You are on page 1of 4

Energy and Buildings 34 (2002) 533536

Extension of the PMV model to non-air-conditioned buildings in warm climates


P. Ole Fanger*, Jrn Toftum
Department of Energy Engineering, International Centre for Indoor Environment and Energy, Technical University of Denmark, Building 402, DK-2800 Lyngby, Denmark

Abstract The PMV model agrees well with high-quality eld studies in buildings with HVAC systems, situated in cold, temperate and warm climates, studied during both summer and winter. In non-air-conditioned buildings in warm climates, occupants may sense the warmth as being less severe than the PMV predicts. The main reason is low expectations, but a metabolic rate that is estimated too high can also contribute to explaining the difference. An extension of the PMV model that includes an expectancy factor is introduced for use in non-air-conditioned buildings in warm climates. The extended PMV model agrees well with quality eld studies in non-air-conditioned buildings of three continents. # 2002 Elsevier Science B.V. All rights reserved.
Keywords: PMV; Thermal sensation; Expectancy; Warm climates; Adaptation

1. The PMV model and the adaptive model The PMV model is based on extensive American and European experiments involving over a thousand subjects exposed to well-controlled environments [1]. The studies showed that the thermal sensation was closely related to the thermal load on the effector mechanisms of the human thermoregulatory system. The PMV model predicts the thermal sensation as a function of activity, clothing and the four classical thermal environmental parameters: air temperature, mean radiant temperature, air velocity and humidity. The advantage of the PMV model is that it is a exible tool that includes all the major variables inuencing thermal sensation. It quanties the absolute and relative impact of these six factors and can therefore, be used in indoor environments with widely differing HVAC systems as well as for different activities and different clothing habits. Considering that the PMV model has been an international standard since the 1980s [2,3], it has been used in practice outdoors and indoors in a wide range of buildings and vehicles in different parts of the world. The PMV model has been validated in climate chamber studies with Asian subjects [4,5] as well as in the eld, most recently in ASHRAE's world-wide research in buildings
Corresponding author. Tel.: 45-4525-4041; fax: 45-4593-2166. E-mail address: fanger@et.dut.dk (P. Ole Fanger). URL: http://www.ie.dtu.dk
*

with HVAC systems that were situated in cold, temperate and warm climates and were studied during both summer and winter [69]. ASHRAE's eld studies are characterized by the requirements for the use of high-quality measurements and procedures. They belong to class I in the quality classication of eld studies introduced by de Dear [10]. Other eld studies belong to class II with less stringent requirements, while the largest number of eld studies have low requirements for measurements and the description of activity and clothing. They belong to the lowest quality category, class III. While the PMV model predicts thermal sensation well in buildings with HVAC systems, eld studies in warm climates in buildings without air-conditioning have shown that it predicts a warmer thermal sensation than the occupants actually feel [11,12]. For such non-air-conditioned buildings an adaptive model has been proposed [13]. This model is a regression equation that relates the neutral temperature indoors to the monthly average temperature outdoors. The only variable is thus the average monthly outdoor temperature, which at its highest may have an indirect impact on the human heat balance. An obvious weakness of the adaptive model is that it does not include human clothing or activity or the four classical thermal parameters that have a wellknown impact on the human heat balance and therefore on the thermal sensation. Although, the adaptive model predicts the thermal sensation quite well for non-air-conditioned buildings of the 1900s located in warm parts of the world,

0378-7788/02/$ see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 7 7 8 8 ( 0 2 ) 0 0 0 0 3 - 8

