Maarten G. Sourbron 1, Lieve M. Helsen 1. Corresponding SUMMARY

Transcription

1 Evaluation of adaptive thermal comfort models in moderate climates and their application to office buildings equipped with Thermally Active Building Systems (TABS) Maarten G. Sourbron 1, Lieve M. Helsen 1 1 Katholieke Universiteit Leuven (K.U.Leuven), Belgium Corresponding Maarten.Sourbron@mech.kuleuven.be SUMMARY For a moderate climate, this paper compares a conventional thermal comfort model (ISO7730) with adaptive models (ISSO74 and EN15251), investigates the energy use with heating/cooling set points set accordingly, and evaluates the calculated number of excess hours in a TABS building. ISSO74 allows higher summer operative zone temperatures (T op,max ) for outdoor temperatures higher than 18.4 C only. Because this rarely occurs in moderate climates, its benefit is lost for the most part. The EN15251 summer T op,max is very high compared to the other models. An office building simulation with ideal heating and cooling shows a higher cooling need for ISSO74 (+30%), and a lower for EN15251 (-20%) compared to ISO7730. The year round cooling need and the lower winter T op,max of ISSO74, cause these unexpected results. For the same building with TABS, only EN15251 results in less excess hours (minus 50%-90%). Using meteo data of warmer years, this conclusion holds. INTRODUCTION Thermal comfort being that condition of mind which expresses satisfaction with the thermal environment [1] is a definition much encountered in the literature. But what is often omitted is the affix and is assessed by subjective evaluation, meaning that translating the conditions for achieving thermal comfort into universally applicable equations a task which engineers very much like to perform is a difficult, if not impossible task. For the design of TABS buildings, of which it is known that indoor temperatures float during the course of the day [2[3], it is important to apply suitable thermal comfort criteria, in order to assess the performance of the building correctly. When a choice has to be made between the conventional static comfort models, with their fixed temperature limits, and the adaptive models, with their temperature limits floating with the outdoor temperature, it seems obvious to choose the latter for TABS buildings. But which indoor temperatures allow the adaptive models for the moderate climate of Western and Central Europe? And which impact has the choice of comfort model on the energy use? May adaptive models be applied to TABS buildings? And how is thermal comfort in TABS buildings evaluated when using the static or adaptive thermal comfort models as guideline? This article intends to answer these questions. Conventional static thermal comfort theory The basic theory of thermal comfort, developed by Fanger and translated in the international standard ISO7730 [4], is based on the prediction of the number of dissatisfied people (PPD or Predicted Percentage of Dissatisfied). 1

2 Usually the equivalent operative temperature is used as defining parameter ([4], Annex A). Nicol and Humphreys [5] defined this operative temperature as a simple but well performing index, while more complex indices show a lower correlation with the comfort votes of the respondents. Depending on the activity and the clothing level of the occupants, an optimal operative temperature exists. Adaptive thermal comfort models The static thermal comfort model described in ISO7730 is often criticized as to recognize too little the outdoor climatic context and a person s ability to fit the indoor climate to its personal requirements [6]. Therefore the static approach would cause an increased reliance on mechanical cooling. Adaptive thermal comfort models assume that people adapt their thermal requirements to e.g. outdoor climatic conditions. Several attempts have been made to incorporate this adaptation into thermal comfort standards. They all relate the indoor operative temperature to a reference outdoor temperature. Operative zone temp. ( C) %_ISO7730 T_room_min 90%_ISSO74_adaptive T_room_max 90%_EN15251_adaptive T_room_min 18.4 C : Intersection point ISSO74 with ISO %_ISO7730 T_room_max 90%_ISSO74_adaptive T_room_min 90%_EN15251_adaptive T_room_max Reference outdoor temperature (different definitions) ( C) Figure 1: Allowed indoor operative temperature as a function of the reference outdoor temperature for different thermal comfort models The Fanger approach implicitly accounts for adaptive behaviour by changing clothing from summer (0.5 clo) to winter period (1 clo). de Dear and Brager s proposal for adaptive comfort criteria [7] is incorporated into the most recent version of the ASHRAE standard 55 on thermal comfort [1]. However, since, according to de Dear and Brager's own discussion [7], this comfort model should not be applied when T ref,out falls below 10 C, it is not evaluated in this paper. In the Dutch guideline ISSO74, described by van der Linden et al. [8], two building types ALPHA and BETA are defined. An ALPHA building (adaptive) has an operable façade with at least one operable window or at least one temperature adjustment tool per two occupants and possibilities to adjust clothing to outdoor and indoor conditions. All other buildings are BETA buildings (non-adaptive). Compared to the de Dear and Brager criteria, they additionally define clear criteria for winter conditions. The European standard EN15251 is based on the conventional comfort model (ISO7730), and is extended with an adaptive model comparable to the de Dear and Brager s model. However, the curve of the recommended indoor operative temperature as a function of the reference outdoor temperature is shifted 1 C higher, resulting in a higher T op,max for summer outdoor temperatures (Figure 1). The question raises whether this is still comfortable. Nevertheless this thermal comfort model is based on extensive measurement data analysis, as described by Nicol and Humphreys [5]. 2

