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 plot

Please login with a confirmed email address before reporting spam

When using a non-linear time dependent solver, convergence plot shows reciprocal of step size vs time step graph.How can I interpret that the solution is approaching convergence?


1 Reply Last Post 16 août 2023, 09:26 UTC−4
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 16 août 2023, 09:26 UTC−4

Like you said, the "convergence" plot is simply showing the reciprocal of the time step size. Depending on your solver settings and the nature of your problem, this may or may not change from one time step to another. In configurations where the time step size is allowed to vary (which is the usual default), if you observe a tendency toward endlessly decreasing step sizes vs. time, then this generally implies the existence of undesirable numerical issues. If you look at the solution log, you can learn more about the seriousness of the errors that are causing the time-stepper to attempt to use shorter time steps (to address its difficulties). Modifying your geometry, mesh, solver settings, boundary and/or initial conditions, discretization, etc., can all impact success (or failure) of time-domain models to complete the calculations. Always prepare and solve simple problems first -- then work your way toward the more complicated problems of interest to you.

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Like you said, the "convergence" plot is simply showing the reciprocal of the time step size. Depending on your solver settings and the nature of your problem, this may or may not change from one time step to another. In configurations where the time step size is allowed to vary (which is the usual default), if you observe a tendency toward endlessly decreasing step sizes vs. time, then this generally implies the existence of undesirable numerical issues. If you look at the solution log, you can learn more about the seriousness of the errors that are causing the time-stepper to attempt to use shorter time steps (to address its difficulties). Modifying your geometry, mesh, solver settings, boundary and/or initial conditions, discretization, etc., can all impact success (or failure) of time-domain models to complete the calculations. Always prepare and solve simple problems first -- then work your way toward the more complicated problems of interest to you.

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.