Investigation on Thermal Comfort and Thermal Adaptive Behaviors of Rural Residents in Suibin Town, China, in Summer

Abstract

With global warming, the climate is becoming hotter, even at higher latitudes. In order to clarify the thermal environment and thermal comfort level of rural residents at higher latitudes in summer, a field survey on thermal comfort was conducted in Suibin Town, China. The results show the following: (1) The correlation between the operative temperature and the mean thermal sensation score is stronger than that between the operative temperature and the thermal sensation score. Moreover, the applicability of the thermal comfort evaluation index of the operative temperature in severely cold areas was verified. The linear regression method showed that the neutral temperature is 25.1 °C and the thermal acceptable range is 22.5–27.6 °C. (2) There is a strong correlation between thermal sensation and thermal acceptability. (3) The preferred temperature in summer is 25.3 °C. Moreover, rural residents prefer cooler indoor temperatures in summer. (4) The clothing insulation of rural residents decreases with an increase in the indoor operative temperature. (5) Rural residents’ acceptance of the indoor temperature in summer is influenced by the economy, psychology, adaptive behaviors, etc. Moreover, rural residents expect cooler indoor temperatures in summer and will adapt to the thermal environment via low-cost adaptive behaviors.

1. Introduction

In September 2018, the Central Committee of the Communist Party of China and the State Council issued the “Strategic Plan for Rural Revitalization (2018–2022)”, which identified villages as regional complexes with natural, social and economic characteristics and multiple functions, such as production, living, ecology and culture. They promote one another and coexist with cities and towns, which together provide the main spaces of human activities. Prosperity in the countryside makes the country prosperous and decline in the countryside makes the country decline. All localities and departments are required to implement the rural revitalization strategy in light of their respective situations. In the same year, Heilongjiang Province launched an investigation and compiled the “Outline of Development, Construction and Planning of Villages and Towns in Heilongjiang Province (2018–2030)”, emphasizing the acceleration of the renovation of dilapidated houses in rural areas and the comprehensive improvement of renovation standards. New (renovated) rural houses must have basic living functions and structures, and their functionality should be reasonable.

During the National People’s Congress in 2021, the government’s report referred to “peak carbon dioxide emissions” and “carbon neutrality”, advocating resource conservation, energy conservation, emission reductions and green and low carbon. With the development of China’s economy, the building sector’s energy consumption is also increasing. According to relevant statistics, as of 2018, the proportion of the building sector’s energy consumption out of China’s total energy consumption has reached 30% [1]. Building projects are a major consumer of resources and energy. In the context of large-scale planning, construction and rural revitalization, newly built and rebuilt houses in rural areas should also seek to be energy-saving and emission-reducing and use little amounts of predominantly green carbon. At the same time, it is necessary to ensure the good thermal comfort of residential buildings.

At present, a good and comfortable indoor environment is pursued at the cost of energy consumption. Reducing energy consumption is contradictory to obtaining a comfortable indoor environment. In the process of promoting the rural revitalization strategy, balancing energy saving in rural houses with improved indoor thermal comfort is an urgent problem that must be solved. Therefore, it is necessary to conduct field research on the thermal comfort and adaptive behaviors of rural residents in order to formulate an energy-saving and emission-reducing scheme that realizes reasonable indoor thermal comfort and comfortable indoor temperatures.

In 1970, Fanger developed a thermal balance equation of the human body according to the conditions of human thermal comfort and proposed the predicted mean vote (PMV) and the predicted percentage of dissatisfied (PPD) based on their thermal balance equation and ASHRAE’s seven-point thermal sensation index [2]. These two indexes were recommended by ISO 7730-1994 for use in thermal comfort evaluations [3]. In the study of human thermal comfort, the PMV index is based entirely on laboratory observations, in which the environmental parameters remain unchanged and the human body is passively stimulated. In addition, the different types of buildings are not considered, and the differences in people’s thermal self-regulation adaptabilities, adaptive behaviors and thermal experiences are ignored, which makes these standards limited. After a large number of field investigations and studies on thermal comfort had been conducted [4,5,6], it was concluded that “The international standard ISO 7730-1994 formulated by PMV index is obviously different from the actual field investigation and research on human thermal comfort”. There are differences between the results of field research and laboratory research on thermal comfort because, under laboratory conditions, the subject is assumed to passively accept changes in climate conditions, and the behavioral, psychological and physiological adaptive adjustments that one makes in a real environment are ignored. Joo-Young [7] confirmed the role of physiological adaptive regulation in human thermal comfort through research. Brager [8] confirmed the effect of psychological adaptive regulation on human thermal comfort through research. Other scholars [9] have confirmed, through surveys in Britain, Pakistan and other places, that self-regulation behaviors can help to improve satisfaction with the thermal environment. Therefore, studies on thermal comfort should be conducted on-site, taking into account human physiological, psychological and adaptive behavioral regulation (see Table 1).

In the research of many international scholars, Oikonomou [10] and others evaluated the thermal comfort of dwellings in northwest Greece via the aspects of architecture and bioclimate through literature analyses and field investigations in 2011. In 2014, Üllar Alev [11] and others measured and simulated the thermal environment of dwellings in Estonia, Finland and Sweden, and the influence of envelope performance on thermal comfort was studied. Nematchoua [12] and others analyzed the thermal comfort satisfaction of Cameroonian residents through a questionnaire survey and evaluated the thermal comfort of local dwellings. Ensieh Ghorbani Nia [13] investigated the architectural elements of “shikili” houses in northern Iran in 2014 and evaluated them from the perspective of bioclimate. In 2019, Zeyad Amin Al-Absi [14] and others studied the adaptive behavior of residents in relation to thermal comfort in high-rise residential building in Malaysia. In 2022, Naja Aqilah [15] and others summarized the thermal comfort of residential buildings and evaluated the current situation of adaptive thermal comfort research and the energy-saving potential of residential buildings. They determined the comfortable temperature ranges in residential buildings and clarified the influence of adaptive measures on residential energy-saving potential.

Table 1. Review of thermal comfort research from different regions.

