced into the pipe. To allow airflow discharge, there is a small orifice on the beam. Due to the pressure difference, the beam oscillates.
Four months ago, I opened a topic about this problem on the forum (link here). I see the overset mesh method as a suitable approach to solving this issue. However, as shown in the images, there is incorrect interpolation occurring.
I use fenics as a solid solver.
What do you think could be the solution to this problem?
@Turan I am not sure I completely understand the geometry from the surface plots. Could you please provide a simplified drawing? Is this a continuous elastic wall, with a hole in the middle?
I would consider overset as a solution to pre-defined motion. This is not the case, here, is it? I guess that overset could complicate a bit the whole situation.
I don’t see any preCICE-related issue here (if “interpolation” means “preCICE mapping”). This is just related to the overset configuration in OpenFOAM, no?
I would approach this problem similarly to the perpendcular-flap tutorial: one fluid mesh, one solid mesh. Why doesn’t that work?
A disk is placed in the middle of the pipe. A hole, almost the same size as the beam, is drilled in the center of the disk. You can think of it as if the beam is placed into this hole. However, the beam does not cover a small portion of the hole in the disk, and this gap is for air evacuation. The small square block seen in the two-dimensional perspective represents this gap.
This case is not predefined. It is essentially a modified configuration of the perpendicular flap tutorial according to the specified conditions. The airflow does not pass freely over the beam but is instead constrained to flow through the hole.
For the solid mechanics simulation, I am using FEniCS, while handling the fluid dynamics in OpenFOAM. Due to the small orifice , mesh deformation issues were occurring, leading to negative volume problems. I considered using overset mesh. I defined the overset mesh region around the beam.
I think what you need is no overset mesh, but just a refinement around that region, no? Or at least make the overset mesh region larger. I would expect all kinds of potential discontinuities by defining the overset mesh to share a boundary with the orifice.
What do you think might be the cause of the problem? Is it due to the effect of high aspect ratios? How much refinement can I do on cell? What do you mean by refinement?
You seem to be using a uniform mesh with stretching (constant number of cells per x and y, just different sizes). With snappyHexMesh, for example, you can have finer cells around the interface.
See, for example, the images in this step of the FSI training course we have. This is from the last preCICE Workshop, and we plan to have it again this year.
