Hi Denis, Thank you for the pointer! This is very helpful.
The below is my understanding (and please let me know if I am wrong): 1. If we do the L2 projection over a domain of elements, then it smooths the stress jumps in the sense of integration. The stress would be continuous over the domain; 2. Or, if we do the L2 projection over single elements (or, we focus on elements on the boundaries between the domains in #1), there will still be jumps over the boundaries. This sometimes is desirable; we can then choose our own method to assign to the node (e.g. peak stress). Thanks, Shawn On Friday, November 4, 2016 at 4:08:00 AM UTC-7, Denis Davydov wrote: > > Hi Lisa, > > > > On 4 Nov 2016, at 11:54, Lisa Collins <lisa.col...@gmail.com > <javascript:>> wrote: > > > > Thanks for the replies. > > The problem I am facing has two parts: > > > > 1- In deal ii, how to calculate stress at nodes based on stress at Gauss > points? Stress at nodal points is not continues. Is there any averaging > function to calculate stress at nodal points? > > Well, that’s exactly my reply above ;-) You can do an L2 projection from a > discontinuous function (defined by values at quadrature points) to a > continuous FE space. This could also be understood as some averaging. > > Cheers, > Denis. > > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
