[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Getfem-users] GetFEM program

From: Yves Renard
Subject: Re: [Getfem-users] GetFEM program
Date: Fri, 7 Oct 2011 11:10:29 +0200
User-agent: KMail/1.13.5 (Linux/2.6.32-34-server; KDE/4.4.5; x86_64; ; )

Dear Danilo,

The definition of the region 10 is in fact inapropriate. Your definition


means that you include faces of elements (edges in 2D), not elements 
themselves. The result is that the elliptic term is indeed integrated on 
edges, not on elements. And of course it makes a difference. You can define 
region as follows

domain = m.cvid()
m.set_region(10, [domain, domain*0-1])

Indeed, a -1 is needed (0 for the matlab interface) to specify that the 
element itself is added. This is not very satisfactory. I think, I will modify 
the set_region method in order to be able to write m.set_region(10, m.cvid()).

Thank you for your report,


On vendredi 23 septembre 2011, Danilo Sergi wrote:
> Dear Prof. Renard,
> I am grateful for your consideration of my problem. I have attached a
> program for the heat equation as implemented in Matlab. By considering
> only conduction (Laplace term), it is already seen a difference between
> the case where the whole mesh is specified explicitly (lines 48  and
> 108) and the default setting.
> I hope that the remark makes sense and can be useful.
> Sincerely yours,
> Danilo Sergi


  Yves Renard (address@hidden)       tel : (33)
  Pole de Mathematiques, INSA-Lyon             fax : (33)
  20, rue Albert Einstein
  69621 Villeurbanne Cedex, FRANCE


reply via email to

[Prev in Thread] Current Thread [Next in Thread]