Hi Chiao,

On 10/11/2017 04:25 PM, benechiao@gmail.com wrote:

Hi,

I am very new to sfepy. I am currently working on an FSI project, and I want to use Sfepy for my simulation (open-source is must, and based on http://feacompare.com/sortbyfeatures.html, it seems that Sfepy is pretty feature-rich and has modern design). My main focuses are on the time-integrator and solution translator between interfaces. I want to use Sfepy as structure and fluid solvers.

First, while that page seems useful, I do not recommend choosing a FE solver just based on that (e.g. SfePy cannot do nonlinear-contact, there are inaccuracies and the evaluation is only binary).

We do use sfepy for solid mechanics including large deformations, but the CFD support is very limited (and we (the man developers) have no practical experience with CFD).

Concerning the time-integrator support, I am working on improving that [1], but it is not ready yet. We plan to support PETSc time-steppers and other solvers. Currently, there is not much besides a simple back-ward difference-based time-stepping.

[1] https://github.com/rc/sfepy/commits/dynamic-ts-solvers

So my questions are: a) is it possible to get data (through memory) in each step? Also, b) is it possible to modify bc's and loads from external sources in each step?

b) is definitely possible. I am not sure what exactly you mean by a).

P.S. I know that I should do more self research before I post something here. But since the project schedule is very tight, and I do hope to get some quick answers. Thank you very much.

It would be great if you were able to implement FSI with sfepy, but, basically, due to our limited knowledge of the topic, you would be on your own. By no means I want to dissuade you, but check also for other options.

Cheers, r.

Best, Chiao

---

SfePy mailing list sfepy@python.org https://mail.python.org/mm3/mailman3/lists/sfepy.python.org/