Thermal performance investigation and optimization of buildings with integrated phase change materials and solar photovoltaic thermal collectors

Accepted Manuscript Title: Thermal Performance Investigation and Optimization of Buildings with Integrated Phase Change Materials and Solar Photovoltaic Thermal Collectors Author: Wenye Lin Zhenjun Ma Paul Cooper M. Imroz Sohel Luwei Yang PII: DOI: Reference:

S0378-7788(16)30040-8 http://dx.doi.org/doi:10.1016/j.enbuild.2016.01.041 ENB 6415

To appear in:

ENB

Received date: Revised date: Accepted date:

27-10-2015 14-1-2016 29-1-2016

Please cite this article as: W. Lin, Z. Ma, P. Cooper, M.I. Sohel, L. Yang, Thermal Performance Investigation and Optimization of Buildings with Integrated Phase Change Materials and Solar Photovoltaic Thermal Collectors, Energy and Buildings (2016), http://dx.doi.org/10.1016/j.enbuild.2016.01.041 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Thermal Performance Investigation and Optimization of Buildings with

2

Integrated Phase Change Materials and Solar Photovoltaic Thermal

3

Collectors

4

Wenye Lin1, Zhenjun Ma*,1, Paul Cooper1, M. Imroz Sohel1, Luwei Yang2

Sustainable Buildings Research Centre (SBRC), University of Wollongong, New South Wales, 2522, Australia

6 2

Technical Institute of Physics and Chemistry (TIPC), Chinese Academy of Sciences (CAS),

us

7

cr

1

5

ip t

1

Beijing, 100190, China

9

*Corresponding Author: Phone: 61 02 4221 4143; Email: [email protected]

an

8

10

Abstract: This paper presents the thermal performance investigation and optimization of

12

buildings with integrated phase change materials (PCMs) and solar photovoltaic thermal

13

(PVT) collectors. PCMs are embedded into building envelopes to increase local thermal mass

14

while PVT collectors are used to generate both electricity and low grade thermal energy for

15

winter space heating. The thermal performance of a typical Australian house with PVT

16

collectors and three different types of PCMs is simulated and analyzed by comparing with

17

that of the house using PVT collectors only, using PCMs only, and without using PVT

18

collectors and PCMs. The results showed that using PVT collectors and PCMs

19

simultaneously can substantially improve the indoor thermal performance of the house. The

20

Coefficients of Thermal Performance Enhancement (CTPE) of the house using PVT

21

collectors and PCMs of RT18HC, SP21E and SP24E with a thickness of 20 mm were

22

improved to 43.4, 48.8 and 46.2% respectively, compared to that of the house using the

23

PCMs only (-9.1, 2.6 and 0.2% for RT18HC, SP21E and SP24E, respectively). The CTPE of

24

the house can be increased to 70.2% if Taguchi method is used to determine the optimal air

25

flow rate of PVT collectors, thickness of PCM layers, PCM type and additional wall

Ac ce p

te

d

M

11

1

Page 1 of 36

26

insulation. The optimization results also showed that the additional wall insulation of the

27

house was a critical factor affecting the thermal performance of the PCM enhanced buildings

28

with PVT collectors.

ip t

29

Keywords: Photovoltaic thermal collectors; Phase change materials; Thermal performance

31

enhancement; Taguchi method; Optimization

cr

30

us

32

Nomenclature

34

A

heat transfer area (m2)

35

CTPE

Coefficient of Thermal Performance Enhancement

36

cp

specific heat capacity [J/(kg·K)]

37

Gt

total solar radiation (W/m2)

38

h

specific enthalpy (J/kg)

39

hamb

convective heat transfer coefficient between ambient and exterior surface of

M

an

33

d

building envelopes [W/(m2·K)]

40

IAM

incident angle modifier

42

m

mass (kg)

te

41

44

n

45

Nf,s

Ac ce p

air mass flow rate (kg/s)

43

number of observations in a trail test

number of factors which significantly affect the objective response heat flow rate (W)

46

heat flux (W/m2)

47 48

S/N

signal to noise ratio (dB)

49

T

50

Ts

51

Tl

upper temperature limit of the hysteresis region

52

t

time (s)

53

tot

total time of concern (s)

54

W

width (m)

55

x

the coordinate direction along the PVT length

56

y

objective response

temperature (oC)

lower temperature limit of the hysteresis region

2

Page 2 of 36

Greek letters

58

ηp

electrical efficiency

59

ρref

reflectance of PV panel

60

φ

weighting factor

61

Subscripts

62

amb

ambient

63

c

convective

64

cooling

cooling curve of PCM h-T relationship

65

eq

equivalent

66

f

working fluid

67

f-r

direction from the working fluid to the roof insulation

68

heating

heating curve of PCM h-T relationship

69

L

heat loss

70

l

liquid

71

opt

optimal level

72

p

PV panel

73

pre

prediction

74

p-f

direction from PV panel to working fluid

75

p-r

direction from PV panel to roof insulation

76

r

roof insulation, or long wave radiation

77

rs

roof space

78

r-rs

direction from the roof insulation to the roof space

79

s

solid

80

Superscripts

81

k

83

cr us

an M d

te

Ac ce p

82

ip t

57

time step

1. Introduction

84

Energy consumption of the building sector represents over 40% of the primary energy

85

usage in many International Energy Agency (IEA) member countries [1]. A significant

86

proportion of this consumption was due to the ever growing demand for better indoor thermal

87

comfort [2, 3]. With the population increase and economic growth, energy use in buildings is

88

predicted to continue to increase [2]. Many efforts have therefore been made on the 3

Page 3 of 36

89

development and deployment of low energy technologies and cost effective solutions to

90

promote building energy efficiency and sustainability [4-7]. Solar photovoltaic thermal (PVT) collectors, which can generate electricity and low grade

92

thermal energy simultaneously, have attracted increasing attention [8, 9]. The utilization of

93

the low grade thermal energy generated by PVT collectors may provide an alternative

94

solution for effective space heating and other functional purposes, and can thereby reduce

95

building energy consumption. Air-based PVT collectors can generate hot air which can be

96

used for direct space heating or can be ducted into Heating, Ventilation and Air Conditioning

97

(HVAC) units for pre-conditioning the ventilation air. Shahsavar et al. [10], for instance,

98

proposed a building integrated photovoltaic thermal (BIPVT) setup to pre-heat the incoming

99

ventilation air prior to entering the air handling unit (AHU) in winter and to cool the PV

100

panels by using the exhaust air from the building in summer. A PVT-heat pump hybrid system

101

was described by Bazilian et al. [11], in which the heated air from the PVT collectors was

102

used as the heat source of the heat pump in winter. Kamthania et al. [12] evaluated the

103

performance of a hybrid PVT double pass facade for space heating. The results showed that

104

the room temperature in a typical winter day was 5-6oC higher than the ambient air

105

temperature. However, the use of PVT alone is hard to continuously control building indoor

106

thermal comfort as solar energy is intermittent [11].

cr

us

an

M

d

te

Ac ce p

107

ip t

91

To overcome the intermittence of solar energy, the integration of phase change materials

108

(PCMs) with PVT collectors may provide an alternative approach to maximizing the

109

utilization of solar energy and improving the building thermal performance. PCMs with high

