Energy & Buildings 220 (2020) 110086

Contents lists available at ScienceDirect

Energy & Buildings

journal homepage: www.elsevier.com/locate/enbuild

Dynamic indoor comfort temperature settings based on the variation

in clothing insulation and its energy-saving potential for an
air-conditioning system
Xinbo Xu a,b, Weiwei Liu a,∗, Zhiwei Lian b
a
School of Energy Science & Engineering, Central South University, Changsha 410083, Hunan, China
b
School of Design, Shanghai Jiao Tong University, Shanghai 200240, China

a r t i c l e i n f o a b s t r a c t

Article history: Saving energy consumption of air-conditioning is essential to building energy conservation while satisfy-
Received 11 February 2020 ing the thermal comfort requirements of indoor environment. On the basis of human clothing adjustment
Revised 14 April 2020
behavior, a dynamic setting temperature (DST) that adapts to climate change was proposed in this paper,
Accepted 22 April 2020
i.e. indoor air-conditioning setting temperature changes dynamically with outdoor air temperature; fur-
Available online 28 April 2020
thermore. A method was established to obtain the DST. The daily mean clothing insulation was predicted
Keywords: according to the historical temperature at first, then indoor thermal comfort temperature was calculated
Thermal comfort based on the variable clothing insulation, and finally the DST was determined. According to the value or
Thermal adaption range of predicted mean vote (0, -0.5~0.5, -1~+1) corresponding to the thermal comfort, three levels of
Clothing adjustment behavior DST were defined. The corresponding thermal acceptability rates were 95%, 90% and 73%, respectively. A
Setting temperature of air-conditioning simulation of DST was conducted in a public building in Changsha, China based on meteorological data
system
of 2017. Three levels of DST ranges were obtained, varied from 20.1 to 27.8 °C in level I, 17.8 to 28.7 °C in
Energy saving
level II and 15.4 to 29.7 °C in level III, respectively. The simulation results of operational air-conditioning
energy consumption shows that the application of DST could save up to 65.5% of energy consumption
compared with the traditional fixed setting temperature (FST), and the operational energy consumption
decreased by 59.1% due to the application of level-III DST compared with level-I DST. Therefore, the DST
proposed in this paper has significant energy saving potential while meeting the thermal comfort re-
quirements.
© 2020 Elsevier B.V. All rights reserved.

1. Introduction Setting temperature, the value of indoor temperature controlled

by the air-conditioning system, influences the energy consumption
The comfortable indoor thermal environment is relevant to peo- of air-conditioning significantly. Liu et al. explored the relationship
ple’s health and working efficiency [1,2,54–56], but it requires a between the office setting temperature and energy consumption
large amount of energy consumption of air-conditioning system of air-conditioning in China in summer [7]. Their results showed
that leads to huge amount of greenhouse gas (GHG) emissions. that energy consumption decreased by 113% and 271% respectively
The air-conditioning system accounts for the major proportion in when the temperature was set at 28 °C compared with 26 °C and
building energy consumption, which can exceed 50% [3,4]. In the 24 °C. Based on different setting temperatures, Wan et al. predicted
recent years, extreme weather occurs more frequently due to cli- the summer energy consumption of air-conditioning system in of-
mate change and global warming [5], which is mostly caused by fice buildings in Hong Kong, China [8]. They found that raising the
GHG emissions. Therefore, reducing energy consumption of the air- setting temperature to 25.5 °C or higher would contribute to at
conditioning system while meeting the thermal comfort require- least 2.9% energy savings annually, compared with 24 °C setting
ments is especially important for reducing GHG emissions and mit- in the cooling season in the current building energy code. Wang
igating extreme weather [6]. et al. simulated the energy consumption of air-conditioning for
a residential building in Hangzhou China in winter [9]. Their re-
sults showed that the energy consumption per unit area decreased
by 43.3% when the setting temperature dropped by 4 °C. To in-
∗
Corresponding author. vestigate the effect of setting temperature on air-conditioning en-
E-mail address: [email protected] (W. Liu). ergy consumption in winter, Ding et al. established three types of

https://doi.org/10.1016/j.enbuild.2020.110086
0378-7788/© 2020 Elsevier B.V. All rights reserved.
2 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

