Contents lists available at ScienceDirect

Applied Thermal Engineering

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

Research Paper

Numerical study on the thermal performance of lightweight temporarybuilding integrated with phase change materials

Li Zhua,b, Yang Yanga, Sarula Chena, Yong Suna,b,⁎

a School of Architecture, Tianjin University, Tianjin 300072, ChinabAPEC Sustainable Energy Center, Tianjin 300072, China

H I G H L I G H T S

• Thermal behavior of lightweight buildings containing PCMs for temporary accommodations were studied.• Experimental validation was conducted for the numerical model at component level.• Influence of location, thickness and orientation of PCM were evaluated with different indicators.• Multi-orientation optimization schemes were put forwarded and further studied.

A R T I C L E I N F O

Keywords:Numerical simulationThermal performanceModular prefabricated PCM panelIndoor thermal environmentMulti-orientation optimization

A B S T R A C T

The phase change materials (PCMs) integrated in building envelope structure can decrease the buildings’ energyconsumption and improve indoor thermal comfort quality. This paper numerically studied the application effectof PCMs as a passive alternative in lightweight building with high shape coefficient for temporary accom-modations under Tianjin climate, and the impacts of some key design parameters, such as the location, thicknessand orientation, on building’s thermal behavior were explored when single orientation layout scheme wasadopted. Besides, the multi-orientation layout schemes were put forward to achieve further optimization onindoor thermal comfort. The simulation results stated that the proper application of PCMs could obviouslyimprove and promote indoor thermal comfort. In detail, the results indicated that PCM layer with a reasonablethickness, e.g. 5.0 mm in this paper, which was positioned to the interior surface was recommend when singleorientation layout was applied. When incorporating the PCM layer in at least five orientations (S5), the proposedmulti-orientation optimization schemes could ensure a comfortable indoor climate under the extreme closedcondition without extra mechanical cooling measures. Thereby, the numerical results in this paper support andhighlight the potential of using PCMs and multi-orientation optimization in lightweight temporary buildings.

1. Introduction

In order to minimize the construction weight, transportation costs aswell as the construction time, and consequently, greenhouse gas emis-sions, the lightweight buildings have received extensive attentions fromstakeholders all over the world [1–4]. Available data shows that themarket for lightweight constructions has increased from 0.5% in 2006to 4.5% in 2015 [5]. As one kind of lightweight buildings, temporarylightweight building comes into practice as an alternative to the on-sitemethod to cope with the demand from temporary accommodations[6–8], such as post-disaster sheltering, refugee camps, builder camps,mining camps, and also from some developing countries with housedelivery problems due to the lack of skills and housing quality [9]. As

one of the important temporary housing solutions, prefabricated, mass-produced and standardized lightweight houses are often provided bygovernment and NGOs after large-scale disasters in China and manyother countries [10]. Usually, they could be identified into two maingroups, including ready-made units and kit supplies [11]. The former isoften referred as the housing solution that totally constructed in factoryand then just need to be transported to the site. However, it is hard tomove this kind of lightweight building to areas with difficult access andthey often need heavy transport system. Instead of providing finishedunits, the kit supplies solution produces small modular elements thatconstitute the unit and then being assembled in the target area. In fact,the study object going to be studied in this paper belongs to kit suppliesand is assembled by modular prefabricated EPS panel covered with

https://doi.org/10.1016/j.applthermaleng.2018.03.103Received 15 December 2017; Received in revised form 25 March 2018; Accepted 29 March 2018

⁎ Corresponding author at: School of Architecture, Tianjin University, Tianjin 300072, China.E-mail address: yong.sun@tju.edu.cn (Y. Sun).

Applied Thermal Engineering 138 (2018) 35–47

Available online 30 March 20181359-4311/ © 2018 Elsevier Ltd. All rights reserved.

T

http://www.sciencedirect.com/science/journal/13594311https://www.elsevier.com/locate/apthermenghttps://doi.org/10.1016/j.applthermaleng.2018.03.103https://doi.org/10.1016/j.applthermaleng.2018.03.103mailto:yong.sun@tju.edu.cnhttps://doi.org/10.1016/j.applthermaleng.2018.03.103http://crossmark.crossref.org/dialog/?doi=10.1016/j.applthermaleng.2018.03.103&domain=pdf

color steel sheet, which has been widely deployed in China [12,13],international organizations like IKEA Foundations and UNCHR inEthiopia and Lebanon [14].

As we all known, the building envelope regulates the heat exchangebetween outdoor and indoor environment and highly affects the energydemands and comfort of the occupants [8]. When it comes to thelightweight temporary building with a high shape coefficient, the im-pacts of low thermal inertia and heat capacity will be more seriously[1,13,15,16]. It is important to underline that the basic purpose of eachtype of lightweight temporary is to offer better indoor climatic condi-tions or at least different from the external environment. When this isnot realized, it is easy that prolonged exposure of occupants to en-vironment conditions, some extreme, may cause sick building syndrome[7,8,17]. In fact, there are two effective ways to improve the thermalperformance of lightweight buildings [18–20]. One approach is to ad-dress the thermal mass of envelope structure, and thermal mass is oftendefined as the ability to store heat and maintain temperature stabili-zation. Another approach is to minimize the thermal resistance ofbuilding envelope, which expresses the ability of heat insulation. It isknown that thermal insulation materials and its applications in buildingsector have been studied for decades [21]. However, some problemsalso caused by insulation materials, such as fire safety, initial invest-ment, and space occupancy in transportation, storage and usage pro-cess, may limit their overall performance [22–24]. Thus, passive oreven active technologies are considered a necessity to cope with theindoor thermal discomfort [7]. However, the operation of active air-conditioning system in disaster areas and some worksites relies on re-liable power supply system, and this reliability is often maintained byusing diesel generator or distributed energy system [25]. In fact, eventechnology advance keeps lowering the cost of renewable energysystem, it still requires a significant initial investment. Meanwhile,though diesel generators proves to be cheap initially, maintaining asecure and consistent supply of diesel fuel proves to be an expensiveeffort and seems to be not eco-friendly.

Recently, thermal mass optimization of building envelope and theapplication of passive technology have been regarded as a good optionin lightweight buildings [12]. The application of PCM in the buildings

can alleviate contradiction of the supply and demand of building energyin time, space and intensity. In fact, direct PCM inclusion through en-capsulation in the building envelope is assumed to be a reasonablemethod [2], and this has been confirmed by both experimental andsimulation studies [26–28]. For example, Medina et al. [26] providedthe foundations for the development of PCM structural insulated panel(PCMSIP) and evaluated their thermal performance, and results showedthe average reductions in daily heat transfer across the PCMSIPsreached to 38% for 20% (concentration) PCM. Vicente et al. [27] testeda PCM wall element with macro-encapsulation, and the results showedthat combination of PCM and 10 cm XPS external insulation resulted in80% reduction of thermal amplitude. Therefore, such optimizations canimprove the thermal properties of building envelope and thus lead tothe usage decrease of lightweight materials [29]. Usually, commercialPCMs could be divided into two main groups, organic material based onparaffin and inorganic material based on salt hydrates [1,30]. From theprospect of physical properties, inorganic PCMs show supercooling andphase segregation during transitional processes and they also exhibitlittle flammability. Differently, organic PCMs are available over a largetemperature range and compatible with conventional constructionmaterials [31]. From the prospect of PCMs encapsulation, macro-encapsulated organic PCMs with stainless steel container [32,33], glasscontainer [34–36] or even plastic bag [37] are easier to realize charge/discharge and avoid leakage problem during operation, which couldhelp facilitating the whole deployment, disassembling, recycle andstorage process. Comparatively, though the salt hydrates are less ex-pensive but they have to be carefully encapsulated due to corrosivenessand hygroscopicity [1]. Therefore, for portable, relocatable and trans-portable lightweight temporary building, macroencapsulated paraffinsseems to be a good choice for its comprehensive advantages. Whenmassive temporary housing demand occurs in a short time, modularcomposite lightweight panels could be rapidly produced and deliveredto the destination to satisfy basic housing demand firstly, and thenfurther improving its thermal comfort by charging PCM into the con-tainer at appropriate time. Considering that there is little interferencebetween PCMs charge/discharge and the house assembling/dis-mantling, this approach will exhibit a high efficiency in the practical

