Menu

Discussion Forum

New Discussion
Filter

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.

How to plot the relative error?

Posted Apr 29, 2013, 3:42 a.m. EDT Results & Visualization, Studies & Solvers 5 Replies

Evgeni Sergeev
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

I've just spent more than a whole day troubleshooting my model.
However, if I could plot the relative error over the geometry, I believe it would have been solved in minutes.
There was a problem localised to a particular junction between domains.

So, is there any way to visualise the relative error in a domain? Maybe via Livelink for Matlab? I couldn't find anything about that.
5 Replies Last Post Nov 23, 2016, 2:04 p.m. EST
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 29, 2013, 6:11 a.m. EDT
Hi

I fully agree with you that a better way to find the variables names, typically of such internal variables that gives us a better understanding where the solver is locked would help for debugging.

Today you need to scan many pages o help files and still after 30 min I did not manage to localise the L2 norm variable name, neither not find it in the "equation view" (and there are no specific equation view for the solvers)

Try to push it over to "support" if you explain clearly why it's useful, then you might see it appear in next revision

--
Good luck
Ivar
Hi I fully agree with you that a better way to find the variables names, typically of such internal variables that gives us a better understanding where the solver is locked would help for debugging. Today you need to scan many pages o help files and still after 30 min I did not manage to localise the L2 norm variable name, neither not find it in the "equation view" (and there are no specific equation view for the solvers) Try to push it over to "support" if you explain clearly why it's useful, then you might see it appear in next revision -- Good luck Ivar
Evgeni Sergeev
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 1, 2013, 1:26 a.m. EDT
Hi Ivar,

I've checked with support. Summarising:
• Currently there is indeed no way to access the variable.
• Others have requested this feature too.
• The variable is planned to be exposed for post-processing in a later version.
• It's not known which version that will be.

Evgeni
Hi Ivar, I've checked with support. Summarising: • Currently there is indeed no way to access the variable. • Others have requested this feature too. • The variable is planned to be exposed for post-processing in a later version. • It's not known which version that will be. Evgeni
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago May 1, 2013, 2:22 a.m. EDT
Hi

then probably its not in next release I have heard will come soon, to bad, would be helpful indeed :)

--
Good luck
Ivar
Hi then probably its not in next release I have heard will come soon, to bad, would be helpful indeed :) -- Good luck Ivar
Camille S.
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago May 21, 2015, 5:50 a.m. EDT
Hello,

I would like to know if it is now possible to plot the relative error.
Actually, my aim is to understand the influence on my result of the factor in error estimate of the Direct Time-Dependant Solver. That is why I need to plot the relative error.

Best regards
Hello, I would like to know if it is now possible to plot the relative error. Actually, my aim is to understand the influence on my result of the factor in error estimate of the Direct Time-Dependant Solver. That is why I need to plot the relative error. Best regards
DW
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 8 years ago Nov 23, 2016, 2:04 p.m. EST
See the blog "plotting-the-algebraic-residual-to-study-model-convergence"
www.comsol.com/blogs/plotting-the-algebraic-residual-to-study-model-convergence/

-DW
See the blog "plotting-the-algebraic-residual-to-study-model-convergence" https://www.comsol.com/blogs/plotting-the-algebraic-residual-to-study-model-convergence/ -DW

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.

Suggested Content