- Email: [email protected]
Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas
Accepted Manuscript Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas Zhongjun Zhang, Yufeng Zhang, Ling Jin PII:
S0360-1323(17)30485-7
DOI:
10.1016/j.buildenv.2017.10.028
Reference:
BAE 5141
To appear in:
Building and Environment
Received Date: 28 July 2017 Revised Date:
18 October 2017
Accepted Date: 21 October 2017
Please cite this article as: Zhang Z, Zhang Y, Jin L, Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas, Building and Environment (2017), doi: 10.1016/ j.buildenv.2017.10.028. 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
Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas Zhongjun Zhang a, b, Yufeng Zhang a, *, Ling Jin c a
RI PT
State Key Laboratory of Subtropical Building Science, Department of Architecture, South China University of Technology, Wushan, Guangzhou, 510640, China b School of Civil and Architectural Engineering, Anyang Institute of Technology, Anyang, 455000, China c Department of Water Conservancy and Civil Engineering, South China Agriculture University, Wushan, Guangzhou, 510640, China * Corresponding author: E-mail address: [email protected](Y.Zhang)
AC C
EP
TE D
M AN U
SC
ABSTRACT Semi-open spaces are a key element in the climate adaptive design of rural folk houses and are important places where rural residents perform daily activities in hot-humid areas. In this study, a year-long thermal comfort field survey was conducted in rural folk houses in 11 rural areas in the Guangdong Province, the hot-humid area of southern China. The subjective responses of residents were recorded via questionnaires, and all ambient environment parameters were measured. The results show that air speed was greater and relative humidity was lower in the semi-open spaces. The clothing insulation of the survey respondents varied with indoor operative temperatures from 0.27 clo to 1.2 clo during the non-summer season and remained steady at 0.30 clo during the summer season. The thermal neutral, acceptable and preferred indoor operative temperatures were determined for the interior and semi-open spaces in the summer and non-summer. Compared to those of the interior spaces, the thermal neutral temperature was 0.6 - 1.3 °C lower and the upper limit of 80% acceptable temperature was decreased by 0.8 - 4.7 °C in the semi-open spaces. The acceptable temperature in the summer for the rural residents was found to be 0.2 - 1 °C higher than that of the urban residents in the same climate. Psychological factors such as local culture, expectations and perceived control might be the reasons for these differences. This study contributes to a better understanding of the thermal comfort of rural people and to the improvement of rural living conditions. Keywords: Rural folk house, Thermal comfort, Hot-humid area, Semi-open space, Field study 1 Introduction
Both climatic conditions and economic level have important effects on building’s performance and energy consumption. In the economically developed areas, HVAC systems are widely used for maintaining comfortable indoor environment in whatever outdoor climate, and the buildings consume lots of energy and discharged amounts of pollutions. Only some vernacular buildings, where HVAC devices and systems are used very less, retain well the architectural characteristic of climatic adaptability and energy efficient, in which man, building and environment develop harmoniously and sustainably.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
The global climate is so diverse and the vernacular buildings vary widely in their functions, forms and materials. In cold climates, the design strategy of vernacular buildings is to increase heat gains and reduce heat loss, while in hot climates the opposite strategy is used. For example, in the northern of Portugal, as the winter is very cold, glass balcony that obtains more solar radiation is the main feature of the traditional architectures; while in the south, small size windows and doors and patios (courtyards) are the main features for minimizing heat gains in summer [1]. In Turkey, in order to adapt to dry-hot climate and reduce solar radiation, the traditional buildings often enclose together, forming a small courtyard that provides good shading and comfortable microclimate. In addition, thermal mass envelopes are often used to cope with the larger diurnal range of outdoor temperature [2, 3]. Similarly in the Eastern Mediterranean region, thick walls are used, open and semi-open spaces are the common architectural forms, and night ventilation cooling is the main passive cooling strategy in some vernacular buildings [4]. In Malaysia it is always high temperature, high humidity and abundant rainfall as it is in the equatorial hot areas. In order to create a good indoor environment, the vernacular houses in Malaysia often raise on the posts above the ground, open many spaces or windows in the wall and have a high ceiling to provide good ventilation and improve comfort [5]. The passive design strategies greatly improve the environmental quality of the vernacular buildings, but with no guarantee for satisfying residents’ thermal comfort requirements. For example, Foruzanmeh conducted a field study on the vernacular dwellings in central Iran and reported that the dwellings could not provide comfortable temperature throughout an entire typical hot summer day [6]. Similar results were found for the modern buildings using the passive strategies of vernacular buildings. Toe et al. investigated the modern Malaysia buildings that used the traditional passive cooling technology. They found that using passive design reduced thermal discomfort, but could not provide thermal comfort for a whole day, and the indoor operative temperature exceeded the 80% upper limit temperature of adaptive thermal comfort standard [7, 8]. Therefore, it is essential to study the thermal comfort characteristics of occupants in the vernacular buildings for well understanding the passive design strategies and creating a comfortable built environment. China stretches across vast area covering the cold, temperate and tropical zones, according to the Chinese National Standard GB 50176 (2016), China is divided into five climatic zones for building thermal design: the severe cold zone and the cold zone in northern China; and the hot summer and cold winter zone, the hot summer and warm winter zone and the temperate zone in southern China, as shown in Fig. 4. Many thermal comfort field studies have been carried out in various climatic zones, both in urban buildings [9-13] and rural buildings [14, 15]. The living conditions are very different in urban and rural buildings in China. Urban residents improve their indoor thermal environments via air-conditioning in the summer and central heating systems in the winter, while rural residents open windows and use electric fans in the summer and stove heating in the winter. During the winter in northern China, changing clothes is the most important adaptive behavior of rural residents besides heating [16-18], while passive building design is particularly important during the summer in southern China. To adapt to the hot-humid climate, the buildings in southern China are generally designed to be lightweight and transparent to facilitate natural ventilation and relieve thermal discomfort in the summer. The typical plans of the traditional folk houses in this area are the
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
‘San-xiang-liang-lang’ in Guangfu and the ‘Si-dian-jin’ and ‘Xia-shan-hu’ in Chaoshan, as shown in Fig. 1. The actual plans are not exactly the same as the typical ones as local residents used to make adjustments according to their family population and economic situation, while the halls and patios are preserved very well in all cases. Typically in an area, many houses with the same plan form are constructed and organized into a settlement, Fig. 2 and Fig. 3 show the representative village street plans and sections. The patio and the semi-open hall are connected to each other and form the core of the houses. They create a complete natural ventilation system in combination with the outdoor "cold lane" and the interior bedrooms. They are categorized as semi-open spaces, as they are semi-closed and semi-open, with features of both indoor and outdoor spaces. These semi-open spaces are key features of the climate adaptive design of folk houses and are important areas where residents perform daily activities. Some studies have confirmed the unique advantages of the semi-open spaces in folk houses for improving the indoor thermal environment from the perspectives of building design and energy savings [22-24]. However, only a few studies have focused on human thermal comfort in such spaces. Rijal and Yoshida conducted a 40-day survey in the interior and semi-open spaces of traditional houses in five regions of Nepal in both summer and winter seasons, and they reported that residents had a high degree of satisfaction with the houses they lived in and that the neutral temperature in the semi-open spaces was higher than that of the interior spaces [25]. Ryu et al. investigated human thermal comfort in the Daechung, a semi-open space between the front and back yards, in Korean traditional houses, and they found that it was cooler in the summer due to good ventilation [26]. Other studies focused on semi-open spaces in urban areas [27-30]. China is a large agricultural country with 45% of its population and 60% of its built-up areas in rural areas. Most of the thermal comfort studies in rural areas were carried out in northern China and few were carried out in southern China; the studies in the hot-humid areas in southern China mainly focused on the interior spaces [31]. The semi-open spaces play an important role in creating a comfortable indoor environment and providing space for daytime activities, but these spaces have not received much attention in the literature. Therefore, the aims of this study were to conduct a year-long thermal comfort field study of the interior and semi-open spaces in rural folk houses in the hot-humid areas of China, to compare the two types of spaces in terms of environmental parameters and thermal comfort responses, and to identify the neutral and acceptable conditions for the rural residents of both spaces. This study contributes to the improvement of rural living conditions, informs the new rural construction campaign and safeguards agricultural development in southern China.
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
Fig. 1. Typical traditional folk house plan (adapted from [19])
AC C
EP
Fig.2. Village street I (the entrance of house is on the front) (adapted from [20, 21])
Fig.3. Village street II (the entrance of house is on both sides) (adapted from [20, 21])
ACCEPTED MANUSCRIPT 2 Methods 2.1 Location and climate
TE D
M AN U
SC
RI PT
The Guangdong Province is a large province located in southern China, and a majority of the province has a hot summer and warm winter climate as shown in Fig. 4. The annual mean air temperature and relative humidity are 10 - 24 °C and 60 - 80%, respectively, and the monthly mean air temperatures are 28 - 29 °C in July and 9 - 16 °C in January. Eleven rural areas were selected from the eastern, western and northern parts of the province, as shown in Fig. 2. They are located at a great distance from central urban areas and have similar climates. The mean air temperature and relative humidity, respectively, were 29.6 ± 0.4 °C and 79 ± 1% in the hottest month of the year, June, and 14.7 ± 0.8 °C and 75 ± 3% in the coldest month, January [32].
Fig. 4. Locations of the rural areas
2.2 Houses and spaces
AC C
EP
Modern folk houses in rural areas evolved mostly from traditional folk houses, and the semi-open spaces of halls and patios have been well retained as shown in Fig. 1. The semi-open hall is usually one-story with a high ceiling, and open at the front to ensure good ventilation and day lighting. The wall materials included bricks and stone, and the roof materials included wood and tile. In the interior spaces like bedrooms, the windows were small to prevent overheat by solar radiation. 2.3 Time of survey
To fully understand the thermal environments and thermal responses of people in the interior and semi-open spaces of rural folk houses, a survey was conducted throughout the year in 2015. The surveys were administered with the help of college students who came from the surveyed rural areas; therefore, the time in which the surveys were completed was mainly during the holidays of various seasons, including the summer and winter vacations for Tomb Sweeping Day, Labor Day, the Dragon Boat Festival, National Day, and the Mid-Autumn Festival.