public building models [10]. They found that the heating energy
consumption of small office buildings would increase by 6.9–9.8%
and that of large office buildings would increase by 6.5–9.7% as
the setting temperature rose by every 1 °C. Yamtraipat et al. cal-
culated the annual air-conditioning energy consumption according
to the setting temperature of 147 rooms in 13 buildings in Thai-
land [11]. It was found that the energy consumption can be re-
duced by about 6.14% for every 1 °C increase in setting tempera-
ture from 20 to 28 °C. With this, the electrical energy can be saved
by 804.60 GWh/year if the setting temperature of all rooms was
adjusted to 26 °C. The results of these above literature show that
the energy consumption of the air-conditioning system can be sig-
nificantly saved by adjusting the setting temperature reasonably,
i.e. increasing the setting temperature in summer or decreasing in
winter.
Setting temperature should be determined based on indoor
thermal comfort; in other words, it is supposed to be within
the temperature range where the human body feels comfortable.
Clothing insulation of building residents is a key factor affecting
indoor thermal comfort temperature [12,13]. The clothing insula-
tion is regarded as a fixed value in existing relevant standards, e.g.,
the ASHRAE Standard 55–2010 and the ISO 7730:2005 assume that
Fig. 1. Rationale for determining the dynamic setting temperature of air-
the typical clothing insulation is 0.5 clo in summer and 1 clo in
conditioning system.
winter [14,15]. Under the condition of settled indoor thermal en-
vironment parameters such as humidity and radiant temperature,
the thermal comfort temperature based on fixed clothing insula- tual setting temperatures are supposed to be higher in summer
tion is also fixed. However, clothing adjustment is a powerful be- and lower in winter compared with the fixed setting temperature
havioral adaption mode, because people can adjust their clothing (FST) corresponding to the constant clothing insulation. These de-
insulation according to the change of outdoor climate so as to suit viations result in more energy consumption of the air-conditioning
their own thermal comfort requirement [16,17]. Field investigations system.
in different regions have shown that thermal comfort temperature In order to save air-conditioning energy consumption while sat-
changes with clothing insulation. Indraganti and Boussaa proposed isfying the thermal comfort requirements of indoor environment,
that the indoor neutral temperature decrease 0.7 °C for every 0.1 this study proposed a new insight for determining the setting tem-
clo increase in clothing insulation, according to the field surveys perature of the air-conditioning system. According to the variable
conducted in nine air-conditioned office buildings in Qatar during clothing insulation changing with climate, indoor setting temper-
the five summer months [18]. Wang et al. conducted a field survey ature changes dynamically with outdoor air temperature, i.e. dy-
on winter thermal comfort in 23 buildings locating in six commu- namic setting temperature (DST) which adapts to climate change.
nities in Harbin, China [19]. The results showed that the clothing The main work done includes: (1) established a method to re-
insulation increased from 0.77 to 0.88 clo while the indoor neu- alize the new idea, in which air-conditioning setting tempera-
tral temperature gradually decreased from 25.1 to 21.1 °C. Hong ture changes with indoor clothing insulation which adapts to cli-
et al. analyzed the current situation of residential thermal comfort mate change; (2) applied the innovative method in a typical pub-
in winter based on an extensive survey of 2500 dwellings selected lic building and analyzed the energy-saving potential of the air-
from five major urban areas in England [20]. They found that the conditioning system.
clothing insulation decreased by 0.035 clo, leading to 0.2 °C in-
creasing in the neutral temperature. Liu et al. conducted a long-
term field survey in two typical naturally ventilated offices located 2. Method
in Changsha, China [17]. It was concluded that the neutral tem-
perature decreased by approximately 0.4 °C in winter and 0.7 °C For the air-conditioning system in buildings, it is necessary to
in summer when the clothing insulation increased by 0.1 clo. In set the control temperature in advance, i.e., the value of setting
addition, there are regional differences in adaptive thermal com- temperature should be determined ahead. This paper is arranged
fort, such as the time-scale, because the same values of clothing as follows (see Fig. 1): (1) predict the daily mean clothing insu-
insulation might not be suitable for all climate zones or countries lation according to the historical outdoor temperatures; (2) calcu-
[17,21]; thus the adaptive thermal comfort model should be ad- late the indoor thermal comfort temperature based on the dynamic
justed accordingly [22]. clothing insulation; (3) determine the dynamic setting tempera-
According to the studies presented above, there are two limi- ture.
tations in the method which determines the setting temperature
using the constant clothing insulation. First, the dynamic thermal 2.1. Prediction of clothing insulation based on historical temperature
comfort requirement caused by variable clothing insulation is dif-
ficult to be satisfied in practice. Since people’s clothing insulation The variation in indoor clothing insulation is closely related to
changes significantly with climate in public buildings such as shop- past outdoor temperatures with more recent experiences of out-
ping malls [23], the thermal comfort temperature will also change door temperature being more influential in the clothing insulation
accordingly. Second, the constant clothing insulation causes higher [23–25], at the same time, according to the thermal comfort theory
operational energy consumption of air-conditioning system. How- put forward by Fanger, clothing insulation is an important factor
ever, in real life, the clothing insulation is likely to be lower than affecting indoor comfort temperature, thus the historical climate
the standard recommended value in hot summer, while it may influences the indoor clothing insulation and indirectly influence
be higher in cold winter [14,15]. Under these conditions, the ac- the air-conditioning setting temperature.
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 3

This paper predicted the change of the daily mean clothing in- Table 1
Definition of thermal comfort temperature levels.
sulation based on the historical temperature for two reasons: (1)
the historical temperature experienced by people has a signifi- Level Thermal sensation ranges
cant impact on clothing adjustment behavior and adaptive ther- I PMV = 0
mal comfort [24,25]; (2) historical temperature provides reliable II −0.5 ≤ PMV ≤ 0.5
weather conditions for the forecast of air-conditioning setting tem- III −1 ≤ PMV ≤ 1
perature. PMV means predicted mean vote, is an evaluation
In the author’s previous research, a longitudinal thermal com- index of human thermal response.
fort survey (13 months) was conducted in two different types of
naturally ventilated building in Changsha China [26]. 15 subjects Table 2
Parameter settings for calculating thermal comfort tempera-
(9 males and 6 females) working in the two offices were recruited.
ture.
They were allowed to use the air-conditioners and set the tem-
perature freely. However, they were always asked to keep energy- Parameter Value Reference
saving in their mind. There was no restriction on their clothing Metabolic rate 100 W/m 2
[34]
at any time. Repeated questionnaires were conducted to obtain Effective mechanical power 0 [34]
their daily mean indoor clothing insulation. The daily mean cloth- Relative humidity 60% [34]
Air velocity 0.2 m/s [34]
ing insulation model was implemented using 1427 useful clothing
records collected during the survey.
The daily mean clothing insulation model is given by, Table 3
Setting temperature values of air-conditioning sys-
U−L tem.
CLO = (1)
1 + BeCTrm Level Value
where CLO is the daily mean clothing insulation (clo); U and L are
I TPMV = 0
the upper and lower limits for clothing adjustment (clo), which II TPMV=−0.5 or TPMV = 0.5
come from the indoor clothing insulation research completed by III TPMV=−1 or TPMV = 1
the author [26]; B and C are the regression coefficients of the
PMV means predicted mean vote.
model; Trm means the running mean (RM) outdoor temperature TPMV = 0 ,TPMV=−0.5 ,TPMV = 0.5 ,TPMV=−1 and TPMV = 1
(°C). mean the value of the thermal comfort temperature
The RM outdoor temperature in Eq. (1) was calculated accord- when PMV are equal to 0,−0.5,0.5,−1 and 1.
ing to the past outdoor temperature, as shown in Eq. (2).
Trm =(1 − α ){Td−1 +α Td−2 +α 2 Td−3 + · · · +α n−1 Td−n } (2) PMV model had suitable accuracy when the predicted mean vote
where Trm means the running mean (RM) outdoor temperature at was between −1 and 1 [30,31]. Therefore, it is reasonable to cal-
time d (°C); Td-n means the daily mean outdoor temperature at the culate the indoor thermal comfort temperature based on the value
nth day before (°C); α is time constant (0 ≤ α ≤ 1) that reflects the of PMV. A similar method has been used in literatures [32,33].
influence degree of historical temperature. ta = F(M, W, Rh , PMV, Icl , νa ) (4)
The running mean outdoor temperature of any day (Trm(i) ) can
be obtained from the daily mean outdoor temperature and the where ta is the thermal comfort temperature ( °C); M is the
running mean outdoor temperature of the previous day (Ti-1 and metabolic rate (W/m2 ); W is the effective mechanical power
Trm(i-1) ), which is calculated as, (W/m2 ); Rh indicates the air relative humidity (%); Icl indicates the
clothing insulation (clo), and va is the air velocity (m/s). The de-
Trm(i) =(1 − α )Ti−1 +α Trm(i−1) (3) tailed calculation process is listed in Appendix A.
A higher value of α indicates that historical temperature has a The good indoor thermal environment has the characteristics of
greater impact on clothing adjustment behavior [27]. no thermal radiation source, gentle breeze velocity, and medium
humidity, and the occupants usually work in light physical labor
2.2. Calculation of dynamic thermal comfort temperature based on intensity. The setting parameter values for Eq. (4) are listed in
variable clothing insulation Table 2.

