Dear Hilde,
this error message can occur when one of the values specified within Type56 (*.bui) equals zero. This can be e.g. a area of a wall and is mostly caused during editing the *.bui - within Prebid the Input of "0" should be prevented by an error message in the user interface.
regards,
Matthias
Dipl.-Ing. Matthias Rudolph
TRANSSOLAR Energietechnik GmbH
Konzepte für energieeffizientes Bauen und Strategies for energy efficient design and thermal
Nutzerkomfort in Gebäuden comfort in buildings
phon ++49-711-67976-192 Curiestrasse 2
fax ++49-711-67976-11 70563 Stuttgart
Germany
Mailto:rudolph@xxxxxxxxxxxxxx
http://www.transsolar.com
-----Ursprüngliche Nachricht-----
Von: Hilde Breesch [mailto:Hilde.Breesch@xxxxxxxx]
Gesendet: Donnerstag, 18. März 2004 18:27
An: trnsys@xxxxxxxxxxxxxxxxxxx
Betreff: star matrix is singular
Dear Trnsys users,
I'm trying to perform a sensitivity analysis with Trnsys15-Comis3.1 on natural night cooling (single sided natural ventilation) in a small office. Ventilation air by day is mechanically blown into the office and extracted in the corridor. Using Matlab I can carry out many simulations at once. The input data change each time for one parameter. The value of the internal convective heat transfer coefficient depends on the surface (wall vs floor), the current time (working hour or not) and the presence of the night ventilation. Night ventilation through louvres in the façade is (manual) controlled by maximum indoor and outdoor temperature during the last day.
Only changing the height of the zone 'kantoor' (= office) a little bit from probe123.bui to probe124.bui, and therefore the areas of the internal walls, causes the next error (see probe124.lst):
***** ERROR ***** TRNSYS ERROR # 185
UNIT 10 TYPE 56 MULTIZONE BUILDING
STAR MATRIX IS SINGULAR - SIMULATION STOPPED.
I tried a lot of things to understand the reason of and find a solution for this problem (e.g. changing values of convective heat transfer coefficients and), but couldn't find it. Has any one had the same problem? Or does somebody know the reason? You can find all the files in attachment.
thanks in advance,
Hilde Breesch
PhD student Buildings & Climatic Control
Ghent University, Department of Architecture & Urban Planning Plateaustraat 22, B-9000 GENT tel. +32 9 264 78 61 fax. +32 9 264 41 85
e-mail: Hilde.Breesch@xxxxxxxx http://allserv.ugent.be/~hbreesch/research < Datei: Helpdesk.zip >>