534

P. Ole Fanger, J. Toftum / Energy and Buildings 34 (2002) 533536

the question remains as to how well it would suit buildings of new types in the future where the occupants may wear different clothing and change their activity pattern. Why then does the PMV model seem to overestimate the sensation of warmth in non-air-conditioned buildings in warm climates? There is general agreement that physiological acclimatization does not play a role [11]. One suggested explanation is that openable windows in naturally ventilated buildings should provide a higher level of personal control than in air-conditioned buildings. We do not believe that this is generally true in warm climates. Although, an openable window sometimes may provide some control of air temperature and air movement, this applies only to the persons who work close to a window [14]. What happens to persons in the ofce who work far away from the window? And in warm climates, the normal strategy in naturally ventilated buildings is to cool the building during the night and then close the windows some time during the morning when the outdoor temperature exceeds the indoor temperature. Another obstacle is of course trafc noise, which makes open windows in many areas impossible. We believe that in warm climates air-conditioning with proper thermostatic control in each space can provide a better perceived control than openable windows. 2. Extension of the PMV model Another factor suggested as an explanation to the difference is the expectations of the occupants. We think this is the major factor explaining why the PMV overestimates the thermal sensation of occupants in non-air-conditioned buildings in warm climates. These occupants are typically people who have been living in warm environments indoors and outdoors, maybe even through generations. They may believe that it is their ``destiny'' to live in environments where they feel warmer than neutral. If given a chance they may not on average prefer an environment that is different from that chosen by people who are used to air-conditioned buildings. But it is likely that they would judge a given warm environment as less severe and less unacceptable than would people who are used to air-conditioning. This may be expressed by an expectancy factor, e, to be multiplied with PMV to reach the mean thermal sensation vote of the occupants of the actual non-air-conditioned building in a warm climate. The factor e is estimated to
Table 1 Expectancy factors for non-air-conditioned buildings in warm climates Expectation Classification of non-air-conditioned buildings Location High Moderate Low In regions where air-conditioned buildings are common In regions with some air-conditioned buildings In regions with few air-conditioned buildings

vary between 1 and 0.5. It is 1 for air-conditioned buildings. For non-air-conditioned buildings, the expectancy factor is assumed to depend on the duration of the warm weather over the year and whether such buildings can be compared with many others in the region that are air-conditioned. If the weather is warm all year or most of the year and there are no or few other air-conditioned buildings, e may be 0.5, while it may be 0.7 if there are many other buildings with airconditioning. For non-air-conditioned buildings in regions where the weather is warm only during the summer and no or few buildings have air-conditioning, the expectancy factor may be 0.70.8, while it may be 0.80.9 where there are many air-conditioned buildings. In regions with only brief periods of warm weather during the summer, the expectancy factor may be 0.91. Table 1 proposes a rst rough estimation of ranges for the expectancy factor corresponding to high, moderate and low degrees of expectation. A second factor that contributes erroneously to the reported difference between the calculated PMV and actual thermal sensation votes in non-air-conditioned buildings is the estimated activity. In many eld studies in ofces, the metabolic rate is estimated on the basis of a questionnaire identifying the percentage of time the person was sedentary, standing, or walking. This mechanistic approach does not acknowledge the fact that people, when feeling warm, unconsciously tend to slow down their activity. They adapt to the warm environment by decreasing their metabolic rate. The lower pace in warm environments should be acknowledged by inserting a reduced metabolic rate when calculating the PMV. To examine these hypotheses further, data were downloaded from the database of thermal comfort eld experiments [10]. All data of at least quality class II and obtained in non-air-conditioned buildings during the summer period in warm climates were used in the analysis. Quality class II comprises data from eld studies in which all physical variables necessary for the calculation of PMV were collected at the same time and place as the questionnaires. However, measurements may not have complied with all recommendations of current thermal comfort standards. Data from four cities (Bangkok, Brisbane, Athens, and Singapore) were included, representing a total of more than 3200 sets of observations [1518]. The data from these four cities with warm climates were also used for the development of the adaptive model [13]. For each set of observations, recorded metabolic rates were reduced by 6.7% for every scale unit of PMV above

Expectancy factor, e Warm periods Occurring briefly during the summer season Summer season All seasons 0.91.0 0.70.9 0.50.7

P. Ole Fanger, J. Toftum / Energy and Buildings 34 (2002) 533536

535

Fig. 1. Thermal sensation in non-air-conditioned buildings in four cities in warm climates. Comparison of observed mean thermal sensation with predictions made using the new extension of the PMV model for non-air-conditioned buildings in warm climates. The lines are based on linear regression analysis weighted according to the number of responses obtained in each building.

