Comparative analyses on dynamic performances of photovoltaic–thermal solar collectors integrated with phase change materials

Energy Conversion and Management 131 (2017) 79–89

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Comparative analyses on dynamic performances of photovoltaic– thermal solar collectors integrated with phase change materials Di Su, Yuting Jia, Guruprasad Alva, Lingkun Liu, Guiyin Fang ⇑ School of Physics, Nanjing University, Nanjing 210093, China

a r t i c l e

i n f o

Article history: Received 19 August 2016 Received in revised form 25 October 2016 Accepted 1 November 2016 Available online 8 November 2016 Keywords: Solar energy Photovoltaic–thermal collectors Electrical and thermal performances Dynamic characteristics Phase change material

a b s t r a c t The operating conditions (especially temperature) of photovoltaic–thermal solar collectors have significant influence on dynamic performance of the hybrid photovoltaic–thermal solar collectors. Only a small percentage of incoming solar radiation can be converted into electricity, and the rest is converted into heat. This heat leads to a decrease in efficiency of the photovoltaic module. In order to improve the performance of the hybrid photovoltaic–thermal solar collector, we performed comparative analyses on a hybrid photovoltaic–thermal solar collector integrated with phase change material. Electrical and thermal parameters like solar cell temperature, outlet temperature of air, electrical power, thermal power, electrical efficiency, thermal efficiency and overall efficiency are simulated and analyzed to evaluate the dynamic performance of the hybrid photovoltaic–thermal collector. It is found that the position of phase change material layer in the photovoltaic–thermal collector has a significant effect on the performance of the photovoltaic–thermal collector. The results indicate that upper phase change material mode in the photovoltaic–thermal collector can significantly improve the thermal and electrical performance of photovoltaic–thermal collector. It is found that overall efficiency of photovoltaic–thermal collector in ‘upper phase change material’ mode is 10.7% higher than that in ‘no phase change material’ mode. Further, for a photovoltaic–thermal collector with upper phase change material, it is verified that 3 cm-thick phase change material layer is excellent both in electrical and thermal performance. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction With population explosion and technological progress, the world demand for energy is increasing. According to the Renewable 2015 Global Status Report, renewable energy represented approximately 59% of net addition to global power capacity in 2014 [1]. As an important renewable energy source, solar energy has been more attractive recently due to its easy availability, cost effectiveness, accessibility [2]. A photovoltaic (PV) cell is able to convert solar energy directly into electricity, because semiconducting materials absorb photons from sunlight and release electrons. However, the typical crystalline silicon cells only absorb incident light in 300–1100 nm range due to the band gap of the thin–film silicon [3]. Meanwhile, the operating temperature affects the electrical performance of silicon–based PV cells. Skoplaki and Palyvos [4] discussed most of the explicit and implicit correlations between module operating temperature and the electrical performance of PV cells. Radziemaka and Klugmann [5] investigated PV cells performance that showed ⇑ Corresponding author. E-mail address: [email protected] (G. Fang). http://dx.doi.org/10.1016/j.enconman.2016.11.002 0196-8904/Ó 2016 Elsevier Ltd. All rights reserved.

a power drop of 0.65% per 1 K temperature increase. PV cells have limited efficiency and the remaining absorbed solar energy has to be transformed into heat. This leads to higher operating temperature and lower cell efficiency. An appealing solution to this problem is to utilize a part of this heat in solar thermal collector [6]. Typical thermal energy absorbers are composed of a black absorber and a heat transfer fluid to remove heat. Usually, there are three types of working fluids: air, water and refrigerant. In other words, a so–called hybrid photovoltaic/thermal (PV/T) collector, which can improve the efficiency of PV cells and reutilize exhaust heat at the same time, has been proposed. Up to now, researchers have done both theoretical explorations and experimental studies on the hybrid PV/T collector, which can be classified as air–based PV/T collector and water–based PV/T collector. New forms of configuration designs of PV/T collector were put forward to satisfy the needs of various applications. Shahsavar and Ameri designed an air–based PV/T system with a thin aluminum sheet, which suspended at the middle of air channel to increase the heat exchange surface. The results showed that there was optimum number of fans, depending on the air mass flow rate, for achieving to maximum electrical efficiency [7]. The influence of air mass flow rate on both electrical and thermal performance of

80

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

Nomenclature c D F H h I L _ m T U

v

W x

a d

l

specific heat capacity, kJ/kg K equivalent diameter, m packing factor height, m convective heat transfer coefficient, W/m2 K solar radiation intensity, W/m2 length, m mass flow rate, kg/s temperature, K overall heat transfer coefficient, W/m2 K flow velocity, m/s width, m distance in flowing direction, m absorption coefficient thickness, m dynamic viscosity