Nomenclature

T temperature, °CTF phase change temperature, °CTe ambient temperature, °CTH indoor design calculation parameter for comfort air con-

ditioning room, °CTsky sky temperature, °CTGS ground surface temperature, °CTin indoor air temperature, °CTo sol–air temperature, °CIsum summer overheating discomfort, °C·hourtsum cumulative discomfort duration, hourcp specific heat, J·kg−1·°C−1

h enthalpy, J·kg−1

r liquid fraction, –L latent heat, J·kg−1

q heat flux, W·m−2

I solar irradiation, W·m−2

RES long wave radiation, W·m−2

t time, minute or hourb thickness, mm or m

Greek symbols

α convective heat transfer coefficient, W·m−2·°C−1

ρ density (kg·m−3)ρs solar absorptivity, –λ thermal conductivity (W·m−1·°C−1)

Subscripts and abbreviation

i indoor airE eastS southW westN northR roofF floorj PCMsl liquids solidref referenceS schemeUDF User-Defined-FunctionCFD computational fluid dynamicSTDkε standard k-ε modelEWT enhanced wall treatment methodEPS expanded polystyrene foam boardsPCMs phase change materialsk steel panel (k= 1) or insulation material (k= 2)

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

36

process.As we all known, the thermal performance of PCM room depends

not only on the properties of PCM but also on building environment[13], such as climatic conditions, orientation and properties of buildingenvelope, etc. When PCMs is macroencapsulated inside of the envelope,the challenge that can be presented is the thickness and location as wellas the orientation [33,37–40]. Meng et al. [12] studied melting tem-perature and thickness effect on thermal behavior of a lightweightbuilding built with prefabricated EPS panel through TRNSYS model.They found that, during summer sunny days, such type of building canbring a temperature drop of 4.28–7.7 °C and reduce the indoor airtemperature fluctuation by up to 28.8–67.8%, but they did not give anyrecommendations on the location and orientation. Lee et al. [41] ex-perimentally studied the optimal PCM location, their results indicatedthat the optimal PCM location was 2.54 cm (south) and 1.27 cm (west)from the wallboard. Jin et al. [37] also investigated the thickness andlocation effect through mathematical model. However, their resultsobtained based on some simplifying assumptions, such as fixed interiorsurface temperature was adopted and the interaction between compo-site PCM envelope and indoor environment was ignored. In short,previous studies mainly focus on heavyweight buildings [31], and theexisted few studies for lightweight buildings are rarely focus on pro-viding emergency accommodation solutions. Also, despite the men-tioned studies on location and thickness, we did not the optimizationwork about multi-orientation scheme in lightweight temporary buildingwith a high shape coefficient [13].

In this work, the thermal behavior of a single PCM room mainlyassembled by modular prefabricated EPS panel outfitted with macro-encapsulated PCM layer for temporary accommodation was studied.The dynamic thermal behavior simulation of the PCM room with a highshape coefficient was performed by the enthalpy-porosity method withANSYS Fluent 15.0 [42]. A comprehensive parametric study was car-ried out to evaluate the influence of location, thickness and orientationof the PCM layer on indoor thermal discomfort in north area of Chinaunder summer condition. Different evaluation indexes, such as indoorair temperature decline, summer overheating discomfort (Isum) andcumulative discomfort duration (tsum), were adopted to evaluate thedynamic thermal performance of PCM room. Unlike most of the pre-vious research work, the multi-orientation optimization scheme of thistype of temporary building were also studied to achieve fully passiveenergy saving.

2. Physical and mathematical models

2.1. Physical model

2.1.1. Geometric descriptionAs shown in Fig. 1(a), the dimension of the model room is

2.40m×2.40m×2.40m. The walls, roof and floor are mainly as-sembled by modular prefabricated EPS panels (thickness: 100mm) withtwo color steel sheets (thickness: 0.5 mm) on each side. The previousstudy of Zhang et al. [43] has suggested that the melting temperatureshould be close to the average room temperature. Therefore, onecommercial organic PCMs (i.e. RT26) provided by local supplier (RUHRTECH) [44] is placed in a 2.20m×2.20m steel container (thickness:2.5 mm, 5.0mm and 7.5mm) inside of the EPS panel, and macro-en-capsulation method of PCM was adopted. With macro-encapsulationPCMs, modular production, rapid charge/discharge and leakage pro-blem can be well coped, and function of the construction structure canalso be less affected. The material properties and geometry size used inthe simulation are shown in Table 1.

In this work, the window(s) and door(s) were not considered, be-cause the main purpose of this study was not to give an accuratelyestimation on energy saving but instead of understanding the potentialof PCM in energy saving, peak load shifting and providing further de-sign suggestions for stakeholders. In fact, the actual influence of solarincident radiation usually depended on different factors, such as or-ientation, shading coefficient, size and quantity of the window(s) aswell as the solar elevation angle [1,12]. In summer, the solar elevationangle in daytime usually was much larger and therefore only part of thesunlight could directly illuminate the inner surface of floor and otherwalls. Besides, for a single lightweight temporary building assembledby modular EPS panel, the solar radiation absorbed by the exteriorsurface in different orientations usually was the main factor that re-sulted in the seriously fluctuation, especially when the lightweighttemporary building had a high shape coefficient. Similar methodologyalso had been adopted by Ye et al. [13].

In the aspect of mesh generation, structural meshing method wasadopted in the computational domain. Besides, mesh independenceverification had been conducted before the simulation. Finally, thesurface and detail mesh distribution of the modelling room wereshowed in Fig. 1(b) and (c). The total element number was 228,600 andthe node number was 241408.

Fig. 1. 3D building model and mesh generation.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

37

2.1.2. Description of the simulation schemeConsidering the disproportion spatiotemporal distribution of solar

irradiance at different orientations, the application effect of PCM inprefabricated EPS panel and its improvement of thermal behavior needto be explored with different location and thickness allocations. For alightweight temporary with a high shape coefficient, it is also necessaryto evaluate the composite PCM envelope at different orientations forfurther optimization. Following that, the multi-orientation schemes ofthis type of temporary building were put forward and discussed in thispaper to realize fully passive energy saving. In order to obtain the op-timized location of PCM layer in single orientation, three different lo-cation allocations with 5.0mm fixed PCM layer at all of the six or-ientations were investigated. Each of the allocations was labeled asCases ‘n’, where ‘Case 1’ referred to the PCM layer to be placed near theinterior surface. ‘Case 2’ represented the PCM layer to be inserted in themiddle position and ‘Case 3’ indicated the PCM layer to be installednear the exterior surface. The detailed allocations for the four walls,roof and floor are illustrated in Fig. 2(a) and (b).

Following the above research, the studies was extended to evaluatethe influence of thickness and orientation on indoor thermal environ-ment. Hence, the estimated thicknesses explored are centered on thevalue of 5.0 mm with a range of 2.5–7.5 mm. Then the simulation arefurther carried out with optimum PCM location at all orientations. Inview of the single orientation installation could not satisfy users well inlightweight buildings with a high shape coefficient, the multi-

orientation layout scheme was proposed to further increase thehomogeneous characters of heat mass. To this end, the multi-orienta-tion scheme were obtained according to the optimized sequence oforientation, which would be elaborated in the following sections.

