Energy 208 (2020) 118409

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Experimental evaluation of novel photovoltaic/thermal systems using

serpentine cooling tubes with different cross-sections of circular,
triangular and rectangular
Amin Shahsavar a, Mehdi Eisapour b, Pouyan Talebizadehsardari c, d, *
a
Department of Mechanical Engineering, Kermanshah University of Technology, Kermanshah, Iran
b
Department of Mechanical Engineering, Mazandaran University of Science and Technology, Babol, Iran
c
Metamaterials for Mechanical, Biomechanical and Multiphysical Applications Research Group, Ton Duc Thang University, Ho Chi Minh City, Vietnam
d
Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City, Vietnam

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

Article history: In this study, an experimental assessment is presented on the effects of employing serpentine tubes with
Received 21 May 2020 three different cross-sections of circular, triangular, and rectangular, on the characteristics of a photo-
Received in revised form voltaic/thermal (PV/T) unit in terms of energy and exergy efficiencies compared with a conventional PV
14 July 2020
system. The influences of adding magnetite nanoparticle to the base-fluid and employing a higher mass
Accepted 15 July 2020
Available online 25 July 2020
flow rate of the cooling fluid are examined. The results demonstrated that by comparing PV and PV/T
units with a circular serpentine tube, the electrical efficiency improves by almost 12% due to adding
coolant tube in addition of getting 22.6W extra thermal energy power. Furthermore, by changing the
Keywords:
Photovoltaic-thermal module
cooling tube configuration from conventional circular form to rectangular, electrical efficiency eases by
Serpentine cooling tubes 2% to reducing PV module temperature. Moreover, it was found that by adding nanoparticle to pure
Magnetite nanoparticle water to employ nanofluid as coolant fluid, overall energy and exergy efficiencies enhance by 6.6% and
Nanofluid 0.7%, respectively, using nanoparticles with a volume concentration of 2% for the case of the rectangular
Mass flow rate serpentine tube for the flow rate of 20 kg/h. Furthermore, enhancing the mass flow rate has a positive
Energy and exergy analysis trend on the PV/T performance in terms of both energy and exergy efficiencies.
© 2020 Published by Elsevier Ltd.

1. Introduction degree of surface temperate of the PV module [9,10]. Thus, the

thermal management of the PV module is very essential.
Recently, sustainable renewable energies have attracted the To mitigate the PV module temperature, channels of heat
researcher’s attention due to fossil fuels’ limitation and also their transfer fluid (HTF) are placed under the PV module to circulate
environmental hazards [1,2]. Different renewable energies such as coolant fluid. Besides, the heated fluid can be employed to provide
water, biomass, solar, geothermal, wind, and nuclear which among the heat load in various applications [11,12]. Therefore, researchers
them, solar energy is considered as the most used source of have been working on the configuration of heat transfer fluid
renewable energy. To harvest and convert solar radiation into channels [13,14] using different types of coolant fluids such as air
electrical energy, the photovoltaic (PV) modules are required. [15], water [15], nanofluids (NFs) [16], microencapsulated PCM
However, some issues are considered regarding PV technology slurry [17], and microencapsulated PCM nano-slurry [18]. The
including efficiency [3], storage as well as environmental condi- thermal performance of PV/T units has been examined in different
tions such as dust [4,5], temperature and humidity [6,7]. Among studies from both energy and exergy points of view toward an
these problems, increasing the surface temperature of PV is proved efficient system [19]. Tomar et al. [20] experimentally and theo-
as a negative factor [8] and reported that the PV electrical energy retically checked out glass-to-glass and glass-to-tedlar PV system
generation is reduced approximately by 0.45% by increasing each with and without duct integration on a test cell to investigate the
efficiency of various layouts of PV/T unit along with their building
implications. It was deduced that the glass to glass PV system has
higher electrical and thermal performance than other tested con-
* Corresponding author. Ton Duc Thang University, Ho Chi Minh City, Vietnam.
figurations. In an experimental study, Aste et al. [21] presented an
E-mail address: [email protected] (P. Talebizadehsardari).

https://doi.org/10.1016/j.energy.2020.118409
0360-5442/© 2020 Published by Elsevier Ltd.
2 A. Shahsavar et al. / Energy 208 (2020) 118409

