Menu

Discussion Forum

New Discussion
Filter

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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 do segregated solvers work ?

Posted Jul 6, 2011, 3:31 a.m. EDT Studies & Solvers Version 4.1 4 Replies

Yoann Many
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Sorry to bother you all one more time, but I think after this I would be done :)

I looked around the forum to obtain information about segregated solvers, but I didn't understand it so far. So here is my (multiple) question :

Can anyone explain how do segregated solvers work? What's the meaning of the segregated steps? is the order of the variables within a particular step important? please...

Thank you in advance!

4 Replies Last Post Oct 11, 2011, 8:14 a.m. EDT
COMSOL Moderator

Hello Yoann Many

Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.

If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.

Daniel Nussbächer
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 9, 2011, 8:49 a.m. EDT
I'm interested in this topic, too.
Is there anyone who could explain it?
Thx so far
I'm interested in this topic, too. Is there anyone who could explain it? Thx so far
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 Sep 10, 2011, 4:25 a.m. EDT
Hi

my understanding of the segragated steps are that you might groupe some dependent variables, solve for these and then use the results to solve again with the other set of comlementary dependent variables and then loop around until all variables have reached a reasonnable solution (global estimator is small).

This is particularly useful when you have either independent physics that are wekly linked, or a fast and a slow process. Basically you need less RAM andit solves quicker than if you put everything into the same pot and take all dependent variables in one go.

Then selecting which variables to choose first, second etc is a little tricky, COMSOL tries and achieves it quite well I find, often I can solve everything in one go, sa I have quite some RAM at hand (48Gb), at least for "small models" other times, when you have linked the variables in a way COMSOl does not see (i.e. via material properties etc) COMSOL suggests to regroup all variables in one step..

hope this helps

By the way do not forget the "LM"s that appear when you add complec BC

--
Good luck
Ivar
Hi my understanding of the segragated steps are that you might groupe some dependent variables, solve for these and then use the results to solve again with the other set of comlementary dependent variables and then loop around until all variables have reached a reasonnable solution (global estimator is small). This is particularly useful when you have either independent physics that are wekly linked, or a fast and a slow process. Basically you need less RAM andit solves quicker than if you put everything into the same pot and take all dependent variables in one go. Then selecting which variables to choose first, second etc is a little tricky, COMSOL tries and achieves it quite well I find, often I can solve everything in one go, sa I have quite some RAM at hand (48Gb), at least for "small models" other times, when you have linked the variables in a way COMSOl does not see (i.e. via material properties etc) COMSOL suggests to regroup all variables in one step.. hope this helps By the way do not forget the "LM"s that appear when you add complec BC -- Good luck Ivar
Sirisha Govindaraju
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 10, 2011, 5:49 p.m. EDT
If for some reason, you end up having only one segregated step for all variables, then will this be the same as using a fully coupled solver?

Thanks,
Sirisha
If for some reason, you end up having only one segregated step for all variables, then will this be the same as using a fully coupled solver? Thanks, Sirisha
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 Oct 11, 2011, 8:14 a.m. EDT
Hi

no in all cases it a fully coupled solver, the only thing is that the solver loops around in a segregated way.

I would rather say one segregated step = direct solving (TBC it's my first reaction)

--
Good luck
Ivar
Hi no in all cases it a fully coupled solver, the only thing is that the solver loops around in a segregated way. I would rather say one segregated step = direct solving (TBC it's my first reaction) -- Good luck Ivar

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