2.2. Mathematical model

2.2.1. Assumptions for the mathematical modelIn order to study the transient heat transfer characteristics of the

model, assumptions have been listed as follows: (1) natural convectioninside of the PCM layer during melting and the super-cooling effectduring solidifying process are both ignored, (2) the density, heat con-duction coefficient, specific heat and phase change temperature of li-quid phase and solid phase are constant, and not change with tem-perature, (3) physical properties of the other building materials areconstant, (4) the PCMs inside of the PCM layer is homogenous andisotropic, (5) the thermal contact resistances between inner layers ofbuilding envelope are ignored, (6) thermal expansion of PCMs is ne-glected.

2.2.2. Governing equationsFor the indoor fluid domain (air), the three-dimensional unsteady

continuity equations, momentum equations and energy governingequations are given as Eq. (1)–(5).

Continuity:

Table 1Physical properties of selected materials used in the simulation.

Material Densityρ (kg·m−3)

Thermal conductivityλ (W·m−1·°C−1)

Specific heatcp (J·kg−1·°C−1)

Latent heatL (kJ·kg−1)

Thicknessb (mm)

Melting/solidificationTF (°C)

EPS 12.7 0.045 1090 / 100 (97.5, 95.0, 92.5) /Steel sheet 8030 16.27 500 / 0.5 /PCMs 870 (l) 0.2 (l) 1800 (l) 180 2.5, 5.0, 7.5 26 (l)(RT27) 750 (s) 0.2 (s) 2400 (s) 26 (s)PCMs 880 (l) 0.21 (l) 3200 220 5.0 41 (l)(RT42) 760 (s) 0.21 (s) 42 (s)

Fig. 2. Location allocation of PCM layer in building envelope system.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

38

∂∂

+∂∂

+ ∂∂

=ux

uy

uz

0x y x(1)

X-momentum:

⎜ ⎟

⎜ ⎟

⎛⎝

∂∂

+ ∂∂

+ ∂∂

+ ∂∂

⎞⎠

= −∂∂

+ ⎛⎝

∂∂

+ ∂∂

+ ∂∂

⎞⎠

ρ ut

u ux

u uy

u uz

px

μ ux

uy

uz

ix

xx

yx

zx

x x x2

2

2

2

2

2 (2)

Y-momentum:

⎜ ⎟

⎜ ⎟⎛⎝

∂∂

+∂∂

+∂∂

+∂∂

⎞⎠

= −∂∂

+ ⎛⎝

∂∂

+∂∂

+∂∂

⎞⎠

+ −

ρut

uux

uuy

uuz

py

μux

uy

uz

ρ gβ T T( )

iy

xy

yy

zy

y y yi

2

2

2

2

2

2 0 (3)

Z-momentum:

⎜ ⎟ ⎜ ⎟⎛⎝

∂∂

+∂∂

+∂∂

+∂∂

⎞⎠

= −∂∂

+ ⎛⎝

∂∂

+∂∂

+∂∂

⎞⎠

ρut

uux

uuy

uuz

pz

μux

uy

uyzi

zx

zy

zz

z z z z2

2

2

2

2

2

(4)

Energy:

⎜ ⎟ ⎜ ⎟⎛⎝

∂∂

+ ∂∂

+ ∂∂

+ ∂∂

⎞⎠

= ⎛⎝

∂∂

+ ∂∂

+ ∂∂

⎞⎠

ρ C Tt

u Tx

u Ty

u Tz

λ Tx

Ty

Tyzi p x y z i

2

2

2

2

2

2i (5)

where T is the temperature (°C). ρ, λ and cp are density (kg·m−3),thermal conductivity (W·m−1·°C−1) and specific heat (J·kg−1·°C−1),respectively. And subscript i represents the indoor air.

For PCM domain, Fluent provides two methods to solve melting andsolidifying, i.e. enthalpy-porosity method and equivalent heat capacitymethod. In this paper, the enthalpy-porosity method which determinesthe enthalpy value in the energy equation is considered to predict thethermal behavior of composite PCM envelope by numerically solvingNavier-Stokes partial differential equations of mass, energy and mo-mentum [34–36,42]. The total enthalpy value covers both sensible andlatent parts in which the latent enthalpy is evaluated as a percentage ofthe latent heat of the PCM in liquid phase state (i.e. liquid fraction). Thespecific energy governing equations are as follows:

⎜ ⎟∂∂

= ⎛⎝

∂∂

+ ∂∂

+ ∂∂

⎞⎠

ρ ht

λ Tx

Ty

Tyzj j

2

2

2

2

2

2 (6)

where h is the enthalpy (J·kg−1). Subscript j denotes the PCMs. Besides,the enthalpy of PCM layer is computed as the sum value of the sensibleenthalpy and latent heat:

∫= + +h h C dT rLref TT

pref (7)

where href represents the reference enthalpy (href , J·kg−1) at the re-ference temperature (Tref , °C). r is the liquid fraction of PCM layer. L isthe latent heat of PCMs (J·kg−1). Among them, the liquid fraction of thePCM layer is defined as:

= ⎧⎨⎩

<>

rif T T solidif T T liquid

0, ( )1, ( )

F

F (8)

where TF is the phase change temperature (°C).For the other computational domain, related energy governing

equation is shown in Eq. (9)

⎜ ⎟∂∂

= ⎛⎝

∂∂

+ ∂∂

+ ∂∂

⎞⎠

=ρC Tt

λ Tx

Ty

Tyz

k( ) 1,2p k k2

2

2

2

2

2 (9)

where subscript k denotes the steel sheet (k=1) and insulation mate-rial (k=2) of the prefabricated EPS panel, respectively.

2.2.3. Initial and boundary conditionsIn this study, the modelling room is located in Tianjin city (a middle

latitude city in China), which represents a typical warm temperatesemi-humid continental monsoon climate with cold winters and hot andrelatively dry summers. The obvious temperature difference betweendaytime and nighttime shows that Tianjin has the application potentialof PCMs as well as the utilization of night cooling.

For the surfaces exposed to indoor air, the boundary condition isgiven by Eq. (10) [13,36]:

∂∂

= −=

=λTx

α T T( )x

in in x0

0(10)

where λ0 (W·m−1·°C−1) is the thermal conductivity of indoor layer (i.e.color steel sheet). αin (W·m−2·°C−1) is the convective heat transfer

Fig. 3. The chosen meteorological data of Tianjin during simulation.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

39

coefficient of interior surface, and the inner surface is coupled with theindoor air. Tin and =Tx 0 (°C) are the indoor air temperature and thetemperature on the interior surface, respectively.

For the surfaces exposed to outdoor environment, the boundarycondition is given by Eq. (11) [36,40]:

∂∂

= − + −=

=λTx

α T T ρ I R( )x L

ex e x L s ES (11)

where λL (W·m−1·°C−1) is the thermal conductivity of outer door layer(i.e. color steel sheet). αex (W·m−2·°C−1) is the convective heat transfercoefficient of exterior surface, and the αex is obtained according to theEq. (12) [45]. Te and =Tx L (°C) are the ambient temperature and thetemperature on the exterior surface, respectively. ρs is the solar ab-sorptivity of exterior surface, which is set to be 0.6 [46]. (W·m−2) is thesolar irradiation. RES (W·m−2) is the radiation heat exchange betweenenvelope and sky/ground surface.