No.	Authors	Year	Region	Gist
1	Zhu et al. [16]	2010	Rural houses in Yinchuan, China	1; 2
2	Huang et al. [17]	2011	Rural houses in Beijing, China	1; 4; 5
3	Yang et al. [18]	2011	Rural houses in Guanzhong area, China	1; 4; 5
4	Oikonomou et al. [10]	2011	Dwellings in Northwest Greece	1; 2
5	Jin et al. [19]	2013	Rural houses in eastern Guangdong, China	1; 4; 5
6	Ge et al. [20]	2013	Rural houses in Weifang, China	1; 4; 5
7	Alev et al. [11]	2014	Dwellings in Estonia, Finland and Sweden	1; 2
8	Nematchoua et al. [12]	2014	Dwellings in Cameroon	2
9	Nia [13]	2014	Houses in northern Iran	3
10	Wang et al. [21]	2014	Rural houses in Harbin, China	1; 4; 5
11	Zhang Qi et al. [22]	2014	Dwellings in Shanghai, Shandong, Xinjiang, Liaoning and Guangxi, China	1; 4; 5; 6
12	Liu et al. [23]	2016	Dwellings in Khampa Tibetan area, China	1; 2
13	Xiong et al. [24]	2016	Rural houses in Hubei, China	1; 2; 7; 8
14	Liu et al. [25]	2018	Rural houses in Central Hainan, China	1; 8
15	Pang et al. [26]	2018	Rural houses in Beihai, China	1; 5; 7
16	Li et al. [27]	2018	Rural houses in western Liaoling, China	2
17	Al-Absi [14]	2019	High-rise residential building in Malaysia	8
18	Ni Sijia [28]	2020	Rural houses in Central Jiangxi, China	2; 7

1. Field investigation. 2. Thermal comfort. 3. Bioclimate. 4. Thermal neutral temperature. 5. Acceptable temperature. 6. Expected temperature. 7. Thermal sensation. 8. Thermal adaptation.

Table 1. Review of thermal comfort research from different regions.

No.	Authors	Year	Region	Gist
1	Zhu et al. [16]	2010	Rural houses in Yinchuan, China	1; 2
2	Huang et al. [17]	2011	Rural houses in Beijing, China	1; 4; 5
3	Yang et al. [18]	2011	Rural houses in Guanzhong area, China	1; 4; 5
4	Oikonomou et al. [10]	2011	Dwellings in Northwest Greece	1; 2
5	Jin et al. [19]	2013	Rural houses in eastern Guangdong, China	1; 4; 5
6	Ge et al. [20]	2013	Rural houses in Weifang, China	1; 4; 5
7	Alev et al. [11]	2014	Dwellings in Estonia, Finland and Sweden	1; 2
8	Nematchoua et al. [12]	2014	Dwellings in Cameroon	2
9	Nia [13]	2014	Houses in northern Iran	3
10	Wang et al. [21]	2014	Rural houses in Harbin, China	1; 4; 5
11	Zhang Qi et al. [22]	2014	Dwellings in Shanghai, Shandong, Xinjiang, Liaoning and Guangxi, China	1; 4; 5; 6
12	Liu et al. [23]	2016	Dwellings in Khampa Tibetan area, China	1; 2
13	Xiong et al. [24]	2016	Rural houses in Hubei, China	1; 2; 7; 8
14	Liu et al. [25]	2018	Rural houses in Central Hainan, China	1; 8
15	Pang et al. [26]	2018	Rural houses in Beihai, China	1; 5; 7
16	Li et al. [27]	2018	Rural houses in western Liaoling, China	2
17	Al-Absi [14]	2019	High-rise residential building in Malaysia	8
18	Ni Sijia [28]	2020	Rural houses in Central Jiangxi, China	2; 7

1. Field investigation. 2. Thermal comfort. 3. Bioclimate. 4. Thermal neutral temperature. 5. Acceptable temperature. 6. Expected temperature. 7. Thermal sensation. 8. Thermal adaptation.

In recent years, some scholars in China have studied the indoor thermal environment and thermal comfort of rural houses in different climatic zones. Wang Zhaojun [21] and others conducted a field investigation of rural houses in Harbin in the winter of 2012. It was suggested that the operative temperature should be used as the evaluation index of indoor thermal comfort of rural houses in severely cold areas in winter. In the winter of 2012 and the summer of 2013, Zhang Qi [22] and others conducted field measurements and investigations of rural houses. The acceptable temperature ranges, thermal neutral temperatures and expected temperatures of residents were obtained. In the summer of 2014 and the winter of 2015, Xiong Yan [24] and others conducted a field investigation of the thermal environment and thermal comfort in rural areas of Hubei Province. It was found that the indoor thermal environment was extremely poor, but the findings regarding local residents’ thermal sensations were relatively good, which indicates that rural residents have good adaptability to the environment they have been living in for a long time. In the summer of 2016, Liu Yanfeng [25] and others investigated the indoor thermal environment of rural houses in central Hainan in summer. The results show that residents were heat-resistant and adapted well to the environment by changing their behaviors to regulate the thermal conditions. In addition, Deng Yuwan [29] and others studied the outdoor environment of residential areas in Changsha in 2022 and found that the local neutral temperature was 26.2 °C, and 90% of the acceptable temperature range was lower than 28.5 °C. Residents adapted to the thermal environment by adjusting their activity times and locations in summer.

Although some Chinese scholars have undertaken useful research on the indoor thermal environment and thermal comfort of rural houses in different climatic zones, the indoor thermal comfort levels of rural houses in different regions and climatic zones are quite different because of China’s large size and many rural areas. Therefore, the database of thermal comfort in rural houses in China is incomplete and needs to be improved. In severely cold zones, scholars have been mainly engaged in studying the thermal comfort of rural houses in winter, but with climate change and global warming, the thermal comfort of rural houses in summer is also now worth studying and discussing.

In view of the above problems, this paper mainly studies the indoor thermal comfort of rural residents at higher latitudes in summer and what thermal adaptive behaviors rural residents adopt to adapt to this environment.

Therefore, this study takes rural houses in Suibin Town, China as an example. A field investigation of thermal comfort in summer is conducted. A thermal sensation vote (TSV) was used, and the operative temperature was selected as the thermal comfort evaluation index in this study. The regression of the thermal neutral temperature by the operative temperature (t_o) was the method used to measure the thermal comfort. The purpose of this paper is to explore the indoor thermal comfort level of rural residents in high-latitude and severely cold areas in China in summer, determine the thermal preferences of rural residents, and reveal the thermal adaptive behaviors of rural residents by which they adapt to the indoor thermal environment in summer. The results can be used to estimate the indoor thermal neutral temperatures, acceptable thermal temperature ranges and thermal preference temperatures of rural residents in high-latitude and severely cold areas of China in summer. This approach will also be helpful for improving some of the thermal comfort databases in China. Moreover, this study provides a reference for studies of indoor thermal environments of rural houses in the same climatic zone and provides a logical method for the study of the indoor thermal environments of rural houses in other climatic zones. It also improves the replicability and potential impact of the investigation approach overall. In addition, this study provides design guidance for improving the indoor thermal environments of rural houses.

2. Background Information

2.1. Village Location and Climate

The research site was Suibin Town, China. Suibin Town is located at 47°17′25″ N latitude and 131°51′37″ E longitude. Suibin Town is in Suibin County, Hegang City, Heilongjiang Province, and is located in the south of Suibin County, in the northwest of Sanjiang Plain, on the alluvial plain at the confluence between the Heilongjiang River and Songhuajiang River. It faces Qinghua Township in Fujin City across the Songhuajiang River in the east and south, borders Beigang Township in the west and north, and connects to Liansheng Township in the northeast. In 2018, the area was 251.4 km², and the registered population was 40,464 [30]. At the end of 2011, Suibin Town had jurisdiction over four communities—Xingmin, Jiangcheng, Huayuan and Dongli—and 17 administrative villages—Qingfa, Jili, Jichang, Fengyi, Hechun, Suibin, Qinglian, Minzhu, Aolai, Shengli, Jizhen, Zhonghe, Jifu, Datong, Zhenrong, Jicheng and Xiangri [31].