3 In order to compare different models in this paper, the 10% dissatisfied people criterion is used. Figure 1 summarizes the different thermal comfort models as a function of the specific reference outdoor temperature. It should be noted that different methods use different definitions for the reference outdoor temperature. Thermal comfort models in TABS buildings According to the different standards and guidelines concerning adaptive thermal comfort, the application of adaptive models should be restricted to: 1. buildings without mechanical cooling systems (air conditioning), 2. buildings used mainly for sedentary activity, 3. buildings where occupants have easy access to operable windows or some other occupants intervening possibility on ventilation, 4. buildings where occupants must be allowed to adapt their clothing or work schedule to the indoor or outdoor thermal conditions. Neither of these restrictions is necessarily satisfied by TABS buildings. While in heating regime heat loss and heat input are almost equal, the cooling case is more problematic: the available cooling power from the TABS can be 2 to 3 times lower than the occurring heat gains. Consequently, when TABS buildings do not have an additional HVAC system to substantially cover the peak load, a TABS system cannot be regarded as a full air conditioning system, capable of keeping the temperature at a specified set point. Often they are defined as passively cooled buildings, where the heat load during the day is stored and removed by means of ground cooling, night ventilation or a cooling tower [9]. Regarding the third requirement, Pfafferott et al. [9] describe in their large survey of lowenergy office buildings that the occupants perceive their impact to control or change the working environment by the manual operation of windows in summer as limited : Since the temperature difference in summer between inside and outside is smaller, the manual control of windows changes the room temperature not as significant as in winter and, hence, the occupant cannot perceive its interaction directly. This is also confirmed by Wagner et al. [10]. They therefore conclude that the adaptive comfort criteria are applicable to both naturally ventilated as well as to, what they call, mixed-mode buildings. Buildings with TABS fall under this last category in their view. On the other hand, regarding the fourth requirement, clothing and work schedule adaptation is perhaps not possible in every cultural and economical context, as Barlow et al. [11] suggest. For TABS buildings, with their changing room temperatures during the day, this all leads to the conclusion that adaptive thermal comfort models are applicable if occupants are allowed to change their clothing and/or activity level freely and if they are well aware of the characteristic features of their HVAC system with respect to thermal comfort. METHODS Thermal comfort limits for a moderate climate With the limits for thermal comfort defined by the different comfort models (see Figure 1), it is possible to produce an annual thermal comfort limits profile based on a typical meteorological year (TMY), in this case for Maastricht (the Netherlands). This site is chosen 3

