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.

Problem in Boundary Integration

Please login with a confirmed email address before reporting spam

Hi,

I am having some issues with the command POSTINT. I am tyring to solve a much more complex problem than outlined in the attached document. However, if somebody can point out what is the issue in the sample problem attched, I can go figure out the issue in my original problem.

3
==============\
|*********************\
|****del^u=0*********\ 4 (u=1)
|***********************\
| 1 (u=0) *************/
|**********************/ 5 (u=1)
|********************/
=============/
2


I am solving Laplace's equation in the above 2D-domain where all the edges except 1,4 and 5 are insulated. Thus the current density (=-del(u)) when integrated over any boundary edge should give me the total current passing through the edge. I use:

% Current passing into the system through edges 4 and 5

v1 = postint(fem,'-(ux*(-nx)+uy*(-ny))','dl',[4,5],'edim',1)

% Current passing out of the system through edge 1

v2 = postint(fem,'-(ux*(nx)+uy*(ny))','dl',[1],'edim',1)

Since the edges 3 and 2 are insulated (i.e. del(u)=0), and the gometry is symmetric I should expect v1 = v2. But I get,

v1 = 0.8553
v2 = 0.8776

Why this discrepancy ? I have attached the code for your perusal and also the results.

Regards
Murali


4 Replies Last Post 8 ago 2012, 01:01 GMT-4
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 16 feb 2012, 02:30 GMT-5
Hi

have you checked how this difference changes with the mesh, some % difference in flux estimation is not necesarily surprising.

Check also the KB on the amin site about flux estimation

--
Good luck
Ivar
Hi have you checked how this difference changes with the mesh, some % difference in flux estimation is not necesarily surprising. Check also the KB on the amin site about flux estimation -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 16 feb 2012, 18:57 GMT-5
Hi Ivar,

Yes I have checked the dependence on meshing. It is still the same. I did not quite understand what you meant by "KB" and "amin site" ? Can you please explain ?

Thanks
Murali
Hi Ivar, Yes I have checked the dependence on meshing. It is still the same. I did not quite understand what you meant by "KB" and "amin site" ? Can you please explain ? Thanks Murali

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 17 feb 2012, 02:42 GMT-5
I think the issues discussed in chapter "Computing Accurate Fluxes" of the Comsol (3.5a) User's Guide apply to your problem. You should be able to get much more precision in post-processing by "summing the reaction force" along the boundary rather than integrating the flux.
I think the issues discussed in chapter "Computing Accurate Fluxes" of the Comsol (3.5a) User's Guide apply to your problem. You should be able to get much more precision in post-processing by "summing the reaction force" along the boundary rather than integrating the flux.

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 8 ago 2012, 01:01 GMT-4
Hi

KB is the Knowledge Base of the "main" (typo sorry) COMSOL site.

see www.comsol.eu/support/knowledgebase/browse/900/

that is where you find most tricks directly commented from COMSOL

--
Good luck
Ivar
Hi KB is the Knowledge Base of the "main" (typo sorry) COMSOL site. see http://www.comsol.eu/support/knowledgebase/browse/900/ that is where you find most tricks directly commented from COMSOL -- 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.