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.

The convergency of result in differnent order of BC

Please login with a confirmed email address before reporting spam

Here we will show two different results in simulation but in the same boundary condition(BC). So it must be something wrong. To the geomertry, two Al cicles r=0.43m in a EP rectangular 1*2m, we apply the prediscribed displacement at the left boundary of the rectangular, and apply the periodic boundary condition(PBC) at the up and down boundaries. In the end, we caculate the frequency domain problem, and discuss the deformed mode in 1440Hz. There are two PBC we can consided, the continuity and the Floquet periodicity BC. And we know that when kx=ky=0 in the Floquet periodity, the two BC are equivalent and we can get a same result, a same deformed mode. Now, we will apply each BC, the continuity or the Floquet periodicity(kx=ky=0) in each calculation. It means, we make the simulation with the continuity BC firstly, and then we only exchang the Floquet periodicity BC from the continuity BC. we can predict the two results are similar. But when we exchange the continuity BC from the Floquet periodicity BC again. It shows a very different result. So we suspect the fault occures in the echanging process, maybe it's a bug of the software. I'm not sure. So we ask for help here, to discuss this problem. We get the same results in different comsol version 4.4, 5.2 or 5.2a. Thanks a lot, if you can give ant help.



4 Replies Last Post 2017年10月17日 GMT-4 05:06
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2017年10月16日 GMT-4 09:00

Hi Jiachen,

This is indeed a bug.

If you switch on Advanced Physics Options (in the Show menu at the top of the Model Builder pane), you will see that the setting of Apply reaction terms on in the Constraint Settings section is not propely reset when flipping back and forth.

You need manually to set Apply reaction terms on when moving back to Continuity.

Regards,
Henrik

-------------------
Henrik Sönnerlind
COMSOL
Hi Jiachen, This is indeed a bug. If you switch on **Advanced Physics Options** (in the **Show** menu at the top of the **Model Builder** pane), you will see that the setting of **Apply reaction terms on** in the **Constraint Settings** section is not propely reset when flipping back and forth. You need manually to set **Apply reaction terms on** when moving back to **Continuity**. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2017年10月16日 GMT-4 11:02

Hi Henrik, Thanks for you rapid and correct reply. I've test your method. It can perfect overcome this problem. Thank you very much! In fact, I'm not undestand the constraint settings setting. I means when should I set this configuration? Thanks.

Hi Henrik, Thanks for you rapid and correct reply. I've test your method. It can perfect overcome this problem. Thank you very much! In fact, I'm not undestand the constraint settings setting. I means when should I set this configuration? Thanks.

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2017年10月17日 GMT-4 03:52

Hi Jiachen,

The description in the Symmetric and Nonsymmetric Constraints section in the COMSOL Multiphysics Reference Manual gives a theoretical background.

Some other comments:

  • The Floquet conditions actually use a special type of reaction forces which is hardwired, so in this case the Individual Dependent Variables setting is more of an indication.
  • The main purpose of these settings it to make it possible for one physics to drive a boundary condition in another physics, without the connection being reflected back.
  • If you enable Equation View, it is possible to see how the constraints are implemented.

Regards, Henrik

-------------------
Henrik Sönnerlind
COMSOL
Hi Jiachen, The description in the **Symmetric and Nonsymmetric Constraints** section in the **COMSOL Multiphysics Reference Manual** gives a theoretical background. Some other comments: * The Floquet conditions actually use a special type of reaction forces which is hardwired, so in this case the **Individual Dependent Variables** setting is more of an indication. * The main purpose of these settings it to make it possible for one physics to drive a boundary condition in another physics, without the connection being reflected back. * If you enable **Equation View**, it is possible to see how the constraints are implemented. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 2017年10月17日 GMT-4 05:06

Hi Henrik, Thanks again!

Regards, Jiachen

Hi Henrik, Thanks again! Regards, Jiachen

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.