Daily mean clothing insulation changes with the influence of 2.3. Determination of DST based on DTCT
outdoor air temperature, which leads to the daily mean ther-
mal comfort temperature changes dynamically. With the predicted The purpose of controlling the air-conditioning system is to
mean vote (PMV) model established by Fanger [13], the daily mean maintain the indoor temperature in the setting thermal comfort
thermal comfort temperature is calculated according to the pre- zone. The setting temperature values are divided into three levels
dicted clothing insulation, i.e. the dynamic thermal comfort tem- according to the three levels of thermal comfort temperature (see
perature (DTCT). Section 2.2), as shown in Table 3.
Generally, people feel comfortable when the thermal sensation The setting temperature of level I is the most stringent, which
is between slightly cool and slightly warm [28,29]. Thus three lev- controls the indoor temperature within a range of no more than
els of thermal comfort temperature are defined in accordance with 0.5 °C from the neutral temperature, because the temperature con-
PMV (Table 1). The thermal comfort temperature of level I refers to trol precision is ± 0.5 °C in existing standards [35]. The setting
the neutral temperature (PMV = 0); thermal comfort temperature temperatures of level II and level III are the boundary values of
range of level II is relatively looser, which refers to the correspond- thermal comfort temperature at the corresponding level, respec-
ing indoor temperature range when PMV is between −0.5 and tively. When the indoor temperature is higher than the upper limit
0.5; thermal comfort temperature range of level III is the broad- of thermal comfort temperature, the refrigeration system works
est, in which people’s thermal sensation is between slightly cool until the indoor temperature is stable at the upper limit. When the
and slightly warm (−1 ≤ PMV ≤ 1). indoor temperature is lower than the lower limit of thermal com-
Then, calculate the air temperature value according to the set- fort temperature, the heating system works until the indoor tem-
ting value of PMV as Eq. (4). Existing studies had verified that the perature is stable at the lower limit. When the indoor temperature
4 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

is within the range of thermal comfort temperature, no operation 3.1. The daily mean outdoor temperatures and running mean outdoor
is needed. The temperature control precision is also set to ± 0.5 °C, temperatures
the same as level I.
The daily mean thermal comfort temperature is calculated According to the standard about the climate season division
based on the variable clothing insulation, and three levels of daily in China [38], the distribution of air-conditioning season days in
setting temperature are determined based on the definition given Changsha 2017 was 127 days in summer (from May 28th to Oc-
in Table 3, i.e., the dynamic setting temperature. tober 1st) and 89 days in winter (from November 18th to Febru-
If the thermal comfort temperature is calculated based on the ary 15th). As shown in Fig. 3, the daily mean air temperature var-
constant clothing insulation recommended by relevant standards, ied between 21 °C and 39 °C in summer, with an average value of
the three levels of setting temperature will not change in summer 29.4 °C; the daily mean air temperature varied between 2 °C and
and winter, respectively, i.e., the fixed setting temperature. 17 °C in winter, with an average value of 10 °C.
The running mean outdoor temperature was calculated by
2.4. Energy-saving potential evaluation method for DST Eq. (2). According to the published results in the reference [39], the
value of α was 0.8 and the starting value was the average temper-
2.4.1. Energy calculation of air-conditioning system in a public ature of the previous seven days. As shown in Fig. 3, the slow vari-
building ation and small fluctuation in the RM outdoor temperature were
The paper took a typical public building as an example, and an- observed compared with the daily mean outdoor temperature. The
alyzed the impact of adopting DST on the air-conditioning energy RM outdoor temperature varied between 25 and 36 °C in summer
consumption through the calculation of operational energy con- and between 6 and 15 °C in winter, respectively.
sumption in summer and winter.
The central air-conditioning system is widely used in public 3.2. The daily mean indoor clothing insulation
buildings, which consists of four parts as shown in Fig. 2: cold
source (water chillers), heat source, water pumps and fans. The en- Indoor daily mean clothing insulation was determined accord-
ergy consumption of air-conditioning system can be calculated as ing to Eq. (1), in which the coefficient values were chosen from
follows. published research results [26]. Fig. 4 shows the day to day change
ES =EWC +EWP +EF (5) of daily mean clothing insulation in summer and winter. The cloth-
ing insulation changed obviously with the RM outdoor temperature
EW =EHS +EWP +EF (6)
and had an opposite trend, it ranged from 0.19 to 0.55 clo in sum-
where ES and EW mean energy consumption of air-conditioning mer and 1.09 to 1.51 clo in winter, with the average value reached
system in summer and winter (kWh); EWC means energy con- 0.39 clo in summer, while increased to 1.33 clo in winter. However,
sumption of water chillers (kWh); EHS is energy consumption of the clothing insulation was constant in some standards (0.5 clo in
heat source (kWh); EWP is energy consumption of water pumps summer and 1 clo in winter) [14,15]. In 89% of summer days, the
(kWh), and EF means energy consumption of fans (kWh). The cal- daily mean clothing insulation was lower than 0.5 clo; while in all
culation equations for the energy consumption of each part are winter days, the daily mean clothing insulation was higher than 1
listed in Appendix B. clo.
DeST was used to simulate the annual energy consumption. It
was developed by the Institute of Environment and Equipment of
3.3. The daily indoor comfort setting temperature
the Department of Building Technology Science of Tsinghua Uni-
versity in the early 1980s and was widely used in building energy
The change of the indoor thermal comfort temperature and dy-
simulations [36]. For example, Yao used DeST for simulating the
namic setting temperature are shown in Fig. 5. The DST of level I,
energy consumption of various typical residential buildings in hot
II and III varied from 20.1 to 27.8 °C, 17.8 to 28.7 °C and 15.4 to
summer and cold winter areas, and for evaluating the energy con-
29.7 °C, respectively.
sumption in different residence units [37].
The number of days when DST was higher than the correspond-
ing level of FST in summer was level-I 66 days, level-II 85 days and
2.4.2. Calculation of energy-saving rate
level-III 113 days, respectively; the number of days when DST was
In order to evaluate the energy-saving potential of DST, the
lower than the corresponding level of FST in winter was level-I 88
energy-saving rate index is defined as.
days, level-II 88 days and level-III 89 days, respectively.
ECI −EVI The value of FST was level-I 26.4 °C, level-II 27.7 °C and level-III
η= (7)
ECI 28.9 °C in summer, and level-I 22.4 °C, level-II 20.6 °C and level-III
where η indicates annual energy saving rate (%); ECI and EVI mean 18.8 °C in summer, respectively.
annual energy consumption of the air-conditioning system corre-
sponding to FST and DST (kWh). A higher value of η indicates a 3.4. The energy-saving rate of DST
higher energy-saving potential.
Using the method presented in Section 2.2.1, a public build-
3. Results ing was used as an example and the annual energy consumption
of air-conditioning system with two kinds of setting temperature
We took the meteorological data of Changsha China in 2017 as (DST and FST) was calculated. The interrelated calculation infor-
an example, and calculated the DST and analyzed its energy-saving mation includes the plan layout, thermal physical parameters of
potential compared with adopting the FST. Changsha is the capi- enclosure, etc. (see Appendix C). The calculated values of daily en-
tal of Hunan province located in the south central area of China ergy consumption are shown in Appendix D.
(28°11 N, 112°58 E), with a climate of hot summer and cold winter. Table 4 lists the annual operational energy consumption values
The air-conditioning system needs to be turned on to maintain in- of the air-conditioning system. It indicates that the looser the set-
door thermal comfort for both summer and winter. Therefore, the- ting temperature range, the lower the energy consumption. The
ses climatic characteristics provide good calculation conditions for air-conditioning energy consumption at level-III setting tempera-
our study. ture is about 40.9% of that at level-I. Compared with adopting the
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 5