4 because of the availability of real weather data from the Royal Dutch Meteorological Institute [12]. The number of days that the reference outdoor temperature is in a certain temperature range can be calculated for the intersection temperature between the adaptive models and the ISO7730 model. Figure 1 shows that this intersection point appears at T ref,out = 18.4 C for the ISSO74 model. If a certain climate shows little or no reference outdoor temperatures above this intersection point, using the corresponding adaptive comfort model will hardly allow higher operative temperatures in summer time than the ISO7730 model would. On the contrary, with the lower winter set points for cooling for the ISSO74 model, an office building with a cooling need in winter and mid season will be cooled to lower operative temperatures than with the ISO7730 model. Building zone energy use as a function of the thermal comfort model applied To evaluate the effect of using different thermal comfort models to define the heating and cooling set points on the energy use of an office building, a TRNSYS [13] simulation is run on a typical office building zone. The 2-zone building section is a cut-out of a typical office building with offices at the south and the north separated by a corridor. The outside wall has a 10 cm PUR insulation. This is a high level of insulation compared to standard building practice [14]. Since this building model will further in this paper be used to evaluate TABS, which are limited to buildings with a high quality envelope [15], the building quality is chosen to be high in this case. The ceiling is covered with gypsum. The raised floor is 20 cm high and tiles are made of gypsum related material. The office space is separated from the corridor by a light gypsum wall. The outside window has a thermally insulated frame and is placed without thermal bridges. The corridor is provided with a suspended ceiling used for hot and cold water supply, air ducts, electricity and ICT. Building zones parameters and simulation assumptions are presented in Table 1. Table 1 : Parameters of the 2-zone building section and simulation assumptions Building and simulation parameters Heated volume (m³) Heated area (m²) 37.8 Transmission area (m²) 10.8 U-value external wall (W/m²K) 0.29 U-value total (W/m²K) 0.96 Glazing : U-value (W/m²K); g-value (%) 1.6; 40 Percentage of glazing (%) 50 Office occupation 8AM 12AM; 1PM 6PM Lunch break Appliances and lights remain on People sensible heat output (no latent heat) 90 W/person (40% convective, 60% radiative) Appliances (PC s, printers, etc ) 150 W/person Lights 10 W/m² (20% convective, 80% radiative) Residual electric consumption during night 1 W/m² (100% convective) Hygienic ventilation 36 m³/h.person at 18 C supply temperature In this simulation, the office zones are heated and cooled by means of ideal heat and cold emitters with sufficient power to keep the room continuously at the required temperature. No night setback is used. The set points for heating and cooling are defined by the respectively lower and upper temperature limits as imposed by the different thermal comfort models. For this simulation it is explicitly chosen not to use real heating and cooling equipment in the 4

5 building. The energy consumption would largely depend on the applied control strategy. This would hamper the allocation of the differences in energy use to the thermal comfort models applied. Evaluation of the number of exceeding hours for the different thermal comfort models in TABS buildings The third part investigates how the different thermal comfort models evaluate the resulting thermal comfort prevailing in the office zones. The office building zones defined in the previous section are used, but with a TABS heating and cooling system and with a fixed control strategy: the TABS are controlled with a constant water flow at a temperature determined by the cooling curve as given by Tödtli et al. [16]. This control strategy is far from ideal. But, since the focus of these simulation is not the search for the optimal control strategy, but the comparison of comfort evaluation by the different comfort models, this simple control strategy can be used. RESULTS Thermal comfort limits for a moderate climate The results of the thermal comfort models analysis method, described in the METHODS section, to the Maastricht TMY, are presented in Figure 2. It is clear that the definitions of the reference outdoor temperature for both thermal comfort models does not differ much. 120% T_ref ISSO74 T_ref EN15251 Days in temperature range (% of a year) 100% 80% 60% 40% 20% ISSO74 intersection point 0% Reference outdoor temperature ( C) Figure 2 : Number of days the reference outdoor temperature (different definitions for different models) is lower than the abscissa temperature for the Maastricht TMY Only 6% of a year (21 days) has a higher reference outdoor temperature than the 18.4 C intersection point of the ISSO74 model. The adaptive EN15251 model has no intersection point with ISO7730, but is characterized by a higher operative temperature for reference outdoor temperatures higher than 10 C. 167 days (46%) are above this limit of 10 C. This analysis shows that the benefit of the ISSO74 adaptive model allowing higher summer indoor temperatures is very limited if applied to a moderate climate. This was already indicated by van Hoof et al. [17]. EN15251, with a maximum operative temperature always being higher than ISO7730 will obviously show benefit with respect to the required cooling energy for the building. 5