neutral, i.e. a PMV of 1.5 corresponded to a reduction in the metabolic rate of 10%. Next, the PMV was recalculated with reduced metabolic rates using ASHRAE's thermal comfort tool [19]. The resulting PMV values were then adjusted for expectation by multiplication with expectancy factors estimated to be 0.9 for Brisbane, 0.7 for Athens and Singapore and 0.6 for Bangkok. As an average for each building included in the eld studies, Fig. 1 and Table 2 compare the observed thermal sensation with predictions using the new extended PMV model for warm climates. The new extension of the PMV model for non-airconditioned buildings in warm climates predicts the actual votes well. The extension combines the best of the PMV and the adaptive model. It acknowledges the importance of expectations already accounted for by the adaptive model, while maintaining the PMV model's classical thermal parameters that have direct impact on the human heat balance. It should also be noted that the new PMV extension predicts a higher upper temperature limit when the expectancy factor is low. People with low expectations are ready to accept a warmer indoor environment, which also agrees well
Table 2 Non-air-conditioned buildings in warm climates City Expectancy PMV adjusted factor, e to proper activity 0.6 0.7 0.7 0.9 2.0 1.2 1.0 0.9 PMVe adjusted for expectation 1.2 0.8 0.7 0.8 Observed mean vote 1.3 0.7 0.7 0.8

with the observations behind the adaptive model. An example: the relation between PMV and PPD predicts that 10% of a large group of occupants will be dissatised at a PMV of 0.5. Assuming that this relation remains unchanged also with occupants of low expectations, an expectancy factor will result in a higher upper temperature limit corresponding to a PMV of 0.5/e, but with no more than 10% dissatised. Table 3 shows the upper temperature limits adjusted for expectation, calculated for both 10 and 20% dissatised. A low expectation may change the upper temperature limit by 2 8C. It is only the upper temperature limit of the comfort zone that is affected by the expectancy factor, and no adjustment of the lower limit is expected. The database of quality eld studies in non-air-conditioned buildings in warm climates is still rather modest. Additional eld studies in such buildings in different parts of the world would therefore be useful to further rene the
Table 3 Example of upper temperature limits of the comfort zone adjusted for expectation in non-air-conditioned buildings in warm climates Expectancy factor, e 1 0.7 0.5 PPD (%) PMVe Unadjusted PMV 0.5 0.85 0.7 1.2 1 1.7 Upper temperature limit (8C) 27.9 28.7 28.3 29.4 29.0 30.5

10 20 10 20 10 20

0.5 0.85 0.5 0.85 0.5 0.85

Bangkok Singapore Athens Brisbane

Comparison of observed thermal sensation votes and predictions made using the new extension of the PMV model.

The calculations were made with a metabolic rate of 1 met, a clothing insulation of 0.5 clo, an air velocity of 0.3 m/s, and a relative humidity of 70%.

536

P. Ole Fanger, J. Toftum / Energy and Buildings 34 (2002) 533536

extended PMV model and to determine expectancy factors at different locations in warm climate regions. Such studies with measurements of class II, or even better class I, should include questions on occupants' expectations, acceptability and previous experience. It would also be useful to study the impact of warm ofce environments on work pace and metabolic rate. 3. Static and steady-state model The PMV model has been referred to as a static model, indicating that it should prescribe one constant temperature. But this is not true. The PMV model may actually predict any air temperature between 10 and 35 8C as neutral, depending on the other ve variables in the model. And it prescribes a certain range around the neutral temperature as acceptable, depending on the permitted percent dissatised. It is also sometimes claimed that the model applies only under steady-state conditions. It is correct that the PMV model was developed for steady-state conditions. It has, however, been shown to apply with good approximation at the relatively slow uctuations of the environmental parameters typically occurring indoors. It may even be applied as a rst approximation after step changes of the variables. Immediately after an upward stepwise change of temperature, the PMV model thus predicts well the thermal sensation, while it takes around 20 min at temperature down-steps [20]. 4. Conclusions The PMV model agrees well with high-quality eld studies in buildings with HVAC systems, situated in cold, temperate and warm climates and studied during both summer and winter. In non-air-conditioned buildings in warm climates, occupants may, however, perceive the warmth as being less severe than the PMV predicts. This is mainly caused by low expectations, but a metabolic rate that is estimated too high under warm conditions also contributes to explain the difference. An extension of the PMV model that includes an expectancy factor is proposed for use in non-air-conditioned buildings in warm climates. The extended PMV model agrees well with available quality eld studies in non-air-conditioned buildings in warm climates of three continents. Acknowledgements Financial support for this study from the Danish Technical Research Council is gratefully acknowledged.

