Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Different results in different COMSOL versions

Shumsun Nahar Siddique

Please login with a confirmed email address before reporting spam

Hi there

I have been used COMSOL version 3.5a for my work. But now I am trying to use COMSOL version 4.2a. I rebuilt one of my old version model in new versions but new version gave me different results compared to that of my old version. Two screen shoot of same point, (x,y)=(65m, 14.75m) of my domain has been attached with this mail. Conceptually, my old model result is correct.

I use Darcy law, Subsurface flow-solute transport and one PDE helping equation. All the equations and constants are exactly same in both versions. The boundary conditions are also same (as far as I can understand new version). Why these results are so different for each other? Can someone help me regarding this issue? I have attached both versions models for further information. I really want suggestion.

Thank you,
Shums


5 Replies Last Post Jul 1, 2012, 4:36 p.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 29, 2012, 1:36 p.m. EDT
I have just briefly looked through your models and Smax values are different in the models.

When I have more time I will take a closer look. Some time ago I had a similar problem, in my case it was not a COMSOL error rather that the model in 4.2 had to be configured differently - I think your model also suffers from this issue.

regards,
I have just briefly looked through your models and Smax values are different in the models. When I have more time I will take a closer look. Some time ago I had a similar problem, in my case it was not a COMSOL error rather that the model in 4.2 had to be configured differently - I think your model also suffers from this issue. regards,

Shumsun Nahar Siddique

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 29, 2012, 1:54 p.m. EDT
Hello Mahtadin

Thank you very much for your response. The good news is the results of the both versions are similar after changing some issue in mesh and discretization orders. By adding a boundary layer mesh to boundary number 6, and change the discretization order for solute transport from linear to quadratic, the results are much similar now.

But unfortunately, the new version does not show smooth graph. I have tried ''time discrete solver'' and changing Relative tolerance and absolute tolerance. But till do not get smooth graph (attached: screen shoot of this graph ). Can you please suggest something about this issue?

I have also attached the latest version of my model.

Thank you.
Shums
Hello Mahtadin Thank you very much for your response. The good news is the results of the both versions are similar after changing some issue in mesh and discretization orders. By adding a boundary layer mesh to boundary number 6, and change the discretization order for solute transport from linear to quadratic, the results are much similar now. But unfortunately, the new version does not show smooth graph. I have tried ''time discrete solver'' and changing Relative tolerance and absolute tolerance. But till do not get smooth graph (attached: screen shoot of this graph ). Can you please suggest something about this issue? I have also attached the latest version of my model. Thank you. Shums


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 30, 2012, 10:32 a.m. EDT
Hm, to my understanding the little jump at the top of the sigmoid is because of saturation - mostly relating to immobile fluid. Why it occurs in one version and not on the other, I have no clue.

It would seem that the issue lies in solute transport. I used S01 in solute source 1 and S02 as solute source 2. I got the results found in the attachment - it gives a smooth curve, I leave it to you to determine the physical meaning.
Hm, to my understanding the little jump at the top of the sigmoid is because of saturation - mostly relating to immobile fluid. Why it occurs in one version and not on the other, I have no clue. It would seem that the issue lies in solute transport. I used S01 in solute source 1 and S02 as solute source 2. I got the results found in the attachment - it gives a smooth curve, I leave it to you to determine the physical meaning.


Shumsun Nahar Siddique

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 30, 2012, 2:23 p.m. EDT
Dear Mohtadin

Thank you very much for your suggestion. Yes, the graph is smooth now. And surly I will find the physical meaning behind it and let you know.

Thank you once again.

Shums
Dear Mohtadin Thank you very much for your suggestion. Yes, the graph is smooth now. And surly I will find the physical meaning behind it and let you know. Thank you once again. Shums

Shumsun Nahar Siddique

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jul 1, 2012, 4:36 p.m. EDT
Dear Mohtadin

According to your suggestion, I have changed the S01 and S02 terms and got smooth curves for 10years with time step 10days. But when I run the simulations for 25 years with time step 10days, the graphs are not smooth. The model can generate smooth graphs at y=14.75m , but when I check the graphs below that y, do not get any smooth graphs (attached screenshot).

I have tried different versions to solve this issue (like-version 4.2a and version 4.2), but face same problem. I have also tried different solvers. Can you please suggest something about this issue?

The latest model has been also attached with this mail for your kind consideration.

Thank you very much.
Shums
Dear Mohtadin According to your suggestion, I have changed the S01 and S02 terms and got smooth curves for 10years with time step 10days. But when I run the simulations for 25 years with time step 10days, the graphs are not smooth. The model can generate smooth graphs at y=14.75m , but when I check the graphs below that y, do not get any smooth graphs (attached screenshot). I have tried different versions to solve this issue (like-version 4.2a and version 4.2), but face same problem. I have also tried different solvers. Can you please suggest something about this issue? The latest model has been also attached with this mail for your kind consideration. Thank you very much. Shums

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.