- Email: [email protected]
Investigation of winter indoor thermal environment and heating demand of urban residential buildings in China's hot summer – Cold winter climate region
Accepted Manuscript Investigation of Winter Indoor Thermal Environment and Heating Demand of Urban Residential Buildings in China’s Hot Summer – Cold Winter Climate Region Borong Lin, Zhe Wang, Yanchen Liu, Yingxin Zhu, Qin Ouyang PII:
S0360-1323(16)30066-X
DOI:
10.1016/j.buildenv.2016.02.022
Reference:
BAE 4408
To appear in:
Building and Environment
Received Date: 5 January 2016 Revised Date:
21 February 2016
Accepted Date: 22 February 2016
Please cite this article as: Lin B, Wang Z, Liu Y, Zhu Y, Ouyang Q, Investigation of Winter Indoor Thermal Environment and Heating Demand of Urban Residential Buildings in China’s Hot Summer – Cold Winter Climate Region, Building and Environment (2016), doi: 10.1016/j.buildenv.2016.02.022. 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 Investigation of Winter Indoor Thermal Environment and Heating Demand of
Urban Residential Buildings in China’s Hot Summer – Cold Winter Climate Region Borong Lin1, 2*, Zhe Wang1, 2, 3, Yanchen Liu1, 2, 3, Yingxin Zhu1, 2 and Qin Ouyang1, 2 1
Department of Building Science, Tsinghua University, Beijing 100084, China Key Laboratory of Eco Planning & Green Building, Ministry of Education, Tsinghua University 3 Beijing Key Laboratory of Indoor Air Quality Evaluation and Control (Tsinghua University), Beijing, China
Corresponding email: [email protected] Corresponding phone (+86) 010 62785691
SC
*
RI PT
2
EP
TE D
M AN U
Abstract In this paper, the winter indoor thermal environment of dwellings in China’s Hot Summer – Cold Winter (HSCW) climate region has been investigated by on-site measurement. And the driving forces of the heating requirements of residents in HSCW area has been scrutinized by heteroskedasticity-robust Ordinary Least Square analysis. It is found that in HSCW area, the average internal temperature is 13.5 oC for the living room and 12.7 oC for the bedroom, leading to an uncomfortable indoor thermal environment, markedly colder than that in the UK, which has similar winter climate. Through the investigation of the occupant heating behavior, it was found that domestic heating is operated part-time-part-space and triggered by both time and temperature in the HSCW area. 10 oC is found to be the lowest acceptable internal temperature without heating. The poor indoor thermal environment observed in HSCW area is mainly due to the short heating operation duration (averagely 1.9 hours per day for the living room and 1.4 hours for the bedroom). Income, presence of child, heating system and thermal experience markedly influence occupants’ heating requirements. However, the education is not a statistically significant factor. The rebound effect is observed in HSCW area when analyzing the effects of building energy efficiency on indoor thermal environment.
AC C
Keywords Indoor thermal environment; Occupant behavior; Residential buildings; China’s Hot Summer – Cold Winter area; Field measurement and questionnaire survey 1. Introduction Domestic heating in China’s Hot Summer – Cold Winter (HSCW) 1 climate region has attracted increasing attention not only in China but also worldwide [1]. Chinese government is responsible for the supplying of central heating to dwellings only north of the Qin-Huai line. However HSCW area lies south of the line, 1
According to the Standard of Climatic Regionalization for Architecture [3], the winter climate of HSCW climate region has the
following two characteristics. Firstly, the average temperature in the coldest month is between 0 and 10 oC. Secondly, the annual days with daily average temperature less than 5 oC are no more than 90 days. According to ASHRAE Standard 90.1 -2010 [4], the HSCW area can be classified into the Climate Zone 3. 1
RI PT
ACCEPTED MANUSCRIPT therefore no central heating is provided by the government in this region. On one hand, residents in HSCW area keep complaining about the poor winter indoor thermal environment resulting from the absence of central heating. On the other hand, with the economic growth and accompanying increase of disposable income, individual heating facilities, such as air source heat pump, gas fueled floor heating and etc., have been widely installed and utilized for domestic heating. Over the last 15 years, the diffusion of individual heating facilities leads to a 575 times growth of residential heating energy consumption in this region [2], constituting a major contributor to the increase of residential energy consumption and associated carbon emission of China. It is worthwhile investigating the internal temperature profiles of dwellings in the HSCW area of China and scrutinizing the driving forces behind that.
M AN U
SC
Indoor thermal environment investigation is a hot research area since it closely related to the building energy consumption and plays a significant role to create a comfort, healthy and efficient built environment. In terms of the investigation methodologies, Alfano et al. [5] proposed a four-step method to assess the indoor thermal environment of existing buildings: firstly cognitive survey, secondly subjective investigation, thirdly instrumental survey and lastly calculation of the indices. Dell'Isola [6] discussed the influence of measurement uncertainties on thermal environment assessment and found that the ambiguous class attribution due to the measurement uncertainties is hard to avoid, however the use of different instruments consistent with ISO 7726 [7] requirements results in compatible values. In addition, many researchers performed field investigation to assess the indoor thermal environment of commercial buildings, including office buildings [8-10], airport terminal buildings [11-13], school buildings [14], [15] and etc.
AC C
EP
TE D
As for the winter thermal environment of residential buildings in the HSCW area of China, which is the topic of this paper, Yoshino et al. [16], [17] measured the winter indoor thermal environment of 26 households in Shanghai, Chongqing and Changsha, three major cities of HSCW area, during the year between 1998 and 2003. It was found that the winter indoor environment of Shanghai, Chongqing and Changsha is far from ASHRAE comfort zone, with internal temperatures averagely 3-6 oC lower than cities located in other parts of China. These research supported the complaints of HSCW residents about the poor winter indoor environment. Yoshino predicted that with the economic growth and the accompanying increase of household income, the poor indoor environment will be improved. Shipworth [18] found no significant change of thermostat settings in English houses between 1984 and 2007 after a careful analysis of two cross-sectional housing survey database, INT84 and CARB07. What is the case in China’s HSCW area, whether the poor indoor thermal environment has been improved during the past 15 years still remains unclear. The household thermal comfort requirements, including the heating temperature, heating duration and etc., significantly influence the residential heating energy consumption [19-22]. Scrutinizing the driving forces behind the household heating requirements is helpful to predict the residential heating energy consumption trend, to identify potential ways to curtail the rising residential heating energy consumption and to develop related policies. Frederiks et al.’s review illustrated that the occupants’ thermal comfort requirements are determined by both individual and contextual variables [23]. How the socio-demographic, psychological and building efficiency variables influenced the internal temperature and household energy consumption has been investigated [24-26]. It was found that the number of occupants, household income, occupant age, building efficiency level all have statistically significant effects on daily mean internal temperature. Shipworth et al.’s research [27] demonstrated that adding heating control systems did not reduce the maximum internal temperature or the duration of heating operation, which is contrary to what the UK government assumed when developing related regulations (e.g. SAP), policies (e.g. DEFRA) and designing 2
ACCEPTED MANUSCRIPT programmers (e.g. of the Energy Saving Trust).
RI PT
The goals of this research are twofold. Firstly, the winter indoor thermal environment and occupants’ heating behavior of urban residential buildings in HSCW area will be investigated by field measuring of 54 households. This part will be presented in Section 2. Secondly, how the dwelling internal temperature is determined, or what factors significantly affect HSCW residents’ thermal comfort demands will be explored by statistics analysis of the survey and measurement data. This research will be presented in Section 3. Conclusions are drawn in Section 4 of the paper.
SC
2. Winter Indoor Thermal Environment Investigation Field measurements of winter indoor thermal environment were performed in 16 residential communities of three major cities in the HSCW climate region. All the dwellings surveyed in this research are high rise apartments, with the dwelling area varying between 80 and 120 m2. Table 1 summarizes the key information of this field study. Figure 1 compares the weather condition of Nanjing and Shanghai. Suzhou locates between Nanjing and Shanghai. It can be observed that the outdoor climate of Nanjing and Shanghai is very similar.
TE D
Number of houses for measurement 10 8 5 4 3 2 2 3 3 4 4 2 4
40 Daily Average Outdoor Temperature
Heating system
City
Individual heating Individual heating Individual heating Individual heating District heating Individual heating District heating Individual heating Individual heating Individual heating Individual heating Individual heating Individual heating
Shanghai Shanghai Suzhou Suzhou Suzhou Suzhou Nanjing Nanjing Nanjing Nanjing Suzhou Suzhou Suzhou
AC C
EP
Residential community No.1 No.2 No.3 No.4 No.5 No.6 No.7 No.8 No.9 No.10 No.11 No.12 No.13
M AN U
Table 1 Summary of Winter Indoor Thermal Environment Filed Measurement
30
Nanjing Shanghai
20 10 0 1/1
2/1
3/1
4/1
5/1
6/1
7/1
-10
8/1
9/1
10/1
11/1
12/1
Date Figure 1 Temperature of the Typical Meteorological Year of Nanjing and Shanghai (data source: [28]) 3
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
This measurement, which lasted throughout the whole heating season of 2014/15 (from late November to middle March), was performed under the guidance of ASHRAE Standard 55-2013 [29]. AOSONG GSP 958 temperature/humidity sensors were utilized to record the internal temperature and relative humidity every five minutes. Table 2 compared the instrumentation valid range and accuracy of AOSONG GSP 958 and the criteria required by ASHRAE Standard 55-2013. It can be seen that AOSONG GSP 958 basically meets the requirement of both ASHRAE [29] and ISO [33] standard, except for the air temperature measurement accuracy, which is ±0.5oC rather than the required value ±0.2oC. 27 households were selected randomly for heating behavior survey. Around 50% of households in each community were selected to guarantee the representativeness of the occupant behavior survey. Only the households equipped with the individual heating facilities were selected for this survey since the central heating is operated full-time-full-space. EFRD S-300 and S-350 smart meters, which were able to measure the real-time electricity consumption of one single electrical appliance, were utilized to record the time when the heating is on. The temperature/humidity sensors were installed at around 1.1 meter height and far away from electrical appliances, without exposure to heat sources. In this research, local discomfort including radiant asymmetries, vertical air temperature gradients, draught rate and etc. has not been measured since local discomfort is not a significant factor in severe thermal environment which is the case in this research since the indoor thermal environment is far from comfort during the majority of occupied hours. In the ISO 7730 [30] and EN 15251 [31] Standards, local discomfort is not considered when the absolute value of PMV (Predicted Mean Vote) [32] is beyond 0.7. Table 2 Key Parameters of Instrument
Instrument
AOSONG GSP 958
Relative humidity Electricity consumption Electricity consumption
AOSONG GSP 958 EFRD S-300 EFRD S-350
Valid range by ISO Standard 7726-2001
Uncertainty by ISO Standard 7726-2001
[33]
[33]
±0.5oC
10 ~ 40 oC
±0.2oC
10 ~ 40 oC
Required: ±0.5oC; Desirable: ±0.2oC
±5%
25% ~ 95%
±5%
/
/
0 ~ 3000 W
±5 W
/
/
/
/
0 ~ 3500 W
±5 W
/
/
/
/
-20 ~ 70 oC
EP
Air temperature
Valid range
TE D
Measured parameter
Valid range Uncertainty by by Uncertai ASHRAE ASHRAE nty Standard Standard [29] 55-2013 55-2013 [29]
AC C
0 ~ 99.9%
2.1 Overview In figure 2, the distributions of internal and external temperature/relative humidity readings during occupied hours2 are presented. Bimodal distributions of internal temperature can be observed in both the living room and bedroom, one for the absence of heating and the other for the operation of heating. However the fact that only a small proportion of internal temperature readings are higher than 18 oC (23% for the living room and 2
The occupied hours are set from 17 PM to 8 AM for the working days and 24 hours for the holidays, 4
ACCEPTED MANUSCRIPT 13% for the bedroom) illustrates that the duration of heating operation is not long. 25%
30%
Percentage
20% 10%
10%
20% 15% 10%
M AN U
Percentage
30%
100
90
80
70
60
50
40
SC
25%
10%
RI PT
Living Room Relative Humidity/%
40%
20%
30
28
20
0 4 8 12 16 20 24 Living Room Temperature/oC
10
0% 0
5%
100
90
80
70
60
50
10% 5% 100
90
80
70
60
50
22
40
-2 2 6 10 14 18 Outdoor Temperature/o C
15%
0%
AC C
-6
20%
30
0%
25%
20
10%
40
Bedroom Relative Humidity/%
0
20%
EP
30%
30
28
TE D
40%
4 8 12 16 20 24 Bedroom Temperature/o C
20
0
0
10
0%
0%
Percentage
Percentage
15%
5%
0%
Percentage
20%
10
Percentage
40%
Outdoor Relative Humidity/%
Figure 2 Internal and External Temperature and Relative Humidity Distributions
2.2 Indoor thermal comfort The methods to assess the indoor thermal environment depend on whether the building is mechanically or naturally conditioned. For mechanically conditioned environment, steady state thermal comfort model, proposed by Fanger [32], is widely accepted in related comfort standards [29], [30]. For naturally conditioned buildings, de Dear and Brager [34] proposed the adaptive thermal comfort model, which was also utilized in ASHRAE thermal comfort standard [29]. The majority of the cases we investigated are neither mechanically conditioned nor naturally conditioned. Basically, they belong to mix-mode buildings [35], which means they utilized natural ventilation when natural ventilation is able to provide adequate indoor thermal comfort and 5
switched to air condition when naturalACCEPTED ventilation is MANUSCRIPT insufficient to guarantee comfort. Different researchers have different opinions when the steady state or the adaptive model should be utilized in mix-mode buildings [36], [37] . However this debate is very complicated and beyond the topic of this paper. In this research, ASHRAE comfort zone [29], based on the steady state thermal comfort model, is used to evaluate the indoor thermal environment.