= +α v5.62 3.9e (12)

The heat exchange caused by solar radiation and heat convectionbetween ambient and envelope is substituted by the sol–air temperature(To, °C), which could be defined as [28,47,48]:

= + =TI ρα

T·

os

ex L (13)

For the contact surface between different layers, including insula-tion material, steel sheet, PCM container and PCM (when PCM layer isnot undergoing a phase transition), the boundary is given as Eq. (14)[49]:

− ∂∂

= − ∂∂

λ Tx

λ Txx layer x layer, 1 , 2 (14)

Below is the boundary condition for contact surface between thePCM and PCM container when a phase transition occurs inside of thePCM layer, the boundary is given as Eq. (15) [49]:

− ∂∂

= ∂∂

− ∂∂

λ Tx

Lρ rt

λ Txx layer

jx layer, 1 , 2 (15)

where and (W·m−1·°C−1) are the thermal conductivity of layer 1 andlayer 2. Besides, the exterior surface of the floor is assumed to be theadiabatic boundary.

As to the simulation time, the study work of Elnajjar [50] showedthat a one day evaluation for PCM room can be misleading. Therefore, afive day’s evaluation simulation was adopted in this paper and only theresults of the fifth day was presented and used for further analysis. Thedynamic monitor value of solar radiation at different orientations, skytemperature, ground surface temperature and wind velocity of Tianjinin summer could be extracted from the Special Meteorological Data Setfor Building Thermal Environment Analysis of China [51]. The data ofJuly 06–10 used in simulation were shown in Fig. 3.

2.3. Model validation

2.3.1. Experimental setupThe experimental devices as well as the schematic diagram and the

thermocouples’ distribution used to validate the model were shown inFig. 4. The experimental system mainly consisted of: (1) test sample ofmodular EPS panel; (2) on-site weather station (Jinzhou Sunshine/ PC-4); (3): computer (Pavilion15/ HP); (4) data acquisition instrumentwith an accuracy of± 0.5% (Changhui/ SWP-MD809); (5) 5.0mmmacroencapsulated PCM layer (RT42/ RUHR TECH); and several K-type thermocouples (OMEGA Company/5TC-TT-K-36-36). In the testsample, the PCM layer with an equivalent size of 0.5m in length andwidth was placed near the exterior surface of the composite panel. Fivethermocouples distributed as Fig. 4(c) was fixed at the backside of thePCM layer to monitor the dynamic temperature profile during the test.The experimental setup is located in Tianjin of China and, specially,located besides the building of school of architecture in Tianjin Uni-versity. Considering that experimental PCM panel directly exposed in

Fig. 4. Experimental test rig, schematic diagram and thermocouples’ distribution.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

40

the hot ambient environment, RT 42 with a melting temperature of42 °C was adopted in the validation test to ensure the PCM in solid statebefore the test. Compared with RT26 used in the simulation, we hadtaken into account the fact that the correctness of the heat transfermodel would not be changed with the change of physical parameter ofmaterials [13,39,52].

2.3.2. Validation of numerical procedureThe validation experiment was carried out on June 27, 2017, the

meteorological data was shown in Fig. 5(a). As shown in Fig. 5(b), thesimulation results of the back-side temperature of the PCM layer agreedwell with the experimental data. However, there is also some differencebetween the experimental and numerical results, and the numericalresults firstly was a little higher than experimental value during the firsthalf of the studied period. Based on the experimental result, the max-imum temperature difference between the numerical value and theexperimental value was 1.9 °C, therefore, the corresponding maximumrelative error during this whole test was 4.1%. This was due to theactual environmental condition fluctuated seriously during the first halfof the experiment, especially the solar radiation, which resulted in thenumerical results higher than that of the measurement. In view of that,the simulation results have good identity with the experimental testresults. Besides, similar models in have also been well verified in largeamounts of literatures, e.g. Kong et al. [39], Faheem et al. [53] andother researchers [35,37,54], their simulation results are in extremelygood agreement with experiments.

The standard k-ε model (STDkε) with enhanced wall treatmentmethod (EWT) was adopted in this paper. The governing equationswere discretized by the second order upwind scheme for advectionterms and central differencing scheme (implicit scheme with second-order accuracy) for diffusion terms, and the coupling of the velocity andpressure fields was carried out by using SIMPLE scheme. Moreover,default residual convergence criteria for each governing equations wereemployed for all simulations, and the time step and the simulation timewere respectively set at 60 s and 120 h, and the total iteration stepswere 7220.

3. Evaluation indicator of indoor thermal discomfort

To evaluate promotion of indoor thermal comfort after in-corporating PCMs into the envelope, two different indoor thermal dis-comfort indicators, including summer overheating discomfort (Isum) andcumulative discomfort duration (tsum), are introduced and given as Eqs.(16) and (17):

∫= − >I T T dτ T T( ) ,sum in H in H024

(16)

∫= >t dτ T T,sum in H024

(17)

where TH is the indoor design calculation parameter for comfort airconditioning room (°C). Isum is the time integral of temperature differ-ence between indoor temperature and maximum indoor design calcu-lation temperature (°C·hour). Where tsum is the time integral of indoorthermal discomfort time (hour). Obviously, a smaller value of Isum ortsum indicates a better indoor thermal comfort. When Isum is equal orclose to zero, it means that the building can maintain indoor thermalcomfort without additional active cooling measures, and therefore thiskind of building could be described as an ideal passive building. Withrespect to TH , Chinese design code for heating ventilation and air con-ditioning of civil buildings (GB50736-2012) recommends that it shouldbe controlled in the range of 24–28 °C. Therefore, the value of 28 °C wasused as the reference temperature in this paper.

4. Results and discussions

4.1. Location effect of PCM layer

In the present work, the location effect of PCM layer on indoorthermal environment was considered at all six orientations. The indoortemperature response and the corresponding liquid fraction under dif-ferent location allocations, i.e. No PCM, Case 1, Case 2 and Case 3, wereshown in Figs. 6 and 7. Compared with reference room, the PCM roomshowed an obvious improvement, which proved the function of PCM inlifting indoor thermal comfort and reducing energy consumption. Fur-ther, when the PCM layer was placed far from the environment heatsource (i.e. Case 1), it was found that its impact on reducing the indoortemperature was better than the other two configurations, and thisphenomenon was irrelevant to the orientation except for the roof. Whenit was placed close to the exterior surface, the PCM layer could reach toits phase transition temperature more quickly in both diurnal andnocturnal time, and then a more fully melt process than Case 1 and Case2 consequently occurred. However, the PCM in Case 3 didn’t increasethe heat absorbed from the room transferred from other orientations,and it actually only decreased the heat release to the room in orienta-tion applied PCM. From the prospect of balance between effective heatabsorbed from room and heat release to the room, Case 1 could achievea maximum benefit when integrated in most orientations.

Fig. 8 further addressed the location effect on indoor thermal be-havior, including the maximum indoor temperature, correspondingdecline and indoor thermal discomfort. As depicted in Fig. 8(a), themaximum indoor temperature of the ordinary room reached to an un-bearable value of 42.68 °C. Meanwhile, it could be seen that the PCMroom exhibited much lower maximum indoor temperature, and the

Fig. 5. Meteorological data and back-side temperature comparison of PCM layer between numerical results and experimental results.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

41

maximum temperature decline of Case 1 was 5.07 °C in roof and thecorresponding maximum indoor temperature was 37.61 °C. However,the maximum temperature decline of Case 3 was only 1.80 °C in floorand the corresponding maximum indoor temperature reached to40.88 °C. Since the indoor temperature severely exceeded the ambientand there was almost no difference between the reference room andPCM room, the Case 3 was proved to be the least effective location.

The effect of location on Isum and tsum were summarized and listed inFig. 8(b). It was presented that the two indicators calculated in Case 1were both the smallest among the three cases within the same

orientation. This phenomenon also indicated that the study room out-fitted with PCM layer being composited near the inner surface had abetter lifting effect throughout the whole day. The minimum tsum andIsum value of Case 1 were 11.1 h (in east) and 72.1 °C·hour (in roof),respectively. Therefore, the Case 1 was recommended among the threelocation allocations in this study.

4.2. Thickness effect of PCM layer

Generally, if the PCM layer embedded in the building envelope

Fig. 6. Location effect of PCM layer on indoor temperature.

