- Email: [email protected]
Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment
Accepted Manuscript Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment Wenjie Ji, Bin Cao, Maohui Luo, Yingxin Zhu PII:
S0360-1323(16)30513-3
DOI:
10.1016/j.buildenv.2016.12.021
Reference:
BAE 4751
To appear in:
Building and Environment
Received Date: 11 September 2016 Revised Date:
28 November 2016
Accepted Date: 12 December 2016
Please cite this article as: Ji W, Cao B, Luo M, Zhu Y, Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment, Building and Environment (2017), doi: 10.1016/j.buildenv.2016.12.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment Wenjie Ji a, b, Bin Cao a, b, * Maohui Luo a, b, Yingxin Zhu a, b Department of Building Science, Tsinghua University, Beijing 100084, China b Key Laboratory Eco Planning & Green Building, Ministry of Education, Tsinghua University, China a
Corresponding email: [email protected]
RI PT
*
Corresponding phone (+86) 010 62782746
EP
TE D
M AN U
SC
ABSTRACT The purpose of this study is to explore how a short-term thermal experience influences thermal comfort evaluation. Thermal experience, which refers to the previous thermal environment, may result in the formation of some “memory” on humans. When people enter another environment where the temperature is different from the previous one, the previous experience may result in some different feelings and changes on the evaluations of thermal comfort, comparing with staying in a steady state condition. In this paper, we mainly focus on short-term thermal experience within the time scale of minutes to hours. Climate chamber experiments were conducted for analysis and discussion. The experiment we designed had three sets of conditions: 1) started and ended at an air temperature of 20oC, and experienced higher temperatures in between; 2) started and ended at an air temperature of 25oC, and experienced higher or lower temperatures in between, and 3) started and ended at an air temperature of 30oC, and experienced lower temperatures in between. The evaluations of thermal comfort of the subjects at different temperature conditions were recorded by questionnaires. We found that both comfort and discomfort resulted from the contrast between the current and previous conditions. Even though the initially poor thermal environment was improved a little bit, the evaluation of the thermal comfort would be improved a lot. Additionally, the decrease of thermal sensation caused by cold stimulation was more obvious than the increase due to hot stimulation. People’s the evaluations could be considered as a combination of both the past and the present feelings.
AC C
KEYWORDS Thermal experience, chamber experiment, thermal sensation, step-changed temperature, adaptive thermal comfort 1.INTRODUCTION Indoor thermal environmental quality contributes to the thermal comfort perception, wellbeing and performance of the occupant [1]. Human thermal comfort is defined as “that condition of mind which expresses satisfaction with the thermal environment” [2][3]. Thermal comfort conditions can be evaluated by using a lot of indices based on environmental variables such as air temperature, humidity, wind speed as well as behavioral ones such as clothing and activities [4]. What kind of indoor climate should be created and how to implement it are closely connected with human health and the building energy consumption. Accordingly, it is essential to examine our requirements for indoor space conditioning, and to judge if it is necessary to maintain what are now regarded as comfortable environments in some current standards.
ACCEPTED MANUSCRIPT
RI PT
1.1.Current development of adaptive thermal comfort There are two classic theories to evaluate indoor thermal comfort. One of them is Fanger’s predicted mean vote (PMV) model, which is based on heat balance calculation of the human body [5]. The other theory is the adaptive thermal comfort model, which emphasizes that not only the physical environment influences thermal comfort, but also the human body itself which has the ability to adapt to the surroundings [6]. Ever since the concept of adaptive thermal comfort was announced in the 1960s [7], it has been continually improved and developed. In 1998, De Dear and Brager proposed the adaptive thermal comfort model where they claimed that people can interact with the environment in three ways: physiological acclimatization, behavioural adjustment, and psychological habituation [8]. Their model provided us with a new perspective to understand thermal comfort.
M AN U
SC
Some studies indicate that in air-conditioned buildings, the comfortable temperature range is getting narrower nowadays than before [9]. However, Arens found that occupants’ satisfaction did not show obvious differences in the indoor environments of different levels according to the current standards [10]. It was suggested that thermal comfort and satisfaction can hardly be improved by steady-state control even with high control precision, thus adaptive thermal comfort is expected to become the prevalent trend in the future [11]. As this theory became accepted and widely used by more and more people, it was also introduced into some standards, including ASHRAE Standard 55 [12], EN 15251 [13] and China GB/T 50785 [14], etc.
TE D
Nevertheless, some issues of the adaptive thermal comfort theory remain to be solved and explained in depth [15]. As Halawaand van Hoof stated [16], the underlying premises and the logic of adaptive processes still need rethinking. Additionally, the individual roles in the behavioral, physiological and psychological adaptive processes should be further identified [17].
AC C
EP
1.2. Statement of the problem The adaptive thermal comfort model is usually described as a “black box” model, as its mechanism is not clear to some extent. Thermal experience is an important part of adaptive thermal comfort. There are some current research studies focusing on thermal experience. Gagge [18], who was one of the earliest scholars to study on short-term thermal experience, found some laws of human body reactions in variable temperature environment. Van der Lans et al. [19] reported that after a 10-day cold acclimation, subjects perceived the environment with a warmer sensation, felt more comfortable in the cold, and reported less shivering. Chun et al. [20] conducted a comparative study using Korean and Japanese subjects, who had a 1day different thermal experiences. Their results showed that when entering a similar environment, the subjects who previously experienced higher temperature reported a 0.44 lower thermal sensation than those with cooler initial experience. Kelly Lisa and Ken Parsons [21] found that thermal sensation changed immediately along with the change of environmental thermal parameters, and put forward a transient PMV model. Nagano et al. [22] did experiments about step-changed temperature, and found that the neutral temperature changed. Yu et al. [23] showed that subjects who were used to naturally ventilated environments had a better physiological thermoregulation when exposed to high temperatures. Addiotional studies also discussed the effects of short-term hot or cold stimulus on the human body and found that people have the capacity of self-adaptation and adjustment [24].
ACCEPTED MANUSCRIPT Short-term thermal experience refers to hot or cold stimulus within the time scale of minutes to hours. To date, there are hardly any consensus about how the short-term thermal experiences influence thermal comfort and thermal adaptation.
RI PT
1.3. Objective This study focuses on the impacts of short-term thermal experience on thermal comfort evaluations, considering different directions and ranges of temperature change. We tried to find out whether the thermal comfort evaluations simultaneously resulted from the current surrounding environments, as well as the previous thermal experience, and found some quantitative relationship between temperature change and thermal evaluations.
AC C
EP
TE D
M AN U
SC
2. METHODS 2.1. Climate chamber The experiment was conducted in an artificial climate chamber (5m×3m×2.7m), where the temperature and humidity can be accurately controlled. There are two rooms in the chamber. The environments in the two rooms can be controlled separately by an automatic control system, which means the temperature and humidity can be different in the two rooms. There is a door between the two rooms. When needed, subjects can enter one room from the other using this door. In this experiment, we mainly observed how the temperature changes influenced the human body comfort and thermal sensation. The air humidity was controlled to be in the range of 45~55% in both rooms. When the air conditioning system is on, the average air velocity in the chamber is 0.05m/s, which could be considered to have no impact on human thermal comfort. The automatic control interface of the climate chamber is shown in Figure 1.
Figure 1. The automatic control interface of the artificial climate chamber
2.2. Experimental protocol The experiment was carried out in May of 2015 in Beijing, which is located in the Cold Zone according to China’s national standard [27] . The whole climate chamber was in a separate room, as a result, the indoor thermal environment in the chamber couldn’t be influenced by the outdoor temperature. In order to verify if the previous short-term thermal experience has an impact on the thermal evaluation of the current environment by the occupants, the experimental conditions in this study were set as shown in Figure 2. In total, there were three groups of conditions, including: 1) started and ended at an air temperature of 20 oC, and
ACCEPTED MANUSCRIPT experienced higher temperatures of 22, 25 and 30 oC in between; 2) started and ended at an air temperature of 25 oC, and experienced higher temperature (30 oC) or lower temperatures (20 o C) in between; 3) started and ended at an air temperature of 30oC, and experienced lower temperatures of 20, 25 and 28 oC in between. There were 8 experimental conditions in all.
22
20
25 30
25
20
25
30oC started and ended at an air temperature of 30℃ 20 20
30
25oC
22oC
28
30
30
25
28o C
RI PT
started and ended at an air temperature of 20℃
started and ended at an air temperature of 25℃
20oC
SC
Figure 2. Three experimental conditions
M AN U
We recruited 8 subjects for the experiments. Each subject participated in all conditions. Totally, there were 64 samples. The subjects were college students in good health conditions. They were not allowed to have intense physical exercise before the experiments. During the experiments, all subjects wore the same standard clothes (short T-shirt, thin trousers and sneakers), with a thermal insulation of 0.57 clo, and their metabolic rate was 1.1 met. The subjects were seated and did some light work, such as reading and typing. The subjects had to fill in a questionnaire several times during each experimental condition. There were mainly three questions in the questionnaire, which were Thermal Sensation Vote (TSV), Thermal Comfort Vote (TCV), and thermal expectation. The scale for each question is shown in Table 1.