Suibin Town has a cold, temperate, monsoon climate, which is characterized by four distinct seasons, with a winter that is controlled by polar continental air masses and is cold, dry and long. The summer is determined by subtropical ocean air masses, during which the climate is warm and humid, but only for a short amount of time. Spring and autumn are alternately cold and warm, and the climate is changeable [31]. According to the Köppen climate classification, Suibin Town belongs to the Dwb classification, and according to the GB 50176-2016 [32] climate zoning standard, is a severely cold zone.

2.2. Village Residential Appearance

Fengyi Village is in Suibin Town, Suibin County, Hegang City, Heilongjiang Province. In November 2021, Fengyi Village was identified as a typical case of rural industrial revitalization by the Agricultural Office of the Heilongjiang Provincial Committee. Figure 1 shows an aerial photograph of the whole of Fengyi Village. Figure 2 shows the appearance of rural houses in Fengyi Village. The buildings in the village are arranged neatly, and the village layout, residential appearance and building materials conform to the typical characteristics of village construction in cold areas. The village is representative.

During the field investigation in Fengyi Village, it was found that the buildings were mostly north–south oriented; most of them were “I”-shaped, and they were all one-story buildings with sloping roofs. No buildings with two or more floors were found. Among the 144 rural houses in Fengyi Village, the number of brick houses was 104, accounting for 72.22%. The rural building materials were mainly red bricks. Figure 3 shows the plan of a typical brick house in Fengyi Village. See

Table 2. Detailed information of typical rural houses in Fengyi.

Parameter	Structure	U-Value
roof	iron sheet + timber	2.4 W/m2·K
suspended ceiling	plant ash + timber + plastic board	0.7 W/m2·K
external wall	red brick + cement mortar	1.5 W/m2·K
internal wall	red brick + white lime	2.6 W/m2·K
ground	ceramic tile + cement mortar	3.9 W/m2·K
door	wooden door with iron sheet	2.8 W/m2·K
window	wooden window with glass (SHGC = 0.87)	5.9 W/m2·K

Note: SHGC (solar heat gain coefficient) refers to the ratio of the indoor heat gain (related to the solar radiation passing through the transparent exterior protective elements (doors, windows or transparent curtain walls) to the amount of solar radiation projected outwards through them.

for details of typical rural houses in Fengyi Village.

During the investigation, it was found that only one resident in the village had installed an air conditioner. However, the owner did not use the air conditioner even during the hot summer months, choosing instead to occasionally use an electric fan that consumes less energy. This phenomenon is related to the residents’ long-term living habits and the economic conditions in this area, which has seen the development of unique personal adaptation methods. Their thermal comfort range and acceptance are greatly influenced by their residential buildings, as well as their own thermal experiences, psychological factors and economic conditions.

3. Research Methods

3.1. Investigation Methods

The subjective questionnaire survey was conducted, and field environmental parameters were measured for 7 consecutive days in Suibin Town, from 12–19 June 2022 (summer), in order to collect subjective perceptions of indoor thermal comfort and data of indoor and outdoor environmental parameters. In addition, the subjective questionnaire survey’s results and the data on environmental parameters were statistically analyzed. The statistical methods mainly comprised linear and nonlinear fitting regression.

3.2. Measurement of Indoor and Outdoor Environmental Parameters

The outdoor air temperature and relative humidity were measured using measuring instruments. The measuring point was set at 1.1 m from the ground. At the same time, three automatic temperature and humidity recorders were set up, set to automatically record outdoor temperature and humidity values every 5 min for 7 consecutive days. The outdoor temperature and humidity values that were ultimately used for the statistical analysis are the averages of the values measured by the three instruments.

Indoor air temperature, relative humidity, air velocity and black globe temperature were measured using measuring instruments. The measuring point was set at 1.1 m from the ground and 1 m from the subject. During the interviews with each subject, the indoor air temperature, relative humidity, air velocity and black globe temperature were recorded four times. The values ultimately used for the statistical analysis are the averages of the four measured values.

The measuring instruments are shown in

Table 3. Properties of measuring instruments.

Instrument	Testo 435 Anemometer	JTR04 Black Ball Temperature Instrument	BES-02B Temperature and Humidity Recorder
Measured parameter	Temperature, humidity and air velocity	Black globe temperature	Temperature and humidity
Range	−20~+70 °C 0~+100% RH 0~+20 m/s	+5~+120 °C	−30~+50 °C 0~+99% RH
Accuracy	±0.3 °C ±2% RH ±0.03 m/s	±0.5 °C	≤0.5 °C ≤3% RH
Characteristic	—	Globe diameter (150 mm) and emissivity (>0.95)	Automatic temperature and humidity acquisition

. The instruments met the precision requirements of international standards (ISO 7726 [33], ASHRAE 55-2020 [34]) and the Chinese standard (GB/T 50785-2012 [35]).

Figure 4a shows the point at which the outdoor environment measurements were taken, and Figure 4b shows the indoor measurement points. Figure 5a shows the parameters of the outdoor environmental measurements. In order to preclude the influence of direct sunlight, a BES-02B automatic temperature and humidity recorder was wrapped in a special box with aluminum foil attached to the inside and outside. Figure 5b shows the parameters of the field measurements inside the rural houses.

3.3. Subjective Questionnaire Survey

By interviewing subjects and asking them to fill out questionnaires, we collected the subjective perceptions of indoor thermal comfort of the rural residents in Suibin Town, and at the same time, the indoor environmental parameters of rural houses were measured in the field. According to the central limit theorem, the distribution of human thermal responses approximates a normal distribution when the sample size is sufficiently large—a sample size greater than 100 [36] is generally considered appropriate. During this investigation, 296 valid questionnaires were collected. Among all the subjects, 92.9% had lived in the area for more than 5 years, 88.17% for more than 10 years, 71.28% for more than 20 years and 52.36% for more than 30 years. The data of their subjective evaluations of indoor thermal comfort are representative.

Table 4. Basic information of subjects.

Male (%)	Female (%)	Age Distribution (%)
		0–18	19–29	30–39	40–49	50–59	60–69	70+
50.68	49.32	2.36	12.84	11.15	19.93	28.38	14.53	10.81

shows the basic information of the subjects. As can be seen from Table 4, the proportions of male and female subjects were relatively balanced. The subjects’ ages were concentrated in the 40–49 and 50–59 years old ranges—mainly middle-aged and elderly people. Subjects over 40 years old accounted for 73.65% of the total.

