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.
nonlinear refractive index
Posted Dec 2, 2015, 3:14 p.m. EST RF & Microwave Engineering, Wave Optics, Materials, Modeling Tools & Definitions, Parameters, Variables, & Functions Version 5.0 2 Replies
Please login with a confirmed email address before reporting spam
Hello COMSOL Community,
How can I determine the value of the nonlinear refractive index for an geometry i build in COMSOL?
Best wishes,
Fabian
How can I determine the value of the nonlinear refractive index for an geometry i build in COMSOL?
Best wishes,
Fabian
2 Replies Last Post Jan 4, 2016, 3:34 a.m. EST
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi
I suppose you are referring to an optimization model there, that you can do in different ways in COMSOL, that you have a geometry, you know the results but you want to i.e. define the index of refraction such that it behaves as n=f(r). Where r being some spatial coordinate as r=sqrt((x-x0)^2+(y-y0)^2+(z-z0)^2) and you want to identify f().
If you start with a liner relationship you need to find a dependent scalar variable, that you define in a global value node in your model with a condition to be minimized (the error between the know result and the calculated one with your new variable.
You will find several models that uses this approach in the COMSOL examples, if your function is more complex you should consider to study and use the optimization module of COMSOL
--
Good luck
Ivar
I suppose you are referring to an optimization model there, that you can do in different ways in COMSOL, that you have a geometry, you know the results but you want to i.e. define the index of refraction such that it behaves as n=f(r). Where r being some spatial coordinate as r=sqrt((x-x0)^2+(y-y0)^2+(z-z0)^2) and you want to identify f().
If you start with a liner relationship you need to find a dependent scalar variable, that you define in a global value node in your model with a condition to be minimized (the error between the know result and the calculated one with your new variable.
You will find several models that uses this approach in the COMSOL examples, if your function is more complex you should consider to study and use the optimization module of COMSOL
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi Ivar,
I am also working on a similar kind of problem. I want to have some clarifications on my simulation. I am started with eps0 and got value of "emw.normE" over the domains of my model. In this case the simulation took "linear solver" mode. Now i want to simulate intensity dependent epsilon. So i replaced my eps simply with eps=eps0+A*mod(emw.normE)^2. In this case the solver automatically switched to nonlinear solver and converged solution obtained after few iterations. I have not changed anything in my study node especially to call the earlier simulated "emw.normE" at eps0 into the expression eps=eps0+A*mod(emw.normE)^2. Am i doing the right way or should i define somewhere in the study node?
Please also clarify how does it differ for a single wavelength and a certain sweep of wavelengths. if possible, please specify the earlier optimized model that available in COMSOL RF module.
Thanks in advance,
sriram
I am also working on a similar kind of problem. I want to have some clarifications on my simulation. I am started with eps0 and got value of "emw.normE" over the domains of my model. In this case the simulation took "linear solver" mode. Now i want to simulate intensity dependent epsilon. So i replaced my eps simply with eps=eps0+A*mod(emw.normE)^2. In this case the solver automatically switched to nonlinear solver and converged solution obtained after few iterations. I have not changed anything in my study node especially to call the earlier simulated "emw.normE" at eps0 into the expression eps=eps0+A*mod(emw.normE)^2. Am i doing the right way or should i define somewhere in the study node?
Please also clarify how does it differ for a single wavelength and a certain sweep of wavelengths. if possible, please specify the earlier optimized model that available in COMSOL RF module.
Thanks in advance,
sriram