| Author |
Message |
milli
Joined: 22 May 2008
Posts: 13
|
 13 Aug 2008 18:12 Efficiency calculation of PML Boundary.. |
|
|
|
hi everyone..
in my project i am working with FDTD simulations with PML boundaries..
i am new to these topics..
can anyone explain me... how to find "EFFICIENCY OF PML"..
any material related to this will also be appreciated
thanks in advance..
|
|
| Back to top |
|
 |
milli
Joined: 22 May 2008
Posts: 13
|
| 20 Aug 2008 9:45 Re: Efficiency calculation of PML Boundary.. |
|
|
|
| someone reply to this plzzz
|
|
| Back to top |
|
|
rrumpf
Joined: 06 Jul 2007
Posts: 80
Helped: 10
|
| 23 Aug 2008 20:21 Re: Efficiency calculation of PML Boundary.. |
|
|
|
The "efficiency" of a PML is simply how much energy it absorbs (does not reflect) relative to the energy incident on it. You probably want to study how its efficiency is affected by angle of incidence, background material properties, frequency, etc.
While there are many way to set this up, here is one possible recipe. For simplicity I am assuming you want to do a 2D analysis of PML efficiency.
1. Setup a small problem space where the dielectric constant and permeability are all set to 1.0.
2. Use periodic boundary conditions on your x-low and x-high boundaries.
3. Use a PML boundary condition on your y-low and y-high boundaries. This will be what you test.
4. Use the total-field/scattered-field formulation to inject a one-way source into your problem. Set the amplitude 1.
5. Run your simulation and record the power reflected by the PML, Pref.
6. Calculate the power in your source (Pinc).
7. Calculate the "efficiency" of your PML: Eff = 1 - (Pref/Pinc)
8. Run your model to sweep frequency, angle of incidence, background material properties, etc.
Hope this helps!
-Tip
|
|
| Back to top |
|
|
milli
Joined: 22 May 2008
Posts: 13
|
| 24 Aug 2008 13:35 Re: Efficiency calculation of PML Boundary.. |
|
|
|
Thanks for the reply......
and i would like to know few things like....
what are the factors tht effect the efficiency..?
and may i know whether you have worked on this topic before ??
thanks in advance...
Regards,
milli
|
|
| Back to top |
|
|
