Energy and Buildings 159 (2018) 558–571

Contents lists available at ScienceDirect

Energy and Buildings

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

Experimental study of a local ventilation strategy to protect

semi-exposed relics in a site museum
Xilian Luo a,b,∗ , Zhaolin Gu a,b,∗ , Wei Tian b,c , Yin Xia d , Tao Ma e
a
School of Human Settlements and Civil Engineering, Xi’an Jiaotong University, Xi’an 710049, China
b
Shaanxi Energy environmental & Building Energy Conservation Engineering Technology Research Center, Xi’an Jiaotong University, Xi’an 710049, China
c
China United Northwest Institute for Engineering Design & Research Co., Ltd,No.81 Laodong Road, Xi’an, 710082, China
d
Emperor Qinshihuang’s Mausoleum Site Museum, Xi’an 710600, China
e
Shaanxi Provincial Institute for the Conservation of Cultural Heritage, Xi’an 710075, China

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

Article history: Indoor environmental controls are critical considerations when preserving artifacts in museums. Con-
Received 30 June 2017 trolling the indoor environment of site museum buildings with large space layout typically consumes
Received in revised form 1 December 2017 large amounts of energy. Most of the energy load needed to preserve relics is dissipated throughout
Accepted 1 December 2017
the non-occupied space, because the environments that contain the relics occupy relatively little space
Available online 5 December 2017
compared to the entire exhibition hall. In this study, an energy efficient local ventilation system was
proposed to independently control the environment of a funerary pit, which was preserved within a
Keywords:
museum separate from the large exhibition hall. An experimental hall, which consists of a funerary pit
Local ventilation system
Relic preservation
and an experimental local ventilation system, was built to simulate a site museum building. Experiments
Energy saving were conducted to investigate the system’s performance and to validate its ability to preserve historical
Site museum terracotta figurines in their semi-exposed pits within the exhibition hall. The study evaluated the stability
Local environmental control of the conservation environment and the system’s energy usage by assessing the effects of the ventila-
tion rate. The experimental results showed that the local ventilation system provides an energy-saving
strategy supporting the environmental control of the funerary pit. The LV system significantly decreased
temperature fluctuations in the funerary pit and temperature differences across semi-exposed relics. The
system could be used to control the local preservation environment of the funerary pit separately from
a large space exhibition hall.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction
Energy consumption within buildings is responsible for 40% of
the global energy requirement in the EU, the USA, and the UK [1–3],
while it is responsible for more than one-third in China [4,5]. One-
fifth of utilized energy in China is consumed by public buildings
[5]. Fossil fuel depletion, growing energy demands from build-
ing users, and governmental policies for reducing environmental
pollution emissions have increased the urgency for finding tech-
nological solutions to reduce energy consumption and increase
energy efficiency. These solutions are particularly important for
public buildings with high energy consumption, such as office
buildings, schools, hotels, and museums [6,7].
∗ Corresponding authors at: School of Human Settlements and Civil Engineering,
Xi’an Jiaotong University, Xi’an 710049, China
E-mail addresses: xlluo@mail.xjtu.edu.cn (X. Luo), guzhaoln@mail.xjtu.edu.cn
(Z. Gu).
China is an ancient civilized country with more than 4500 muse-
ums [8]; many of which are site museums, indicating that these
museums are built directly at the place where the relics were found.
Site museums preserve the original panorama of immovable his-
toric sites in an exhibition hall, which generally means that most
of the museum has a large open space layout [9]. However, this
type of special building consumes significant amounts of energy for
air conditioning (AC) systems, to create a stable and clean preser-
vation microenvironment for the preservation of relics. Energy
consumption in a museum building is typically of less importance in
comparison to the demands of providing the appropriate preserva-
tion environment for collections [10]. However, wherever possible,
energy use should be minimized, because cultural relics require
a continuously working AC system throughout the year. This sig-
nificantly increases the economic burden of the site museum’s
operational costs.
Given their significant energy consumption, more studies are
required to fully understand microclimate control within site
museums. Many recent studies have identified different ther-