TE D
Table 1. the scales of three questions TSV, TCV and Thermal expectation Scale TSV TCV Thermal expectation Hot
Very comfortable
--
2
Slightly hot
Comfortable
--
1
Warm
Just comfortable
Warmer
EP
3
Neutral
Neutral
Unchanged
-1
Cool
Just uncomfortable
Cooler
-2
Slightly Cold
Uncomfortable
--
Very uncomfortable
--
AC C
0
-3
Cold
There are three phases in each experimental condition. Taking the first group, which started and ended at an air temperature of 20 oC, as an example: a) the adaptive phase (1 to 20 min). Subjects stayed in a 20 oC environment for 20 minutes, and filled in the questionnaires 3 times (10 min, 5 min, 5 min); b) the temperature change phase (21 to 60 min). Subjects entered a 22 (25 and 30) oC environment and stayed for 40 minutes, in the meantime they filled in the questionnaires 10 times (1min, 1min, 1min, 1min, 1min, 5 min, 5 min, 5 min, 10 min 10 min); c) the rebound phase (61 to 90 min). Subjects went back to the 20 oC environment and stayed for 30 minutes, filled in the questionnaires 8 times (1min, 1min, 1min, 1min, 1min, 5 min, 10 min, 10 min). In phases b) and c), the questionnaires were conducted more frequently when the subjects just entered that phase. The three phases and the times when subjects filling in the questionnaires are shown in Figure 3.
ACCEPTED MANUSCRIPT
Figure 3. The time length of the three phases in each experimental condition (The red marks are the moment when subjects need to fill in the questionnaires)
RI PT
3. RESULTS 3.1. Condition that started and ended at an air temperature of 20 oC The results from the condition where air temperature started and ended at 20 oC are shown in Figures 4-8. All the results shown in these figures are expressed as the mean values of TSV, TCV or thermal expactation of all the 8 subjects.
TE D
M AN U
SC
The solid lines in Figure 4 correspond to the TSV from 20 oC to the other temperatures then back to 20 oC , and the dotted lines correspond to the PMV which is calculated using the known parameters. From a) the adaptive phase to b) the temperature variation phase: The thermal sensations of the subjects who went into a higher temperature from 20 oC, were all improved significantly, even though they went into 22 oC, which was still a somewhat cold environment, they felt slightly warm. And with the passage of time, TSV in the 22, 25 and 30 o C temperatures were all slowly decreased. A comparison of the TSV and PMV, revealed that the bigger the temperature changed, the smaller the difference between the two. And in the case of 22 and 25 oC, with temperature differences of 2 and 5 degrees, TSV is higher than PMV, but TSV is lower than PMV in the 30 oC condition. Traditionally, we thought that the current thermal sensation would be amplified with the contrast of thermal experience before and after. However, it seems to be different in the 20 oC to 30 oC experimental conditions, with a temperature difference of 10 oC.
AC C
EP
Additionally, from b) the temperature variation phase to c) the rebound phase: The thermal sensation of the subjects who went back to 20 oC from higher temperature decreased quickly, but the TSV were all higher than the PMV in the 20 oC condition, the TSV in the 30 oC condition is the highest in the rebound phase. It seems that when people come from a hot environment to a cold environment, they may not feel very cold.
ACCEPTED MANUSCRIPT 3
a） ）
c） ）
b） ）
TSV22 TSV25 TSV30 PMV22 PMV25 PMV30
2
0 0
10
20
30
40
50
-1
70
80
90
100
SC
-2
60
RI PT
TSV
1
-3
M AN U
time(min) Figure 4. TSV and PMV of the condition that started and ended at an air temperature of 20 oC
1.5
EP
TE D
In addition, the results presented in Figure 5 show how the deviation of the TSV and the PMV [△(TSV-PMV)] further changed in the process. At the moment of going from 20 oC to other environment, the TSV is higher than the PMV in the 22 oC and 25 oC conditions. Additionally, △(TSV-PMV) is bigger at first, and then slowly goes down; in the 30 oC condition, the TSV is almost equal to PMV at the beginning of the change, then the deviation gradually grows, and it is opposite to 22 and 25 oC, whereas the TSV is lower than PMV throughout the process. In the dynamic process, the difference is obvious between the predicted value of PMV and the real value of TSV. Accordingly, using the PMV model to estimate thermal sensation is not accurate, as it will be lower than the actual sensation in a moderately cold environment and higher than the actual in a moderately hot environment. That is the reason why the deviations in this experiment have both positive and negative values, mainly due to the contrast between the before and after environments, namely the effects of short-term thermal experience.
22
25
30
(TSV-PMV)
AC C
1
0.5
0
10 15 20 21 22 23 24 25 30 35 40 50 60 61 62 63 64 65 70 80 90
-0.5
-1
time to vote
Figure 5. The deviation of the TSV and PMV in the whole process of the condition that started and ended at an air temperature of 20 oC
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
The disparity between the TSV and PMV is obvious in different temperature conditions. And, at the moment of going into another environment, the TSV changes quickly. Evidently, at the same time, we have a new PMV value with new parameters, so we have the variation of TSV (△TSV) and the variation of PMV (△PMV) at the moment the environment changes, using the average value of the first 5 min. The relationship between the △TSV and △PMV with the temperature difference in an instant is shown in Figure 6. The two dotted lines are the △PMVs, namely the △PMV from 20 oC to other temperatures and the △PMV back to 20 oC, after the stimuli are exactly symmetrical, they have a linear relationship with the temperature difference, with a slope of ± 0.3266. Moreover, the solid lines are △TSVs, apparently there is a scissors difference between the △TSV and the △PMV. The trends of the △TSV from 20 oC to other temperatures and the △TSV back to 20 oC after stimuli are similar. As we can see, when the temperature difference is small, the variation of TSV is more obvious than the variation of PMV, both in two conditions. However, when the temperature difference is large, the △TSV is less obvious than △PMV. It is noteworthy that the cut-off points of the scissors difference are different. From 20 oC to other conditions, namely from cold to warm, the cutpoint is about 7.5 oC, while for the other one having a warmer or hot stimulus, then going back to 20 oC is about 5 oC. This means that from 20 oC to other hot environments, the variation of the TSV is bigger than that of the PMV when the temperature difference is less than 7.5 oC, and the actual feeling will be hotter. However, when the temperature difference is greater than 7.5 oC, the actual feeling will not be that hot. On the other hand, after a hot stimulus and back to 20 oC, the actual thermal sensation will be colder when the temperature difference is less than 5 oC, and it will not be so cold when the temperature difference is greater than 5 oC. 4
y = 0.3266x - 0.0007
1 0 0 -1 -2 -3 -4
2
AC C
△TSV, △PMV
2
EP
TE D
3
4
6
y = 0.1254x + 1.5199 TSV 20℃ to others
TSV others back to 20℃ 8
10
12 PMV 20℃ to others
y = -0.1185x - 1.0184 PMV others back to 20℃ y = -0.3266x + 0.0007
temperature difference (℃)
Figure 6. The relationship between △TSV and △PMV with the change of the temperature in an instant (the first 5 minutes) in the condition that started and ended at an air temperature of 20 oC
In addition, the TCV and thermal expectation in the whole process were also recorded. In Figure 7, from a) the adaptive phase to b) the temperature variation phase, the TCVs were all increased from 20 oC to other temperatures, and going into 22 oC was the most comfortable. Additionally, from b) the temperature variation phase to c) the rebound phase, at the moment of going back to 20 oC, the TCV were all decreased, even lower than those in a) the adaptive phase at 22 oC and 25 oC. It is interesting that from 30 oC to 20 oC, the uncomfortable feeling
ACCEPTED MANUSCRIPT was not so obvious under a temperature difference of 10 oC, which coincided with the analysis of the TSV before. 3
c） ）
b） ）
a） ）
22
2
25 30
0 0
10
20
30
40
50
60
-2
-3
80
90
100
M AN U
time(min)
70
SC
-1
RI PT
TCV
1
Figure 7. The TCV of the condition that started and ended at an air temperature of 20 oC
TE D
Similarly, the thermal expectation shown in Figure 8, from a) the adaptive phase to b) the temperature variation phase, the expectation of warmness reduced, and to c) the rebound phase, they felt cold and hoped for warmer temperature, the expectation was more intense than a) the adaptive phase, and in the 30 oC condition, the expectation of coolness was not so apparent as in the 22 and 25 oC conditions as well. All the variation of the feelings was significant in the instant of changing the environment, and after adapting to the new temperature, they would be the steady state.
AC C
thermal expectation
0.5
0
0
-0.5
-1
22
EP
1
10
a） ）
20
30
25 30
40
50
60
70
80
90
100
c） ）
b） ） time(min)
Figure 8. The thermal expectation of the condition that started and ended at an air temperature of 20 oC
The trend from 30 oC to 20 oC was intriguing with a temperature difference of 10 oC. Maybe having the short-term thermal experience of relative coldness will dull the perception of relative hot. Alternatively, people may have formed some impression in their brain, like
ACCEPTED MANUSCRIPT memories of short-term history, when going into another environment where the temperature contrast is bigger, the thermal inertia will lead them to an erroneous judgement of the current environmental temperature.