RI PT
Previous research [16], [38-40] found that residents living in HSCW area are accustomed to wearing heavy clothes indoors in winter, with clothing insulation levels varying between 1.15 and 1.42 Clo units. Therefore in this research, the comfort zone is calculated setting the Clo unit at 1.2. To draw Figure 3, operative temperature, rather than air temperature, is needed. The operative temperature can be calculated by Equation (1) as below,
to = Ata + (1 − A)tr
SC
Equation (1) [29]
Where, to is the operative temperature, ta is the average air temperature and tr is the mean radiant
M AN U
temperature. The value of A is determined according to the average air speed, i.e. A equals to 0.5 when the air speed is less than 0.2m/s, equals to 0.6 when the air speed is between 0.2m/s and 0.6m/s, and equals to 0.7 when the air speed is between 0.6m/s and 1.0m/s.
AC C
EP
TE D
In this research, the operative temperature is assumed to approximate air temperature. There are two reasons behind this approximation. Firstly, from the occupant heating behavior survey (presented in detail in Section 2.3), it was found in the HSCW area of China, the residential domestic heating facilities is turned off during the most period of time. When the heating facilities is off, the average air temperature is approximate to the mean radiant temperature, accordingly approximate to the operative temperature, i.e. ta=tr=to. Secondly, compared with radiant heating terminals like floor heating or radiator, convective terminals were much more widely utilized in this area (81% vs. 19% in this research). When the convective terminals is on, the average air speed is relatively high, likely to exceed 0.6m/s, which means the operative temperature is significantly depend on the air temperature, since the weighting factors of ta and tr is 0.7 and 0.3 respectively when the air speed is above 0.6m/s. Figure 3 is drawn based on this approximation.
Figure 3 ASHRAE Steady State Thermal Comfort Model [29]
As shown in Figure 3, even the Clo. adjustment has already been considered, the indoor thermal 6
ACCEPTED MANUSCRIPT environment of dwellings in the HSCW area during occupied hours is far from comfort given the ASHRAE steady state thermal comfort model. The majority of temperature/humidity readings fall outside the comfort zone due to the low internal temperature.
Bedroom
30% Percentage
30% 20%
20%
Living room Bedroom
M AN U
Percentage
40%
Living room
SC
40%
RI PT
2.3 Occupant heating behavior Unlike the full-time-full-space operated district heating, the individual heating is operated part-time-partspace. Figure 4 illustrates when and under what temperature, domestic heating is triggered. For the living room, the majority of heating (30%) is triggered between 16 PM to 19 PM, the time when residents are back home from work. While for the bedroom, heating is most likely (34%) to be triggered between 19 PM to 22 PM, the time when residents plan to go to sleep. As for the triggering temperature, the majority of residents choose to turn on heating when the temperature is between 10 to 14 oC. 10 oC seems to be the lowest acceptable internal temperature without heating.
10%
10%
0%
0% 6 and below
6-8
8-10
1-4
10-12 12-14 14-16 16-18 18 and above Heating Triggering Temperature/o C
4-7
7-10 10-13 13-16 16-19 19-22 Heating Triggering Time
22-1
TE D
Figure 4 Heating Triggering Temperature and Time
Heating duration/h
AC C
10
EP
The heating duration is presented in Figure 5. The average heating duration per operation is 2.3 hours for the living room and 2.8 hours for the bedroom. From this it can be deduced that occupants tend to turn off heating when they fall asleep. The average heating duration per day is 1.9 hours for the living room and 1.4 hours for the bedroom. It can be predicted that the residential heating energy usage would be lifted significantly if the current part-time-part-space heating operation mode was changed to the full-time-fullspace operation, which is the case in north China
Max
8
+STD
6
Average
4
-STD
2
Min
0 Living room
Bedroom
Living room
per Operation
Bedroom
per Day
Figure 5 Duration of Heating Operation 7
ACCEPTED Figure 6 presents the internal temperatures during MANUSCRIPT heating operation. The maximum temperature during heating is determined by the heating set-point temperature. The average maximum temperature is 20.6 oC for the living room and 21.1 oC for the bedroom. The average difference between the maximum temperature and average temperature during heating operation is 5.5 oC for the living room and 5.6 oC for the bedroom. The operation of heating lifts the average internal temperature 1.6 oC (13.5 to 15.1 oC) for the living room and 2.8 oC (12.7 to 15.5 oC) for the bedroom, which is not as significant as we assume. One explanation for this is the short duration of heating operation.
RI PT
35
+STD
25
Average -STD
15
SC
Temperature/oC
Max
Min
5
Living room Bedroom
M AN U
Living room Bedroom Maximum Temperature during Heating
Average Temperature during Heating
Figure 6 Internal Temperature during Heating Operation
EP
TE D
Another reason for this significant difference between maximum temperature and average temperature during heating and the not marked internal temperature elevation by heating is the slow thermal response of the heating system. Whether the indoor thermal environment can be elevated quickly to the comfort zone is crucial for the intermittent heating especially in HSCW area since the average duration of heating operation is short in this region. Wang et al. [41] utilized thermal time constant to evaluate the thermal dynamic performance of intermittent heating system. The intermittent heating system with small thermal constant time is needed in HSCW region owing to its ability to elevate indoor temperature quickly.
AC C
2.4 Comparison with the past and the UK The UK has similar climate condition with HSCW region in winter, however the average internal temperature of UK is 6 oC higher (19 oC [18] compared with 13 oC), as shown in Figure 7. Comparing the internal temperature measured in 2014 with the data measured in 1998 by Yoshino et al. [16], it seems that the indoor thermal environment became worse during the past 15 years. Limited sampling size in Yoshino et al’s research might be a major reason behind this. Only 6 households were measured in Yoshino et al.’s research, some errors are likely to exist. Besides, in this field survey, the low income group (who live in the affordable house) has been added in the sample, who rarely utilize heating facilities, lowering the average internal temperature value. For these two reasons it is hard to say whether there is significant improvement of the indoor thermal environment in the HSCW area during the past 15 years.
8
ACCEPTED MANUSCRIPT 35 Max
Internal Temperature/o C
30
+STD
25 20
Average
15
-STD
10
Min
0 Internal space
Living room
UK, 2007
Bedroom
Shanghai, 1998
Living room
RI PT
5 Bedroom
Shanghai, Suzhou and Nanjing, 2014
SC
Figure 7 Internal Temperature Comparison [16], [18]
M AN U
3. Analysis of the Driving Forces of Heating Requirements In this section, statistical analysis has been utilized to scrutinize the factors significantly influencing the indoor thermal requirements of residents in HSCW area.
TE D
3.1 Methodology The usual Ordinary Least Square (OLS) method, which is the most widely used in statistical analysis, is based on the homoskedasticity assumption. However, when the heteroskedasticity is present amongst residuals, the usual OLS test is invalid [42]. In this research, the White Test is firstly performed to check for the presence of heteroskedasticity amongst residuals, to decide whether the heteroskedasticity-robust statistical analysis method is needed in this analysis.
AC C
EP
Given the presence of heteroskedasticity, OLS is no longer the Best Linear Unbiased Estimator (BLUE) and no longer asymptotically efficient [42]. If the form of heteroskedasticity is known, the Weighted Least Square (WLS) Estimator is more efficient than the OLS Estimator. However, the form of heteroskedasticity is unknown in this research. Though the Feasible Generalized Least Squares (FGLS) method is able to handle the problem of unknown heteroskedasticity form by assuming a form of the heteroskedasticity. However, if the assumption of the heteroskedasticity form is not correct, it is not guaranteed that FGLS is more efficient than OLS [42]. Besides, since the OLS remains consistent even when the heteroskedasticity is present [42]. The consistence of OLS guarantees that OLS approaches to the true value in large sample size. Therefore in this research, the OLS method is utilized to estimate the effect of variant factors on heating requirements, and heteroskedasticity-robust test method is utilized to test whether this effect is statistically significant. Two datasets have been used to perform the statistical analysis in this research. One dataset is from the field measurement. The description of the explained and explanatory variables of the on-site measurement dataset is shown in Table 3. The Degree of Freedom (DF) of this analysis equals to 46 (n=54, k=7, DF=n-k-1). The sample size can be considered as large when DF is more than 30 [42]. The other dataset is from the survey. This survey is carried through website. Heating set-point temperature and heating duration were asked to characterize occupants’ heating demand. Socio-demographic and occupancy related questions including education background, household income, previous thermal experiences, and building and heating system related questions, including year of construction and heating system were asked to characterize the subjects 9
ACCEPTED MANUSCRIPT surveyed. The description of the statistics of the survey dataset is shown in Table 4, with the Degree of Freedom equaling to 515 (n=525, k=9, DF=n-k-1). Table 3 Statistics of the On-site Measurement Dataset
Variable name
Type
O
Dummy
Standard deviation 3.87
Proportion
66.7% 33.3%
HH_C
T
GB
Dummy
91.5% 18.5%
Dummy
R
77.8% 22.2%
Dummy 38.9% 61.1%
HS
Dummy 9.2% 77.8% 13.0%
EP
AC C
7.5% 92.5%
Dummy
TE D
Building and heating system Building not a certificated green building certificated green building3 Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
Mean 13.16
SC
Explanatory variables Variable description Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied
Number of obs. 54
RI PT
Type scale
M AN U
Explained variable Variable description Average internal temperature
Table 4 Statistics of the Survey Dataset
Explained variable Variable description
Type
Number of obs.
Mean
Heating set-point temperature Heating duration_Weekday Heating duration_Weekend
scale scale scale
525 525 525
19.38 5.23 7.85
Standard deviation 3.19 4.24 5.15
Explanatory variables 3
Certificated by Chinese, including one-star label, two-star label and three-star label. 10
5.6% 85.3% 9.1% Dummy 14.3% 45.3% 40.4%
SC
Dummy
YC
HS
33.2% 19.4% 21.7% 25.7%
Dummy 50.1% 49.9% Dummy
TE D
Building and heating system Year of construction Before 2000 After 2001 5 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
TE
Proportion
Dummy
M AN U
Thermal experience Experience of living in the north 4 No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
Type
RI PT
ACCEPTED MANUSCRIPT Variable description Variable name Socio-demographic and occupancy Education E Middle school/vocational school and less Junior college/college Graduates Income I Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
23.3% 69.1% 7.6%
AC C
EP
3.2 Results and discussion Data analysis toolkit STATA 11 is utilized to perform the statistical analysis in this research. The results are shown in Table 5 and Table 6. The heteroskedasticity-robust t-Statistics are shown in parenthesis. If 5% (10%) is chosen as the significant level, the effect is statistically significant only when the absolute value of t-Statistics is no less than 1.67 (1.30). Table 5 Result of the Statistical Analysis of the On-site Measurement Dataset
Explained variable Number of observation R-square Socio-demographic and occupancy Occupation
4 5
Average temperature 54 0.840
Refers to north of the Qin-Huai line, where central heating is provided Year 2001 is the year when the first edition of
cold winter zone> was officially issued and implemented. Therefore buildings constructed after year of 2001 can be seen as with higher energy efficiency, higher thermal insulation level, and etc. 11
-2.63 (1.81)
SC
--0.51 (-1.02)
--11.13 (-7.69) -6.89 (-4.76)
M AN U
Building and heating system Building not a certificated green building certificated green building Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
RI PT
ACCEPTED MANUSCRIPT Not a university faculty -University faculty -0.43 (-0.66) Household structure With child aged < 12 -Without child aged < 12 4.10 (5.37) Tenure type Owner occupied -Tenant occupied -2.33 (-1.46)
Table 6 Result of the Statistical Analysis of the Survey Dataset
Heating set-point temperature 525 0.589
525 0.405
525 0.382
-0.10 (0.16) -0.21 (-0.29)
--0.24 (-0.42) 1.06 (1.07)
-1.67 (2.43) 2.32 (1.08)
-0.60 (1.51) 1.65 (3.82)
-0.37 (0.85) 1.04 (1.89)
-1.42 (2.39) 2.00 (3.02)
-0.01 (0.01) 0.22 (0.55) 0.80 (2.13)
-0.03 (0.06) 1.33 (2.75) 1.74 (3.09)
--0.21 (-0.38) 1.11 (1.76) 1.29 (1.96)
TE D
Explained variable
AC C
EP
Number of observation R-square Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
Experience of living in the north No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
12
Heating Heating duration_Weekday duration_Weekend
ACCEPTED MANUSCRIPT -0.07 (0.24)
-0.05 (0.14)
-0.37 (0.83)
-1.14 (3.40) -0.03 (-0.06)
--1.75 (-2.29) -1.06 (-1.41)
--1.52 (-2.25) -1.83 (-2.10)
RI PT
Year of construction Before 2000 After 2001 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
SC
Socio-demographic and occupancy effects From Table 5 and Table 6, it can be seen that education and occupation do not significantly influence occupants’ indoor thermal requirements. It was believed that increased education results in more pro-environmental behaviors [43], [44]. However this has not been observed in this research. Contrarily, higher education level leads to longer heating durations on weekends. Barr ascribed this phenomenon to the “knowledge-action gap” [45]. Though residents with higher education level care more about the environment and energy issue, however their actions always lag their values.