110

energy storage densities and the characteristics to store thermal energy at a relatively constant

111

temperature have been recognized as one of sustainable and environmentally-friendly

112

technologies to reduce building energy consumption and provide better indoor thermal

113

comfort [13, 14]. PCMs can be incorporated with building envelopes to increase building

4

Page 4 of 36

thermal mass to reduce the indoor temperature fluctuations of passive buildings and HVAC

115

energy consumption of active buildings [15, 16]. PCMs can also be integrated with building

116

HVAC systems as thermal energy storage units to provide functional purposes and enhance

117

the system efficiency [17, 18]. For instance, using PCMs in brick constructive solutions for

118

passive cooling was experimentally studied by Castell et al. [15]. Zhu et al. [16] reported the

119

optimal control of air conditioned buildings with envelopes enhanced by PCMs. A PCM-

120

based variable air volume (VAV) air conditioning system was proposed by Parameshwaran et

121

al. [17], in which a storage tank with spherical PCM encapsulations was used to reduce the

122

peak electricity consumption of the air conditioning system.

an

us

cr

ip t

114

A number of studies have also reported the use of PCMs beneath the PV panels for

124

thermal regulation [19, 20], and the integration of PCMs with PVT collectors for space air

125

conditioning [20, 21]. Huang [20], for example, examined the effect of using two PCMs on

126

the thermal regulation of a building integrated PV system. The results showed that the solar-

127

to-electrical conversion efficiency was improved due to the use of PCMs. Fiorentini et al. [21]

128

reported the development and implementation of an innovative air conditioning system

129

integrated with an air-based PVT collector and a PCM thermal storage. The thermal energy

130

collected from the PVT collector can be temporarily stored in the PCM storage and used later

131

for space air conditioning or can be directly used to condition the indoor space or

132

precondition the ventilation air. A ceiling ventilation system integrated with PVT collectors

133

and PCMs was studied by Lin et al. [22], in which the PCM was integrated into the ceiling as

134

part of the ceiling insulation and, at the same time, as a centralized thermal storage to

135

temporarily store the low grade thermal energy generated by the PVT collectors. The results

136

obtained from the above studies showed that PVT and PCM integrated systems are effective

137

in building thermal management if they are appropriately designed and optimized.

138

Ac ce p

te

d

M

123

Many studies have demonstrated that appropriate optimization is essential to maximize

5

Page 5 of 36

the performance of various engineering systems [23, 24]. Taguchi method as an experimental

140

optimization technique using the orthogonal array to form a matrix of experiments for robust

141

design [25, 26], has been extensively used for experimental design and determination of the

142

optimal combination of multiple influential factors. For instance, Kuo et al. [27] used

143

Taguchi method to optimize the multiple design parameters of a flat-plate solar collector

144

process for manufacturing. Yi et al. [28] optimized the building configuration in the

145

schematic phase by using Taguchi method to achieve maximum energy and emergy benefits.

146

Taguchi method was also used for parameter optimization of a solar assisted ground source

147

heat pump system [25, 26].

an

us

cr

ip t

139

Although there are many studies in the public domain on the use of PCMs in building

149

envelopes and PVT collectors in buildings, it seems that the potential benefits of using PVT

150

collectors in buildings with envelopes enhanced by PCMs have not been reported. This paper

151

therefore presents the investigation on the use of the heated air derived from PVT collectors

152

(named as PVT ventilation hereafter) to enhance the thermal performance of passive

153

buildings (i.e. without using air-conditioning systems and/or other conventional heating

154

devices) with PCMs integrated in the envelopes (named as PCM enhanced buildings in this

155

study). Based on the results of the thermal performance investigation, Taguchi method is used

156

to further optimize the performance of PCM enhanced buildings with PVT ventilation by

157

identifying the near optimal levels of the influential factors, such as PVT air flow rate, PCM

158

type and PCM thickness. The thermal performance investigation and optimization were

159

carried out based on a typical Australian house using TRNSYS simulation program [29].

160

2. Building Description and Research Methodology

161

2.1 Building description

Ac ce p

te

d

M

148

162

The case building concerned in this study is one of the ‘typical’ Australian dwelling

163

designs that were considered as being ‘representative’ of the Australian residential building

6

Page 6 of 36

stock since 5 and 6 star regulations were introduced [30]. It is assumed that there are no air-

165

conditioning systems and/or other conventional heating devices used in this case building.

166

The floor plan of the house is shown in Fig. 1. The major specifications of the house

167

envelopes are summarized in Table 1.

ip t

164

In this study, it is also assumed that the air-based PVT collectors with a total area of 110

169

m2 were installed on the whole north roof of the house with a roof angle of 22.5o, while the

170

PCM layers were integrated into all the walls and ceiling of the house, as illustrated in Fig. 2.

171

During the winter daytime, the heated air from the PVT collectors is directed into the house

172

for space heating if the air temperature is higher than the indoor temperature. While during

173

the night-time, the fan of the PVT system is switched off. As the major benefits of using the

174

low grade heat derived from air-based PVT collectors is for space heating, the thermal

175

performance analysis of the PCM enhanced building with PVT ventilation therefore focused

176

on the winter heating conditions in this study, although night-time radiative cooling effect of

177

PVT collectors in summer can also improve the thermal performance of PCM enhanced

178

buildings.

179

2.2 Outline of the research method

us

an

M

d

te

Ac ce p

180

cr

168

The overall procedure employed to evaluate and optimize the thermal performance of

181

PCM enhanced buildings with PVT ventilation is illustrated in Fig. 3. Based on the modelling

182

system setup, a range of the test cases with different types of PCMs and various thicknesses

183

of the PCM layers, as well as using or without using PVT ventilation were designed. The

184

thermal performance of the house with each case was then simulated and compared with that

185

of the baseline case without using the PVT ventilation and PCMs. According to the overall

186

test results, further optimization using Taguchi method was conducted to improve the thermal

187

benefits by using the PVT ventilation and PCMs in building envelopes. Finally, the

188

confirmation test of using the optimal combination of the factors was conducted. The design

7

Page 7 of 36

189

of the test cases and the descriptions of the key performance indicator and Taguchi method

190

are briefly presented in below sections.

191

2.3 Design of the test cases A range of test cases, as summarized in Table 2, were first designed and tested to

193

compare and evaluate the thermal performance of the PCM enhanced buildings with or

194

without using the PVT ventilation. In these test cases, three different types of PCMs and four

195

different thicknesses of the PCM layers were used. The PCMs considered were commercial

196

PCMs from Rubitherm [31], including the organic PCM of RT18HC, and the inorganic PCMs

197

of SP24E and SP21E.

an

us

cr

ip t

192

The thermo-physical properties of the three types of PCMs used are summarized in

199

Table 3. The phase change temperature ranges of these three selected PCMs are close to the

200

indoor thermal comfort temperature range. SP24E represents the PCMs whose phase change

201

temperature is close to the upper limit of the thermostat temperature setting, and RT18HC has

202

a phase change temperature close to the lower limit of the thermostat temperature setting. The

203

phase change temperature of SP21E is close to the mean value of the upper and lower limits

204

of the thermostat setting temperature. The thicknesses of the PCM layers tested were 5, 10,

205

20 and 30 mm, respectively. In these test cases, the air mass flow rate of the PVT collectors

206

used was 2000 kg/h. It is noteworthy that this study was mainly focused on the thermal

207

performance enhancement of the house in terms of the indoor temperature variation. The

208

effect of the air velocity on human thermal comfort was not considered.

