eScholarship provides open access, scholarly publishing services to the University of California and delivers a dynamic
research platform to scholars worldwide.
Center for the Built Environment
UC Berkeley
Peer Reviewed
Title:
Developing an Adaptive Model of Thermal Comfort and Preference Author:
de Dear, Richard , Macquarie University
Brager, G. S., University of California, Berkeley Publication Date:01-01-1998
Series:
Indoor Environmental Quality (IEQ)
Publication Info:
Indoor Environmental Quality (IEQ), Center for the Built Environment, Center for Environmental Design Research, UC Berkeley
Permalink:
/uc/item/4qq2p9c6
Additional Info:
ORIGINAL CITATION: de Dear, R.J., and G.S. Brager. 1998. “Towards an Adaptive Model of Thermal Comfort and Preference.” ASHRAE Transactions, Vol 104 (1), pp. 145-167.
Abstract:
The adaptive hypothesis predicts that contextual factors and past thermal history modify building occupants' thermal expectations and preferences. One of the predictions of the adaptive hypothesis is that people in warm climate zones prefer warmer indoor temperatures than people living in cold climate zones. This is contrary to the static assumptions underlying the current ASHRAE comfort standard 55-92. To examine the adaptive hypothesis and its implications for Standard 55-92, the ASHRAE RP-884 project assembled a quality-controlled database from thermal comfort field experiments worldwide (circa 21,000 observations from 160 buildings). Our statistical analysis examined the semantics of thermal comfort in terms of thermal sensation,acceptability, and preferenc
e, as a function of both indoor and outdoor temperature. Optimum indoor temperatures tracked both prevailing indoor and outdoor temperatures, as predicted by the adaptive hypothesis. The static predicted means vote (PMV) model was shown to be partially adaptive by accounting for behavioral adjustments, and fully explained adaptation occurring in HVAC buildings. Occupants in naturally ventilated buildings were tolerant of a significantly wider range of temperatures, explained by a combination of both behavioral adjustment and psychological adaptation. These results formed the basis of a proposal for a variable indoor temperature standard.
THIS PREPRINT IS FOR DISCUSSION PURPOSES ONLY , FOR INCLUSION IN ASHRAE TRANSACTIONS 1998, V. 104, Pt. 1. Not to be reprinted in whole or in part without written permission of the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., 1791 Tullie Circle, NE, Atlanta, GA 30329.Opinions, findings, conclusions, or recommendations expressed in this paper are those of the author(s) and do not necessarily reflect the views of ASHRAE. Written ABSTRACT
The adaptive hypothesis predicts that contextual factors and past thermal history modify building occupants' thermal expectations and preferences. One of the predictions of the adaptive hypothesis is that people in warm climate zones prefer warmer indoor temperatures than people living in cold cli
mate zones. This is contrary to the static assumptions underlying the current ASHRAE comfort standard 55-92. To examine the adaptive hypothesis and its implications for Standard 55-92,the ASHRAE RP-884 project assembled a quality-controlled database from thermal comfort field experiments worldwide (circa 21,000 observations from 160 buildings). Our statistical analysis examined the semantics of thermal comfort in terms of thermal sensation, acceptability, and preference, as a func-tion of both indoor and outdoor temperature. Optimum indoor temperatures tracked both prevailing indoor and outdoor temperatures, as predicted by the adaptive hypothesis. The static predicted means vote (PMV) model was shown to be partially adaptive by accounting for behavioral adjustments,and fully explained adaptation occurring in HVAC buildings.Occupants in naturally ventilated buildings were tolerant of a significantly wider range of temperatures, explained by a combination of both behavioral adjustment and psychological adaptation. These results formed the basis of a proposal for a variable indoor temperature standard.INTRODUCTION
Current comfort standards are intended to optimize the thermal acceptability of indoor environments. Unfortunately,they have tended to require energy-intensive environmental control strategies and often preclude thermally variable solu-tions, such as many climate-responsive and energy-conserv-ing designs, or innovative mechanical strategies that allow for personal control. These standards (ASHRAE 1992, ISO 1994)prescribe a narrow band of temperature to be applied
uniformly through space and time. They are based on a static model of thermal comfort that views occupants as passive recipients of thermal stimuli driven by the physics of the body’s thermal balance with its immediate environment, and mediated by autonomic physiological responses. The static model of thermal comfort is represented in contemporary ther-mal comfort standards such as the current ANSI/ASHRAE Standard 55-1992 (1992) that prescribe relatively constant indoor design temperatures with, at most, a slight seasonal difference to accommodate differences in summer and winter clothing patterns. These standards have come to be regarded as universally applicable across all building types, climate zones, and populations (e.g., Parsons 1994). But many researchers are beginning to challenge the assumption of universality, arguing that it ignores important cultural,climatic, social, and contextual dimensions of comfort, lead-ing to an exaggeration of the need for air conditioning (Kemp-ton and Lutzenhiser 1992).
