On 3/12/19 5:34 AM, Apurva Tiwari wrote: > I'm trying to get appropriate convergence rates for a 2D scalar advection > problem. The strategy is quite standard. I start with a coarse grid, > successively refine it globally and generate the convergence table as done in > step-7. As can be seen in the attached convergence tables, the values are far > off. What all could have gone wrong and where do I start debugging my code > from?
Apurva, So many things could be wrong :-) You have an error that has a component from the spatial discretization and one from the temporal discretization. Try to separate these by making either h or delta t small and observe that the convergence rate for the other is correct. You may also still be in the pre-asymptotic range, seeing that your final mesh is only 128x128 -- if the diameter of the Gaussian you have is smaller than, say 1/10th of the domain's diameter, then the results on all of the earlier meshes are unreliable because you are not resolving the solution on them. So make the problem simpler and see which part it is that is wrong (if anything is wrong at all). Best W. -- ------------------------------------------------------------------------ Wolfgang Bangerth email: [email protected] www: http://www.math.colostate.edu/~bangerth/ -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
