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.
Convergence problem in parametric solver
Posted 06.09.2010, 08:45 GMT-4 1 Reply
Please login with a confirmed email address before reporting spam
Hi!
I attempt to solve some choke in 3D where I put as parameters: frequency, current and permeabilities of copper in three directions.
COMSOL is finding convergence for 0Hz and 50Hz but for 150Hz, convergence locks on 0.01 and doesn't change anymore (for infinity iteration I suppose).
Furthermore I have to say that using only one group of parameters (e.g for 150Hz) everything works fine.
So my question is why COMSOL cannot find convergence in parametric solver for 150Hz after computing 0 and 50Hz. But find solution when I apply only 150Hz (with adequate current and permeabilites of course).
Bellow I apply screenshot from solver with convergence plot.
PS: I'm using Comsol 3.5a
I attempt to solve some choke in 3D where I put as parameters: frequency, current and permeabilities of copper in three directions.
COMSOL is finding convergence for 0Hz and 50Hz but for 150Hz, convergence locks on 0.01 and doesn't change anymore (for infinity iteration I suppose).
Furthermore I have to say that using only one group of parameters (e.g for 150Hz) everything works fine.
So my question is why COMSOL cannot find convergence in parametric solver for 150Hz after computing 0 and 50Hz. But find solution when I apply only 150Hz (with adequate current and permeabilites of course).
Bellow I apply screenshot from solver with convergence plot.
PS: I'm using Comsol 3.5a
Attachments:
1 Reply Last Post 11.09.2010, 08:12 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
I suppose there are many possibilities, normally Comsol proposes initial conditions ="0" whatever you have, this is not the best starting point for many cases, so COMSOL spends quite some time to get first some coherent initial conditions. During this process you might get stuck in a "valley" of your solution domain that is a blind end.
In such cases, running first a parametric study with increasing values to get a good initial value might be helpful: you store the first solution set and then use it as the initial conditions for further studies
--
Good luck
Ivar
I suppose there are many possibilities, normally Comsol proposes initial conditions ="0" whatever you have, this is not the best starting point for many cases, so COMSOL spends quite some time to get first some coherent initial conditions. During this process you might get stuck in a "valley" of your solution domain that is a blind end.
In such cases, running first a parametric study with increasing values to get a good initial value might be helpful: you store the first solution set and then use it as the initial conditions for further studies
--
Good luck
Ivar