The research objectives of this study are to develop and validate a detailed simulation model of a test cell which was used to measure heat balances for comfort evaluation.
The Research Centre on Zero Emission Buildings (ZEB) Test Cell Laboratory is a facility designed to carry out experiments on building envelope systems and their interaction with HVAC terminal units. Tests on the interaction between the building envelope, the building equipment, and indoor environmental quality analyses, with or without the presence of users in the test cell room, can be performed. The ZEB Test Cell Laboratory is constituted of two identical and fully independent test cells. Each test cell has one surface of the cell envelope exposed outside (a façade facing south), while the other five surfaces are surrounded by an environmentally-controlled volume. In order to be able to run numerical tests (in advance or with a large number of parameters) a simulation model of one test cell was developed in the programme IDA Indoor Climate and Energy (IDA ICE) v. 4.8.
The calibration of the model after setting it up according to the laboratory’s technical data was carried out in two steps: In the first, no windows were opened so that all deviations could be attributed to the cell’s envelope itself. Before also investigating the opening of windows in the second step, the discharge coefficient cd which describes flow losses in natural ventilation had to be defined. By experiment, it was found to be cd = 0.75 for the tilted window being used in the laboratory which is in accordance with the research by Heiselberg et al. (Heiselberg et al., 2001).
The simulation results of step 1 showed high accordance with the measurements. The simulated zone air temperature and envelope surface temperatures of step one were almost constantly within the measurement inaccuracy, except for the ceiling temperature. Here the deviation was higher compared to the other results. The evaluation of step 2, the test row with an open window, revealed issues with solar gains, which were greatly overestimated in the simulations. The too high solar gains led to a lower agreement of simulated and measured data in the second test row compared to the first. This deviation made it impossible to investigate the applicability of the previously determined cd value for the opened window.