[Getdp] Problems with mixed periodic and Dirichlet boundary conditions

[David Colignon]

Hi Julien,

Just change the EntityType in the constraint (your basis functions are associated with the nodes):

       { NameOfCoef ui; EntityType NodesOf ; NameOfConstraint CL; }

Regards,

Dave

-- 
David Colignon, Ph.D.
1er Logisticien de Recherche
Université de Liège
ACE - Applied & Computational Electromagnetics
Quartier POLYTECH 1 - Montefiore B28
Allée de la découverte 10
4000 Liège - BELGIQUE
Tél: +32 (0)4 366 37 32
http://www.ulg.ac.be/nic4


On 26/11/15 16:47, de Rosny Julien wrote:
> Hello,
>
> I would like to simulate the wave propagation inside a waveguide with a periodic structure inside it.
> I simplified the problem with the simulation of the homogeneous Helmholtz equ. with Dirichlet boundary conditions on the top &
> bottom and periodic on the left and right.
> To that end, I set the following constrains :
>
> Constraint{
>      { Name CL;
>          Case{
>      // periodic BND cond. between TOP & Bottom
>      {Region L1 ;
>          Type LinkCplx ; RegionRef L3;
>          Coefficient  Complex[1,0];
>          Function Vector[$X,$Y+Sqrt[2],$Z] ;
>          }
>
> //     periodic BND cond. between left & right
> //      {Region L4 ;
> //        Type LinkCplx ; RegionRef L2;
> //        Coefficient  Complex[1,0];
> //        Function Vector[$X+1,$Y,$Z] ;
> //        }
>
>           //{ Region L1; Type Assign; Value 000. ;}
>             // { Region L3; Value 000. ;}
>           // Dirichelt BND condition on the bottom and top
>             { Region L2; Type Assign;  Value 0000. ;}
>             { Region L4; Type Assign; Value 0000. ;}
>        }
>      }
> }
>
> But it gives wrong results (not periodic on top/bottom and not null on left/right borders). See attached files.
> The code works well if now I only use  Dirichlet or periodic boundary constrains for the 4 sides.
>
> Any idea what is wrong?
>
> Thanks.
> Regards.
>
> Julien de Rosny
>
>
>
> _______________________________________________
> getdp mailing list
> getdp at geuz.org
> http://www.geuz.org/mailman/listinfo/getdp
>