Menu

Discussion Forum

New Discussion
Filtra

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 The information provided may be out of date.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 values from a PDE

Posted 31 mag 2012, 02:04 GMT-4 Studies & Solvers Version 4.2a 8 Replies

Jens Krause
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hallo Comsol - expert,

I am trying to setup a transient simulation where the initial values of the dofs are
calculated in a dedicated physics node. All is in a 2-D plane.

The transient simulation is a PDE General Form ("g") with Hx, Hy as unknowns (think of
magnetic field strengths). I am interested in the transient behaviour starting from
a particular crafted initial state.

For the initial state I want to solve a stationary magnetis simulation ("mf", unknown Az).
From the solution I can derive the magnetic field strength (mf.Hx and mf.Hy) which are
the desired initial values for "g".

How can I transfer the output data of one physics section as initial values to anoteher
physics section?

I tried:
1. Study1 only solves "mf"
2. Study2 only solves in "g"
3. Study2 in the section "Values of variables that you do not solve for" I refer to Study1
(Method ="Solution" and Study="Study"
4. In "g" I use mf.Hx and mf.Hy as initial values.

BUT the initial values are all zero.

Does anyone have an idea to fix my approach or different ideas?
I could live with saving data to a file and reloading it.

Thanks for any hints.

Jens


8 Replies Last Post 4 giu 2012, 05:27 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 31 mag 2012, 02:15 GMT-4
Hi

check the help and the doc about the Solver Values of Dependent Variables tab, and the "Dependent Variables" sub node here you define where you get the initial values from or the solved and the unsolved variables

--
Good luck
Ivar
Hi check the help and the doc about the Solver Values of Dependent Variables tab, and the "Dependent Variables" sub node here you define where you get the initial values from or the solved and the unsolved variables -- Good luck Ivar
Jens Krause
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 31 mag 2012, 02:29 GMT-4
Hi Ivar,

I did that already and tried all possible settings and all remains zero. The help
refers to using solutions from different study steps but for the same physics.

What I need is cross initialisation.


Jens
Hi Ivar, I did that already and tried all possible settings and all remains zero. The help refers to using solutions from different study steps but for the same physics. What I need is cross initialisation. Jens
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 31 mag 2012, 03:01 GMT-4
Hi

then have you tried to use the variables from the previous models in the initial tab ? If you work on the same geoemtry it should go

--
Good luck
Ivar
Hi then have you tried to use the variables from the previous models in the initial tab ? If you work on the same geoemtry it should go -- Good luck Ivar
Jens Krause
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 31 mag 2012, 03:23 GMT-4

Hi

then have you tried to use the variables from the previous models in the initial tab ? If you work on the same geoemtry it should go

--
Good luck
Ivar




Hi Ivar,

well I used "mf.Hx" in the initial tab. Is that what you mean with "variables"?

I attach a simplified model, which just does initialsation in the transient study.

Thanks

Jens
[QUOTE] Hi then have you tried to use the variables from the previous models in the initial tab ? If you work on the same geoemtry it should go -- Good luck Ivar [/QUOTE] Hi Ivar, well I used "mf.Hx" in the initial tab. Is that what you mean with "variables"? I attach a simplified model, which just does initialsation in the transient study. Thanks Jens

Attachments:

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 giu 2012, 02:55 GMT-4
Hi

I'm used to have both solver sequences under the same study node, then I'm sure the link is done.

But what puzzles me is that whatever I use as initial conditions, your PDE solves always to the same patterns and values, so Ibelieve something is not as you expected in your PDE, and that the transmission of the initial condition is done correctly, either in my way or in yours with 2 studies, if you link them at the dependent variable node


--
Good luck
Ivar
Hi I'm used to have both solver sequences under the same study node, then I'm sure the link is done. But what puzzles me is that whatever I use as initial conditions, your PDE solves always to the same patterns and values, so Ibelieve something is not as you expected in your PDE, and that the transmission of the initial condition is done correctly, either in my way or in yours with 2 studies, if you link them at the dependent variable node -- Good luck Ivar
Jens Krause
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 giu 2012, 03:35 GMT-4
Hi,

I tried both and also when the stationary is in the same study the link is
not done.

On the PDE node I do not change the settings after creating
the node (except for the initial conditions).

I'm asking support,

Thanks,

Jens
Hi, I tried both and also when the stationary is in the same study the link is not done. On the PDE node I do not change the settings after creating the node (except for the initial conditions). I'm asking support, Thanks, Jens
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 giu 2012, 04:07 GMT-4
Hi

For me they are passing correctly (or as COSMOl expect them to), but I suspect you have something wrong in your PDE physics.
Have you tried to solve your PDE alone with different initial values ?

--
Good luck
Ivar
Hi For me they are passing correctly (or as COSMOl expect them to), but I suspect you have something wrong in your PDE physics. Have you tried to solve your PDE alone with different initial values ? -- Good luck Ivar
Jens Krause
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 4 giu 2012, 05:27 GMT-4
Hi,

interessting that it works for you. I tried someting very simple

Hx=x
Hy=y

and these initial conditions are considered correctly. But the
values from the other physics not.

What are your setting in "Values of variables that you do not solve for"?
Is there something users usually do wrong?


Jens
Hi, interessting that it works for you. I tried someting very simple Hx=x Hy=y and these initial conditions are considered correctly. But the values from the other physics not. What are your setting in "Values of variables that you do not solve for"? Is there something users usually do wrong? Jens

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