# Neumann Boundary Condition in FEM

 Register Blogs Members List Search Today's Posts Mark Forums Read

 May 4, 2013, 06:39 Neumann Boundary Condition in FEM #1 New Member   Aditya Join Date: May 2013 Location: Munich Germany Posts: 27 Rep Power: 4 Hello I am trying to implement a small FEM code for solving Poisson equation on arbitrary domains. I understand how to implement Drichlet Boundary conditions after formulating the Global stiffness matrix. But I do not quite understand how to implement Neumann Boundary conditions. For example I have to integrate over the the boundary of the element. But what if a particular element does not have a boundary, like its completely in the interior. Also when we integrate over the boundary what should be done when a element has only one of its nodes on the boundary ? Thanks in advance

 May 29, 2013, 04:21 #2 Senior Member   Rami Ben-Zvi Join Date: Mar 2009 Posts: 147 Rep Power: 8 Hi Aditya, The Poisson equation u,ii=f FEM formulation is derived using the usual Galerkin shape-function-weighted integration and the Gauss divergence theorem. You get the element boundary term as the surface integral over the element of (NI du/dn), where NI is the shape function of node I and du/dn is the derivative of u in the the outward-normal direction to the face. When the element is internal, the same face is shared by two elements, so this term cancels in the assembly. This is also true if a single node of it is on the boundary, but the face is internal. For en external face, Neumann BC means the du/dn is prescribed, so the integral is readily calculated. In the special case du/dn=0 (no flux), this term is 0. I hope this helps, Rami

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Richard Parker FLUENT 2 August 1, 2013 06:44 happy CFX 14 September 6, 2012 01:54 aaev OpenFOAM Bugs 2 December 15, 2011 15:03 PierreM FLUENT 1 December 5, 2008 08:06 ricklee Main CFD Forum 1 May 9, 2006 14:30

All times are GMT -4. The time now is 10:00.