SC
RI PT
3.2. Condition that started and ended at an air temperature of 30 oC The TSV and PMV of the whole process of the condition that started and ended at an air temperature of 30 oC are shown in Figure 9. From a) the adaptive phase to b) the temperature variation phase: at the moment of going into a lower temperature from 30 oC, the thermal sensations were all reduced, and in the 28 oC and 25 oC conditions, the TSVs were lower than PMVs, which was more obvious in the 28 oC condition, but at 20 oC the TSV was higher than the PMV. The deviation of the TSV and PMV was not as the condition that started and ended at an air temperature of 20 oC, along with the temperature changing. With the passage of time, the TSV in the 28, 25 and 20 oC were all slowly increased to the steady state temperature. Then from b) the temperature variation phase to c) the rebound phase: The subjects went from 28 oC back to 30 oC, the TSV was the highest, but from 20 oC to 30 oC, the TSV was lower than the PMV, which implied that from cold to hot environment people did not feel very hot.
a） ）
b） ）
2
0
-1
-2
-3
10
20
30
40
TE D
TSV
1
0
M AN U
3
50
60
70
c） ）
80
90
100
20 25 28 PMV20 PMV25 PMV30
EP
time(min) Figure 9. The TSV and PMV of the condition that started and ended at an air temperature of 30 oC
AC C
The change of the deviation of the TSV and PMV [△(TSV-PMV)] is shown in Figure 10. From 30 oC to other environments, the TSV is higher than the PMV in the 20 oC condition, and it is lower than the PMV in the 25 oC and 28 oC conditions, and there is a peak of the △(TSV-PMV), and then gradually starts to go down.
ACCEPTED MANUSCRIPT 1.5 20
25
28
0.5
RI PT
△(TSV-PMV)
1
0
10 15 20 21 22 23 24 25 30 35 40 50 60 61 62 63 64 65 70 80 90
-1
time to vote
SC
-0.5
M AN U
Figure 10. The deviation of TSV and PMV during the whole process of the condition that started and ended at an air temperature of 30 oC
TE D
It is identical to the condition started and ended at an air temperature of 20 oC, the relationship of △TSV and △PMV with the temperature difference in an instant is shown in Figure 11. The linear relationship of △PMV and the temperature difference are the same, with a slope of ± 0.3266. However, the cut-off points of the scissors difference are different in this condition. From 30 oC to other temperatures, specifically from hot to cool, the cut-off point is about 5.5 o C, while another is about 3.7 oC. This means that from 30 oC to the other temperatures, the variation of the TSV is larger than that of the PMV when the temperature difference is less than 5.5 oC, whereas if it is over 5.5 oC, the actual feeling will not be so cold. Meanwhile, in another condition involving having a cooler or cold stimulus then going back to 30 oC, the actual thermal sensation will be warmer when the temperature difference is less than 3.7 oC, and it will not feel so hot when it is greater than 3.7 oC.
EP
4
△ TSV, △ PMV
2 1 0
AC C
3
0
-1 -2
2
y = 0.3266x - 0.0009
TSV 30℃-others
y = 0.1696x + 0.5804
TSV others- 30℃ 4
6
8
10
12 PMV 30℃-others
y = -0.2091x - 0.6469 PMV others- 30℃
-3 y = -0.3266x + 0.0009 -4
temperature difference (℃ ℃)
ACCEPTED MANUSCRIPT Figure 11. The relationship between the △TSV and △PMV with the change of the temperature in an instant (the first 5 minutes) in the condition that started and ended at an air temperature of 30 oC
RI PT
The TCV and thermal expectation in the whole process were recorded as shown in Figure 12. From the adaptive phase to b) the temperature variation phase, the TCV were all increased from the 30 oC condition to other temperatures, and going into the 25 oC environment was the most comfortable. Additionally, from b) the temperature variation phase to c) the rebound phase, when going into the 20 oC environment, the uncomfortable feeling was not so obvious. Actually, the bigger the temperature difference, the less uncomfortable they felt, and the trend of thermal expectation is the same, as shown in Figure 13. 3
a） ）
c） ）
b） ）
2
25 28
SC
0 0
10
20
30
-1
-2
-3
M AN U
1
TCV
20
40
50
60
70
80
90
100
TE D
time(min)
Figure 12. The TCV of the condition that started and ended at an air temperature of 30 oC
1
a） ）
0
-0.5
-1
20 25
EP AC C
thermal expectation
0.5
0
c） ）
b） ）
10
20
30
28
40
50
60
70
80
90
100
time(min)
Figure 13. The thermal expectation of the condition that started and ended at an air temperature of 30oC
3.3. Condition that started and ended at an air temperature of 25 oC In this condition, when going into the 30 oC environment, the TSV was higher than the PMV, the sense of cold before and hot after under the temperature difference of 5 oC, which may
ACCEPTED MANUSCRIPT
RI PT
amplify the thermal sensation, in accordance with the result in the condition started and ended at an air temperature of 20 oC, indicating that from cold to hot environments, the variation of the TSV is larger than that of the PMV when the temperature difference is less than 7.5 oC, and the actual feeling will be hotter. In contrast, from 25 oC to 20 oC, the PMV was lower than the TSV, people did not feel so cold in 20 oC. Both, at the same temperature difference of 5 oC, there is a disparity in the thermal sensation from “hot to cold” and “cold to hot”. The TCV, thermal expectation and some other analyses were similar to the other two conditions above. 3
a） ）
c） ）
b） ）
2
0
10
20
30
40
50
SC
0
60
70
M AN U
TSV
1
-1
-2
-3
80
90
100
20 30 PMV20 PMV30
time(min) Figure 14. TSV and PMV of the condition that started and ended at an air temperature of 25 oC
AC C
EP
TE D
3.4. Contrast of the three conditions To further compare the three conditions, taking the average of the TSV at a certain temperature, also considering the PMV, and the differences between the evaluations were examined. The real thermal sensation under the three conditions and the predicted thermal sensation at different temperatures are shown in Figure 15. In the condition started and ended at an air temperature of 20 oC, with the short-term thermal experience of cold, specifically 20 o C, as long as the environmental temperature increased a little, the termal sensation would improve, even at the 30 oC condition it was not so hot. In the condition started and ended at an air temperature of 30 oC, the thermal sensation would improve as long as the environmental temperature was lower, and from 30 oC to 20 oC, people would not feel so cold. Futhermore, though the TSV, in the condition started and ended at an air temperature of 25 o C, was still exhibiting some differentces with the PMV, the overall trend was closest to the PMV, while the cold or hot thermal experience both influence the thermal sensative vote. This implies that the real thermal sensation was usually unequal to PMV, and different short-term thermal experience may lead to different deviation.
ACCEPTED MANUSCRIPT 2 1.5
0.5 0 20
22
25
28
30
RI PT
cold→hot
1
-0.5
started and ended at 20℃
-1
started and ended at 25℃
-1.5
started and ended at 30℃ PMV
temperature(℃)
SC
-2
Figure 15. The TSV of three conditions and the PMV at different temperatures
30℃ ℃
EP
TE D
M AN U
4. DISCUSSIONS 4.1. The variation of TSV along with the temperature change In conclusion, there were four types of temperature change in this experiment (25 oC is considered to be the neutral temperature): a. Cold to neutral; b. Cold to hot; c. Hot to cold; d. Hot to neutral, as shown in Figure 16. In the experiment of the short-term thermal experience, we found that the variations of the TSV and TCV are different in various situations, and drew a conclusion that when the environment changed from cold/hot to neutral condition, or in other words from uncomfortable to comfortable, no change in the nature of thermal sensation was reported, the TSV and TCV would be enlarged. However, when temperature changed from one extreme to another with large temperature difference, like from a cold 20 oC to a hot 30 oC condition and vice versa, they would not feel very hot/cold, and the uncomfortable feeling would not be very strong. In addition, taking the condition that started and ended at an air temperature of 20 oC as an example, we found that compared with going to 25 oC, going from 20 oC to 22 oC, both the TSV and TCV had the largest promotion and improvement, this suggested that moderate stimulation would be the most effective way to improve thermal sensation and comfort. 30℃ ℃
30℃ ℃
28℃ ℃
28℃ ℃
28℃ ℃
28℃ ℃
25℃ ℃
25℃ ℃
25℃ ℃
25℃ ℃
22℃ ℃
22℃ ℃
22℃ ℃
22℃ ℃
20℃ ℃
20℃ ℃
20℃ ℃
20℃ ℃
a. Cold to neutral
b. Cold to hot
c. Hot to cold
d. Hot to neutral
AC C
30℃ ℃
Figure 16. Four types of temperature change
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
There is a conjecture that thermal experience can establish a memory, or named habit. It is maybe some nerve impulses in the brain cortex, maybe some psychological adaption, maybe some sensorial characteristic. Whatever happens, something like a memory can influence the sensation of a new environment, from an involuntary comparison with past experience. Longterm history is similar to long-term memory, while short-term history is similar to short-term memory. Some instructed information can be retained by sense organs for a short time period at first, and it is easy to be replaced by other stimulus or be regressed by itself. When people get into an environment where the temperature is relatively higher but still cold, the thermal sensation will improve, they feel moderate in an environment that is still cold with the short memory of relatively more extreme cold at the instant of change. And with the same memory they will not feel so hot in a hotter environment. Gradually, the thermal sensation will tend to become the real feelings in a steady state after a period of time, while the length of the time depends on the temperature. After this experiment, the thermal sensation of these subjects at some certain temperature will be the same as normal. The whole process is just like memorizing words when learning a new language, short-term stimulations create a memory, which can be strengthened by a large number of repetitive long-term stimulations, or else they are quickly forgotten. Maybe there is also a curve like the Ebbinghaus’ forgetting curve in the thermal sensation regions.
