Why is the excess temperature in 2003 smaller than in 2002

With the previous investigations, this question can be answered. Since the room temperature responds to the heat / loss-coefficient and the time constant of the building, this discrepancy at first view is discussed on the basis of the energy balance model according to Eqs. (1) and (2). An investigation concerning the user behaviour concluded that in 2003 the windows were open more often: Fig. 4 shows exemplarily the ambient air temperature in 2002 and 2003 at the Fraunhofer ISE building. While in summer 2002 the daily mean outdoor air temperature sometimes falls below 20 °C, this occurs very seldom in 2003. Taking a typical user behaviour into account, the windows were closed more often in 2002 than in 2003 due to these low ambient air temperatures. This user behaviour have been shown in each of these buildings, cf. [12], [13] and [14].

As the windows are opened more often

in 2003, the heat loss factor H is higher in 2003. The mean indoor temperature Ti>m is calculated according to Eq. (2) from the mean ambient air temperature Ta, m and the gain-to — loss ratio y=Gm/H. In 2003, the heat gains Gm were (almost) identical to the heat gains in 2002 but more windows were open, which suggests that the heat loss factor H was larger in 2003. Thus, the excess temperature у is smaller in 2003 than in 2002.

Fig. 4: Outdoor and mean indoor air temperature during the summer period (June 1 — August 31) in the Fraunhofer ISE building in the summers of 2002 and 2003.

Why do the room temperatures exceed the comfort criteria more often 2003?

1.6 8.6 15.6 22.6 29.6 6.7 13.7 20.7 27.7 3.8 10.8 17.8 24.8 31.8

10 I…… I…… I…… I…… I…. I….. I…….. I…… I…. I….. I…….. I…… I…… I

1.6 8.6 15.6 22.6 29.6 6.7 13.7 20.7 27.7 3.8 10.8 17.8 24.8 31.8

The temperature amplitude ДТ is calculated according to Eq. (2) from the temperature amplitude ДТ, the quotient of the heat gain amplitude and the heat loss factor AG/H and the time constant t=C/H. The heat gain amplitude AG was (almost) identical in 2002 and 2003. The daily temperature amplitudes are similar in both years, too: ATa,2002=3.8K, ATa,2003=3.9K, AT,2002=1.3K and ATi,2003=1.2K. The conversion of Eq. (2) to C shows that C is proportional to H and, hence, increases with H, if all other input parameters are held constant. With the monitored data from 2002 and 2003, the conclusion can be drawn that the daily heat storage capacity is (almost) identical in both years.

Thus, the more frequent occurrence of high room air temperatures corresponds with the long-term behaviour of the building: Due to the continuous thermal exposure in 2003, the long-term heat storage of the building is heated and cannot compensate for temperature changes which continue for several days.

This hypothesis is verified with Fig. 5 in connection with the simulation study according to Fig. 2. The building structure could not thermally regenerate due to long warm periods. This is the reason, why the slope in the regression line in Fig. 3 is steeper and, hence, the room temperature exceeded more often the comfort criteria, cf. Fig. 1, in 2003 than in 2002.

Conclusions

An analysis of monitored data from summer 2002 (typical summer weather) and 2003 (summer weather with long and extremely warm periods) reveals that office buildings in central European climate do not need to be air-conditioned, if they are accurately designed and rationally operated. However, none of the buildings utilised the passive cooling potential completely.

Starting from the statements si — 4, the influence of the chronology of climate situations has been discussed, and the increased failure to meet the comfort standard in 2003 can be actually explained by using smaller time constants than in 2002. The precise calculation of the heat storage capacity is essential for the accurate calculation of the thermal building performance in summer and, especially, for the design of passive cooling concepts, which make use of the heat modulation due to the building’s thermal inertia.

As these conclusions have been drawn from a simplified cross-section analysis, a parametric model, which focuses on the essential building parameters, can be used successfully for data analysis and enhances the reliability concerning the design and
operation of passive cooling systems: The very complex interactions, which influence the thermal building behaviour, can be accurately modelled with a few concise parameters.

Acknowledgement

The research has been funded by the German Ministry of Economics and Labour within the framework of the German research programme SolarBau:Monitor under the reference O335007C.

The author wishes to thank Katrin Schlegel (Zentrum fur Umweltbewusstes Bauen, Kassel) and Peter Seeberger (University of Applied Science, Department of Building Physics) for the provision of data from the long-term monitoring campaign and the good co-operation in the projects and during the short-term measurements.