time dependent neumann boundary condition #1793
Abhishekc95
started this conversation in
General
Replies: 3 comments 4 replies
-
@praksharma If you have a moment, I would appreciate your help |
Beta Was this translation helpful? Give feedback.
0 replies
-
I think your implementation is okay. It just that the baseline PINN really struggles with wave equation. You can look into papers like learning rate annealing PINN, or neural tangent kernel PINN. there are more works, such as time windows scheduling which is done in NVIDIA modulus to specifically solve such problems. |
Beta Was this translation helpful? Give feedback.
1 reply
-
read this paper. yes baseline PINN works with wave equation, but needs a lot of tinkering due to the imbalanced loss terms. |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
@lululxvi thank you for the library for implementing PINN.
Im trying to solve 1D wave propagation problem
boundary conditions
I was able to run the above code but,$t = 1/f$ and the total simulation time say $(T>t)$ .
My question is related to 4th b.c i.e neumann B.C, where i want to apply it for some time
Basically i want to see the wave traveling from one end to other.
What changes should i make to my code.
I didnt understand the naviers stokes discussion, where it sort of seems to have time stepping scheme. My area is in structures. Is there a time stepping algorithm in PINN for solving my problem?
Thanks in advance.
Beta Was this translation helpful? Give feedback.
All reactions