PV/T system is considerable. Bambrook and Sproul [8] found that thermal and electrical efficiencies improved with the increased air mass flow rates (0.03–0.05 kg/s), and the results indicated that thermal efficiency was in the range of 28–55% and electrical efficiency was between 10.6% and 12.2% at midday. However, Teo et al. [9] discovered that when further increasing air mass (beyond 0.055 kg/s) flow rate to a certain value, the heat extracted had reached a saturated level, and both electrical and thermal efficiencies cannot be increased further by increasing the flow rate. Shan et al. [10] developed mathematical models for air–based PV/T collector with five different configurations. The calculation results indicated that PV/T system with a single channel below PV cells got optimal electrical performance, and thermal performance was best in the system with dual channels. Gaur and Tiwari [11] proposed a thermal modeling to research the effect of water–cooling on the performance of commercially available PV cells. The results showed that electrical efficiencies of a–Si PV module with and without water–cooling were 7.36% and 6.85% respectively. Fudholi et al. [12] presented three fabricated design configurations of water–cooling absorbers: web flow absorber, direct flow absorber and spiral flow absorber. The third absorber produced an electrical efficiency of 13.8% and a thermal efficiency of 54.6%, which was the highest among all absorber designs. Su et al. [13] presented four configurations of the hybrid PV/T collectors with dual channels for different fluids, which were able to recover more heat and reach higher efficiency. This work helped to select a more desirable structure based on different needs. Aiming to cool PV/T collector at high solar concentration ratios, Radwan et al. [14] developed a new cooling technique for PV/T systems using a microchannel heat sink with nanofluids. The results indicated that the use of nanofluids achieved about 5 °C lower cell temperature and higher electrical efficiency than pure water cooling. Both water and air have been utilized as heat removal medium for different applications. As solar energy is intermittent, a great disadvantage of this kind of energy is the large discrepancy between the supply and the demand. The heat demand is maximum in the evening while the solar radiation intensity is minimal or even zero [15]. In some applications, the exhaust heat needs to be stored, and water and air aren’t the optimal medium. Thermal energy storage using phase change materials (PCMs) has great potential and vast application prospects, as suitable PCMs have high latent heat with small volume and relatively constant phase transition temperature [16]. However, some drawbacks limit the application of the PCM, such as the leakage of melted PCM and the required phase change temperatures. A theoretical analysis

g k

s c

photovoltaic efficiency thermal conductivity, W/m K transmission coefficient kinematic viscosity, m2/s

Subscripts a airflow b backplane c solar cell e environment g glass cover i insulation layer p PCM layer ref reference value at reference conditions

was presented to estimate the stored energy dependence on time in a tank containing PCM, and the results showed that parameters, like the PCM, cylinder radius, the mass flow rate, and the inlet temperature of the heat transfer fluid, impacted on the performance of the tank [17]. Then Esen et al. [18] undertook a series of numerical tests and put forward the optimal geometric design of the tank depending on these parameters and PCMs. An appealing solution to this problem is to integrate thermal energy storage into the PV/T collector which stores the heat during the day and releases it at night [19]. The PV/T collector with PCM consists of different layers including a glass cover, PV cells, backplane, PCM, heat transfer pipes and an insulation layer, as shown in Fig. 1. Many studies concerning the thermal performance of waterbased PV/T cells integrated with PCMs have been undertaken. Malvi et al. [20] presented an energy balance model to evaluate a PV/T system with PCM and found the optimum applicable PCM melting point. It was shown that 0.03 m-thick PCM layer was optimum, which increased the electrical efficiency by 9%. Ho et al. [21] introduced a water–surface floating PV/T system integrated with water–saturated microencapsulated phase change material layer. They found that 5 cm-thick PCM layer with a melting temperature of 30 °C had the best performance. Browne et al. [22] designed a novel photovoltaic/thermal collector with PCM and investigated the thermal performance of this system. It was shown that the introduction of PCM was an effective method to enhance heat removal in a PV/T system, and the maximum temperature difference of water in the system with and without PCM was 5.5 °C. Qiu et al. [23] presented the hybrid PV/T system using the microencapsulated phase change material as the working fluid. The results indicated that the MPCM slurry based PV/T system was superior to typical air–based and water–based PV/T systems. Yin et al. [24] put forward a PV/T system with a PCM storage unit to provide domestic hot water. Water was directly heated by solar thermal collector and carried away the exhaust heat during the day. The warm water can be used immediately, or the heat can be extracted into the PCM storage for use at nighttime. However, few studies have been devoted to evaluate the dynamic performance of the air-based PV/T collector integrated with PCMs. Air-based systems are practically preferred for lesser use of materials and lower operating cost than water-based, despite of poor thermophysical properties of air. In this work, PCM layer is integrated into the heat transfer pipes of air-based PV/T collector, and both electrical and thermal performance under the operating conditions of Nanjing in China is simulated and analyzed. Table 1 shows design parameters of the PV/T collector. The

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

81

Fig. 1. The schematic diagram of the PV/T collector with PCM.

Table 1 Design parameters of the PV/T collector. Parameters

Symbol

Value

Length of the PV module Width of the PV module Height of the pipe Packing factor of the solar cell Wind velocity Mass flow rate of air Specific heat capacity of air Thickness of the PCM Reference value

L W H F v _ m ca dp

Average efficiency of power plants

gTpower

1.5 m 1m 0.05 m 0.83 [10] 2 m/s 0.05 kg/s 1.005 kJ/kg K 0.05 m 0.12, 0.0045 and 293 K 0.4

gref , bref and T ref

main objectives of the present study are: (1) to use a mathematical model to analyze dynamic performances of air-based PV/T collector, (2) to assess the effects of the position of the PCM layer in collector, and (3) to determine the optimal thickness of PCM layer integrated into air-based PV/T collector.

one dimensional energy balance method to acquire the parameters of PV/T collectors at each hour. The convection in the heat transfer pipe and the phase change process of the PCM are concurrently taken into account. Moreover, each component has its own energy balance equation, and temperatures of all components can be calculated. The simulation is expected to evaluate the parameters like cell temperature, outlet temperature of air, thermal efficiency and electrical efficiency. 2.1. Glass cover The heat absorbed by the glass cover loses to the surrounding in the form of convection and conduction, including dissipation from the top of the glass cover to the ambient and conduction to the PV cell surface. The energy balance of the glass cover can be written as follows:

ag
I ¼ hge ðT g  T e Þ þ U gc ðT g  T c Þ

ð1Þ

It is noted that hge is the overall heat transfer coefficient of the glass to the atmosphere, it can be calculated as:




dg 1 þ kg he

1

2. Mathematical models

hge ¼

In general, a PV/T collector is mainly composed of a PV module and heat transfer pipes. The PV cell is encapsulated between a protective tempered glass cover and an aluminum–alloy backplane. In the present work, the heat transfer pipes integrated with a PCM layer at various positions are adopted, through which air flows as the working fluid. A schematic diagram of the PV/T collectors with PCM is shown in Fig. 2. To develop the proposed numerical model, the following assumptions have been made: (1) Mean temperature is assumed across each layer. (2) The horizontal temperature in each layer is uniform except for the airflow. (3) Heat capacities of the PV/T components are neglected. Although Deng et al. [25] showed that the relative error was approximately 7% if heat capacity was neglected. In the present work, the equivalent heat capacity of the PCM in phase change process is much larger than heat capacity of the components. Hence the error caused by this assumption is in an acceptable range. (4) The radiant heat transfer and reflection of sunlight are neglected. (5) The PV/T components reach quasi–steady state immediately while the environment temperature and solar radiation intensity change, except for the PCM layer. Based on these assumptions, the practical energy transfer processes in air–based PV/T collectors with PCM are modeled using

In addition, U gc is the overall heat transfer coefficients of the glass to the PV module, it can be described as:


U gc ¼

dg dc þ kg kc

ð2Þ

1 ð3Þ

For the windward face of the glass cover, the convective heat transfer coefficient depends on the wind speed, and it is described by an empirical relationship as follows [26]:

he ¼ 5:7 þ 3:8v

ð4Þ

The heat conductivity (k), transmissivity (s), absorptivity (a) and thickness (d) of PV components are listed in Table 2. 2.2. PV cell It is obvious that the total energy of PV module comes from two kinds of sources. One is the solar energy transmitted through the glass cover and absorbed by the cells, and the other is the conductive heat exchange occurring between glass cover and PV cells. The total energy is divided into two parts. The first part is converted to electricity depending on the cell’s efficiency, and the second part is

82

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

Fig. 2. The sectional view of the PV/T collector with PCM at different positions.

Table 2 Physical parameters of the photovoltaic modules [10,29,35,36]. Materials

k (W/m K)

s

a

d (m)

Glass Solar cell Backplane Insulation

1 [35,36] 148 [10] 144 [29] 0.035 [36]

0.91 [29] 0.09 [10] 0.00039 [29] –

0.05 [29] 0.8 [10] 0.4 [29] –

0.003 [29,35] 0.0003 [10,35] 0.0005 [29] 0.05 [36]

converted to heat transmitted to the backplane. The energy balance on the PV cell is given by:

F
ac
sg
I þ U gc ðT g  T c Þ ¼ g
F
ac
sg
I þ U cb ðT c  T b Þ

ð5Þ

where F is the packing factor of solar cell. In the applications, the photovoltaic efficiency as a function of cell temperature is given by Ref. [27]:

g ¼ gref ½1  bref ðT c  T ref Þ

ð6Þ

where U cb is the overall heat transfer coefficient of the cell to the backplane, it can be calculated as:


U cb ¼

dc db þ kc kb

1 ð7Þ

From Eq. (9), the distance–dependent temperature of airflow can be obtained:

T a ðxÞ ¼

and T a;0 is the inlet temperature of air. Suppose the flow distance x ¼ L, the outlet temperature of air can be described as:

T a ðLÞ ¼ T a;0
Q 1 þ

hba
T b þ hai
T i ð1  Q 1 Þ hba þ hai

Ta ¼

1 L

Z

L

T a ðxÞ
dx ¼ T a;0
R1 þ 0

hba
T b þ hai
T i ð1  R1 Þ hba þ hai

ð13Þ

1 where R1 ¼ 1Q is introduced to simplify the equation. P 1
L

Case 2: PV/T collector with upper PCM as shown in Fig. 2(b) For backplane:

F
ab
sc
sg
I þ ð1  FÞ
ab
sg
I þ U cb ðT c  T b Þ ¼ U bp ðT b  T p Þ ð14Þ For airflow:

F
ab
sc
sg
I þ ð1  FÞ
ab
sg
I þ U cb ðT c  T b Þ ¼ hba ðT b  T a Þ ð8Þ For airflow:

_ a
dT a þ hai ðT a  T i ÞW
dx hpa ðT p  T a ÞW
dx ¼ ca
m

ð15Þ

For insulation layer:

ð9Þ

For insulation layer:

hai ðT a  T i Þ ¼ hie ðT i  T e Þ

ð12Þ

where Q 1 ¼ eP1 L is introduced to simplify the equation. With the help of Eqs. (11) and (12), the mean temperature of air flowing over the length of pipe is:

For backplane:

_ a
dT a þ hai ðT a  T i ÞW
dx hba ðT b  T a ÞW
dx ¼ ca
m

ð11Þ

ai ÞW where P 1 ¼ ðhbacaþh is a factor introduced to simplify the equation, _a
m

2.3. Other PV/T components with PCM For the sunlight reaching the backplane, a portion of the light passes through only glass cover, another part passes through glass cover and the cells. As shown in Fig. 2, there are three types of configuration of heat transfer pipes: no PCM, upper PCM, and lower PCM. The complete energy balance equations can be written under three cases. Case 1: PV/T collector without PCM as shown in Fig. 2(a)


 hba
T b þ hai
T i P1 x hba
T b þ hai
T i e T a;0  þ hba þ hai hba þ hai

ð10Þ

hai ðT a  T i Þ ¼ hie ðT i  T e Þ

ð16Þ

From Eq. (14), the distance–dependent temperature of airflow can be obtained:

83

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89


 hpa
T b þ hai
T i P2 x hpa
T p þ hai
T i e T a ðxÞ ¼ T a;0  þ hpa þ hai hpa þ hai where P2 ¼