Growing dissatisfaction with static comfort temperatures and the ensuing environmental impact caused by mismanage-ment of energy resources, has prompted interest in a variable indoor temperature standard to supplement the current Stan-dard 55. A variable indoor temperature standard, based on the adaptive model of thermal comfort, would have particular relevance to naturally ventilated buildings and other situations in which building occupants have some degree of i
ndoor climatic control. A variable temperature standard links indoor temperatures to the climatic context of the building and accounts for past thermal experiences and current thermal expectations of their occupants.
Ideally, a variable temperature standard would be based on an alternative to traditional comfort theory, termed the adaptive model of comfort, in which factors beyond funda-mental physics and physiology interact with thermal percep-tion. An important premise of the adaptive model is that building occupants are no longer regarded as passive recipi-
Developing an Adaptive Model
of Thermal Comfort and Preference
Richard J. de Dear, Ph.D.
Gail Schiller Brager, Ph.D.
Member ASHRAE
Richard de Dear is a the Deputy Director of the Climatic Impacts Centre for the School of Earth Scie
nces, Macquarie University, Sydney,Australia. Gail Schiller Brager is an Associate Professor of Architecture at the Center for Environmental Design Research, University of Cali-fornia, Berkeley.
SF-98-7-3 (4106) (RP-884)© 1998, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
(). Published in ASHRAE Transactions 1998, Vol 104, Part 1. For personal use only.
Additional distribution in either paper or digital form is not permitted without ASHRAE’s permission.
ents of the thermal environment, as in the case of climate chamber experimental subjects, but rather, play an active role in creating their own thermal preferences. Contextual factors and past thermal history are believed to modify expectations and thermal preferences. Satisfaction with an indoor climate results from matching actual thermal conditions in a given context and one’s thermal expectations of what the indoor climate should be like in that same context (Auliciems 1981, 1989, de Dear 1994a, Nicol 1993). In short, satisfaction occurs through appropriate adaptation to the indoor climatic environ-ment.
The generic term adaptation might be interpreted broadly as the gradual diminution of the organism’s
response to repeated environmental stimulation. Within this broad defini-tion it is possible to clearly distinguish three categories of ther-mal adaptation (Folk 1974, 1981, Goldsmith 1974, Prosser 1958, Clark and Edholm 1985):
Behavioral Adjustment. This includes all modifications a person might consciously or unconsciously make that in turn modify heat and mass fluxes governing the body’s thermal balance. Adjustment can be further sub-classified into personal (e.g., removing an item of clothing), technological (e.g., turning on an air conditioner), and cultural responses (e.g., having a siesta in the heat of the day).
Physiological. The most comprehensive definition of physiological adaptation would include changes in the phys-iological responses that result from exposure to thermal envi-ronmental factors, and which lead to a gradual diminution in the strain induced by such exposure. Physiological adaptation can be broken down into genetic adaptation (intergenera-tional) and acclimatization (within the individual’s lifetime).
Psychological. The psychological dimension of thermal adaptation refers to an altered perception of, and reaction to, sensory information due to past experience and expectations. Personal comfort setpoints are far from thermostatic. Relax-ation of expectations can be likened to the notion of habitu
a-tion in psychophysics (Glaser 1966, Frisancho 1981) where repeated exposure to a stimulus diminishes the magnitude of the evoked response.
