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 The information provided may be out of date.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.

Adding artificial diffusion/stabilization to PDE

Please login with a confirmed email address before reporting spam

Hi folks,

can anyone point me to an example or model in the library where stabilization is added to a PDE.

I've seen in the forum someone saying "transport problem" example, but it was in V3.5a, and I am using 4.3
and no access to 3.5a.

Any thoughts?

Thanks in advance.

Cheers.

Rui

5 Replies Last Post Oct 15, 2012, 9:22 a.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 2, 2012, 12:29 p.m. EDT
Anyone?
Anyone?

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2012, 4:41 a.m. EDT
Dear Rui,

I understand that you are looking for some stabiilzation method for the PDE form, either coefficient, or general.
Am iI right ?

As far as I know, at least in 4.2, there is no such ready-to-use stabilization feature for these modes. You can get such
features only in "specialized modes", such as heat transfer, fluid flow or mass diffusion.

This is really too bad, because some models are really easy to program in PDE form, but cannot converge without
stabilization.

I wonder why stabilization was not implemented in PDE modes. Is it a technical lock, or a commercial one ?
If you get some information, I'm interested.

Best regards.
Dear Rui, I understand that you are looking for some stabiilzation method for the PDE form, either coefficient, or general. Am iI right ? As far as I know, at least in 4.2, there is no such ready-to-use stabilization feature for these modes. You can get such features only in "specialized modes", such as heat transfer, fluid flow or mass diffusion. This is really too bad, because some models are really easy to program in PDE form, but cannot converge without stabilization. I wonder why stabilization was not implemented in PDE modes. Is it a technical lock, or a commercial one ? If you get some information, I'm interested. Best regards.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 12, 2012, 6:00 a.m. EDT
Hello Olivier,

it's a shame, really. I think that would push COMSOL to a whole new level, in my opinion.

Maybe in the near future we can get them, however, it does not help us at the moment.

I have not used it, but after contacting the Support they pointed out two examples that use stabilization in PDE's.

The models in the library are:

- transport problem (version 3.5a);
- shock tube (4.2,4.2a,4.3);

I believe it's just Streamline Diffusion for both examples, but it's a start. No Crosswind Diffusion, though.

If you follow the literature reference in the manual and look at the examples, actually it's not that overwhelming.

Hope it helps!

Best Regards,

Rui Silva




Dear Rui,

I understand that you are looking for some stabiilzation method for the PDE form, either coefficient, or general.
Am iI right ?

As far as I know, at least in 4.2, there is no such ready-to-use stabilization feature for these modes. You can get such
features only in "specialized modes", such as heat transfer, fluid flow or mass diffusion.

This is really too bad, because some models are really easy to program in PDE form, but cannot converge without
stabilization.

I wonder why stabilization was not implemented in PDE modes. Is it a technical lock, or a commercial one ?
If you get some information, I'm interested.

Best regards.


Hello Olivier, it's a shame, really. I think that would push COMSOL to a whole new level, in my opinion. Maybe in the near future we can get them, however, it does not help us at the moment. I have not used it, but after contacting the Support they pointed out two examples that use stabilization in PDE's. The models in the library are: - transport problem (version 3.5a); - shock tube (4.2,4.2a,4.3); I believe it's just Streamline Diffusion for both examples, but it's a start. No Crosswind Diffusion, though. If you follow the literature reference in the manual and look at the examples, actually it's not that overwhelming. Hope it helps! Best Regards, Rui Silva [QUOTE] Dear Rui, I understand that you are looking for some stabiilzation method for the PDE form, either coefficient, or general. Am iI right ? As far as I know, at least in 4.2, there is no such ready-to-use stabilization feature for these modes. You can get such features only in "specialized modes", such as heat transfer, fluid flow or mass diffusion. This is really too bad, because some models are really easy to program in PDE form, but cannot converge without stabilization. I wonder why stabilization was not implemented in PDE modes. Is it a technical lock, or a commercial one ? If you get some information, I'm interested. Best regards. [/QUOTE]

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 13, 2012, 7:19 a.m. EDT
Hi

have you tried to send this demand to "support", they are active partner to talk with, OK they decide the priorities but thhey put all interesting suggestions up on the to do list, and I have noticed, one or two releases later many things are there.

Another way is to come to the annual conferences and discuss it directly with the program managers and the COMSOl developpers

--
Good luck
Ivar
Hi have you tried to send this demand to "support", they are active partner to talk with, OK they decide the priorities but thhey put all interesting suggestions up on the to do list, and I have noticed, one or two releases later many things are there. Another way is to come to the annual conferences and discuss it directly with the program managers and the COMSOl developpers -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Oct 15, 2012, 9:22 a.m. EDT
Hello Ivar,

that's a very good suggestion.

I just use the normal support contact to do this suggestion?
It is going to be very difficult in the near future to attend a COMSOL conference, although i'd love to.

Best Regards,

Rui Silva
Hello Ivar, that's a very good suggestion. I just use the normal support contact to do this suggestion? It is going to be very difficult in the near future to attend a COMSOL conference, although i'd love to. Best Regards, Rui Silva

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.