EP
TE D
4.2. Cold and hot are asymmetric in many cases The results presented in Figure 6 andFigure 11 reveal that the trends and values of the △PMV are exactly the same. However, the relationship of the △TSV and temperature difference are slightly different, though the scissors difference is always present. In the condition that started and ended at an air temperature of 20 oC, the cut-off point is about 7.5 oC from 20 oC to other warmer temperatures, and about 5 oC back from other temperatures to 20 oC. Moreover, in the condition that started and ended at an air temperature of 30 oC, the cut-off point is about 5.5 o C from 30 oC to other cooler temperatures, and about 3.7 oC back from other temperatures to 30 oC. Admittedly, the cut-off points from hot to cooler temperatures are similar in the two conditions, one is about 5 oC, the other is about 5.5 oC. However, the cut-off points are greatly different from cold to warmer, one is about 7.5 oC, the other is about 3.7 oC. Overall, the asymmetry of “cold” and “hot” really exists. Different short-term thermal experiences lead to the discrepancy of the cut-off points to some extent.
AC C
The human body is more sensitive to cold stimulation than to hot. In the last century, it was found that human have the perception of cold and hot, because there are some two types of thermoreceptors in the skin layer, namely cold sensors and hot sensors. When some hot and cold stimulations come from outside, the nervous impulses are produced by the cold or hot sensors, and send pulse signals to the brain, which in turn generates the corresponding stress reactions, such as sweating or shiver. Compared with hot sensors, cold sensors are closer to the skin surface, and have advantage in quantity. Accordingly, people is more sensitive to cold stimulation [28].Though the distribution of the thermoreceptors is the important reason why there are some differences between the cold and hot sensations, undoubtedly, many other factors also cause some influences. The principle behind these phenomenon is complex and subtle, and the mechanism needs to be further investigated. 4.3. The action time which influences the current thermal sensation and thermal comfort In the series of experiments presented here, it is obvious that the short-term thermal experience will influence the judgment of the current thermal sensation and thermal comfort. Apparently, the evaluations of thermal environment are based on not only the current feelings, but also the accumulations over a previous period of time. The specific length of time cannot
ACCEPTED MANUSCRIPT be judged hastily. So, should we reflect on the field studies of thermal comfort now? The vast majority of cases are not in a stable environment, and the evaluations that people filled in the questionnaires are not purely about the current environment, maybe mixed with the effects of the conditions before and after.
RI PT
In the present study, at least we recognized this phenomenon and found a way to improve the thermal sensation and thermal comfort associated with the current environment. Furthermore, it can also save energy of air conditioning in a way, because if people like the step-changed temperature and dynamic environment, it is not necessary to maintain the rooms always in a stable state with low temperature in the summer and high in the winter.
SC
4.4. Limitations and future challenges It is no doubt that there are some limitations in our experiments. A larger amount of subjects should be helpful to make the experiment results more solid. The current study concentrated mainly on the thermal evaluations in climate chamber experiments. In the future studies, some other methods, such as the physiological testing as well as simulation could be explored to further elucidate the short-term thermal experience.
TE D
M AN U
This study was a preliminary observation of adaptive thermal comfort within the time scale of minutes to hours. It summarized the variation of subjective feelings in different step-changed temperatures. In practice, people may always experience changed temperature within a short time period. For example, in a typical office building, people always transit between offices, corridors, meeting rooms, restrooms, etc., where temperatures may be all different. For this reason, knowing the laws of the short-term thermal experience can help us to set temperatures for different indoor spaces more accurately according to the real demands, so that to provide comfort while saving energy used in mechanical conditioning.
AC C
EP
5. CONCLUSIONS (1) Both comfort and discomfort come from dynamic contrast. If the poor thermal environment improves a little, people will feel significantly better. (2) Short-term thermal experience would make some excursion of real thermal sensation. In the condition that started and ended at an air temperature of 20 oC condition, people feel more comfortable even in the 22 oC environment, instead of the 25 oC condition which is close to the neutral temperature in the steady state. (3) The variation of sensation caused by cold stimulation is more intense than that caused by hot stimulation. People are more sensitive to cold stimulation, while the effect of temperature reduction is more obvious than temperature increment. (4) The endurance of people in a cold environment is stronger than in a hot environment. In this experiment, we found that people like a little cold instead of neutral, they seem to be more afraid of a hot environment. (5) The evaluations of thermal environment are based on not only the feelings right away, but also the feelings accumulated over a past period of time, short-term thermal experience can influence the evaluations to a great extent. 6. ACKNOWLEDGEMENT This study was funded by the 12th Five-year National Science and Technology Support Program of China (No.2013BAJ15B01). REFERENCES
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[1] W. Fisk, Health and productivity gains from better indoor environments and their relationship with building energy efficiency, Annual Review of Energy and Environment. 25 (2000) 537-566. [2] K.C. Parsons, Human Thermal Environments: The Effects of Hot, Moderate and Cold Environments on Human Health, Comfort and Performance, New York, Taylor and Francis, 2003. [3] G.R. McGregor, Human biometeorology, Progress in Physical Geography, 36(2011) 93109. [4] K. Chronopoulos, A. Kamoutsis, A. Matsoukis and E. Manoli, An artificial neural network model application for the estimation of thermal comfort conditions in mountainous regions, Greece, 25(2012) 171-181 [5] P.O. Fanger, Thermal Comfort: Analysis and Application in Environmental Engineering, Danish Technology Press, Copenhagen, 1970. [6] R.J. de Dear, G.S. Brager, The adaptive model of thermal comfort and energy conservation in the built environment, Int. J. Biometeorol. 45 (2001) 100–108. [7] S. Roaf, F. Nicol, M. Humphreys, P. Tuohy, A. Boerstra, Twentieth century standards for thermal comfort promoting high energy buildings, Archit. Sci. Rev. 53 (1) (2010) 65-77. [8] G. Brager, R.J. de Dear, Thermal adaptation in the built environment: a literature review, Energy Build. 27(1) (1998) 83–96. [9] F. Nicol, M. Humphreys, Adaptive thermal comfort and sustainable thermal standards for buildings, Energy Build. 34(6) (2002) 563-572. [10] E. Arens, M. Humphreys, R. de Dear, H. Zhang, Are ‘class A’ temperature requirements realistic or desirable? Build. Environ. 45 (2010) 4-10. [11] R. de Dear, T. Akimoto, E. Arens, G. Brager, C. Candido, et al., Progress in thermal comfort research over the last twenty years, Indoor Air. 23 (6) (2013) 442-461. [12] ASHRAE, Thermal Environmental Conditions for Human Occupancy, ASHRAE Standard 55, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, Georgia, 2013. [13] CEN, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings: Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, EN 15251, European Committee for Standardization, Brussels, 2007. [14] B. Li, R. Yao, Q. Wang, Y. Pan, An introduction to the Chinese Evaluation Standard for the indoor thermal environment, Energy Build. 82 (2014) 27-36. [15] M. Luo, B. Cao, J. Damiens, B. Lin, Q. Ouyang, et al., Evaluating thermal comfort in mixed-mode buildings: A field study in a subtropical climate, Build. Environ. 88 (2015) 46-54. [16] E. Halawa, J. van Hoof, The adaptive approach to thermal comfort: a critical review, Energy Build. 51 (2012) 101-110. [17] M. Schweiker, S. Brasche, W. Bischof, M. Hawighorst, K. Voss, et al., Development and validation of a methodology to challenge the adaptive comfort model, Build. Environ. 49 (2012) 336-347 [18] A. P. Gagge, J. A. J. Stolwijk and J. D. Hardy, Comfort and thermal sensations and associate physiological responses at various ambient temperatures, Environment Research, 1(1967)1-20. [19] A. van der Lans, J. Hoeks, B. Brans, G. Vijgen, M. Visser, et al., Cold acclimation recruits human brown fat and increases nonshivering thermogenesis, The Journal of Clinical Investigation. 123 (8) (2013) 3395-3403. [20] C. Chun, A. Kwok, T. Mitamura, N. Miwa, A. Tamura, Thermal diary: connecting temperature history to indoor comfort, Build. Environ. 43(2008) 877-885.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[21] Kelly Lisa, Parsons Ken. Thermal comfort when moving from one environment to another, In Adapting to change: new thinking on comfort Cumberland Lodge, Windsor, UK, 4(2010)9-11. [22] Nagano K, Takaki A, Hirakawa M, Tochihara Y. Effects of ambient temperature steps on thermal comfort requirements, International Journal of Biometeorology. 50(2005)33-39. [23] J. Yu, Q. Ouyang, Y.X. Zhu, H.G. Shen, G.G. Cao, et al., A comparison of the thermal adaptability of people accustomed to air-conditioned environments and naturally ventilated environments, Indoor Air. 22 (2012) 110-118. [24] H. Zhang., Human thermal sensation and comfort in transient and non-uniform thermal environments, 2004. [25] Du X. et al., The response of human thermal sensation and its prediction to temperature step change (Cool-Neutral-Cool), Plos One. 9(8) (2014). [26] H. Liu, et al. The response of human thermal sensation perception and skin temperature to step-change transient thermal environments, Build. Environ. 73(2014)232-238. [27] Thermal Design Code for Civil Buildings, GB 50176-1993, Ministry of Housing and Urban-Rural Development of the People’s Republic of China, Beijing, China, 1993 (in Chinese). [28] J. W. Ring, Richard de Dear, Temperature Transients: A Model for Heat Diffusion through the Skin, Thermoreceptor Response and Thermal Sensation, Indoor Air. 1(4) (1991)448-456.