209

2.4 Description of the key performance indicator

Ac ce p

te

d

M

198

210

In order to evaluate the thermal performance of the PCM enhanced building with the

211

PVT ventilation, a performance indicator, named as Coefficient of Thermal Performance

212

Enhancement (CTPE), as shown in Eq. (1), slightly modified from a previous study [22], was

213

used to facilitate the performance analysis. The thermal performance of the house without

8

Page 8 of 36

using the PVT ventilation and PCMs was used as the baseline condition and its CTPE was set

215

as zero. It is worthwhile to note that only the lower limit of the indoor thermal comfort setting

216

was used for the winter space heating conditions. However, the use of PVT ventilation may

217

cause the overheating of the house in winter, which was not considered in this study. The

218

integration term in the equation indicates that the sum of the shaded area below the lower

219

limit of the thermal comfort temperature setting (Tsetting,low) illustrated in Fig. 4. tot

220

indoor

0

tot

indoor

0

]

− Tsetting,low ) }dt

]

− Tsetting,low ) }dt

baseline

us

∫ {− min[0, (T CTPE = 1 − ∫ {− min[0, (T

cr

ip t

214

(1)

where, Tindoor is the indoor air temperature, t is the time, and tot is the total time of concern.

222

2.5 Description of Taguchi method

an

221

Taguchi method is used to design the matrix of experiments for robust design of a system

224

or process. Analysis of variance (ANOVA) and signal-to-noise (S/N) ratio are used for data

225

analysis [32]. ANOVA is commonly applied to investigate the response significance of

226

individual factors and identify the percentage contribution of each factor to the objective

227

function. Based on the factors with high percentage contributions, the optimal performance of

228

the system or process can be predicted by using Eq. (2) [33]. The sum of squares of factors,

229

pure sum of squares of factors, variance of factors and percentage contribution of factors are

230

the key statistic parameters used in ANOVA. The details on how to calculate these parameters

231

can be found in Ref. [34].

Ac ce p

te

d

M

223

f ,s ) y pre = y + ∑ ( yopt,i − y )

N

232

(2)

i =1

233

where,

is the predicted response of the objective function, Nf,s is the number of the

234

factors which significantly affect the objective response,

235

responses of all experiments,

236

factor at the corresponding optimal level. The importance of the factors is ranked by the

is the mean value of the objective

is the mean value of the objective responses of the ith

9

Page 9 of 36

237

percentage contribution. In Taguchi method, S/N ratio, a statistic for performance measurement combining the

239

information of mean and variance [35], is often used to identify the most robust combination

240

of the levels of individual factors. The combination of the factor levels is regarded as the near

241

optimal parameters for the system or process of concern, according to which, the near optimal

242

system or process performance can be achieved. In this study, the Coefficient of Thermal

243

Performance Enhancement is used as the quality characteristic, and the higher-the-better, as

244

illustrated in Eq. (3), is selected to calculate the S/N ratio.

cr

us

1 n 1  1 n 1  S / N = −10 log ∑ 2  = −10 log ∑  n j =1 y   n j =1 CTPE 2  j  j   

an

245

ip t

238

(3)

where, y is the objective response, n is the number of the observations in a trail test, and the

247

subscript j is the jth observed value of the CTPE. The importance of the factors can be ranked

248

by the difference between maximal and minimal S/N ratios of different factor levels.

249

3. Model Description

250

3.1 House model and PVT model

d te

To simplify the simulation, the house was modelled with two thermally separated zones

Ac ce p

251

M

246

252

(i.e. roof space and room space) using Google SketchUp and then imported into the TRNSYS

253

Type 56 Multi-zone building component model.

254

The PVT model used was a slightly updated dynamic model developed in a previous

255

study [36]. The key governing equations used are presented in Eqs. (4)-(6). By using this

256

PVT model, the outlet air temperature, the PV panel temperature and roof insulation

257

temperature, electrical power generation and thermal energy collected can be easily

258

determined.

259

m& c p, f

dTf dx

= Wp (q& p− f − q& f −r )

(4)

10

Page 10 of 36

260

mp c p, p

261

mr c p , r

dTp dt

= (1− ρref )⋅ IAM ⋅ (1 −η p )⋅ A ⋅ Gt − Q& L − Q& p− f − Q& p−r

(5)

dTr = Q& p −r + Q& f −r − Q& r −rs dt

(6)

where,

is the air mass flow rate, cp is the specific heat capacity, m is the mass, A is the heat

263

transfer area, ρref is the reflectance of the PV panel, IAM is the incident angle modifier, Gt is

264

the total solar radiation on the PV panel, ηp is the electrical efficiency, W is the width of the

265

PVT collector,

266

the PVT length, the subscripts f, p, r, L and rs indicate the working fluid, PV panel, roof

267

insulation, heat loss and roof space, and subscripts p-f, p-r, f-r, r-rs indicate the heat transfer

268

directions from the PV panel to the working fluid, from the PV panel to the roof insulation,

269

from the working fluid to the roof insulation, and from the roof insulation to the roof space,

270

respectively.

271

3.2 Envelope-integrated PCM model

cr

ip t

262

is the heat flow rate, x is the coordinate direction along

d

M

an

us

is the heat flux,

In order to integrate the PCM model within the house model, the corresponding

273

envelopes in the house model were reduced to a single layer of the plasterboard, while the

274

brick (or the external plasterboard layer of the ceiling) and additional insulation layer of the

275

original house model were transplanted into the PCM model, as illustrated in Fig. 5. The

276

PCM model integrated with the transplanted brick layer (or the extra plasterboard of the

277

ceiling) and additional insulation layer is named as envelope-integrated PCM model hereafter.

278

The envelope-integrated PCM model was developed using the finite difference method based

279

on an enhanced enthalpy method, in which the heat transfer from outdoor and the heat flux

280

through the building plasterboard layer were used as the exterior and interior boundary

281

conditions, respectively. The governing equations of the energy balance can be obtained and

282

discretised, based on the assumptions employed below.

283

Ac ce p

te

272

The dominant heat transfer within the envelopes is one-dimensional thermal conduction; 11

Page 11 of 36

The shrinkage cavity caused due to the density change during the phase change process is

285

considered to be evenly distributed within each PCM element [37];

286

The additional insulation layer is considered as a massless layer with the pure thermal

287

resistance as only R-value of the additional insulation was provided in the ‘typical’

288

Australian passive dwelling designs [30];

289

The PCM hysteresis phenomenon was simulated based on the method presented by Bony

290

and Cithelet [38] with separated heating and cooling curves to represent the h-T

291

relationships. However, the equivalent specific heat capacity (cp,eq) within the hysteresis

292

region is considered to be varied with the PCM thermal dynamic state, as shown in Fig. 6.

293

The liquid and solid specific heat capacities of the PCM beyond the phase change region

294

are assumed to be constant.