ACCEPTED MANUSCRIPT
RI PT
The summer is long, hot and humid, and the winter is short and temperate. There is never any cold weather in the hot-humid areas. The survey data were divided accordingly into two groups: the summer group and the non-summer group. According to the standard QX/T 152 (2012) [33], the summer starts when the mean outdoor air temperature is over 22 °C for 5 consecutive days, and it ends when the mean outdoor air temperature is below 22 °C for 5 consecutive days. Therefore, based on analysis of the 2015 meteorological data, the summer began on April 24 and ended on November 21. 2.4 Respondents
M AN U
SC
The respondents were randomly selected, without any limits on gender, age or location (in interior or semi-open spaces). A total of 1657 sets of raw data was obtained, 51.5% of which were from male respondents and 48.5% were from female respondents. A large proportion (62.2%) of respondents were between the ages of 20 to 50 years old, while only a small proportion (5.4%) of respondents were less than 10 years old and more than 70 years old, as Table 1 shows. The sample size was 1369 for interior spaces and 288 for semi-open spaces. Table 1 Age distribution of respondents. Age group (years)
0 - 10
Number Proportion (%)
10 0.6
*
10 - 20
20 - 30
30 - 40
40 - 50
50 - 60
60 - 70
> 70
239 14.4
330 20.0
321 19.4
378 22.8
189 11.4
111 6.7
79 4.8
*including the upper limit value
Gender Number Height (cm)
Interior space
Weight (kg) BMI (kg/m²) a
AD (m2)
(clo)
Male
359
169.0 ± 5.9
60.5 ± 7.0
21.1 ± 0.8
1.69 ± 0.12
0.31 ± 0.12
Female
382
157.9 ± 5.4
51.7 ± 6.7
21.9 ± 2.0
1.50 ± 0.10
0.35 ± 0.13
All
741
163.7 ± 7.7
56.0 ± 7.6
21.3 ± 1.6
1.60 ± 0.13
0.32 ± 0.12
71
168.8 ± 4.8
59.8 ± 5.0
21.2 ± 0.8
1.68 ± 0.08
0.29 ± 0.10
71
158.3 ± 5.0
51.6 ± 4.9
21.6 ± 2.1
1.51 ± 0.07
0.32 ± 0.12
142
163.5 ± 7.2
55.7 ± 6.4
21.4 ± 1.6
1.59 ± 0.12
0.31 ± 0.11
AC C
Space
EP
TE D
Basic information on the respondents is presented for the summer and non-summer groups and for the interior and semi-open spaces categories, as shown in Tables 2 and 3. The t-test indicated that the respondents’ details were not significantly different between the interior and semi-open spaces categories for the summer group (p > 0.313). For the non-summer group, the height, body weight, body surface area, and clothing insulation were not significantly different (p > 0.272), while the BMI of the respondents was significantly lower in the interior spaces category (p < 0.0001). Table 2 Basic information on respondents in summer.
Mae Semi-open Female space All a
2
2
Notes: Body mass index (BMI) = weight/height (kg/m ） AD: DuBois surface area : The clothing insulation Table 3 Basic information on respondents in non-summer. Space
Gender Number
Height (cm)
Weight (kg) BMI (kg/m²) a
AD (m2)
(clo)
ACCEPTED MANUSCRIPT
Semi-o Mae pen Female space All
344
168.5 ± 6.3
61.6 ± 9.0
21.0 ± 0.6
1.70 ± 0.14
0.63 ± 0.25
284
157.4 ± 5.1
51.0 ± 7.3
21.4 ± 1.1
1.49 ± 0.11
0.70 ± 0.30
628
163.5 ± 8.3
57.0 ± 9.5
21.4 ± 1.1
1.60 ± 0.16
0.65 ± 0.27
79
168.9 ± 7.5
62.8 ± 8.4
21.4 ± 1.2
1.71 ± 0.15
0.62 ± 0.23
67
158.3 ± 4.6
52.1 ± 6.5
22.8 ± 1.0
1.51 ± 0.09
0.62 ± 0.23
146
164.0 ± 8.3
57.9 ± 9.3
22.0 ± 1.3
1.62 ± 0.16
0.62 ± 0.23
a
2
2
Notes: Body mass index (BMI) = weight/height (kg/m ） AD: DuBois surface area : The clothing insulation 2.5 Field survey methods
RI PT
Male Interior Female space All
TE D
M AN U
SC
The field survey was conducted by the college students that resided in the rural areas. Upon approval, the students visited the respondents’ houses. After observing the layout of the house and the position of the respondents, the students installed the instruments close to the respondents and recorded the data after 5 - 10 minutes when the readings were stable. The questionnaire was conducted at the same time as the data recording, during which the students read the questions one by one and recorded the responses of the respondents on paper (Fig. 5).
EP
Fig. 5. Field survey
AC C
The main contents of the questionnaire include the following: 1) personal information, such as gender, age, height, weight, and time living at current location; 2) personal factors, such as clothing, current activity and activity within 20 minutes prior to beginning the survey; and 3) subjective responses, such as thermal sensation, comfort, acceptability and preference. The length of time living in the area was used to confirm that the respondent had acclimated to the local climate. Respondents’ clothing type was recorded in detail during the survey, and the insulation of the ensemble was estimated as the sum of the individual values listed in the ASHRAE standard 55 [34]. Since chairs in the rural areas were mostly wooden, the additional thermal insulation value for a chair was set to zero. Some of the respondents engaged in heavy physical work just before the survey, which may have affected their comfort perceptions. Therefore, we excluded those samples, and retained the sample that respondents engaged in sedentary or light activities. The metabolic rate of the respondents was estimated to be 1.2 met, considering both the current and prior activities.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Fig. 6 shows the rating scales used for the subjective responses in the survey. Considering the hot and humid climate, a nine-point extended scale from ISO 15001 (ISO, 2002) was used for the thermal sensation rating. Traditional scales were used for thermal comfort and acceptability ratings. The questionnaire was written in Chinese, and the Chinese wording for the scale degrees from the Chinese national standard GB/T 18977 (AQSIQ, 2003) was used.
Fig. 6 Scales used for thermal response in the survey
=
(1)
EP
TE D
Thermal environment parameters of air temperature (ta), globe temperature (tg), air speed (v) and relative humidity (RH) were measured by the HD 32.3 instrument. The measuring range and accuracy agreed with ISO 7726, as shown in Table 4. The measuring point was set up near the respondents, and it was placed to avoid obvious heat or cold sources in accordance with the ASHRAE standard 55. The height of the sensors was 0.6 m (sedentary) or 1.1 m (standing) above the floor. The operative temperature (top) was calculated according to the ASHRAE Handbook-Fundamentals.
AC C
Measurement content Air temperature Relative humidity Air speed
Globe temperature
Table 4 Instrument parameters. Operation range Accuracy -10 to 80 °C ± 0.2 °C 5% - 98% ± 2% (15 - 90%) ± 2.5% (other scope) 0.05 - 5 m/s ± 0.05 m/s (0 - 0.99 m/s) ± 0.15 m/s (1 - 5 m/s) -10 to 100 °C ± 0.2 °C
2.6 Data process and analysis An independent sample t-test was used to test the environmental differences between interior and semi-open spaces. Linear regression was adopted to derive the relationship between the operative temperature and the thermal sensation, and covariance analysis was used to test the significant differences between various regression lines. Polynomial
ACCEPTED MANUSCRIPT regression was used to analyze the quantitative relationship between the operative temperature and the percentage dissatisfied. In addition, the “bin” method was repeatedly used in this study, i.e., the data were divided into bins by operative temperature in 0.5 °C steps, the values of the variables were averaged for each bin, and the averages were analyzed by their weighted sample size. All the statistical analyses were done with SPSS v 22.0 software (IBM, New York, NY, US), and all the differences were accepted as significant at a 0.05 level.
RI PT
3 Results 3.1 Thermal environment
M AN U
SC
Table 5 shows the thermal environment results in the summer for the interior and semi-open spaces. Compared with those of the interior spaces, no significant differences were found in the semi-open spaces in air temperature, mean radiant temperature and relative humidity (p > 0.096), while air speed was significantly greater (p = 0.014). The detailed air speed distribution is shown in Fig. 7. The proportion of air speed below 0.4 m/s accounted for 58.4% in the interior spaces, while the proportion of air speed between 0.2 - 0.7 m/s accounted for 50.7% in the semi-open spaces. Table 5 Thermal environment parameters in summer. Interior space (°C) (°C)
(m/s)
RH (%)
(°C)
Semi-open space
(°C)
(m/s)
RH (%)
29.5
29.7
0.54
76.1
29.9
30.1
0.68
74.7
Standard deviation
3.0
3.0
0.61
9.8
2.5
2.5
0.63
8.8
Minimum
18.1
18.2
0.01
48.7
18.8
19
0.01
45.4
Maximum
38.3
38.9
2.93
99.9
39
37.8
2.90
99.9
AC C
EP
TE D
Mean
Fig. 7. Air speed distribution in summer Table 6 shows the thermal environment results in the non-summer. Compared with those of the interior spaces, no significant differences were found in the semi-open spaces for air
ACCEPTED MANUSCRIPT
RI PT
temperature or mean radiant temperature (p > 0.058), while relative humidity was significantly lower (p < 0.0001), and air speed was significantly greater (p < 0.0001). From the relative humidity distribution (see Fig. 8), it can be seen that the proportion of relative humidity between 45% - 65% was larger, while the proportion of relative humidity between 65% - 90% was smaller in the semi-open spaces. From the air speed distribution (see Fig. 9), the air speed was mostly below 0.5 m/s in the interior spaces, while it was 0.3 -1.1 m/s in the semi-open spaces. It can also be seen from Tables 5 and 6 that the standard deviations of thermal environment parameters were quite similar between the interior and semi-open spaces. The standard deviation of air speed was relatively greater than the other parameters, indicating that air speed was more variable in the rural houses.
Interior space (°C) (°C)
Semi-open space
RH (%)
(°C)
(°C)
(m/s) RH (%)
21.5
21.4
0.18
70.2
22
22.2
0.38
65.2
3.7
3.8
0.31
12.3
4.3
4.4
0.38
13.4
Minimum
13.4
13.5
0.01
26.1
12.4
12.7
0.01
28.1
Maximum
32.3
32.5
2.78
99.9
30.4
31.1
2.96
99.9
AC C
EP
TE D
Standard deviation
M AN U
Mean
(m/s)
SC
Table 6 Thermal environment parameters in non-summer.
Fig. 8. Relative humidity distribution in non-summer
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Fig. 9. Air speed distribution in non-summer 3.2 Clothing insulation
AC C
EP
TE D
In rural areas, clothing adjustment is one of the most important thermal adaptation methods, and people are accustomed to adding or removing clothes to meet their thermal comfort requirements. It was hot and humid during the summer data collection, and rural residents generally wore short-sleeved shirts, shorts and sandals. The relationship between the operative temperature and clothing insulation in summer is shown in Fig. 10. When the operative temperature was below 28 °C, the clothing insulation decreased with increasing temperature, and when the operative temperature was above 28 °C, the clothing insulation was basically stable at the same level. The stable clothing insulations were 0.30 ± 0.09 clo and 0.29 ± 0.09 clo in the interior and semi-open spaces, respectively, and they were not significantly different (p = 0.813).
Fig. 10. Clothing insulation changes with top in summer The clothing insulation varied greatly between 0.27 clo - 1.2 clo and changed linearly with the operative temperature in the non-summer group, as shown in Fig. 11. The higher the
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
operative temperature, the smaller the clothing insulation was. The regression equations for the interior and semi-open spaces are shown below: Icl = -0.040 top + 1.517 (R2 = 0.808) (Interior space) (2) 2 (Semi-open space) (3) Icl = -0.026 top + 1.193 (R = 0.524) A significant difference (analysis of covariance, p = 0.003) was found between the two regression lines, indicating that the residents changed clothing less sensitively in the semi-open spaces.