6 Building zone energy use as a function of the thermal comfort model applied Table 2 shows that regarding cooling the EN15251 model results in a substantially lower energy consumption, caused by the much higher maximum operative temperatures compared to the other comfort models (see Figure 1). The ISSO74 model results in a higher energy consumption. Not only are there few warm days in the Maastricht TMY (as shown by Figure 2), but more important, the higher energy consumption is caused by the lower operative temperatures during winter and mid-season period. Because this office building requires cooling practically the whole year round, these winter and mid-season limits effect the final result largely. This was not expected for the ISSO74 model, because it is specially designed for use in a moderate thermal climate. A warm year like 2003 (high mean temperature and high amount of solar radiation) can largely affect the heating and cooling set points determined by the adaptive thermal comfort model applied. This, on its turn, will influence the required heating and cooling energy to condition the building. When the same procedure is applied to the 2-zone office building for the year 2003, this leads to similar conclusions as for the TMY of Maastricht, although the differences between ISSO74 and ISO7730 are smaller. Table 2 : Cooling energy consumption for the south and north zone for the Maastricht TMY (with Q cool in kwh/m² and relative to ISO7730) South zone North zone Total Q cool Q cool Q cool ISO (100%) 26 (100%) 42 (100%) ISSO74 69 (119%) 34 (132%) 51 (123%) EN15251 adaptive 52 (91%) 21 (80%) 36 (88%) Applying other lower internal and external loads to the zone, obviously has an effect on the energy use. The change is more or less equal for all seasons though, so this will not influence the conclusions with regard to the adaptive thermal comfort models. Evaluation of the number of exceeding hours for the different thermal comfort models in TABS buildings The number of hours trespassing the lower and upper comfort limits is calculated in Kelvinhours (Kh), with only the hours during occupation of the office zones counted in. Although expected from the basic principle of the adaptive methods, ISSO74 does not evaluate the thermal comfort as better, compared to the non-adaptive method ISO7730. Again, this is caused by the low maximum winter and mid season limits and the heat gains remaining important during this period. The results shown in Table 3 lead to similar conclusions as in the energy consumption analysis. Table 3: Exceeding hours as evaluated by the different thermal comfort models for the Maastricht TMY, relative to the ISO7730 result South zone (Kh) ISO % 100% ISSO74 122% 117% EN15251 adaptive 54% 12% North zone (Kh) 6

7 DISCUSSION The perception exists that adaptive thermal comfort models are usually less conservative and therefore less energy consuming than conventional static models. However, the study presented in this paper shows that this is not generally true. A lot of factors play an important role, such as the type of building (and corresponding cooling and heating loads throughout the year), the outdoor climate, and the set points for heating and cooling in all seasons. Thorough analysis of the different thermal comfort models applied to a moderate climate, reveals that there are very few periods for which the ISSO74 model has higher maximum operative temperatures than the non adaptive ISO7730 model. This shows the low potential of the ISSO74 model to allow higher indoor temperatures or to lower the cooling energy due to higher temperature set points. On the contrary, the ISSO74 model shows lower winter and midseason maximal operative temperatures compared to the ISO7730 model, increasing the cooling energy need. The adaptive EN15251 model deviates from this, but it allows such high indoor temperatures that comfort may be questionable. A cooling energy consumption analysis of a two zone office building with a sufficiently large heating and cooling power in a moderate climate, reveals that the application of adaptive thermal comfort models for defining the set points for heating and cooling, does not result in relevant energy savings. This is caused by the winter regime of the office building zones: the temperature limits are more stringent for cooling in the case of the adaptive models while even during winter time a large cooling need remains. Applying the same procedure to weather data of exceptionally warm years like 2003, does not change the conclusions. Simulating the office zones with lower but still realistic internal or external gains, again leads to the same conclusions. Analysis of the adaptive thermal comfort models shows that they, although by definition not applicable to TABS buildings, can be applied to these buildings. When TABS are used to condition the office zones, the number of excess hours as evaluated by ISSO74 is higher than when evaluated by the non-adaptive ISO7730 model. From this analysis, for a moderate climate and for office buildings with a cooling load during the whole year, it can be concluded that there is no good reason to use existing adaptive thermal comfort models instead of the basic non-adaptive thermal comfort model ISO7730, not only in the case of TABS buildings, but also in general. ACKNOWLEDGEMENT The authors greatly acknowledge the University of Leuven (K.U.Leuven) for providing the financial means to conduct this research. 7

