GiD forum

I am using GiD to view results of two-dimensional shallow water
equations modelling. I use a finite volume method so I have one result
at every mesh element (quadrilaters). Until now I used to interpolate to
the nodes and use GiD to view these results. With the new version I
tried to use the gauss points result option, considering one gauss point
at every element.

According to what's written in 'what's new in version 6.0':
'ContourFill/ContorLines and others will do a local interpolation from
gauss points with internal coordinates to the nodes to draw the
results.', but I cannot manage GiD to do so and my results do not
appear in a smooth way.

Is there any way to use the Gauss points results option and have the
results automatically smoothed by GiD, so I don't need to do the
interpolation myself?

Ernest Bladé
ernest.blade at upc.es

Ernest Bladé i Castellet wrote:

I am using GiD to view results of two-dimensional shallow water
equations modelling. I use a finite volume method so I have one result
at every mesh element (quadrilaters). Until now I used to interpolate to
the nodes and use GiD to view these results. With the new version I
tried to use the gauss points result option, considering one gauss point
at every element.

According to what's written in 'what's new in version 6.0':
'ContourFill/ContorLines and others will do a local interpolation from
gauss points with internal coordinates to the nodes to draw the
results.', but I cannot manage GiD to do so and my results do not
appear in a smooth way.

GiD does an 'extrapolation' of the results on the gauss points to the
nodes. Thus, if only one gauss point is present, the nodal values will be
the same.


Is there any way to use the Gauss points results option and have the
results automatically smoothed by GiD, so I don't need to do the
interpolation myself?


It would be a good idea to implement a better, perhaps global, smoothing
algorithm, but there are several ways to do this smoothing, almost
each research group/person prefers its own smoothing algorithm.
It is subject that needs a more accurate study.

Ernest Bladé
ernest.blade at upc.es

--
--------------------------------------------------------------------------------
Miguel A. de Riera Pasenau miguel at cimne.upc.es http://gid.cimne.upc.es