Fig. 11. Clothing insulation changes with top in non-summer
TE D
3.3 Thermal sensation
AC C
EP
Good linear relationships between thermal sensation (TS) and operative temperature were established in the summer for the interior and semi-open spaces, as shown in Fig. 12 and the following equations: TS = 0.232 top - 5.579 (R² = 0.722) (Interior space) (4) TS = 0.255 top - 5.976 (R² = 0.652) (Semi-open space) (5) -1 The thermal sensitivity, namely, the slope of the regression line, was 0.232 °C for the interior spaces and 0.255 °C-1 for the semi-open spaces; that is, for each 1 °C increase of top, thermal sensation increased by 0.232 and 0.255 units of scale for the interior spaces and semi-open spaces, respectively. No significant difference was found between the slopes of the two regression lines (analysis of covariance, p = 0.701). The summer thermal neutral temperature was determined to be 24.0 °C for the interior spaces and 23.4 °C for the semi-open spaces using Eqs. (4) and (5), and the latter was 0.6 °C lower than the former. Additionally, the size was smaller and the points were much more scattered for the data with operative temperatures below 26 °C.
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 12. The relationship between thermal sensation votes and top in summer
AC C
EP
TE D
M AN U
Good linear relationships between thermal sensation and operative temperature were established in the non-summer sample, as shown in Fig. 13 and the following equations: TS = 0.238 top - 4.700 (R² = 0.841) (Interior space) (6) (Semi-open space) (7) TS = 0.200 top - 3.673 (R² = 0.670) -1 -1 The thermal sensitivity was 0.238 °C in the interior spaces and 0.200 °C in the semi-open spaces, and no significant difference was found between the two (analysis of covariance, p = 0.246). The non-summer thermal neutral temperatures were determined to be 19.7 °C and 18.4 °C for the interior and semi-open spaces, respectively, and the latter was 1.3 °C lower than the former.
Fig. 13. The relationship of thermal sensation voting and top in non-summer
3.4 Percentage dissatisfied The percentage dissatisfied was the percentage of respondents who reported a negative thermal acceptability vote. Second-order polynomial functions relating the percentage dissatisfied (PD) and the operative temperature were established in the summer for the interior and semi-open spaces, as shown in Fig. 14 and the following equations: (Interior space) (8) PD = 0.326 top2 -15.144 top + 175.884 (R2 = 0.621) 2 2 PD = 0.363 top -16.328 top + 181.641 (R = 0.415) (Semi-open space) (9)
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
The percentage dissatisfied increased with temperature when the operative temperature was higher than 26 °C. The upper limit of the 80% thermal acceptable operative temperature was 31.1 °C for the interior spaces and 30.3 °C for the semi-open spaces, and the latter was 0.8 °C lower than the former.
Fig. 14. Percentage dissatisfied changes with top in summer
AC C
EP
TE D
The results for the non-summer group are shown in Fig. 15 and the following equations: PD = 0.218 top2 - 9.264 top + 99.118 (R2 = 0.395) (Interior space) (10) PD = 0.352 top2 - 13.411 top + 129.999 (R2 = 0.322) (Semi-open space) (11) Only weak relationships between the percentage dissatisfied and the operative temperature were found in non-summer group. When the operative temperature was above 21 °C, the percentage dissatisfied increased with the temperature, and this trend was more obvious in the semi-open spaces. The upper limit of the 80% thermal acceptable operative temperature was 30.7 °C for the interior spaces and 26.0 °C for the semi-open spaces, and the latter was 4.7 °C lower than the former. The percentage dissatisfied was below 20% for most of the data when the operative temperature was below 21 °C, indicating that there was no lower temperature limit.
Fig. 15. Percentage dissatisfied changes with top in non-summer
ACCEPTED MANUSCRIPT 3.5 Thermal preference
M AN U
SC
RI PT
The votes for the "no change" preference were equally divided into the votes expecting "cooler" and "warmer", and the relationship between the proportions of "cooler" and "warmer" preferences and operative temperature was obtained, as shown in Figs. 14 and 15. In the summer, the thermal preference changed obviously with the operative temperature only when the operative temperature was above 25 °C (Fig. 16). Probit regression was thus used for those data, showing that the preferred operative temperatures were 25.6 °C and 25.3 °C for the interior and semi-open spaces, respectively. These values were similar to each other and were higher than the thermal neutral temperatures in the spaces in the summer.
(a) In interior space (b) In semi-open space Fig. 16. Thermal preference changes with top in summer
AC C
EP
TE D
In the non-summer season, the thermal preference proportions changed obviously with operative temperature for all temperature ranges (Fig. 17). Preferred temperatures were determined to be 21.4 °C and 20.8 °C for the interior and semi-open spaces, respectively. These values were similar to each other and were higher than the thermal neutral temperatures in the spaces in the non-summer group.
(a) In interior space (b) In semi-open space Fig. 17. Thermal preference changes with top in non-summer 3.6 Airflow preference By using the same method for thermal preference, the relationship between the percentage
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
of airflow preference and the operative temperature was obtained, as shown in Figs. 18 and 19. In the summer, the proportion of respondents wanting “more airflow” increased rapidly with increasing temperature when the operative temperature was above 20 °C for the interior spaces and when the operative temperature was above 26 °C for the semi-open spaces. In the non-summer, the proportion of respondents wanting “more airflow” increased obviously with temperature when the operative temperature was above 24 °C for the interior spaces and above 28 °C for the semi-open spaces.
AC C
EP
TE D
(a) In interior space (b) In semi-open space Fig. 18. Airflow preference changes with top in summer
(a) In interior space (b) In semi-open space Fig. 19. Airflow preference changes with top in non-summer
4 Discussion
4.1 Interior vs semi-open spaces The results show that the thermal neutral temperature of residents in the semi-open spaces was lower than that in the interior spaces, both in the summer (0.6 °C) and non-summer (1.3 °C) seasons. Following the same trend, the results show that compared with those of the interior spaces, the upper limit of the 80% thermal acceptable range was offset toward a lower temperature by 0.8 °C in the summer and by 4.7 °C in the non-summer for the semi-open
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
spaces. These findings are different from the findings of Rijal et al., who conducted a 40-day field investigation of interior and semi-open spaces in five regions of Nepal during the summer and winter seasons and found that the thermal neutral temperature of the residents in semi-open spaces was higher than that of the residents in interior spaces [24]. Thermal neutral and acceptable temperatures are affected by many factors, including physical, physiological and psychological factors. Compared with those of the interior spaces, air speed was greater and relative humidity was lower, while the other parameters were the same in the semi-open spaces. According to the heat balance principle, the human body can maintain thermal neutrality at a higher operative temperature under the conditions of higher air speed or lower relative humidity. The present results differ from this, and it is therefore reasonable to conclude that physical factors are not relevant to the present findings. Literature on heat acclimatization indicates that changes in the human thermoregulatory mechanism can be triggered by repeated exposure to extreme hot and humid conditions. This is obviously not the case in the present study. Therefore, psychological factors are the main reasons for the differences in neutral and acceptable temperatures between the two spaces. Studies have claimed that the indoor thermal history in all seasons of a year played a key role in shaping the subjects’ sensations in a wide range of thermal conditions [35]. In this study, the times that the respondents stayed in the interior and semi-open spaces were approximately equal, and the respondents’ experiences with the two spaces contributed equally to the indoor thermal history. A previous study indicated that perceived control had a marked effect on comfort and satisfaction [36]. One reason for the present findings could be that the residents had more adaptive opportunities in the interior spaces, such as using fans and moving from one room to another, and hence could accept the higher temperatures. In addition, there was a large difference in lighting between the two environments; it was dark in the interior spaces and bright in the semi-open spaces, and the interaction between light exposure and thermal perceptions may have impacted the results as well.
EP
4.2 Summer vs non-summer
AC C
The difference between the summer and non-summer samples in the thermal sensation and operative temperature relationship was observed by transferring Figs. 12 and 13 to Fig. 20. In both the interior and semi-open spaces, residents’ thermal sensations in the summer season were always lower than they were in the non-summer season. This finding may be caused by clothing adjustment. According to Figs. 10 and 11, the residents’ clothing insulation was lower in the summer than in the non-summer, and therefore their thermal sensation was lower.
RI PT
ACCEPTED MANUSCRIPT
(a) In interior space
(b) In semi-open space
SC
Fig. 20. The relationship between thermal sensation and operative temperature with seasons
TE D
4.3 Rural vs urban residents
M AN U
Eqs. (3) and (5) show that the thermal sensitivity of residents in the interior spaces was 0.232 °C-1 in the summer and 0.238 °C-1 in the non-summer. The same findings were found for the semi-open spaces, showing that the summer thermal sensitivity was 0.255 °C-1 and the non-summer thermal sensitivity was 0.200 °C-1. Further analysis indicated that thermal sensitivity was not significantly different in either interior spaces or in semi-open spaces (analysis of covariance, p > 0.262). In summary, the thermal sensitivity of the rural residents did not change with the seasons regardless of whether they were in interior or semi-open spaces.
AC C
EP
Zhang et al. [37] and Chen et al. [38] conducted field studies on the thermal comfort of urban residents in hot-humid areas. The survey was conducted from May to October, which was consistent with the summer season in this study. With the permission of the authors, the raw data of the urban residents were collected, and a good linear relationship between thermal sensation and operative temperature was obtained as follows: TS = 0.396 top - 10.626 (R² = 0.949) (12) The equation shows that the thermal neutral temperature was 26.8 °C and the thermal sensitivity was 0.396 °C-1 for the urban residents, which were both higher than those of the rural residents in the summer. A good second-order polynomial function was established relating percentage dissatisfied with operative temperature as follows: PD = 1.477 top2 - 78.806 top + 1054.157 (R² = 0.965) (13) The equation shows that the upper limit of the 80% acceptable operative temperature was 30.1 °C for the urban residents, which was lower than that of the rural residents by 0.2-1 °C. A recent climatic chamber study in hot-humid areas found that rural subjects felt more comfortable and were more satisfied under identical cool or warm conditions and that they had a much wider acceptable temperature range than the urban subjects [39]. This finding is consistent with the present findings. The possible reasons for these findings were also proposed as differences in local culture, expectations and environmental cognition between the rural and urban residents.
