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 The information provided may be out of date.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.

In parametric sweep, is it possible to use the eigenfrequency of last solution as the seed to the current eigenvalue problem?

Please login with a confirmed email address before reporting spam

Hi,

I'm doing a parametric sweep over some parameter (kx). For each value of kx, I trying to find one eigenfrequency. But my problem is that I want to use the solution for the previous value of kx as an input to the current value of kx.

Specifically, I want to use the value of the previous eigenfrequency as an input to the "search around frequency... " field of the eigenvalue solver for the current kx. Is this possible?

I know that you can use the previous solution as the initial value of the new solution but it doesn't let me specify the eigenfrequency of the previous solution as the initial guess of eigenfrequency of the new problem.

Thank you.

4 Replies Last Post 30 oct. 2012, 08:31 UTC−4
COMSOL Moderator

Hello Hossein Mousavi

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.


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 12 déc. 2011, 18:36 UTC−5
Hi, I have exacly the same problem to solve. Do you know already how to do it?

Hi, I have exacly the same problem to solve. Do you know already how to do it?

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 13 déc. 2011, 01:16 UTC−5
Hi

my way is the manual one, first an eigenfrequency analysis, then a frequency sweep around one of the modes I identified previously. As often damping used are low, the "range()" definition around a peak is slightly delicate, one should get the points on both side of the resonance, but not just on, as the amplitude tend to make the solver diverge

--
Good luck
Ivar
Hi my way is the manual one, first an eigenfrequency analysis, then a frequency sweep around one of the modes I identified previously. As often damping used are low, the "range()" definition around a peak is slightly delicate, one should get the points on both side of the resonance, but not just on, as the amplitude tend to make the solver diverge -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30 oct. 2012, 07:35 UTC−4
Hi,

I have the same question. For computing a mechanical band structure I do a parametric sweep of the wave vector k. Now I want that, for every k, the computed eigenfrequency of the step before is used as "search around frequency". In that way I want to make sure that only one band is calculated and that I do not jump to another band while varying k.

Does anyone have an idea how to do that?

Thanks and best regards
Felix
Hi, I have the same question. For computing a mechanical band structure I do a parametric sweep of the wave vector k. Now I want that, for every k, the computed eigenfrequency of the step before is used as "search around frequency". In that way I want to make sure that only one band is calculated and that I do not jump to another band while varying k. Does anyone have an idea how to do that? Thanks and best regards Felix

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 30 oct. 2012, 08:31 UTC−4
Hi

that looks like my issue about how to define simply the time steps from a periodic function, such to have closely defined points around steep gradients.
Currently there is no easy function built-in COMSOl that I know about.
So probably one would have to call some outside Matlab function that can read in a time series and output a time stepping with intervalles inversely proportional to the slope of the input fucntion.

In your case it would be frequency steps based on a frequency domain amplitude signal previously calculated

Sorry i have no ready made answer for this one.

--
Good luck
Ivar
Hi that looks like my issue about how to define simply the time steps from a periodic function, such to have closely defined points around steep gradients. Currently there is no easy function built-in COMSOl that I know about. So probably one would have to call some outside Matlab function that can read in a time series and output a time stepping with intervalles inversely proportional to the slope of the input fucntion. In your case it would be frequency steps based on a frequency domain amplitude signal previously calculated Sorry i have no ready made answer for this one. -- 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.