Fig. 2. Structure of air-conditioning system in summer and winter.

6 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

Fig. 3. The change of the daily mean outdoor temperature and the running mean outdoor temperature.

Table 4
Annual energy consumption and energy saving rate at all levels.

Level Annual energy consumption of DST (kWh) Annual energy consumption of FST (kWh) η (%)
I 493,574.0 585,240.1 15.7%
II 347,333.4 483,876.9 28.2%
III 201,833.8 303,738.3 33.6%

DST: dynamic setting temperature of air-conditioning system, FST: fixed setting temperature of air-conditioning
system; η: energy saving rate of dynamic setting temperature compared with fixed setting temperature.
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 7

Fig. 4. The change of the daily mean clothing insulation.

corresponding level of FST, the energy-savings can be up to 33.6% [42,43,57], the method proposed in this paper focused on the air-
at level-III DST and down to 15.7% at level-I DST. conditioned environment based on the theory of thermal comfort.
The developed method determined the setting temperature dy-
4. Discussion namically which adapts to climate change. Thermal adaptation is
an important approach to maintain thermal comfort in response to
This study proposed a new method to realize the energy sav- the climate change [23,44,58]. Nowadays, in order to reduce energy
ings of the air-conditioning system by adjusting the setting tem- consumption in buildings, dynamic thermal comfort models and
perature dynamically. Different from the traditional measures to standards for naturally ventilated environment were also based on
reduce air-conditioning energy consumption that focused on build- the thermal adaptation theory [45,46].
ing enclosure structure [40,41] or air-conditioning components
8 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

Fig. 5. The daily setting temperature of air-conditioning system. DTCT means dynamic thermal comfort temperature, DST dynamic setting temperature of air-conditioning
system and FST fixed setting temperature of air-conditioning system.
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 9

The theoretical basis of DST is that thermal comfort temper- in some periods of summer, while it is lower in winter. In the pre-
ature is affected by clothing insulation which adapts to climate sented example, there are 113 days of DST higher than FST in sum-
change. Some results of field investigation revealed that clothing mer and 89 days lower in winter. On the other hand, the indoor
adjustment is a powerful behavioral adaption mode for the cli- temperature can be maintained within the range of DTCT with-
mate change [47–49,59]. However, in the current standards, cloth- out air-conditioning operation in some time periods, due to the
ing insulation is assumed to be constant (0.5 clo in summer and 1 DTCT changes with climate, which means that the thermal com-
clo in winter) when calculate thermal comfort temperature [14,15]. fort can be guaranteed only by clothing adjustment behavior. From
Such assumption does not reflect the clothing adjustment behav- the calculation, it is seen that when applying the DST of level I, II
ior which adapts to thermal comfort though it may be suitable and III, the corresponding number of days without air-conditioning
for some uniformed staff. If the clothing insulation is lower than operation is 1, 10 and 29 days respectively. As shown in Table 4,
0.5 clo in summer and higher than 1 clo in winter, it means that the energy-saving of Level III is the highest, mainly because the
the setting temperature can be higher in summer and lower in difference between DST and FST increases as the level rises. The
winter compared with the recommended values, which leads to maximum difference between DST and FST of level I, II, and III, as
a corresponding reduction in operational energy consumption. In shown in Fig. 5, is 2.8, 3.3, and 3.9 °C, respectively; meanwhile,
fact, this is a common occurrence as pointed out by some pre- the number of days without air-conditioning operation is the most
vious researches. In Changsha, China, the indoor clothing insula- when adopting DST of level III, accounting for 13.4% of the total
tion ranged from 0.2 to 1.25 clo in summer and 0.98 to 1.88 clo days.
in winter. Compared with the recommended value, 75% of dates Theoretically, the indoor environment can adapt to the annual
were lower in summer and 90% of dates were higher in winter. In climate change after applying DST to the automatic control of the
Shanghai, China, the clothing insulation ranged from 0.22 to 0.76 air-conditioning system. But how to achieve this goal still needs
clo in summer, with an average value of 0.45 clo; while it ranged to be investigated further. Here we present a method to realize
from 0.75 to 2.34 clo in winter, with an average value of 1.39 clo it: (1) monitor and record the thermal environment parameters
[50]. Similar results also appeared in the study conducted in Rome, and outdoor air temperature through measuring points; (2) cal-
Italy, in which 80% of men and 70% of women’s wore less indoor culate setting temperature of the next day in advance with the
clothing in summer, and 75% of men and women wore more in method proposed in Section 2; (3) set the control temperature of
winter, compared with the recommended value (0.5 clo in sum- air-conditioning system automatically.
mer and 1 clo in winter) [51]. In the typical indoor thermal envi- Reasonable prediction of clothing insulation based on outdoor
ronment (Table 2), the neutral temperature decreased by approx- air temperature is essential to determine the DST. The clothing in-
imately 0.6 °C when the clothing insulation increased by 0.1 clo, sulation prediction model were determined based on the survey
according to the PMV model. Thus it is reasonable to adopt DST for conducted in Changsha China [26]. Thus it must be validated and
air-conditioning energy savings. In addition, due to the considera- adjusted according to different climate conditions and areas, con-
tion of variable clothing insulation which adapts to climate change, sidering the differences in climate and habits of clothing adjust-
the DST is more in line with real thermal comfort requirements. ment behavior.
This paper defined three levels of air-conditioning setting tem-
perature (Table 3), corresponding to different boundary values of
thermal comfort temperature ranges (Table 1). Level I is expected 5. Conclusions
to stabilize the ambient temperature at the neutral temperature, so
as to maintain the human sensation at thermal neutrality, which is This paper proposed a method for reducing energy consump-
also recommended in relevant standards [46]. Level III is expected tion of air-conditioning system based on dynamic setting tempera-
to keep the ambient temperature in a large range as far as possible, ture. The following main conclusions were obtained.
while the human sensation is between slightly cooler and slightly
warmer. The control intensity of level II for the air-conditioning en- (1) The DST varied with outdoor air temperature, and its determi-
vironment is between the two levels mentioned before. With the nation method was based on the theory of clothing adjustment
expansion of allowed ambient temperature range, the setting tem- behavior which adapts to thermal comfort. The calculation re-
perature can be higher in summer and lower in winter, leading to sults showed that under the climatic conditions of Changsha,
a significant reduction in air-conditioning energy consumption. As China, the DST varied from 22.6 to 30.2 °C in summer and 14.9
shown in Table 4, the setting temperature of level III can reduce to 26.8 °C in winter.
the energy consumption by 59.1% compared with level I. On the (2) The adoption of DST could reduce the operational energy con-
other hand, according to Fanger’s PMV-PPD model [13], the pre- sumption of the air-conditioning system. The calculation results
dicted percentage of dissatisfied corresponding to thermal comfort showed that its energy-saving can be up to 65.5% compared
temperature of level I, II and III are 5%, 10% and 27%, respectively. with the FST recommended by relevant standards.
In other words, even the setting temperature of level III which is (3) three levels of air-conditioning setting temperature were de-
determined on the basis of the least restrictive thermal comfort fined according to the corresponding thermal comfort tempera-
temperature can satisfy most people (over 73%) and meet the re- ture ranges. The calculation results showed that 59.1% of energy
quirements of standard [14]. Some studies conducted by Martin consumption can be saved with the application of level-III set-
and Cottafava et al. also found that properly reduced the percent- ting temperature compared with level-I.
age of thermal acceptability votes can decrease the energy con- (4) The air-conditioning environment which adapts to climate
sumption of air-conditioning system [52,53]. change can be achieved through the combination of DST and
The calculation results of air-conditioning energy consumption automatic control.
are listed in Table 4, which shows the significant energy saving
potential of DST. The energy-saving of adopting DST ranges from
15.7 to 33.6% compared with FST at the same level. The level-I CRediT authorship contribution statement
FST is often used in the current air-conditioning control scheme
and the reduced energy consumption of DST can be up to 65.5% Xinbo Xu: Formal analysis, Writing - original draft, Data cu-
as a comparison. There are mainly two reasons for adopting DST ration. Weiwei Liu: Conceptualization, Methodology, Writing - re-
to save energy. On the one hand, the DST is higher than the FST view & editing, Supervision. Zhiwei Lian: Resources, Supervision.
10 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