https://doi.org/10.1016/j.enbuild.2017.12.001
0378-7788/© 2017 Elsevier B.V. All rights reserved.
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571
mal requirements for artifact preservation compared to visitor
comfort, concluding that both should be treated independently
[11–14]. In general, environmental controls in visitor areas can
be conveniently implemented using traditional central AC systems
[15,16]. However, environmental controls for the preservation of
unearthed relics remain a scientific and technological challenge,
due to their stringent environmental requirements and unique
preservation surroundings. Different requirements for visitor com-
fort and artifact preservation may cause museum staff to overlook
the particular requirements of the relics. As a result, significant fluc-
tuations can be observed in the indoor environment, i.e., the daily
temperature fluctuations reach more than 10 ◦ C in the Emperor
Qin’s Mausoleum Site Museum (China) and the Jinsha Site Museum
(China) [17]. Many relics may suffer deterioration or ruin due to the
adoption of improper preservation environments [18].
The Emperor Qin’s Mausoleum Site Museum in Xi’an, China is a
world-famous historical site, designated by United Nations Educa-
tional, Scientific, and Cultural Organization as one of the world’s
cultural legacies. To provide an energy efficient AC system that
protects the immobile relics, the Ministry of Science and Tech-
nology of China approved a project for the development of an air
curtain system and a capillary radiation cooling system. Research
on these systems has generated promising results [12,14,19,20].
Both systems (air curtain and capillary radiation) have specific fea-
tures that can create a steady preservation environment for the
relics, allowing independent control and local preservation of arti-
fact environments. However, both systems have limitations: The
capillary radiation system belongs to an all-water AC system. The
water pipeline for the chilled water is directly fixed in the preserva-
tion area, which causes the risks of water leakage within exhibition
areas. Furthermore, the RH of the preservation area is typically very
high, so that moisture easily condensates on the surface of water
pipe and flows into the fragile earthen ruins. Moreover, the capillary
radiant system may encourage fungal growth due to the beneficial
high moisture, lack of air purification, and poor ventilation. The air
curtain system belongs to an all-air AC system. The high air veloc-
ity at the jet outlet may cause vibration and noise. The symmetric
arrangement of the jet outlet and return air grille of the air curtain
system cause difficulties in the popularization of funerary pits with
irregular geometrical structure.
By fixing both the air supply outlet and return air inlet at
the bottom and top of the funerary pit, this study proposes an
energy efficient local ventilation (LV) system to control the specific
preservation environment for semi-exposed relics. This ventilation
system is essentially a type of displacement ventilation without
internal heat sources. Displacement ventilation systems are known
to effectively control visitor-occupied environments in museums at
a relatively low energy cost, and are therefore widely used in pub-
lic buildings with large space layouts. DV systems can also improve
the indoor air quality in occupied areas by separating contaminated
air from clean air via stratification. The systems effectively achieve
energy savings as well as high indoor air quality [21–23].
This research evaluated LV system performance when control-
ling the local preservation area in an experimental exhibition hall
with a large space layout. The study specifically explored both the
environmental stability and energy consumption of the LV system
by evaluating the effects of the ventilation flow rate.
2. Preliminary energy consumption analysis and system
description
China has many terracotta pottery relics; most of these have
been unearthed and preserved in funerary pits. The Emperor Qin’s
Mausoleum Site Museum displays these types of relics, which are
located in pits that form a reservoir at the ground of the exhibition
559
hall (see Fig. 1a). This site museum generally faces the problem of
either having no AC system, or having an AC system that wastes
too much energy for cooling and heating the entire large space (see
Fig. 1b). Efficient AC systems, such as stratified air conditioning and
local heating systems are therefore attracting increased attention.
The total energy usage of a building’s AC system (Q) can be
estimated with Eq. (1).
Q = qA A (1)
In this expression, qA represents the cooling load per unit floor
area (Q-Index) andA represents the total air-conditioned area of the
building. A site museum’s exhibition hall generally has a high ceil-
ing height (≥ 10 m), a large ratio of external building envelope area
to wall area, large interior volumes, and complicated functions. A
traditional central AC system that cools the entire space (see Fig. 1b)
will be energy-hungry. The predicted cooling energy consumption
and cooling load index of the No. 1 exhibition hall of the Emperor
Qin’s Mausoleum Site Museum would amount to 2860.66 kW and
179 W/m2 [24,25]. In addition, an exhibition hall typically has a very
large floor area (A), where only a small proportion is occupied by
collections (see Fig. 1a). For example, at the Emperor’s Mausoleum
Site Museum, the area occupied by relics occupies only 34% of the
total indoor space of the No. 1 pit exhibition hall [26]. This indicates
that most of the energy load is likely wasted in areas that are not
occupied by relics.
Eq. (1) highlights two ways to reduce the energy consumption:
reducing the cooling load per unit area (qA ), or decreasing the air-
conditioned area (A). The local environmental strategy achieves
both; it reduces the height of the air-conditioned space to decrease
qA and provides local environmental control within the occupied
area to decrease A. In general, it is difficult to implement a local
heating in the funerary pit for this tall exhibition hall since hot
air will rise up and move out of the funerary pit [12]. However,
local cooling is particularly viable for the funerary pit because the
soil temperature of the underlying funerary pit is typically lower
(cooler) than the air temperature of the space above, once cool air
is supplied to the pit, a steady thermal layer is created, which main-
tains a stable preservation environment (see Fig. 1c). This type of
local cooling system is widely used in commercial buildings with
large space layouts, including airport terminals, large conference
halls, and large open plan offices. This further supports the suc-
cessful implementation of independent controls for both the relic
environment and the visitor environment [14].
3. Climatic recommendations for semi-exposed terracotta
relics
The UNI 10829:1999 standard [27] indicates that terracotta
relics are insensitive to ambient temperature and RH. Therefore,
temperature and RH values are not provided as optimal conserva-
tion parameters, thus implying that environmental controls may
not be necessary for terracotta relics in site museums. However,
several studies have analyzed the primitive environments of relics
in site museums, and recommended that environmental conditions
should be reconsidered in a more differentiated and qualified way
because these relics are semi-exposed complex (see Fig. 2) [20].
Environmental control systems are therefore required to reduce
differences in temperature and RH between the surrounding soil
and air, since a lack of equilibrium would cause fluctuations in envi-
ronmental parameters and thus allow heat and moisture to transfer
from the soil into the air [20,28].
In general, the soil environment underneath the relics is more
stable than the air environment. As such, the natural soil parame-
ters should be considered as the target values for relic conservation.
The soil temperature at depths of less than 10 m below the ground
560 X. Luo et al. / Energy and Buildings 159 (2018) 558–571

Fig. 1. Schematic diagrams of environmental control systems for site museums with funerary pits. (a) The No. 1 exhibition hall of Emperor Qin’s Mausoleum Site Museum
(courtesy of the Emperor Qin’s Mausoleum Site Museum). (b) Environmental control provided by a traditional central AC system. (c) The local environmental control provided
in a preservation area using a LV system.