M

an

us

cr

ip t

284

Based on the assumptions, the relationship between the enthalpy and temperature of the

296

PCM at the time step k+1 can be determined by Eq. (7), while the enthalpy of the PCM at the

297

time step k+1 can be determined based on the energy balance using the PCM thermodynamic

298

state at the time step k.

te

d

295

Ac ce p

( )

299

( )

 if h k = hheating T k , T k +1 ≥ T k , Ts < T k < Tl k +1 hheating T ,  or if T k ≥ Tl   if h k = hheating T k , T k +1 < T k , Ts < T k < Tl    h k +1 = c kp,eq T k +1 − T k + h k , or if hheating(T k ) < h k < hcooling(T k ) , Ts < T k < Tl   k k k +1 k k or if h = hcooling T , T > T , Ts < T < Tl   if h k = hcooling T k , T k +1 ≤ T k , Ts < T k < Tl  k +1 hcooling T ,  or if T k ≤ Ts 

(

( )

)

( )

( )

(7)

( )

300

where, h is the specific enthalpy, hheating and hcooling represent the heating curve and cooling

301

curve respectively, cp,eq is the equivalent specific heat capacity within the hysteresis region, Ts

302

and Tl are the lower and upper temperature limits of the hysteresis region respectively, the

303

superscript k represents the time step k, and the subscripts s and l indicate the solid and liquid,

12

Page 12 of 36

305 306 307

respectively. The equivalent specific heat capacity is calculated by Eq. (8), in which the weighting factor (i.e. φ) is related to the PCM thermodynamic state and can be determined by Eq. (9).

ckp,eq = ϕ k ⋅ c p,s + (1 − ϕ k )⋅ c p,l , if hheating(T k ) < h k < hcooling(T k ) , Ts < T k < Tl

(8)

ip t

304

308

ϕ k = [c p,l T k + (hl − c p ,l Tl ) − h k ]/[c p,l T k + (hl − c p ,l Tl ) − c p , sT k ]

309

For a given PCM, its heating and cooling h-T relationships can be obtained from

311 312

4. Performance Testing and Discussions

313

4.1 Setup of the test

cr

Differential Scanning Calorimetry (DSC) test data.

an

us

310

(9)

As the living and sleeping spaces of the house were combined into one single air-node

315

(i.e. the room space) in the house model, the higher value of the lower limit of the thermal

316

comfort temperature settings for the living space and sleeping space defined in Australian

317

National House Energy Rating Scheme [39] for winter conditions was used as the lower

318

temperature requirement to maintain the thermal comfort of the house, as shown in Fig. 7.

te

d

M

314

The test results from the whole winter period (i.e. from June to August) in Sydney

320

weather conditions were used to analyze the Coefficient of Thermal Performance

321

Enhancement of the house while the test results from seven consecutive winter days were

322

selected to illustrate the indoor temperature variations of the house due to the use of PCMs

323

and PVT collectors for space heating. The solar irradiations, ambient air temperatures and

324

outdoor air humidity ratios in these seven selected test days are illustrated in Fig. 8.

325

4.2 Validation of the envelope-integrated PCM model

Ac ce p

319

326

The experimental data reported by Kuznik and Virgone [40] was used to validate the

327

effectiveness of the envelope-integrated PCM model. In their experiments, the inside air

328

temperature at the center of the enclosure was measured by a PT100 sensor with a calculated

329

resolution of ±0.25oC. A model for the cubical enclosure integrated with the PCM tested in 13

Page 13 of 36

Ref. [40] was developed for model validation. The long wave radiations among the interior

331

walls of the enclosure, and between the exterior surface of the enclosure and the internal

332

walls of the climatic chamber, were considered as the extra heat flux on the boundaries of the

333

envelope-integrated PCM model. Fig. 9 shows the validation results. It can be seen that the

334

simulation data agreed well with the experimental data reported in [40] with relatively

335

smaller deviations when compared to the results using a single h-T relationship for both

336

heating and cooling processes presented in Kuznik et al. [41]. The maximum deviation

337

between the model prediction and the experimental data reported in [40] was around 0.4oC

338

and the root mean square of the deviation was around 0.2oC when the sinusoidal temperature

339

setting was applied for the climatic chamber. The validation results indicated that this

340

envelope-integrated PCM model can simulate the phase change process of PCMs with an

341

acceptable accuracy.

342

4.3 Thermal performance of the house using the PCMs only

d

M

an

us

cr

ip t

330

The Coefficients of Thermal Performance Enhancement (CTPEs) of the house by only

344

using the different types of PCMs with different thicknesses (i.e. Test cases 2-5) were

345

compared with that of the original house without using the PVT ventilation and PCMs (i.e.

346

Test case 1) and the results are summarized in Table 4. The CTPEs of the house by using

347

RT18HC were -8.6, -13.6, -9.1 and -5.3% when the thicknesses of the PCM layer were 5, 10,

348

20 and 30 mm, respectively. The negative values of the CTPE indicated a decrease in indoor

349

thermal performance of the house as compared to that of the original house. This is mainly

350

due to the non-existence of air-conditioning systems and/or other conventional heating

351

devices in the house. Consequently,

352

daytime periods and night-time periods was lower than that of the thermal comfort

353

temperature setting, and the average temperature increase during the night-time was less than

354

the average temperature decrease during the daytime due to the use of PCMs.

Ac ce p

te

343

the indoor temperature of the house during some

14

Page 14 of 36

The CTPEs of the house by using SP21E were -0.6, 0.7, 2.6 and 4.7% when the

356

thicknesses of the PCM layer were 5, 10, 20 and 30 mm, respectively. Similar values can also

357

be observed when SP24E was applied. Compared to RT18HC, the indoor thermal

358

performance was slightly improved. The above results indicated that using PCMs alone

359

cannot improve the indoor thermal performance of the house under Sydney winter heating

360

conditions, which coincided with the conclusion reported by Guichard et al. [42].

cr

ip t

355

The indoor temperatures of the house by using the PCMs only and that of the original

362

house without using the PVT ventilation and PCMs (i.e. Test case 1) under the seven selected

363

consecutive winter days are shown in Fig. 10. To save the page size, the results for the

364

thicknesses of the PCM layer of 5 and 30 mm (i.e. Test case 2 and Test case 5) are presented

365

only. Compared to the baseline case (i.e. Test case 1), the indoor temperature fluctuations

366

decreased due to the use of the PCMs in the building walls and ceiling. However, there was

367

no clear difference in the indoor temperature during the night-time when the three different

368

types of PCMs with a thickness of 5 mm were used.

te

d

M

an

us

361

Difference among the indoor temperatures by using SP24E, SP21E and RT18HC can be

370

observed when the thickness of the PCM layer increased to 30 mm. A clear discharging

371

process during the night-time can be observed when RT18HC was used as the indoor

372

temperature obviously increased when compared to that of the original house. However, the

373

storage capacity of SP24E may not be fully utilized due to its higher phase change

374

temperature. It is worthwhile to mention that as the phase change temperature of RT18HC

375

was lower than that of the thermal comfort temperature setting specified for the daytime, the

376

contribution of RT18HC for the indoor thermal performance enhancement was therefore

377

mainly from the night-time.