M AN U
Income is a key determinant of residents’ heating requirements. The households with income more than 100,000 RMB set averagely 1.65 oC higher heating temperature and 1-2 hours longer heating duration compared with those earning less than 50,000 RMB. Tenure type can be considered as the instrumental variable of the household income. It can be seen that the average difference in internal temperatures between an owner occupied dwelling and a tenant occupied dwelling is approximately 2.33 oC.
TE D
The presence of a child less than 12 years old is also a significant driver of higher heating requirements, increasing the mean internal temperature by an average of 4.10 oC compared with those households without a child.
AC C
EP
Building and heating system effects Since no physical survey of building energy efficiency has been performed in this research, other variables are used as instrumental variables, to characterize the building efficiency. The first instrumental variable of building efficiency we chose is whether it is a certificated green building. Since the Design Standard for Energy Efficiency of Residential Buildings in Hot Summer and Cold Winter Zone [46] was put into effect at the year of 2001, the variable, whether it is constructed before or after 2001, is able to identify the building efficiency. The rebound effect is observed in HSCW area. Since it was found that the building energy efficiency does not influence the residents heating requirements (heating set-point temperature and heating duration), but statistically significantly improve the indoor thermal environment (averagely 2.63 oC higher internal temperatures). This result confirms Sorrell et al.’s finding that half of the temperature rebound effect is not due to households demanding higher standard of comfort, but simply because energy efficiency measures such as higher insulation, increasing air-tightness level and etc. lead to a warmer indoor environment [47]. Compared with the central heating, if the air source heat pump is utilized in the dwelling, the mean internal temperature will be lowered by 11.13 oC, and if the gas fueled floor heating is utilized, the mean internal temperature will be lowered by 6.89 oC. This huge variance is due to the fact that central heating is operated full-time-full-space to satisfy the heating requirements of all the households. However, the air source heat 13
ACCEPTED MANUSCRIPT pump or gas fueled floor heating is operated part-time-part-space according to the household’s individual heating demand. The set-point temperature of central heating is 1.14 oC lower than that of the air source heat pump, which indicates that the central heating is likely to supply more heat than that actual needed by the occupants.
RI PT
Thermal experience effects The thermal adaptive model predicted that the previous thermal experience is likely to alter residents’ thermal comfort sensation and thermal requirements through behavioral adjustment, physiological acclimatization and adjustment of psychological expectations [48], [49]. Yu’s chamber experiment confirmed this prediction [50] and found that subjects coming from Beijing, where central heating is provided, complained of greater cold discomfort than those from Shanghai, where central heating is unavailable, even when they are exposed to the same environment.
M AN U
SC
For this analysis thermal experience is separated into four discrete bands, no, short, medium and long experience of living in the north where central heating is provided. It is found that living in the north for a short period of time (less than 1 month) has statistically insignificant impact on residents’ heating requirements, indicating that the thermal experience effects take time to come about. However, living in the north for a long period of time (more than 3 months) markedly lifted residents’ heating requirements, resulting in a higher heating set-point temperature and longer heating duration.
EP
TE D
4. Conclusion In this research, the winter indoor thermal environment of dwellings in China’s Hot Summer – Cold Winter climate region has been investigated. The on-site investigation measured the internal temperature and humidity of 54 households, 27 of them were selected for heating behavior survey, in which the heating operation time and duration were recorded by smart meters. The average internal temperature is 13.5 oC for the living room and 12.7 oC for the bedroom. The majority of the temperature/humidity readings fall outside the ASHRAE steady state thermal comfort zone due to the low internal temperatures. The average internal temperature of dwellings in HSCW climate region is 6 oC lower than that in the UK, which has similar winter climate condition. Besides, no significant improvement of the indoor thermal environment has been observed in the HSCW area during the past 15 years.
AC C
Domestic heating is triggered by both time and temperature in HSCW area. The majority of heating is consumed between 16PM and 22PM (55% for the living room and 46% for the bedroom). In terms of the heating triggering temperature, the majority of residents choose to turn on heating when the temperature is between 10 to 14 oC (65% for the living room and 50% for the bedroom), indicating that 10 oC seems to be the lowest acceptable internal temperature without heating. Domestic heating is operated part-time-part-space in HSCW area. Each heating operation lasts averagely 2.3 hours for the living room and 2.8 hours for the bedroom. The average heating duration per day is 1.9 hours for the living room and 1.4 hours for the bedroom throughout the whole heating season. The short heating operation duration is the main reason for the poor indoor thermal environment as well as the relative low residential heating energy consumption in the HSCW area. Additionally, the driving forces of heating requirements of residents in HSCW area have been scrutinized. It is found that higher household income, the presence of child and not short experience (no less than 3 months) 14
ACCEPTED of living in the north, where central heating is MANUSCRIPT provided, significantly influence occupants’ heating requirements, indicating that with the further economic growth and more frequent population mobility, a higher heating set-point temperature and longer heating duration will be demanded in this area. It was also found that a short living experience in the north will not significantly improve occupant’s heating demand, demonstrating that thermal experience takes time to affect occupants’ thermal preferences.
RI PT
The rebound effect is observed in this research. Higher building energy efficiency results in a warmer winter indoor environment. However, the building energy efficiency level does not influence the residents heating requirements.
SC
The mean internal temperature of the residence equipped with central heating is approximately 11 oC higher than those equipped with individual air source heat pump. This variance is due to the longer heating operation duration rather than the higher heating set-point temperature of the central heating system. In fact, the set-point temperature of central heating is 1.14 oC, which is statistically significantly lower than that of the air source heat pump.
M AN U
Acknowledgement This research is supported by the National Natural Science Foundation of China (Grant number 51561135001) and the Innovative Research Groups of the National Natural Science Foundation of China (Grant number 51521005). Great thanks to Zhe Li, Zhen Zhao and Zufeng Pei for their contribution to the field measurement and survey. Special thanks to all the residents who provide kind help for the on-site measurement.
TE D
Reference
AC C
EP
[1] Los Angeles Times. For central heat, China has a north-south divide at Qin-Huai line. Retrieved June 20, 2015, from http://www.latimes.com/world/asia/la-fg-china-heat-20141115-story.html [2] Tsinghua University Building Energy Research Center 2013. 2013 Annual Report on China Building Energy Efficiency. Beijing: China Architecture & Building Press. (in Chinese) [3] State Bureau of Technical Supervision, Minister of Housing and Urban-Rural Development, Standard of Climatic Regionalization for Architecture, GB 50178-93. (in Chinese) [4] American Society of Heating, Refrigerating and Air-Conditioning Engineers, Thermal Environmental Conditions for Human Occupancy, ANSI/ASHRAE Standard 55-2013 [5] D'Ambrosio Alfano, F.R., Olesen, B.W., Palella B.I., Riccio G. Thermal Comfort: Design and Assessment for Energy Saving. Energy and Buildings 81 (2014), 326-336. [6] M. Dell'Isola, A. Frattolillo, B.I. Palella, G. Riccio, Measurement uncertainties influence on the thermal environment assessment, International Journal of Thermophysics 33 (8-9) (2012) 1616-1620 [7] International Standardization Organization, Ergonomics of the Thermal Environment—Instruments for Measuring Physical Quantities, ISO 7726: 2002 [8] Z.F. Pei, B.R. Lin, Y.C. Liu, Y.X. Zhu, Comparative study on the indoor environment quality of green office buildings in China with a long-term field measurement and investigation, Building and Environment 84 (2015) 80-88 [9] T. Goto, T. Mitamura, H. Yoshino, A. Tamura, E. and Inomata, Long term field survey on thermal adaptation in office buildings in Japan, Building and Environment, 42 (2007), 3944–3954 [10] M.H. Luo, B. Cao, J. Damiens, B.R. Lin, Y.X. Zhu, Evaluating thermal comfort in mixed-mode buildings: A field study in a subtropical climate, Building and Environment, 88 (2014), 46–54 15
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[11] Z. Wang, H.T. Zhao, B.R. Lin, Y.X. Zhu, Q. Ouyang, J. Yu, Investigation of indoor environment quality of Chinese large-hub airport terminal buildings through longitudinal field measurement and subjective survey, Building and Environment 94 (2015) 593-605 [12] A. Kotopouleas and M. Nikolopoulou, Understanding thermal comfort conditions in airport terminal Buildings, Proceedings of 8th Windsor Conference: Counting the Cost of Comfort in a changing world, Cumberland Lodge, Windsor, UK, 10-13 April 2014 [13] C.A. Balaras, E. Dascalaki, A. Gaglia, K. Droutsa, Energy conservation potential, HVAC installations and operational issues in Hellenic airports, Energy and Buildings 35 (2003) 1105–1120 [14] L.D. Pereira, D. Raimondo, S.P. Corgnati, M.G. da Silva, Assessment of indoor air quality and thermal comfort in Portuguese secondary classrooms: Methodology and results, Building and Environment 81 (2014) 69-80 [15] R. Yao, J. Liu, B. Li, Occupants’ adaptive responses and perception of thermal environment in naturally conditioned university classrooms, Applied Energy, 87 (2010), 1015–1022. [16] H. Yoshino,S. Guan, Y.F. Lun, A. Mochida, T. Shigeno, Y. Yoshino, Q.Y. Zhang, Indoor thermal environment of urban residential buildings in China: winter investigation in five major cities, Energy and Buildings 36 (2004) 1227–1233 [17] H. Yoshino, Y. Yoshino, Q.Y. Zhang, A. Mochida, N.P. Li, Z.H. Li, H. Miyasaka, Indoor thermal environment and energy saving for urban residential buildings in China, Energy and Buildings 38 (2006) 1308–1319 [18] M. Shipworth, Thermostat settings in English houses: No evidence of change between 1984 and 2007, Building and Environment 46 (2011) 635-642 [19] A.H. Neto, A.S. Fiorelli, Comparison between detailed model simulation and artificial neural network for forecasting building energy consumption, Energy and Buildings 40 (2008) 2169–2176 [20] Z. M. Gill, M. J. Tierney, I. M. Pegg, N. Allan, Low-energy dwellings: the contribution of behaviours to actual performance. Build Res Inform 2010 38(5): 491–508. [21] O.G. Santin, Behavioural Patterns and User Profiles related to energy consumption for heating, Energy and Building, 43 (2011): 2662–2672 [22] Z. Yu, B.C.M. Fung, F. Haghighat, H. Yoshino, E. Morofsky, A systematic procedure to study the influence of occupant behavior on building energy consumption, Energy and Building, 43 (2011): 1409–1417 [23] E. R. Frederiks, K. Stenner and E. V. Hobman, The Socio-Demographic and Psychological Predictors of Residential Energy Consumption: A Comprehensive Review, Energies 8 (2015), 573-609 [24] S. Kelly, M. Shipworth, D. Shipworth, M. Gentry, A. Wright, M. Pollitt, D. Crawford-Brown, K. Lomas, Predicting the diversity of internal temperatures from the English residential sector using panel methods, Applied Energy 102 (2013) 601–621 [25] W. Abrahamse, L. Steg, How do socio-demographic and psychological factors relate to households’ direct and indirect energy use and savings, Journal of Economic Psychology, 30 (2009) 711-720 [26] E.R. Frederiks, K. Stenner, E.V. Hobman, Household energy use: Applying behavioural economics to understand consumer decision-making and behavior, Renewable & Sustainable Energy Review, 41 (2015) 1385-1394 [27] M. Shipworth, S. K. Firth, M. I. Gentry, A. J. Wright, D. T. Shipworth and K. J. Lomas, Central heating thermostat settings and timing: building demographics, BUILDING RESEARCH & INFORMATION (2010) 38(1), 50–69 [28] Meteorological Information Center of China Meteorological Administration, Department of Building Science Tsinghua University, (2005). Meteorological data set for built thermal environment analysis in China. Beijing: China Architecture & Building Press. (in Chinese) [29] American Society of Heating, Refrigerating and Air-Conditioning Engineers, Thermal Environmental Conditions for Human Occupancy, ANSI/ASHRAE Standard 55-2013 [30] International Standardization Organization, Ergonomics of the Thermal Environment – Analytical Determination and Interpretation of Thermal Comfort Using Calculation of the PMV and PPD Indices and Local Thermal Comfort, ISO 7730: 2005. 16