In many commentators’ minds there is a belief that the static and “adaptive” schools of thought are irreconcilable (e.g., Auliciems 1989, Nicol 1993). The static heat balance models are grounded in a fairly linear, deterministic logic, and are tested with extensive and rigorous laboratory experiments yielding fairly consistent, reproducible results. But the simplistic cause-and-effect approach embodied in the static approach is not so easily applied to the more complex envi-ronments within real buildings populated by real occupants as opposed to subjects. Our opinion is that the adaptive perspec-tive complements rather than contradicts the static heat-balance view. The heat-balance model is more correctly regarded as a partially adaptive model, since it acknowledges the effects of behavioral adjustments made by occupants to thermal environmental parameters, clothing, and metabolic rate. We believe that a variable indoor temperature standard can successfully combine features of both the static and adap-tive models by incorporating behavioral, physiological, and psychological modes of thermal adaptation.
This paper reports results from the ASHRAE RP-884 project—Developing an Adaptive Model of Thermal Comfort and Preference. The research is premised on the development and analysis of a qu
ality-controlled, cumulative database of thermal comfort field experiments worldwide (see de Dear 1998 for more details on the RP-884 database). The specific objectives of RP-884 were to use this global database to: 1.Elaborate and define adaptive processes in the context of
indoor climatic perception.
2.Examine the semantics of thermal sensation, acceptability,
and preference scales within the context of an adaptive model of thermal comfort.
3.Develop statistical models of thermal comfort based on the
various processes of adaptation, including adjustment, acclimatization, and habituation.
4.Compare these adaptive models with predictions of the so-
called static models across the database.
5.Propose a variable temperature standard that, in time, might
eventually supplement and/or modify Standard 55.
This paper highlights the most significant findings of RP-884, while a more detailed treatment can be found in the project’s final report (de Dear et al., 1997). BACKGROUND
Brager and de Dear (1998) present an extensive literature review on thermal adaptation in the built environment, elab-orating the different mechanisms of adaptation, linking the static vs. adaptive comfort theories through a conceptual model with interactive feedback loops, and presenting a wide range of both climate chamber and field evidence for the different modes of adaptation. Many of the highlights of that previous work helped to clarify the conceptual approach and analysis of RP-884, and are presented here for background.
Of the three types of adaptation, behavioral adjustment of the body’s heat-balance probably offers the greatest opportu-nity for people to play an active role in maintaining their own comfort (Nicol and Humphreys 1972, Humphreys 1994a). The extent to which contextual factors offer building occu-pants scope to behaviorally interact with their indoor climate can be described in terms of adaptive opportunity (Baker and Standeven 1994). This concept helps to differentiate those buildings in which a deterministic relationship between the thermal environment and human response is applicable, and those in which an adaptive feedback loop is fully operational. Adaptive opportunity can be thought of as a continuum—at one extreme is the climate chamber, and at the other extreme we find the single-
occupant room with full adaptive possibil-ities from operable windows through to task-ambient air conditioning.
The evidence for physiological acclimatization is more thoroughly documented for heat exposure than for cold, and
for prolonged heat stress induced by a regimen of work in heat (Folk 1974, 1981, Fox 1974, Bruce 1960, Berglund and McNall 1973, Givoni and Goldman 1973). Unlike most behavioral adaptation, where a person consciously takes corrective action when uncomfortable, acclimatization is an unconscious feedback loop mediated by the autonomic nervous system. As shown later in this section, a review of the literature (Brager and de Dear 1998) demonstrated that accli-matization is not likely to be a factor for the moderate range of conditions found in most buildings.
Psychological adaptation encompasses the effects of cognitive and cultural variables, and describes the extent to which habituation and expectation alter thermal perceptions. The role of expectation in thermal comfort research was acknowledged in the earlier work of McIntyre (1980), who stated that “a person’s reaction to a temperature, which is less than perfect will depend very much on his expectations, personality, and what else he is doing at the time.” Although the least studied of the thr
ee adaptive mechanisms, psycho-logical adaptation might actually play the most significant role in explaining the differences between observed and predicted thermal responses. This can be seen particularly in light of different environmental contexts, such as the laboratory vs. home vs. office, or when comparing responses in air-condi-tioned vs. naturally-ventilated buildings (Fishman and Pimbert 1982, Heijs and Stringer 1988, Busch 1990, de Dear et al. 1991c, Rowe 1995, Oseland 1995).
