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.

Acoustic resonances in a room with a Perfectly Matched Layer

Please login with a confirmed email address before reporting spam

Hello. I'm using COMSOL for an acoustics project, and I have trouble setting up the PML to work perfectly.

My case: A rectangular room with totally absorbing ceiling. The room is simply a block, and I have added another block on top which simulates the ceiling as a PML boundary.

However, when simulating, the resonance frequencies are somewhat higher than the theoretical resonances. They clearly depend on the PML set-up. Theoretically I believe the resonance frequencies should be reduced when adding a resistive element (the PML).

I've tried both cartesian and spherical spreading for the PML, the spherical works best with the sphere centre placed a bit under the floor, but this is still not ideal.

Any suggestions? See attached screenshots.


3 Replies Last Post 25 févr. 2013, 08:59 UTC−5
Sergei Yushanov Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25 févr. 2013, 08:38 UTC−5
Jakob,

PML means that you have infinite domain. But what is the definition of the resonant for infinite domain? I had the impression the resonance has physical meaning for finite domain only…

Regards,
Sergei
Jakob, PML means that you have infinite domain. But what is the definition of the resonant for infinite domain? I had the impression the resonance has physical meaning for finite domain only… Regards, Sergei

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25 févr. 2013, 08:42 UTC−5
Hi Sergei,

I am examining the 2D resonances in a 3D room (i.e. the horizontal components of the wave propagation). This cannot be done in a 2D simulation because the energy must be dissipated in the 3rd dimension (the ceiling). You can see the 2D resonances in the screenshot.

Basically the PML would be a sound absorbing material in the ceiling.

Regards,
Jakob
Hi Sergei, I am examining the 2D resonances in a 3D room (i.e. the horizontal components of the wave propagation). This cannot be done in a 2D simulation because the energy must be dissipated in the 3rd dimension (the ceiling). You can see the 2D resonances in the screenshot. Basically the PML would be a sound absorbing material in the ceiling. Regards, Jakob

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25 févr. 2013, 08:59 UTC−5
PML is a wrong boundary condition in this case.
If you want to model an absorbing material you have to add a Biot boundary condition, or D-B not a PML, as they are not meant to do what you want them to do.
What you are simulating is a room with no ceiling, it's not wrong but I guess it's not what you are interested in, usually rooms with no ceiling are rarely used... ;)
Take a look at this article.

www.stanford.edu/class/ee256/Berenger1994.pdf
PML is a wrong boundary condition in this case. If you want to model an absorbing material you have to add a Biot boundary condition, or D-B not a PML, as they are not meant to do what you want them to do. What you are simulating is a room with no ceiling, it's not wrong but I guess it's not what you are interested in, usually rooms with no ceiling are rarely used... ;) Take a look at this article. http://www.stanford.edu/class/ee256/Berenger1994.pdf

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.