Acknowledgements [2] D.H. Kang, P.H. Mo, D.H. Choi, S.Y. Song, M.S. Yeo, K.W. Kim,
Effect of MRT variation on the energy consumption in a PMV-
The project was financially supported by the National Natural controlled office, Building and Environment 45(9) (2010) 1914–
Science Foundation of China (No. 51778625 and 51478471). The au- 1922.
thors would like to thank the participants who volunteered for this
study. Appendix B. Energy consumption formulas for each part of
air-conditioning system
Supplementary materials
(a) Energy consumption of water chillers
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.enbuild.2020.110086. The water chillers provided cooling capacity for air-conditioning
system in summer, and its input power and running time were de-
Appendix A. Calculation of thermal comfort temperature ta termined according to different load rates. The energy consump-
tion can be calculated as
According to the thermal comfort theory put forward by Fanger

EWC = Pi ×Ti (1)
[1], PMV can be calculated as
 
P MV = 0.303exp−0.036M + 0.0275 ∗ {M − W − 3.05[5.733 − 0.007(M − W ) − Pa ]}
−0.42(M − W − 58.2) − 0.0173M (5.867 − Pa ) − 0.0014M (34 − ta ) − 3.96 ∗ 10−8
 
fcl (tcl + 273 )4 − (tr + 273 )4 − fcl hc (tcl − ta ) (1)

Where ta is the air temperature (°C); M is the metabolic rate

(W/m2 ); W is the effective mechanical power (W/m2 ); Pa indicates where EWC means the energy consumption of water chillers (kWh),
the water vapor partial pressure (Pa ); fcl indicates the clothing sur- Pi is the input power under relevant partial load rate (kW) and Ti
face area factor; tr is the mean radiant temperature (°C); tcl is the is the operating time for partial load (h).
clothing surface temperature, and hc is the convective heat transfer
coefficient [W/(m2 •K)]. (b) Energy consumption of heat source
The clothing surface area factor fcl can be calculated as
 The fuel consumed by heat source during the whole heating
1.00 + 1.290Icl , Icl ≤ 0.078m2 · C/W season can be calculated as following,
fcl = (2)
1.05 + 0.645Icl , Icl > 0.078m2 · C/W
QH
C= (2)
Where Icl is the clothing insulation (m2 •K/W), the calculation q c ×η 1 ×η 2
process see Section 2.1. where C means the fuel consumption (kg), QH is the heat supplied
The clothing surface temperature tcl can be calculated as by the heat source (kJ), qc is the calorific value of fuel equivalent
tcl = 35.7 − 0.028(M − W ) (kJ/kg), η1 means the efficiency of heat exchanger and η2 means
   the conversion efficiency from fuel to heat. In order to compare
−Icl 3.96 × 10−8 fcl × (tcl + 273 )4 − (tr + 273 )4
the energy consumption of various parts throughout the year, the
+ fcl hc (tcl − ta )} (3) unified energy consumption unit was shown as,
The convective heat transfer coefficient hc can be calculated as C
EHS = (3)
 √ G
2.38(tcl − ta ) , 2.38(tcl − ta )
0.25 0.25
> 12.1 va where EHS means the energy consumption of heat source and G is
hc = √ √ (4)
12.1 va , 2.38(tcl − ta )0.25 < 12.1 va the conversion factor, that is to say, G kg fuel equivalent for 1 kWh
standard electricity. The values of all conversion factors in the heat
The mean radiant temperature tr can be calculated as
source calculation process were listed in Table B1.
tr =ta + 0.5 (5)
(c) Energy consumption of water pumps
Combined with Eq. (1)–(5), the calculation equation of PMV can
be expressed as
Two pumps were set to work in parallel with constant volumes
PMV = F(M, W, Rh , ta , Icl , νa ) (6) in winter and variable volumes in summer. If the partial loads
were less than 50%, then only one pump would be operated. The
This manuscript uses the method of determining the ther-
energy consumption of water pumps were calculated as follows,
mal comfortable temperature proposed by Kang et al. [2]. The
 Qi ×Hi
Eq. (6) can be conversed to Eq. (7). EWP = · ti (4)
ηi
ta = F(M, W, Rh , PMV, Icl , νa ) (7)
When the variables in Eq. (7) are determined (see Table 2), ta is Table B1
a single-value function of PMV. In the practical process of calculat- Coefficient value in heat source calculating.
ing the boundary of the comfort temperature range in this study, Coefficient Gas steam boiler Reference
ta is calculated by multiple iterations after the error range is de-
η1 0.94 [1]
termined.
η2 0.92 [2]
G 9.98 [3]
References
η1 means the efficiency of heat exchanger, η2 the effi-
[1] P.O. Fanger, Thermal Comfort: Analysis and Applications in En- ciency of converting fuel into heat and G the refraction
vironmental Engineering, McGraw-Hill, New York, 1972. coefficient of fuel to electric energy.
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 11