ðhpa þhai ÞW _a . ca
m

ð17Þ

And the outlet temperature of air can be

The forced convective heat transfer coefficient of air in pipes is given by,

Re ¼

described as:

T a ðLÞ ¼ T a;0
Q 2 þ

hpa
T p þ hai
T i ð1  Q 2 Þ hpa þ hai

ð18Þ

where Q 2 ¼ eP2 L . With the help of Eqs. (16) and (17), the mean temperature of air flowing over the length of pipe is:

1 T a ¼ L

Z

L

0

hpa
T p þ hai
T i T a ðxÞ
dx ¼ T a;0
R2 þ ð1  R2 Þ hpa þ hai

ð19Þ

2 where R2 ¼ 1Q . P 2
L

tf
D c

ð28Þ

where c is the kinematic viscosity and D is the equivalent diameter. For a flat rectangular tube:

D¼

2H
W HþW

ð29Þ

By quantitative calculation, the airflow in this work is in turbulent state (Re > 5000). The Nusselt number of airflow can be estimated by the following correlation [30]: 0:8

Nu ¼ 0:0214ðRe

 100ÞPr

0:4

Case 3: PV/T collector with lower PCM as shown in Fig. 2(c)

F
ab
sc
sg
I þ ð1  FÞab
sg
I þ U cb ðT c  T b Þ ¼ hba ðT b  T a Þ ð20Þ For airflow:

ð21Þ

ð22Þ

From Eq. (21), the distance–dependent temperature of airflow can be obtained:


 hba
T b þ hap
T p P3 x hba
T b þ hap
T p e T a ðxÞ ¼ T a;0  þ hba þ hap hba þ hap where P3 ¼

ðhba þhap ÞW . _a ca
m

ð23Þ

And the outlet temperature of air can be

described as:

T a ðLÞ ¼ T a;0
Q 3 þ

hba
T b þ hap
T p ð1  Q 3 Þ hba þ hap

ð24Þ

where Q 3 ¼ eP3
L . With the help of Eqs. (23) and (24), the mean temperature of air flowing over the length of pipe is:

1 T a ¼ L

Z

L

T a ðxÞ
dx ¼ T a;0
R3 þ 0

hba
T b þ hap
T p ð1  R3 Þ hba þ hap

ð25Þ

ð31Þ

With the help of Eqs. (26), (27) and (31), the overall heat transfer coefficient in the mathematical model between each component of the PV/T collector with PCM can be calculated. 2.5. PCM layer During the melting and solidification processes, the heat exchange between PCM layer and the contiguous layer is mainly in the form of conduction. Even a very strong natural convection in the solid-liquid interface has a negligible effect on the heat transfer compared to the effect of heat conduction [31]. Table 3 shows a list of the PCM thermal parameters used by the simulation [32]. As shown in Table 3, the PCM has different heat properties in solid and liquid phase. During the phase change process, the temperature of the PCM layer is taken as the melting point and the thermal conductivity of the PCM depends on the ratio of the solid and liquid phases [20].

ð32Þ

  d
Q PCM ¼ W
L hpa ðT a  T p Þ  U pi ðT p  T i Þ dt

ð33Þ

1 ð26Þ

In addition, the overall heat transfer coefficient of airflow to the contiguous layer can be described as:

h¼

ka D

For case 3:

In three cases, conductive heat exchange occurs between each component of the PV/T collector with PCM. For neighboring component layers 1 and 2, the overall heat transfer coefficient is given by,

d1 d2 þ k1 k2

So the convective heat transfer coefficient between air in pipes and neighboring component is given by:

d
Q PCM ¼ W
L½U bp ðT b  T p Þ  hpa ðT p  T a Þdt

2.4. Thermal exchange coefficients




Ta < 1:5 T

For case 2:

3 where R3 ¼ 1Q . P 3
L

U 12 ¼

2300 < Re < 106 ; 0:6 < Pr < 1:5; 0:5 <

ha ¼ Nu


For insulation layer:

U pi ðT p  T i Þ ¼ hie ðT i  T e Þ

ð30Þ

where Pr is the Prandtl number. T a and T denote to the mean air temperature and the wall temperature of the tube respectively. This correlation can meet the precision for engineering calculation quite well under the following conditions:

For backplane:

_ a
dT a þ hap ðT a  T p ÞW
dx hba ðT b  T a ÞW
dx ¼ ca
m


2=3 #
0:4 " Ta D 1þ L T


1 d 1 þ k ha

ð27Þ

Furthermore, the convective heat transfer coefficient inside the air duct was assumed as a constant in some Refs. [28,29]. However, in this work ha is also calculated according to flow regime.

Table 3 Thermal properties of the PCM [20]. Thermal property

Symbol

Value

Thermal conductivity (solid) Thermal conductivity (liquid) Specific heat capacity (solid) Specific heat capacity (liquid) Density (solid) Density (liquid) Latent heat Melting point

kPCMs kPCMl cPCMs cPCMl

0.24 W/m K 0.15 W/m K 2.9 kJ/kg K 2.1 kJ/kg K 860 kg/m3 780 kg/m3 210 kJ/kg 28 °C

qPCMs qPCMl LH T mp

84

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

If Q PCM < Q sol ,

T p ¼ T mp 

thermal efficiency together. The concept of conversion factor is then introduced, in that

Q sol  Q PCM ; kp ¼ kPCMs cPCMs

go ¼ gt þ

If Q sol < Q PCM < Q liq ,

T p ¼ T mp ; kp ¼ kPCMs
If Q liq < Q PCM ,

3. Results and discussion

Q PCM  Q liq T p ¼ T mp þ ; kp ¼ kPCMl cPCMl where Q PCM is the cumulative heat flow into the PCM layer, Q sol and Q liq refer to the cumulative heat flows required to complete solid and liquid phase changes respectively, and T mp is the PCM melting point. 2.6. Evaluation of performance parameters The conversion efficiency of the photovoltaic panel can be obtained from Eq. (6). According to Ref. [1,28], the rate of useful thermal energy obtained from the PV/T collector with PCM is given by:

ð34Þ

where Q_ PCM is the heat generation rate of the PCM. The electrical power of the PV/T collector can be described as follows:

Q_ p ¼ g
ac
sg
F
A
I

ð38Þ

where gpower is the average efficiency of power plant at national level and its value is taken as 0.4 in this work, due to the fact that the efficiency of thermal power plant in China is around 40–42%.

