Hi! Does compas_fea on opensees support the analysis of shell elements like this? If not, would adding this feature entail a major extension of the current wrappers/functionaliy around opensees? I’m thinking about it from the perspective of a fellow pull-requester.
hi rafael,
we are currently preparing to release compas_fea2 which will completely replace version 1.
not sure how much effort it is to still add something like this to FEA1.
deferring to @franaudo 
Hi Rafael,
compas_fea(1) does not currently support what you are asking. Adding it now would make little (if no) sense, since compas_fea2 is almost ready and it is designed to ease outside contributions.
That’s good news, @franaudo @tomvanmele! With what (cool) new features the next release will surprise us?
It is a complete restructure to have a clear API (for both users and developers). This should promote more contributions and also help to expand the range of softwares supported (currently only abaqus, ansys and limited opensees).
On top of that, there are already a few upgrades, such as for example the possibility to use user-defined material subroutines for advanced material modelling.
But stay tuned, it should be ready in a few weeks… 
1 Like
Hi @franaudo
any news on Compas_Fea_v2?
I am curious to understand how you are dealing with OpenSees
I am the developer of Alpaca4d https://www.food4rhino.com/app/alpaca4d which is a Grasshopper Wrapper for a limited OpenSees (I have still implemented a lot)
Best,
Marco
Ciao Marco,
fea2 is in the making…we expanded the team quite a lot and decided to add more stuff before releasing. I wish I could give you a date, but unfortunately I don’t have it. What I can tell you is that I have added a 3D topology optimization feature and we are working on the integration with gmesh and compas_view2 so that we are independent from the CAD environment (rhino/blender will still be supported though).
Meanwhile, we are still maintaining compas_fea (now fully compatible with the latest version of compas).
Regarding OpenSees, the approach with compas_fea and compas_fea2 is always to generate an input file, because in this way we are consistent across all supported software (abaqus, ansis, sofistik for now): moreover, the bottleneck in terms of time is (almost always) on the analysis part, hence the additional I/O time to generate and read the input file is marginal.
A different strategy would be to use the opensees python interpreter and generate the model in runtime. Probably you are already aware of it, but I thought it was worth mentioning it.
Please let me know if you have any other questions!
Ciao,
Francesco
PS: I haven’t used alpaca4d but looks cool! I will give it a try at some point
1 Like
Hi @franaudo ,
Is there any news about compas_fea2 ?
Is it still under active development?
I would love to give it a try at some point.
Best,
Nicolas
Hello Nicolas!
I know that I keep saying this…but we are really close 
If you want to have a preview, here is the repo: compas-dev/compas_fea2: FEA for the COMPAS framework. (github.com)
Please note that it is an alpha version and that we will soon publish major updates, but at least you can have an idea 
Best,
Francesco
super nice!
Thanks a lot for the repo link, I will definitely have a look. 
Cheers,
Nicolas