On 23 April 2014 15:18, Garth N. Wells <[email protected]> wrote: > > > On 23 Apr 2014, at 14:52, David Ham <[email protected]> wrote: > > > > > > > On Wednesday, April 23, 2014, Garth N. Wells <[email protected]> wrote: > > > > On 16 Apr 2014, at 12:08, David Ham <[email protected]> wrote: > > > > > Hi Jack, > > > > > > The student who was working on isoparametric mappings last summer didn't > > > get very far, however we were very fortunate to host Martin Alnæs for six > > > weeks this spring. We designed, and he implemented, the UFL layer changes > > > needed to get isoparametric in. We still have some legwork to do in our > > > FFC branch to get those changes into Firedrake and I believe the > > > equivalent work for Dolfin is still underway, but there is progress and > > > we hope to have isoparametric by the summer.
Thanks David. I'll be honest and say that I haven't had a play with Firedrake yet, some of the features like the OP2 integration look very cool. I saw that Patrick is using it with his adjoint now. I think he will come to visit University of Luxembourg in the autumn so maybe we will play with it then. > > > > > > On the shell subject, Firedrake now supports simulation on extruded > > > shells formed of triangular prism elements. I'm not sure if that helps > > > you but you're welcome to give it a try. I will give it a go, but I suspect this is an example of point 3 of my original post regarding terminology and generality! There are something called faceted shell elements in solid mechanics which are essentially a composition of flat plate elements... They have issues, particularly I think Corrado is interested in stability analysis and snap-through problems which faceted shells are not good for. Smooth curvature and the correct interpolation of the direction vector are very important. > > > > > > > Jack is referring to a different type of shell - he’s referring to shells > > in the structural mechanics sense rather than a spherical shell. Indeed! Garth are you aware of another class of PDEs which are similar to mechanical shells? My idea for this project is to have just enough 'general' functions within UFL/FFC/FIAT to support the functionality within an additional package which specifically targets mechanical shell problems. I believe this fits in with the philosophy of the project. It would be helpful as a thought exercise to see other PDEs with similar properties to mechanical shells > > > > > > Firedrake is not limited to spherical shells. One can take any immersed > > manifold and extrude in an arbitrary direction. > > > > Nice, but the thread is about solving the equations for structural shell > models and not extruding meshes, e.g. solving the Koiter or Naghdi type shell > equations. A major issue for these types of models is supporting function > spaces and/or geometric concepts in UFL to resolve/distinguish between > in-plane and out-of-plane displacement components, and in cases how to > perform the local element tricks used some shell formulations. I will be visiting Corrado at Pierre and Marie Curie to discuss drawing up an exact blueprint for how this might be done. MITC operators are a 'must' for Naghdi type shells, but I suspect this might involve all the way back up to FIAT... Perhaps Koiter shells might be the easiest way forward, as suggested in my original post. I will collect references over the next few weeks and I will post them here when done. _______________________________________________ fenics mailing list [email protected] http://fenicsproject.org/mailman/listinfo/fenics