ACCEPTED MANUSCRIPT 4.4 Practical implication
AC C
EP
TE D
M AN U
SC
RI PT
Many studies have revealed that residents in the traditional buildings were more satisfied with the thermal environment than those in the modern buildings. For example, Dili confirmed that in Kerala the traditional residential buildings were more effective in providing comfortable indoor environment than the modern buildings [40]; Nematchoua conducted a survey in five towns of four climate zones in Cameroon and reported that the traditional buildings were more comfortable than the modern buildings [41]; and Subramanian reported that 88% of the residents in traditional buildings feel comfortable with indoor air temperature in Thanjavur, while the proportion was only 54% for modern buildings [42]. Nevertheless, the above studies fail to provide the reasons behind the higher satisfaction in traditional buildings: due to architectural passive design or human thermal adaptation. On the other hand, at the initial phase of architectural design, architects used to determine the passive and climate responsive strategies by using the Givoni's or Olgyay's bioclimatic charts. However, the thermal comfort requirements of people from various areas might be different due to thermal history, lifestyle or culture background. This difference has not yet been reflected in the current bioclimatic design. This study found that, in the hot and humid areas of southern China, the thermal neutral and acceptable temperatures were significantly different between the rural and urban people, and the upper limit of 80% acceptable temperature was higher by 0.2-1℃ for the rural people in summer. This finding provides an important clue for explaining the higher satisfaction of residents in traditional buildings, and presents as a key basis for bioclimatic design of both rural and urban buildings. In addition, this study also found that the residents’ thermal demands were quite difference in the two kinds of spaces in the rural buildings. Compared to those of the interior spaces, both the thermal neutral temperature and upper limit of acceptable temperatures were lower in the semi-open spaces. This provides valuable indications that residents were more rigorous to the semi-open spaces, and architects need to design the interior and semi-open spaces based on their own thermal comfort standard. In conclusion, this paper clarifies the varied thermal comfort standards for the rural and urban areas and the interior and semi-open spaces, supports the design of various types of buildings and spaces, and greatly helps to improve their indoor environmental qualities. 5 Conclusions
A year-long thermal comfort field study was conducted in rural folk houses in the hot-humid areas of China. A total of 1657 sets of raw data that included both thermal environment parameters and respondents’ subjective responses were collected regarding interior and semi-open spaces. The main conclusions are shown below. 1. Air speed was greater and relative humidity was lower in the semi-open spaces. 2. The residents’ clothing insulation varied with indoor operative temperature from 0.27 clo to 1.2 clo in the non-summer season, and it was steady at 0.30 clo during the summer season when indoor operative temperature was above 28 °C. The clothing change with temperature was less sensitive in the semi-open spaces.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
3. In the interior spaces, the thermal neutral, upper limit of the 80% acceptable and preferred operative temperatures, respectively, were determined to be 24.0 °C, 31.1 °C and 25.6 °C in the summer and 19.7 °C, 30.7 °C and 21.4 °C in the non-summer. 4. In the semi-open spaces, the thermal neutral, upper limit of the 80% acceptable and preferred operative temperatures, respectively, were determined to be 23.4 °C, 30.3 °C and 25.3 °C in the summer and 18.4 °C, 26.0 °C and 20.8 °C in the non-summer. 5. Compared to the interior spaces, the thermal neutral temperature was lower by 0.6 °C in the summer and 1.3 °C in the non-summer, and the upper limit of 80% acceptable temperature decreased by 0.8 °C in the summer and 4.7 °C in the non-summer in the semi-open spaces. The differences might be due to the decrease in adaptive opportunities and the brighter lighting environment of the semi-open spaces. 6. Compared to the non-summer samples, the thermal sensitivity remained the same, and the thermal neutral temperature was much higher in the summer. This difference might be caused by decreased clothing insulation in the summer. 7. Compared to the urban residents, the upper limit of the 80% acceptable temperature was 0.2 -1 °C higher for the rural residents. This finding is consistent with the present findings and might be caused by differences in local culture, expectations and environmental cognition between rural and urban residents. Acknowledgements
References
TE D
This work was supported by the National Natural Science Foundation of China under Grant (No. 50708038，No.51708228); the Program for New Century Excellent Talents in University under Grant (No. NCET-10-0373).
AC C
EP
[1] Fernandes J, Pimenta C, Mateus R, et al. Contribution of Portuguese Vernacular Building Strategies to Indoor Thermal Comfort and Occupants' Perception. Buildings, 2015, 2015(5):1242-1264. [2] Manioğlu. G, Yılmaz.Z. Energy efficient design strategies in the hot dry area of Turkey. Building & Environment, 2008, 43(7):1301-1309. [3] Baran M, Yıldırım M, Yılmaz A. Evaluation of ecological design strategies in traditional houses in Diyarbakir, Turkey. Journal of Cleaner Production, 2011, 19(6–7):609-619. [4] Michael A, Demosthenous D, Philokyprou M. Natural Ventilation for Cooling in Mediterranean Climate: A Case Study in Vernacular Architecture of Cyprus. Energy & Buildings, 2017, 144. [5] Hidayahtuljamilah N, Kassim N, Zafrullah H, et al. Re-adaptation of Malay vernacular architecture thermal comfort elements:towards sustainable design in Malasia. (A literature review). International Conference on Innovation and Technology for Sustainable Built Environment. 2012. [6] Foruzanmehr D. Summer-time Thermal Comfort in Vernacular Earth Dwellings in Yazd, Iran. International Journal of Sustainable Design, 2012, 2(1). [7] Toe D, Kubota T. Comparative assessment of vernacular passive cooling techniques for improving indoor thermal comfort of modern terraced houses in hot–humid climate of Malaysia. Solar Energy, 2015, 114:229-258. [8] Kubota T, Toe D. Application of Passive Cooling Techniques in Vernacular Houses to Modern Urban Houses: A Case Study of Malaysia. Procedia - Social & Behavioral Sciences, 2015, 179:29-39.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[9] Liu W, Zheng Y, Deng Q, et al. Human thermal adaptive behaviour in naturally ventilated offices for different outdoor air temperatures: A case study in Changsha China. Building & Environment, 2012, 50:76-89. [10] Yao R, Liu J, Li B. Occupants’ adaptive responses and perception of thermal environment in naturally conditioned university classrooms. Applied Energy, 2010, 87(3):1015-1022. [11] Wang Z, Zhang L, Zhao J, He Y. Thermal comfort for naturally ventilated residential buildings in Harbin. Energy & Buildings, 2010, 42:2406-15. [12] Zhang Y, Wang J, Chen H, et al. Thermal comfort in naturally ventilated buildings in hot-humid area of China. Heating Ventilating & Air Conditioning, 2011, 45(11):2562-2570. [13] Zhang Y, Chen H, Meng Q. Thermal comfort in buildings with split air-conditioners in hot-humid area of China. Building & Environment, 2013, 64(6):213-224. [14] Yan H, Yang L, Zheng W, et al. Analysis of behaviour patterns and thermal responses to a hot-arid climate in rural China. Journal of Thermal Biology, 2016, 59:92-102. [15] Du X, Bokel R, Dobbelsteen A. Building microclimate and summer thermal comfort in free-running buildings with diverse spaces: A Chinese vernacular house case. Building & Environment, 2014, 82:215-227. [16] Sheng X. Filed test and analysis of thermal comfort in rural house and urban residential building in sever cold region. Harbin,China: Harbin Institute of Technology; 2013. [17] Huang L, Zhu Y, Ouyang Q, Cao B. Field survey of indoor thermal comfort in rural housing of northern China in heating season. Journal of Southeast University (English Edition), 2010, 26(2):169-72. [18] Liu Y, Qian Y, Yan H, et al. Field study on thermal comfort of rural houses in winter in a the Guanzhong region,Shaanxi Province. Journal of Xian University of Architecture & Technology, 2011, 43(4):551-556. [19] Lu Q. Guangdong folk house. Beijing: China Architecture & Building Press, 2008. [20] Chen J. Research on Natural Ventilation System of Traditional Village in Chaoshan Area.Guanghzou, China: South China university of Technology; 2014. [21] Hui X. Study on climate adaptation of cold lane-courtyard space system in Guangfu traditonal villages. Guanghzou, China: South China university of Technology; 2016. [22] Huang Y, Ma J. Analysis of Measured Thermal Environment of Traditional Architecture Lengxiang in Lingnan. Energy Conservation in Buildings, 2014, 6 (42): 75-78. [23] Xie H. Naturally Ventilation of Lingnan Residential Houses. Small Town Construction, 2007, 12: 58-62 (in Chinese). [24] Zhou X, Ma J, Zhou X, Measured Analysis of Lingnan Residential Houses Indoor Thermal Environment. Heating, Ventilation & Air Conditioning, 2011, 41 (5): 94-97 (in Chinese). [25] Rijal HB, Yoshida H, Umemiya N. Seasonal and regional differences in neutral temperatures in Nepalese traditional vernacular houses. Building & Environment, 2010, 45(12):2743-2753. [26] Ryu Y, Kim S, Lee D. The influence of wind flows on thermal comfort in the Daechung, of a traditional Korean house. Building & Environment, 2009, 44(1):18-26. [27] Lu S, Fang K, Qi Y, et al. Influence of Natural Ventilation on Thermal Comfort in Semi-open Building under Early Summer Climate in the Area of Tropical Island . Procedia Engineering, 2015, 121:944-951. [28] Hwang R, Lin T. Thermal Comfort Requirements for Occupants of Semi-Outdoor and Outdoor Environments in Hot-Humid Regions. Architectural Science Review, 2007, 50(4):357-364.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[29] Spagnolo J, Dear RD. A field study of thermal comfort in outdoor and semi-outdoor environments in subtropical Sydney Australia. Building & Environment, 2003, 38(5):721-738. [30] Nakano J, Tanabe S I. Thermal comfort and adaptation in semi-outdoor environments. Ashrae Transactions, 2004, 110:543-553. [31] Jin L, Meng Q, Zhao L. Indoor Environment and Thermal Comfort in Rural Houses in East Guangdong of China. Journal of Civil, Architectural & Environmental Engineering, 2013, 35 (2):105-12(in Chinese). [32] China Meteorological Administation.China Meteorological Data Service Center. http://data.cma.cn/site/index.html. [33] China Meteorological Administration.Division of climatic season, QX/T 152. Beijing, China, 2012 (in Chinese). [34] ASHRAE. Thermal enviromental conditions for human occupancy, Standard 55. Atlanta, USA, 2004. [35] Zhang Y, Chen H, Wang J, et al. Thermal comfort of people in the hot and humid area of China—impacts of season, climate, and thermal history. Indoor Air, 2016, 26(5):820-830. [36] Paciuk M. The role of personal control of the environment in thermal comfort and satisfaction at the workplace. Coming of age, EDRA 21. Oklahoma city. OK, USA, 1990, 303-12. [37] Zhang Y, Liu Q, Meng Q. Airflow utilization in buildings in hot and humid areas of China. Building & Environment, 2015, 87:207-214. [38] Chen H, Zhang Y, Wang J, Meng Q, Research on Summer Thermal Comfort of Naturally Ventilation Buildings in Hot and Humid Region of China - Taking Guangzhou as an Example. Heating, Ventilation & Air Conditioning, 2010.40(2): 96-101 (in Chinese). [39] Zhang Z, Zhang Y, Jin L. Thermal comfort of rural residents in a hot–humid area. Building Research & Information, 2017, 45(1-2):209-221. [40] Dili A, Naseer M, Varghese T. Thermal comfort study of Kerala traditional residential buildings based on questionnaire survey among occupants of traditional and modern buildings. Energy & Buildings, 2010, 42(11):2139-2150. [41] Nematchoua M, Tchinda R, Orosa J. Thermal comfort and energy consumption in modern versus traditional buildings in Cameroon: A questionnaire-based statistical study. Applied Energy, 2014, 114(2):687-699. [42] Subramanian C, Ramachandran N, Kumar S. Performance evaluatuion of traditonal and mordern residential buildings for thermal comfort by questionare survey in Thanjavur. International Journal of Civil Engineering & Technology, 2017, 8(82):440-451.
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
Thermal sensitivity of people kept the same regardless of spaces and seasons. Thermal neutral temperature was lower in semi-open spaces than interior spaces. Acceptable temperature range was narrower in semi-open spaces than interior spaces. Acceptable temperature range of rural residents was wider than urban residents.