Fig. 7. Location effect of PCM layer on liquid fraction.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

42

structure was too thin, the substantially melt and almost no solidifica-tion process would happen [7]. On the contrary, a thick enough PCMlayer was also not recommended due to its unsatisfactory thermalperformance and waste. Therefore, the thickness of PCM layer wasanother key design parameter for PCM room. Thus, following the pre-viously study relative to the location, the further simulation was ex-tended to evaluate the thickness effect.

Figs. 9 and 10 illustrated the indoor temperature and liquid fractioninside of the PCM layer with different thickness, respectively. As shownin Fig. 10, when b=2.5mm, the melting and solidification processinside of the PCM layer all happened, but the melting process fullycompleted in a fast speed (about 5–7 h) and the partly solidificationprocess was in a relative slow speed (about 7–12 h) in most orienta-tions, including roof, east, south and west. When b= 5.0mm or7.5 mm, the melting and solidification speed could basically achievebalance. However, the fully melt state never occurred in 7.5mm con-dition even after 5 days’ operation, which indicated that there existedan slightly overuse of PCMs.

Moreover, the influence of thickness on different indoor thermalbehavior indicators like maximum indoor temperature, decline andindoor thermal discomfort were listed in Fig. 11. In summer, as com-pared with ordinary room, the lowest maximum indoor temperatureand the largest indoor temperature decline of PCM room were 37.54 °Cand 5.14 °C, respectively, when 7.5mm PCM layer was embedded in theroof. Although simulation results showed that increasing the thicknesscould decrease the maximum indoor temperature, the temperaturedifference between 5.0 mm and 7.5 mm was not obvious. The calcu-lated temperature difference mentioned above was maintained withinthe range of 0.05–0.22 °C. In addition, the similar conclusion could alsobe obtained from the calculated values of Isum and tsum of each or-ientation in Fig. 11(b). It can be seen that the minimum value of Isumand tsum value were 71.5 °C·hour (in roof) and 10.9 h (in east),

respectively. As compared with the PCM room outfitted with 5.0mmPCM layer, the Isum value of the study room with 7.5 mm PCM layerdeclined less than 5.6%.

As a result, on the one hand, the results indicated that less and in-sufficient PCM would reduce its thermal storage capacity and thereforeplayed limited role in relieving temperature fluctuation. On the otherhand, the indoor thermal environment of lightweight temporarybuilding didn’t improve significantly when 7.5 mm PCM layer adopted,which mean that 5.0mm PCM layer was enough for the target room.Besides, as we all know, the lower thermal conductivity was better forenergy conservation for a normal wall due to the thermal resistance wasincreased. Therefore, in this situation, the composite PCM panel out-fitted with 5.0 mm PCM layer not only could use the sufficient latentheat to absorb the heat, but also could utilize the low thermal con-ductivity of the composite panel to block the heat transfer and pos-sessed a better economical feature, which were consistent with thework of Ye et al [13] and Zhou et al [55].

4.3. Orientation effect and optimization layout scheme

It should be noted that though the PCM layer embedded in eachorientation could strengthen indoor thermal comfort effectively, therestill existed a huge improvement space for this type of buildings. Asmentioned above, the minimum value of Isum and tsum value were71.5 °C·hour and 10.9 h respectively, which would make users un-comfortable or even cause healthy problem. In view of the insignif-icance of increasing PCM in one orientation, an effective way was toincrease the homogeneous characters of heat mass in building envelopestructure.

Due to the fact that different orientation has different solar-airtemperature, the thermal performance of the composite PCM panel indifferent orientation was also different. The Isum and tsum value of

Fig. 8. Location effect of PCM layer on indoor thermal behavior.

Fig. 9. Thickness effect of PCM layer on indoor temperature.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

43

5.0 mm case at different orientation were calculated and shown inFig. 12. It could be seen an obvious distinction between the two in-dictors from Fig. 12. Compared with the reference room, the tsum valueof east-facing panel was reduced by 1 h, however, almost no reductionof tsum was found when roof-facing panel adopted. On the other hand,the Isum value of roof-facing panel was declined by 31.2 °C·hour and theIsum value of north-facing panel still showed a reduction of 18.1 °C·hour.The distinction between the two indicators proved the PCM roompossessed a good peak load shifting effect, but only single orientationoutfitted with PCM layer could not realize fully indoor comfort.Therefore, further optimum design scheme was selected based on theindicator of Isum value. According to the value of Isum, the other fourorientations were sorted as per the sequence of west (76.5 °C·hour),south (77.9 °C·hour), floor (79.0 °C·hour) and east (80.9 °C·hour). Thus,all the five multi-orientation optimization layout schemes were labeledas S2-S6 and listed in Table 2 for further study.

4.4. Multi-orientation layout optimization

Figs. 13 and 14 described the indoor temperature and liquid fractionprofiles of single orientation scheme and the other five multi-orienta-tion optimization schemes in PCM room. Fig. 13 showed that themaximum indoor temperature declined rapidly with the quantity in-crease of PCM layer in different orientations. That is to say, increasingthe quantity of PCM layer in different orientations could increase theoverall heat absorbed from the room and decrease the overall heatreleased to the room. Further observation showed that, when orienta-tion numbers exceeded four, the indoor temperature decreased belowambient. When orientation numbers exceeded five, the indoor tem-perature of PCM room could be maintained within 28 °C throughout thewhole day. Meanwhile, as could be seen in Fig. 14, the melting andsolidification process all happened in multi-orientations scheme and

Fig. 10. Thickness effect of PCM layer on liquid fraction.

Fig. 11. Thickness effect of PCM layer on indoor thermal behavior.

Fig. 12. Indoor thermal discomfort of the modelling room with 5.0 mm PCMlayer.

Table 2Multi-orientation optimization scheme.

Optimization scheme South East North West Roof Floor

S2 ○ ○ ○ √ √ ○S3 √ ○ ○ √ √ ○S4 √ ○ ○ √ √ √S5 √ √ ○ √ √ √S6 √ √ √ √ √ √

Footnotes: ‘○’ (or ‘√’) represents the PCM layer was (or was not) applied inassigned orientation.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

44

could basically achieve a dynamic balance.The influence of multi-orientation scheme on indoor thermal dis-

comfort were illustrated in Fig. 15. It was shown in Fig. 15(a) that thePCM room exhibited a continuous temperature falling as the quantity ofPCM layer increased in different orientations. In detail, the maximumindoor temperature was reduced to 27.96 °C when S5 was applied,which was lower than the setting upper limits of indoor thermal com-fort temperature mentioned above. As the quantity of PCM panel fur-ther grows (i.e. S6), the maximum temperature decline was 15.52 °C,which only increased by 0.80 °C.

Fig. 15(b) illustrated the expected improvement in the specificconditions of the study from the prospect of indoor thermal discomfort.Clearly, the Isum and tsum value of the study room were both reduced to0 when composite PCM panel exceeded five, which proved again thatthere existed an optimum multi-orientation design scheme for light-weight temporary building, and the suitable multi-orientation schemein this paper was S5.

Therefore, when incorporating the PCM layer in at least five or-ientations (S5), the proposed multi-orientation optimization schemescould ensure a comfortable indoor climate under the extremely closedcondition, i.e. no window(s) and door(s) in this paper, without extramechanical cooling measures. That is to say, the application effect ofPCM in this type of building would be better when combined withnatural ventilation cooling in nighttime. Besides, the results also re-vealed that it was not necessary to further increase the quantity of PCMcomposite panel in more orientations.