Fig. 2. Schematic diagram of semi-exposed relics in a site museum. (a) Unearthed semi-exposed terracotta relics of the Hanyangling Museum. (b) The soil-air coupled
preservation environment.
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571
Fig. 3. The LV system experimental device. (a) System flow chart. (b) Environmental controls for the pit. (c) Interior view of the experimental facility.
surface in Xi’an (the museum and study site) has been reported to
be approximately 20 ◦ C [29]; therefore, the control temperature of
LV system was set at 20–21 ◦ C. It is more difficult to identify a spe-
cific target RH range; however, to mitigate the desiccation cracking
of relics, the preservation environment should be as moist as pos-
sible. This is required because the RH of soil pore air is almost 100%,
with an RH balance point of 99.9% [28].
However, spore germination and mycelium growth both benefit
from humidity; a highly humid environment may damage in-situ
relics due to mold growth. The ambient RH requirement for mold
fungus development has been estimated to range between 80 and
95% [30]. With increasing humidity, the probability that mold fun-
gus will form increases as well. The optimal RH for mold fungus
development ranges between 90 and 95%. In addition, once the RH
exceeds 90%, temperature fluctuations may cause water conden-
sation on all semi-exposed relic surfaces. Therefore, the LV system
humidifier was not switched on during this study and the average
RH of the funerary pit was sustained between 80 and 90% approx-
imately.
561
4. Experimental setup and procedure
4.1. Experimental LV system
All conservation measures have to undergo extensive and
detailed investigation, testing, and evaluation; only those demon-
strated to be effective and safe may then be used [31]. Therefore, an
experimental building was built in the campus of the Xi’an Jiaotong
University to simulate a site museum exhibition hall. The distance
between the experimental building and the Emperor Qin’s Terra-
Cotta Warriors and Horses Museum is only 31.4 km. The area and
height of the experimental building are 100 m2 and 5 m, respec-
tively. A cuboid-shaped pit with a length, width, and depth of 4 m,
2.8 m, and 2.0 m, respectively, was excavated in the experimental
hall. Three soil columns with heights of 0.5 m were placed in the
funerary pit to investigate the desiccation cracking process of the
earthen site (see Fig. 3). In general, the desiccation cracking process
takes place continuously and may take several years to complete.
The floors of the indoor gallery of the experimental building for
visitors are paved with floor-bricks; however, in the funerary pit
562 X. Luo et al. / Energy and Buildings 159 (2018) 558–571
Fig. 4. Layout of the monitoring points in the experimental hall.
the soil ground is directly exposed to the air. The width and depth
of the funerary pits, and soil type, and the climatic conditions of
the experimental site are similar to those at the Emperor Qin’s
mausoleum Site Museum. A LV system was designed and estab-
lished in the experimental funerary pit (see Fig. 3). The air was
supplied at the ground level of the pit into a plenum chamber with
a length and depth of 2.5 m and 0.7 m, respectively. One façade of
the chamber served as the air diffuser, which was composed of a
perforated panel with 2.0 mm diameter circular holes. The perfo-
rated plate serves as a type of air supply outlet that is widely used in
clean air-conditioning systems, since it can effectively decrease the
air velocity and temperature gradients in the air-conditioned area
[32]. The returned air was purified and cooled in the air-handling
unit and then directed into the plenum chamber using a centrifu-
gal fan. The air pressure was equalized in the plenum chamber
to ensure a uniform air supply through the perforated panels. A
5100 W air-cooled water chiller was used as the cold source of the
air-handling unit[33].Although there is a significant length differ-
ence between the experimental funerary pit and the real funerary
pits of the Emperor Qin’s mausoleum Site Museum, it will not affect
the performance of the ventilation system since the supply outlet
and return air inlet of the proposed AC system were installed at
both sides of the width direction and the critical dimensions that
affect the performance of the DV system are the width and depth
of the funerary pit.
4.2. Layout of the measuring points
A series of LV system tests were performed in the experimental
building to monitor environmental parameters, understand local
environmental controls in the funerary pit, and evaluate LV system
performance.
Parameters monitored during the study (see Fig. 4) included:
• The temperature and RH of the outdoor environment were
automatically measured every 10 min by placing the TR-72Ui
thermos-recorder in a thermometer shelter. The accuracy of the
TR-72Ui was ± 0.3 ◦ C for temperature and ± 5% for RH.
• Vertical temperature and RH profiles in the integrated pit envi-
ronment were measured using the same model meter at five
observation points (labeled T1 to T5) along a central line in the
pit. The distances from the ground level of the pit upward to
points T1 through T5 were 0.2 m, 0.5 m, 1.3 m, 1.9 m, and 2.6 m,
respectively. Of these points, T4 was the dividing point between
the preservation zone and the visitor zone. T5 was in the visitor
area, at a height of 0.6 m above the pit. To evaluate the effect of
the LV system on the visitor passageway away from the funerary
pit, temperature was also monitored at a sixth observation point
(T6). T6 was located at the same height as T5 (2.6 m) and was 2 m
away from the funerary pit. Data from all six points (T1-T6) were
automatically collected and recorded every 2 min.
• The soil temperature and the temperature of the relic body were
monitored using thermocouples at three measuring points: Ts1,
Tr1, and Tr2. Ts1 was placed 0.05 m below the base of the pit.
Tr1 and Tr2 were inserted into the ceramic pottery block; both
monitoring points were 0.25 m from the air-soil interface. Mea-
surements were collected every 2 min; thermocouple accuracy
was ± 0.004|T|◦ C.
• The air velocity in the center of the funerary pit (TaV) was
measured every 2 min at a height of 0.35 m using a Swema 03
micro-velocity sensor, with an accuracy of ± 0.03 m/s. To calcu-
late the cooling load of the LV system, the volume flow (Lw) and
the temperature of the supply and return water (Tws and Twr)
flowing through the air-handling unit were also measured. The
water pump for the chilled water was working at a fixed con-
dition. The volumetric flow of the chilled water was measured
via turbine flow meter and recorded manually. The duration
and sampling intervals of the acquisition of volumetric flow
were 15 min and 30 s. The accuracy of the turbine flow meter
was ± 1%|L| L/h; the temperatures of the supply and return water
were monitored via T-type thermocouples recorded with an
ADAM 4018. The sampling intervals were 2 min. The accuracy
of the thermocouple was ± 0.004|T|◦ C.
• A temperature controller was placed in the pit at a height of 0.2 m
(Tc) to manage the automatic start and stop of the water pump.
The control temperature was set to 20–21 ◦ C. When the measured
temperature at Tc exceeded 21 ◦ C, the water pump switched on,
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571 563

Table 1
Summary of the test cases.