Table B2 (d) Energy consumption of fans

Rated performance parameters of water chillers.

Er Qr1 Qr2 Pr1 Pr2 The energy consumption of fans were determined by working
Category (kW) (m3 /h) (m3 /h) COP (kPa) (kPa) states which depend on the characteristics of fans and air ducts.
Centrifugal refrigerator 1182.94 203.96 244.75 5 76.16 73.86 When the constant air volume system was used, the characteristics
of ducts were unchanged. The energy consumption of fans were
Er means refrigerating effect, Qr1 chilled water flow, Qr2 cooling water flow, COP
coefficient of performance, Pr1 pressure drop in chilled water side and Pr2 pressure calculated as follows,
drop in cooling water side.  Qj ×Pj
EF = · tj (5)
ηj
Table B3 where EF means the energy consumption of fans (m3 /s), Pj is the
Rated performance parameters of heat source. practical pressure (kPa ), ηj means efficiencies of fans and tj means
Eh Qh Ph Th1 Th2 ηh actual work time of fans (s).
Category (kW) (m3 /h) (kPa) (°C) (°C) (%) The types and performance parameters of air conditioning com-
Gas steam boiler 644.82 9.26 55.16 125 65 84 ponents are listed in Table B2, B3, B4 and B5.

Eh means calorific production, Qh steam flow, Ph pressure drop, Th1 outlet temper-
ature, Th2 inlet temperature and η efficiency of heat source.
References
Table B4 [1] Y. Ding, Q. Fu, Z. Tian, M. Li, N. Zhu, Influence of indoor design
Rated performance parameters of air supply fans.
air parameters on energy consumption of heating and air condi-
Category Hf (Pa) hf (kJ/kg) ηf (%) R (%) tioning, Energy and Buildings 56 (2013) 78–84.
Non-frequency conversion 1300 20 70 30 [2] TSG G002-2010, Supervision Administration Regulation on En-
ergy Conservation Technology for Boiler, Xinhua Publishing House,
Hf means head of fan, hf enthalpy difference of air supply, ηf efficiency of
Beijing, 2010.
fan and R minimum air volume ratio.
[3] GB/T 2589–2008, General principles for calculation of total
production energy consumption, Standards Press of China, Beijing,
where EWP means the energy consumption of water pumps, Qi
2008.
is the practical water volume (m3 /s), Hi is the practical pressure
(kPa ), ηi means efficiencies of water pumps and ti means actual
work time of water pumps (s).

Table B5
Rated performance parameters of pumps.

Position Category Qp (m3 /h) Hp (kPa) ηp (%)

Cold source primary pump Constant speed pump 224.35 303.78 65
Cooling water pump Constant speed pump 269.22 238.40 65
Water supply pump Constant speed pump 431.00 175.00 75
Heat source primary pump Constant speed pump 61.15 275.27 65
Heat pump Constant speed pump 24.46 225.36 65

Qp means flow rate of pump, Hp head of pump and ηp efficiency of pump.

12 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

Fig. C1. Architectural plan.

Table C1 References
Building related parameters.
[1] JGJ 48–2014, Design Code for Store Building, China Architecture
Shopping mall Reference
& Building Press, Beijing, 2014.
Floor height 6m [2] JGJ 134–2010, Design Standard for Energy Efficiency of Residen-
Building area 5000 m2 tial Buildings in Hot Summer and Cold Winter Zone, China Archi-
HVAC system All-air system
Lighting load 14.5 W/m2 [1]
tecture & Building Press, Beijing, 2010.
People density 6 m2 /p [1]
Facility load 16.5 W/m2 [1]
Fresh air volume 30 m3 /(h.p.) [1] Appendix D. Energy consumption of air-conditioning system
Schedule Whole year: 7:00–20:00 [1]
The change of daily air-conditioning energy consumption are
shown in Fig. D1. The air-conditioning energy consumption adopt-
ing FST of level I, II and III varied from 145.3 to 5461 kWh, 0 to
Appendix C. Building information 4678.6 kWh and 0 to 3896.3 kWh, respectively. The maximum val-
ues of air-conditioning energy consumption adopting DST of level
The typical business building for the simulation study was lo- I, II and III are 4344.7 kWh, 3510.2 kwh and 2763.7 kwh, respec-
cated in Wangchengpo District, Changsha, China. The one-story tively, while the minimum value is 0.
building with south-north orientations, it covered a total floor area The change of daily air-conditioning energy consumption in var-
of 50 0 0 m2 s and is 6 m high, had no basement and adopted flat ious parts are shown in Fig. D2. The energy consumption of differ-
roof. Plan structure of the building was shown in Fig. C1. ent positions in air-conditioning system decrease with the increase
Table C1 lists the values of load indicators used in this building. of temperature control level.
The external loads of the building are mainly transmitted into As shown in Fig. D3, compared with FST, the air-conditioning
the room through the enclosure structure. In the hot summer and energy consumption of different positions adopting DST are lower
cold winter areas such as Changsha, the materials of enclosure in different levels, such as the energy consumption of water pumps
structure will significantly affect indoor thermal comfort. In order reduced by 40.1% in level III. The cold and heat source con-
to enhance the thermal insulation performance of enclosure struc- sume the most energy, and the fans consume the least. Com-
ture, selecting materials with suitable heat transfer coefficients in pared with level I, the energy consumption of cold and heat
accordance with the national energy-saving standard [2]. The pa- source in level II and level III is reduced by 30.9% and 62.1%,
rameters are shown in Table C2. respectively.

