Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Feb 11, 2013, 7:06 p.m. EST
Model attached.
Model attached.
Nils Malm
COMSOL Employee
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Feb 12, 2013, 3:14 a.m. EST
Hi Will,
While PDE interfaces are not disabled in domains with PMLs or Infinite Elements (as other physics not supporting them), the PML/IE does not actually apply. That is, the PDE is not sensitive to the domain's PML status in any way. The underlying reason for this is that some people may want to, e.g., set up an equation solving for a material property or computing some kind of auxiliary quantity - which would not have been possible if the PDEs had been disabled in PML and IE domains. Since the PDE interfaces are meant for advanced modeling, we cannot assume anything about what users want to use them for. Therefore, we allow almost everything and do not try to interpret or transform the equations in any way. So for the PDE interfaces, what you specify is essentially always what you get.
If you are modeling an equation which is not covered by any COMSOL module supporting PMLs, you can still use the transformation matrix and Jacobian determinant defined by the built-in PMLs in the PML domains to define your own scaled equations. This, however, is only possible (or at least straight-forward) when modeling in the weak form. Another alternative, which lets you hand over the PML transformations to the COMSOL machinery, is to set up your equation as a new physics interface using the Physics Builder.
Hi Will,
While PDE interfaces are not disabled in domains with PMLs or Infinite Elements (as other physics not supporting them), the PML/IE does not actually apply. That is, the PDE is not sensitive to the domain's PML status in any way. The underlying reason for this is that some people may want to, e.g., set up an equation solving for a material property or computing some kind of auxiliary quantity - which would not have been possible if the PDEs had been disabled in PML and IE domains. Since the PDE interfaces are meant for advanced modeling, we cannot assume anything about what users want to use them for. Therefore, we allow almost everything and do not try to interpret or transform the equations in any way. So for the PDE interfaces, what you specify is essentially always what you get.
If you are modeling an equation which is not covered by any COMSOL module supporting PMLs, you can still use the transformation matrix and Jacobian determinant defined by the built-in PMLs in the PML domains to define your own scaled equations. This, however, is only possible (or at least straight-forward) when modeling in the weak form. Another alternative, which lets you hand over the PML transformations to the COMSOL machinery, is to set up your equation as a new physics interface using the Physics Builder.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
May 6, 2014, 11:15 a.m. EDT
Hi Will,
Are you able to implement PMLs in PDE now? I am having the same problem now.
Thanks,
Xiang
Hi Will,
Are you able to implement PMLs in PDE now? I am having the same problem now.
Thanks,
Xiang
Please login with a confirmed email address before reporting spam
Posted:
10 years ago
Apr 26, 2015, 1:50 p.m. EDT
Hi everybody,
Are you able to implement PMLs in a weak form? I am having the same problem now.
Thanks,
Ulac.
Hi everybody,
Are you able to implement PMLs in a weak form? I am having the same problem now.
Thanks,
Ulac.