Nomenclature Tsun Sun temperature (K)

V output voltage of the PV module (V)
Apv surface area of PV module (m2)
cp Specific heat capacity (J/kg.K) Greeks
E_ p pumping power (W) helI electrical energy efficiency (%)
Ir solar radiation intensity (W/m2) helII electrical exergy efficiency (%)
I output current of the PV module (A) hovI overall energy efficiency (%)
m_ mass flow rate (kg/s) hovII overall exergy efficiency (%)
Dp Pressure drop (Pa) hp Pump efficiency (%)
Ta ambient temperature (K) hthI thermal energy efficiency (%)
Tf ;m mean fluid temperature (K) hthI thermal exergy efficiency (%)
Tin inlet temperature of fluid (K) rnf Nanofluid density (kg/m3)
Tout outlet temperature of fluid (K) f Nanofluid concentration (%)

innovative design of a glazed PV/T unit cooled with water. They PV/T units. Besides, a theoretical optimization for nanoparticles
found that both electrical and thermal efficiencies of a simple concentration was performed to achieve the highest PV/T perfor-
module are lower than the novel proposed PV/T unit. Siddiqui et al. mance. It was concluded that the higher heat absorption from the
[22] assessed numerically the produced electrical and thermal PV surface was obtained using Ag/CoSO4 NF in comparison with
energies of a 3D cooled PV module. They showed higher overall pure water. It was theoretically found that when the amount of Ag
efficiency using a cooling system. Three various PV/T unit with nanoparticle is 37 ppm, the best merit function was obtained. To
various water heating systems including direct-coupled PV pump, evaluate the electrical and thermal energies of a PV/T unit, three
traditional pump, and natural flow were conducted experimentally different types of fluids including water, Ag/water, and Alumina/
by Fu et al. [23]. It was demonstrated that the natural circulation water were studied numerically by Khanjari et al. [30]. It was
has a higher effect on the exergy and energy efficiencies of the indicated that using NFs enhances both thermal and electrical ef-
system compared with the other proposed systems which were ficiencies. Besides, the effect of Ag-water NF is much more than
equal to 21.84% and 69.63%, respectively. Furthermore, the results water/alumina NF for different volume concentrations from 28% to
illustrated that utilizing a PV pump results in higher thermal per- 45%. Al-Waeli et al. [31] employed nano-PCM/NF to assess techno-
formance. In another experimental study, the space between the economically the performance of a PV/T unit experimentally. The
absorbing plate and the fluid tube is filled with iron instead of air in maximum thermal and electrical efficiencies were achieved 72%
an innovative tube plate PV/T designed performed by Huo et al. and 13.7%, respectively, and also the output power was found
[24]. From their experiments, it was derived that the photoelectric 12.7 W. In addition, the efficiency of the inverter was 97.3%. In
efficiency of the studied system is 20% higher than the traditional another paper, Al-Waeli et al. [31] evaluated experimentally and
one due to the fact that by employing iron fillings, the surface mathematically a PV/T unit performance using NF coolant and
temperature of the PV system was reduced 3.5e6.5
C compared nano-PCM. Results showed that there was a negligible difference
with that for the air-based system. Eisapour et al. [18] numerically between the experimental and mathematical models. For the
investigated exergy and energy efficiencies of a PV/T unit using a experimental study, the electrical and thermal efficiencies were
new configuration of the PV/T unit including wavy tubes with 13.7% and 72%, respectively whereas they were 13.2% and 71.3% for
different wavelengths and wave amplitudes. It was deduced that the mathematical investigation, respectively.
when wavelength and amplitude were 200 mm and 16 mm, In this paper, a comparative study between traditional PV
respectively, the highest electrical efficiency, 11.18%, and thermal module and PV module with a cooling system known as PV/T unit is
efficiency, 62.4%, were achieved. drawn experimentally using a serpentine tube for the cooling
To achieve higher performance, in combination with other system. Moreover, an investigation on the PV/T unit performance
enhancement techniques, one promising method is to enhance the with different coolant tube cross-sections including circular,
properties of working fluid to enhance the heat transfer rate from triangular and rectangular based on the first and second law of
the PV module [25]. NFs are widely employed in various heat thermodynamics is carried out. To better understand the important
transfer applications to enhance the conductivity of the heat role of cooling fluid on the PV/T units, water as a base fluid is
transfer fluid (HTF) [26]. The thermophysical properties, viscosity substituted by different types of NFs with magnetite nanoparticles
and density of three different sorts of NFs were investigated volume concentration of 0.5%, 1%, 1.5% and 2%. Furthermore, the
experimentally by Al-Waeli et al. [27]. It was reported that the response of the system to the variation of inlet mass flow rate is also
density and viscosity of all examined NFs are more than pure water. examined by increasing the mass flow rate of the coolant from
Moreover, they mentioned that the rise of viscosity and density of 20 kg/h to 80 kg/h for all the proposed cases. To the best of the
the propylene glycol-water and ethylene glycol-water, respectively, authors’ knowledge, this work is the first experimental investiga-
are more than other studied ones. In an experimental study, Multi- tion on the energy and exergy performances of NF-based PV/T units
walls Carbon Nano-Tubes (MWCNTs) in water was considered as equipped with sheet-and-serpentine tube collector with triangular
cooling fluid in a PV/T unit by Abdallah et al. [28]. They reported and rectangular cross-sections.
that the hybrid PV/T unit overall efficiency improves by adding
nanoparticles to pure water due to an increase in thermal con-
ductivity. Besides, various concentrations of NFs were examined 2. System description
and it was found that when the volume concentration of MWCNTs
was 0.075, the highest efficiency was achieved. Han et al. [29] In this study, three different PV/T units including (I) PV/T unit
experimentally explored a new type of COSO4-based Ag for hybrid with a sheet-and-circular serpentine tube collector (case 1), (II) PV/
T unit with a sheet-and-triangular serpentine tube collector (case
 A. Shahsavar et al. / Energy 208 (2020) 118409 3