Q liq  Q PCM Q  Q sol þ kPCMl
PCM Q liq  Q sol Q liq  Q sol

_ a ðLÞ  T a0 Þ þ Q_ PCM Q_ u ¼ ca
mðT

gp gpower

ð35Þ

And three kinds of efficiencies are defined for a PV/T collector with PCM. The both thermal and electrical efficiency can be described respectively as follows [33]:

gt ¼

Q_ u A
I

ð36Þ

gp ¼

Q_ p A
I

ð37Þ

Researchers consider that electricity is in a higher grade form since it is converted from thermal energy [6,7,33]. Hence, the overall efficiency cannot be obtained by adding electrical efficiency and

The hourly variation of solar radiation intensity and ambient temperature from 8:00 to 17:00 on a typical summer day in Nanjing is shown in Fig. 3. During this time period, the radiation intensity initially rises to a maximum value of approximately 1000 W/ m2 at 12:00, then decreases. The ambient temperature rises to a peak value at 14:00 and then decreases. In this work, the incident sunlight is assumed to remain perpendicular to the surface of PV cell during the day with a purpose to simplify calculations. This is a reasonable assumption. Mousazadeh et al. [34] discussed different types of sun-tracker which could adjust the solar systems perpendicularly to the direction of radiation. And the wind speed over the glass cover is a constant, and inlet air temperature is equal to ambient temperature. The design parameters of the PV/T collector with PCM are shown in Tables 1–3. All simulations are performed in MATLAB software. A series of simulation results, including the photovoltaic efficiency, temperature of solar cell, the photovoltaic power, the rate of useful heat obtained and the overall efficiency will be analyzed in detail. 3.1. Model validation Stropnik and Stritih [37] constructed a PV panel integrated with phase change material. The focus of their work was experimental setup and simulation of heat extraction process from the PV panel, using TRNSYS software. To verify the simulation method of the PCM layer in this work, the calculation was conducted under the same conditions (shown in Table 4 and Fig. 4) and compared with the actual experimental data. The simulation process of air-based PV/T collector without PCM was verified in our previous work [13].

Fig. 3. The hourly variation of solar radiation intensity and ambient temperature of Nanjing.

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89 Table 4 The design parameters of the PV panel with PCM [37]. Parameters

Value

Length of the PV module Width of the PV module Thickness of the glass cover Thickness of the EVA Thickness of the PV cells Thickness of the Tedlar Thickness of the PCM Latent heat of PCM Melting peak of the PCM Thermal conductivity of the PCM Specific heat capacity of the PCM Density of the PCM (solid) Density of the PCM (liquid)

1638 mm 982 mm 3.2 mm 0.5 mm 0.3 mm 0.5 mm 35 mm 245 kJ/kg 28 °C 0.2 W/m K 2 kJ/kg K 880 kg/m3 770 kg/m3

The hourly variation of PV cells temperature in experimental results and in numerical results is shown in Fig. 5. It is observed that experimental curve and simulation curve rise from 10:00 to 14:00 and drop after 14:00. The simulation results nicely follow the measurement results during some periods (10:00–11:00 and 14:00–15:00). It is shown in Fig. 4 that during these periods, solar radiation is below 400 W/m2. During 11:00–14:00, the simulation results are slightly above experiment results, and the difference is maximum 3.2% at 12:00 when solar radiation is the strongest. What probably causes this difference is the fact that the radiant heat exchange between the collector and the ambient has been neglected. In addition, the heat capacities of PV components have been neglected. If radiant heat exchange and heat capacities of all components are taken into account, the PV modules would release an extra amount of heat to ambient, and the temperature of PV cells may be closer to the experimental results during 11:00–14:00. In general, it is clear that simulation results are in good agreement with measurement results. And the correlation coefficient between two curves is 0.992.

Fig. 5. Comparisons of the calculated results with experimental results.

3.2. Electrical performance of the PV/T collector with PCM The hourly variation of solar cell temperature throughout the day is shown in Fig. 6. The curve of no PCM mode and curve of lower PCM mode have entirely uniform trend, which rise from

Fig. 6. The hourly variation of solar cell temperature.

Fig. 4. The hourly variation of solar radiation intensity and ambient temperature [37].