5. Conclusions

This research aimed to give a detailed understanding of the influ-ence of some key parameter, such as location, thickness and orienta-tion, on summer thermal behavior of lightweight temporary buildingoutfitted with PCM in North China. Following that, multi-orientationoptimization schemes were put forward to achieve the goal of fullypassive energy saving. The useful conclusions are shown as follow:

(1) For lightweight temporary building assembled by modular pre-fabricated EPS panel with a high shape coefficient, the PCM layerplaced next to the interior environment was recommended amongthree studied location allocations. This conclusion was in agree-ment with the previous studies of Jin et al. [37] and Gounni et al.[38]. It could also be concluded that it was the overall ability toblock the heat release to the room and accelerate the heat absorbedfrom the room that determined the optimum location.

(2) The composite PCM envelope structure with a reasonable PCMdosage, e.g. 5.0mm in this paper, was recommended among thethree studied thickness conditions. The results indicated that lessand insufficient PCM would reduce its thermal storage capacity andtherefore could only play limited role in relieving temperaturefluctuation. Also, further increasing the thickness of PCM layer waseffective but limited, which required further consideration of itsinvestment costs and the economic benefits it can bring.

(3) When incorporating the PCM layer in at least five orientations (S5),the proposed multi-orientation optimization schemes could ensure

Fig. 13. Indoor temperature profile with PCM layer in multi-orientation.

Fig. 14. Liquid fraction profile under different multi-orientation layout scheme.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

45

a comfortable indoor climate under the extremely closed conditionwithout extra mechanical cooling measures. That is to say, themodular prefabricated composite PCM panel will play a morepromising role in reducing the peak heat load during summer sunnydays when combined with natural ventilation cooling in nighttime.

Acknowledgement

This work was sponsored by Natural Science Foundation of China(Grant No. 51478297) and the Programme of Introducing Talents(Grant No. B13011).

References

[1] S. Sonnick, L. Erlbeck, K. Schlachter, et al., Temperature stabilization using salthydrate storage system to achieve thermal comfort in prefabricated wooden houses,Energy Build. (2018).

[2] L. Erlbeck, P. Schreiner, K. Schlachter, et al., Adjustment of thermal behavior bychanging the shape of PCM inclusions in concrete blocks, Energy Convers. Manage.158 (2018) 256–265.

[3] V.V. Tyagi, S.C. Kaushik, S.K. Tyagi, et al., Development of phase change materialsbased microencapsulated technology for buildings: A review, Renew. Sustain.Energy Rev. 15 (2) (2011) 1373–1391.

[4] F. Rouault, D. Bruneau, P. Sebastian, et al., Use of a latent heat thermal energystorage system for cooling a light-weight building: Experimentation and co-simu-lation, Energy Build. 127 (2016) 479–487.

[5] C. Tonelli, M. Grimaudo, Timber buildings and thermal inertia: Open scientificproblems for summer behavior in Mediterranean climate, Energy Build. 83 (Part2)(2014) 89–95.

[6] D. Félix, D. Monteiro, J.M. Branco, R. Bologna, A. Feio, The role of temporary ac-commodation buildings for post-disaster housing reconstruction, J. House Built.Environ. 30 (2014) 683–699.

[7] C. Cornaro, D. Sapori, F. Bucci, et al., Thermal performance analysis of an emer-gency shelter using dynamic building simulation, Energy Build. 88 (2015) 122–134.

[8] P. Marin, M. Saffari, A.D. Gracia, et al., Energy savings due to the use of PCM forrelocatable lightweight buildings passive heating and cooling in different weatherconditions, Energy Build. 129 (2016) 274–283.

[9] S.J. Kolo, F.P. Rahimian, J.S. Goulding, Offsite manufacturing construction: a bi-gopportunity for housing delivery in Nigeria, Procedia Eng. 85 (2014) 319–327.

[10] Y. Peng, A comparison of two approaches to develop concentrated rural settlementsafter the 5.12 Sichuan Earthquake in China, Habitat Int. 49 (2015) 230–242.

[11] D. Félix, J.M. Branco, A. Feio, Temporary housing after disasters: A state of the artsurvey, Habitat Int. 40 (2013) 136–141.

[12] E. Meng, H. Yu, B. Zhou, Study of the thermal behavior of the composite phasechange material (PCM) room in summer and winter, Appl. Therm. Eng. 126 (2017).

[13] R. Ye, W. Lin, X. Fang, et al., A numerical study of building integrated withCaCl2•6H2O/expanded graphite composite phase change material, Appl. Therm.Eng. 126 (2017) 480–488.

[14] J. Eischeid, Flat-pack shelter, Green Futures 90 (2013) 11.[15] Y. Zhang, K. Du, J. He, Optimized analysis of lightweight wall outfitted with PCM in

hot summer zone, Proceedings of the 8th International Symposium on Heating,Ventilation and Air Conditioning, Springer, Berlin Heidelberg, 2014, pp. 233–240.

[16] W. Xiao, X. Wang, Y.P. Zhang, Thermal analysis of lightweight wall with shapest-abilized PCM panel for summer insulation, J. Eng. Thermophys. 30 (2009)1561–1563.

[17] Y. Sun, Y. Zhang, L. Bao, Z. Fan, D. Wang, J. Sundell, Effects of gender and dor-mitory environment on sick building syndrome symptoms among college studentsin Tianjin, China, Build. Environ. 68 (2013) 134–139.

[18] B.L. Gowreesunker, S.B. Stankovic, S.A. Tassou, P.A. Kyriacou, Experimental andnumerical investigations of the optical and thermal aspects of a PCM-glazed unit,Energy Build. 61 (2013) 239–249.

[19] M. Arıcı, H. Karabay, Determination of optimum thickness of double-glazed win-dows for the climatic regions of Turkey, Energy Build. 42 (2010) 1773–1778.

[20] C. Liu, Y. Wu, D. Li, et al., Effect of PCM thickness and melting temperature onthermal performance of double glazing units, J. Build. Eng. (2017).

[21] S.B. Sadineni, S. Madala, R.F. Boehm, Passive building energy savings: a review ofbuilding envelope components, Renew. Sustain. Energy Rev. 15 (2011) 3617–3631.

[22] C. Shen, X. Li, Energy saving potential of pipe-embedded building envelope uti-lizing low-temperature hot water in the heating season, Energy Build. 138 (2016).

[23] J.P. Hidalgo, S. Welch, J.L. Torero, Performance criteria for the fire safe use ofthermal insulation in buildings, Constr. Build. Mater. 100 (2015) 285–297.

[24] V.V. Tyagi, D. Buddhi, PCM thermal storage in buildings: a state of art, Renew.Sustain. Energy Rev. 11 (6) (2007) 1146–1166.

[25] B. Severson, A. St. Leger, Feasibility study of photovoltaic panels in military tem-porary housing structures, IEEE Green Technologies Conference, IEEE ComputerSociety, 2013, pp. 78–84.

[26] M.A. Medina, E.J.B. King, M. Zhang, On the heat transfer rate reduction of struc-tural insulated panels (SIPs) outfitted with phase change materials (PCMs), Energy33 (2008) 667–678.

[27] R. Vicente, T. Silva, Brick masonry walls with PCM macrocapsules: an experimentalapproach, Appl. Therm. Eng. 67 (2014) 24–34.

[28] T. Silva, R. Vicente, C. Amaral, A. Figueiredo, Thermal performance of a windowshutter containing PCM: numerical validation and experimental analysis, Appl.Energy 179 (2016) 64–84.

[29] B.M. Diaconu, M. Cruceru, Novel concept of composite phase change material wallsystem for year-round thermal energy savings, Energy Build. 42 (10) (2010)1759–1772.

[30] X. Wang, H. Yu, L. Li, M. Zhao, Experimental assessment on the use of phase changematerials (PCMs)-bricks in the exterior wall of a full-scale room, Energy Convers.Manage. 120 (2016) 81–89.