S0360-1323(17)30485-7
DOI:
10.1016/j.buildenv.2017.10.028
Reference:
BAE 5141
To appear in:
Building and Environment
Received Date: 28 July 2017 Revised Date:
18 October 2017
Accepted Date: 21 October 2017
Please cite this article as: Zhang Z, Zhang Y, Jin L, Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas, Building and Environment (2017), doi: 10.1016/ j.buildenv.2017.10.028. 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
Thermal comfort in interior and semi-open spaces of rural folk houses in hot-humid areas Zhongjun Zhang a, b, Yufeng Zhang a, *, Ling Jin c a
RI PT
State Key Laboratory of Subtropical Building Science, Department of Architecture, South China University of Technology, Wushan, Guangzhou, 510640, China b School of Civil and Architectural Engineering, Anyang Institute of Technology, Anyang, 455000, China c Department of Water Conservancy and Civil Engineering, South China Agriculture University, Wushan, Guangzhou, 510640, China * Corresponding author: E-mail address: [email protected](Y.Zhang)
AC C
EP
TE D
M AN U
SC
ABSTRACT Semi-open spaces are a key element in the climate adaptive design of rural folk houses and are important places where rural residents perform daily activities in hot-humid areas. In this study, a year-long thermal comfort field survey was conducted in rural folk houses in 11 rural areas in the Guangdong Province, the hot-humid area of southern China. The subjective responses of residents were recorded via questionnaires, and all ambient environment parameters were measured. The results show that air speed was greater and relative humidity was lower in the semi-open spaces. The clothing insulation of the survey respondents varied with indoor operative temperatures from 0.27 clo to 1.2 clo during the non-summer season and remained steady at 0.30 clo during the summer season. The thermal neutral, acceptable and preferred indoor operative temperatures were determined for the interior and semi-open spaces in the summer and non-summer. Compared to those of the interior spaces, the thermal neutral temperature was 0.6 - 1.3 °C lower and the upper limit of 80% acceptable temperature was decreased by 0.8 - 4.7 °C in the semi-open spaces. The acceptable temperature in the summer for the rural residents was found to be 0.2 - 1 °C higher than that of the urban residents in the same climate. Psychological factors such as local culture, expectations and perceived control might be the reasons for these differences. This study contributes to a better understanding of the thermal comfort of rural people and to the improvement of rural living conditions. Keywords: Rural folk house, Thermal comfort, Hot-humid area, Semi-open space, Field study 1 Introduction
Both climatic conditions and economic level have important effects on building’s performance and energy consumption. In the economically developed areas, HVAC systems are widely used for maintaining comfortable indoor environment in whatever outdoor climate, and the buildings consume lots of energy and discharged amounts of pollutions. Only some vernacular buildings, where HVAC devices and systems are used very less, retain well the architectural characteristic of climatic adaptability and energy efficient, in which man, building and environment develop harmoniously and sustainably.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
The global climate is so diverse and the vernacular buildings vary widely in their functions, forms and materials. In cold climates, the design strategy of vernacular buildings is to increase heat gains and reduce heat loss, while in hot climates the opposite strategy is used. For example, in the northern of Portugal, as the winter is very cold, glass balcony that obtains more solar radiation is the main feature of the traditional architectures; while in the south, small size windows and doors and patios (courtyards) are the main features for minimizing heat gains in summer [1]. In Turkey, in order to adapt to dry-hot climate and reduce solar radiation, the traditional buildings often enclose together, forming a small courtyard that provides good shading and comfortable microclimate. In addition, thermal mass envelopes are often used to cope with the larger diurnal range of outdoor temperature [2, 3]. Similarly in the Eastern Mediterranean region, thick walls are used, open and semi-open spaces are the common architectural forms, and night ventilation cooling is the main passive cooling strategy in some vernacular buildings [4]. In Malaysia it is always high temperature, high humidity and abundant rainfall as it is in the equatorial hot areas. In order to create a good indoor environment, the vernacular houses in Malaysia often raise on the posts above the ground, open many spaces or windows in the wall and have a high ceiling to provide good ventilation and improve comfort [5]. The passive design strategies greatly improve the environmental quality of the vernacular buildings, but with no guarantee for satisfying residents’ thermal comfort requirements. For example, Foruzanmeh conducted a field study on the vernacular dwellings in central Iran and reported that the dwellings could not provide comfortable temperature throughout an entire typical hot summer day [6]. Similar results were found for the modern buildings using the passive strategies of vernacular buildings. Toe et al. investigated the modern Malaysia buildings that used the traditional passive cooling technology. They found that using passive design reduced thermal discomfort, but could not provide thermal comfort for a whole day, and the indoor operative temperature exceeded the 80% upper limit temperature of adaptive thermal comfort standard [7, 8]. Therefore, it is essential to study the thermal comfort characteristics of occupants in the vernacular buildings for well understanding the passive design strategies and creating a comfortable built environment. China stretches across vast area covering the cold, temperate and tropical zones, according to the Chinese National Standard GB 50176 (2016), China is divided into five climatic zones for building thermal design: the severe cold zone and the cold zone in northern China; and the hot summer and cold winter zone, the hot summer and warm winter zone and the temperate zone in southern China, as shown in Fig. 4. Many thermal comfort field studies have been carried out in various climatic zones, both in urban buildings [9-13] and rural buildings [14, 15]. The living conditions are very different in urban and rural buildings in China. Urban residents improve their indoor thermal environments via air-conditioning in the summer and central heating systems in the winter, while rural residents open windows and use electric fans in the summer and stove heating in the winter. During the winter in northern China, changing clothes is the most important adaptive behavior of rural residents besides heating [16-18], while passive building design is particularly important during the summer in southern China. To adapt to the hot-humid climate, the buildings in southern China are generally designed to be lightweight and transparent to facilitate natural ventilation and relieve thermal discomfort in the summer. The typical plans of the traditional folk houses in this area are the
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
‘San-xiang-liang-lang’ in Guangfu and the ‘Si-dian-jin’ and ‘Xia-shan-hu’ in Chaoshan, as shown in Fig. 1. The actual plans are not exactly the same as the typical ones as local residents used to make adjustments according to their family population and economic situation, while the halls and patios are preserved very well in all cases. Typically in an area, many houses with the same plan form are constructed and organized into a settlement, Fig. 2 and Fig. 3 show the representative village street plans and sections. The patio and the semi-open hall are connected to each other and form the core of the houses. They create a complete natural ventilation system in combination with the outdoor "cold lane" and the interior bedrooms. They are categorized as semi-open spaces, as they are semi-closed and semi-open, with features of both indoor and outdoor spaces. These semi-open spaces are key features of the climate adaptive design of folk houses and are important areas where residents perform daily activities. Some studies have confirmed the unique advantages of the semi-open spaces in folk houses for improving the indoor thermal environment from the perspectives of building design and energy savings [22-24]. However, only a few studies have focused on human thermal comfort in such spaces. Rijal and Yoshida conducted a 40-day survey in the interior and semi-open spaces of traditional houses in five regions of Nepal in both summer and winter seasons, and they reported that residents had a high degree of satisfaction with the houses they lived in and that the neutral temperature in the semi-open spaces was higher than that of the interior spaces [25]. Ryu et al. investigated human thermal comfort in the Daechung, a semi-open space between the front and back yards, in Korean traditional houses, and they found that it was cooler in the summer due to good ventilation [26]. Other studies focused on semi-open spaces in urban areas [27-30]. China is a large agricultural country with 45% of its population and 60% of its built-up areas in rural areas. Most of the thermal comfort studies in rural areas were carried out in northern China and few were carried out in southern China; the studies in the hot-humid areas in southern China mainly focused on the interior spaces [31]. The semi-open spaces play an important role in creating a comfortable indoor environment and providing space for daytime activities, but these spaces have not received much attention in the literature. Therefore, the aims of this study were to conduct a year-long thermal comfort field study of the interior and semi-open spaces in rural folk houses in the hot-humid areas of China, to compare the two types of spaces in terms of environmental parameters and thermal comfort responses, and to identify the neutral and acceptable conditions for the rural residents of both spaces. This study contributes to the improvement of rural living conditions, informs the new rural construction campaign and safeguards agricultural development in southern China.
RI PT
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
Fig. 1. Typical traditional folk house plan (adapted from [19])
AC C
EP
Fig.2. Village street I (the entrance of house is on the front) (adapted from [20, 21])
Fig.3. Village street II (the entrance of house is on both sides) (adapted from [20, 21])
ACCEPTED MANUSCRIPT 2 Methods 2.1 Location and climate
TE D
M AN U
SC
RI PT
The Guangdong Province is a large province located in southern China, and a majority of the province has a hot summer and warm winter climate as shown in Fig. 4. The annual mean air temperature and relative humidity are 10 - 24 °C and 60 - 80%, respectively, and the monthly mean air temperatures are 28 - 29 °C in July and 9 - 16 °C in January. Eleven rural areas were selected from the eastern, western and northern parts of the province, as shown in Fig. 2. They are located at a great distance from central urban areas and have similar climates. The mean air temperature and relative humidity, respectively, were 29.6 ± 0.4 °C and 79 ± 1% in the hottest month of the year, June, and 14.7 ± 0.8 °C and 75 ± 3% in the coldest month, January [32].
Fig. 4. Locations of the rural areas
2.2 Houses and spaces
AC C
EP
Modern folk houses in rural areas evolved mostly from traditional folk houses, and the semi-open spaces of halls and patios have been well retained as shown in Fig. 1. The semi-open hall is usually one-story with a high ceiling, and open at the front to ensure good ventilation and day lighting. The wall materials included bricks and stone, and the roof materials included wood and tile. In the interior spaces like bedrooms, the windows were small to prevent overheat by solar radiation. 2.3 Time of survey
To fully understand the thermal environments and thermal responses of people in the interior and semi-open spaces of rural folk houses, a survey was conducted throughout the year in 2015. The surveys were administered with the help of college students who came from the surveyed rural areas; therefore, the time in which the surveys were completed was mainly during the holidays of various seasons, including the summer and winter vacations for Tomb Sweeping Day, Labor Day, the Dragon Boat Festival, National Day, and the Mid-Autumn Festival.