References
[1] P.O. Fanger, Thermal comfort, Danish Technical Press, Copenhagen, Denmark, 1970. [2] CEN ISO 7730, Moderate thermal environmentsdetermination of the PMV and PPD indices and specification of the conditions for thermal comfort, International Organization for Standardization, Geneva, 1994. [3] Thermal environmental conditions for human occupancy, American Society of Heating, Refrigerating and Air-conditioning Engineers Inc. ASHRAE 55-1992R. [4] R. de Dear, K.G. Leow, A. Ameen, Thermal comfort in the humid tropics. Part I. Climate chamber experiments on temperature preferences in Singapore, ASHRAE Transactions 97 (1) (1991) 874879. [5] S. Tanabe, K. Kimura, T. Hara, Thermal comfort requirements during the summer season in Japan, ASHRAE Transactions 93 (1) (1987) 564577. [6] K.M. Cena, in: K. Cena, R. de Dear (Eds.), Field study of occupant comfort and office thermal environments in a hot-arid climate, Final report ASHRAE 921-RP, ASHRAE Inc., Atlanta, 1998. [7] G. Donini, J. Molina, C. Martello, D. Ho Ching Lai, K. Ho Lai, C. Yu Chang, M. La Flamme, V.H. Nguyen, F. Haghihat, Field study of occupant comfort and office thermal environments in a cold climate, Final report ASHRAE 821 RP, ASHRAE Inc., Atlanta, 1996. [8] R. de Dear, M. Fountain, S. Popovic, S. Watkins, G. Brager, E. Arens, C. Benton, A field study of occupant comfort and office thermal environments in a hot-humid climate, Final report ASHRAE 702 RP, ASHRAE Inc., Atlanta, 1996. [9] G.E. Schiller, E. Arens, F. Bauman, C. Benton, M. Fountain, T. Doherty, A field study of thermal environments and comfort in office buildings, Final report ASHRAE 462 RP, ASHRAE Inc., Atlanta, 1988. [10] R. de Dear, A global database of thermal comfort field experiments, ASHRAE Transactions 104 (1b) (1998) 11411152. [11] G.S. Brager, R. de Dear, Thermal adaptation in the built environment: a literature review, Energy and Buildings 27 (1998) 8396. [12] M.A. Humphreys, Outdoor temperatures and comfort indoors, Building Research and Practice 6 (2) (1978) 92105. [13] R. de Dear, G.S. Brager, Developing an adaptive model of thermal comfort and preference, ASHRAE Transactions 104 (1a) (1998) 145167. [14] J.F. Nicol, M.R.B. Kessler, Perception of comfort in relation to weather and adaptive opportunities, ASHRAE Transactions 104 (1b) (1998) 10051017. [15] J.F. Busch, A tale of two populations: thermal comfort in airconditioned and naturally ventilated offices in Thailand, Energy and Buildings 18 (1992) 235249. [16] R. de Dear, A. Auliciems, Validation of the predicted mean vote model of thermal comfort in six Australian field studies, ASHRAE Transactions 91 (2) (1985) 452468. [17] N. Baker, M. Standeven, in: Proceedings from CIBSE National Conference on A Behavioural Approach to Thermal Comfort Assessment in Naturally Ventilated Buildings, 1995, pp. 7684. [18] R. de Dear, K.G. Leow, S.C. Foo, Thermal comfort in the humid tropics: field experiments in air-conditioned and naturally ventilated buildings in Singapore, International Journal of Biometeorology 34 (1991) 259265. [19] M.E. Fountain, C. Huizenga, A thermal sensation prediction tool for use by the profession, ASHRAE Transactions 103 (2) (1997) 130136. [20] R. de Dear, J.W. Ring, P.O. Fanger, Thermal sensations resulting from sudden ambient temperature changes, Indoor Air 3 (1993) 181192.

You might also like