Table C2
Basic information of building enclosure structure.

Position Type Area (m2 ) Heat transfer coefficient (W/m2 · K) Reference

Exterior wall 24 Brick wall/Polyphenyl board insulation 1580 0.4 [2]

Outer door Glass outdoor door of shopping mall 120 5.2 [2]
External windows Vacuum/Low-E film glass plating 100 1.8 [2]
Ground 40 mm concrete floor 5000 1.7 [2]
Roof Aerated concrete thermal insulation roof 5000 0.6 [2]
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 13

Fig. D1. The change of daily air-conditioning energy consumption. DST means dynamic setting temperature of air-conditioning system and FST fixed setting temperature of
air-conditioning system. The daily energy consumption of the fresh air system is 397.22 kWh, which is not included in the energy consumption of air-conditioning system.
14 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

Fig. D2. The change of daily air-conditioning energy consumption in various parts. DST means dynamic setting temperature of air-conditioning system and FST fixed setting
temperature of air-conditioning system.
 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086 15

Fig. D3. Energy consumption of the different positions in air-conditioning system. DST means dynamic setting temperature of air-conditioning system and FST fixed setting
temperature of air-conditioning system.

References fore and after energy efficient refurbishment, Build. Environ. 44 (6) (2009)
1228–1236.
[1] F. Zhang, R. de Dear, P. Hancock, Effects of moderate thermal environments [21] W. Liu, Y. Zheng, Q. Deng, L. Yang, Human thermal adaptive behaviour in nat-
on cognitive performance: a multidisciplinary review, Appl. Energy 236 (2019) urally ventilated offices for different outdoor air temperatures: a case study in
760–777. Changsha China, Build. Environ. 50 (2012) 76–89.
[2] G. Calis, M. Kuru, Assessing user thermal sensation in the Aegean region [22] D. Yang, J. Xiong, W. Liu, Adjustments of the adaptive thermal comfort model
against standards, Sustain. Cities Soc. 29 (2017) 77–85. based on the running mean outdoor temperature for Chinese people: a case
[3] W. Wu, H.M. Skye, P.A. Domanski, Selecting HVAC systems to achieve comfort- study in Changsha China, Build. Environ. 114 (2017) 357–365.
able and cost-effective residential net-zero energy buildings, Appl. Energy 212 [23] C. Morgan, R. de Dear, Weather, clothing and thermal adaptation to indoor
(2018) 577–591. climate, Climate Res. 24 (3) (2003) 267–284.
[4] M. Nemś, J. Kasperski, A. Nemś, A. Bać, Validation of a new concept of a solar [24] F. Haldi, D. Robinson, Modelling occupants’ personal characteristics for thermal
air heating system with a long-term granite storage bed for a single-family comfort prediction, Int. J. Biometeorol. 55 (5) (2011) 681–694.
house, Appl. Energy 215 (2018) 384–395. [25] P.M. de Carvalho, M.G. da Silva, J.E. Ramos, Influence of weather and indoor
[5] K. Lundgren, T. Kjellstrom, Sustainability challenges from climate change and climate on clothing of occupants in naturally ventilated school buildings, Build.
air conditioning use in urban areas, Sustainability 5 (7) (2013) 3116–3128. Environ. 59 (2013) 38–46.
[6] J.L. Ouyang, J. Ge, T.L. Shen, K. Hokao, J.L. Gao, The reduction potential of en- [26] W. Liu, D. Yang, X. Shen, P. Yang, Indoor clothing insulation and thermal his-
ergy consumption, CO2 emissions and cost of existing urban residential build- tory: a clothing model based on logistic function and running mean outdoor
ings in Hangzhou city, China, J. Asian Archit. Build. Eng. 7 (1) (2008) 139–146. temperature, Build. Environ. 135 (2018) 142–152.
[7] H. Liu, X.-l. Ma, Y.-f. Gao, Impact of summer office set air-conditioning temper- [27] F. Nicol, M. Humphreys, Derivation of the adaptive equations for thermal com-
ature on energy consumption and thermal comfort, J. Cent. South Univ. Tech- fort in free-running buildings in European standard EN15251, Build. Environ.
nol. 16 (2009) 6–11. 45 (1) (2010) 11–17.
[8] K.K.W. Wan, D.H.W. Li, J.C. Lam, Assessment of climate change impact on [28] A.P. Gegge, Introduction to Thermal Comfort, INSERM, 1977.
building energy use and mitigation measures in subtropical climates, Energy [29] P.O. Fanger, Thermal Comfort: Analysis and Applications in Environmental En-
36 (3) (2011) 1404–1414. gineering, McGraw-Hill, New York, 1972.
[9] X. Wang, D. Wang, S. Chen, J. Wu, Simulation of low energy consumption strat- [30] R. de Dear, J. Kim, C. Candido, M. Deuble, Adaptive thermal comfort in Aus-
egy for residential buildings in hangzhou based on clustering behaviour, in: tralian school classrooms, Build. Res. Inf. 43 (3) (2015) 383–398.
J. Xiao, J. Ke, Z. You (Eds.), 2018 International Symposium on Mechanics, Struc- [31] J. van Hoof, Forty years of Fanger’s model of thermal comfort: comfort for all?
tures and Materials Science, 2018. Indoor Air 18 (3) (2008) 182–201.
[10] Y. Ding, Q. Fu, Z. Tian, M. Li, N. Zhu, Influence of indoor design air parame- [32] D.H. Kang, P.H. Mo, D.H. Choi, S.Y. Song, M.S. Yeo, K.W. Kim, Effect of MRT vari-
ters on energy consumption of heating and air conditioning, Energy Build. 56 ation on the energy consumption in a PMV-controlled office, Build. Environ. 45
(2013) 78–84. (9) (2010) 1914–1922.
[11] N. Yamtraipat, J. Khedari, J. Hirunlabh, J. Kunchornrat, Assessment of Thailand [33] Y. Zhang, Z. Wu, M. Zhang, J. Mai, L. Jin, F. Wang, Smart indoor humidity and
indoor set-point impact on energy consumption and environment, Energy Pol- condensation control in the spring in hot-humid areas, Build. Environ. 135
icy 34 (7) (2006) 765–770. (2018) 42–52.
[12] K.H. Lee, S. Schiavon, Influence of three dynamic predictive clothing insulation [34] GB/T 18049-2017, Ergonomics of the Thermal Environment-Analytical Deter-
models on building energy use, HVAC sizing and thermal comfort, Energies 7 mination and Interpretation of Thermal Comfort Using Calculation of the PMV
(4) (2014) 1917–1934. and PPD Indices and Local Thermal Comfort Criteria, Standards Press of China,
[13] P.O. Fanger, Thermal comfort, Danish Technical Press, Copenhagen, 1970. Beijing, 2017.
[14] ISO 7730, Moderate Thermal Environments-Determination of the PMV and PPD [35] GB/T 25858-2010, Performance Testing Method of Precision Air Conditioning
Indices and Specification of the Conditions for Thermal Comfort, International Unit, Standards Press of China, Beijing, 2017.
Organization for Standardization, Geneve, Switzerland, 1994. [36] Manuals for DeST-C. DeST group of Tsinghua University. http://dest.tsinghua.
[15] ASHRAE, ANSI/ASHRAE Standard 55-2013: Thermal Environmental Conditions edu.cn/
for Human Occupancy, ASHRAE, Atlanta, USA, 2013. [37] J. Yao, Energy optimization of building design for different housing units in
[16] K.C. Parsons, The effects of gender, acclimation state, the opportunity to adjust apartment buildings, Appl. Energy 94 (2012) 330–337.
clothing and physical disability on requirements for thermal comfort, Energy [38] Q.X./.T. 152-2012, Division of Climatic Season, Meteorological Publishing
Build. 34 (6) (2002) 593–599. House, Beijing, 2012.
[17] W. Liu, Q. Deng, W. Ma, H. Hao, J. Zhao, Feedback from human adaptive be- [39] F. Nicol, M. Humphreys, Derivation of the adaptive equations for thermal com-
haviour to neutral temperature in naturally ventilated buildings: physical and fort in free-running buildings in European standard EN15251, Build. Environ.
psychological paths, Build. Environ. 67 (2013) 240–249. 45 (1) (2010) 11–17.
[18] M. Indraganti, D. Boussaa, Comfort temperature and occupant adaptive be- [40] F. Goia, M. Haase, M. Perino, Optimizing the configuration of a facade module
haviour in offices in Qatar during summer, Energy Build. 150 (2017) 23–36. for office buildings by means of integrated thermal and lighting simulations in
[19] Z. Wang, L. Zhang, J. Zhao, Y. He, A. Li, Thermal responses to different residen- a total energy perspective, Appl. Energy 108 (2013) 515–527.
tial environments in Harbin, Build. Environ. 46 (11) (2011) 2170–2178. [41] S. Obyn, G. van Moeseke, Variability and impact of internal surfaces convective
[20] S.H. Hong, J. Gilbertson, T. Oreszczyn, C. Green, I. Ridley, G. Warm Front Study, heat transfer coefficients in the thermal evaluation of office buildings, Appl.
A field study of thermal comfort in low-income dwellings in England be- Therm. Eng. 87 (2015) 258–272.
16 X. Xu, W. Liu and Z. Lian / Energy & Buildings 220 (2020) 110086