ACCEPTED MANUSCRIPT
RI PT
The summer is long, hot and humid, and the winter is short and temperate. There is never any cold weather in the hot-humid areas. The survey data were divided accordingly into two groups: the summer group and the non-summer group. According to the standard QX/T 152 (2012) [33], the summer starts when the mean outdoor air temperature is over 22 °C for 5 consecutive days, and it ends when the mean outdoor air temperature is below 22 °C for 5 consecutive days. Therefore, based on analysis of the 2015 meteorological data, the summer began on April 24 and ended on November 21. 2.4 Respondents
M AN U
SC
The respondents were randomly selected, without any limits on gender, age or location (in interior or semi-open spaces). A total of 1657 sets of raw data was obtained, 51.5% of which were from male respondents and 48.5% were from female respondents. A large proportion (62.2%) of respondents were between the ages of 20 to 50 years old, while only a small proportion (5.4%) of respondents were less than 10 years old and more than 70 years old, as Table 1 shows. The sample size was 1369 for interior spaces and 288 for semi-open spaces. Table 1 Age distribution of respondents. Age group (years)
0 - 10
Number Proportion (%)
10 0.6
*
10 - 20
20 - 30
30 - 40
40 - 50
50 - 60
60 - 70
> 70
239 14.4
330 20.0
321 19.4
378 22.8
189 11.4
111 6.7
79 4.8
*including the upper limit value
Gender Number Height (cm)
Interior space
Weight (kg) BMI (kg/m²) a
AD (m2)
(clo)
Male
359
169.0 ± 5.9
60.5 ± 7.0
21.1 ± 0.8
1.69 ± 0.12
0.31 ± 0.12
Female
382
157.9 ± 5.4
51.7 ± 6.7
21.9 ± 2.0
1.50 ± 0.10
0.35 ± 0.13
All
741
163.7 ± 7.7
56.0 ± 7.6
21.3 ± 1.6
1.60 ± 0.13
0.32 ± 0.12
71
168.8 ± 4.8
59.8 ± 5.0
21.2 ± 0.8
1.68 ± 0.08
0.29 ± 0.10
71
158.3 ± 5.0
51.6 ± 4.9
21.6 ± 2.1
1.51 ± 0.07
0.32 ± 0.12
142
163.5 ± 7.2
55.7 ± 6.4
21.4 ± 1.6
1.59 ± 0.12
0.31 ± 0.11
AC C
Space
EP
TE D
Basic information on the respondents is presented for the summer and non-summer groups and for the interior and semi-open spaces categories, as shown in Tables 2 and 3. The t-test indicated that the respondents’ details were not significantly different between the interior and semi-open spaces categories for the summer group (p > 0.313). For the non-summer group, the height, body weight, body surface area, and clothing insulation were not significantly different (p > 0.272), while the BMI of the respondents was significantly lower in the interior spaces category (p < 0.0001). Table 2 Basic information on respondents in summer.
Mae Semi-open Female space All a
2
2
Notes: Body mass index (BMI) = weight/height (kg/m ） AD: DuBois surface area : The clothing insulation Table 3 Basic information on respondents in non-summer. Space
Gender Number
Height (cm)
Weight (kg) BMI (kg/m²) a
AD (m2)
(clo)
ACCEPTED MANUSCRIPT
Semi-o Mae pen Female space All
344
168.5 ± 6.3
61.6 ± 9.0
21.0 ± 0.6
1.70 ± 0.14
0.63 ± 0.25
284
157.4 ± 5.1
51.0 ± 7.3
21.4 ± 1.1
1.49 ± 0.11
0.70 ± 0.30
628
163.5 ± 8.3
57.0 ± 9.5
21.4 ± 1.1
1.60 ± 0.16
0.65 ± 0.27
79
168.9 ± 7.5
62.8 ± 8.4
21.4 ± 1.2
1.71 ± 0.15
0.62 ± 0.23
67
158.3 ± 4.6
52.1 ± 6.5
22.8 ± 1.0
1.51 ± 0.09
0.62 ± 0.23
146
164.0 ± 8.3
57.9 ± 9.3
22.0 ± 1.3
1.62 ± 0.16
0.62 ± 0.23
a
2
2
Notes: Body mass index (BMI) = weight/height (kg/m ） AD: DuBois surface area : The clothing insulation 2.5 Field survey methods
RI PT
Male Interior Female space All
TE D
M AN U
SC
The field survey was conducted by the college students that resided in the rural areas. Upon approval, the students visited the respondents’ houses. After observing the layout of the house and the position of the respondents, the students installed the instruments close to the respondents and recorded the data after 5 - 10 minutes when the readings were stable. The questionnaire was conducted at the same time as the data recording, during which the students read the questions one by one and recorded the responses of the respondents on paper (Fig. 5).
EP
Fig. 5. Field survey
AC C
The main contents of the questionnaire include the following: 1) personal information, such as gender, age, height, weight, and time living at current location; 2) personal factors, such as clothing, current activity and activity within 20 minutes prior to beginning the survey; and 3) subjective responses, such as thermal sensation, comfort, acceptability and preference. The length of time living in the area was used to confirm that the respondent had acclimated to the local climate. Respondents’ clothing type was recorded in detail during the survey, and the insulation of the ensemble was estimated as the sum of the individual values listed in the ASHRAE standard 55 [34]. Since chairs in the rural areas were mostly wooden, the additional thermal insulation value for a chair was set to zero. Some of the respondents engaged in heavy physical work just before the survey, which may have affected their comfort perceptions. Therefore, we excluded those samples, and retained the sample that respondents engaged in sedentary or light activities. The metabolic rate of the respondents was estimated to be 1.2 met, considering both the current and prior activities.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Fig. 6 shows the rating scales used for the subjective responses in the survey. Considering the hot and humid climate, a nine-point extended scale from ISO 15001 (ISO, 2002) was used for the thermal sensation rating. Traditional scales were used for thermal comfort and acceptability ratings. The questionnaire was written in Chinese, and the Chinese wording for the scale degrees from the Chinese national standard GB/T 18977 (AQSIQ, 2003) was used.
Fig. 6 Scales used for thermal response in the survey
=
(1)
EP
TE D
Thermal environment parameters of air temperature (ta), globe temperature (tg), air speed (v) and relative humidity (RH) were measured by the HD 32.3 instrument. The measuring range and accuracy agreed with ISO 7726, as shown in Table 4. The measuring point was set up near the respondents, and it was placed to avoid obvious heat or cold sources in accordance with the ASHRAE standard 55. The height of the sensors was 0.6 m (sedentary) or 1.1 m (standing) above the floor. The operative temperature (top) was calculated according to the ASHRAE Handbook-Fundamentals.
AC C
Measurement content Air temperature Relative humidity Air speed
Globe temperature
Table 4 Instrument parameters. Operation range Accuracy -10 to 80 °C ± 0.2 °C 5% - 98% ± 2% (15 - 90%) ± 2.5% (other scope) 0.05 - 5 m/s ± 0.05 m/s (0 - 0.99 m/s) ± 0.15 m/s (1 - 5 m/s) -10 to 100 °C ± 0.2 °C
2.6 Data process and analysis An independent sample t-test was used to test the environmental differences between interior and semi-open spaces. Linear regression was adopted to derive the relationship between the operative temperature and the thermal sensation, and covariance analysis was used to test the significant differences between various regression lines. Polynomial
ACCEPTED MANUSCRIPT regression was used to analyze the quantitative relationship between the operative temperature and the percentage dissatisfied. In addition, the “bin” method was repeatedly used in this study, i.e., the data were divided into bins by operative temperature in 0.5 °C steps, the values of the variables were averaged for each bin, and the averages were analyzed by their weighted sample size. All the statistical analyses were done with SPSS v 22.0 software (IBM, New York, NY, US), and all the differences were accepted as significant at a 0.05 level.
RI PT
3 Results 3.1 Thermal environment
M AN U
SC
Table 5 shows the thermal environment results in the summer for the interior and semi-open spaces. Compared with those of the interior spaces, no significant differences were found in the semi-open spaces in air temperature, mean radiant temperature and relative humidity (p > 0.096), while air speed was significantly greater (p = 0.014). The detailed air speed distribution is shown in Fig. 7. The proportion of air speed below 0.4 m/s accounted for 58.4% in the interior spaces, while the proportion of air speed between 0.2 - 0.7 m/s accounted for 50.7% in the semi-open spaces. Table 5 Thermal environment parameters in summer. Interior space (°C) (°C)
(m/s)
RH (%)
(°C)
Semi-open space
(°C)
(m/s)
RH (%)
29.5
29.7
0.54
76.1
29.9
30.1
0.68
74.7
Standard deviation
3.0
3.0
0.61
9.8
2.5
2.5
0.63
8.8
Minimum
18.1
18.2
0.01
48.7
18.8
19
0.01
45.4
Maximum
38.3
38.9
2.93
99.9
39
37.8
2.90
99.9
AC C
EP
TE D
Mean
Fig. 7. Air speed distribution in summer Table 6 shows the thermal environment results in the non-summer. Compared with those of the interior spaces, no significant differences were found in the semi-open spaces for air
ACCEPTED MANUSCRIPT
RI PT
temperature or mean radiant temperature (p > 0.058), while relative humidity was significantly lower (p < 0.0001), and air speed was significantly greater (p < 0.0001). From the relative humidity distribution (see Fig. 8), it can be seen that the proportion of relative humidity between 45% - 65% was larger, while the proportion of relative humidity between 65% - 90% was smaller in the semi-open spaces. From the air speed distribution (see Fig. 9), the air speed was mostly below 0.5 m/s in the interior spaces, while it was 0.3 -1.1 m/s in the semi-open spaces. It can also be seen from Tables 5 and 6 that the standard deviations of thermal environment parameters were quite similar between the interior and semi-open spaces. The standard deviation of air speed was relatively greater than the other parameters, indicating that air speed was more variable in the rural houses.
Interior space (°C) (°C)
Semi-open space
RH (%)
(°C)
(°C)
(m/s) RH (%)
21.5
21.4
0.18
70.2
22
22.2
0.38
65.2
3.7
3.8
0.31
12.3
4.3
4.4
0.38
13.4
Minimum
13.4
13.5
0.01
26.1
12.4
12.7
0.01
28.1
Maximum
32.3
32.5
2.78
99.9
30.4
31.1
2.96
99.9
AC C
EP
TE D
Standard deviation
M AN U
Mean
(m/s)
SC
Table 6 Thermal environment parameters in non-summer.
Fig. 8. Relative humidity distribution in non-summer
SC
RI PT
ACCEPTED MANUSCRIPT
M AN U
Fig. 9. Air speed distribution in non-summer 3.2 Clothing insulation
AC C
EP
TE D
In rural areas, clothing adjustment is one of the most important thermal adaptation methods, and people are accustomed to adding or removing clothes to meet their thermal comfort requirements. It was hot and humid during the summer data collection, and rural residents generally wore short-sleeved shirts, shorts and sandals. The relationship between the operative temperature and clothing insulation in summer is shown in Fig. 10. When the operative temperature was below 28 °C, the clothing insulation decreased with increasing temperature, and when the operative temperature was above 28 °C, the clothing insulation was basically stable at the same level. The stable clothing insulations were 0.30 ± 0.09 clo and 0.29 ± 0.09 clo in the interior and semi-open spaces, respectively, and they were not significantly different (p = 0.813).
Fig. 10. Clothing insulation changes with top in summer The clothing insulation varied greatly between 0.27 clo - 1.2 clo and changed linearly with the operative temperature in the non-summer group, as shown in Fig. 11. The higher the
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
operative temperature, the smaller the clothing insulation was. The regression equations for the interior and semi-open spaces are shown below: Icl = -0.040 top + 1.517 (R2 = 0.808) (Interior space) (2) 2 (Semi-open space) (3) Icl = -0.026 top + 1.193 (R = 0.524) A significant difference (analysis of covariance, p = 0.003) was found between the two regression lines, indicating that the residents changed clothing less sensitively in the semi-open spaces.
