Re: [deal.II] Re: QProjector::project_to_face and Gauss Lobatto

2017-04-19 Thread Vishal Ganpat
GLL points include the endpoints of the domain...they are useful when you
are trying to solve large-scale problems and have integration points same
as the interpolation points...e.g. Lagrange interpolation with gll points
etc, a very standard technique used in spectral element methods. I'm
curious abt my original question on how it is done in deal ii.

On Apr 19, 2017 4:43 AM, "Daniel Arndt" 
wrote:

vganpat,



I am trying to solve a problem that has both the volume integral and a
> surface integral.
>
You would normally just create a quadrature rule for the volume and the
surface integral separately.


> If I am using a higher order polynomial approximation and use Gauss
> Lobatto quadrature,
> the project_to_face() function seems to generate new points on the faces.
> But, with GLL points,
> there are already some points on the face that also contribute to the
> volume integral.
> What is not clear to me is, through this projection operation, I will lose
> the unique global number
> for these coincident points (those from Quadrature and then the
> SubQuadrature).
> If I were to record responses at the Quadrature points, with this
> duplication of nodes,
> how can I uniquely capture the response. Any help is appreciated.
>
What exactly are you trying to achieve? In general, the quadrature points
do not coincide with the support points of the used finite element
on a cell. Why are you using Gauss-Lobatto quadrature? Can you explain a
bit more what you are doing?

Best,
Daniel

-- 
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 a topic in the
Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/
topic/dealii/ZLJQo5m85WM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to
dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

-- 
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 dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


[deal.II] Re: QProjector::project_to_face and Gauss Lobatto

2017-04-19 Thread Daniel Arndt
vganpat,


I am trying to solve a problem that has both the volume integral and a 
> surface integral.
>
You would normally just create a quadrature rule for the volume and the 
surface integral separately.
 

> If I am using a higher order polynomial approximation and use Gauss 
> Lobatto quadrature, 
> the project_to_face() function seems to generate new points on the faces. 
> But, with GLL points, 
> there are already some points on the face that also contribute to the 
> volume integral. 
> What is not clear to me is, through this projection operation, I will lose 
> the unique global number 
> for these coincident points (those from Quadrature and then the 
> SubQuadrature). 
> If I were to record responses at the Quadrature points, with this 
> duplication of nodes, 
> how can I uniquely capture the response. Any help is appreciated. 
>
What exactly are you trying to achieve? In general, the quadrature points 
do not coincide with the support points of the used finite element
on a cell. Why are you using Gauss-Lobatto quadrature? Can you explain a 
bit more what you are doing?

Best,
Daniel

-- 
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 dealii+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.