Climate chamber evidence against the effects of acclima-tization on thermal comfort in moderate thermal environments comes from an experimental research design known as the preferred temperature method, in which the temperature within the chamber is directly controlled by its single subject. Using this technique, Fanger (1972 et al., 1977) tested subjects with differing climatic experiences (winter swimmers, work-ers from a refrigerated storeroom, long-term inhabitants of the tropics, and control groups), and found that their temperature preferences were all approximately the same. de Dear et al. (1991a) replicated Fanger’s tropical experiment with heat acclimated students on location in Singapore, and produced similar results. Gonzalez (1979) also studied the role of natu-ral heat acclimatization during a five day humid heat wave in New Haven, Connecticut. He found that for exercising subjects there was a discernible increase in preferred temper-ature after the heat wave
(Gonzalez 1979), but there were no statistically significant differences in resting subjects. In conclusion, on the basis of the majority of experimental evidence published to date, subjective discomfort and thermal acceptability under conditions most typically encountered in residences and office buildings, by resting or lightly active building occupants, appear to be unaffected by the physiolog-ical processes of acclimatization.
Although chamber studies have the advantage of careful control, field research is best for assessing the potential impacts of behavioral or psychological adaptations as they occur in realistic settings. Humphreys’ (1975) early review of 36 thermal comfort field studies worldwide produced one of the first, and most widely referenced, statistical relationships between indoor thermal neutralities and prevailing indoor temperatures. He found that building occupants were able to find comfort in indoor temperatures covering a broad band of more than 13 K, and attributed this to the adaptive processes, concluding that “. . . the range of recent experience is better regarded as one of the factors that will contribute to the accept-ability of the environment to which the respondent is exposed.” Subsequent work by both Humphreys (1978) and Auliciems (1981) found convincing evidence for a relation-ship between indoor thermal neutralities and outdoor climate, particularly in so-called free running buildings that had no centralized heating or cooling plant (i.e., naturally ventilated).
While this work has been widely cited as the first to reveal a strong statistical association between neutralities and outdoor climate, the actual causal mechanisms were left unclear. To more rigorously test the relative influences of behavioral, physiological, and psychological adaptive influ-ences, field researchers have to collect simultaneous measure-ments of all of the input variables to Fanger’s predicted mean vote (PMV) model (ISO 1994). de Dear (1994a) and Brager and de Dear (1998) present a meta-analysis of results from such field experiments conducted in both climate-controlled (air-conditioned) and free-running (naturally ventilated) buildings located in a broad spectrum of climates and seasons (Busch 1990, de Dear and Auliciems 1985, de Dear and Foun-tain 1994b, de Dear et al. 1991c, Donnini et al. 1996, Schiller et al. 1988). The purpose of the meta-analysis was to compare observed comfort temperatures (based on sensation votes) with those predicted by the static heat balance model (Fanger’s PMV index). The PMV model predicted comfort temperatures with reasonable accuracy in most air-condi-tioned buildings, but failed significantly in the naturally venti-lated buildings, with the magnitude of the discrepancy increasing in the more extreme climate zones of the meta-anal-ysis. Since all basic physical parameters governing the body’s heat balance were included in PMV’s calculations, including the previously ignored contribution of the insulating value of the chair, the mismatch between observation and prediction in naturally ventilated buildings implicate adaptive factors beyond the body’s heat-balance.adaptive
While we have known for a long time that clothing was a key input to the comfort problem (e.g., the clo inputs to Fanger’s 1970 PMV model), only a few studies have exam-ined field evidence of behavioral adjustment in the form of clothing changes. Fishman and Pimbert (1982) found that clo values had a strong linear dependence on outdoor weather and season, especially for women. Humphreys (1994b) and Nicol et al. (1994) concluded that as much as one-half the seasonal changes in comfort temperature could be attributed to clothing flexibility. In a longitudinal study, Nicol and Raja (1996) found that clothing changes were more strongly dependent on the succession of outdoor temperatures that occurred prior to the measurement, compared to the instantaneous or daily
mean outdoor temperature, or for that matter, the instanta-neous indoor temperature, implying that we dress more for outdoor climate than indoor climate. By asking separate ques-tions about availability, use, and effectiveness of a variety of behavioral adaptive mechanisms, Benton and Brager (1994) found that clothing adjustments were given one of the highest effectiveness ratings. These findings all support the hypothe-sis that the statistical dependence of indoor neutrality on outdoor climate may, in part, be due to behavioral adjustments that directly affect the heat balance, rather than acclimatiza-tion or habituation.
