Forum Message

 

 

We have moved the forum to https://forum.ngsolve.org . This is an archived version of the topics until 05/05/23. All the topics were moved to the new forum and conversations can be continued there. This forum is just kept as legacy to not invalidate old links. If you want to continue a conversation just look for the topic in the new forum.

Notice

The forum is in read only mode.

Suitability for anisotropic electrodynamics problem

More
2 years 5 months ago #4064 by Lange
Dear Forum,

I plan to compute the electromagnetic field distribution of an arrangement of dielectric elements each characterized by an anisotropic permittivity (including off-diagonal tensor components), and located in a box-shaped volume, in the frequency domain. The materials are non-magnetic, so \mu = 1. The side walls of the box should be periodic boundary conditions, while top and bottom should scatter an external wave of a defined frequency in and out of the volume.
Although I have browsed the website and forum for quite a while, I have not found a suitable example. Can anyone confirm that the above is feasible with ngsolve, and possibly point to an example file? Any hint would be highly appreciated.

Best regards,
Christoph Lange
More
2 years 5 months ago #4065 by joachim
Hi Christoph,

I think this paper deals with a similar setup as you have described:
M. Huber et al: "Simulation of Diffraction in periodic Media with a coupled Finite Element and Plane Wave Approach"
epubs.siam.org/doi/10.1137/070705118


Solving frequency domain Maxwell, with tensorial eps is standard in NGSolve, just write down the equations.
How big is the problem electrically ? 10 waves in every direction is easy, maybe 50 is doable with more or less standard methods on a large computer, but more waves require alternatives like time domain solvers.
Periodic boundary conditions are fine with CSG geometry (see docu), and coming soon for OCC geometry.
The coupling to plane waves (as shown in the paper) has required an extension code in C++.
Nowadays it is not a big deal to extend NGSolve such that also the plane waves in the exterior domain can modelled from Python, we can do this together, in the time-scale of a few weeks.

Looks interesting and doable,

Joachim
Time to create page: 0.170 seconds