Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Phạm Ngọc Kiên
Dear Prof. Wolfgang Bangerth, I would like to thank you very much. I am reading through the FE_NedelecSZ codes in Deal.II library. I currently try to solve the electromagnetics problem, followed by the research of Ross Kynch. My goal is to compute the right hand side for the problem: curl curl E +

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Wolfgang Bangerth
On 2/18/19 10:47 PM, Phạm Ngọc Kiên wrote: > > 2. I have tried some thing with the third method. And here-below is my code: > > Point p{0.5,0.5,0.5};//position in reference cell > Quadrature q(p); > FEValues fe_values_q(fe, q,update_quadrature_points); > fe_values_q.reinit(cell); > > //position

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Phạm Ngọc Kiên
I am sorry as my question is not clear enough. What I mean is when we loop over all degree of freedom in a cell, for example, for (unsigned int i = 0; i < dofs_per_cell; ++i) { .. } Do we have any method to get the position of the i_th degree of freedom inside this loop? I would like to

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Phạm Ngọc Kiên
Hi, I have tried with the methods described in the FAQ and figured out that: 1. As I used edge-based element (FE_NedelecSZ), I queried if my element had support points by *fe.has_support_points()*. As a consequence, it returned zero. The *FiniteElement::get_unit_support_points() *also returned an

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Phạm Ngọc Kiên
Yes, thank you very much. I will dig deeper into this. Best, Pham Ngoc Kien Vào Th 3, 19 thg 2, 2019 vào lúc 11:23 Bruno Turcksin < bruno.turck...@gmail.com> đã viết: > Pham Ngoc Kien, > > Le lun. 18 févr. 2019 à 19:33, Phạm Ngọc Kiên > a écrit : > >> With the DoFTools::map_dofs_to_support_poi

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Bruno Turcksin
Pham Ngoc Kien, Le lun. 18 févr. 2019 à 19:33, Phạm Ngọc Kiên a écrit : > With the DoFTools::map_dofs_to_support_points() , I saw in the library > manual that it can be used with no edge elements or the like. > As my finite element is edge-based one, I wonder if this function work? > No the func

Re: [deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Phạm Ngọc Kiên
Hi Bruno, With the DoFTools::map_dofs_to_support_points() , I saw in the library manual that it can be used with no edge elements or the like. As my finite element is edge-based one, I wonder if this function work? I have check the function fe.has_support_points() in my code and it returned 0. That

[deal.II] Re: Sparsematrix initialization in P4est program

2019-02-18 Thread David Wells
Hi Gabriel, I believe that you are still using the standard deal.II SparseMatrix class; is this correct? If so, this class won't work with fully distributed calculations. You will need to use either the Trilinos or PETSc wrappers since not all information is available on the local processor wit

[deal.II] Re: How to get the coordinates of a given degree of freedom on an edge?

2019-02-18 Thread Bruno Turcksin
Hi, Take a look at this entry in the FAQ. Best, Bruno On Monday, February 18, 2019 at 12:20:48 AM UTC-5, Phạm Ngọc Kiên wrote: > > Dear colleagues, > I am working

[deal.II] Re: Calculate error values as percentages

2019-02-18 Thread Daniel Arndt
Maxi, Usually when calculating error values I use > VectorTools::integrate_difference(). Still, that gives me an absolute > value. Now I would like to calculate the relative value, i.e. > |u-u_h|_{H^1}/|u|_{H^1}. Based on what I could find until now that means > that I have to create a vector

[deal.II] Calculate error values as percentages

2019-02-18 Thread 'Maxi Miller' via deal.II User Group
Usually when calculating error values I use VectorTools::integrate_difference(). Still, that gives me an absolute value. Now I would like to calculate the relative value, i.e. |u-u_h|_{H^1}/|u|_{H^1}. Based on what I could find until now that means that I have to create a vector containing the