Case Operational model Date f (Hz) La (m3 /h) ACH (ac/h)

1 The LV system switched off July 17–19, 2015 – – –

2 The LV system switched on August 1–3, 2015 15 181.6 8.1
3 The LV system switched on July, 23–25, 2015 20 267.1 11.9
4 The LV system switched on July 29–31, 2015 30 370.4 16.5
5 Both the LV and comfort AC systems switched on July 16–17, 2016 20 267.1 11.9

providing cold water to the air handling unit at a temperature Table 2

Summary of the outdoor temperature (◦ C) and RH (%).
below 20 ◦ C.
• The air velocities in the funerary pit were monitored along the Case Duration Temperature RH
horizontal direction using a Swema 03 micro-velocity sensor at
Max Min Avg Max Min Avg
the five measuring points of Tv1 to Tv5. The distances of the
1 July, 17–19, 2015 35.1 21.7 28.1 83 31 54
five monitoring points from the air supply outlet were 0.05 m,
2 August, 1–3, 2015 42.0 21.3 31.2 100 26 59
0.2 m, 0.4 m 0.8 m, and 1.4 m, respectively. The air velocities were 3 July, 23–25, 2015 39.2 22.5 30.1 64 13 38
manually recorded with duration and sampling interval of each 4 July, 29–31, 2015 41.5 27.5 34.3 59 23 38
monitoring point of 5 min and 5 s. 5 July, 16–17, 2016 35.3 23.3 29.5 99 32 54

4.3. Test cases

Five tests were conducted during the summer months of 2015
and 2016 (see Table 1). The LV system was initially turned off in case
1 to measure the results of the natural ventilation condition of the
exhibition hall. Then, three experiments were conducted where the
LV system for the preservation area was only switched on in case
2 to case 4. Different ventilation rates were established in these
three cases to determine a suitable fan frequency for the LV system.
The air conditioning systems in newly built archaeology museums
are typically installed to satisfy the thermal requirements of visi-
tors. A coupling operational mode of the LV system with a comfort
AC system was performed in case 5 to validate the simultaneous
environmental control for both visitors and relics area.
• In cases 2–4, the temperature of the chilled water (see Tws in
Fig. 4) was set to 10–12 ◦ C. As soon as the water temperature
exceeded 12 ◦ C, the refrigeration unit started until the water tem-
perature had dropped below 10 ◦ C. Within this temperature range
of the chilled water, the average temperature of the supply air
varied between 19.6 ◦ C and 20.1 ◦ C.
• In case 5, the traditional comfort AC system for the visitors’ envi-
ronment was operated from 9 a.m. to 6 p.m., while the LV system
for the preservation area worked 24 h per day.
• The fan frequency (f), air volume flow (La), and air change rate
are also listed in Table 1; the air changes per hour (ACH) were
calculated as the supply air volume divided by the pit volume.
5. Results and discussion
Tables 1 summarizes the experimental conditions used to vali-
date the ability of the LV system to support the independent control
of the preservation environment. The equilibrium between the air-
soil interface and the LV system energy consumption was also
analyzed.
5.1. Outdoor environment
Summer is relatively hot in Xi’an, with significant tempera-
ture swings between day and night. Table 2 summarizes both the
outdoor temperatures and RH during the test case periods. The out-
door temperature varied between 21.3 ◦ C and 42 ◦ C, with a mean
value of 30.9 ◦ C during the test period. These high outdoor tem-
perature and diurnal temperature differences have been known to
accelerate the chemical process of relic deterioration [33], and sig-
nificantly impact the energy consumption required to control the
indoor environment of the site museum. Consequently, a suitable
AC system is required to create a stable conservation environment
for the semi-exposed relics.
5.2. Indoor preservation environment
Figs. 5 and 6 present the indoor temperatures and RH distri-
butions, respectively, across the test cases of 1–4. Tables 3 and 4
present the mean temperature and RH values (T 1-T 6and RH1-
RH6) as well as daily ranges (T1-T4 and RH1-RH4 for the
funerary pit), respectively. During test case 1, when the LV sys-
tem was turned off, the preservation environment of the pit was
unstable. The diurnal temperature fluctuation at observation points
T1-T4 (in the conservation area) were 2.5 ◦ C, 3.9 ◦ C, 5.7 ◦ C, and
8.0 ◦ C, respectively. These fluctuations were larger than the recom-
mended range (≤ 1.5 ◦ C) for the conservation of cultural relics in
museums [34]. Outdoor climatic conditions significantly affected
the preservation environment; large fluctuations of climatic fac-
tors within the funerary pit corresponded with periodic variations
in the outdoor environment. Higher outdoor temperatures further
aggravated high temperatures and fluctuations in the funerary pit.
In test cases 2–4, the LV system created a suitable conservation
environment in the pit. Fig. 5 shows that the LV system significantly
reduced temperature fluctuations in the funerary pit. Even in test
case 2, when only a low level of air volume was supplied to the
funerary pit, the preservation environment significantly improved,
when a stable and uniformly cool air reservoir had formed in the
conservation area. The temperature fluctuation at T4 declined by
approximately 40% during test case 2 compared to test case 1.
Increasing the volume flow of the supplied air in test case 3 and
4 minimized temperature fluctuations in the pit. The temperature
distribution in the funerary pit became more uniform during these
test cases; the vertical temperature difference between the top and
bottom of the pit (T 4 − T 1) was reduced from 3.0 ◦ C in case 2 to
approximately 2 ◦ C in cases 3 and 4. When the volume air increased
by approximately 39.0% from 267.1 m3 /h in case 3–370.4 m3 /h in
case 4, there was almost no variation in the mean temperature at
T1-T4. This indicates that case 3 was sufficient to provide a stable
preservation environment for the funerary pit.
The daily changes in RH aligned with the temperature changes
during the test cases. The RH at the base of the preservation area
(T1) was high during test case 1, at approximately 100%. This almost
saturated RH near the air-soil interface efficiently reduced the
moisture transport from the soil into the air; however, such a high
level caused moisture condensation on relic surfaces and acceler-
ated bacterial and fungal growth in the funerary pit. This would
564 X. Luo et al. / Energy and Buildings 159 (2018) 558–571