378

4.4 Thermal performance of the house using PCMs and PVT ventilation simultaneously

379

Table 5 summarizes the CTPEs of the house by using the PVT ventilation and PCMs

Ac ce p

369

15

Page 15 of 36

with different thicknesses (i.e. Test cases 7-10), using the PVT ventilation only (i.e. Test case

381

6), and that of the baseline case of the original house (i.e. Test case 1). The CTPE of the

382

house increased significantly when the PVT ventilation was used. Compared to the use of

383

RT18HC and SP24E, higher CTPEs of the house were achieved when SP21E was used. The

384

resulted CTPEs of the house were 43.3, 46.3, 48.8 and 53.1% when the thicknesses of SP21E

385

were 5, 10, 20 and 30 mm, respectively. The use of RT18HC showed a better performance in

386

terms of the CTPE (i.e.52.1%) than that of using SP24E (i.e. 48.4%) when the thickness of

387

the PCM layer was 30 mm. The use of the PVT ventilation and PCMs simultaneously can

388

greatly improve the indoor thermal performance, when compared with the house using the

389

PCMs only and using the PVT ventilation only. The above results indicated that the PVT

390

ventilation can effectively enhance the PCM storage performance through charging more

391

thermal energy into the PCMs during the daytime and using it later during the night-time for

392

space heating in winter.

d

M

an

us

cr

ip t

380

The comparison of the indoor temperatures of the house by using the PVT ventilation

394

only (i.e. Test case 6) and using the PVT ventilation and PCMs simultaneously with that of

395

the baseline case of the original house (i.e. Test case 1) under the same seven winter test days

396

are presented in Fig. 11. For the same reason, the results with the thicknesses of the PCM

397

layer of 5 and 30 mm (i.e. Test case 7 and Test case 10) are presented only. Compared to the

398

results presented in Fig. 10 by using the PCMs only, the indoor temperature of the house

399

increased significantly during the daytime when the heated air from the PVT collectors was

400

supplied to the house. However, during the night-time, the indoor temperature was only

401

slightly higher than that of using the PCMs only when the thickness of the PCM layer was 5

402

mm, due to the limited storage capacity of the PCMs used. When the PCM thickness

403

increased to 30 mm, an obvious increase in the indoor temperature can be observed during

404

the night-time as compared to that of using the PCMs only. However, SP24E still could not

Ac ce p

te

393

16

Page 16 of 36

be fully charged during the daytime even with the PVT ventilation and its thermal benefit to

406

improve the indoor temperature during night-time was therefore limited.

407

In order to understand the thermodynamic state of the PCMs, the surface temperatures of the

408

PCM layer close to the plasterboard of the house model with the PVT ventilation are

409

presented and illustrated in Fig. 12. Only the results for the north wall, south wall and ceiling

410

with the PCM thickness of 20 mm are provided, as the north wall and south wall represent the

411

sunny side and shady side of the building envelopes respectively, and the ceiling of the house

412

was well insulated in comparison to the walls of the house (see Table 1). It can be seen that a

413

higher PCM surface temperature can be found in the ceiling when RT18HC was used. In the

414

wall, the temperature fluctuation was relative small when RT18HC was used, as compared to

415

that using SP21E and SP24E. This is because that a large amount of heat was charged in and

416

discharged from RT18HC at around its phase change temperature of 18oC, which is lower

417

than the phase change temperatures of SP21E and SP24E. Smaller PCM surface temperature

418

fluctuations in the ceiling were resulted when SP21E and SP24E were used, as compared to

419

that in the north wall and south wall, due to the better insulation of the ceiling. More clear

420

PCM charging and discharging processes can be found in the north wall and ceiling when

421

SP21E and SP24E were used, as compared to that of the south wall. This is due to the fact

422

that the north wall exposed to strong solar radiation during the daytime and more heat was

423

therefore charged into the PCM layers which reduced the temperature fluctuations while the

424

ceiling was better insulated. The above results illustrate that the insulation of the building

425

envelopes plays an important role in ensuring the effectiveness of the PCMs used.

Ac ce p

te

d

M

an

us

cr

ip t

405

426

Based on the performance evaluation above, it is shown that using PVT ventilation can -

427

improve the CTPE of PCM enhanced buildings under Sydney winter heating conditions. The

428

thickness of the PCM layer and PCM types should be selected carefully to make full use of

429

the PCM thermal storage capacity. The optimization of the additional wall insulation of the

17

Page 17 of 36

430

original house may also assist in improving the performance of PCM enhanced buildings.

431

4.5 Optimization by using Taguchi method Based on the results from thermal performance investigation, further optimization using

433

Taguchi method was designed and conducted in order to achieve a higher CTPE through fully

434

utilizing the advantages of PVT ventilation and PCMs. The air flow rate of the PVT

435

ventilation, thickness of the PCM layer, PCM types and additional wall insulation are

436

considered as the control factors in the optimization, as shown in Table 6. Three levels were

437

considered for each individual factor. The air flow rates of the PVT collectors used were 1000,

438

2000 and 3000 kg/h and the thicknesses of the PCM layer were 10, 20 and 30 mm,

439

respectively. Three PCMs considered were RT18HC, SP21E and SP24E. The additional wall

440

insulation of the house used were 1.0, 2.0 and 3.0 m2·K/W, respectively. As there are four

441

control factors and three factor levels, Standard L9 (34) Taguchi Orthogonal Array was

442

selected to design the matrix of experiments, as shown in Table 7. The resulted CTPEs and

443

S/N ratios of each trail test are also provided in Table 7.

te

d

M

an

us

cr

ip t

432

The mean S/N ratios for each level of the control factors are summarized in Table 8. The

445

best combination of the control factor levels can be identified by selecting the levels with

446

highest S/N ratios. The identified optimal levels for the air flow rate of the PVT ventilation,

447

the thickness of the PCM layer, PCM type and additional wall insulation were 3000 kg/h, 30

448

mm, SP21E and 3.0 m2·K/W, respectively. To further investigate the significance of each

449

control factor to the objective response, the ANOVA was conducted and the results are

450

presented in Table 9. It can be seen that the additional wall insulation of the house and the

451

thickness of the PCM layer were the key factors which significantly influence the indoor

452

thermal performance of the PCM enhanced buildings. The percentage contributions of the

453

house additional wall insulation and the thickness of the PCM layer were 56.6% and 18.4%,

454

respectively.

Ac ce p

444

18

Page 18 of 36

The predicted objective response using Eq. (2) was 63.3%, when only considering the

456

significant factors of the additional wall insulation and the thickness of the PCM layer. A

457

conformation test was then conducted by using the optimal factor levels identified in Table 8.

458

The resulted CTPE of the house was 70.2%, which was higher than the predicted value due to

459

the consideration of all factors with their optimal levels. It was also much higher than the

460

highest CTPE presented in Table 5 (i.e. 53.1% in Test case 10 with SP21E PCM), indicating

461

the effectiveness of the optimization carried out.

462

5. Conclusions

us

cr

ip t

455

This paper investigated the thermal performance of buildings with integrated air-based

464

solar photovoltaic thermal collectors and phase change materials. The thermal energy derived

465

from the PVT collectors was used for space heating in winter conditions while PCMs were

466

laminated onto the building envelopes to increase local thermal mass. The thermal

467

performance of a typical Australian house with three different types of PCMs (i.e. organic

468

PCM of RT18HC, and the inorganic PCMs of SP24E and SP21E) and PVT collectors was

469

simulated and compared with that of the house using the PVT ventilation only, using the

470

PCMs only, and without using the PVT ventilation and PCMs.