[31] M.A. Izquierdo-Barrientos, J.F. Belmonte, D. Rodríguez-Sánchez, et al., A numericalstudy of external building walls containing phase change materials (PCM), Appl.Therm. Eng. 47 (2012) 73–85.

[32] M.C. Browne, D. Quigley, H.R. Hard, et al., Assessing the thermal performance ofphase change material in a photovoltaic/thermal system, Energy Proc. 91 (2016)113–121.

[33] Y. Li, S. Liu, J. Lu, Effects of various parameters of a PCM on thermal performanceof a solar chimney, Appl. Therm. Eng. 127 (2017).

[34] S. Li, K. Zou, G. Sun, et al., Simulation research on the dynamic thermal perfor-mance of a novel triple-glazed window filled with PCM, Sustain. Cities Soc. (2018).

[35] C. Liu, Y. Zhou, D. Li, et al., Numerical analysis on thermal performance of a PCM-filled double glazing roof, Energy Build. 125 (2016) 267–275.

[36] D. Li, Y. Zheng, C. Liu, et al., Numerical analysis on thermal performance of roofcontained PCM of a single residential building, Energy Convers. Manage. 100(2015) 147–156.

[37] X. Jin, M.A. Medina, X. Zhang, Numerical analysis for the optimal location of a thinPCM layer in frame walls, Appl. Therm. Eng. 103 (2016) 1057–1063.

[38] A. Gounni, M.E. Alami, The optimal allocation of the PCM within a composite wallfor surface temperature and heat flux reduction: an experimental approach, Appl.Therm. Eng. (2017).

[39] X. Kong, S. Lu, Y. Li, et al., Numerical study on the thermal performance of buildingwall and roof incorporating phase change material panel for passive cooling ap-plication, Energy Build. 81 (2014) 404–415.

[40] H. Hamza, N. Hanchi, B. Abouelkhayrat, et al., Location and thickness effect of twophase change materials between layers of roof on energy consumption for air-

Fig. 15. Indoor thermal behavior under different multi-orientation layout scheme.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

46

http://refhub.elsevier.com/S1359-4311(17)37939-5/h0005http://refhub.elsevier.com/S1359-4311(17)37939-5/h0005http://refhub.elsevier.com/S1359-4311(17)37939-5/h0005http://refhub.elsevier.com/S1359-4311(17)37939-5/h0010http://refhub.elsevier.com/S1359-4311(17)37939-5/h0010http://refhub.elsevier.com/S1359-4311(17)37939-5/h0010http://refhub.elsevier.com/S1359-4311(17)37939-5/h0015http://refhub.elsevier.com/S1359-4311(17)37939-5/h0015http://refhub.elsevier.com/S1359-4311(17)37939-5/h0015http://refhub.elsevier.com/S1359-4311(17)37939-5/h0020http://refhub.elsevier.com/S1359-4311(17)37939-5/h0020http://refhub.elsevier.com/S1359-4311(17)37939-5/h0020http://refhub.elsevier.com/S1359-4311(17)37939-5/h0025http://refhub.elsevier.com/S1359-4311(17)37939-5/h0025http://refhub.elsevier.com/S1359-4311(17)37939-5/h0025http://refhub.elsevier.com/S1359-4311(17)37939-5/h0030http://refhub.elsevier.com/S1359-4311(17)37939-5/h0030http://refhub.elsevier.com/S1359-4311(17)37939-5/h0030http://refhub.elsevier.com/S1359-4311(17)37939-5/h0035http://refhub.elsevier.com/S1359-4311(17)37939-5/h0035http://refhub.elsevier.com/S1359-4311(17)37939-5/h0040http://refhub.elsevier.com/S1359-4311(17)37939-5/h0040http://refhub.elsevier.com/S1359-4311(17)37939-5/h0040http://refhub.elsevier.com/S1359-4311(17)37939-5/h0045http://refhub.elsevier.com/S1359-4311(17)37939-5/h0045http://refhub.elsevier.com/S1359-4311(17)37939-5/h0050http://refhub.elsevier.com/S1359-4311(17)37939-5/h0050http://refhub.elsevier.com/S1359-4311(17)37939-5/h0055http://refhub.elsevier.com/S1359-4311(17)37939-5/h0055http://refhub.elsevier.com/S1359-4311(17)37939-5/h0060http://refhub.elsevier.com/S1359-4311(17)37939-5/h0060http://refhub.elsevier.com/S1359-4311(17)37939-5/h0065http://refhub.elsevier.com/S1359-4311(17)37939-5/h0065http://refhub.elsevier.com/S1359-4311(17)37939-5/h0065http://refhub.elsevier.com/S1359-4311(17)37939-5/h0070http://refhub.elsevier.com/S1359-4311(17)37939-5/h0075http://refhub.elsevier.com/S1359-4311(17)37939-5/h0075http://refhub.elsevier.com/S1359-4311(17)37939-5/h0075http://refhub.elsevier.com/S1359-4311(17)37939-5/h0080http://refhub.elsevier.com/S1359-4311(17)37939-5/h0080http://refhub.elsevier.com/S1359-4311(17)37939-5/h0080http://refhub.elsevier.com/S1359-4311(17)37939-5/h0085http://refhub.elsevier.com/S1359-4311(17)37939-5/h0085http://refhub.elsevier.com/S1359-4311(17)37939-5/h0085http://refhub.elsevier.com/S1359-4311(17)37939-5/h0090http://refhub.elsevier.com/S1359-4311(17)37939-5/h0090http://refhub.elsevier.com/S1359-4311(17)37939-5/h0090http://refhub.elsevier.com/S1359-4311(17)37939-5/h0095http://refhub.elsevier.com/S1359-4311(17)37939-5/h0095http://refhub.elsevier.com/S1359-4311(17)37939-5/h0100http://refhub.elsevier.com/S1359-4311(17)37939-5/h0100http://refhub.elsevier.com/S1359-4311(17)37939-5/h0105http://refhub.elsevier.com/S1359-4311(17)37939-5/h0105http://refhub.elsevier.com/S1359-4311(17)37939-5/h0110http://refhub.elsevier.com/S1359-4311(17)37939-5/h0110http://refhub.elsevier.com/S1359-4311(17)37939-5/h0115http://refhub.elsevier.com/S1359-4311(17)37939-5/h0115http://refhub.elsevier.com/S1359-4311(17)37939-5/h0120http://refhub.elsevier.com/S1359-4311(17)37939-5/h0120http://refhub.elsevier.com/S1359-4311(17)37939-5/h0125http://refhub.elsevier.com/S1359-4311(17)37939-5/h0125http://refhub.elsevier.com/S1359-4311(17)37939-5/h0125http://refhub.elsevier.com/S1359-4311(17)37939-5/h0130http://refhub.elsevier.com/S1359-4311(17)37939-5/h0130http://refhub.elsevier.com/S1359-4311(17)37939-5/h0130http://refhub.elsevier.com/S1359-4311(17)37939-5/h0135http://refhub.elsevier.com/S1359-4311(17)37939-5/h0135http://refhub.elsevier.com/S1359-4311(17)37939-5/h0140http://refhub.elsevier.com/S1359-4311(17)37939-5/h0140http://refhub.elsevier.com/S1359-4311(17)37939-5/h0140http://refhub.elsevier.com/S1359-4311(17)37939-5/h0145http://refhub.elsevier.com/S1359-4311(17)37939-5/h0145http://refhub.elsevier.com/S1359-4311(17)37939-5/h0145http://refhub.elsevier.com/S1359-4311(17)37939-5/h0150http://refhub.elsevier.com/S1359-4311(17)37939-5/h0150http://refhub.elsevier.com/S1359-4311(17)37939-5/h0150http://refhub.elsevier.com/S1359-4311(17)37939-5/h0155http://refhub.elsevier.com/S1359-4311(17)37939-5/h0155http://refhub.elsevier.com/S1359-4311(17)37939-5/h0155http://refhub.elsevier.com/S1359-4311(17)37939-5/h0160http://refhub.elsevier.com/S1359-4311(17)37939-5/h0160http://refhub.elsevier.com/S1359-4311(17)37939-5/h0160http://refhub.elsevier.com/S1359-4311(17)37939-5/h0165http://refhub.elsevier.com/S1359-4311(17)37939-5/h0165http://refhub.elsevier.com/S1359-4311(17)37939-5/h0170http://refhub.elsevier.com/S1359-4311(17)37939-5/h0170http://refhub.elsevier.com/S1359-4311(17)37939-5/h0175http://refhub.elsevier.com/S1359-4311(17)37939-5/h0175http://refhub.elsevier.com/S1359-4311(17)37939-5/h0180http://refhub.elsevier.com/S1359-4311(17)37939-5/h0180http://refhub.elsevier.com/S1359-4311(17)37939-5/h0180http://refhub.elsevier.com/S1359-4311(17)37939-5/h0185http://refhub.elsevier.com/S1359-4311(17)37939-5/h0185http://refhub.elsevier.com/S1359-4311(17)37939-5/h0190http://refhub.elsevier.com/S1359-4311(17)37939-5/h0190http://refhub.elsevier.com/S1359-4311(17)37939-5/h0190http://refhub.elsevier.com/S1359-4311(17)37939-5/h0195http://refhub.elsevier.com/S1359-4311(17)37939-5/h0195http://refhub.elsevier.com/S1359-4311(17)37939-5/h0195http://refhub.elsevier.com/S1359-4311(17)37939-5/h0200http://refhub.elsevier.com/S1359-4311(17)37939-5/h0200