Fig. 5. Variations in air temperature. (a) The LV system switched off. (b) The LV system switched on at f = 15 Hz. (c) The LV system switched on at f = 20 Hz. (d) The LV system
switched on at f = 30 Hz.

Table 3
Summary of the indoor air temperature (◦ C).

Case Preservation area Visitor’s area

T1 T2 T3 T4 T1−4 T1 T2 T3 T4 T5 T6

1 23.0 24.0 25.1 26.2 24.6 2.5 3.9 5.7 8.0 27.7 28.0
2 20.7 21.4 22.6 23.7 22.1 2.0 2.9 3.4 4.6 29.8 30.5
3 20.5 21.1 21.9 22.8 21.6 1.3 2.0 2.6 4.0 28.3 28.9
4 20.5 20.9 21.7 22.5 21.5 1.2 1.8 2.4 3.2 30.6 31.2
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571 565

Fig. 6. Variation of air RH. (a) The LV system switched off. (b) The LV system switched on at f = 15 Hz. (c) The LV system switched on at f = 20 Hz. (d) The LV system switched
on at f = 30 Hz.

Table 4
Summary of the indoor air RH (◦ C).

Cases Preservation area Visitor’s area

RH1 RH2 RH3 RH4 RH1−4 RH1 RH2 RH3 RH4 RH5 RH6

1 98 95 82 71 87 3 11 19 30 59 60
2 82 76 69 70 74 23 28 29 36 60 59
3 88 82 76 77 81 27 36 36 41 56 56
4 86 81 72 73 78 23 24 26 31 50 51
566 X. Luo et al. / Energy and Buildings 159 (2018) 558–571

Fig. 7. Air velocity variations in the pit. (a) The LV system switched off. (b) The LV system switched on at f = 15 Hz. (c) The LV system switched on at f = 20 Hz. (d) The LV
system switched on at f = 30 Hz.

Table 5
Summary of the air velocity.

Cases 1 2 3 4

UMax 0.013 0.106 0.182 0.280

UAVG 0.004 0022 0.026 0.070

result in relic deterioration, especially in the absence of air ventila-

tion. In test cases 2–4, the funerary pit experienced high humidity,
but at a RH lower than 90%, which could prevent moisture conden-
sation on relic surfaces.
In all test cases (with and without operation of the LV system),
the temperatures and RH at T5 and T6 were almost approximately
equal for all cases. This means that the LV system created a cool
air lake in the pit without influencing the visitor area. Although
the environment of the funerary pits was mainly governed by the
LV system, the outdoor climatic conditions still posed an impor-
Fig. 8. Attenuation curves of the air supply velocity along the along the horizontal
tant influence on the operation of the system since the visitors’
direction of the funerary pit.
area was naturally ventilated and could be regarded as an upper
boundary condition of the funerary pit. The return air inlet will
entrain air from the visitors’ area, which will lead to an increment significantly larger than a traditional grill opening. Significant fluc-
in environmental fluctuation and energy consumption. tuations were observed in the velocity profiles when the LV system
Air motion in the preservation area typically attracts little atten- was switched on; however, average velocities were lower than the
tion; however, air velocity significantly impacts the desiccation recommended range (≤0.15 m/s) for museum collections [37].
cracking of earthen sites, since the soil’s drying rate is proportional The air supply outlet of the LV system was fixed in the funer-
to the air velocity of the preservation environment [35,36]. Fig. 7 ary pit. The flow features of the regions adjacent to the outlet were
and Table 5 provide the air velocities at the monitoring point TaV critical for the evaluation of LV system performance. The attenua-
(see Fig. 4) in different test conditions. The air velocities varied little, tion curves of the air supply velocity along the horizontal direction
regardless of whether the LV system was operating or not; the mean in the funerary pit are shown in Fig. 8. The velocities attenuated
values of case 1–3 were even below the manufacturer’s specified quickly and reached the recommended range (≤0.15 m/s) at dis-
low value (0.040 m/s) of the air velocity sensor. The preservation tances of less than 0.1 m, 0.2 m, and 0.4 m for case 2, case 3, and case
area may have had negligible air velocities since the effective aer- 4, respectively. Considering that case 3 was sufficient to provide a
ated area provided by the perforated panel of the LV system was stable preservation environment for the funerary pit, a distance of
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571 567

Fig. 9. The soil-air preservation environments of semi-exposed relics. (a) Unearthed terra-cotta relics in the Hanyangling Museum, China. (b) Unearthed ivory artifacts in
the Jinsha Site Museum, China.

Fig. 10. Temperature profiles of the soil-air environment. (a) The LV system switched off. (b) The LV system switched on at f = 15 Hz. (c) The LV system switched on at f = 20 Hz.
(d) The LV system switched on at f = 30 Hz.

