[TRNSYS-users] Problems with detailed longwave radiation and view factor matrix...

[Arnaud Candaele]

Hello everbody,

I am currently trying to use the "detailed model" of longwave radiation
exchange within a zone, but when I run TRNSYS, I get this error message :
"Internal error of detailed longwave radiationexchange of zone 1. Energy
balance error of xxx% at time 0.05". The thing is that I created my view
factor matrix with Fluent since my room is concave. This matrix is well
defined, but Fluent gave me this matrix in a different order than TRNSYS
regarding the surfaces ID's. So for example, if the element (0,0) of this
matrix is referring to the view factor of "surface 1 to surface 1", it is
read by TRNSYS as the view factor of "surface 12 to surface 12", since in
this example surface 12 is the first surface listed in the bui file for
this zone. As a first trial of the detailed longwave radiation mode, I
didn't correct the matrix.

My first question is : could that error message be linked to the fact that
the matrix is incorrect ? I mean, of course the results of my calculation
would be completely wrong since some walls that weren't supposed to receive
radiation would get it, etc.. but shouldn't it still compute, even with
wrong results at the end ? The sum of all viewfactors from a surface to all
other is well 1 (or just below), so I guess it shouldn't lead to energy
balance error ...?

Anyway, adressing the correction of the viewfactors, I have 2 options :

   1. Rewriting my viewfactor matrix so that the surfaces ID's match
   between the vfm file and the bui file.
   2. Trying to modify the bui file, so that the surfaces order of that
   zone match the order in the vfm file.

I tried the second option for now since my matrix is quite large and
changing it manually could take some time... Of course I could use a
spreadsheet macro but I have to write/find it, and I don't have it right
now..

So I changed 2 things in the .bui file : 1) the ZONE1 (which is the zone in
which I want to use detailed mode) faces listing (in "BUILDING"), so that
the faces are listed in the same order as in my vfm file, and 2) the faces
listing in ZONE1, in "_EXTENSION_BuildingGeometry_" at the end of the bui
file, even though I don't think this is important..

When I do these 2 changes in my bui file, and with the same faces' ID's
listing in my vfm file, I still get the error message that the surfaces
ID's in the vfm file mismatch the surfaces ID's in the bui file.

My question is : is there some other part of the bui file that I need to
change in the bui file ?

Or finally, could the longwave radiation error message be caused by
something else ? I've seen another post on that subject last year, and
there was a suggestion that it might be caused by the initial temperature
of the walls that would be too close to each other. I tried setting
different initial temperatures in the room, but I still get the same error
(except that the percentage of Energy balance error changes...).

If anyone could help, it would be appreciated !

Thank you,

Arnaud.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/trnsys-users-onebuilding.org/attachments/20130329/70025806/attachment-0001.htm>