In this study, we designed the Survey Questionnaire on Indoor Thermal Comfort of Rural Houses (see the Appendix A for the questionnaire’s contents). The purpose of the questionnaire’s design is to obtain more convenient and accurate data on subjective evaluations, such as the basic information, activity behavior information, clothing information and indoor thermal sensation and thermal adaptive behavior information of rural residents. Therefore, when designing the questions in the questionnaire, the main information being sought included the following:

(1)

Sex, age and residence time of the subject;

(2)

Subject’s activities within a set 15 min. There were some common activities to choose from listed in the questionnaire;

(3)

Subject’s clothing, from underwear to outerwear. There was a detailed list of clothing types in the questionnaire to choose from;

(4)

Subject’s subjective feelings, including thermal sensation, thermal comfort, thermal acceptability, thermal preference, humidity sensations, humidity preferences, air movement sensations and air movement preferences;

(5)

Subject’s window-opening behaviors and thermal adaptive behaviors. There were some common thermal adaptive behaviors to choose from in the questionnaire.

The inquiry of subjective sensation in the questionnaire adopted the seven-point scale of thermal sensation used in the international standards (ASHRAE 55-2020 standard [34], ASHRAE Handbook 2021 [37] and ISO 7730 [38]) and the Chinese standard (GB/T 33658-2017 [39]), as shown in

Table 5. Seven-point scale used in the questionnaire.

	−3	−2	−1	0	+1	+2	+3
Thermal sensation vote (TSV)	Cold	Cool	Slightly cool	Neutral	Slightly warm	Warm	Hot
Humidity sensation vote (HSV)	Very dry	Dry	Slightly dry	Neutral	Slightly humid	Humid	Very humid
Air movement sensation vote (AMSV)	Very low	Low	Slightly low	Neutral	Slightly high	High	Very high

.

The inquiry into preferences regarding air temperature, relative humidity and air velocity in the questionnaire used a three-point scale, as shown in

Table 6. Three-point scale used in the questionnaire.

	−1	0	+1
Thermal preference vote (TPV)	Be cooler	No change	Be warmer
Humidity preference vote (HPV)	Be more dryer	No change	Be more humid
Air movement preference vote (AMPV)	Be lower	No change	Be higher

.

The thermal comfort vote (TCV) and thermal acceptability vote (TAV) in the questionnaire were ranked on a five-point scale and a two-point scale, respectively, as shown in

Table 7. Five-point scale and two-point scale used in the questionnaire.

	0	1	2	3	4
Thermal comfort vote (TCV)	Comfortable	Slightly uncomfortable	Uncomfortable	Very uncomfortable	Extremely uncomfortable
Thermal acceptability vote (TAV)	Acceptable	Unacceptable	—	—	—

.

3.4. Calculation Methods

In this study, standardized thermal comfort questionnaires and measuring instruments were used. All statistical significance levels were set to p < 0.05 to ensure the accuracy of the results. The regression equations between operative temperature and the related parameters of the thermal environment were established and analyzed by linear regression.

The operative temperature (t_o) comprehensively considers the influence of air temperature (t_ɑ) and mean radiant temperature (${\overline{t}}_r$) on human thermal sensation, and the calculation method given in the ASHRAE Handbook 2021 [37] is shown in Equation (1).

$$t_o=\frac{h_r{\overline{t}}_r+h_c{t}_a}{h_r+h_c}$$

(1)

where

• t_o—operative temperature (°C);
• t_ɑ—air temperature (°C);
• ${\overline{t}}_r$—mean radiant temperature (°C);
• h_r—linear radiative heat transfer coefficient (W/m²·°C);
• h_c—convective heat transfer coefficient (W/m²·°C).

According to the ASHRAE 55-2020 standard [34], the operative temperature (t_o) can be calculated using Equation (2).

$$t_o=A{t}_a+\left(1-A\right){\overline{t}}_r$$

(2)

where

• t_o—operative temperature (°C);
• t_ɑ—air temperature (°C);
• ${\overline{t}}_r$—mean radiant temperature (°C);
• A—correction coefficient. When the air velocity is less than 0.2 m/s, it takes the value of 0.5, but when it is 0.2 m/s to 0.6 m/s, it takes 0.6, and when it is 0.6 m/s to 1.0 m/s, it takes the value 0.7.

The mean radiant temperature (${\overline{t}}_r$) was measured using a black globe thermometer, and the mean radiant temperature was calculated by measuring the black globe temperature, air temperature and air velocity according to Equations (3) and (4), given in the international standard (ISO 7726 [33]) and the Chinese standard (GB/T 50785-2012 [35]).

By natural convection:

$${\overline{t}}_r={\left[{\left(t_g+273\right)}^4+\frac{0.25\times {10}^8}{{\epsilon }_g}{\left(\frac{\left|t_g-t_a\right|}{D}\right)}^{\frac{1}{4}}\times \left(t_g-t_a\right)\right]}^{\frac{1}{4}}-273$$

(3)

By forced convection:

$${\overline{t}}_r={\left[{\left(t_g+273\right)}^4+\frac{1.1\times {10}^8\times v_a{}^{0.6}}{{\epsilon }_g\times D^{0.4}}\left(t_g-t_a\right)\right]}^{\frac{1}{4}}-273$$

(4)

where

• ${\overline{t}}_r$—mean radiant temperature (°C);
• t_ɡ—black globe temperature (°C);
• t_ɑ—air temperature (°C);
• v_ɑ—air velocity (m/s);
• D—the diameter of the black globe (m);
• ε_ɡ—the emissivity of the black globe.

The diameter of a standard black globe is 0.15 m, and its emissivity is 0.95. The JTR04 black ball temperature instrument used in this study had the properties of a standard black globe. Therefore, according to the international standard (ISO 7726 [33]), the equation for calculating the mean radiant temperature can be simplified to Equations (5) and (6).

By natural convection:

$${\overline{t}}_r={\left[{\left(t_g+273\right)}^4+0.4\times {10}^8{\left|t_g-t_a\right|}^{\frac{1}{4}}\times \left(t_g-t_a\right)\right]}^{\frac{1}{4}}-273$$

(5)

By forced convection:

$${\overline{t}}_r={\left[{\left(t_g+273\right)}^4+2.5\times {10}^8\times v_a{}^{0.6}\left(t_g-t_a\right)\right]}^{\frac{1}{4}}-273$$

(6)

where

• ${\overline{t}}_r$—mean radiant temperature (°C);
• t_ɡ—black globe temperature (°C);
• t_ɑ—air temperature (°C);
• v_ɑ—air velocity (m/s).

Before calculating the average radiation temperature, it is necessary to judge the coefficient of heat transfer via convection at the level of the black globe (h_cɡ). According to the international standard (ISO 7726 [33]), the calculation can be carried out as follows below.