0.2 m between the relics and the air supply outlet could ensure air Fig. 10 shows the soil temperatures (Ts) and air temperatures
velocities below 0.15 m/s. (T1) detected during the experimental tests. The mean temperature
differences (T = |T1-Ts|) for test cases 1–4 were 1.14 ◦ C, 0.47 ◦ C,
0.33 ◦ C, and 0.27 ◦ C, respectively (see Fig. 11). The outdoor climatic
5.3. Balance of the soil-air coupled preservation environment
conditions during test case 1 were relatively cool, with an average
outdoor temperature of 28.1 ◦ C (see Table 3); a significant temper-
Fig. 9 shows in-situ cultural relics in site museums that remain
ature difference was observed at the air-soil interface (Fig. 10a).
partly buried in the soil environment. In such a scenario, semi-
Additionally, both soil and air temperatures became unstable, fol-
exposed relics are surrounded by both soil and air. Moisture
lowing an upward trend. When the LV system was switched on
transport in the soil-air environment is closely correlated with tem-
for test cases 2–4, the temperature differences were significantly
perature differences at the soil-air interface and this difference is
lower than for test case 1. Soil and air temperatures at the inter-
a primary driver of desiccation cracking and salt damage to semi-
face almost overlapped during test cases 3 and 4, suggesting that
exposed relics [20,28]. A uniform temperature distribution in the
the in-situ relics were preserved in a stable and balanced soil-air
air-soil preservation environment is required to delay relic deteri-
environment.
oration.
568 X. Luo et al. / Energy and Buildings 159 (2018) 558–571

Fig. 11. The average temperature difference across the soil-air interface in cases Fig. 13. The average temperature difference between parts of the semi-exposed
1–4. relics in cases 1–4.

Fig. 12 shows the time-series for temperatures at Tr1 and Tr2

across all four test cases. The differences in mean temperature
(T = |Tr2-Tr1|) along the terracotta pottery block placed in the
test pit in cases 1–4 were 2.09 ◦ C, 0.80 ◦ C, 0.48 ◦ C, and 0.43 ◦ C,
respectively (see Fig. 13). The daily temperature variations were
analogous to the variations at the soil-air interface. The tempera-
ture profiles were lower, and the temperature differences along the
ceramic pottery block were also lower during LV system operation.
The LV system provided a balanced conservation environment for
semi-exposed relics, as the lower temperature differences across
the air-soil environment and relic bodies mitigated heat transport
and moisture transfer.
The temperature differences were very small and were almost
identical in test cases 3 and 4. As noted above, this confirmed that
case 3 would provide a uniform and stable local environment for
the funerary pit.

Fig. 12. Temperature profiles along the relic body. (a) The LV system switched off. (b) The LV system switched on at f = 15 Hz. (c) The LV system switched on at f = 20 Hz. (d)
The LV system switched on at f = 30 Hz.
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571
Table 6
Summary of energy consumption.
Cases Case 2 Case 3
Tw (◦ C) 1.1 1.2
Lw (m3 /h) 1.0 1.0
Q (W) 1279.9 1396.2
Q-Index (W/m2 ) 114.3 124.7
Table 7
Summary of the coupled experimental case.
AC system Air conditioned zones Operational time
Comfort AC system Visitors’ area Switched on from 9 a.m. to 6 p.m.
LV system Funerary pit Remained on 24 h per day
5.4. Energy consumption
Energy consumption is an essential variable when selecting an
HVAC system for a museum building. The energy consumption of
an environmental control system (Q) can be determined via Eq. (2):
Q = Lw w C Pw Tw = Lw w C Pw (Tws − Twr )
In this expression, Lw , CPw , w , and w represent the volume flow
rate, specific heat, and water density, respectively. The variables Tws
and Twr represent the supply and return water temperatures in the
air handing unit, respectively (Table 6). The cooling capacity was
then transferred from the chilled water to the supplied air in the air
handling unit. A 120 W supply fan was used to convey recirculat-
ing air. During the tests, the volumetric flow remained constant at
Lw = 1.0m3 /h. Table 7 summarizes the energy consumption of the
LV system when Tws , Twr , and Lw values and the physical parame-
ters of water are substituted into Eq. (2). The differences of energy
consumption in case 2–4 were mainly caused by the increasing
recirculating air volume of the return air inlet and enhanced heat
transfer of surface-type heat exchanger in the air-handling unit. As
discussed above, test case 3 achieved to provide a suitable envi-
ronment for the pit area and the LV system energy used in this
case was 1396.2 W. The floor area of the preservation zone was
11.2 m2 and as such, the cooling load per unit floor area (Q-index)
was 124.7 W/m2 . This cooling load per unit floor area is signifi-
cantly lower than the predicted energy consumption required to
operate traditional central AC systems such as those in use at the
No. 1 exhibition hall of the Emperor Qin’s Mausoleum Site Museum
[24,25]. The LV system also saves significant energy because the air-
conditioned area is quite small compared to the large area covered
by a traditional central AC system.
5.5. Coupling operational mode of the LV system with comfort AC
system
For many newly built exhibition halls of site museums (such as
the K0006 and K9901 funerary pits of the of the Emperor Qin’s Mau-
soleum Site Museum), traditional central AC systems have been
installed for the visitors’ area. These existing AC systems operate
from 9 a.m. to 6 p.m. to provide a comfortable thermal environment
(24 ◦ C ≤ T ≤ 28 ◦ C, 40 ≤ RH ≤ 70) [38] for the visitors during open-
ing time. However, the specialized needs of the unearthed relics
are typically ignored in these site museums due to financial and
technological reasons.
To evaluate the performance of the LV system in the case where
a traditional comfort AC system has been equipped to the visitors’
area, an experimental test that separately employs an AC system
for the visitors’ area and a LV system for the funerary pit was per-
formed in July 16–17, 2016 (see Fig. 14). The operating parameters
of both systems are listed in Table 7. The arrangement of both the
temperature and RH sensors are identical to that of case 1 to case 4.
569
Case 4
1.78
1.0
2094.3
187.0
(2)
Fig. 14. The operating model of the independent environmental control systems.
The air volume flow and working temperature of the LV were iden-
tical to those of case 3. The working temperature of the comfort AC
system was set to 26 ◦ C.
Fig. 15 shows the temperature and RH distribution in the funer-
ary pit (T1 to T4) and visitor area (T5) during the test. The statistics
for the mean temperature and RH are listed in Table 8. The mean
temperature and RH at the preservation area (T1 to T4) were almost
identical to those of case 3: a steady preservation environment was
created in the funerary pit throughout all days. Furthermore, dur-
ing the opening hours of the site museum (from 9 a.m. to 6 p.m.),
the temperature and RH at the visitors’ area (T5) varied between
25.4 ◦ C and 27.5 ◦ C and 46% and 63%, respectively, which remains
entirely within the comfort level of humans[38]. The requirements
for both visitors and relics are simultaneously satisfied, indicating
that such a LV system for the relic preservation area could be oper-
ated simultaneously with and independently from already existing
comfort AC systems.
6. Conclusion
Designing environmental controls to preserve site museum
relics remains a technical challenge for environmental scientists
and technologists; all currently feasible solutions consume high
amounts of energy. This study involved testing and validating a LV
system, which resulted in a tentative protocol that allows for local
environmental control of semi-exposed relics in site museums. This
is an efficient energy control strategy, because the conservation
area is separately regulated from the large exhibition hall.
LV system performance was studied by investigating the effects
of different ventilation flow rates. Experimental tests assessed
climatic conditions, environmental parameters in the pit, tem-
perature differences across the soil-air environment, and energy
consumption. The results demonstrate that the LV system could
efficiently create a cool reservoir in the funerary pit and maintain
an adequate environment for local preservation, independent of
the visitor environment. LV system energy consumption was lower
than the consumption of a traditional central AC system used in site
museum, because the area being cooled and the cooling load per
unit floor area were significantly reduced.
Further research will consider the relationship between the
height of air supply outlets and the temperature gradient along
the vertical direction of the funerary pit. This will inform different
570 X. Luo et al. / Energy and Buildings 159 (2018) 558–571

