Bachelor's thesis Mike Dressler


Verification of a Low Order Model with Measurement Data of a Single Building

Simulation models are used to calculate the thermal load of single buildings or entire city districts in
order to assess their demand for energy and possible savings. Simplification of the model and along
with it inaccuracies have to be accepted for the benefit of saving time for simulations and increasing
the economic efficiency of the calculation. It is the object of this study to verify a reduced-ordermodel
(ROM) of the Institute for Energy Efficient Buildings and Indoor Climate’s AixLib by comparison
to measured thermal data from real buildings. These data have been derived from the Annex 58
project of the International Energy Acengy, from two identical buildings under study. Additionally,
the AixLib ROM is being modified, aiming to examine boundary conditions in more detail. To this
end ROM data for ceiling and floor are decoupled so they can be charged with individual boundary
conditions. Both models of simulation presented in this study can be validated by demonstrating
sufficient accuracy in examining boundary conditions.
For the purpose of comparing the ROM to the modified model both are implemented using different
variations, depicting one thermal zone of the entire building or four distinct sections of it. It
can be demonstrated that heating load and the effect of oscillating temperatures are calculated more
precisely in the modified model. Furthermore, higher resolution of climatic influences through
distinct thermal zones leads to lower simulation errors. However, disregarding the exchange of heat
between these thermal zones leads to errors in calculating temperatures.