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 The information provided may be out of date.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.

cannot use old values as the initial values for new step

Please login with a confirmed email address before reporting spam

Dear sir or madam,

I am using comsol to simulate a stationary model. I had convergence problems so that I was forced to vary the inlet BCs gradually to converge to the real case and I had to use the last value of previous step as the initial value in the new step. But it was always 0 when I showed the initial values !!! I don't know what happend and could anybody help out, please ?

Thank you very much.
Best wishes

2 Replies Last Post 27.06.2013, 16:39 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 30.03.2013, 06:50 GMT-4
Hi

to help the solver convergence it is often useful to do a parametric sweep, to reuse last calculation value you must ensure you are using a "continuation" sweep and not a full geometrical parameter sweep (the latter restarts at original initial conditions, mostly all "0". The continuation sweep at is at the bottom at the solver node.

If you have several solver steps, you need to ensure the solver nodes store the previous solution and that next solver node points to the stored solution for its initial conditions, and not the initial physics node initial condition, to get the solvers to chain one from the other (open up all the solver node tabs)

I agree that the solver log, in detailed mode could give more info, such as where the solver red th initial conditions, hope COMSOL developers are listening ;)

--
Good luck
Ivar
Hi to help the solver convergence it is often useful to do a parametric sweep, to reuse last calculation value you must ensure you are using a "continuation" sweep and not a full geometrical parameter sweep (the latter restarts at original initial conditions, mostly all "0". The continuation sweep at is at the bottom at the solver node. If you have several solver steps, you need to ensure the solver nodes store the previous solution and that next solver node points to the stored solution for its initial conditions, and not the initial physics node initial condition, to get the solvers to chain one from the other (open up all the solver node tabs) I agree that the solver log, in detailed mode could give more info, such as where the solver red th initial conditions, hope COMSOL developers are listening ;) -- Good luck Ivar

Josh Thomas Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 27.06.2013, 16:39 GMT-4
Ivar,

I just read this post from March. Good idea to have the log include which study and/or steps the initial values were pulled from for a particular solution.

I would definitely check this and consider it to be a nice upgrade to the functionality in newer versions.

BTW - Thanks COMSOL for including the RAM usage statistics in the log in v4.3b!

Best regards,
Josh
Ivar, I just read this post from March. Good idea to have the log include which study and/or steps the initial values were pulled from for a particular solution. I would definitely check this and consider it to be a nice upgrade to the functionality in newer versions. BTW - Thanks COMSOL for including the RAM usage statistics in the log in v4.3b! Best regards, Josh

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.