Fig. 15. Variation of the air temperature and RH for the coupled operated test.

Table 8
Statistics of indoor air temperature (◦ C) and RH (%) of case 5.

Zone Preservation area Visitor’s area

T1 T2 T3 T4 T1−4 T1 T2 T3 T4 T5 T5

T 20.4 21.3 21.7 22.5 21.5 1.5 2.2 2.8 3.3 26.1 3.4
RH 85 81 79 78 81 23 28 31 24 58 21

approaches for preserving relics at different heights. In addition,
future studies should consider coupled operation of a LV system
and another HVAC system, as such an approach may accommodate
the different thermal environments needed by relics and visitors.
Acknowledgements
This study was supported by the Key Scientific Research
Innovation Team Project of Shaanxi Province (under grant No.
2016KCT-14), the China Postdoctoral Science Foundation (under
grant No. 2016T90923), and the Fundamental Research Funds for
the Central Universities (under grant No. zrzd2017003).
References
[1] M. Zhao, Z.L. Gu, W.B. Kang, X. Liu, L.Y. Zhang, L.W. Jin, Q.L. Zhang,
Experimental investigation and feasibility analysis on a capillary radiant
heating system based on solar and air source heat pump dual heat source,
Appl. Energy 185 (2016) 2094–2105.
[2] R. Barzin, et al., Application of PCM underfloor heating in combination with
PCM wallboards for space heating using price based control system, Appl.
Energy 148 (2015) 39–48.
[3] G.M. Huebner, et al., Explaining domestic energy consumption–the
comparative contribution of building factors, socio-demographics, behaviours
and attitudes, Appl. Energy 159 (2015) 589–600.
[4] Z.J. Liu, W. Xu, X. Zhai, Q. Cheng, X. Chen, Feasibility and performance study of
the hybrid ground-source heat pump system for one office building in
Chinese heating dominated areas, Renew. Energy 101 (2017) 1131–1140.
[5] G.W. Liu, Z.Z. Wu, M.M. Hu, Energy consumption and management in public
buildings in China: an investigation of Chongqing, Energy Procedia 14 (2012)
1925–1930.
[6] J.N. Huang, H.L. Lv, T. Gao, W. Feng, Y.X. Chen, T. Zhou, Thermal properties
optimization of envelope in energy-saving renovation of existing public
buildings, Build. Environ. 75 (2014) 504–510.
[7] X.T. Li, C.W. Yu, China’s building energy efficiency targets: challenges or
opportunities, Indoor Built Environ. 21 (2012) 609–613.
[8] Xinhuanet.com.,
http://news.xinhuanet.com/politics/2015-05/18/c 1115323223. htm
(Accessed: 3rd October 2017).
[9] Z.L. Gu, X.L. Luo, X.Z. Meng, Z.S. Wang, T. Ma, C. Yu, B. Rong, W.W. Li, Y. Tan,
Primitive environment control for preservation of pit relics in archaeology
museum of China, Environ. Sci. Technol. 47 (2013) 1504–1509.
[10] B. Bøhm, M. Ryhl-svendsen, Analysis of the thermal conditions in an unheated
museum store in a temperate climate. On the thermal interaction of earth and
store, Energ Build. 43 (2011) 3337–3342.
[11] M.L. Gennusa, G. Lascari, G. Rizzo, G. Scaccianoce, Conflicting needs of the
thermal indoor environment of museums: in search of a practical
compromise, J. Cult. Herit. 9 (2008) 125–134.
[12] X.L. Luo, Z.L. Gu, C.W. Yu, K. Li, B. Xiao, Preservation of in situ artefacts by local
heating in earthen pit in archaeology museum in cold winter, Build. Environ.
99 (2016) 29–43.
[13] D. Camuffo, E. Pagan, S. Rissanen, L. Bratasz, R. Kozlowski, M. Camuffo, A.D.
Valle, An advanced church heating system favourable to art works: a
contribution to European standardization, J. Cult. Herit. 11 (2010) 205–219.
[14] X.L. Luo, Z.L. Gu, Z.S. Wang, W. Tian, K. Li, An independent and simultaneous
operational mode of air conditioning systems for visitors and relics in
archaeology museum, Appl. Therm. Eng. 100 (2016) 911–924.
[15] H.E. Silva, F.M.A. Henriques, T.A.S. Henriques, G. Coelho, A sequential process
to assess and optimize the indoor climate in museums, Build. Environ. 104
(2016) 21–34.
[16] F. Sciurpi, C. Carletti, G. Cella, L. Pirrangioli, Environmental monitoring and
microclimatic control strategies in La specola museum of Florence, Energy
Build. 95 (2015) 190–201.