By natural convection:

$$h_{cg}=1.4{\left(\frac{\Delta T}{D}\right)}^{\frac{1}{4}}$$

(7)

By forced convection:

$$h_{cg}=6.3\frac{v_a{}^{0.6}}{D^{0.4}}$$

(8)

where

• h_cɡ—coefficient of heat transfer by convection of the black globe (W/m²·K);
• ΔT—difference between black globe temperature and air temperature (°C);
• v_ɑ—air velocity (m/s);
• D—the diameter of the black globe (m).

Next, we calculate the h_cɡ values of natural convection and forced convection. Assuming that the h_cɡ of natural convection is greater than that of forced convection, the mean radiant temperature should be calculated according to the equation regarding natural convection. On the contrary, the mean radiant temperature should be calculated according to the equation regarding forced convection.

The data of indoor air temperature, relative humidity, air velocity and black globe temperature were obtained via field measurements. Operative temperature and mean radiant temperature were calculated according to the above equations and used for the statistical analysis.

4. Results and Analysis

4.1. Outdoor Physical Parameters

See Figure 6 for the outdoor environmental parameters during the 7-day field measurement period from 12–19 June 2022.

The outdoor environmental parameters during the field measurement period are shown in

Table 8. Outdoor environmental parameters.

Parameter	Max	Min	Mean	S
Outdoor air temperature (°C)	35.77	15.56	23.75	5.47
Outdoor relative humidity (%)	92.05	22.20	57.03	18.94

. The maximum value of outdoor air temperature was 35.77 °C, while the minimum value was 15.56 °C, and the average value was 23.75 °C. The maximum outdoor relative humidity was 92.05%, the minimum was 22.20%, and the average was 57.03%.

Because of the occurrence of rain on 17 June, the maximum outdoor relative humidity was high, and it remained high during the rain.

4.2. Thermal Sensation

In field investigations of thermal comfort, there are two ways to express human thermal sensation, namely, thermal sensation vote (TSV) and mean thermal sensation vote (MTS). Thermal sensation vote (TSV) refers to the human thermal sensation at a certain temperature, and mean thermal sensation vote (MTS) refers to the average of human thermal sensations within a certain temperature range [40]. The neutral operative temperature is the operative temperature at which the TSV or MTS is equal to 0. At this time, the rural residents feel neither hot nor cold; this, therefore, gives the subjective thermal demand temperature of rural residents and can be used as an indoor thermal environment calculation parameter. TSV/MTS must equal [−0.5, 0.5] in order to reach a 90% acceptable temperature range, and it must equal [−0.85, 0.85] to reach an 80% acceptable temperature range, in order to effectively describe the subjective acceptable temperature range of rural residents.

Thermal sensation vote (TSV) describes subjects’ actual subjective feelings about the indoor thermal environment of a rural house. Figure 7 shows the distribution of TSV of subjects. According to the figure, the highest value of distribution appeared when the value of TSV was 0 (neutral), accounting for 35.47%, followed by 1 (slightly warm), accounting for 23.65%, and then −1 (slightly cool), accounting for 14.87%. The values of −1 (slightly cool), 0 (neutral) and 1 (slightly warm) accounted for 73.98% of the total, indicating that most of the subjects actually had relatively good subjective feelings regarding their thermal sensations, but 14.53% of the subjects still felt warm (2), while 7.09% felt hot (3), and 4.39% of the subjects actually felt cool (−2).

4.2.1. Analysis of Thermal Sensation Vote

Thermal sensation vote (TSV) is the subject’s actual subjective feelings in relation to their body’s thermal sensation. A regression analysis was performed on the TSV and operative temperature (t_o). Assuming that the regression result was statistically significant, the operative temperature obtained when the TSV calculated by the regression equation result was equal to zero was the thermal neutral operative temperature based on the TSV.

The operation proceeded as follows:

①

Taking the operative temperature of each subject, calculated according to the measured data, as the independent variable and the value of the TSV of each subject as the dependent variable, the linear regression equation was obtained through regression analysis: TSV = a × t_o + b;

②

A significance test of the linear regression relation was necessary. Generally speaking, when p is less than 0.05, the regression relationship is usually considered to be significant. If the regression relation of equation TSV = a × t_o + b was significant, this showed that the TSV yielded by this regression equation can be used to predict human thermal sensation well.

The regression result showed that p was less than 0.001, and there was a statistically significant correlation between the operative temperature (t_o) and the TSV. A regression analysis of the TSV and operative temperature (t_o) was carried out. The fitting results are shown in Figure 8, and the regression equation is as follows:

TSV = 0.3300t_o − 8.5293 (R² = 0.299, p < 0.001)

(9)

By setting the TSV to 0, the thermal neutral operative temperature was determined to be 25.8 °C. By setting the TSV to [−0.5, 0.5], the 90% acceptable temperature range was determined to be [24.3 °C, 27.4 °C]. By setting the TSV to [−0.85, 0.85], the 80% acceptable temperature range was determined to be [23.3 °C, 28.4 °C].

4.2.2. Analysis of Mean Thermal Sensation Vote

The mean thermal sensation vote (MTS) is used to describe human thermal sensation. By adopting a binning method, we performed a regression analysis on the MTS and operating temperature (t_o) of subjects in different temperature ranges. Assuming that the regression result was statistically significant, the operative temperature obtained when the MTS calculated by the regression equation result was equal to zero was the thermal neutral operative temperature based on the MTS.

The operation process was as follows:

①

An operative temperature within the range calculated according to the measured data was divided into several operative temperature intervals separated by ±0.25 °C, so the Δt_o value was 0.5 °C for each operative temperature interval. For example, [22.5 °C, 23.0 °C] [23.0 °C, 23.5 °C] [23.5 °C, 24.0 °C] [24.0 °C, 24.5 °C] [24.5 °C, 25.0 °C] [25.0 °C, 25.5 °C] [25.5 °C, 26.0 °C] [26.0 °C, 26.5 °C] [26.5 °C, 27.0 °C] [27.0 °C, 27.5 °C] [27.5 °C, 28.0 °C] [28.0 °C, 28.5 °C] [28.5 °C, 29.0 °C] [29.0 °C, 29.5 °C] [29.5 °C, 30.0 °C] [30.0 °C, 30.5 °C] [30.5 °C, 31.0 °C] [31.0 °C, 31.5 °C] [31.5 °C, 32.0 °C] [32.0 °C, 32.5 °C] and [32.5 °C, 33.0 °C]. The number of operative temperature intervals was 21 in total;

②

We then calculated the average of the thermal sensation vote values given by all subjects in each operative temperature interval;

③

Taking the operative temperature at the midpoint of each operative temperature interval as the independent variable and the MTS value of each operative temperature interval as the dependent variable, the linear regression relation was obtained through regression analysis—MTS = a × t_o + b;

④

A significance test of the linear regression relation was necessary.