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[31] European Committee for Standardization, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, EN 15251: 2007. [32] P.O. Fanger, Calculation of thermal comfort: introduction of a basic comfort equation, ASHRAE Transactions 73 (2) (1967) 1–20. [33] International Standardization Organization, Ergonomics of the thermal environment - Instruments for measuring physical quantities, ISO 7726: 2001. [34] R. de Dear, G.S. Brager, The adaptive model of thermal comfort and energy conservation in the built environment, Int J Biometeorol 45 (2001) 100–108 [35] G. Brager, Mixed-mode cooling, ASHRAE Journal., 48 (2006) 30–37. [36] D. Rowe, Thermal comfort in a naturally ventilated environment with supplementary cooling and heating, Architectural Science Review, 47 (2004), 131–140. [37] M.P. Deuble, and R.J. de Dear, Mixed-mode buildings: a double standard in occupants’ comfort expectations, Building and Environment, 54 (2012), 53–60. [38] B. Z. Li, J. Liu, R. M. Yao, 2007, investigation and analysis on classroom thermal environment in winter in Chongqing, HV&AC, 2007(5): 115-117 (in Chinese) [39] J. H. Li, L. Yang, J. P. Liu, 2008, adaptive thermal comfort model for hot summer cold and winter zone, HV&AC 2008(7): 20-24. (in Chinese) [40] J.S. Lan (2013) Human Adaptation to Thermal Environment in Free Running Building, Ph.D. Thesis: Chongqing (in Chinese) [41] Z. Wang, B.R. Lin, Y.X. Zhu, Modeling and measurement study on an intermittent heating system of a residence in Cambridgeshire, Building and Environment 92 (2015) 380-386 [42] Wooldridge JM. Introductory econometrics: a modern approach. China Renmin University Press; 2010. [43] W. Poortinga, L. Steg, C. Vlek, C. Values, environmental concern and environmental behavior: A study into household energy use. Environment Behavior. 36 (2004), 70–93. [44] G. Nair, L. Gustavsson, K. Mahapatra, Factors influencing energy efficiency investments in existing Swedish residential buildings. Energy Policy 38 (2010), 2956–2963. [45] S. Barr, A.W. Gilg, N. Ford, The household energy gap: Examining the divide between habitual- and purchase-related conservation behaviours. Energy Policy 33 (2005), 1425–1444. [46] MOHURD, 2001, Design standard for energy efficiency of residential buildings in hot summer and cold winter zone, JGJ 134-2001 (in Chinese) [47] S. Sorrell, J. Dimitropoulos, M. Sommerville, Empirical estimates of the direct rebound effect: a review. Energy Policy 37 (2009): 1356-71. [48] G.S. Brager, R. de Dear, (1998) Thermal adaptation in the built environment: a literature review, Energy and Buildings 27 (1998): 83-96. [49] R. J. de Dear et al., Progress in thermal comfort research over the last twenty years, Indoor Air 2013; 23: 442–461 [50] J. Yu, G. Cao, W. Cui, Q. Ouyang, Y. Zhu People who live in a cold climate: thermal adaptation differences based on availability of heating, Indoor Air 2013; 23: 303–310
17
ACCEPTED MANUSCRIPT
Table 1 Summary of Winter Indoor Thermal Environment Filed Measurement
Heating system
City
Individual heating Individual heating Individual heating Individual heating District heating Individual heating District heating Individual heating Individual heating Individual heating Individual heating Individual heating Individual heating
Shanghai Shanghai Suzhou Suzhou Suzhou Suzhou Nanjing Nanjing Nanjing Nanjing Suzhou Suzhou Suzhou
SC
RI PT
Number of houses for measurement 10 8 5 4 3 2 2 3 3 4 4 2 4
M AN U
Residential community No.1 No.2 No.3 No.4 No.5 No.6 No.7 No.8 No.9 No.10 No.11 No.12 No.13
Table 2 Key Parameters of Instrument
Instrument
AOSONG GSP 958
Relative humidity Electricity consumption Electricity consumption
AOSONG GSP 958 EFRD S-300 EFRD S-350
Valid range by ISO Standard 7726-2001
Uncertainty by ISO Standard 7726-2001
[33]
[33]
-20 ~ 70 oC
±0.5oC
10 ~ 40 oC
±0.2oC
10 ~ 40 oC
Required: ±0.5oC; Desirable: ±0.2oC
0 ~ 99.9%
±5%
25% ~ 95%
±5%
/
/
0 ~ 3000 W
±5 W
/
/
/
/
0 ~ 3500 W
±5 W
/
/
/
/
AC C
EP
Air temperature
Valid range
TE D
Measured parameter
Valid range Uncertainty by by Uncertai ASHRAE ASHRAE nty Standard Standard [29] 55-2013 [29] 55-2013
Table 3 Statistics of the On-site Measurement Dataset
Explained variable Variable description Average internal temperature
Type scale
Explanatory variables Variable description
Variable name
Number of obs. 54
1
Mean 13.16
Standard deviation 3.87
Type
Proportion
O
Dummy 66.7% 33.3%
HH_C
Dummy 7.5% 92.5%
T
Dummy
RI PT
91.5% 18.5%
GB
Dummy
SC
Building and heating system Building not a certificated green building certificated green building1 Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
ACCEPTED MANUSCRIPT
R
HS
77.8% 22.2%
Dummy
M AN U
Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied
38.9% 61.1%
Dummy 9.2% 77.8% 13.0%
Explained variable Variable description
EP
Heating set-point temperature Heating duration_Weekday Heating duration_Weekend
TE D
Table 4 Statistics of the Survey Dataset
AC C
Explanatory variables Variable description Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB 1
Type
Number of obs.
Mean
scale scale scale
525 525 525
19.38 5.23 7.85
Standard deviation 3.19 4.24 5.15
Variable name
Type
Proportion
E
Dummy 5.6% 85.3% 9.1%
I
Dummy 14.3% 45.3% 40.4%
Certificated by Chinese, including one-star label, two-star label and three-star label. 2
ACCEPTED MANUSCRIPT Dummy 33.2% 19.4% 21.7% 25.7%
YC
Dummy
50.1% 49.9%
Dummy
SC
HS
23.3% 69.1% 7.6%
M AN U
Building and heating system Year of construction Before 2000 After 2001 3 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
TE
RI PT
Thermal experience Experience of living in the north 2 No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
Table 5 Result of the Statistical Analysis of the On-site Measurement Dataset
AC C
EP
TE D
Explained variable Number of observation R-square Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied Building and heating system Building not a certificated green building certificated green building Room Structure Living room Bedroom
2 3
Average temperature 54 0.840
--0.43 (-0.66) -4.10 (5.37) --2.33 (-1.46)
-2.63 (1.81) --0.51 (-1.02)
Refers to north of the Qin-Huai line, where central heating is provided Year 2001 is the year when the first edition of
cold winter zone> was officially issued and implemented. Therefore buildings constructed after year of 2001 can be seen as with higher energy efficiency, higher thermal insulation level, and etc. 3
ACCEPTED MANUSCRIPT Heating system Central heating -Non-central air source heat pump -11.13 (-7.69) Non-central gas fueled floor heating -6.89 (-4.76) Table 6 Result of the Statistical Analysis of the Survey Dataset
525 0.405
525 0.382
-0.10 (0.16) -0.21 (-0.29)
--0.24 (-0.42) 1.06 (1.07)
-1.67 (2.43) 2.32 (1.08)
-0.60 (1.51) 1.65 (3.82)
-0.37 (0.85) 1.04 (1.89)
-1.42 (2.39) 2.00 (3.02)
-0.01 (0.01) 0.22 (0.55) 0.80 (2.13)
-0.03 (0.06) 1.33 (2.75) 1.74 (3.09)
--0.21 (-0.38) 1.11 (1.76) 1.29 (1.96)
-0.07 (0.24)
-0.05 (0.14)
-0.37 (0.83)
-1.14 (3.40) -0.03 (-0.06)
--1.75 (-2.29) -1.06 (-1.41)
--1.52 (-2.25) -1.83 (-2.10)
TE D
Experience of living in the north No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
AC C
EP
Year of construction Before 2000 After 2001 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
RI PT
Heating Heating duration_Weekday duration_Weekend
SC
Number of observation R-square Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
Heating set-point temperature 525 0.589
M AN U
Explained variable
4
ACCEPTED MANUSCRIPT Nanjing 30
Shanghai
20 10 0 1/1
2/1
3/1
4/1
5/1
6/1
7/1
-10
8/1
9/1
10/1
11/1
12/1
RI PT
Daily Average Outdoor Temperature（℃）
40
Date
Figure 1 Temperature of the Typical Meteorological Year of Nanjing and Shanghai (data source: [28])
20% 10%
15% 10% 5%
100
90
80
70
60
50
40
5%
Bedroom Relative Humidity/%
1
100
90
80
70
60
50
28
40
4 8 12 16 20 24 Bedroom Temperature/o C
10%
0%
AC C
0
15%
30
0%
20%
20
10%
25%
0
20%
EP
30%
30
Living Room Relative Humidity/%
TE D
40%
28
10
4 8 12 16 20 24 Living Room Temperature/oC
20
0
0
10
0%
0%
Percentage
20%
M AN U
Percentage
30%
SC
25%
Percentage
Percentage
40%
ACCEPTED MANUSCRIPT 25%
30%
Percentage
20% 10%
20% 15% 10% 5% 100
90
80
70
60
50
40
22
RI PT
-2 2 6 10 14 18 Outdoor Temperature/o C
30
0
-6
20
0%
0%
10
Percentage
40%
Outdoor Relative Humidity/%
M AN U
SC
Figure 2 Internal and External Temperature and Relative Humidity Distributions
TE D
Figure 3 ASHRAE Steady State Thermal Comfort Model [29]
40%
Bedroom
10% 0% 6 and below
Percentage
AC C
20%
6-8
Living room Bedroom
30%
EP
30% Percentage
40%
Living room
20% 10% 0% 1-4
8-10
10-12 12-14 14-16 16-18 18 and above Heating Triggering Temperature/o C
4-7
7-10 10-13 13-16 16-19 19-22 Heating Triggering Time
Figure 4 Heating Triggering Temperature and Time
2
22-1
ACCEPTED MANUSCRIPT 10
Heating duration/h
Max 8 +STD 6 Average 4
-STD
2
Min
Living room
Bedroom
Living room
per Operation
RI PT
0 Bedroom
per Day
SC
Figure 5 Duration of Heating Operation
35
25
15
5 Living room Bedroom
+STD Average -STD Min
Living room Bedroom
TE D
Maximum Temperature during Heating
M AN U
Temperature/oC
Max
Average Temperature during Heating
35
25
Max +STD
AC C
Internal Temperature/o C
30
EP
Figure 6 Internal Temperature during Heating Operation
20
Average
15
-STD
10
Min
5 0
Internal space
UK, 2007
Living room
Bedroom
Shanghai, 1998
Living room
Bedroom
Shanghai, Suzhou and Nanjing, 2014
Figure 7 Internal Temperature Comparison [16], [18]
3
ACCEPTED MANUSCRIPT
RI PT
SC M AN U
TE D
EP
The residential winter thermal environment in the HSCW area has been investigated The heteroskedasticity-robust Ordinary Least Square analysis has been utilized The average indoor temperature is 13.5 oC for living rooms and 12.7 oC for bedrooms Heating is operated part-time-part-space and triggered by both time and temperature Income, presence of child and thermal experience markedly influence heating demands
AC C
S0360-1323(16)30066-X
DOI:
10.1016/j.buildenv.2016.02.022
Reference:
BAE 4408
To appear in:
Building and Environment
Received Date: 5 January 2016 Revised Date:
21 February 2016
Accepted Date: 22 February 2016
Please cite this article as: Lin B, Wang Z, Liu Y, Zhu Y, Ouyang Q, Investigation of Winter Indoor Thermal Environment and Heating Demand of Urban Residential Buildings in China’s Hot Summer – Cold Winter Climate Region, Building and Environment (2016), doi: 10.1016/j.buildenv.2016.02.022. 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 Investigation of Winter Indoor Thermal Environment and Heating Demand of
Urban Residential Buildings in China’s Hot Summer – Cold Winter Climate Region Borong Lin1, 2*, Zhe Wang1, 2, 3, Yanchen Liu1, 2, 3, Yingxin Zhu1, 2 and Qin Ouyang1, 2 1
Department of Building Science, Tsinghua University, Beijing 100084, China Key Laboratory of Eco Planning & Green Building, Ministry of Education, Tsinghua University 3 Beijing Key Laboratory of Indoor Air Quality Evaluation and Control (Tsinghua University), Beijing, China
Corresponding email: [email protected] Corresponding phone (+86) 010 62785691
SC
*
RI PT
2
EP
TE D
M AN U
Abstract In this paper, the winter indoor thermal environment of dwellings in China’s Hot Summer – Cold Winter (HSCW) climate region has been investigated by on-site measurement. And the driving forces of the heating requirements of residents in HSCW area has been scrutinized by heteroskedasticity-robust Ordinary Least Square analysis. It is found that in HSCW area, the average internal temperature is 13.5 oC for the living room and 12.7 oC for the bedroom, leading to an uncomfortable indoor thermal environment, markedly colder than that in the UK, which has similar winter climate. Through the investigation of the occupant heating behavior, it was found that domestic heating is operated part-time-part-space and triggered by both time and temperature in the HSCW area. 10 oC is found to be the lowest acceptable internal temperature without heating. The poor indoor thermal environment observed in HSCW area is mainly due to the short heating operation duration (averagely 1.9 hours per day for the living room and 1.4 hours for the bedroom). Income, presence of child, heating system and thermal experience markedly influence occupants’ heating requirements. However, the education is not a statistically significant factor. The rebound effect is observed in HSCW area when analyzing the effects of building energy efficiency on indoor thermal environment.
