Transient simulation on porous media

William Anthonius Wiyono

Please login with a confirmed email address before reporting spam

When I first run a shorter model (0, 1, 10) days my model works fine.

However, when I tried to conduct the complete modeling, I ran into this issue, any work around on this particular problem? (0, 1, 8035) days


3 Replies Last Post May 2, 2024, 4:18 p.m. EDT
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 months ago May 2, 2024, 10:25 a.m. EDT

Hello William,

You wrote "I ran into this issue". What is the issue you are speaking of, precisely?

Jeff

-------------------
Jeff Hiller
Hello William, You wrote "I ran into this issue". What is the issue you are speaking of, precisely? Jeff

William Anthonius Wiyono

Please login with a confirmed email address before reporting spam

Posted: 6 months ago May 2, 2024, 1:59 p.m. EDT

Hi Jeff,

Thank you for responding, I received this error message earlier..

*Feature: Time-Dependent Solver 1 (sol1/t1) Repeated error test failures. May have reached a singularity. Time: 5227199.99999916s Last time step is not converged. * I actually fixed this issue by changing the solver from MUMPS to PARDISO. But, then I faced a new issue... When I try to scale up my model (previously its in a smaller scale for predicting the end results), I ran to a similar issue.

*Repeated error test failures. May have reached a singularity. Time: 0 s. Last time step is not converged. *

Here, I have tried many things, such as changing the solver back from PARDISO to MUMPS, changing the newton itteration from automatic to constant. But still no luck.

Seeking for some guidance and help.


William A

Hi Jeff, Thank you for responding, I received this error message earlier.. *Feature: Time-Dependent Solver 1 (sol1/t1) Repeated error test failures. May have reached a singularity. Time: 5227199.99999916s Last time step is not converged. * I actually fixed this issue by changing the solver from MUMPS to PARDISO. But, then I faced a new issue... When I try to scale up my model (previously its in a smaller scale for predicting the end results), I ran to a similar issue. *Repeated error test failures. May have reached a singularity. Time: 0 s. Last time step is not converged. * Here, I have tried many things, such as changing the solver back from PARDISO to MUMPS, changing the newton itteration from automatic to constant. But still no luck. Seeking for some guidance and help. --- William A

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 6 months ago May 2, 2024, 4:18 p.m. EDT

Regarding the first error message: I would suggest looking closely at what happens to your model's inputs at t=5227200[s] (a.k.a. 60.5 days). Does something change abruptly? That could be the source of the problem.

Regarding the second error, it's harder to tell without the mph file what the problem could be, but certainly it is not shocking that by changing the scale of a model you may have to modifying other settings as well: the physics may be quite different at different scales and therefore everything from physics settings to mesh settings to solver settings need to be revisited.

Jeff

-------------------
Jeff Hiller
Regarding the first error message: I would suggest looking closely at what happens to your model's inputs at t=5227200[s] (a.k.a. 60.5 days). Does something change abruptly? That could be the source of the problem. Regarding the second error, it's harder to tell without the mph file what the problem could be, but certainly it is not shocking that by changing the scale of a model you may have to modifying other settings as well: the physics may be quite different at different scales and therefore everything from physics settings to mesh settings to solver settings need to be revisited. Jeff

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.