S0360-1323(16)30513-3
DOI:
10.1016/j.buildenv.2016.12.021
Reference:
BAE 4751
To appear in:
Building and Environment
Received Date: 11 September 2016 Revised Date:
28 November 2016
Accepted Date: 12 December 2016
Please cite this article as: Ji W, Cao B, Luo M, Zhu Y, Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment, Building and Environment (2017), doi: 10.1016/j.buildenv.2016.12.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Influence of short-term thermal experience on thermal comfort evaluations: A climate chamber experiment Wenjie Ji a, b, Bin Cao a, b, * Maohui Luo a, b, Yingxin Zhu a, b Department of Building Science, Tsinghua University, Beijing 100084, China b Key Laboratory Eco Planning & Green Building, Ministry of Education, Tsinghua University, China a
Corresponding email: [email protected]
RI PT
*
Corresponding phone (+86) 010 62782746
EP
TE D
M AN U
SC
ABSTRACT The purpose of this study is to explore how a short-term thermal experience influences thermal comfort evaluation. Thermal experience, which refers to the previous thermal environment, may result in the formation of some “memory” on humans. When people enter another environment where the temperature is different from the previous one, the previous experience may result in some different feelings and changes on the evaluations of thermal comfort, comparing with staying in a steady state condition. In this paper, we mainly focus on short-term thermal experience within the time scale of minutes to hours. Climate chamber experiments were conducted for analysis and discussion. The experiment we designed had three sets of conditions: 1) started and ended at an air temperature of 20oC, and experienced higher temperatures in between; 2) started and ended at an air temperature of 25oC, and experienced higher or lower temperatures in between, and 3) started and ended at an air temperature of 30oC, and experienced lower temperatures in between. The evaluations of thermal comfort of the subjects at different temperature conditions were recorded by questionnaires. We found that both comfort and discomfort resulted from the contrast between the current and previous conditions. Even though the initially poor thermal environment was improved a little bit, the evaluation of the thermal comfort would be improved a lot. Additionally, the decrease of thermal sensation caused by cold stimulation was more obvious than the increase due to hot stimulation. People’s the evaluations could be considered as a combination of both the past and the present feelings.
AC C
KEYWORDS Thermal experience, chamber experiment, thermal sensation, step-changed temperature, adaptive thermal comfort 1.INTRODUCTION Indoor thermal environmental quality contributes to the thermal comfort perception, wellbeing and performance of the occupant [1]. Human thermal comfort is defined as “that condition of mind which expresses satisfaction with the thermal environment” [2][3]. Thermal comfort conditions can be evaluated by using a lot of indices based on environmental variables such as air temperature, humidity, wind speed as well as behavioral ones such as clothing and activities [4]. What kind of indoor climate should be created and how to implement it are closely connected with human health and the building energy consumption. Accordingly, it is essential to examine our requirements for indoor space conditioning, and to judge if it is necessary to maintain what are now regarded as comfortable environments in some current standards.
ACCEPTED MANUSCRIPT
RI PT
1.1.Current development of adaptive thermal comfort There are two classic theories to evaluate indoor thermal comfort. One of them is Fanger’s predicted mean vote (PMV) model, which is based on heat balance calculation of the human body [5]. The other theory is the adaptive thermal comfort model, which emphasizes that not only the physical environment influences thermal comfort, but also the human body itself which has the ability to adapt to the surroundings [6]. Ever since the concept of adaptive thermal comfort was announced in the 1960s [7], it has been continually improved and developed. In 1998, De Dear and Brager proposed the adaptive thermal comfort model where they claimed that people can interact with the environment in three ways: physiological acclimatization, behavioural adjustment, and psychological habituation [8]. Their model provided us with a new perspective to understand thermal comfort.
M AN U
SC
Some studies indicate that in air-conditioned buildings, the comfortable temperature range is getting narrower nowadays than before [9]. However, Arens found that occupants’ satisfaction did not show obvious differences in the indoor environments of different levels according to the current standards [10]. It was suggested that thermal comfort and satisfaction can hardly be improved by steady-state control even with high control precision, thus adaptive thermal comfort is expected to become the prevalent trend in the future [11]. As this theory became accepted and widely used by more and more people, it was also introduced into some standards, including ASHRAE Standard 55 [12], EN 15251 [13] and China GB/T 50785 [14], etc.
TE D
Nevertheless, some issues of the adaptive thermal comfort theory remain to be solved and explained in depth [15]. As Halawaand van Hoof stated [16], the underlying premises and the logic of adaptive processes still need rethinking. Additionally, the individual roles in the behavioral, physiological and psychological adaptive processes should be further identified [17].
AC C
EP
1.2. Statement of the problem The adaptive thermal comfort model is usually described as a “black box” model, as its mechanism is not clear to some extent. Thermal experience is an important part of adaptive thermal comfort. There are some current research studies focusing on thermal experience. Gagge [18], who was one of the earliest scholars to study on short-term thermal experience, found some laws of human body reactions in variable temperature environment. Van der Lans et al. [19] reported that after a 10-day cold acclimation, subjects perceived the environment with a warmer sensation, felt more comfortable in the cold, and reported less shivering. Chun et al. [20] conducted a comparative study using Korean and Japanese subjects, who had a 1day different thermal experiences. Their results showed that when entering a similar environment, the subjects who previously experienced higher temperature reported a 0.44 lower thermal sensation than those with cooler initial experience. Kelly Lisa and Ken Parsons [21] found that thermal sensation changed immediately along with the change of environmental thermal parameters, and put forward a transient PMV model. Nagano et al. [22] did experiments about step-changed temperature, and found that the neutral temperature changed. Yu et al. [23] showed that subjects who were used to naturally ventilated environments had a better physiological thermoregulation when exposed to high temperatures. Addiotional studies also discussed the effects of short-term hot or cold stimulus on the human body and found that people have the capacity of self-adaptation and adjustment [24].
ACCEPTED MANUSCRIPT Short-term thermal experience refers to hot or cold stimulus within the time scale of minutes to hours. To date, there are hardly any consensus about how the short-term thermal experiences influence thermal comfort and thermal adaptation.
RI PT
1.3. Objective This study focuses on the impacts of short-term thermal experience on thermal comfort evaluations, considering different directions and ranges of temperature change. We tried to find out whether the thermal comfort evaluations simultaneously resulted from the current surrounding environments, as well as the previous thermal experience, and found some quantitative relationship between temperature change and thermal evaluations.
AC C
EP
TE D
M AN U
SC
2. METHODS 2.1. Climate chamber The experiment was conducted in an artificial climate chamber (5m×3m×2.7m), where the temperature and humidity can be accurately controlled. There are two rooms in the chamber. The environments in the two rooms can be controlled separately by an automatic control system, which means the temperature and humidity can be different in the two rooms. There is a door between the two rooms. When needed, subjects can enter one room from the other using this door. In this experiment, we mainly observed how the temperature changes influenced the human body comfort and thermal sensation. The air humidity was controlled to be in the range of 45~55% in both rooms. When the air conditioning system is on, the average air velocity in the chamber is 0.05m/s, which could be considered to have no impact on human thermal comfort. The automatic control interface of the climate chamber is shown in Figure 1.
Figure 1. The automatic control interface of the artificial climate chamber
2.2. Experimental protocol The experiment was carried out in May of 2015 in Beijing, which is located in the Cold Zone according to China’s national standard [27] . The whole climate chamber was in a separate room, as a result, the indoor thermal environment in the chamber couldn’t be influenced by the outdoor temperature. In order to verify if the previous short-term thermal experience has an impact on the thermal evaluation of the current environment by the occupants, the experimental conditions in this study were set as shown in Figure 2. In total, there were three groups of conditions, including: 1) started and ended at an air temperature of 20 oC, and
ACCEPTED MANUSCRIPT experienced higher temperatures of 22, 25 and 30 oC in between; 2) started and ended at an air temperature of 25 oC, and experienced higher temperature (30 oC) or lower temperatures (20 o C) in between; 3) started and ended at an air temperature of 30oC, and experienced lower temperatures of 20, 25 and 28 oC in between. There were 8 experimental conditions in all.
22
20
25 30
25
20
25
30oC started and ended at an air temperature of 30℃ 20 20
30
25oC
22oC
28
30
30
25
28o C
RI PT
started and ended at an air temperature of 20℃
started and ended at an air temperature of 25℃
20oC
SC
Figure 2. Three experimental conditions
M AN U
We recruited 8 subjects for the experiments. Each subject participated in all conditions. Totally, there were 64 samples. The subjects were college students in good health conditions. They were not allowed to have intense physical exercise before the experiments. During the experiments, all subjects wore the same standard clothes (short T-shirt, thin trousers and sneakers), with a thermal insulation of 0.57 clo, and their metabolic rate was 1.1 met. The subjects were seated and did some light work, such as reading and typing. The subjects had to fill in a questionnaire several times during each experimental condition. There were mainly three questions in the questionnaire, which were Thermal Sensation Vote (TSV), Thermal Comfort Vote (TCV), and thermal expectation. The scale for each question is shown in Table 1.