AC C
Keywords Indoor thermal environment; Occupant behavior; Residential buildings; China’s Hot Summer – Cold Winter area; Field measurement and questionnaire survey 1. Introduction Domestic heating in China’s Hot Summer – Cold Winter (HSCW) 1 climate region has attracted increasing attention not only in China but also worldwide [1]. Chinese government is responsible for the supplying of central heating to dwellings only north of the Qin-Huai line. However HSCW area lies south of the line, 1
According to the Standard of Climatic Regionalization for Architecture [3], the winter climate of HSCW climate region has the
following two characteristics. Firstly, the average temperature in the coldest month is between 0 and 10 oC. Secondly, the annual days with daily average temperature less than 5 oC are no more than 90 days. According to ASHRAE Standard 90.1 -2010 [4], the HSCW area can be classified into the Climate Zone 3. 1
RI PT
ACCEPTED MANUSCRIPT therefore no central heating is provided by the government in this region. On one hand, residents in HSCW area keep complaining about the poor winter indoor thermal environment resulting from the absence of central heating. On the other hand, with the economic growth and accompanying increase of disposable income, individual heating facilities, such as air source heat pump, gas fueled floor heating and etc., have been widely installed and utilized for domestic heating. Over the last 15 years, the diffusion of individual heating facilities leads to a 575 times growth of residential heating energy consumption in this region [2], constituting a major contributor to the increase of residential energy consumption and associated carbon emission of China. It is worthwhile investigating the internal temperature profiles of dwellings in the HSCW area of China and scrutinizing the driving forces behind that.
M AN U
SC
Indoor thermal environment investigation is a hot research area since it closely related to the building energy consumption and plays a significant role to create a comfort, healthy and efficient built environment. In terms of the investigation methodologies, Alfano et al. [5] proposed a four-step method to assess the indoor thermal environment of existing buildings: firstly cognitive survey, secondly subjective investigation, thirdly instrumental survey and lastly calculation of the indices. Dell'Isola [6] discussed the influence of measurement uncertainties on thermal environment assessment and found that the ambiguous class attribution due to the measurement uncertainties is hard to avoid, however the use of different instruments consistent with ISO 7726 [7] requirements results in compatible values. In addition, many researchers performed field investigation to assess the indoor thermal environment of commercial buildings, including office buildings [8-10], airport terminal buildings [11-13], school buildings [14], [15] and etc.
AC C
EP
TE D
As for the winter thermal environment of residential buildings in the HSCW area of China, which is the topic of this paper, Yoshino et al. [16], [17] measured the winter indoor thermal environment of 26 households in Shanghai, Chongqing and Changsha, three major cities of HSCW area, during the year between 1998 and 2003. It was found that the winter indoor environment of Shanghai, Chongqing and Changsha is far from ASHRAE comfort zone, with internal temperatures averagely 3-6 oC lower than cities located in other parts of China. These research supported the complaints of HSCW residents about the poor winter indoor environment. Yoshino predicted that with the economic growth and the accompanying increase of household income, the poor indoor environment will be improved. Shipworth [18] found no significant change of thermostat settings in English houses between 1984 and 2007 after a careful analysis of two cross-sectional housing survey database, INT84 and CARB07. What is the case in China’s HSCW area, whether the poor indoor thermal environment has been improved during the past 15 years still remains unclear. The household thermal comfort requirements, including the heating temperature, heating duration and etc., significantly influence the residential heating energy consumption [19-22]. Scrutinizing the driving forces behind the household heating requirements is helpful to predict the residential heating energy consumption trend, to identify potential ways to curtail the rising residential heating energy consumption and to develop related policies. Frederiks et al.’s review illustrated that the occupants’ thermal comfort requirements are determined by both individual and contextual variables [23]. How the socio-demographic, psychological and building efficiency variables influenced the internal temperature and household energy consumption has been investigated [24-26]. It was found that the number of occupants, household income, occupant age, building efficiency level all have statistically significant effects on daily mean internal temperature. Shipworth et al.’s research [27] demonstrated that adding heating control systems did not reduce the maximum internal temperature or the duration of heating operation, which is contrary to what the UK government assumed when developing related regulations (e.g. SAP), policies (e.g. DEFRA) and designing 2
ACCEPTED MANUSCRIPT programmers (e.g. of the Energy Saving Trust).
RI PT
The goals of this research are twofold. Firstly, the winter indoor thermal environment and occupants’ heating behavior of urban residential buildings in HSCW area will be investigated by field measuring of 54 households. This part will be presented in Section 2. Secondly, how the dwelling internal temperature is determined, or what factors significantly affect HSCW residents’ thermal comfort demands will be explored by statistics analysis of the survey and measurement data. This research will be presented in Section 3. Conclusions are drawn in Section 4 of the paper.
SC
2. Winter Indoor Thermal Environment Investigation Field measurements of winter indoor thermal environment were performed in 16 residential communities of three major cities in the HSCW climate region. All the dwellings surveyed in this research are high rise apartments, with the dwelling area varying between 80 and 120 m2. Table 1 summarizes the key information of this field study. Figure 1 compares the weather condition of Nanjing and Shanghai. Suzhou locates between Nanjing and Shanghai. It can be observed that the outdoor climate of Nanjing and Shanghai is very similar.
TE D
Number of houses for measurement 10 8 5 4 3 2 2 3 3 4 4 2 4
40 Daily Average Outdoor Temperature
Heating system
City
Individual heating Individual heating Individual heating Individual heating District heating Individual heating District heating Individual heating Individual heating Individual heating Individual heating Individual heating Individual heating
Shanghai Shanghai Suzhou Suzhou Suzhou Suzhou Nanjing Nanjing Nanjing Nanjing Suzhou Suzhou Suzhou
AC C
EP
Residential community No.1 No.2 No.3 No.4 No.5 No.6 No.7 No.8 No.9 No.10 No.11 No.12 No.13
M AN U
Table 1 Summary of Winter Indoor Thermal Environment Filed Measurement
30
Nanjing Shanghai
20 10 0 1/1
2/1
3/1
4/1
5/1
6/1
7/1
-10
8/1
9/1
10/1
11/1
12/1
Date Figure 1 Temperature of the Typical Meteorological Year of Nanjing and Shanghai (data source: [28]) 3
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
This measurement, which lasted throughout the whole heating season of 2014/15 (from late November to middle March), was performed under the guidance of ASHRAE Standard 55-2013 [29]. AOSONG GSP 958 temperature/humidity sensors were utilized to record the internal temperature and relative humidity every five minutes. Table 2 compared the instrumentation valid range and accuracy of AOSONG GSP 958 and the criteria required by ASHRAE Standard 55-2013. It can be seen that AOSONG GSP 958 basically meets the requirement of both ASHRAE [29] and ISO [33] standard, except for the air temperature measurement accuracy, which is ±0.5oC rather than the required value ±0.2oC. 27 households were selected randomly for heating behavior survey. Around 50% of households in each community were selected to guarantee the representativeness of the occupant behavior survey. Only the households equipped with the individual heating facilities were selected for this survey since the central heating is operated full-time-full-space. EFRD S-300 and S-350 smart meters, which were able to measure the real-time electricity consumption of one single electrical appliance, were utilized to record the time when the heating is on. The temperature/humidity sensors were installed at around 1.1 meter height and far away from electrical appliances, without exposure to heat sources. In this research, local discomfort including radiant asymmetries, vertical air temperature gradients, draught rate and etc. has not been measured since local discomfort is not a significant factor in severe thermal environment which is the case in this research since the indoor thermal environment is far from comfort during the majority of occupied hours. In the ISO 7730 [30] and EN 15251 [31] Standards, local discomfort is not considered when the absolute value of PMV (Predicted Mean Vote) [32] is beyond 0.7. Table 2 Key Parameters of Instrument
Instrument
AOSONG GSP 958
Relative humidity Electricity consumption Electricity consumption
AOSONG GSP 958 EFRD S-300 EFRD S-350
Valid range by ISO Standard 7726-2001
Uncertainty by ISO Standard 7726-2001
[33]
[33]
±0.5oC
10 ~ 40 oC
±0.2oC
10 ~ 40 oC
Required: ±0.5oC; Desirable: ±0.2oC
±5%
25% ~ 95%
±5%
/
/
0 ~ 3000 W
±5 W
/
/
/
/
0 ~ 3500 W
±5 W
/
/
/
/
-20 ~ 70 oC
EP
Air temperature
Valid range
TE D
Measured parameter
Valid range Uncertainty by by Uncertai ASHRAE ASHRAE nty Standard Standard [29] 55-2013 55-2013 [29]
AC C
0 ~ 99.9%
2.1 Overview In figure 2, the distributions of internal and external temperature/relative humidity readings during occupied hours2 are presented. Bimodal distributions of internal temperature can be observed in both the living room and bedroom, one for the absence of heating and the other for the operation of heating. However the fact that only a small proportion of internal temperature readings are higher than 18 oC (23% for the living room and 2
The occupied hours are set from 17 PM to 8 AM for the working days and 24 hours for the holidays, 4
ACCEPTED MANUSCRIPT 13% for the bedroom) illustrates that the duration of heating operation is not long. 25%
30%
Percentage
20% 10%
10%
20% 15% 10%
M AN U
Percentage
30%
100
90
80
70
60
50
40
SC
25%
10%
RI PT
Living Room Relative Humidity/%
40%
20%
30
28
20
0 4 8 12 16 20 24 Living Room Temperature/oC
10
0% 0
5%
100
90
80
70
60
50
10% 5% 100
90
80
70
60
50
22
40
-2 2 6 10 14 18 Outdoor Temperature/o C
15%
0%
AC C
-6
20%
30
0%
25%
20
10%
40
Bedroom Relative Humidity/%
0
20%
EP
30%
30
28
TE D
40%
4 8 12 16 20 24 Bedroom Temperature/o C
20
0
0
10
0%
0%
Percentage
Percentage
15%
5%
0%
Percentage
20%
10
Percentage
40%
Outdoor Relative Humidity/%
Figure 2 Internal and External Temperature and Relative Humidity Distributions
2.2 Indoor thermal comfort The methods to assess the indoor thermal environment depend on whether the building is mechanically or naturally conditioned. For mechanically conditioned environment, steady state thermal comfort model, proposed by Fanger [32], is widely accepted in related comfort standards [29], [30]. For naturally conditioned buildings, de Dear and Brager [34] proposed the adaptive thermal comfort model, which was also utilized in ASHRAE thermal comfort standard [29]. The majority of the cases we investigated are neither mechanically conditioned nor naturally conditioned. Basically, they belong to mix-mode buildings [35], which means they utilized natural ventilation when natural ventilation is able to provide adequate indoor thermal comfort and 5
switched to air condition when naturalACCEPTED ventilation is MANUSCRIPT insufficient to guarantee comfort. Different researchers have different opinions when the steady state or the adaptive model should be utilized in mix-mode buildings [36], [37] . However this debate is very complicated and beyond the topic of this paper. In this research, ASHRAE comfort zone [29], based on the steady state thermal comfort model, is used to evaluate the indoor thermal environment.