M

d

te

Ac ce p

471

an

463

The results showed that using the PCMs in building envelops can reduce the indoor

472

temperature fluctuations. However, using the PCMs alone showed limited even no benefits to

473

the indoor thermal performance enhancement under Sydney winter heating conditions. The

474

Coefficients of Thermal Performance Enhancement (CTPEs) of the house due to the use of

475

RT18HC, SP21E and SP24E with the thickness of 20 mm were -9.1, 2.6 and 0.2%,

476

respectively. The use of the heated air from the PVT collectors can significantly increase the

477

indoor air temperature of the house during the winter daytime. The use of the PVT ventilation

478

and PCMs can substantially increase the indoor thermal performance of the house. The

479

CTPEs of the house increased to 43.4, 48.8, and 46.2% when the PVT ventilation with an air

19

Page 19 of 36

480

flow rate of 2000 kg/h and the PCMs of RT18HC, SP21E and SP24E with a thickness of 20

481

mm were used. The CTPE of the house was further improved through optimization using Taguchi

483

method. The optimal CTPE of the house was 70.2% when the PVT air flow rate was 3000

484

kg/h, the PCM thickness was 30 mm, the PCM type was SP21E and the additional wall

485

insulation was 3.0 m2·K/W, respectively. The additional wall insulation of the house was the

486

critical factor affecting the thermal performance of the PCM enhanced building with PVT

487

ventilation. Although both additional wall insulation and PCM layers can improve the

488

building thermal performance, additional wall insulation cannot replace the PCM layers as

489

PCMs can store a large amount of thermal energy and use it later and can therefore greatly

490

reduce the indoor temperature fluctuations.

M

an

us

cr

ip t

482

491

References

493

[1] IEA, ‘Modernising building energy codes to secure our global energy future’,

494

International Energy Agency, , [accessed 04.11.13].

495

[2] Perez-Lombard L., Ortiz J. and Pout C. 2008, ‘A review on buildings energy consumption

496

information’, Energy and Buildings, vol.40, pp.394-398.

497

[3] Lam J.C., Wan K.K.W., Tsang C.L., and Yang L. 2008, ‘Building energy efficiency in

498

different climates’, Energy Conversion and Management, vol.49, pp.2354-2366.

499

[4] Buker M.S. and Riffat S.B. 2015, ‘Recent developments in solar assisted liquid desiccant

500

evaporative cooling technology—A review’, Energy and Buildings, vol.96, pp.95-108.

501

[5] Ma Z.J. and Wang S.W. 2011, ‘Enhancing the performance of large primary-secondary

502

chilled water systems by using bypass check valve’, Energy, vol.36, pp.268-276.

503

[6] Kaln s S. and Jelle B.P. 2015, ‘Phase change materials and products for building

504

applications: A state-of-the-art review and future research opportunities’, Energy and

Ac ce p

te

d

492

20

Page 20 of 36

Buildings, vol.94, pp.150-176.

506

[7] Niu X.F., Xiao F. and Ma Z.J. 2012, ‘Investigation on capacity matching in liquid

507

desiccant and heat pump hybrid air-conditioning systems’, International Journal of

508

Refrigeration, vol.35, pp. 160-170.

509

[8] Zondag H.A. 2008, ‘Flat-plate PV-thermal collectors and systems: a review’, Renewable

510

and Sustainable Energy Review, vol.12, pp.891-959.

511

[9] Kumar R. and Rosen M.A. 2011, ‘A critical review of photovoltaic–thermal solar

512

collectors for air heating’, Applied Energy, vol.88, pp.3603-3614.

513

[10] Shahsavar A., Salmanzadeh M., Ameri M. and Talebizadeh P. 2011, ‘Energy saving in

514

building by using the exhaust and ventilation air for cooling of photovoltaic panels’, Energy

515

and Buildings, vol.43, pp.2219-2226.

516

[11] Bazilian M.D., Leenders F., van der Ree B.G.C. and Prasad D. 2001, ‘Photovoltaic

517

cogeneration in the building environment’, Solar Energy, vol.71, pp.57-69.

518

[12] Kamthania D. Nayak S. and Tiwari G.N. 2011, ‘Performance evaluation of a hybrid

519

photovoltaic thermal double pass facade for space heating’, Energy and Buildings, vol.43,

520

pp.2274-2281.

521

[13] Soares N., Costa J.J., Gaspar A.R. and Santos P. 2013, ‘Review of passive PCM latent

522

heat thermal energy storage systems towards buildings’ energy efficiency’, Energy and

523

Buildings, vol.59, pp.82-103.

524

[14] Zhu N., Wang S.W., Xu X.H. and Ma Z.J. 2010, ‘A simplified dynamic model of

525

building structures integrated with shaped-stabilized phase change materials’, International

526

Journal of Thermal Science, vol.49, pp.1722-1731.

527

[15] Castell A., Martorell I., Medrano M., Perez G. and Cabeza L.F. 2010, ‘Experimental

528

study of using PCM in brick constructive solutions for passive cooling’, Energy and

529

Buildings, vol.42, pp.534-540.

Ac ce p

te

d

M

an

us

cr

ip t

505

21

Page 21 of 36

[16] Zhu N., Wang S.W., Ma Z.J. and Sun Y.J. 2011, ‘Energy performance and optimal

531

control of air-conditioned buildings with envelopes enhanced by phase change materials’,

532

Energy conversion and Management, vol.52, pp.3197-3205.

533

[17] Parameshwaran R., Harikrishnan S., and Kalaiselvam S. 2010, ‘Energy efficient PCM-

534

based variable air volume air conditioning system for modern buildings’, Energy and

535

Buildings, vol.42, pp.1353-1360.

536

[18] Al-Abidi A.A., Mat S., Sopian K., Sulaiman M.Y., Mohammad A.T. 2013, ‘Experimental

537

study of PCM melting in triplex tube thermal energy storage for liquid desiccant air

538

conditioning system’, Energy and Buildings, vol.60, pp.270-279.

539

[19] Ho C.J., Tanuwijava A.O. and Lai C.M. 2012, ‘Thermal and electrical performance of a

540

BIPV integrated with a microencapsulated phase change material layer’, Energy and

541

Buildings, vol.50, pp.331-338.

542

[20] Huang M.J. 2011, ‘The effect of using two PCMs on the thermal regulation performance

543

of BIPV systems’, Solar Energy Mater Solar Cells, vol.95, pp.957-963.

544

[21] Fiorentini M., Cooper P. and Ma Z.J. 2015, ‘Development and optimization of an

545

innovative HVAC system with integrated PVT and PCM thermal storage for a net-zero

546

energy retrofitted house’, Energy and Buildings, vol.94, pp.21-32.

547

[22] Lin W.Y., Ma Z.J., Sohel M. I. and Cooper P. 2014, ‘Development and evaluation of a

548

ceiling ventilation system enhanced by solar photovoltaic thermal collectors and phase

549

change materials’, Energy Conversion and Management, vol.88, pp.218-230.

550

[23] Padovan R. and Manzan M. 2014, ‘Genetic optimization of a PCM enhanced storage

551

tank for solar domestic hot water systems’, Solar Energy, vol.103, pp.563-573.