TE D
Table 1. the scales of three questions TSV, TCV and Thermal expectation Scale TSV TCV Thermal expectation Hot
Very comfortable
--
2
Slightly hot
Comfortable
--
1
Warm
Just comfortable
Warmer
EP
3
Neutral
Neutral
Unchanged
-1
Cool
Just uncomfortable
Cooler
-2
Slightly Cold
Uncomfortable
--
Very uncomfortable
--
AC C
0
-3
Cold
There are three phases in each experimental condition. Taking the first group, which started and ended at an air temperature of 20 oC, as an example: a) the adaptive phase (1 to 20 min). Subjects stayed in a 20 oC environment for 20 minutes, and filled in the questionnaires 3 times (10 min, 5 min, 5 min); b) the temperature change phase (21 to 60 min). Subjects entered a 22 (25 and 30) oC environment and stayed for 40 minutes, in the meantime they filled in the questionnaires 10 times (1min, 1min, 1min, 1min, 1min, 5 min, 5 min, 5 min, 10 min 10 min); c) the rebound phase (61 to 90 min). Subjects went back to the 20 oC environment and stayed for 30 minutes, filled in the questionnaires 8 times (1min, 1min, 1min, 1min, 1min, 5 min, 10 min, 10 min). In phases b) and c), the questionnaires were conducted more frequently when the subjects just entered that phase. The three phases and the times when subjects filling in the questionnaires are shown in Figure 3.
ACCEPTED MANUSCRIPT
Figure 3. The time length of the three phases in each experimental condition (The red marks are the moment when subjects need to fill in the questionnaires)
RI PT
3. RESULTS 3.1. Condition that started and ended at an air temperature of 20 oC The results from the condition where air temperature started and ended at 20 oC are shown in Figures 4-8. All the results shown in these figures are expressed as the mean values of TSV, TCV or thermal expactation of all the 8 subjects.
TE D
M AN U
SC
The solid lines in Figure 4 correspond to the TSV from 20 oC to the other temperatures then back to 20 oC , and the dotted lines correspond to the PMV which is calculated using the known parameters. From a) the adaptive phase to b) the temperature variation phase: The thermal sensations of the subjects who went into a higher temperature from 20 oC, were all improved significantly, even though they went into 22 oC, which was still a somewhat cold environment, they felt slightly warm. And with the passage of time, TSV in the 22, 25 and 30 o C temperatures were all slowly decreased. A comparison of the TSV and PMV, revealed that the bigger the temperature changed, the smaller the difference between the two. And in the case of 22 and 25 oC, with temperature differences of 2 and 5 degrees, TSV is higher than PMV, but TSV is lower than PMV in the 30 oC condition. Traditionally, we thought that the current thermal sensation would be amplified with the contrast of thermal experience before and after. However, it seems to be different in the 20 oC to 30 oC experimental conditions, with a temperature difference of 10 oC.
AC C
EP
Additionally, from b) the temperature variation phase to c) the rebound phase: The thermal sensation of the subjects who went back to 20 oC from higher temperature decreased quickly, but the TSV were all higher than the PMV in the 20 oC condition, the TSV in the 30 oC condition is the highest in the rebound phase. It seems that when people come from a hot environment to a cold environment, they may not feel very cold.
ACCEPTED MANUSCRIPT 3
a） ）
c） ）
b） ）
TSV22 TSV25 TSV30 PMV22 PMV25 PMV30
2
0 0
10
20
30
40
50
-1
70
80
90
100
SC
-2
60
RI PT
TSV
1
-3
M AN U
time(min) Figure 4. TSV and PMV of the condition that started and ended at an air temperature of 20 oC
1.5
EP
TE D
In addition, the results presented in Figure 5 show how the deviation of the TSV and the PMV [△(TSV-PMV)] further changed in the process. At the moment of going from 20 oC to other environment, the TSV is higher than the PMV in the 22 oC and 25 oC conditions. Additionally, △(TSV-PMV) is bigger at first, and then slowly goes down; in the 30 oC condition, the TSV is almost equal to PMV at the beginning of the change, then the deviation gradually grows, and it is opposite to 22 and 25 oC, whereas the TSV is lower than PMV throughout the process. In the dynamic process, the difference is obvious between the predicted value of PMV and the real value of TSV. Accordingly, using the PMV model to estimate thermal sensation is not accurate, as it will be lower than the actual sensation in a moderately cold environment and higher than the actual in a moderately hot environment. That is the reason why the deviations in this experiment have both positive and negative values, mainly due to the contrast between the before and after environments, namely the effects of short-term thermal experience.
22
25
30
(TSV-PMV)
AC C
1
0.5
0
10 15 20 21 22 23 24 25 30 35 40 50 60 61 62 63 64 65 70 80 90
-0.5
-1
time to vote
Figure 5. The deviation of the TSV and PMV in the whole process of the condition that started and ended at an air temperature of 20 oC
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
The disparity between the TSV and PMV is obvious in different temperature conditions. And, at the moment of going into another environment, the TSV changes quickly. Evidently, at the same time, we have a new PMV value with new parameters, so we have the variation of TSV (△TSV) and the variation of PMV (△PMV) at the moment the environment changes, using the average value of the first 5 min. The relationship between the △TSV and △PMV with the temperature difference in an instant is shown in Figure 6. The two dotted lines are the △PMVs, namely the △PMV from 20 oC to other temperatures and the △PMV back to 20 oC, after the stimuli are exactly symmetrical, they have a linear relationship with the temperature difference, with a slope of ± 0.3266. Moreover, the solid lines are △TSVs, apparently there is a scissors difference between the △TSV and the △PMV. The trends of the △TSV from 20 oC to other temperatures and the △TSV back to 20 oC after stimuli are similar. As we can see, when the temperature difference is small, the variation of TSV is more obvious than the variation of PMV, both in two conditions. However, when the temperature difference is large, the △TSV is less obvious than △PMV. It is noteworthy that the cut-off points of the scissors difference are different. From 20 oC to other conditions, namely from cold to warm, the cutpoint is about 7.5 oC, while for the other one having a warmer or hot stimulus, then going back to 20 oC is about 5 oC. This means that from 20 oC to other hot environments, the variation of the TSV is bigger than that of the PMV when the temperature difference is less than 7.5 oC, and the actual feeling will be hotter. However, when the temperature difference is greater than 7.5 oC, the actual feeling will not be that hot. On the other hand, after a hot stimulus and back to 20 oC, the actual thermal sensation will be colder when the temperature difference is less than 5 oC, and it will not be so cold when the temperature difference is greater than 5 oC. 4
y = 0.3266x - 0.0007
1 0 0 -1 -2 -3 -4
2
AC C
△TSV, △PMV
2
EP
TE D
3
4
6
y = 0.1254x + 1.5199 TSV 20℃ to others
TSV others back to 20℃ 8
10
12 PMV 20℃ to others
y = -0.1185x - 1.0184 PMV others back to 20℃ y = -0.3266x + 0.0007
temperature difference (℃)
Figure 6. The relationship between △TSV and △PMV with the change of the temperature in an instant (the first 5 minutes) in the condition that started and ended at an air temperature of 20 oC
In addition, the TCV and thermal expectation in the whole process were also recorded. In Figure 7, from a) the adaptive phase to b) the temperature variation phase, the TCVs were all increased from 20 oC to other temperatures, and going into 22 oC was the most comfortable. Additionally, from b) the temperature variation phase to c) the rebound phase, at the moment of going back to 20 oC, the TCV were all decreased, even lower than those in a) the adaptive phase at 22 oC and 25 oC. It is interesting that from 30 oC to 20 oC, the uncomfortable feeling
ACCEPTED MANUSCRIPT was not so obvious under a temperature difference of 10 oC, which coincided with the analysis of the TSV before. 3
c） ）
b） ）
a） ）
22
2
25 30
0 0
10
20
30
40
50
60
-2
-3
80
90
100
M AN U
time(min)
70
SC
-1
RI PT
TCV
1
Figure 7. The TCV of the condition that started and ended at an air temperature of 20 oC
TE D
Similarly, the thermal expectation shown in Figure 8, from a) the adaptive phase to b) the temperature variation phase, the expectation of warmness reduced, and to c) the rebound phase, they felt cold and hoped for warmer temperature, the expectation was more intense than a) the adaptive phase, and in the 30 oC condition, the expectation of coolness was not so apparent as in the 22 and 25 oC conditions as well. All the variation of the feelings was significant in the instant of changing the environment, and after adapting to the new temperature, they would be the steady state.
AC C
thermal expectation
0.5
0
0
-0.5
-1
22
EP
1
10
a） ）
20
30
25 30
40
50
60
70
80
90
100
c） ）
b） ） time(min)
Figure 8. The thermal expectation of the condition that started and ended at an air temperature of 20 oC
The trend from 30 oC to 20 oC was intriguing with a temperature difference of 10 oC. Maybe having the short-term thermal experience of relative coldness will dull the perception of relative hot. Alternatively, people may have formed some impression in their brain, like
ACCEPTED MANUSCRIPT memories of short-term history, when going into another environment where the temperature contrast is bigger, the thermal inertia will lead them to an erroneous judgement of the current environmental temperature.