RI PT
Previous research [16], [38-40] found that residents living in HSCW area are accustomed to wearing heavy clothes indoors in winter, with clothing insulation levels varying between 1.15 and 1.42 Clo units. Therefore in this research, the comfort zone is calculated setting the Clo unit at 1.2. To draw Figure 3, operative temperature, rather than air temperature, is needed. The operative temperature can be calculated by Equation (1) as below,
to = Ata + (1 − A)tr
SC
Equation (1) [29]
Where, to is the operative temperature, ta is the average air temperature and tr is the mean radiant
M AN U
temperature. The value of A is determined according to the average air speed, i.e. A equals to 0.5 when the air speed is less than 0.2m/s, equals to 0.6 when the air speed is between 0.2m/s and 0.6m/s, and equals to 0.7 when the air speed is between 0.6m/s and 1.0m/s.
AC C
EP
TE D
In this research, the operative temperature is assumed to approximate air temperature. There are two reasons behind this approximation. Firstly, from the occupant heating behavior survey (presented in detail in Section 2.3), it was found in the HSCW area of China, the residential domestic heating facilities is turned off during the most period of time. When the heating facilities is off, the average air temperature is approximate to the mean radiant temperature, accordingly approximate to the operative temperature, i.e. ta=tr=to. Secondly, compared with radiant heating terminals like floor heating or radiator, convective terminals were much more widely utilized in this area (81% vs. 19% in this research). When the convective terminals is on, the average air speed is relatively high, likely to exceed 0.6m/s, which means the operative temperature is significantly depend on the air temperature, since the weighting factors of ta and tr is 0.7 and 0.3 respectively when the air speed is above 0.6m/s. Figure 3 is drawn based on this approximation.
Figure 3 ASHRAE Steady State Thermal Comfort Model [29]
As shown in Figure 3, even the Clo. adjustment has already been considered, the indoor thermal 6
ACCEPTED MANUSCRIPT environment of dwellings in the HSCW area during occupied hours is far from comfort given the ASHRAE steady state thermal comfort model. The majority of temperature/humidity readings fall outside the comfort zone due to the low internal temperature.
Bedroom
30% Percentage
30% 20%
20%
Living room Bedroom
M AN U
Percentage
40%
Living room
SC
40%
RI PT
2.3 Occupant heating behavior Unlike the full-time-full-space operated district heating, the individual heating is operated part-time-partspace. Figure 4 illustrates when and under what temperature, domestic heating is triggered. For the living room, the majority of heating (30%) is triggered between 16 PM to 19 PM, the time when residents are back home from work. While for the bedroom, heating is most likely (34%) to be triggered between 19 PM to 22 PM, the time when residents plan to go to sleep. As for the triggering temperature, the majority of residents choose to turn on heating when the temperature is between 10 to 14 oC. 10 oC seems to be the lowest acceptable internal temperature without heating.
10%
10%
0%
0% 6 and below
6-8
8-10
1-4
10-12 12-14 14-16 16-18 18 and above Heating Triggering Temperature/o C
4-7
7-10 10-13 13-16 16-19 19-22 Heating Triggering Time
22-1
TE D
Figure 4 Heating Triggering Temperature and Time
Heating duration/h
AC C
10
EP
The heating duration is presented in Figure 5. The average heating duration per operation is 2.3 hours for the living room and 2.8 hours for the bedroom. From this it can be deduced that occupants tend to turn off heating when they fall asleep. The average heating duration per day is 1.9 hours for the living room and 1.4 hours for the bedroom. It can be predicted that the residential heating energy usage would be lifted significantly if the current part-time-part-space heating operation mode was changed to the full-time-fullspace operation, which is the case in north China
Max
8
+STD
6
Average
4
-STD
2
Min
0 Living room
Bedroom
Living room
per Operation
Bedroom
per Day
Figure 5 Duration of Heating Operation 7
ACCEPTED Figure 6 presents the internal temperatures during MANUSCRIPT heating operation. The maximum temperature during heating is determined by the heating set-point temperature. The average maximum temperature is 20.6 oC for the living room and 21.1 oC for the bedroom. The average difference between the maximum temperature and average temperature during heating operation is 5.5 oC for the living room and 5.6 oC for the bedroom. The operation of heating lifts the average internal temperature 1.6 oC (13.5 to 15.1 oC) for the living room and 2.8 oC (12.7 to 15.5 oC) for the bedroom, which is not as significant as we assume. One explanation for this is the short duration of heating operation.
RI PT
35
+STD
25
Average -STD
15
SC
Temperature/oC
Max
Min
5
Living room Bedroom
M AN U
Living room Bedroom Maximum Temperature during Heating
Average Temperature during Heating
Figure 6 Internal Temperature during Heating Operation
EP
TE D
Another reason for this significant difference between maximum temperature and average temperature during heating and the not marked internal temperature elevation by heating is the slow thermal response of the heating system. Whether the indoor thermal environment can be elevated quickly to the comfort zone is crucial for the intermittent heating especially in HSCW area since the average duration of heating operation is short in this region. Wang et al. [41] utilized thermal time constant to evaluate the thermal dynamic performance of intermittent heating system. The intermittent heating system with small thermal constant time is needed in HSCW region owing to its ability to elevate indoor temperature quickly.
AC C
2.4 Comparison with the past and the UK The UK has similar climate condition with HSCW region in winter, however the average internal temperature of UK is 6 oC higher (19 oC [18] compared with 13 oC), as shown in Figure 7. Comparing the internal temperature measured in 2014 with the data measured in 1998 by Yoshino et al. [16], it seems that the indoor thermal environment became worse during the past 15 years. Limited sampling size in Yoshino et al’s research might be a major reason behind this. Only 6 households were measured in Yoshino et al.’s research, some errors are likely to exist. Besides, in this field survey, the low income group (who live in the affordable house) has been added in the sample, who rarely utilize heating facilities, lowering the average internal temperature value. For these two reasons it is hard to say whether there is significant improvement of the indoor thermal environment in the HSCW area during the past 15 years.
8
ACCEPTED MANUSCRIPT 35 Max
Internal Temperature/o C
30
+STD
25 20
Average
15
-STD
10
Min
0 Internal space
Living room
UK, 2007
Bedroom
Shanghai, 1998
Living room
RI PT
5 Bedroom
Shanghai, Suzhou and Nanjing, 2014
SC
Figure 7 Internal Temperature Comparison [16], [18]
M AN U
3. Analysis of the Driving Forces of Heating Requirements In this section, statistical analysis has been utilized to scrutinize the factors significantly influencing the indoor thermal requirements of residents in HSCW area.
TE D
3.1 Methodology The usual Ordinary Least Square (OLS) method, which is the most widely used in statistical analysis, is based on the homoskedasticity assumption. However, when the heteroskedasticity is present amongst residuals, the usual OLS test is invalid [42]. In this research, the White Test is firstly performed to check for the presence of heteroskedasticity amongst residuals, to decide whether the heteroskedasticity-robust statistical analysis method is needed in this analysis.
AC C
EP
Given the presence of heteroskedasticity, OLS is no longer the Best Linear Unbiased Estimator (BLUE) and no longer asymptotically efficient [42]. If the form of heteroskedasticity is known, the Weighted Least Square (WLS) Estimator is more efficient than the OLS Estimator. However, the form of heteroskedasticity is unknown in this research. Though the Feasible Generalized Least Squares (FGLS) method is able to handle the problem of unknown heteroskedasticity form by assuming a form of the heteroskedasticity. However, if the assumption of the heteroskedasticity form is not correct, it is not guaranteed that FGLS is more efficient than OLS [42]. Besides, since the OLS remains consistent even when the heteroskedasticity is present [42]. The consistence of OLS guarantees that OLS approaches to the true value in large sample size. Therefore in this research, the OLS method is utilized to estimate the effect of variant factors on heating requirements, and heteroskedasticity-robust test method is utilized to test whether this effect is statistically significant. Two datasets have been used to perform the statistical analysis in this research. One dataset is from the field measurement. The description of the explained and explanatory variables of the on-site measurement dataset is shown in Table 3. The Degree of Freedom (DF) of this analysis equals to 46 (n=54, k=7, DF=n-k-1). The sample size can be considered as large when DF is more than 30 [42]. The other dataset is from the survey. This survey is carried through website. Heating set-point temperature and heating duration were asked to characterize occupants’ heating demand. Socio-demographic and occupancy related questions including education background, household income, previous thermal experiences, and building and heating system related questions, including year of construction and heating system were asked to characterize the subjects 9
ACCEPTED MANUSCRIPT surveyed. The description of the statistics of the survey dataset is shown in Table 4, with the Degree of Freedom equaling to 515 (n=525, k=9, DF=n-k-1). Table 3 Statistics of the On-site Measurement Dataset
Variable name
Type
O
Dummy
Standard deviation 3.87
Proportion
66.7% 33.3%
HH_C
T
GB
Dummy
91.5% 18.5%
Dummy
R
77.8% 22.2%
Dummy 38.9% 61.1%
HS
Dummy 9.2% 77.8% 13.0%
EP
AC C
7.5% 92.5%
Dummy
TE D
Building and heating system Building not a certificated green building certificated green building3 Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
Mean 13.16
SC
Explanatory variables Variable description Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied
Number of obs. 54
RI PT
Type scale
M AN U
Explained variable Variable description Average internal temperature
Table 4 Statistics of the Survey Dataset
Explained variable Variable description
Type
Number of obs.
Mean
Heating set-point temperature Heating duration_Weekday Heating duration_Weekend
scale scale scale
525 525 525
19.38 5.23 7.85
Standard deviation 3.19 4.24 5.15
Explanatory variables 3
Certificated by Chinese
5.6% 85.3% 9.1% Dummy 14.3% 45.3% 40.4%
SC
Dummy
YC
HS
33.2% 19.4% 21.7% 25.7%
Dummy 50.1% 49.9% Dummy
TE D
Building and heating system Year of construction Before 2000 After 2001 5 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
TE
Proportion
Dummy
M AN U
Thermal experience Experience of living in the north 4 No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
Type
RI PT
ACCEPTED MANUSCRIPT Variable description Variable name Socio-demographic and occupancy Education E Middle school/vocational school and less Junior college/college Graduates Income I Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
23.3% 69.1% 7.6%
AC C
EP
3.2 Results and discussion Data analysis toolkit STATA 11 is utilized to perform the statistical analysis in this research. The results are shown in Table 5 and Table 6. The heteroskedasticity-robust t-Statistics are shown in parenthesis. If 5% (10%) is chosen as the significant level, the effect is statistically significant only when the absolute value of t-Statistics is no less than 1.67 (1.30). Table 5 Result of the Statistical Analysis of the On-site Measurement Dataset
Explained variable Number of observation R-square Socio-demographic and occupancy Occupation
4 5
Average temperature 54 0.840
Refers to north of the Qin-Huai line, where central heating is provided Year 2001 is the year when the first edition of
cold winter zone> was officially issued and implemented. Therefore buildings constructed after year of 2001 can be seen as with higher energy efficiency, higher thermal insulation level, and etc. 11
-2.63 (1.81)
SC
--0.51 (-1.02)
--11.13 (-7.69) -6.89 (-4.76)
M AN U
Building and heating system Building not a certificated green building certificated green building Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
RI PT
ACCEPTED MANUSCRIPT Not a university faculty -University faculty -0.43 (-0.66) Household structure With child aged < 12 -Without child aged < 12 4.10 (5.37) Tenure type Owner occupied -Tenant occupied -2.33 (-1.46)
Table 6 Result of the Statistical Analysis of the Survey Dataset
Heating set-point temperature 525 0.589
525 0.405
525 0.382
-0.10 (0.16) -0.21 (-0.29)
--0.24 (-0.42) 1.06 (1.07)
-1.67 (2.43) 2.32 (1.08)
-0.60 (1.51) 1.65 (3.82)
-0.37 (0.85) 1.04 (1.89)
-1.42 (2.39) 2.00 (3.02)
-0.01 (0.01) 0.22 (0.55) 0.80 (2.13)
-0.03 (0.06) 1.33 (2.75) 1.74 (3.09)
--0.21 (-0.38) 1.11 (1.76) 1.29 (1.96)
TE D
Explained variable
AC C
EP
Number of observation R-square Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
Experience of living in the north No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
12
Heating Heating duration_Weekday duration_Weekend
ACCEPTED MANUSCRIPT -0.07 (0.24)
-0.05 (0.14)
-0.37 (0.83)
-1.14 (3.40) -0.03 (-0.06)
--1.75 (-2.29) -1.06 (-1.41)
--1.52 (-2.25) -1.83 (-2.10)
RI PT
Year of construction Before 2000 After 2001 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
SC
Socio-demographic and occupancy effects From Table 5 and Table 6, it can be seen that education and occupation do not significantly influence occupants’ indoor thermal requirements. It was believed that increased education results in more pro-environmental behaviors [43], [44]. However this has not been observed in this research. Contrarily, higher education level leads to longer heating durations on weekends. Barr ascribed this phenomenon to the “knowledge-action gap” [45]. Though residents with higher education level care more about the environment and energy issue, however their actions always lag their values.
