Hi Markus
Thanks. I'm not sure if that note is in the tut. Perhaps it could just be added
as a note under possible extensions.
Best
Andrew
On 15.04.2011, at 08:27, Markus Bürg wrote:
> Hello Andrew,
>
> it depends on the geometry you have. If you have a simple geometry, then all
> faces are oriented in the same way. Thus the dofs are ordered the same way,
> too. If you have a complex geometry and the face orientation changes, then
> you have to check how the orientation of the two faces is and take into
> account this rotation. However, in both cases the dofs are ordered with
> respect to the local coordinate system of the face.
>
> Best Regards,
> Markus
>
>
>
> Am 15.04.11 07:50, schrieb Andrew McBride:
>> Hi
>>
>> I have one question about how the periodic boundary conditions are
>> implemented in the tut. The faces on opposite sides are matched and then
>> constraints imposed between the degrees of freedom (see below). Can one
>> assume the the ordering of dof on the opposite faces (f1 and f2) match?
>>
>> Andrew
>>
>>
>>>>> {
>>>>> const unsigned int nDofsPerFace =
>>>>> dofHandler.get_fe().dofs_per_face;
>>>>>
>>>>> std::vector<unsigned int> localDofs1(nDofsPerFace);
>>>>> f1->get_dof_indices(localDofs1);
>>>>>
>>>>> std::vector<unsigned int> localDofs2(nDofsPerFace);
>>>>> f2->get_dof_indices(localDofs2);
>>>>>
>>>>> for(unsigned int i = 0; i< nDofsPerFace; ++i)
>>>>> {
>>>>> if(!constraints.is_constrained(localDofs1[i])&&
>>>>> !constraints.is_constrained(localDofs2[i]))
>>>>> {
>>>>> constraints.add_line(localDofs1[i]);
>>>>> constraints.add_entry(localDofs1[i],
>>>>> localDofs2[i], 1.0);
>>>>> }
>>>>> }
>>>>>
>>>>>
>>>
>> _______________________________________________
>> dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
>>
>>
> _______________________________________________
> dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
_______________________________________________
dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