85

86

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

7:00 to 12:00 and drop from 12:00 to 18:00. Solar cell temperature in no PCM case is slightly higher than the temperature in lower PCM case, and average temperature difference of solar cells is 1.05 °C. The reason is that lower PCM layer and the solar cells are separated by the airflow, and the lower PCM layer has a minimal impact on performance of the PV cells. Besides, the temperature of solar cells in upper PCM case rises from 27.2 °C to a peak value of 70.0 °C at 13:00, and then drops to 37.0 °C. Comparing curve of no PCM mode with curve of upper PCM mode, the maximum temperature difference of solar cells is 7.4 °C at 11:00. It is found that upper PCM mode causes a delay in temperature rise compared to the no PCM mode, because the extra heat of the PV cells is absorbed by PCM layer. As the PCM starts to melt, which further slows down the rate of temperature increase, it leads to the maximum difference at 11:00 and delays peak temperature to 13:00. And curve of upper PCM mode declines more slowly than curve of no PCM mode, two curves intersect at 16:00, because of the larger thermal mass of the PCM. The heat absorbed by the PCM returns to the solar cell, which causes the cells warmer than no PCM case. The hourly variation of photovoltaic efficiency is shown in Fig. 7. Eq. (6) shows that the photovoltaic efficiency and the temperature are negative correlation. Curve of no PCM mode and curve of lower PCM mode drop from 7:00 to 12:00 and rise from 12:00 to

Fig. 7. The hourly variation of photovoltaic efficiency.

Fig. 8. The hourly variation of electrical power.

Fig. 9. The hourly variation of electrical efficiency.

18:00, and the average difference in photovoltaic efficiency is 0.06%. Besides, the photovoltaic efficiency in upper PCM case drops from 11.88% to a least value of 9.57% at 13:00, and then rises to 11.35%. Comparing curve of no PCM mode with curve of upper PCM mode, the maximum difference in the photovoltaic efficiency is 0.4% at 12:00. The hourly variation of electrical power is shown in Fig. 8. The variation trend of electrical power in each case is similar to the others. All curves rise from 7:00 to 12:00, and reach peak value at 12:00, then drop from 12:00 to 18:00. In other words, the variation trend of electrical power is consistent with the change in the solar radiation intensity. It is because electrical power mainly depends on the solar radiation intensity when the photovoltaic efficiency stays largely constant, shown in Eq. (35). It is noted that the maximum difference between the electrical power of no PCM case and upper PCM case is 4.2 W at 12:00. Hence, increasing photovoltaic efficiency can still improve the electrical power especially at midday. The hourly variation of electrical efficiency is shown in Fig. 9. According to Eqs. (35) and (37), it is concluded that electrical efficiency just depends on photovoltaic efficiency, and all curves of electrical efficiency keep in step with the hourly variation of photovoltaic efficiency. After 16:00, it is clear that upper PCM mode has the lowest efficiency. The reason is that the PCM melts completely, and temperature of PCM rises caused by the accumulating heat.

Fig. 10. The hourly variation of outlet temperature of air.

87

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

In conclusion, upper PCM mode improves the electrical performance of the PV/T collector, while lower PCM mode has little effect on it. 3.3. Thermal performance of PV/T collector with PCM The hourly variation of outlet temperature of air is shown in Fig. 10. Curve of no PCM mode and curve of lower PCM mode have entirely uniform trend, which rise from 7:00 to 14:00 and drop from 14:00 to 18:00. Outlet temperature in no PCM mode is slight higher than the temperature in lower PCM mode, and average difference in outlet temperature is 0.97 °C. It is because the lower PCM layer absorbs heat from the airflow and reduces the outlet temperature of air. In addition, the outlet temperature of air in upper PCM case rises from 28.3 °C to a maximum value of 36.7 °C at 14:00, and then drops to 33.5 °C. The maximum difference in outlet temperature of upper PCM mode and lower PCM mode is 6.2 °C at 12:00. After 12:00, the curve of upper PCM mode declines slowly and intersects with the curve of lower PCM mode during 16:00–17:00. Thermal energy absorbed by the upper PCM is conducted to airflow, which causes the outlet temperature of upper PCM mode is higher than that of no PCM mode. The results indicate that air-based PV/T collector with PCM can be used in building surface to utilize solar energy and provide comfortable indoor temperature. The hourly variation of the rate of useful thermal energy obtained is shown in Fig. 11. The curve of no PCM mode and curve of lower PCM mode are almost the same at a given time, and rise from 8:00 to 12:00, reach the peak value of 289 W and drop from 12:00 to 17:00. It is clear that upper PCM mode obtains more thermal energy than the others, and reaches a maximum value of 435 W at midday. After the PCM layer completely melts, the curve of upper PCM mode suffers a sudden drop, and it appears that the outlet temperature of air increases. The useful thermal energy is represented by the area under curves. The thermal energy obtained

Fig. 12. The hourly variation of thermal efficiency.

in no PCM case is around 6:6  106 J and the thermal energy in upper PCM case is approximately 9:8  106 J, increasing by about 48%. The hourly variation of thermal efficiency is shown in Fig. 12. The values of no PCM mode and lower PCM mode almost keep at a constant level of 19%. It is noted that the curve of upper PCM mode reduces quickly before 9:00, remains relatively stable around 28% during 10:00–14:00, and resumes the fast downward trend after 15:00. It means that the melting process of the PCM layer in PV/T collector can be divided into three steps: heating to phase transition temperature, beginning to melt and completely melted.

Fig. 13. The hourly variation of overall efficiency.

The hourly variation of overall efficiency is shown in Fig. 13. The overall efficiency of no PCM mode and lower PCM mode is relatively stable. During 10:00–14:00, average difference in overall efficiency between no PCM mode and upper PCM mode is approximately 10.7%. It is clear that the upper PCM layer can significantly improve thermal performance of the PV/T collector, while the lower PCM has little impact on it. As shown in Table 5, the present work has relatively higher thermal efficiency as compared with the others, while cell efficiency is the lowest in the three studies. It is obvious that introduction of the PCM layer significantly improves the thermal property of the air-based PV/T collector.

Table 5 Comparison of the efficiency of the air-based PV/T collector in literatures with the present study.

Fig. 11. The hourly variation of rate of useful thermal energy.