552

[24] Summer E.K., Antar M.A. and Lienhard V J.H. 2012, ‘Design and optimization of an air

553

heating solar collector with integrated phase change material energy storage for use in

554

humidification-dehumidification desalination’, Solar Energy, vol.86, pp.3417-3429.

Ac ce p

te

d

M

an

us

cr

ip t

530

22

Page 22 of 36

[25] Sivasakthivel T., Murugesan K. and Thomas H.R. 2014, ‘Optimization of operating

556

parameters of ground source heat pump system for space heating and cooling by Taguchi

557

method and utility concept’, Applied Energy, vol.116, pp.76-85.

558

[26] Verma V. and Murugesan K. 2014, ‘Optimization of solar assisted ground source heat

559

pump system for space heating application by Taguchi method and utility concept’, Energy

560

and Buildings, vol.82, pp.296-309.

561

[27] Kuo C.F.J, Su T.L., Jhang P.R., Huang C.Y. and Chiu C.H. 2011, ‘Using the Taguchi

562

method and grey relational analysis to optimize the flat-plat collector process with multiple

563

quality characteristics in solar energy collector manufacturing’, Energy, vol.36, pp.3554-3562.

564

[28] Yi H., Srinivasan R.S. and Braham W.W. 2015, ‘An integrated energy-emergy approach

565

to building form optimization: use of EnergyPlus, emergy analysis and Taguchi-regression

566

method’, Energy and Buildings, vol.84, pp.89-104.

567

[29] Beckman W.A. 2001, ‘TRNSYS reference manual’, vol.2. Solar Energy Laboratory.

568

[30] Australian Government Department of Industry. 2013, ‘Representative dwelling models:

569

industry consultation summary paper for survey participants’.

570

[31] Rubitherm PCMs, < www.rubitherm.com>, [accessed 14.07.2014].

571

[32] Taguchi G., Chowdhury S. and Wu Y. 2004, ‘Taguchi’s quality engineering handbook’,

572

Wiley, New Jersey, USA.

573

[33] Milos M. and Miroslav R. 2013, ‘Application of the Taguchi method for optimization of

574

laser cutting: a review’, Nonconventional Technologies Review, Romania, December, pp.50-

575

57.

576

[34] Roy K.R. 2010, ‘A primer of the Taguchi method’, Society of Manufacturing Engineers,

577

Michigan, USA.

578

[35] Velibor M. and Miloš M. 2011, ‘Optimization of surface roughness in turning alloy steel

579

by using Taguchi method’, Scientific Research and Essays, vol.6, pp.3473-3483.

Ac ce p

te

d

M

an

us

cr

ip t

555

23

Page 23 of 36

[36] Sohel M.l., Ma Z.J., Cooper P., Adams J. and Scott R.A. 2014, ‘A dynamic model for

581

air-based photovoltaic thermal systems working under real operation conditions’, Applied

582

Energy, vol.123, pp.216-225.

583

[37] Shamsundar N. and Sparrow E.M. 1976, ‘Effect of density change on multidimensional

584

conduction phase change’, Journal of Heat Transfer, vol.98, pp.550-557.

585

[38] Bony J. and Citherlet S. 2007, ‘Numerical model and experimental validation of heat

586

storage with phase change materials’, Energy and Buildings, vol.39, pp.1065-1072.

587

[39] National House Energy Rating Scheme, Australia, ,

588

[accessed 15.01.14].

589

[40] Kuznik F. and Virgone J. 2009, ‘Experimental investigation of wallboard containing

590

phase change material: data for validation of numerical modelling’, Energy and Buildings,

591

vol.41, pp.561-570.

592

[41] Kuznik F., Virgone J. and Johannes K. 2010, ‘Development and validation of a new

593

TRNSYS type for the simulation of external building walls containing PCM’, Energy and

594

Buildings, vol.42, pp.1004-1009.

595

[42] Guichard S., Miranville F., Bigot D., Malet-Damour B. and Boyer H., ‘Experimental

596

investigation on a complex roof incorporating phase-change material’, Energy and Buildings

597

vol.108, pp.36-43.

cr

us

an

M

d

te

Ac ce p

598

ip t

580

24

Page 24 of 36

599 600

Table 1 Specifications of the house envelopes [30]

601

Materials used

ip t

12 mm plasterboard Additional wall insulation with a R-value of 1.0 m2K/W 110 mm brick 12 mm plasterboard Additional ceiling insulation with a R-value of 3.0 m2K/W 12 mm plasterboard*

Wall (from interior to exterior)

* An extra plasterboard layer was added to facilitate the use of the PCM model developed.

us

602 603

cr

Ceiling (from interior to exterior)

604

an

Table 2 Summary of the test cases studied

605

Test cases

PVT ventilation used

Without using PVT ventilation and PCMs

1

No

te

Using PVT ventilation only

606 607 608

Ac ce p

Using PVT ventilation and PCMs simultaneously

No

d

Using PCMs only

2 3 4 5 6 7 8 9 10

M

Test cases

Yes Yes

PCM layer thickness (mm) RT18HC SP21E SP24E 0

0

0

5 10 20 30 0 5 10 20 30

5 10 20 30 0 5 10 20 30

5 10 20 30 0 5 10 20 30

Table 3 Thermo-physical properties of the PCMs used [31]

Liquid/solid density (kg/m3) SP24E 1400/1500 SP21E 1400/1500 RT18HC 770/880 PCMs

Thermal conductivity [W/(m·K] 0.6 0.6 0.2

Melting temperature range (oC) 24-25 22-23 17-19

Congealing temperature range (oC) 23-21 21-19 19-17

Specific heat capacity [kJ/(kg·K)] 2.0 2.0 2.0

609 610 611 612 613 25

Page 25 of 36

614 615

Table 4 Coefficients of Thermal Performance Enhancement of the house by using PCMs with different thicknesses PCM layer thickness (mm)

Test case 1 Test case 2 Test case 3 Test case 4 Test case 5

0 5 10 20 30

RT18HC 0 -8.6% -13.6% -9.1% -5.3%

CTPE SP21E 0 -0.6% 0.7% 2.6% 4.7%

SP24E 0 -2.1% -1.1% 0.2% 1.8%

ip t

Test cases

cr

616 617

Table 5 Coefficients of Thermal Performance Enhancement of the house by using the PVT

622

ventilation and PCMs

625

an

PCM layer thickness (mm)

Test case 1 Test case 6 Test case 7 Test case 8 Test case 9 Test case 10

0 0 5 10 20 30

te

d

M

Test cases

Ac ce p

623 624

us

618 619 620 621

CTPE RT18HC SP21E 0 0 38.1% 38.1% 40.0% 43.3% 40.1% 46.3% 43.4% 48.8% 52.1% 53.1%

SP24E 0 38.1% 40.3% 42.8% 46.2% 48.4%

Table 6 Control factors and corresponding levels in Taguchi design

Factor

Level

1 2 3

Air flow rate of PVT collectors (kg/h) 1000 2000 3000

Thickness of PCM layer (mm)

PCM type

10 20 30

SP21E RT18HC SP24E

Additional wall insulation (m2·K/W) 1.0 2.0 3.0

626 627 628 629 630 26

Page 26 of 36

Table 7 Taguchi L9 (34) test plan Thickness of PCM layer (mm)

