Page 1 of 1

convergence criterion "derive" ?

Posted: Tue Oct 26, 2010 11:30 am
by Chebli Rezki
Hello,
If the "Norm. residual" correspond to the relative precision for the solution of the linear system ("epsilo"). Then how to impose the global convergence criterion for the solution of the algorithm for each time step (derive) ?
 
Best regards.
Rezki

Re: convergence criterion "derive" ?

Posted: Tue Oct 26, 2010 12:05 pm
by Alexandre Douce
Hi,
I suppose that you want to reach a steady state for your simulation. Actually, the simplest way is to use monitoring points. We plan to add residuals for steady algorithm in a next version of Code_Saturne.
In order to have more information about the derive, please consult this thread:
https://code-saturne.info/products/code-saturne/forums/general-usage/778113951/754418244