Menu

Discussion Forum

New Discussion
Filtrer

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 Optics not converge

Posted 8 déc. 2015, 19:03 UTC−5 Studies & Solvers 3 Replies

Group Liu
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hello everyone,

I am trying to use Comsol to simulate a nonlinear optics problem using in the frequency domain. More specifically, I use n=n_0+gamma*power as the form to include third nonlinearity. Comsol can handle the simulation when nonlinearity is not very strong (\Delta_n=gamma*power<0.01), but when the nonlinearity is increased, Comsol cannot converge. Anyone has met similar problem, and could you guys give me some suggestions?

Thanks!
3 Replies Last Post 13 déc. 2015, 01:10 UTC−5
Maxim Electromobility
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 10 déc. 2015, 10:44 UTC−5
Hi Group Liu,

Solution of nonlinear problem depends on the initial conditions. If you know more or less how your solution should look like, you can use this field distribution as an initial conditions (initial guess) to improve the convergence. For instance a solution of linear problem with gamma=0 could be a good initial guess. Comsol uses zero fields by default.

And one has to keep in mind, that not every nonlinear problem has a stable solution in principle.

What can help is a solution tracing during calculations. Comsol can display how the field distribution evolved after each iteration step. In this way you will see how exactly Comsol tries to find the solution, and take it in to the account by adjusting the initial conditions accordingly.

Best regards,
Maxim
Hi Group Liu, Solution of nonlinear problem depends on the initial conditions. If you know more or less how your solution should look like, you can use this field distribution as an initial conditions (initial guess) to improve the convergence. For instance a solution of linear problem with gamma=0 could be a good initial guess. Comsol uses zero fields by default. And one has to keep in mind, that not every nonlinear problem has a stable solution in principle. What can help is a solution tracing during calculations. Comsol can display how the field distribution evolved after each iteration step. In this way you will see how exactly Comsol tries to find the solution, and take it in to the account by adjusting the initial conditions accordingly. Best regards, Maxim
Group Liu
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 11 déc. 2015, 12:41 UTC−5
Hi Maxim,

Thanks for your suggestions. I have tried to search this topic, but never find a general solution to this problem. I guess it might because each nonlinear problem is not the same. I would look into my problem more carefully and see what is the hidden cause for the non-convergence.

Thanks!
Hi Maxim, Thanks for your suggestions. I have tried to search this topic, but never find a general solution to this problem. I guess it might because each nonlinear problem is not the same. I would look into my problem more carefully and see what is the hidden cause for the non-convergence. Thanks!
Pu
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 9 years ago 13 déc. 2015, 01:10 UTC−5
Hi, Maxim!

Good to see you here.
--
Pu, ZHANG
School of Physics, HUST.
Hi, Maxim! Good to see you here. -- Pu, ZHANG School of Physics, HUST.

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.

Suggested Content