The operative temperature ranges and values of the MTS are shown in

Table 9. Operative temperature range and mean thermal sensation vote values.

to (°C)	MTS	to (°C)	MTS	to (°C)	MTS
22.75	0	26.25	−0.1	29.75	1.1
23.25	0	26.75	−0.2	30.25	2
23.75	0	27.25	0.2	30.75	2
24.25	0	27.75	0.1	31.25	2.5
24.75	0.1	28.25	0	31.75	3
25.25	0.1	28.75	0.6	32.25	3
25.75	0.1	29.25	1.1	32.75	3

.

The regression results show that p was less than 0.001, and there was a statistically significant correlation between the operative temperature (t_o) and MTS. A regression analysis was performed on the MTS and operative temperature (t_o). The fitting results are shown in Figure 9. The regression equation is as follows:

MTS = 0.3335t_o − 8.3691 (R² = 0.773, p < 0.001)

(10)

By setting the MTS to 0, the thermal neutral operative temperature was determined to be 25.1 °C. By setting the MTS to [−0.5, 0.5], the 90% acceptable temperature range was determined to be [23.6 °C, 26.6 °C]. By setting the MTS to [−0.85, 0.85], the 80% acceptable temperature range was determined to be [22.5 °C, 27.6 °C]. It can also be seen from Figure 9 that when the value of MTS is close to or equal to 0, it is concentrated in the temperature range [23.0 °C, 28.0 °C], indicating that the human body feels comfortable in this range. The thermal neutral operative temperature obtained (25.1 °C) was close to the thermal neutral operative temperature (25.4 °C) in summer in severely cold zones that was described in reference [41].

4.2.3. Comparative Analysis of the TSV and MTS

The comparison in

Table 10. Comparison of thermal sensation vote and mean thermal sensation vote.

Regression Equation	Thermal Neutral Operative Temperature	90% Acceptable Temperature Interval	80% Acceptable Temperature Interval
TSV = 0.3300to − 8.5293 (R2 = 0.299, p < 0.001)	25.8 °C	[24.3 °C, 27.4 °C]	[23.3 °C, 28.4 °C]
MTS = 0.3335to − 8.3691 (R2 = 0.773, p < 0.001)	25.1 °C	[23.6 °C, 26.6 °C]	[22.5 °C, 27.6 °C]

shows that the fitting degree of the MTS equation (Equation (10)) is obviously better than that of the TSV equation (Equation (9)). The reason is that the TSV refers to the subjective feeling of each subject in relation to the indoor thermal environment. Because of the different subjective feelings of different people, there were differences among individuals, and the thermal sensations of different people were not the same at the same temperature. MTS gives the average thermal sensation of all subjects in a certain temperature range, which reduces the errors arising in relation to the different thermal sensations of different people at the same temperature, to some extent. Therefore, compared with the TSV equation (Equation (9)), it is more accurate to evaluate thermal comfort using the thermal neutral operative temperature and acceptable operative temperature obtained by the MTS equation (Equation (10)). Due to the differences in thermal sensations among individuals, and the adjustments made in the psychology and adaptive behaviors of some residents, a slightly higher temperature was tolerated by some, and these subjects also felt comfortable at a slightly higher temperature. Therefore, the temperature obtained by the TSV equation (Equation (9)) was relatively higher than that obtained by the MTS equation (Equation (10)). Critical discomfort among rural residents in the Suibin area in summer was reached when the indoor temperature was less than 22.5 °C and more than 27.6 °C.

In summer, the thermal neutral temperature of the TSV in Beijing (39°54′27″ N, 116°23′17″ E) was 27.2 °C [41], while that in Nanjing (32°02′38″ N, 118°46′43″ E) was 28 °C [42] and that in Guangzhou (23°06′32″ N, 113° 15′53″ E) was 29.3 °C [43]. Compared with the thermal neutral temperature of the TSV (25.8 °C) in Suibin Town (47°17′25″ N, 131°51′37″ E) in summer, people living in severely cold zones adapted to a lower air temperature. Therefore, they expected a lower air temperature during summer.

4.3. Thermal Comfort and Thermal Acceptability

The thermal comfort vote (TCV) and thermal acceptability vote (TAV) jointly describe a subject’s comfort in and satisfaction with the indoor thermal environment of a rural house. Figure 10a shows the distribution of the TCV of the subjects. According to this figure, 48.31% of the subjects felt comfortable (0), and 40.54% felt slightly uncomfortable (1). Ratings of uncomfortable (2), very uncomfortable (3) and extremely uncomfortable (4) accounted for 6.42%, 3.04% and 1.69%, respectively. It was shown that during the test period, nearly 40% of the subjects were slightly uncomfortable, and nearly 10% of the subjects were more uncomfortable, indicating that rural residents had obvious demands related to indoor thermal comfort. Figure 10b shows the distribution of the TAV of the subjects. According to the figure, 79.05% of the subjects thought that the thermal environment was acceptable (0), and 20.95% thought it was unacceptable (1). Overall, rural residents showed a high degree of acceptance of the indoor thermal environment.

We calculated the percentages of subjects with thermal acceptability inputs corresponding to each value of TSV out of the total number of subjects with this value of TSV, thus deriving the thermal acceptable percentage (TAP), as shown in

Table 11. Correspondence of thermal sensation vote and thermal acceptable percentage.

TSV	TAP (%)
−3 (Cold)	—
−2 (Cool)	69.23
−1 (Slightly cool)	75
0 (Neutral)	100
1 (Slightly warm)	81.43
2 (Warm)	58.14
3 (Hot)	23.81

. We then performed a regression analysis with the TSV.

The regression analysis was carried out with the TSV as the independent variable and TAP as the dependent variable. The regression results show that p was 0.016 and less than 0.05; thus, there was a statistically significant correlation between the TSV and TAP. The fitting results are shown in Figure 11, and the regression equation is as follows:

TAP = −7.0296TSV² − 1.4346TSV + 90.9129 (R² = 0.938, p < 0.05)

(11)

In Figure 11, the maximum acceptability (or satisfaction) occurred on the cold side of the symmetrical TSV scale. That is, rural residents were more accepting of and satisfied with lower temperatures in summer. According to Figure 11, an acceptability of 90% occurred between TSVs of −0.47 and 0.27], and an acceptability of 80% occurred between TSVs of −1.35 and 1.15. Because the MTS equation was more accurate, the above values were brought into the linear regression equation of the MTS and operative temperature, and the 90% acceptable temperature range was [23.7 °C, 25.9 °C], while the 80% acceptable temperature range was [21.0 °C, 28.5 °C]. Compared with the 90% acceptable temperature range [23.6 °C, 26.6 °C] and the 80% acceptable temperature range [22.5 °C, 27.6 °C] of the MTS, there was a slight difference. The reason for this was that the thermal adaptation behaviors of the subjects had a regulating effect on their thermal acceptance, correlating with better thermal acceptability at the same temperature.

4.4. Thermal Preference

Thermal preference vote (TPV) describes how the indoor temperature of the rural house in which the subject lives can make them feel more comfortable. Figure 12 shows the distribution of the TPV scores of the subjects. According to the figure, 47.63% of the subjects wanted the indoor temperature to be cooler (−1) than the measured indoor temperature, 41.22% wanted the temperature to remain unchanged (0), and 11.15% wanted the temperature to be warmer (1).

