Learning Forum

[LCP]

Hello,

I am creating a semi-2D FSI problem. For FSI, a 3D model is mandatory, but to simulate a 2D problem anyway, I am using one element in the Z-direction. Then, I added to symmetry regions on the two faces in the XY-plane. This, if I understand correctly, prolongs the geometry infinitesimally in the Z-direction.

When looking for confirmation that this is the case for symmetry regions, though, I am running into other people using the symmetry region to just model half of the geometry in order to save computational time (see for instance the "Properly implementing symmetric boundary conditions" post on the forum).

So what does the symmetry region actually do, and can I use it to model a 2D model in the way I prescribed it?

Also, does it then make sense that there is a plane strain assumption applicable since the z-axis is infinitesimally long?

[1shan]

,nWhen you define a symmetry plane, the translational dof's perpendicular to the plane are constrained, so yes this will result in zero strain along the plane normal(z-axis). nRegards,nIshan.n

[LCP]

Thank you !!n