Menu

Discussion Forum

New Discussion
Filtrer

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

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.

Initial conditions (steady state) to time dependent model

Posted 14 août 2014, 16:29 UTC−4 Studies & Solvers Version 4.4 5 Replies

Aaron Shifman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hello All,

I have a model where I want to find the steady-state solution to a time dependent model, I let it run with a large time block. Now I want to use this as the initial condition to the next study step. The issue I have is that there is a massive solution stored in memory. Is there a way of taking the last time step of a study, saving it as a steady state solution and forgetting the rest (or better yet, as the time dependent solver progresses, only keep the previous solution in memory while solving)?

Thanks,
Aaron R. Shifman
5 Replies Last Post 18 août 2014, 02:19 UTC−4
Lasse Murtomäki
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 15 août 2014, 04:20 UTC−4
I would add a stationary study and use that as an initial condition.
I would add a stationary study and use that as an initial condition.
Pieter Westerik
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 15 août 2014, 05:39 UTC−4
I believe the solution is only stored at the time intervals that you specify in the time dependent study. COMSOL will calculate time steps in between if that is necessary for accuracy. So you can specify that you want only the solution at the end time, everyting in between is not saved I believe. But indeed, since you want a steady state solution anyways, why not use the steady state solver?
I believe the solution is only stored at the time intervals that you specify in the time dependent study. COMSOL will calculate time steps in between if that is necessary for accuracy. So you can specify that you want only the solution at the end time, everyting in between is not saved I believe. But indeed, since you want a steady state solution anyways, why not use the steady state solver?
Aaron Shifman
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 15 août 2014, 08:09 UTC−4
Thanks for the input,

I would like to use a stationary solver, but I don't think I can since I have temporal derivatives built into the model, and I don't know of a way to fake time in a stationary model, is there?

Thanks,
Aaron
Thanks for the input, I would like to use a stationary solver, but I don't think I can since I have temporal derivatives built into the model, and I don't know of a way to fake time in a stationary model, is there? Thanks, Aaron
Pieter Westerik
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 15 août 2014, 08:11 UTC−4
In steady state nothing changes with time and therefore time derivatives are by definition zero. I am not sure if COMSOL handles this automatically, otherwise you have to make a copy of your physics for the steady state study where you put the time derivatives to zero.
In steady state nothing changes with time and therefore time derivatives are by definition zero. I am not sure if COMSOL handles this automatically, otherwise you have to make a copy of your physics for the steady state study where you put the time derivatives to zero.
Lasse Murtomäki
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 18 août 2014, 02:19 UTC−4

Thanks for the input,

I would like to use a stationary solver, but I don't think I can since I have temporal derivatives built into the model, and I don't know of a way to fake time in a stationary model, is there?

Thanks,
Aaron


Of course you can, just add another study and look what happens. Steady-state solutions do not take very long time.

Lasse
[QUOTE] Thanks for the input, I would like to use a stationary solver, but I don't think I can since I have temporal derivatives built into the model, and I don't know of a way to fake time in a stationary model, is there? Thanks, Aaron [/QUOTE] Of course you can, just add another study and look what happens. Steady-state solutions do not take very long time. Lasse

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