By adopting the above-mentioned binning method, we calculated the percentages of subjects who wanted the temperature to be cooler (−1) and warmer (1) in each operative temperature interval out of the total number of subjects within this interval and obtained the preferred thermal percentage (TPP). The TPP was divided into TPP_Cooler and TPP_Warmer, as shown in

Table 12. Operative temperature range and thermal preferred percentage value.

to (°C)	TPPCooler (%)	TPPWarmer (%)	to (°C)	TPPCooler (%)	TPPWarmer (%)	to (°C)	TPPCooler (%)	TPPWarmer (%)
22.75	0.00	100.00	26.25	23.53	20.59	29.75	71.43	0.00
23.25	0.00	100.00	26.75	30.30	21.21	30.25	78.38	0.00
23.75	42.86	14.29	27.25	38.71	16.13	30.75	100.00	0.00
24.25	33.33	0.00	27.75	42.86	14.29	31.25	100.00	0.00
24.75	33.33	9.52	28.25	25.00	18.75	31.75	100.00	0.00
25.25	52.00	20.00	28.75	58.33	4.17	32.25	100.00	0.00
25.75	38.89	5.56	29.25	62.50	0.00	32.75	100.00	0.00

. We then performed a regression analysis with the operative temperature (t_o).

Taking the operative temperature (t_o) as the independent variable and the TPP as the dependent variable, a regression analysis was carried out. The regression results show that the p was less than 0.001 for TPP_Cooler and was 0.002 (less than 0.05) for TPP_Warmer. There was also a statistically significant correlation between the operating temperature (t_o) and the TPP. A regression analysis was carried out on the TPP and the operative temperature (t_o). The fitting results are shown in Figure 13. The regression equations of the TPP_Cooler and TPP_Warmer are as follows:

TPP_Cooler = 9.4679t_o − 208.8563 (R² = 0.810, p < 0.001)

(12)

TPP_Warmer = −5.9400t_o + 181.2410 (R² = 0.405, p = 0.002 < 0.05)

(13)

As shown in Figure 13, the temperature corresponding to the intersection of two fitting curves is the preferred temperature. Combining the above two regression equations (Equations (12) and (13)), the thermal preference temperature at the intersection of the two fitting curves could be calculated. Through this calculation, it was concluded that the preferred temperature amongst local rural residents was 25.3 °C in summer. The preference temperature obtained (25.3 °C) was very close to the neutral temperature (25.1 °C) of the MTS. This indicates the reliability of the neutral temperature result for the MTS.

4.5. Humidity Sensation and Humidity Preference

The humidity sensation vote (HSV) describes the subject’s actual subjective feelings about the indoor humidity of a rural houses. Figure 14a shows the distribution of the HSV scores of the subjects. According to the figure, the highest value of distribution appeared when the value of the HSV was 0 (neutral), accounting for 34.80%; this was followed by −1 (slightly dry), accounting for 20.27%, and 1 (slightly humid), accounting for 17.57%. The values of −1 (slightly dry), 0 (neutral) and 1 (slightly humid) accounted for 72.64% of the total, indicating that most of the subjects actually had relatively positive subjective feelings regarding humidity. However, 9.12% of the subjects still felt humid (2), and 4.05% felt very humid (3), while 8.45% of the subjects actually felt dry (−2), and 5.74% felt very dry (−3).

The humidity preference vote (HPV) describes how the humidity of the rural houses in which the subjects want to live can make them feel more comfortable. Figure 14b shows the distribution of the HPV scores of the subjects. According to the figure, 27.70% of the subjects wanted the indoor humidity to be dryer (−1) than the measured value, 54.06% wanted the humidity to remain unchanged (0), and 18.24% wanted the humidity to increase (1).

4.6. Air Movement Sensation and Air Movement Preference

Air movement sensation vote (AMSV) describes the actual subjective feelings of the subjects regarding the indoor air movement in a rural house. Figure 15a shows the distribution of the AMSV scores of the subjects. According to the figure, the greatest distribution appeared when the value of the AMSV was 0 (neutral), accounting for 32.77%. This was followed by −2 (low), accounting for 22.64%, and then by −1 (slightly low), accounting for 22.30%. The values of −1 (slightly low), 0 (neutral) and 1 (slightly high) accounted for 63.85% of the total, which indicates that more than half of the subjects actually had relatively positive subjective feelings regarding air movement sensations. However, there were still many subjects who actually felt that the air movement was low or very low. Among these, 22.64% felt that it was low (−2), and 9.46% felt it was very low (−3). Few subjects actually felt it to be high or very high—3.04% felt it was high (2), and 1.01% felt it was very high (3).

Air movement preference vote (AMPV) describes how the indoor air movement of a rural house can make one feel more comfortable. Figure 15b shows the distribution of the AMPV scores of the subjects. According to the figure, only 5.07% of the subjects wanted the indoor air movement to be reduced (−1), while 46.28% wanted the air movement to remain unchanged (0), and 48.65% wanted it to be increased (1).

Nearly half of the subjects wanted the air movement to be increased, which is close to the number of subjects who gave scores of −1, −2 and −3 in the AMSV. This shows that nearly half of the subjects thought that the measured indoor air movement was low, and they wanted it to be higher.

4.7. Thermal Adaptive Behavior

Given the heat during summer, the restrictions of the rural residents’ economic conditions and the influence of their living habits, when the indoor thermal environment of rural houses cannot meet the residents’ comfort needs, the residents will spontaneously adopt different behaviors in order to adapt to the thermal environment. In the field investigation, a number of multiple-choice questions were presented to the subjects, so as to determine the common thermal adaptive behaviors of rural residents and their usage ratios. Among them, the most common thermal adaptive behaviors in summer were staying in the shade, changing clothing (reducing), drinking cold drinks, sitting quietly or reducing activity, opening doors and windows for ventilation, using hand-cranked fans, using electric fans, using air conditioners, etc. Figure 16 shows the distributions of different thermal adaptive behaviors in summer.

When the subjects felt that the indoor environment was hot, they adopted some adaptive behaviors. As shown in Figure 16, the most common thermal adaptive behaviors included opening doors and windows for ventilation (73.65%), staying in the shade (48.65%), sitting quietly or reducing activity (43.92%) and changing clothing (39.19%), followed by using hand-cranked fans (31.42%) and drinking cold drinks (22.64%). The common feature of these behaviors is that they are cheap and convenient. Few rural residents used electric fans, air conditioners or other refrigeration equipment to adapt to the hot indoor environment because these have high operating costs. If rural residents were to frequently use such electric equipment, their electricity bills would increase, which is uneconomical. Therefore, in order to adapt to the hot indoor environment, local rural residents were more inclined to adopt more traditional and convenient behaviors than relatively troublesome and expensive ones. Generally speaking, rural residents adopt different thermal adaptation behaviors to adjust their thermal comfort according to a change in indoor temperature and have shown certain environmental adaptation and adjustment abilities.