Fig. 11. Clothing insulation changes with top in non-summer
TE D
3.3 Thermal sensation
AC C
EP
Good linear relationships between thermal sensation (TS) and operative temperature were established in the summer for the interior and semi-open spaces, as shown in Fig. 12 and the following equations: TS = 0.232 top - 5.579 (R² = 0.722) (Interior space) (4) TS = 0.255 top - 5.976 (R² = 0.652) (Semi-open space) (5) -1 The thermal sensitivity, namely, the slope of the regression line, was 0.232 °C for the interior spaces and 0.255 °C-1 for the semi-open spaces; that is, for each 1 °C increase of top, thermal sensation increased by 0.232 and 0.255 units of scale for the interior spaces and semi-open spaces, respectively. No significant difference was found between the slopes of the two regression lines (analysis of covariance, p = 0.701). The summer thermal neutral temperature was determined to be 24.0 °C for the interior spaces and 23.4 °C for the semi-open spaces using Eqs. (4) and (5), and the latter was 0.6 °C lower than the former. Additionally, the size was smaller and the points were much more scattered for the data with operative temperatures below 26 °C.
RI PT
ACCEPTED MANUSCRIPT
SC
Fig. 12. The relationship between thermal sensation votes and top in summer
AC C
EP
TE D
M AN U
Good linear relationships between thermal sensation and operative temperature were established in the non-summer sample, as shown in Fig. 13 and the following equations: TS = 0.238 top - 4.700 (R² = 0.841) (Interior space) (6) (Semi-open space) (7) TS = 0.200 top - 3.673 (R² = 0.670) -1 -1 The thermal sensitivity was 0.238 °C in the interior spaces and 0.200 °C in the semi-open spaces, and no significant difference was found between the two (analysis of covariance, p = 0.246). The non-summer thermal neutral temperatures were determined to be 19.7 °C and 18.4 °C for the interior and semi-open spaces, respectively, and the latter was 1.3 °C lower than the former.
Fig. 13. The relationship of thermal sensation voting and top in non-summer
3.4 Percentage dissatisfied The percentage dissatisfied was the percentage of respondents who reported a negative thermal acceptability vote. Second-order polynomial functions relating the percentage dissatisfied (PD) and the operative temperature were established in the summer for the interior and semi-open spaces, as shown in Fig. 14 and the following equations: (Interior space) (8) PD = 0.326 top2 -15.144 top + 175.884 (R2 = 0.621) 2 2 PD = 0.363 top -16.328 top + 181.641 (R = 0.415) (Semi-open space) (9)
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
The percentage dissatisfied increased with temperature when the operative temperature was higher than 26 °C. The upper limit of the 80% thermal acceptable operative temperature was 31.1 °C for the interior spaces and 30.3 °C for the semi-open spaces, and the latter was 0.8 °C lower than the former.
Fig. 14. Percentage dissatisfied changes with top in summer
AC C
EP
TE D
The results for the non-summer group are shown in Fig. 15 and the following equations: PD = 0.218 top2 - 9.264 top + 99.118 (R2 = 0.395) (Interior space) (10) PD = 0.352 top2 - 13.411 top + 129.999 (R2 = 0.322) (Semi-open space) (11) Only weak relationships between the percentage dissatisfied and the operative temperature were found in non-summer group. When the operative temperature was above 21 °C, the percentage dissatisfied increased with the temperature, and this trend was more obvious in the semi-open spaces. The upper limit of the 80% thermal acceptable operative temperature was 30.7 °C for the interior spaces and 26.0 °C for the semi-open spaces, and the latter was 4.7 °C lower than the former. The percentage dissatisfied was below 20% for most of the data when the operative temperature was below 21 °C, indicating that there was no lower temperature limit.
Fig. 15. Percentage dissatisfied changes with top in non-summer
ACCEPTED MANUSCRIPT 3.5 Thermal preference
M AN U
SC
RI PT
The votes for the "no change" preference were equally divided into the votes expecting "cooler" and "warmer", and the relationship between the proportions of "cooler" and "warmer" preferences and operative temperature was obtained, as shown in Figs. 14 and 15. In the summer, the thermal preference changed obviously with the operative temperature only when the operative temperature was above 25 °C (Fig. 16). Probit regression was thus used for those data, showing that the preferred operative temperatures were 25.6 °C and 25.3 °C for the interior and semi-open spaces, respectively. These values were similar to each other and were higher than the thermal neutral temperatures in the spaces in the summer.
(a) In interior space (b) In semi-open space Fig. 16. Thermal preference changes with top in summer
AC C
EP
TE D
In the non-summer season, the thermal preference proportions changed obviously with operative temperature for all temperature ranges (Fig. 17). Preferred temperatures were determined to be 21.4 °C and 20.8 °C for the interior and semi-open spaces, respectively. These values were similar to each other and were higher than the thermal neutral temperatures in the spaces in the non-summer group.
(a) In interior space (b) In semi-open space Fig. 17. Thermal preference changes with top in non-summer 3.6 Airflow preference By using the same method for thermal preference, the relationship between the percentage
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
of airflow preference and the operative temperature was obtained, as shown in Figs. 18 and 19. In the summer, the proportion of respondents wanting “more airflow” increased rapidly with increasing temperature when the operative temperature was above 20 °C for the interior spaces and when the operative temperature was above 26 °C for the semi-open spaces. In the non-summer, the proportion of respondents wanting “more airflow” increased obviously with temperature when the operative temperature was above 24 °C for the interior spaces and above 28 °C for the semi-open spaces.
AC C
EP
TE D
(a) In interior space (b) In semi-open space Fig. 18. Airflow preference changes with top in summer
(a) In interior space (b) In semi-open space Fig. 19. Airflow preference changes with top in non-summer
4 Discussion
4.1 Interior vs semi-open spaces The results show that the thermal neutral temperature of residents in the semi-open spaces was lower than that in the interior spaces, both in the summer (0.6 °C) and non-summer (1.3 °C) seasons. Following the same trend, the results show that compared with those of the interior spaces, the upper limit of the 80% thermal acceptable range was offset toward a lower temperature by 0.8 °C in the summer and by 4.7 °C in the non-summer for the semi-open
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
spaces. These findings are different from the findings of Rijal et al., who conducted a 40-day field investigation of interior and semi-open spaces in five regions of Nepal during the summer and winter seasons and found that the thermal neutral temperature of the residents in semi-open spaces was higher than that of the residents in interior spaces [24]. Thermal neutral and acceptable temperatures are affected by many factors, including physical, physiological and psychological factors. Compared with those of the interior spaces, air speed was greater and relative humidity was lower, while the other parameters were the same in the semi-open spaces. According to the heat balance principle, the human body can maintain thermal neutrality at a higher operative temperature under the conditions of higher air speed or lower relative humidity. The present results differ from this, and it is therefore reasonable to conclude that physical factors are not relevant to the present findings. Literature on heat acclimatization indicates that changes in the human thermoregulatory mechanism can be triggered by repeated exposure to extreme hot and humid conditions. This is obviously not the case in the present study. Therefore, psychological factors are the main reasons for the differences in neutral and acceptable temperatures between the two spaces. Studies have claimed that the indoor thermal history in all seasons of a year played a key role in shaping the subjects’ sensations in a wide range of thermal conditions [35]. In this study, the times that the respondents stayed in the interior and semi-open spaces were approximately equal, and the respondents’ experiences with the two spaces contributed equally to the indoor thermal history. A previous study indicated that perceived control had a marked effect on comfort and satisfaction [36]. One reason for the present findings could be that the residents had more adaptive opportunities in the interior spaces, such as using fans and moving from one room to another, and hence could accept the higher temperatures. In addition, there was a large difference in lighting between the two environments; it was dark in the interior spaces and bright in the semi-open spaces, and the interaction between light exposure and thermal perceptions may have impacted the results as well.
EP
4.2 Summer vs non-summer
AC C
The difference between the summer and non-summer samples in the thermal sensation and operative temperature relationship was observed by transferring Figs. 12 and 13 to Fig. 20. In both the interior and semi-open spaces, residents’ thermal sensations in the summer season were always lower than they were in the non-summer season. This finding may be caused by clothing adjustment. According to Figs. 10 and 11, the residents’ clothing insulation was lower in the summer than in the non-summer, and therefore their thermal sensation was lower.
RI PT
ACCEPTED MANUSCRIPT
(a) In interior space
(b) In semi-open space
SC
Fig. 20. The relationship between thermal sensation and operative temperature with seasons
TE D
4.3 Rural vs urban residents
M AN U
Eqs. (3) and (5) show that the thermal sensitivity of residents in the interior spaces was 0.232 °C-1 in the summer and 0.238 °C-1 in the non-summer. The same findings were found for the semi-open spaces, showing that the summer thermal sensitivity was 0.255 °C-1 and the non-summer thermal sensitivity was 0.200 °C-1. Further analysis indicated that thermal sensitivity was not significantly different in either interior spaces or in semi-open spaces (analysis of covariance, p > 0.262). In summary, the thermal sensitivity of the rural residents did not change with the seasons regardless of whether they were in interior or semi-open spaces.
AC C
EP
Zhang et al. [37] and Chen et al. [38] conducted field studies on the thermal comfort of urban residents in hot-humid areas. The survey was conducted from May to October, which was consistent with the summer season in this study. With the permission of the authors, the raw data of the urban residents were collected, and a good linear relationship between thermal sensation and operative temperature was obtained as follows: TS = 0.396 top - 10.626 (R² = 0.949) (12) The equation shows that the thermal neutral temperature was 26.8 °C and the thermal sensitivity was 0.396 °C-1 for the urban residents, which were both higher than those of the rural residents in the summer. A good second-order polynomial function was established relating percentage dissatisfied with operative temperature as follows: PD = 1.477 top2 - 78.806 top + 1054.157 (R² = 0.965) (13) The equation shows that the upper limit of the 80% acceptable operative temperature was 30.1 °C for the urban residents, which was lower than that of the rural residents by 0.2-1 °C. A recent climatic chamber study in hot-humid areas found that rural subjects felt more comfortable and were more satisfied under identical cool or warm conditions and that they had a much wider acceptable temperature range than the urban subjects [39]. This finding is consistent with the present findings. The possible reasons for these findings were also proposed as differences in local culture, expectations and environmental cognition between the rural and urban residents.