PCM type

1 2 3 4 5 6 7 8 9

1000 1000 1000 2000 2000 2000 3000 3000 3000

10 20 30 10 20 30 10 20 30

SP21E RT18HC SP24E RT18HC SP24E SP21E SP24E SP21E RT18HC

632 633

S/N ratio -8.643 -4.979 -4.637 -4.836 -6.701 -4.262 -5.656 -4.122 -5.455

Thickness of PCM layer (mm)

CTPE S/N 50.7% -6.086 54.9% -5.267 55.9% -5.084 Level 3

CTPE S/N 48.8% -6.385 54.9% -5.267 57.7% -4.785 Level 3

CTPE S/N 53.5% -5.676 55.7% -5.090 52.3% -5.671 Level 2

Additional wall insulation (m2·K/W) CTPE S/N 45.5% -6.933 56.5% -4.972 59.4% -4.532 Level 3

1.600

0.586

2.401

2

4

1

3

Ac ce p

634

d

1.002

M

Air flow rate of PVT ventilation (kg/h)

te

1 2 3 Optimal S/Nmax-S/Nmin Rank

an

Table 8 Response table of the CTPE

Level

635

Additional wall CTPE insulation (m2·K/W) 1 37.0% 2 56.4% 3 58.6% 3 57.3% 1 46.2% 2 61.2% 2 52.0% 3 62.2% 1 53.4%

ip t

Air flow rate of PVT ventilation (kg/h)

us

Trail test No.

cr

631

PCM types

Table 9 Analysis of variance table

Control factors (i.e. Source) Air flow rate of PVT ventilation Thickness of PCM layer PCM type Additional wall insulation All other/error Total

Degree of freedom

Sum of squares

Variance

Variance ratio

Pure sum of squares

Percentage contribution

(2)

0.0046

Pooled

2

0.0126

0.0063

3.9375

0.0094

18.4%

(2)

0.0018

Pooled

2

0.0322

0.0161

10.0625

0.0290

56.6%

4 8

0.0064 0.0512

0.0016

1.0

0.0128

25.0% 100.0%

636

27

Page 27 of 36

Figure Captions Fig. 1 Floor plan of a typical Australian house concerned in this study. Fig. 2 Illustration of the PCM enhanced house with the PVT ventilation.

ip t

Fig. 3 Overall procedure employed to evaluate and improve the thermal performance of the PCM enhanced building with PVT ventilation.

cr

Fig. 4 Illustration of the working principles expressed in Eq. (1). Fig. 5 Envelope-integrated PCM model.

us

Fig. 6 Enthalpy-temperature relationship.

an

Fig. 7 Thermal comfort temperature settings used in winter conditions.

Fig. 8 Solar irradiations, ambient air temperatures and outdoor air humidity ratios during the

M

seven selected test days.

Fig. 9 Validation of the envelope-integrated PCM model.

d

Fig. 10 Indoor temperatures of the house using and without using the PCMs.

te

Fig. 11 Indoor temperatures of the house using the PVT ventilation and PCMs simultaneously, using the PVT ventilation only and that of the original house without using the PVT

Ac ce p

ventilation and PCMs.

Fig. 12 Surface temperatures of the PCM layers close to the plasterboard of the house model.

28

Page 28 of 36

ip t cr us an M d

Ac ce p

te

Fig. 1 Floor plan of a typical Australian house concerned in this study [30].

Fig. 2 Illustration of the PCM enhanced house with the PVT ventilation.

29

Page 29 of 36

ip t cr us an M d

te

Fig. 3 Overall procedure employed to evaluate and optimize the thermal performance of the

Ac ce p

PCM enhanced house with PVT ventilation.

Fig. 4 Illustration of the working principles expressed in Eq. (1).

30

Page 30 of 36

ip t cr us

Ac ce p

te

d

M

an

Fig. 5 Envelope-integrated PCM model.

Fig. 6 Enthalpy-temperature relationship.

25

20

Lower temperature settings for living space Lower temperature settings for sleeping space

T/ °C

Lower temperature settings used in this study

15

10

0

4

8

12 Time / h

16

20

24

Fig. 7 Thermal comfort temperature settings used in winter conditions.

31

Page 31 of 36

20 Ambient temperature Humidity ratio

16

ip t

450

12

cr

300

150

0

Temperature / °C Humidity ratio / g/(kg dry air)

Solar irradiation

0

12

24

36

48

60

72

84 96 time / h

us

Total horizontal solar irradiation / W/m 2

600

108

120

132

144

156

8

4 168

an

Fig. 8 Solar irradiations, ambient air temperatures and outdoor air humidity ratios during the seven selected test days. 35

M

Temperature setting of the climatic chamber Enclosure temperature response - experimental Enclosure temperature response - modelling

d

25

20

8

13

Ac ce p

15

te

T / °C

30

18

23

28

33

38

43

time / h

Fig. 9 Validation of the envelope-integrated PCM model.

32

Page 32 of 36

22 Baseline RT18HC SP21E SP24E

20

ip t

T / °C

18

16

0

12

24

36

48

60

72

84 96 time / h

108

120

132

144

156

168

156

168

us

12

cr

14

22

Baseline RT18HC SP21E SP24E

M

20

d

T / °C

18

te

16

Ac ce p

14

12

an

a) PCM layer thickness of 5 mm

0

12

24

36

48

60

72

84 96 time / h

108

120

132

144

b) PCM layer thickness of 30 mm

Fig. 10 Indoor temperatures of the house using and without using the PCMs.

33

Page 33 of 36

30 Baseline PVT PVT-RT18HC PVT-SP21E PVT-SP24E

28 26

22

ip t

T / °C

24

20 18

14 0

12

24

36

48

60

72

84 96 time / h

108

120

132

144

156

168

156

168

us

12

cr

16

30 Baseline PVT PVT-RT18HC PVT-SP21E PVT-SP24E

28

M

26

22 20

d

T / °C

24

te

18 16

Ac ce p

14 12

an

a) PCMs thickness of 5 mm

0

12

24

36

48

60

72

84 96 time / h

108

120

132

144

b) PCMs thickness of 30 mm

Fig. 11 Indoor temperatures of the house using the PVT ventilation and PCMs simultaneously, using the PVT ventilation only and that of the original house without using the PVT ventilation and PCMs.

34

Page 34 of 36

24 North wall South wall Ceiling

22

ip t

T / °C

20

18

0

12

24

36

48

60

72

84 96 time / h

a) RT18HC

132

144

156

168

M

20

T / °C

18

d

16

12

24

36

48

Ac ce p

0

te

14

12

120

an

22

108

us

14

cr

16

24 22

60

North wall South wall Ceiling

72

84 96 time / h

108

120

132

144

156

168

108

120

132

144

156

168

b) SP21E North wall South wall Ceiling

T / °C

20 18 16 14 12

0

12

24

36

48

60

72

84 96 time / h

c) SP24E Fig. 12 Surface temperatures of the PCM layers close to the plasterboard of the house model. 35

Page 35 of 36

Highlights: Thermal performance investigation of buildings with integrated PCMs and PVT systems

ip t

Taguchi method is used to determine the optimal combination of the control factors

Using PVT and PCMs can improve building performance under winter heating

Ac ce p

te

d

M

an

us

cr

conditions

36

Page 36 of 36