SC
RI PT
3.2. Condition that started and ended at an air temperature of 30 oC The TSV and PMV of the whole process of the condition that started and ended at an air temperature of 30 oC are shown in Figure 9. From a) the adaptive phase to b) the temperature variation phase: at the moment of going into a lower temperature from 30 oC, the thermal sensations were all reduced, and in the 28 oC and 25 oC conditions, the TSVs were lower than PMVs, which was more obvious in the 28 oC condition, but at 20 oC the TSV was higher than the PMV. The deviation of the TSV and PMV was not as the condition that started and ended at an air temperature of 20 oC, along with the temperature changing. With the passage of time, the TSV in the 28, 25 and 20 oC were all slowly increased to the steady state temperature. Then from b) the temperature variation phase to c) the rebound phase: The subjects went from 28 oC back to 30 oC, the TSV was the highest, but from 20 oC to 30 oC, the TSV was lower than the PMV, which implied that from cold to hot environment people did not feel very hot.
a） ）
b） ）
2
0
-1
-2
-3
10
20
30
40
TE D
TSV
1
0
M AN U
3
50
60
70
c） ）
80
90
100
20 25 28 PMV20 PMV25 PMV30
EP
time(min) Figure 9. The TSV and PMV of the condition that started and ended at an air temperature of 30 oC
AC C
The change of the deviation of the TSV and PMV [△(TSV-PMV)] is shown in Figure 10. From 30 oC to other environments, the TSV is higher than the PMV in the 20 oC condition, and it is lower than the PMV in the 25 oC and 28 oC conditions, and there is a peak of the △(TSV-PMV), and then gradually starts to go down.
ACCEPTED MANUSCRIPT 1.5 20
25
28
0.5
RI PT
△(TSV-PMV)
1
0
10 15 20 21 22 23 24 25 30 35 40 50 60 61 62 63 64 65 70 80 90
-1
time to vote
SC
-0.5
M AN U
Figure 10. The deviation of TSV and PMV during the whole process of the condition that started and ended at an air temperature of 30 oC
TE D
It is identical to the condition started and ended at an air temperature of 20 oC, the relationship of △TSV and △PMV with the temperature difference in an instant is shown in Figure 11. The linear relationship of △PMV and the temperature difference are the same, with a slope of ± 0.3266. However, the cut-off points of the scissors difference are different in this condition. From 30 oC to other temperatures, specifically from hot to cool, the cut-off point is about 5.5 o C, while another is about 3.7 oC. This means that from 30 oC to the other temperatures, the variation of the TSV is larger than that of the PMV when the temperature difference is less than 5.5 oC, whereas if it is over 5.5 oC, the actual feeling will not be so cold. Meanwhile, in another condition involving having a cooler or cold stimulus then going back to 30 oC, the actual thermal sensation will be warmer when the temperature difference is less than 3.7 oC, and it will not feel so hot when it is greater than 3.7 oC.
EP
4
△ TSV, △ PMV
2 1 0
AC C
3
0
-1 -2
2
y = 0.3266x - 0.0009
TSV 30℃-others
y = 0.1696x + 0.5804
TSV others- 30℃ 4
6
8
10
12 PMV 30℃-others
y = -0.2091x - 0.6469 PMV others- 30℃
-3 y = -0.3266x + 0.0009 -4
temperature difference (℃ ℃)
ACCEPTED MANUSCRIPT Figure 11. The relationship between the △TSV and △PMV with the change of the temperature in an instant (the first 5 minutes) in the condition that started and ended at an air temperature of 30 oC
RI PT
The TCV and thermal expectation in the whole process were recorded as shown in Figure 12. From the adaptive phase to b) the temperature variation phase, the TCV were all increased from the 30 oC condition to other temperatures, and going into the 25 oC environment was the most comfortable. Additionally, from b) the temperature variation phase to c) the rebound phase, when going into the 20 oC environment, the uncomfortable feeling was not so obvious. Actually, the bigger the temperature difference, the less uncomfortable they felt, and the trend of thermal expectation is the same, as shown in Figure 13. 3
a） ）
c） ）
b） ）
2
25 28
SC
0 0
10
20
30
-1
-2
-3
M AN U
1
TCV
20
40
50
60
70
80
90
100
TE D
time(min)
Figure 12. The TCV of the condition that started and ended at an air temperature of 30 oC
1
a） ）
0
-0.5
-1
20 25
EP AC C
thermal expectation
0.5
0
c） ）
b） ）
10
20
30
28
40
50
60
70
80
90
100
time(min)
Figure 13. The thermal expectation of the condition that started and ended at an air temperature of 30oC
3.3. Condition that started and ended at an air temperature of 25 oC In this condition, when going into the 30 oC environment, the TSV was higher than the PMV, the sense of cold before and hot after under the temperature difference of 5 oC, which may
ACCEPTED MANUSCRIPT
RI PT
amplify the thermal sensation, in accordance with the result in the condition started and ended at an air temperature of 20 oC, indicating that from cold to hot environments, the variation of the TSV is larger than that of the PMV when the temperature difference is less than 7.5 oC, and the actual feeling will be hotter. In contrast, from 25 oC to 20 oC, the PMV was lower than the TSV, people did not feel so cold in 20 oC. Both, at the same temperature difference of 5 oC, there is a disparity in the thermal sensation from “hot to cold” and “cold to hot”. The TCV, thermal expectation and some other analyses were similar to the other two conditions above. 3
a） ）
c） ）
b） ）
2
0
10
20
30
40
50
SC
0
60
70
M AN U
TSV
1
-1
-2
-3
80
90
100
20 30 PMV20 PMV30
time(min) Figure 14. TSV and PMV of the condition that started and ended at an air temperature of 25 oC
AC C
EP
TE D
3.4. Contrast of the three conditions To further compare the three conditions, taking the average of the TSV at a certain temperature, also considering the PMV, and the differences between the evaluations were examined. The real thermal sensation under the three conditions and the predicted thermal sensation at different temperatures are shown in Figure 15. In the condition started and ended at an air temperature of 20 oC, with the short-term thermal experience of cold, specifically 20 o C, as long as the environmental temperature increased a little, the termal sensation would improve, even at the 30 oC condition it was not so hot. In the condition started and ended at an air temperature of 30 oC, the thermal sensation would improve as long as the environmental temperature was lower, and from 30 oC to 20 oC, people would not feel so cold. Futhermore, though the TSV, in the condition started and ended at an air temperature of 25 o C, was still exhibiting some differentces with the PMV, the overall trend was closest to the PMV, while the cold or hot thermal experience both influence the thermal sensative vote. This implies that the real thermal sensation was usually unequal to PMV, and different short-term thermal experience may lead to different deviation.
ACCEPTED MANUSCRIPT 2 1.5
0.5 0 20
22
25
28
30
RI PT
cold→hot
1
-0.5
started and ended at 20℃
-1
started and ended at 25℃
-1.5
started and ended at 30℃ PMV
temperature(℃)
SC
-2
Figure 15. The TSV of three conditions and the PMV at different temperatures
30℃ ℃
EP
TE D
M AN U
4. DISCUSSIONS 4.1. The variation of TSV along with the temperature change In conclusion, there were four types of temperature change in this experiment (25 oC is considered to be the neutral temperature): a. Cold to neutral; b. Cold to hot; c. Hot to cold; d. Hot to neutral, as shown in Figure 16. In the experiment of the short-term thermal experience, we found that the variations of the TSV and TCV are different in various situations, and drew a conclusion that when the environment changed from cold/hot to neutral condition, or in other words from uncomfortable to comfortable, no change in the nature of thermal sensation was reported, the TSV and TCV would be enlarged. However, when temperature changed from one extreme to another with large temperature difference, like from a cold 20 oC to a hot 30 oC condition and vice versa, they would not feel very hot/cold, and the uncomfortable feeling would not be very strong. In addition, taking the condition that started and ended at an air temperature of 20 oC as an example, we found that compared with going to 25 oC, going from 20 oC to 22 oC, both the TSV and TCV had the largest promotion and improvement, this suggested that moderate stimulation would be the most effective way to improve thermal sensation and comfort. 30℃ ℃
30℃ ℃
28℃ ℃
28℃ ℃
28℃ ℃
28℃ ℃
25℃ ℃
25℃ ℃
25℃ ℃
25℃ ℃
22℃ ℃
22℃ ℃
22℃ ℃
22℃ ℃
20℃ ℃
20℃ ℃
20℃ ℃
20℃ ℃
a. Cold to neutral
b. Cold to hot
c. Hot to cold
d. Hot to neutral
AC C
30℃ ℃
Figure 16. Four types of temperature change
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
There is a conjecture that thermal experience can establish a memory, or named habit. It is maybe some nerve impulses in the brain cortex, maybe some psychological adaption, maybe some sensorial characteristic. Whatever happens, something like a memory can influence the sensation of a new environment, from an involuntary comparison with past experience. Longterm history is similar to long-term memory, while short-term history is similar to short-term memory. Some instructed information can be retained by sense organs for a short time period at first, and it is easy to be replaced by other stimulus or be regressed by itself. When people get into an environment where the temperature is relatively higher but still cold, the thermal sensation will improve, they feel moderate in an environment that is still cold with the short memory of relatively more extreme cold at the instant of change. And with the same memory they will not feel so hot in a hotter environment. Gradually, the thermal sensation will tend to become the real feelings in a steady state after a period of time, while the length of the time depends on the temperature. After this experiment, the thermal sensation of these subjects at some certain temperature will be the same as normal. The whole process is just like memorizing words when learning a new language, short-term stimulations create a memory, which can be strengthened by a large number of repetitive long-term stimulations, or else they are quickly forgotten. Maybe there is also a curve like the Ebbinghaus’ forgetting curve in the thermal sensation regions.