Most rural residents adapt to their indoor environment by opening doors and windows for ventilation, which is consistent with the observation that nearly half of the subjects wanted the movement of indoor air to be higher. During the field surveys, 60.81% of the subjects opened the windows or doors of their houses; 65.56% of these did so because the room was too hot, 52.22% needed ventilation, and 6.67% did so because the room smelled of smoke.

Besides staying in the shade, rural residents adapt to their indoor environment by reducing their activities. The distributions of subjects’ activities are shown in Figure 17. At the time of the interview, 83.78% of the subjects undertook minor activities, such as sleeping, reclining, sitting quietly, standing quietly, reading and writing, and only 5.07% of the subjects engaged in moderate activities, while no subjects were engaged in high-intensity activities.

Rural residents adapt to their indoor environment by increasing or decreasing their clothing, which leads to changes in their insulation. Therefore, the operative temperature (t_o) was taken as the independent variable and clothing insulation (I_cl) was taken as the dependent variable in the regression analysis.

The levels of insulation of common, single-piece clothing items are summarized according to international standards (ASHRAE 55-2020 standard [34], ASHRAE Handbook 2021 [37] and ISO 7730 [38]) and Chinese standards (GB/T 50785-2012 [35] and GB/T 33658-2017 [39]), as well as “the Practical Design Manual of Heating and Air Conditioning” [44]. By adding together the insulation offered by each item worn by a subject, the total clothing insulation value can be obtained, which is referred to as the clothing insulation (I_cl) in this regression analysis. The average value of clothing insulation for the subjects was 0.40 clo in summer.

The regression results show that p was 0.001 (less than 0.05), and there was a statistically significant correlation between the operative temperature (t_o) and the clothing insulation (I_cl). A regression analysis was carried out on the clothing insulation (I_cl) and the operative temperature (t_o). The fitting results are shown in Figure 18, and the regression equation is as follows:

I_cl = −0.0137t_o + 0.7777 (R² = 0.036, p = 0.001 < 0.05)

(14)

As shown in Figure 18, clothing insulation (I_cl) had a negative correlation with the operative temperature (t_o). As the indoor thermal conditions of rural houses gradually became warmer, the clothing insulation of rural residents gradually decreased. Substituting the neutral temperature of the MTS (25.1 °C) into the above regression equation (Equation (14)), it was found that the value contributed by clothing insulation to the neutral temperature of the MTS (25.1 °C) was 0.43 clo, which is 0.03 clo higher than the average value of clothing insulation for the subjects (0.40 clo). The average value of clothing insulation for the subjects (0.40 clo) was substituted into the regression equation (Equation (14)), and the operative temperature was 27.6 °C. Afterwards, when the indoor thermal environment of rural houses gradually became warmer, the rural residents chose to wear lighter clothes in order to adapt. This is also why the neutral temperature of the TSV (25.8 °C) was 0.7 °C higher than that of the MTS (25.1 °C); that is, at a higher temperature, rural residents can adapt to the indoor thermal environment by adopting certain behaviors, such as reducing the amount of clothes, enabling them to feel comfortable at higher temperatures.

5. Conclusions

In rural areas at high latitudes in severely cold areas in China, where the climate in summer is now becoming hotter, the thermal comfort of subjects was investigated using 296 questionnaires and the actual measurement of indoor and outdoor thermal environments. The purpose of this study was to develop a general understanding of indoor thermal comfort during summer in rural areas at high latitudes and in severely cold zones in China and to reveal the behaviors of rural residents by which they adapt to their indoor thermal environment in summer. Through the analysis of the subjective and objective data of indoor and outdoor thermal environments, the following conclusions can be drawn:

• The regression analysis method was used to analyze and define the neutral temperature and acceptable temperature ranges. It was found that the correlation between operative temperature and the MTS is stronger than that between operative temperature and the TSV, and the thermal neutral temperature is 25.1 °C, while 80% of the acceptable temperature range [22.5 °C, 27.6 °C] was obtained. This shows that the rural residents in severely cold areas have a wide range of acceptable temperatures, and most of them feel comfortable in and have a high degree of acceptance of their indoor thermal environment. Critical discomfort was reached in rural residents in the Suibin area in summer when the indoor temperature was less than 22.5 °C and more than 27.6 °C. The applicability of the thermal comfort evaluation index to the operative temperature in a severely cold area was thus verified;
• Rural residents in severely cold areas hope for a cooler indoor temperature in summer, of which they are more accepting and with which they are more satisfied. The preferred temperature is 25.3 °C in summer, which is very close to the neutral temperature (25.1 °C). In addition, there is a strong correlation between thermal sensation and thermal acceptability;
• Rural residents in severely cold areas are more satisfied with their experienced humidity than with the air movement in summer, which indicates that the influence of air movement on indoor thermal comfort is higher than that of humidity. In total, 72.64% of the subjects expressed a relatively positive feeling of humidity in summer; 54.4% of the subjects scored the AMSV less than 0, stating that indoor air flow was low, and 48.65% of the subjects stated a preference for increased indoor air flow in summer. This also explains why 73.65% of the subjects adapt to their indoor environment in summer by opening doors and windows for ventilation;
• Rural residents in severely cold areas describe opening windows (73.65%), staying in the shade (48.65%), reducing activities (43.92%), changing clothes (39.19%), etc., as means of increasing their indoor thermal comfort in summer. These preferred thermal adaptive behaviors have the common characteristics of tradition, convenience and low price. Rural residents do not tend to undertake relatively cumbersome and expensive thermal adaptive behaviors. In addition, the clothing insulation of rural residents in severely cold areas decreases with the increase in indoor operative temperature. Further, their clothing insulation decreases by 0.0137 clo for every 1 °C increase in the indoor operative temperature. Therefore, rural residents’ acceptance of indoor thermal comfort in summer in severely cold areas is influenced by economic, psychological and adaptive behaviors.

This study will help us to better understand the summertime thermal comfort of rural residents at higher latitudes. It will also help to improve rural living conditions and the indoor thermal comfort database of rural houses in China, so as to more effectively complete the energy-saving renovation of rural houses and provide a reference for designs in relation to indoor thermal comfort. It is suggested that strategies for the energy-saving construction of rural residential buildings at higher latitudes should consider the indoor thermal comfort level of rural residents and their thermal adaptive behaviors. At the same time, this study provides a reference for the study of indoor thermal environments in rural houses in the same climatic zone and provides a logical method for the study of indoor thermal environments in rural houses in other climatic zones. It also shows the replicability and the potential impact of the overall investigation approach. In future research work, we will select more typical villages at higher latitudes and in severely cold areas to study indoor thermal comfort, compare the differences between different villages and assess these issues in other seasons, so as to further improve the indoor thermal comfort database of rural houses.