8 REFERENCES [1] ANSI/ASHRAE Standard , Thermal Environmental Conditions for Human Occupancy [2] Lehmann B., Dorer V. and Koschenz M., Application range of thermally activated building systems tabs, Energy and Buildings, 2007, 39: p [3] Koschenz M. and Lehmann B., Thermoaktive Bauteilsysteme tabs, 2000, Duebendorf (Switzerland): EMPA Energiesysteme/Haustechnik. [4] ISO 7730:2005, Ergonomics of the thermal environment Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria. 2005, International Organization for Standardization, Switzerland. [5] Nicol F. and Humphreys M., Derivation of the adaptive equations for thermal comfort in freerunning buildings in European standard EN15251, Building and Environment, 2010, 45: p [6] de Dear R., Brager G. and Cooper D., Developing an Adaptive Model of Thermal Comfort and Preference, FINAL REPORT ASHRAE RP , ASHRAE. [7] de Dear R.J. and Brager G.S., Thermal comfort in naturally ventilated buildings : revisions to ASHRAE Standard 55, Energy and Buildings, 2002, Vol. 34: p [8] van der Linden A.C., et al., Adaptive temperature limits: A new guideline in The Netherlands (A new approach for the assessment of building performance with respect to thermal indoor climate), Energy and Buildings, 2006, Vol. 38: p [9] Pfafferott J.Ü., et al., Comparison of low-energy office buildings in summer using different thermal comfort criteria, Energy and Buildings, 2007, Vol. 39: p [10] Wagner A., et al., Thermal comfort and workplace occupant satisfaction--results of field studies in German low energy office buildings, Energy and Buildings, 2007, 39(7): p [11] Barlow S. and Fiala D., Occupant comfort in UK offices--how adaptive comfort theories might influence future low energy office refurbishment strategies, Energy and Buildings, 2007, 39(7): p [12] KONINKLIJK NEDERLANDS METEOROLOGISCH INSTITUUT (KNMI), [cited 2008 March]; Available from: [13] SEL-University of Wisconsin-USA and TRANSSOLAR-Stuttgart, TRNSYS 16, Transient System Simulation Program [14] Neumann C. and Herkel S., Langzeitmonitoring der Demonstrationsgebäude im Rahmen des Forschungsprogramms EnBau:MONITOR. 2. August 2005, Fraunhofer ISE, Gruppe Solares Bauen: Freiburg. [15] Koschenz M. and Lehmann B., Thermoaktive Bauteilsysteme tabs, 2000, Duebendorf (Switzerland): EMPA Energiesysteme/Haustechnik. [16] Tödtli J., et al., TABS Control, Steuerung und Regelung von thermoaktiven Bauteilsystemen, 2009, Zürich: Faktor Verlag. [17] van Hoof J. and Hensen J.L.M., Quantifying the relevance of adaptive thermal comfort models in moderate thermal climate zones, Building and Environment, 2007, Vol. 42: p