2

Let's say I create a .poly detailing the information of the my domain geometry. I also add node boundary markers and face boundary markers.

Question 1:

Do I have to be consistent between the boundary markers of nodes and faces? I mean, if a node belongs to a face with boundary marker 1, do I have to label it with 1 too?

Question 2:

In the case there's no need to be consistent, what will happen when I create the mesh? Specifically talking, which boundary markers will appear in .face and .node files? The boundary markers of the faces or the nodes specified in the .poly file? How should I proceed to propagate the boundary markers in a good way (I'm doing Finite element method, so I want to know which are my Dirichlet and Neumann faces)?

Community
  • 1
  • 1
Gonzalo Benavides
  • 223
  • 1
  • 6

1 Answers1

0

Dirichlet boundary conditions are associated with nodal degrees of freedom; Neumann boundary conditions describe fluxes/tractions on element faces.

Dirichlet boundary conditions will be specified at nodes. The node ids are pertinent here.

Neumann boundary conditions are specified on element faces.

They are separate by definition and necessity. I am not sure about the Tetgen generation tool, but my guess is that node and element numbering should not be affected by your boundary conditions.

duffymo
  • 305,152
  • 44
  • 369
  • 561