search

CUQI-DTU / CUQIpy-FEniCS

CUQIpy plug-in for FEniCS PDE modeling
GNU General Public License v3.0
5 stars 2 forks source link

Navier Stokes steady implementation #75

Open EzequielS2 opened 6 months ago

EzequielS2 commented 6 months ago

Hi @amal-ghamdi @jakobsj @nabriis @chaozg

How to implement the below problem in Fenics?

The given expression will be interpreted at run-time so that vx = 6(y − 1)(2 − y). As y ∈ [1, 2] thereby creating a parabolic velocity profile with a mean velocity of unity.

amal-ghamdi commented 6 months ago

Hi @amal-ghamdi @jakobsj @nabriis @chaozg

How to implement the below problem in Fenics?

The given expression will be interpreted at run-time so that vx = 6(y − 1)(2 − y). As y ∈ [1, 2] thereby creating a parabolic velocity profile with a mean velocity of unity.

5

Hi @EzequielS2 , Thank you for reaching out and raising this issue! May I ask you please a bit more about the problem context. Are you using this model as part of an inverse problem? If this is the case, what are the parameters you are trying to invert for?

Thank you

chaozg commented 6 months ago

Hi @EzequielS2, since you're looking for help on implementing your forward/physics model in FEniCS, I strongly suggest you to post your question at the FEniCS forum, which is a more suitable place to ask questions about FEniCS.

I'd like to remind you that this project/repository is more on interfacing CUQIpy and FEniCS for inverse problems. Personally, I'm more than happy to help, but I really don't have much experience with the Navier-Stokes equations. What do you think @amal-ghamdi?

EzequielS2 commented 6 months ago

Thanks!!

amal-ghamdi commented 6 months ago

Thank you @EzequielS2 so much for the details about your problem and thank you @chaozg for your reply. I agree with @chaozg, I have a very good experience using FEniCS forum mentioned by @chaozg above. I would recommend posting your problem there.

This repository, as @chaozg mentioned is related to solving PDE-based Bayesian inverse problems by combining CUQIpy and FEniCS. If you are interested to know more about the type of problems we address, feel free to check our recently published paper https://iopscience.iop.org/article/10.1088/1361-6420/ad22e8. I hope this address the issue you raised and feel free to follow up with related questions.

Best, Amal