Evidence for psychological adaptation examines how contextual factors influence one’s perception of control and expectation, which in turn affect thermal response. Paciuk’s (1990) analysis of available control (adaptive opportunity), exercised control (behavioral adjustment), and perceived control (expectation) revealed that perceived degree of control was one of the strongest predictors of thermal comfort in office buildings, and had a significant impact in shaping both ther-mal comfort and satisfaction. This finding was also supported by the work of Williams (1995), in which office workers expressed higher levels of satisfaction when they perceived themselves to have more control over their environment. The effect of air conditioning on perceived control, expectation, and resulting thermal response has been investigated by several other researchers as well (Rowe et al. 1995, Fishman and Pimbert 1982, Black and Milroy 1966, Rohles et al. 1977). Their findings consistently indicate that people have a wider tolerance of variations in indoor thermal conditions if they can exert some control over them, such as in naturally ventilated buildings. In contrast, people in large open-plan air-condi-tioned buildings, typically devoid of any individualized control, had higher expectations for homogeneity and cool temperatures, and soon became critical if thermal conditions did not match these expectations.
Methods
Our literature review (Brager and de Dear 1998) indicated that the overwhelming weight of evidence supporting human thermal adaptation came from field research, rather than climate chamber laboratory experiments. Therefore, the RP-884 approach focused exclusively on field data, and began the process of assembling a database by sending a three-page questionnaire on field research methods to most of the thermal comfort research community currently or recently active in field research. On the basis of the questionnaire returns, we requested data from researchers whose:
came as close as possible to laboratory-grade,
2.data were structured to allow each set of questionnaire
responses to be linked to a concurrent set of indoor and outdoor climate observations, and 3.indoor climatic observations were comprehensive enough
to enable heat-balance indices (the static model) to be calculated for each questionnaire respondent.
A primary goal was to keep the internal consistency of the database as high as possible. To this end,
the RP-884 database was assembled from raw field data files instead of processed or published findings, enabling us to apply a variety of quality controls and standardized data processing techniques. Since the database is described in detail in de Dear 1998, the purpose of the next section is to briefly outline its contents and the basic steps taken to ensure its integrity.
Assembling the World Comfort Database
The raw data comprising the RP-884 database came from four continents and a broad spectrum of climatic zones. Nearly 21,000 sets of raw data were compiled from several locations in England and Wales, Bangkok, Thailand, several Califor-nian locations, Montreal and Ottawa in Canada, six cities across Australia, five cities in Pakistan, Athens in Greece, Singapore, and Grand Rapids in Michigan.
Each complete set of raw data was structured within the database using the template developed in previous ASHRAE-funded research projects, particularly RP-462 in a Mediterra-nean climate (Schiller et al. 1988), RP-702 in a hot-humid climate (de Dear and Fountain 1994c), and RP-821 in a cold climate (Donnini et al. 1996). The data fields included:•thermal questionnaire responses (sensation, acceptabil-ity, and preference),
•clothing and metabolic estimates,
•concurrent indoor climate observations (air and globe temperatures, air velocity, dew point, and plane radiant asymmetry temperature),
•thermal indices (mean radiant temperature, operative temperature, turbulence intensity, ET*, SET*, TSENS, DISC, PMV/PPD, and PD draft risk) were recalculated for each set of observations using the ASHRAE RP-781 software package known as the ASHRAE Thermal Comfort Tool (Fountain and Huizenga 1996),•outdoor meteorological observations including daily temperatures and relative humidities at 600 hours and 1500 hours, and daily effective temperatures (ET*) also calculated with the software package(excluding the effects of solar radiation).
Of all these variables, it was the clothing insulation esti-mate that provided the RP-884 team with the most difficulties, since a variety of estimation methods were used in the various database contributions. To standardize the database, clo unit estimates based on either the Sprague and Munson (1974) method (also described in McIntyre 1980), the ISO Standard 7730 (1984) method, or the ISO Standard 7730 (1994) method were converted into their equivalents under the Standard 55 technique by using a set of conversion coefficients described in de Dear (1998). Accompanying each clothing insulation
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论