EP
TE D
4.2. Cold and hot are asymmetric in many cases The results presented in Figure 6 andFigure 11 reveal that the trends and values of the △PMV are exactly the same. However, the relationship of the △TSV and temperature difference are slightly different, though the scissors difference is always present. In the condition that started and ended at an air temperature of 20 oC, the cut-off point is about 7.5 oC from 20 oC to other warmer temperatures, and about 5 oC back from other temperatures to 20 oC. Moreover, in the condition that started and ended at an air temperature of 30 oC, the cut-off point is about 5.5 o C from 30 oC to other cooler temperatures, and about 3.7 oC back from other temperatures to 30 oC. Admittedly, the cut-off points from hot to cooler temperatures are similar in the two conditions, one is about 5 oC, the other is about 5.5 oC. However, the cut-off points are greatly different from cold to warmer, one is about 7.5 oC, the other is about 3.7 oC. Overall, the asymmetry of “cold” and “hot” really exists. Different short-term thermal experiences lead to the discrepancy of the cut-off points to some extent.
AC C
The human body is more sensitive to cold stimulation than to hot. In the last century, it was found that human have the perception of cold and hot, because there are some two types of thermoreceptors in the skin layer, namely cold sensors and hot sensors. When some hot and cold stimulations come from outside, the nervous impulses are produced by the cold or hot sensors, and send pulse signals to the brain, which in turn generates the corresponding stress reactions, such as sweating or shiver. Compared with hot sensors, cold sensors are closer to the skin surface, and have advantage in quantity. Accordingly, people is more sensitive to cold stimulation [28].Though the distribution of the thermoreceptors is the important reason why there are some differences between the cold and hot sensations, undoubtedly, many other factors also cause some influences. The principle behind these phenomenon is complex and subtle, and the mechanism needs to be further investigated. 4.3. The action time which influences the current thermal sensation and thermal comfort In the series of experiments presented here, it is obvious that the short-term thermal experience will influence the judgment of the current thermal sensation and thermal comfort. Apparently, the evaluations of thermal environment are based on not only the current feelings, but also the accumulations over a previous period of time. The specific length of time cannot
ACCEPTED MANUSCRIPT be judged hastily. So, should we reflect on the field studies of thermal comfort now? The vast majority of cases are not in a stable environment, and the evaluations that people filled in the questionnaires are not purely about the current environment, maybe mixed with the effects of the conditions before and after.
RI PT
In the present study, at least we recognized this phenomenon and found a way to improve the thermal sensation and thermal comfort associated with the current environment. Furthermore, it can also save energy of air conditioning in a way, because if people like the step-changed temperature and dynamic environment, it is not necessary to maintain the rooms always in a stable state with low temperature in the summer and high in the winter.
SC
4.4. Limitations and future challenges It is no doubt that there are some limitations in our experiments. A larger amount of subjects should be helpful to make the experiment results more solid. The current study concentrated mainly on the thermal evaluations in climate chamber experiments. In the future studies, some other methods, such as the physiological testing as well as simulation could be explored to further elucidate the short-term thermal experience.
TE D
M AN U
This study was a preliminary observation of adaptive thermal comfort within the time scale of minutes to hours. It summarized the variation of subjective feelings in different step-changed temperatures. In practice, people may always experience changed temperature within a short time period. For example, in a typical office building, people always transit between offices, corridors, meeting rooms, restrooms, etc., where temperatures may be all different. For this reason, knowing the laws of the short-term thermal experience can help us to set temperatures for different indoor spaces more accurately according to the real demands, so that to provide comfort while saving energy used in mechanical conditioning.
AC C
EP
5. CONCLUSIONS (1) Both comfort and discomfort come from dynamic contrast. If the poor thermal environment improves a little, people will feel significantly better. (2) Short-term thermal experience would make some excursion of real thermal sensation. In the condition that started and ended at an air temperature of 20 oC condition, people feel more comfortable even in the 22 oC environment, instead of the 25 oC condition which is close to the neutral temperature in the steady state. (3) The variation of sensation caused by cold stimulation is more intense than that caused by hot stimulation. People are more sensitive to cold stimulation, while the effect of temperature reduction is more obvious than temperature increment. (4) The endurance of people in a cold environment is stronger than in a hot environment. In this experiment, we found that people like a little cold instead of neutral, they seem to be more afraid of a hot environment. (5) The evaluations of thermal environment are based on not only the feelings right away, but also the feelings accumulated over a past period of time, short-term thermal experience can influence the evaluations to a great extent. 6. ACKNOWLEDGEMENT This study was funded by the 12th Five-year National Science and Technology Support Program of China (No.2013BAJ15B01). REFERENCES
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[1] W. Fisk, Health and productivity gains from better indoor environments and their relationship with building energy efficiency, Annual Review of Energy and Environment. 25 (2000) 537-566. [2] K.C. Parsons, Human Thermal Environments: The Effects of Hot, Moderate and Cold Environments on Human Health, Comfort and Performance, New York, Taylor and Francis, 2003. [3] G.R. McGregor, Human biometeorology, Progress in Physical Geography, 36(2011) 93109. [4] K. Chronopoulos, A. Kamoutsis, A. Matsoukis and E. Manoli, An artificial neural network model application for the estimation of thermal comfort conditions in mountainous regions, Greece, 25(2012) 171-181 [5] P.O. Fanger, Thermal Comfort: Analysis and Application in Environmental Engineering, Danish Technology Press, Copenhagen, 1970. [6] R.J. de Dear, G.S. Brager, The adaptive model of thermal comfort and energy conservation in the built environment, Int. J. Biometeorol. 45 (2001) 100–108. [7] S. Roaf, F. Nicol, M. Humphreys, P. Tuohy, A. Boerstra, Twentieth century standards for thermal comfort promoting high energy buildings, Archit. Sci. Rev. 53 (1) (2010) 65-77. [8] G. Brager, R.J. de Dear, Thermal adaptation in the built environment: a literature review, Energy Build. 27(1) (1998) 83–96. [9] F. Nicol, M. Humphreys, Adaptive thermal comfort and sustainable thermal standards for buildings, Energy Build. 34(6) (2002) 563-572. [10] E. Arens, M. Humphreys, R. de Dear, H. Zhang, Are ‘class A’ temperature requirements realistic or desirable? Build. Environ. 45 (2010) 4-10. [11] R. de Dear, T. Akimoto, E. Arens, G. Brager, C. Candido, et al., Progress in thermal comfort research over the last twenty years, Indoor Air. 23 (6) (2013) 442-461. [12] ASHRAE, Thermal Environmental Conditions for Human Occupancy, ASHRAE Standard 55, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, Georgia, 2013. [13] CEN, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings: Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, EN 15251, European Committee for Standardization, Brussels, 2007. [14] B. Li, R. Yao, Q. Wang, Y. Pan, An introduction to the Chinese Evaluation Standard for the indoor thermal environment, Energy Build. 82 (2014) 27-36. [15] M. Luo, B. Cao, J. Damiens, B. Lin, Q. Ouyang, et al., Evaluating thermal comfort in mixed-mode buildings: A field study in a subtropical climate, Build. Environ. 88 (2015) 46-54. [16] E. Halawa, J. van Hoof, The adaptive approach to thermal comfort: a critical review, Energy Build. 51 (2012) 101-110. [17] M. Schweiker, S. Brasche, W. Bischof, M. Hawighorst, K. Voss, et al., Development and validation of a methodology to challenge the adaptive comfort model, Build. Environ. 49 (2012) 336-347 [18] A. P. Gagge, J. A. J. Stolwijk and J. D. Hardy, Comfort and thermal sensations and associate physiological responses at various ambient temperatures, Environment Research, 1(1967)1-20. [19] A. van der Lans, J. Hoeks, B. Brans, G. Vijgen, M. Visser, et al., Cold acclimation recruits human brown fat and increases nonshivering thermogenesis, The Journal of Clinical Investigation. 123 (8) (2013) 3395-3403. [20] C. Chun, A. Kwok, T. Mitamura, N. Miwa, A. Tamura, Thermal diary: connecting temperature history to indoor comfort, Build. Environ. 43(2008) 877-885.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[21] Kelly Lisa, Parsons Ken. Thermal comfort when moving from one environment to another, In Adapting to change: new thinking on comfort Cumberland Lodge, Windsor, UK, 4(2010)9-11. [22] Nagano K, Takaki A, Hirakawa M, Tochihara Y. Effects of ambient temperature steps on thermal comfort requirements, International Journal of Biometeorology. 50(2005)33-39. [23] J. Yu, Q. Ouyang, Y.X. Zhu, H.G. Shen, G.G. Cao, et al., A comparison of the thermal adaptability of people accustomed to air-conditioned environments and naturally ventilated environments, Indoor Air. 22 (2012) 110-118. [24] H. Zhang., Human thermal sensation and comfort in transient and non-uniform thermal environments, 2004. [25] Du X. et al., The response of human thermal sensation and its prediction to temperature step change (Cool-Neutral-Cool), Plos One. 9(8) (2014). [26] H. Liu, et al. The response of human thermal sensation perception and skin temperature to step-change transient thermal environments, Build. Environ. 73(2014)232-238. [27] Thermal Design Code for Civil Buildings, GB 50176-1993, Ministry of Housing and Urban-Rural Development of the People’s Republic of China, Beijing, China, 1993 (in Chinese). [28] J. W. Ring, Richard de Dear, Temperature Transients: A Model for Heat Diffusion through the Skin, Thermoreceptor Response and Thermal Sensation, Indoor Air. 1(4) (1991)448-456.