ACCEPTED MANUSCRIPT 4.4 Practical implication
AC C
EP
TE D
M AN U
SC
RI PT
Many studies have revealed that residents in the traditional buildings were more satisfied with the thermal environment than those in the modern buildings. For example, Dili confirmed that in Kerala the traditional residential buildings were more effective in providing comfortable indoor environment than the modern buildings [40]; Nematchoua conducted a survey in five towns of four climate zones in Cameroon and reported that the traditional buildings were more comfortable than the modern buildings [41]; and Subramanian reported that 88% of the residents in traditional buildings feel comfortable with indoor air temperature in Thanjavur, while the proportion was only 54% for modern buildings [42]. Nevertheless, the above studies fail to provide the reasons behind the higher satisfaction in traditional buildings: due to architectural passive design or human thermal adaptation. On the other hand, at the initial phase of architectural design, architects used to determine the passive and climate responsive strategies by using the Givoni's or Olgyay's bioclimatic charts. However, the thermal comfort requirements of people from various areas might be different due to thermal history, lifestyle or culture background. This difference has not yet been reflected in the current bioclimatic design. This study found that, in the hot and humid areas of southern China, the thermal neutral and acceptable temperatures were significantly different between the rural and urban people, and the upper limit of 80% acceptable temperature was higher by 0.2-1℃ for the rural people in summer. This finding provides an important clue for explaining the higher satisfaction of residents in traditional buildings, and presents as a key basis for bioclimatic design of both rural and urban buildings. In addition, this study also found that the residents’ thermal demands were quite difference in the two kinds of spaces in the rural buildings. Compared to those of the interior spaces, both the thermal neutral temperature and upper limit of acceptable temperatures were lower in the semi-open spaces. This provides valuable indications that residents were more rigorous to the semi-open spaces, and architects need to design the interior and semi-open spaces based on their own thermal comfort standard. In conclusion, this paper clarifies the varied thermal comfort standards for the rural and urban areas and the interior and semi-open spaces, supports the design of various types of buildings and spaces, and greatly helps to improve their indoor environmental qualities. 5 Conclusions
A year-long thermal comfort field study was conducted in rural folk houses in the hot-humid areas of China. A total of 1657 sets of raw data that included both thermal environment parameters and respondents’ subjective responses were collected regarding interior and semi-open spaces. The main conclusions are shown below. 1. Air speed was greater and relative humidity was lower in the semi-open spaces. 2. The residents’ clothing insulation varied with indoor operative temperature from 0.27 clo to 1.2 clo in the non-summer season, and it was steady at 0.30 clo during the summer season when indoor operative temperature was above 28 °C. The clothing change with temperature was less sensitive in the semi-open spaces.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
3. In the interior spaces, the thermal neutral, upper limit of the 80% acceptable and preferred operative temperatures, respectively, were determined to be 24.0 °C, 31.1 °C and 25.6 °C in the summer and 19.7 °C, 30.7 °C and 21.4 °C in the non-summer. 4. In the semi-open spaces, the thermal neutral, upper limit of the 80% acceptable and preferred operative temperatures, respectively, were determined to be 23.4 °C, 30.3 °C and 25.3 °C in the summer and 18.4 °C, 26.0 °C and 20.8 °C in the non-summer. 5. Compared to the interior spaces, the thermal neutral temperature was lower by 0.6 °C in the summer and 1.3 °C in the non-summer, and the upper limit of 80% acceptable temperature decreased by 0.8 °C in the summer and 4.7 °C in the non-summer in the semi-open spaces. The differences might be due to the decrease in adaptive opportunities and the brighter lighting environment of the semi-open spaces. 6. Compared to the non-summer samples, the thermal sensitivity remained the same, and the thermal neutral temperature was much higher in the summer. This difference might be caused by decreased clothing insulation in the summer. 7. Compared to the urban residents, the upper limit of the 80% acceptable temperature was 0.2 -1 °C higher for the rural residents. This finding is consistent with the present findings and might be caused by differences in local culture, expectations and environmental cognition between rural and urban residents. Acknowledgements
References
TE D
This work was supported by the National Natural Science Foundation of China under Grant (No. 50708038，No.51708228); the Program for New Century Excellent Talents in University under Grant (No. NCET-10-0373).
AC C
EP
[1] Fernandes J, Pimenta C, Mateus R, et al. Contribution of Portuguese Vernacular Building Strategies to Indoor Thermal Comfort and Occupants' Perception. Buildings, 2015, 2015(5):1242-1264. [2] Manioğlu. G, Yılmaz.Z. Energy efficient design strategies in the hot dry area of Turkey. Building & Environment, 2008, 43(7):1301-1309. [3] Baran M, Yıldırım M, Yılmaz A. Evaluation of ecological design strategies in traditional houses in Diyarbakir, Turkey. Journal of Cleaner Production, 2011, 19(6–7):609-619. [4] Michael A, Demosthenous D, Philokyprou M. Natural Ventilation for Cooling in Mediterranean Climate: A Case Study in Vernacular Architecture of Cyprus. Energy & Buildings, 2017, 144. [5] Hidayahtuljamilah N, Kassim N, Zafrullah H, et al. Re-adaptation of Malay vernacular architecture thermal comfort elements:towards sustainable design in Malasia. (A literature review). International Conference on Innovation and Technology for Sustainable Built Environment. 2012. [6] Foruzanmehr D. Summer-time Thermal Comfort in Vernacular Earth Dwellings in Yazd, Iran. International Journal of Sustainable Design, 2012, 2(1). [7] Toe D, Kubota T. Comparative assessment of vernacular passive cooling techniques for improving indoor thermal comfort of modern terraced houses in hot–humid climate of Malaysia. Solar Energy, 2015, 114:229-258. [8] Kubota T, Toe D. Application of Passive Cooling Techniques in Vernacular Houses to Modern Urban Houses: A Case Study of Malaysia. Procedia - Social & Behavioral Sciences, 2015, 179:29-39.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[9] Liu W, Zheng Y, Deng Q, et al. Human thermal adaptive behaviour in naturally ventilated offices for different outdoor air temperatures: A case study in Changsha China. Building & Environment, 2012, 50:76-89. [10] Yao R, Liu J, Li B. Occupants’ adaptive responses and perception of thermal environment in naturally conditioned university classrooms. Applied Energy, 2010, 87(3):1015-1022. [11] Wang Z, Zhang L, Zhao J, He Y. Thermal comfort for naturally ventilated residential buildings in Harbin. Energy & Buildings, 2010, 42:2406-15. [12] Zhang Y, Wang J, Chen H, et al. Thermal comfort in naturally ventilated buildings in hot-humid area of China. Heating Ventilating & Air Conditioning, 2011, 45(11):2562-2570. [13] Zhang Y, Chen H, Meng Q. Thermal comfort in buildings with split air-conditioners in hot-humid area of China. Building & Environment, 2013, 64(6):213-224. [14] Yan H, Yang L, Zheng W, et al. Analysis of behaviour patterns and thermal responses to a hot-arid climate in rural China. Journal of Thermal Biology, 2016, 59:92-102. [15] Du X, Bokel R, Dobbelsteen A. Building microclimate and summer thermal comfort in free-running buildings with diverse spaces: A Chinese vernacular house case. Building & Environment, 2014, 82:215-227. [16] Sheng X. Filed test and analysis of thermal comfort in rural house and urban residential building in sever cold region. Harbin,China: Harbin Institute of Technology; 2013. [17] Huang L, Zhu Y, Ouyang Q, Cao B. Field survey of indoor thermal comfort in rural housing of northern China in heating season. Journal of Southeast University (English Edition), 2010, 26(2):169-72. [18] Liu Y, Qian Y, Yan H, et al. Field study on thermal comfort of rural houses in winter in a the Guanzhong region,Shaanxi Province. Journal of Xian University of Architecture & Technology, 2011, 43(4):551-556. [19] Lu Q. Guangdong folk house. Beijing: China Architecture & Building Press, 2008. [20] Chen J. Research on Natural Ventilation System of Traditional Village in Chaoshan Area.Guanghzou, China: South China university of Technology; 2014. [21] Hui X. Study on climate adaptation of cold lane-courtyard space system in Guangfu traditonal villages. Guanghzou, China: South China university of Technology; 2016. [22] Huang Y, Ma J. Analysis of Measured Thermal Environment of Traditional Architecture Lengxiang in Lingnan. Energy Conservation in Buildings, 2014, 6 (42): 75-78. [23] Xie H. Naturally Ventilation of Lingnan Residential Houses. Small Town Construction, 2007, 12: 58-62 (in Chinese). [24] Zhou X, Ma J, Zhou X, Measured Analysis of Lingnan Residential Houses Indoor Thermal Environment. Heating, Ventilation & Air Conditioning, 2011, 41 (5): 94-97 (in Chinese). [25] Rijal HB, Yoshida H, Umemiya N. Seasonal and regional differences in neutral temperatures in Nepalese traditional vernacular houses. Building & Environment, 2010, 45(12):2743-2753. [26] Ryu Y, Kim S, Lee D. The influence of wind flows on thermal comfort in the Daechung, of a traditional Korean house. Building & Environment, 2009, 44(1):18-26. [27] Lu S, Fang K, Qi Y, et al. Influence of Natural Ventilation on Thermal Comfort in Semi-open Building under Early Summer Climate in the Area of Tropical Island . Procedia Engineering, 2015, 121:944-951. [28] Hwang R, Lin T. Thermal Comfort Requirements for Occupants of Semi-Outdoor and Outdoor Environments in Hot-Humid Regions. Architectural Science Review, 2007, 50(4):357-364.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[29] Spagnolo J, Dear RD. A field study of thermal comfort in outdoor and semi-outdoor environments in subtropical Sydney Australia. Building & Environment, 2003, 38(5):721-738. [30] Nakano J, Tanabe S I. Thermal comfort and adaptation in semi-outdoor environments. Ashrae Transactions, 2004, 110:543-553. [31] Jin L, Meng Q, Zhao L. Indoor Environment and Thermal Comfort in Rural Houses in East Guangdong of China. Journal of Civil, Architectural & Environmental Engineering, 2013, 35 (2):105-12(in Chinese). [32] China Meteorological Administation.China Meteorological Data Service Center. http://data.cma.cn/site/index.html. [33] China Meteorological Administration.Division of climatic season, QX/T 152. Beijing, China, 2012 (in Chinese). [34] ASHRAE. Thermal enviromental conditions for human occupancy, Standard 55. Atlanta, USA, 2004. [35] Zhang Y, Chen H, Wang J, et al. Thermal comfort of people in the hot and humid area of China—impacts of season, climate, and thermal history. Indoor Air, 2016, 26(5):820-830. [36] Paciuk M. The role of personal control of the environment in thermal comfort and satisfaction at the workplace. Coming of age, EDRA 21. Oklahoma city. OK, USA, 1990, 303-12. [37] Zhang Y, Liu Q, Meng Q. Airflow utilization in buildings in hot and humid areas of China. Building & Environment, 2015, 87:207-214. [38] Chen H, Zhang Y, Wang J, Meng Q, Research on Summer Thermal Comfort of Naturally Ventilation Buildings in Hot and Humid Region of China - Taking Guangzhou as an Example. Heating, Ventilation & Air Conditioning, 2010.40(2): 96-101 (in Chinese). [39] Zhang Z, Zhang Y, Jin L. Thermal comfort of rural residents in a hot–humid area. Building Research & Information, 2017, 45(1-2):209-221. [40] Dili A, Naseer M, Varghese T. Thermal comfort study of Kerala traditional residential buildings based on questionnaire survey among occupants of traditional and modern buildings. Energy & Buildings, 2010, 42(11):2139-2150. [41] Nematchoua M, Tchinda R, Orosa J. Thermal comfort and energy consumption in modern versus traditional buildings in Cameroon: A questionnaire-based statistical study. Applied Energy, 2014, 114(2):687-699. [42] Subramanian C, Ramachandran N, Kumar S. Performance evaluatuion of traditonal and mordern residential buildings for thermal comfort by questionare survey in Thanjavur. International Journal of Civil Engineering & Technology, 2017, 8(82):440-451.
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
Thermal sensitivity of people kept the same regardless of spaces and seasons. Thermal neutral temperature was lower in semi-open spaces than interior spaces. Acceptable temperature range was narrower in semi-open spaces than interior spaces. Acceptable temperature range of rural residents was wider than urban residents.