[17] W.T. Jia, The Physical and Chemical Relationship Between Indoor Air Quality
and Atmospheric Corrosion in Site Museums. Master Thesis, The University of
Chinese Academy of Sciences, 2015.
[18] J. Ferdyn-Grygierek, Indoor environment quality in the museum building and
its effect on heating and cooling demand, Energ Build. 85 (2014) 32–44.
 X. Luo et al. / Energy and Buildings 159 (2018) 558–571
[19] X.L. Luo, Z.L. Gu, C. Yu, T. Ma, K. Kase, Efficacy of an air curtain system for local
pit environmental control for relic preservation in Archaeology Museums,
Indoor Built Environ. 25 (2016) 29–40.
[20] X.L. Luo, Z.L. Gu, T.Y. Li, X.Z. Meng, T. Ma, C. Yu, Environmental control
strategies for the in situ preservation of unearthed relics in archaeology
museums, J. Cult. Herit. 16 (2015) 790–797.
[21] Y. Wang, F.Y. Zhao, J. Kuckelkorn, D. Liu, J. Liu, J.L. Zhang, Classroom energy
efficiency and air environment with displacement natural ventilation in a
passive public school building, Energ Build. 70 (2014) 258–270.
[22] T. Gil-Lopez, M.A. Galvez-Huerta, P.G. O’Donohoe, J. Castejon-Navas, P.M.
Dieguez-Elizondo, Analysis of the influence of the return position in the
vertical temperature gradient in displacement ventilation systems for large
halls, Energ Build. 140 (2017) 371–379.
[23] A. Makhoul, K. Ghali, N. Ghaddar, The Energy saving potential and the
associated thermal comfort of displacement ventilation systems assisted by
personalised ventilation, Indoor Built Environ. 22 (2013) 508–519.
[24] C. Cao, X. Meng, X. Liu, X.H. Yang, W.Y. Hu, L.W. Jin, Energy analysis of relics
museum buildings, Energy Procedia 75 (2015) 1809–1818.
[25] C. Cao, Energy Consumption Analysis and Energy Conservation Strategies
Research of a Typical Relics Museum, Thesis, Master Degree, Xi’an Jiaotong
University, 2015.
[26] The No.1 pit exhibition hall of Emperor Qinshihuang’s Mausoleum Site
Museum, http://www.qinling360.com/bingmayong/yihaokeng/ (Accessed:
3rd October 2017).
[27] UNI 10829, Works of Art of Historical Importance. Ambient Conditions for the
Conservation, Measurement and Analysis, UNI-Ente Italiano di Unificazione,
Milano, 1999 (In Italian).
571
[28] X.L. Luo, Z.L. Gu, C. Yu, Desiccation cracking of earthen sites in archaeology
museum-a viewpoint of chemical potential difference of water content,
Indoor Built Environ. 24 (2015) 147–152.
[29] X. Zhang, The Theory on Heat Transfer of Vertical Ground Heat Exchanger and
the Test Method for Parameter of the Rock–Soil Thermal Properties Thesis,
Master Degree, Chang’ an University, 2012.
[30] H. Viitanen, J. Vinha, K. Salminen, T. Ojanen, R. Peuhkuri, Moisture and
bio-deterioration risk of building materials and structures, J. Build. Phys. 33
(2010) 201–224.
[31] Chinese National Committee for the International Council on Monuments and
Sites. Principles for Conservation of Heritage Site in China, Cultural Relics
Press, Beijing, 2015.
[32] S.C. Hu, Y.K. Chuah, M.C. Yen, Design and evaluation of a minienvironment for
semiconductor manufacture process, Build. Environ. 37 (2016) 201–208.
[33] G. Pavlogeoratos, Environmental parameters in museums, Build. Environ. 38
(2003) 1457–1462.
[34] G. Thomson, The Museum Environment, second ed., Butterworld-Heinemann,
London, 1986.
[35] A.W. Fan, W. Liu, C.Q. Wang, Simulation on the daily change of soil
temperature under various environment conditions, Acta Energi Sin. 24
(2003) 167–171.
[36] J. Rogers, M. Kaviany, Funicular and evaporative-front regimes in convective
drying of granular beds, Int. J. Heat Mass Transf. 35 (1992) 469–480.
[37] Y.H. Yau, B.T. Chew, A.Z.A. Saifullah, A field study on thermal comfort of
occupants and acceptable neutral temperature at the National Museum in
Malaysia, Indoor built Environ. 22 (2013) 433–444.
[38] GB 50736-2012, Design Code for Heating Ventilation and Air Conditioning of
Civil Buildings, The Standardization Administration of the People’s Republic
of China (SAC), Beijing, 2012.