Air-based PV/T collector

Cell efficiency (%)

Thermal efficiency (%)

Reference

Single micro-channel Channel under PV cell Integrated with PCM

14.3 10.9 10.5

18 22.5 29.5

[26] [10] Present study

88

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

3.4. Influence of thickness of PCM on performance of PV/T collector with upper PCM layer From the above work, it can be determined that the PV/T collector with upper PCM is excellent in both electrical and thermal performance. Then simulation is performed by changing the thickness of the PCM layer to select the right size based on upper PCM mode. The hourly variation of temperature of solar cells is shown in Fig. 14. Except for d = 2 cm curve, the solar cells temperature value rises from 7:00 to 12:00, and reaches peak value at 12:00 and then drops from 12:00 to 18:00. According to Eq. (26), lesser the thickness of the PCM layer, better the heat exchange rate. Hence the rate of temperature increase in solar cell is the slowest when d = 2 cm. However, the thin PCM layer leads to small heat storage capacity. After 12:00, 2 cm-thick PCM layer is overheating and returns thermal energy back to the solar cell, which warms the solar cell to 77.2 °C at 14:00. Therefore 2 cm-thick PCM layer is not the optimal choice. The hourly variation of electrical efficiency is shown in Fig. 15. All curves except d = 2 cm curve have similar trend, which drop from 7:00 to 12:00, and rise from 12:00 to 18:00. It is clear that the electrical efficiency of d = 3 cm curve is the highest before

Fig. 16. The hourly variation of heat absorbed by PCM layer with different PCM thickness.

Fig. 17. The hourly variation of outlet temperature with different PCM thickness.

Fig. 14. The hourly variation of solar cell temperature with different PCM thickness.

Fig. 15. The hourly variation of electrical efficiency with different PCM thickness.

16:00. After 16:00, solar radiation levels are low and the PV cell generates only a small fraction of its total daily electricity. So, 3 cm-thick PCM layer is excellent in electrical performance. The hourly variation of the rate of heat absorbed by the PCM layer is shown in Fig. 16. d = 2 cm curve rises from 7:00 to 14:00, reaches the peak value of 8750 kJ, and drops to 6950 kJ at 18:00. d = 5 cm curve consistently rises and reaches the maximum value of 9770 kJ at 18:00. d = 3 cm curve rises from 7:00 to 16:00, reaches the peak value of 10,590 kJ, and drops to 10,270 kJ at 18:00. It is obvious that 3 cm-thick PCM layer can absorb the most amount heat during the daytime. The hourly variation of outlet temperature of air is presented in Fig. 17. Before 12:00, the outlet temperature values of three cases are almost the same value at a given time. And then 2 cm-thick PCM layer melts completely, which leads to the outlet temperature rising abruptly in d = 2 cm curve. At 15:00, 3 cm-thick PCM layer is in liquid state and begins to warm the airflow. From the above results, it can be concluded that 3 cm-thick PCM layer is optimal for the PV/T collector with upper PCM in this work. This optimal value helps to make full use of solar energy and minimize the cost for the application of air-based PV/T collector with PCM on buildings.

D. Su et al. / Energy Conversion and Management 131 (2017) 79–89

4. Conclusions In this work, PCM layer is integrated into the heat transfer pipes of air-based PV/T collector. Both electrical and thermal performances under the operating conditions (as shown Table 1) are analyzed and discussed. The parameters of the air-based PV/T collector refer to: (1) the position of the PCM layer in collector, which includes no PCM case, upper PCM case and lower PCM case, and (2) the thickness of PCM layer integrated into collector, which includes 2 cm, 3 cm and 5 cm thickness. The following conclusions can be drawn: (1) With the increase of solar radiation intensity, the solar cell temperature increased to 76.7 °C and the electrical power increased to 106.1 W, respectively, while the photovoltaic efficiency decreased to 9.2% and electrical efficiency decreased to 6.7%, respectively. The solar cell temperature reached a peak value of 70 °C at 13:00 in upper PCM mode, and the electrical efficiency got a minimum of 6.7% at 12:00 in no PCM mode. (2) The rate of useful thermal energy followed the variation of solar radiation intensity and reached a peak value of 435 W at 12:00 in upper PCM mode. Besides, the outlet temperature followed the variation of ambient temperature and reached maximum 42.4 °C at 12:00 in no PCM mode. (3) It is demonstrated that the comprehensive performance of the PV/T collector in upper PCM mode is optimal among three structures, and overall efficiency of the PV/T collector in upper PCM mode is 10.7% higher than that in no PCM mode. (4) For a PV/T collector with upper PCM, it is verified that 3 cmthick PCM layer is excellent both in electrical and thermal performance. Acknowledgements This work was supported by the National Natural Science Foundation of China (Grant nos. 51376087, 51676095) and the Priority Academic Program Development of Jiangsu Higher Education Institutions. The authors also wish to thank reviewers and editor for kindly giving revising suggestions. References [1] Hassania S, Taylor RA, Mekhilef S, Saidur R. A cascade nanofluid-based PV/T system with optimized optical and thermal properties. Energy 2016;112:963–75. [2] Kannan N, Vakeesan D. Solar energy for future world: a review. Renew Sustain Energy Rev 2016;62:1092–105. [3] Xu Y, Xuan Y, Yang L. Full-spectrum photon management of solar cell structures for photovoltaic–thermoelectric hybrid systems. Energy Convers Manage 2015;103:533–41. [4] Skoplaki E, Palyvos JA. Operating temperature of photovoltaic modules: a survey of pertinent correlations. Renew Energy 2009;34:23–9. [5] Radziemska E, Klugmann E. Thermally affected parameters of the current– voltage characteristics of silicon photocell. Energy Convers Manage 2002;43:1889–900. [6] Chow TT. A review on photovoltaic/thermal hybrid solar technology. Appl Energy 2010;87:365–79. [7] Shahsavar A, Ameri M. Experimental investigation and modeling of a direct– coupled PV/T air collector. Sol Energy 2010;84:1938–58. [8] Bambrook SM, Sproul AB. Maximising the energy output of a PVT air system. Sol Energy 2012;86:1857–71.

