Questions - Valparaíso - Chile

Christophe Geuzaine Christophe.Geuzaine at ulg.ac.be
Wed Aug 9 18:56:05 CEST 2000


Alejandro Angulo wrote:
> 
> Christophe:
> 
> Thanks for your help, now my fomulation is worked
> quite
> nicely. But now, too, I need to calculate some
> specific quantities like: stored energy, total losses,
> flux cross some section, resistence, inductance....
> How I make this (calculate integral quantities)?
> 

There is a beginning of support for computing such integral quantities
in post-processing, but it is really still "beta" -- or even "alpha".
(What I mean is that you can try it out, but it will change greatly in
future versions of the code).

For the moment, the trick is to define the terms as "Integral" in the
"PostProcessing" field, instead of "Local", and to use the type
"Cumulative" to sum up all contributions from each of the elements in
the domain of integration. Beware that the domain of integration is the
one specified in the "PostProcessing" (and not in the PostOperation,
which will only specify where to plot the integral quantity. OK, this is
dumb, since the quantity is a number and not a field, but, as I
said...). 

For example,

{ Name Dual; NameOfFormulation Test;
    Quantity {
      { Name n2; Type Cumulative;
        Value { Integral { [ nu[] * ({d a}+mu[]*{d phi})^2 ]; 
                           In Vol_Mag; Integration I1; Jacobian JVol; }
}
      } 
    }
  }

will compute the integral of (nu[] * ({d a}+mu[]*{d phi})^2) over region
Vol_Mag. There is no way to integrate quantities on something else than
a Region group for the moment.

Another solution is to plot *local* results in a file (e.g. in "Table"
format), and to make (or use) a small program to compute the integrals
you want...

-- 
Christophe Geuzaine

Tel: +32-(0)4-366.37.10    mailto:Christophe.Geuzaine at ulg.ac.be
Fax: +32-(0)4-366.29.10    http://www.geuz.org