M AN U
Income is a key determinant of residents’ heating requirements. The households with income more than 100,000 RMB set averagely 1.65 oC higher heating temperature and 1-2 hours longer heating duration compared with those earning less than 50,000 RMB. Tenure type can be considered as the instrumental variable of the household income. It can be seen that the average difference in internal temperatures between an owner occupied dwelling and a tenant occupied dwelling is approximately 2.33 oC.
TE D
The presence of a child less than 12 years old is also a significant driver of higher heating requirements, increasing the mean internal temperature by an average of 4.10 oC compared with those households without a child.
AC C
EP
Building and heating system effects Since no physical survey of building energy efficiency has been performed in this research, other variables are used as instrumental variables, to characterize the building efficiency. The first instrumental variable of building efficiency we chose is whether it is a certificated green building. Since the Design Standard for Energy Efficiency of Residential Buildings in Hot Summer and Cold Winter Zone [46] was put into effect at the year of 2001, the variable, whether it is constructed before or after 2001, is able to identify the building efficiency. The rebound effect is observed in HSCW area. Since it was found that the building energy efficiency does not influence the residents heating requirements (heating set-point temperature and heating duration), but statistically significantly improve the indoor thermal environment (averagely 2.63 oC higher internal temperatures). This result confirms Sorrell et al.’s finding that half of the temperature rebound effect is not due to households demanding higher standard of comfort, but simply because energy efficiency measures such as higher insulation, increasing air-tightness level and etc. lead to a warmer indoor environment [47]. Compared with the central heating, if the air source heat pump is utilized in the dwelling, the mean internal temperature will be lowered by 11.13 oC, and if the gas fueled floor heating is utilized, the mean internal temperature will be lowered by 6.89 oC. This huge variance is due to the fact that central heating is operated full-time-full-space to satisfy the heating requirements of all the households. However, the air source heat 13
ACCEPTED MANUSCRIPT pump or gas fueled floor heating is operated part-time-part-space according to the household’s individual heating demand. The set-point temperature of central heating is 1.14 oC lower than that of the air source heat pump, which indicates that the central heating is likely to supply more heat than that actual needed by the occupants.
RI PT
Thermal experience effects The thermal adaptive model predicted that the previous thermal experience is likely to alter residents’ thermal comfort sensation and thermal requirements through behavioral adjustment, physiological acclimatization and adjustment of psychological expectations [48], [49]. Yu’s chamber experiment confirmed this prediction [50] and found that subjects coming from Beijing, where central heating is provided, complained of greater cold discomfort than those from Shanghai, where central heating is unavailable, even when they are exposed to the same environment.
M AN U
SC
For this analysis thermal experience is separated into four discrete bands, no, short, medium and long experience of living in the north where central heating is provided. It is found that living in the north for a short period of time (less than 1 month) has statistically insignificant impact on residents’ heating requirements, indicating that the thermal experience effects take time to come about. However, living in the north for a long period of time (more than 3 months) markedly lifted residents’ heating requirements, resulting in a higher heating set-point temperature and longer heating duration.
EP
TE D
4. Conclusion In this research, the winter indoor thermal environment of dwellings in China’s Hot Summer – Cold Winter climate region has been investigated. The on-site investigation measured the internal temperature and humidity of 54 households, 27 of them were selected for heating behavior survey, in which the heating operation time and duration were recorded by smart meters. The average internal temperature is 13.5 oC for the living room and 12.7 oC for the bedroom. The majority of the temperature/humidity readings fall outside the ASHRAE steady state thermal comfort zone due to the low internal temperatures. The average internal temperature of dwellings in HSCW climate region is 6 oC lower than that in the UK, which has similar winter climate condition. Besides, no significant improvement of the indoor thermal environment has been observed in the HSCW area during the past 15 years.
AC C
Domestic heating is triggered by both time and temperature in HSCW area. The majority of heating is consumed between 16PM and 22PM (55% for the living room and 46% for the bedroom). In terms of the heating triggering temperature, the majority of residents choose to turn on heating when the temperature is between 10 to 14 oC (65% for the living room and 50% for the bedroom), indicating that 10 oC seems to be the lowest acceptable internal temperature without heating. Domestic heating is operated part-time-part-space in HSCW area. Each heating operation lasts averagely 2.3 hours for the living room and 2.8 hours for the bedroom. The average heating duration per day is 1.9 hours for the living room and 1.4 hours for the bedroom throughout the whole heating season. The short heating operation duration is the main reason for the poor indoor thermal environment as well as the relative low residential heating energy consumption in the HSCW area. Additionally, the driving forces of heating requirements of residents in HSCW area have been scrutinized. It is found that higher household income, the presence of child and not short experience (no less than 3 months) 14
ACCEPTED of living in the north, where central heating is MANUSCRIPT provided, significantly influence occupants’ heating requirements, indicating that with the further economic growth and more frequent population mobility, a higher heating set-point temperature and longer heating duration will be demanded in this area. It was also found that a short living experience in the north will not significantly improve occupant’s heating demand, demonstrating that thermal experience takes time to affect occupants’ thermal preferences.
RI PT
The rebound effect is observed in this research. Higher building energy efficiency results in a warmer winter indoor environment. However, the building energy efficiency level does not influence the residents heating requirements.
SC
The mean internal temperature of the residence equipped with central heating is approximately 11 oC higher than those equipped with individual air source heat pump. This variance is due to the longer heating operation duration rather than the higher heating set-point temperature of the central heating system. In fact, the set-point temperature of central heating is 1.14 oC, which is statistically significantly lower than that of the air source heat pump.
M AN U
Acknowledgement This research is supported by the National Natural Science Foundation of China (Grant number 51561135001) and the Innovative Research Groups of the National Natural Science Foundation of China (Grant number 51521005). Great thanks to Zhe Li, Zhen Zhao and Zufeng Pei for their contribution to the field measurement and survey. Special thanks to all the residents who provide kind help for the on-site measurement.
TE D
Reference
AC C
EP
[1] Los Angeles Times. For central heat, China has a north-south divide at Qin-Huai line. Retrieved June 20, 2015, from http://www.latimes.com/world/asia/la-fg-china-heat-20141115-story.html [2] Tsinghua University Building Energy Research Center 2013. 2013 Annual Report on China Building Energy Efficiency. Beijing: China Architecture & Building Press. (in Chinese) [3] State Bureau of Technical Supervision, Minister of Housing and Urban-Rural Development, Standard of Climatic Regionalization for Architecture, GB 50178-93. (in Chinese) [4] American Society of Heating, Refrigerating and Air-Conditioning Engineers, Thermal Environmental Conditions for Human Occupancy, ANSI/ASHRAE Standard 55-2013 [5] D'Ambrosio Alfano, F.R., Olesen, B.W., Palella B.I., Riccio G. Thermal Comfort: Design and Assessment for Energy Saving. Energy and Buildings 81 (2014), 326-336. [6] M. Dell'Isola, A. Frattolillo, B.I. Palella, G. Riccio, Measurement uncertainties influence on the thermal environment assessment, International Journal of Thermophysics 33 (8-9) (2012) 1616-1620 [7] International Standardization Organization, Ergonomics of the Thermal Environment—Instruments for Measuring Physical Quantities, ISO 7726: 2002 [8] Z.F. Pei, B.R. Lin, Y.C. Liu, Y.X. Zhu, Comparative study on the indoor environment quality of green office buildings in China with a long-term field measurement and investigation, Building and Environment 84 (2015) 80-88 [9] T. Goto, T. Mitamura, H. Yoshino, A. Tamura, E. and Inomata, Long term field survey on thermal adaptation in office buildings in Japan, Building and Environment, 42 (2007), 3944–3954 [10] M.H. Luo, B. Cao, J. Damiens, B.R. Lin, Y.X. Zhu, Evaluating thermal comfort in mixed-mode buildings: A field study in a subtropical climate, Building and Environment, 88 (2014), 46–54 15
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[11] Z. Wang, H.T. Zhao, B.R. Lin, Y.X. Zhu, Q. Ouyang, J. Yu, Investigation of indoor environment quality of Chinese large-hub airport terminal buildings through longitudinal field measurement and subjective survey, Building and Environment 94 (2015) 593-605 [12] A. Kotopouleas and M. Nikolopoulou, Understanding thermal comfort conditions in airport terminal Buildings, Proceedings of 8th Windsor Conference: Counting the Cost of Comfort in a changing world, Cumberland Lodge, Windsor, UK, 10-13 April 2014 [13] C.A. Balaras, E. Dascalaki, A. Gaglia, K. Droutsa, Energy conservation potential, HVAC installations and operational issues in Hellenic airports, Energy and Buildings 35 (2003) 1105–1120 [14] L.D. Pereira, D. Raimondo, S.P. Corgnati, M.G. da Silva, Assessment of indoor air quality and thermal comfort in Portuguese secondary classrooms: Methodology and results, Building and Environment 81 (2014) 69-80 [15] R. Yao, J. Liu, B. Li, Occupants’ adaptive responses and perception of thermal environment in naturally conditioned university classrooms, Applied Energy, 87 (2010), 1015–1022. [16] H. Yoshino,S. Guan, Y.F. Lun, A. Mochida, T. Shigeno, Y. Yoshino, Q.Y. Zhang, Indoor thermal environment of urban residential buildings in China: winter investigation in five major cities, Energy and Buildings 36 (2004) 1227–1233 [17] H. Yoshino, Y. Yoshino, Q.Y. Zhang, A. Mochida, N.P. Li, Z.H. Li, H. Miyasaka, Indoor thermal environment and energy saving for urban residential buildings in China, Energy and Buildings 38 (2006) 1308–1319 [18] M. Shipworth, Thermostat settings in English houses: No evidence of change between 1984 and 2007, Building and Environment 46 (2011) 635-642 [19] A.H. Neto, A.S. Fiorelli, Comparison between detailed model simulation and artificial neural network for forecasting building energy consumption, Energy and Buildings 40 (2008) 2169–2176 [20] Z. M. Gill, M. J. Tierney, I. M. Pegg, N. Allan, Low-energy dwellings: the contribution of behaviours to actual performance. Build Res Inform 2010 38(5): 491–508. [21] O.G. Santin, Behavioural Patterns and User Profiles related to energy consumption for heating, Energy and Building, 43 (2011): 2662–2672 [22] Z. Yu, B.C.M. Fung, F. Haghighat, H. Yoshino, E. Morofsky, A systematic procedure to study the influence of occupant behavior on building energy consumption, Energy and Building, 43 (2011): 1409–1417 [23] E. R. Frederiks, K. Stenner and E. V. Hobman, The Socio-Demographic and Psychological Predictors of Residential Energy Consumption: A Comprehensive Review, Energies 8 (2015), 573-609 [24] S. Kelly, M. Shipworth, D. Shipworth, M. Gentry, A. Wright, M. Pollitt, D. Crawford-Brown, K. Lomas, Predicting the diversity of internal temperatures from the English residential sector using panel methods, Applied Energy 102 (2013) 601–621 [25] W. Abrahamse, L. Steg, How do socio-demographic and psychological factors relate to households’ direct and indirect energy use and savings, Journal of Economic Psychology, 30 (2009) 711-720 [26] E.R. Frederiks, K. Stenner, E.V. Hobman, Household energy use: Applying behavioural economics to understand consumer decision-making and behavior, Renewable & Sustainable Energy Review, 41 (2015) 1385-1394 [27] M. Shipworth, S. K. Firth, M. I. Gentry, A. J. Wright, D. T. Shipworth and K. J. Lomas, Central heating thermostat settings and timing: building demographics, BUILDING RESEARCH & INFORMATION (2010) 38(1), 50–69 [28] Meteorological Information Center of China Meteorological Administration, Department of Building Science Tsinghua University, (2005). Meteorological data set for built thermal environment analysis in China. Beijing: China Architecture & Building Press. (in Chinese) [29] American Society of Heating, Refrigerating and Air-Conditioning Engineers, Thermal Environmental Conditions for Human Occupancy, ANSI/ASHRAE Standard 55-2013 [30] International Standardization Organization, Ergonomics of the Thermal Environment – Analytical Determination and Interpretation of Thermal Comfort Using Calculation of the PMV and PPD Indices and Local Thermal Comfort, ISO 7730: 2005. 16