89

[9] Teo HG, Lee PS, Hawlader MNA. An active cooling system for photovoltaic modules. Appl Energy 2012;90:309–15. [10] Shan F, Tang F, Cao L, Fang GY. Comparative simulation analyses on dynamic performances of photovoltaic–thermal solar collectors with different configurations. Energy Convers Manage 2014;87:778–86. [11] Gaur A, Tiwari GN. Performance of a–Si thin film PV modules with and without water flow: an experimental validation. Appl Energy 2014;128:184–91. [12] Fudholi A, Sopian K, Yazdi MH, Ruslan MH, Ibrahim A, Kazem HA. Performance analysis of photovoltaic thermal (PVT) water collectors. Energy Convers Manage 2014;78:641–51. [13] Su D, Jia YT, Huang X, Alva G, Tang YJ, Fang GY. Dynamic performance analysis of photovoltaic–thermal solar collector with dual channels for different fluids. Energy Convers Manage 2016;120:13–24. [14] Radwan A, Ahmed M, Ookawara S. Performance enhancement of concentrated photovoltaic systems using a microchannel heat sink with nanofluids. Energy Convers Manage 2016;119:289–303. [15] Esen M. Thermal performance of a solar-aided latent heat store used for space heating by heat pump. Sol Energy 2000;69:15–25. [16] Lin WY, Ma ZJ. Using Taguchi-Fibonacci search method to optimize phase change materials enhanced buildings with integrated solar photovoltaic thermal collectors. Energy 2016;106:23–37. [17] Esen M, Ayhan T. Development of a model compatible with solar assisted cylindrical energy storage tank and variation of stored energy with time for different phase change materials. Energy Convers Manage 1996;37:1775–85. [18] Esen M, Durmus A, Durmus A. Geometric design of solar-aided latent heat store depending on various parameters and phase change materials. Sol Energy 1998;62:19–28. [19] Hengstbergera F, Zaunera C, Reschb K, Holperc S, Grobbauer M. High temperature phase change materials for the overheating protection of facade integrated solar thermal collectors. Energy Build 2016;124:1–6. [20] Malvi CS, Dixon-Hardy DW, Crook R. Energy balance model of combined photovoltaic solar–thermal system incorporating phase change material. Sol Energy 2011;85:1440–6. [21] Ho CJ, Chou WL, Lai CM. Thermal and electrical performance of a water-surface floating PV integrated with a water-saturated MEPCM layer. Energy Convers Manage 2015;89:862–72. [22] Browne MC, Norton B, McCormack SJ. Heat retention of a photovoltaic/thermal collector with PCM. Sol Energy 2016;133:533–48. [23] Qiu ZZ, Ma XL, Zhao XD, Li P, Ali S. Experimental investigation of the energy performance of a novel micro–encapsulated phase change material (MPCM) slurry based PV/T system. Appl Energy 2016;165:260–71. [24] Yin HM, Yang DJ, Kelly G, Garant J. Design and performance of a novel building integrated PV/thermal system for energy efficiency of buildings. Sol Energy 2013;87:184–95. [25] Deng J, Ma RJ, Yuan GF, Chang C, Yang XD. Dynamic thermal performance prediction model for the flat-plate solar collectors based on the two-node lumped heat capacitance method. Sol Energy 2016;135:769–79. [26] Agrawal S, Tiwari GN. Energy and exergy analysis of hybrid micro–channel photovoltaic thermal module. Sol Energy 2011;85:356–70. [27] Skoplaki E, Palyvos JA. On the temperature dependence of photovoltaic module electrical performance: a review of efficiency/power correlations. Sol Energy 2009;83:614–24. [28] Bahaidarah H, Subhan A, Gandhidasan P, Rehman S. Performance evaluation of a PV (photovoltaic) module by back surface water cooling for hot climatic conditions. Energy 2013;59:445–53. [29] Shan F, Cao L, Fang GY. Dynamic performances modeling of a photovoltaic– thermal collector with water heating in buildings. Energy Build 2013;66:485–94. [30] Gnielinski V. New equations for heat and mass transfer in turbulent pipe and channel flows. Int Chem Eng 1976;16:359–68. [31] Lamberg P, Lehtiniemi R, Henell AM. Numerical and experimental investigation of melting and freezing processes in phase change material storage. Int J Therm Sci 2004;43:277–87. [32] Smith CJ, Forster PM, Crook R. Global analysis of photovoltaic energy output enhanced by phase change material cooling. Appl Energy 2014;126:21–8. [33] Aste N, Del Pero C, Leonforte F. Water flat plate PV–thermal collectors: a review. Sol Energy 2014;102:98–115. [34] Mousazadeh H, Keyhani A, Javadi A, Movli H, Abrinia K, Sharifi A. A review of principle and sun-tracking methods for maximizing solar systems output. Renew Sustain Energy Rev 2009;13:1800–18. [35] Dubey S, Tay AAO. Testing of two different types of photovoltaic–thermal (PVT) modules with heat flow pattern under tropical climatic conditions. Energy Sustain Develop 2013;17:1–12. [36] Tiwari A, Sodha MS. Performance evaluation of solar PV/T system: an experimental validation. Sol Energy 2006;80:751–9. [37] Stropnik R, Stritih U. Increasing the efficiency of PV panel with the use of PCM. Renew Energy 2016;97:671–9.