[Getdp] material property defined in each element

Jens Trommler trommler at temf.tu-darmstadt.de
Tue Apr 5 18:51:53 CEST 2011


Really Great! It worked from first try. Thanks a lot for this really
quick answer!

Just a short question for better understanding:
Why do I have to introduce a SubSpace in the FunctionSpace. What is
this meant for?


2011/4/5 Ruth V. Sabariego <r.sabariego at ulg.ac.be>:
> You could fill a function space with the values of sigma by means of a
> constraint.
> You start reading the values from a text file "mysigma.txt" with the
> following structure:
> TotalNumberOfNodes
> numberOfNode SigmaValue
> .                         .
> .                         .
> numberOfNode SigmaValue
>
>
> // get (nodal values of) sigma from .txt file:
> listsigma = ListFromFile["mysigma.txt"];
> mysigma[] = ValueFromIndex[]{listsigma{}};
>
> ============================================
> Constraint {
>  // completely fix sigma function space
>  { Name MySigma ; Type Assign ;
>    Case {
>      { Region MyDomain  ; Value mysigma[] ; }
>    }
>  }
>
> ============================================
> FunctionSpace {
>  { Name BF_sigma ; Type Form0 ;
>    BasisFunction {
>      // interpolate each component of sigma using standard nodal
>      // shape functions; you could also define other fuction space
>      { Name sn ; NameOfCoef s ; Function BF_Node ;
>        Support Region[{MyDomain}] ; Entity NodesOf[ All] ; }
>    }
>    SubSpace {
>      // make subspaces available in formulation
>      { Name sub ; NameOfBasisFunction sn; }
>    }
>    Constraint {
>      { NameOfCoef s ;
>        EntityType NodesOf ; NameOfConstraint MySigma; }
>    }
>  }
> }
> ============================================
> And then you can use it in a formulation as:
>
> Formulation {
>  { Name Electrostatics_vf ; Type FemEquation ;
>    Quantity {
>      { Name v ; Type Local  ; NameOfSpace Hgrad_vf_Ele ; }
>      // get each sigma tensor component
>      { Name sig ; Type Local  ; NameOfSpace BF_sigma[sub] ; }
>
>    }
>    Equation {
>
>      Galerkin { [ {sig} * Dof{Grad v} , {Grad v} ] ;
>                 In MyDomain ; Jacobian Vol ; Integration GradGrad ; }
>
>      // hack: null terms necessary to force initialization of fully
>      // fixed function space
>      Galerkin { [ Dof{sig} * 0 , {v} ] ;
>                 In MyDomain ; Jacobian Vol ; Integration GradGrad ; }
>    }
>  }
> }
>
> Regards,
> Ruth
>
>
>
> On 05/04/11 14:44, Jens Trommler wrote:
>>
>> Hi,
>>
>> I have the conductivity of each element given from another simulation
>> and now want to use this conductivity in GetDP for some
>> quasistationary simulations.
>> Is there an easy way to set the conductivity element-wise?
>> I guess that I need to write an .res file but don't really know how
>> exactly it should look like and how I can use it in a GetDP script.
>> Looked already in the examples but couldn't find any similar problem.
>>
>> Thank you for any help.
>> Regards,
>> Jens.
>>
>> _______________________________________________
>> getdp mailing list
>> getdp at geuz.org
>> http://www.geuz.org/mailman/listinfo/getdp
>>
>
> --
> Dr. Ir. Ruth V. Sabariego
> University of Liege, Dept. of Electrical Engineering&  Computer Science,
> Applied&  Computational Electromagnetics (ACE),
> phone: +32-4-3663737 - fax: +32-4-3662910 - http://ace.montefiore.ulg.ac.be/
>
>
> _______________________________________________
> getdp mailing list
> getdp at geuz.org
> http://www.geuz.org/mailman/listinfo/getdp
>
>