[42] D. Zhang, X. Zhang, N. Cai, Study on energy saving possibility of digital variable [52] E. Barbadilla-Martin, J. Guadix Martin, J.M. Salmeron Lissen, J. Sanchez Ramos,
multiple air conditioning system in three office buildings in Shanghai, Energy S. Alvarez Dominguez, Assessment of thermal comfort and energy savings in
Build. 75 (2014) 23–28. a field study on adaptive comfort with application for mixed mode offices,
[43] L. Mucke, D. Fleig, K. Vajen, U. Jordan, Hybrid liquid desiccant air-conditioning Energy Build. 167 (2018) 281–289.
systems: a conceptual study with respect to energy saving potentials, Int. J. [53] D. Cottafava, S. Magariello, R. Ariano, et al., Crowdsensing for a sustainable
Refrig.-Rev. Int. Du Fr. 69 (2016) 64–73. comfort and for energy saving, Energy Build. 186 (2019) 208–220.
[44] G.S. Brager, R.J de Dear, Thermal adaption in the built environment: a litera- [54] Xiaojun Fan, Weiwei Liu, Wargocki Pawel, et al., Physiological and psycholog-
ture review, Energy Build. 27 (1) (1998) 83e96. ical reactions of sub-tropically acclimatized subjects exposed to different in-
[45] ASHRAE, Thermal Comfort. ASHRAE Handbook: Fundamentals, American So- door temperatures at a relative humidity of 70%, Indoor Air 29 (2019) 215–230.
ciety of Heating, Refrigerating and Air-Conditioning Engineers, Inc., Atlanta, [55] Jingyun Shen, Xi Zhang, Zhiwei Lian, Impact of wooden versus nonwooden in-
2013. terior designs on office workers’ cognitive performance, Perceptual and Motor
[46] GB/T 33658-2017, Thermal Comfort Requirements and Evaluation For Indoor Skills (2020) 36–51.
Environment, Standards Press of China, Beijing, 2017. [56] Minghui Zhu, Weiwei Liu, Wargocki Pawel, Changes in EEG signals during the
[47] J.F. Nicol, I.A. Raja, A. Allaudin, G.N. Jamy, Climatic variations in comfortable cognitive activity at varying air temperature and relative humidity, Journal of
temperatures: the Pakistan projects, Build. Environ. 30 (1999) 261–279. Exposure Science & Environmental Epidemiology 30 (2020) 285–298.
[48] F. Haldi, D. Robinson, Modelling occupants’ personal characteristics for thermal [57] Jialin Wu, Zhiwei Lian, et al., A method to evaluate building energy consump-
comfort prediction, Int. J. Biometeorol. 55 (5) (2011) 681–694. tion based on energy use index of different functional sectors, Sustainable
[49] S. Schiavon, K.H. Lee, Dynamic predictive clothing insulation models based Cities and Society 53 (2020).
on outdoor air and indoor operative temperatures, Build. Environ. 59 (2013) [58] Jing Xiong, Perceptual and physiological responses of elderly subjects to mod-
250–260. erate temperatures, Building and Environment 156 (2019) 117–122.
[50] Y. Jiao, H. Yu, T. Wang, Y. An, Y. Yu, The relationship between thermal envi- [59] Manchen He, Zhiwei Lian, et al., Evaluation on the performance of quilts based
ronments and clothing insulation for elderly individuals in Shanghai, China, J. on young people’s sleep quality and thermal comfort in winter, Energy Build
Therm. Biol. 70 (2017) 28–36. 183 (2019) 174–183.
[51] F. Salata, I. Golasi, V. Ciancio, F. Rosso, Dressed for the season: clothing and
outdoor thermal comfort in the Mediterranean population, Build. Environ. 146
(2018) 50–63.