Your assumption is correct, I checked the shape of the surface after initialization. It looks (for OFFSET=2 and refine_global(2)) like
<https://lh3.googleusercontent.com/-7NpXDgl1JwU/WcoOM9Guc_I/AAAAAAAACTw/O0Jv2XoGvVQCg6Kxr7e2FWXpR6peIJVbQCLcBGAs/s1600/visit0001.png> Setting the OFFSET-value to 10 while leaving refine_global at 2 results in <https://lh3.googleusercontent.com/-5qd18qRW9pw/WcoOPnnH2kI/AAAAAAAACT0/fX-9QbsEsysyH6Z9PdXwmdAIOggs9jOGgCLcBGAs/s1600/visit0000.png> Here the drop is clearly visible (which presumably results in the calculation problems), but can not be removed with higher values for refine_global. As example using refine_global(4) and OFFSET=10: <https://lh3.googleusercontent.com/-2YTcx4C86R4/WcoOTRRZawI/AAAAAAAACT4/sZEBzwVaqhYEl1TFvHf9sRv-eRyCigWMgCLcBGAs/s1600/visit0002.png> I initialize my solution using the code shown above, but still the sharp drop remains. Why? Thank you! Am Dienstag, 26. September 2017 03:15:47 UTC+2 schrieb Wolfgang Bangerth: > > On 09/25/2017 01:25 PM, 'Maxi Miller' via deal.II User Group wrote: > > > > I do not understand the reason why that happens. Is it a mathematical > > problem? Or rather a problem in my code? Afaik the gradients should not > > depend on the offset, thus I do not know where to look for the problem > here. > > Did you adjust the boundary values in the same way as the initial > values? Because if you don't, your initial guess will have a sharp > boundary layer where it drops from OFFSET to zero, and that leads to a > very large non-linearity. > > Best > W. > > -- > ------------------------------------------------------------------------ > Wolfgang Bangerth email: [email protected] > <javascript:> > 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.
