Support for self-contact in dealii

[Anirban Pal]

Hello,

I want to solve a elasto-statics problem on a domain which is highly cracked. 

The cracks do not propagate. I'm interested in computing the elastic behavior

of the overall domain.. For this, I need some self contact routines to ensure

contact (pressure-overclosure relationships) at the internal cracks,

I was wondering if deal ii is suited for this purpose. And if not, what would be 

the potential amount of work required to add the required features. My domain

needs to be very large and have many cracks, so I need to use parallelism.

Thank you for your attention.

Anirban

[Wolfgang Bangerth]

It's certainly going to be work, but I think it can be done.

There are a number of publications that appeared with deal.II and that
deal with contact problems -- search for "contact" here:
http://dealii.org/publications.html

I think you probably also want to look through the PhD thesis by Damien
Lebrun-Grandie:
http://oaktrust.library.tamu.edu/bitstream/handle/1969.1/152540/LEBRUN-GRANDIE-DISSERTATION-2014.pdf;sequence=1
He didn't use deal.II, but the techniques he describes in his thesis are
all applicable immediately with deal.II as well.

Best
W.

--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@colostate.edu
www: http://www.math.colostate.edu/~bangerth/