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[31] European Committee for Standardization, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, EN 15251: 2007. [32] P.O. Fanger, Calculation of thermal comfort: introduction of a basic comfort equation, ASHRAE Transactions 73 (2) (1967) 1–20. [33] International Standardization Organization, Ergonomics of the thermal environment - Instruments for measuring physical quantities, ISO 7726: 2001. [34] R. de Dear, G.S. Brager, The adaptive model of thermal comfort and energy conservation in the built environment, Int J Biometeorol 45 (2001) 100–108 [35] G. Brager, Mixed-mode cooling, ASHRAE Journal., 48 (2006) 30–37. [36] D. Rowe, Thermal comfort in a naturally ventilated environment with supplementary cooling and heating, Architectural Science Review, 47 (2004), 131–140. [37] M.P. Deuble, and R.J. de Dear, Mixed-mode buildings: a double standard in occupants’ comfort expectations, Building and Environment, 54 (2012), 53–60. [38] B. Z. Li, J. Liu, R. M. Yao, 2007, investigation and analysis on classroom thermal environment in winter in Chongqing, HV&AC, 2007(5): 115-117 (in Chinese) [39] J. H. Li, L. Yang, J. P. Liu, 2008, adaptive thermal comfort model for hot summer cold and winter zone, HV&AC 2008(7): 20-24. (in Chinese) [40] J.S. Lan (2013) Human Adaptation to Thermal Environment in Free Running Building, Ph.D. Thesis: Chongqing (in Chinese) [41] Z. Wang, B.R. Lin, Y.X. Zhu, Modeling and measurement study on an intermittent heating system of a residence in Cambridgeshire, Building and Environment 92 (2015) 380-386 [42] Wooldridge JM. Introductory econometrics: a modern approach. China Renmin University Press; 2010. [43] W. Poortinga, L. Steg, C. Vlek, C. Values, environmental concern and environmental behavior: A study into household energy use. Environment Behavior. 36 (2004), 70–93. [44] G. Nair, L. Gustavsson, K. Mahapatra, Factors influencing energy efficiency investments in existing Swedish residential buildings. Energy Policy 38 (2010), 2956–2963. [45] S. Barr, A.W. Gilg, N. Ford, The household energy gap: Examining the divide between habitual- and purchase-related conservation behaviours. Energy Policy 33 (2005), 1425–1444. [46] MOHURD, 2001, Design standard for energy efficiency of residential buildings in hot summer and cold winter zone, JGJ 134-2001 (in Chinese) [47] S. Sorrell, J. Dimitropoulos, M. Sommerville, Empirical estimates of the direct rebound effect: a review. Energy Policy 37 (2009): 1356-71. [48] G.S. Brager, R. de Dear, (1998) Thermal adaptation in the built environment: a literature review, Energy and Buildings 27 (1998): 83-96. [49] R. J. de Dear et al., Progress in thermal comfort research over the last twenty years, Indoor Air 2013; 23: 442–461 [50] J. Yu, G. Cao, W. Cui, Q. Ouyang, Y. Zhu People who live in a cold climate: thermal adaptation differences based on availability of heating, Indoor Air 2013; 23: 303–310
17
ACCEPTED MANUSCRIPT
Table 1 Summary of Winter Indoor Thermal Environment Filed Measurement
Heating system
City
Individual heating Individual heating Individual heating Individual heating District heating Individual heating District heating Individual heating Individual heating Individual heating Individual heating Individual heating Individual heating
Shanghai Shanghai Suzhou Suzhou Suzhou Suzhou Nanjing Nanjing Nanjing Nanjing Suzhou Suzhou Suzhou
SC
RI PT
Number of houses for measurement 10 8 5 4 3 2 2 3 3 4 4 2 4
M AN U
Residential community No.1 No.2 No.3 No.4 No.5 No.6 No.7 No.8 No.9 No.10 No.11 No.12 No.13
Table 2 Key Parameters of Instrument
Instrument
AOSONG GSP 958
Relative humidity Electricity consumption Electricity consumption
AOSONG GSP 958 EFRD S-300 EFRD S-350
Valid range by ISO Standard 7726-2001
Uncertainty by ISO Standard 7726-2001
[33]
[33]
-20 ~ 70 oC
±0.5oC
10 ~ 40 oC
±0.2oC
10 ~ 40 oC
Required: ±0.5oC; Desirable: ±0.2oC
0 ~ 99.9%
±5%
25% ~ 95%
±5%
/
/
0 ~ 3000 W
±5 W
/
/
/
/
0 ~ 3500 W
±5 W
/
/
/
/
AC C
EP
Air temperature
Valid range
TE D
Measured parameter
Valid range Uncertainty by by Uncertai ASHRAE ASHRAE nty Standard Standard [29] 55-2013 [29] 55-2013
Table 3 Statistics of the On-site Measurement Dataset
Explained variable Variable description Average internal temperature
Type scale
Explanatory variables Variable description
Variable name
Number of obs. 54
1
Mean 13.16
Standard deviation 3.87
Type
Proportion
O
Dummy 66.7% 33.3%
HH_C
Dummy 7.5% 92.5%
T
Dummy
RI PT
91.5% 18.5%
GB
Dummy
SC
Building and heating system Building not a certificated green building certificated green building1 Room Structure Living room Bedroom Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
ACCEPTED MANUSCRIPT
R
HS
77.8% 22.2%
Dummy
M AN U
Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied
38.9% 61.1%
Dummy 9.2% 77.8% 13.0%
Explained variable Variable description
EP
Heating set-point temperature Heating duration_Weekday Heating duration_Weekend
TE D
Table 4 Statistics of the Survey Dataset
AC C
Explanatory variables Variable description Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB 1
Type
Number of obs.
Mean
scale scale scale
525 525 525
19.38 5.23 7.85
Standard deviation 3.19 4.24 5.15
Variable name
Type
Proportion
E
Dummy 5.6% 85.3% 9.1%
I
Dummy 14.3% 45.3% 40.4%
Certificated by Chinese
ACCEPTED MANUSCRIPT Dummy 33.2% 19.4% 21.7% 25.7%
YC
Dummy
50.1% 49.9%
Dummy
SC
HS
23.3% 69.1% 7.6%
M AN U
Building and heating system Year of construction Before 2000 After 2001 3 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
TE
RI PT
Thermal experience Experience of living in the north 2 No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
Table 5 Result of the Statistical Analysis of the On-site Measurement Dataset
AC C
EP
TE D
Explained variable Number of observation R-square Socio-demographic and occupancy Occupation Not a university faculty University faculty Household structure With child aged < 12 Without child aged < 12 Tenure type Owner occupied Tenant occupied Building and heating system Building not a certificated green building certificated green building Room Structure Living room Bedroom
2 3
Average temperature 54 0.840
--0.43 (-0.66) -4.10 (5.37) --2.33 (-1.46)
-2.63 (1.81) --0.51 (-1.02)
Refers to north of the Qin-Huai line, where central heating is provided Year 2001 is the year when the first edition of
cold winter zone> was officially issued and implemented. Therefore buildings constructed after year of 2001 can be seen as with higher energy efficiency, higher thermal insulation level, and etc. 3
ACCEPTED MANUSCRIPT Heating system Central heating -Non-central air source heat pump -11.13 (-7.69) Non-central gas fueled floor heating -6.89 (-4.76) Table 6 Result of the Statistical Analysis of the Survey Dataset
525 0.405
525 0.382
-0.10 (0.16) -0.21 (-0.29)
--0.24 (-0.42) 1.06 (1.07)
-1.67 (2.43) 2.32 (1.08)
-0.60 (1.51) 1.65 (3.82)
-0.37 (0.85) 1.04 (1.89)
-1.42 (2.39) 2.00 (3.02)
-0.01 (0.01) 0.22 (0.55) 0.80 (2.13)
-0.03 (0.06) 1.33 (2.75) 1.74 (3.09)
--0.21 (-0.38) 1.11 (1.76) 1.29 (1.96)
-0.07 (0.24)
-0.05 (0.14)
-0.37 (0.83)
-1.14 (3.40) -0.03 (-0.06)
--1.75 (-2.29) -1.06 (-1.41)
--1.52 (-2.25) -1.83 (-2.10)
TE D
Experience of living in the north No experience of living in the north Have lived in the north less than 1 month Have lived in the north between 1 and 3 months Have lived in the north more than 3 months
AC C
EP
Year of construction Before 2000 After 2001 Heating system Central heating Non-central air source heat pump Non-central gas fueled floor heating
RI PT
Heating Heating duration_Weekday duration_Weekend
SC
Number of observation R-square Socio-demographic and occupancy Education Middle school/vocational school and less Junior college/college Graduates Income Household income < 50,000 RMB Household income ~ (50,000, 100,000) RMB Household income > 100,000 RMB
Heating set-point temperature 525 0.589
M AN U
Explained variable
4
ACCEPTED MANUSCRIPT Nanjing 30
Shanghai
20 10 0 1/1
2/1
3/1
4/1
5/1
6/1
7/1
-10
8/1
9/1
10/1
11/1
12/1
RI PT
Daily Average Outdoor Temperature（℃）
40
Date
Figure 1 Temperature of the Typical Meteorological Year of Nanjing and Shanghai (data source: [28])
20% 10%
15% 10% 5%
100
90
80
70
60
50
40
5%
Bedroom Relative Humidity/%
1
100
90
80
70
60
50
28
40
4 8 12 16 20 24 Bedroom Temperature/o C
10%
0%
AC C
0
15%
30
0%
20%
20
10%
25%
0
20%
EP
30%
30
Living Room Relative Humidity/%
TE D
40%
28
10
4 8 12 16 20 24 Living Room Temperature/oC
20
0
0
10
0%
0%
Percentage
20%
M AN U
Percentage
30%
SC
25%
Percentage
Percentage
40%
ACCEPTED MANUSCRIPT 25%
30%
Percentage
20% 10%
20% 15% 10% 5% 100
90
80
70
60
50
40
22
RI PT
-2 2 6 10 14 18 Outdoor Temperature/o C
30
0
-6
20
0%
0%
10
Percentage
40%
Outdoor Relative Humidity/%
M AN U
SC
Figure 2 Internal and External Temperature and Relative Humidity Distributions
TE D
Figure 3 ASHRAE Steady State Thermal Comfort Model [29]
40%
Bedroom
10% 0% 6 and below
Percentage
AC C
20%
6-8
Living room Bedroom
30%
EP
30% Percentage
40%
Living room
20% 10% 0% 1-4
8-10
10-12 12-14 14-16 16-18 18 and above Heating Triggering Temperature/o C
4-7
7-10 10-13 13-16 16-19 19-22 Heating Triggering Time
Figure 4 Heating Triggering Temperature and Time
2
22-1
ACCEPTED MANUSCRIPT 10
Heating duration/h
Max 8 +STD 6 Average 4
-STD
2
Min
Living room
Bedroom
Living room
per Operation
RI PT
0 Bedroom
per Day
SC
Figure 5 Duration of Heating Operation
35
25
15
5 Living room Bedroom
+STD Average -STD Min
Living room Bedroom
TE D
Maximum Temperature during Heating
M AN U
Temperature/oC
Max
Average Temperature during Heating
35
25
Max +STD
AC C
Internal Temperature/o C
30
EP
Figure 6 Internal Temperature during Heating Operation
20
Average
15
-STD
10
Min
5 0
Internal space
UK, 2007
Living room
Bedroom
Shanghai, 1998
Living room
Bedroom
Shanghai, Suzhou and Nanjing, 2014
Figure 7 Internal Temperature Comparison [16], [18]
3
ACCEPTED MANUSCRIPT
RI PT
SC M AN U
TE D
EP
The residential winter thermal environment in the HSCW area has been investigated The heteroskedasticity-robust Ordinary Least Square analysis has been utilized The average indoor temperature is 13.5 oC for living rooms and 12.7 oC for bedrooms Heating is operated part-time-part-space and triggered by both time and temperature Income, presence of child and thermal experience markedly influence heating demands
AC C