conditioned room, J. Thermal Sci. Eng. Appl. 8 (2) (2016) 021009-1–021009-7.[41] K.O. Lee, M.A. Medina, E. Raith, X. Sun, Assessing the integration of a thin phase

change material (PCM) layer in a residential building wall for heat transfer re-duction and management, Appl. Energy 137 (2015) 699–706.

[42] W. Youssef, Y.T. Ge, S.A. Tassou, CFD modelling development and experimentalvalidation of a phase change material (PCM) heat exchanger with spiral-wiredtubes, Energy Convers. Manage. 157 (2018) 498–510.

[43] Y. Zhang, G. Zhou, K. Lin, et al., Application of latent heat thermal energy storage inbuildings: State-of-the-art and outlook, Build. Environ. 42 (6) (2007) 2197–2209.

[44] http://www.ruhrtech.com/.[45] J.M. Blanco, P. Arriaga, E. Rojí, et al., Investigating the thermal behavior of double-

skin perforated sheet façades: Part A: Model characterization and validation pro-cedure, Build. Environ. 82 (82) (2014) 50–62.

[46] China Academy of Building Research (CABR). Code for thermal design of civilbuildings (GB50176-2016), China Building Industry Press, Beijing, 2016

[47] P.W. O'Callaghan, S.D. Probert, Sol-air temperature, Appl. Energy 3 (4) (2011)307–311.

[48] Y. Tham, T. Muneer, Sol-air temperature and daylight illuminance profiles for theUKCP09 data sets, Build. Environ. 46 (6) (2011) 1243–1250.

[49] S.D. Zwanzig, Y. Lian, E.G. Brehob, Numerical simulation of phase change materialcomposite wallboard in a multi-layered building envelope, Energy Convers.Manage. 69 (3) (2013) 27–40.

[50] E. Elnajjar, P.C.M. Using, Embedded in building material for thermal management:performance assessment study, Energy Build. 151 (2017).

[51] China Meteorological Administration, Tsinghua University. Special MeteorologicalData set for Building Thermal Environment Analysis of China, China BuildingIndustry Press, Beijing, 2005.

[52] R. Ye, W. Lin, K. Yuan, et al., Experimental and numerical investigations on thethermal performance of building plane containing CaCl2·6H2O/expanded graphitecomposite phase change material, Appl. Energy 193 (2017) 325–335.

[53] A. Faheem, G. Ranzi, F. Fiorito, et al., A numerical study on the thermal perfor-mance of night ventilated hollow core slabs cast with micro-encapsulated PCMconcrete, Energy Build. 127 (2016) 892–906.

[54] B.L. Gowreesunker, S.A. Tassou, Effectiveness of CFD simulation for the perfor-mance prediction of phase change building boards in the thermal environmentcontrol of indoor spaces, Build. Environ. 59 (328) (2013) 612–625.

[55] D. Zhou, G. Shire, Y. Tian, Parametric analysis of influencing factors in PhaseChange Material Wallboard (PCMW), Appl. Energy 119 (2014) 33–42.

L. Zhu et al. Applied Thermal Engineering 138 (2018) 35–47

47

http://refhub.elsevier.com/S1359-4311(17)37939-5/h0200http://refhub.elsevier.com/S1359-4311(17)37939-5/h0205http://refhub.elsevier.com/S1359-4311(17)37939-5/h0205http://refhub.elsevier.com/S1359-4311(17)37939-5/h0205http://refhub.elsevier.com/S1359-4311(17)37939-5/h0210http://refhub.elsevier.com/S1359-4311(17)37939-5/h0210http://refhub.elsevier.com/S1359-4311(17)37939-5/h0210http://refhub.elsevier.com/S1359-4311(17)37939-5/h0215http://refhub.elsevier.com/S1359-4311(17)37939-5/h0215http://www.ruhrtech.com/http://refhub.elsevier.com/S1359-4311(17)37939-5/h0225http://refhub.elsevier.com/S1359-4311(17)37939-5/h0225http://refhub.elsevier.com/S1359-4311(17)37939-5/h0225http://refhub.elsevier.com/S1359-4311(17)37939-5/h0235http://refhub.elsevier.com/S1359-4311(17)37939-5/h0235http://refhub.elsevier.com/S1359-4311(17)37939-5/h0240http://refhub.elsevier.com/S1359-4311(17)37939-5/h0240http://refhub.elsevier.com/S1359-4311(17)37939-5/h0245http://refhub.elsevier.com/S1359-4311(17)37939-5/h0245http://refhub.elsevier.com/S1359-4311(17)37939-5/h0245http://refhub.elsevier.com/S1359-4311(17)37939-5/h0250http://refhub.elsevier.com/S1359-4311(17)37939-5/h0250http://refhub.elsevier.com/S1359-4311(17)37939-5/h0260http://refhub.elsevier.com/S1359-4311(17)37939-5/h0260http://refhub.elsevier.com/S1359-4311(17)37939-5/h0260http://refhub.elsevier.com/S1359-4311(17)37939-5/h0265http://refhub.elsevier.com/S1359-4311(17)37939-5/h0265http://refhub.elsevier.com/S1359-4311(17)37939-5/h0265http://refhub.elsevier.com/S1359-4311(17)37939-5/h0270http://refhub.elsevier.com/S1359-4311(17)37939-5/h0270http://refhub.elsevier.com/S1359-4311(17)37939-5/h0270http://refhub.elsevier.com/S1359-4311(17)37939-5/h0275http://refhub.elsevier.com/S1359-4311(17)37939-5/h0275

Numerical study on the thermal performance of lightweight temporary building integrated with phase change materialsIntroductionPhysical and mathematical modelsPhysical modelGeometric descriptionDescription of the simulation scheme

Mathematical modelAssumptions for the mathematical modelGoverning equationsInitial and boundary conditions

Model validationExperimental setupValidation of numerical procedure

Evaluation indicator of indoor thermal discomfortResults and discussionsLocation effect of PCM layerThickness effect of PCM layerOrientation effect and optimization layout schemeMulti-orientation layout optimization

ConclusionsAcknowledgementReferences