I am developping a library of mesh mover for FSI, but I got some problem to keep valid mesh for movement of high lift components of aircraft (from takeoff to cruise). I use an interpolation function (IDW) to move the mesh but when two bodies are close of each other the mesh in between get distorted to generate negative elements.
I think you are solving a CFD problem with "Moving meshes". In such a case, you meshes need to be "regenerated" at every time step. You cannot simply keep "valid mesh" for the entire simulation. In such a case, you can restructure your mesh so that no distortion occurs. The motion of the solid bodies must play an important role in the mesh restructuring.
Regards...
Lattice Boltzmann currently has acquired a lot of attention , because it overcomes the problems of the structure complexity . and can be used as long as you are working in the low mach number region.
To my knowledge there are currently only 2 possible ways of ensuring mesh quality: remeshing or multiple configurations (like in combustion engine simulation).
Hello Jonathan
I have no experience with the IDW method you are using. But for my FSI work I am using a pseudo linear structural mesh motion solver. I inversely scale the material properties of the element so as to reduce distortion of small elements. This works for me even for considerable movement of the structure. (I use a finite element based solver for Navier-Stokes equations). I can expect a better mesh quality and for longer time using a non-linear pseudo structural solver.
I also think what makes a big difference is the boundary conditions you apply for the mesh motion solver (at least in my case it did). Allowing slip on certain boundaries (depends on the problem) may help.
To give you an idea .. I am attaching a picture with the mesh movement solved as mentioned above.
- Badges
- Science topic