2), and (III) PV/T unit with a sheet-and-rectangular serpentine tube

collector (case 3) are experimentally designed and built to compare
their performances in terms of energy and exergy. To attach the
collector made from the copper to the back of the PV module,
thermal grease was employed and also serpentine coolant tubes are
welded at the bottom of the collector. Besides, stone wool was used
as thermal insulation to prevent heat dissipation from the tube to
the surrounding. Fig. 1(aec) present the schematic of the studied
PV/T units. For case 1, the inner diameter of the tube is 8 mm with a
thickness of 1 mm. For triangular and rectangular tubes, the di-
mensions are considered so that their hydraulic diameter is equal to
the circular tube.

3. Experimental setup

Indoor tests were chosen and conducted due to variation of

weather conditions and also having quite similar conditions for
different cases. Therefore, solar simulator was developed. The solar
simulator was utilized to provide a controllable indoor test facility
under conditions of a laboratory. It consists of four 10 W LED lamps,
four 400 W metal halide lamps (Reflector Sunlight Dysprosium
Lamp, RSDL), an aluminium box, four DC fans and an AC phase-cut
dimmer. For measuring the radiation intensity, a Trotec BF06 lux
meter was employed. In all the experiments, the incoming solar
radiation was 1000 W/m2. Note that a K-type thermocouple was
used to measure the ambient temperature. For all the experiments,
the ambient temperature inside the simulator was recorded at
22
C. In the present study, the PV module (Eco-Worthy, USA) made
of 72 solar cells with the dimensions of 32 cm  21.5 cm and
nominal output power of 10 W is used. Other characteristics of the
employed PV modules are tabulated in Table 1. Fig. 2 (a) illustrates a
photograph of the PV module. For circulation of the coolant fluid in
the serpentine tubes, a 20 V DC pump with a head of 3.5 m and the
adjustable pumping speed was used. A Malema-M10000 rotary
flow meter located after the pump was employed to measure the
flow rate of the cooling fluid. Also, a Rosemount 3051CD differential
pressure transmitter was employed to measure the pressure drop
of NF inside the serpentine tube. In the present study, the fluid
seemed to move in a closed cycle. The pump was utilized for
circulating fluid into a serpentine tube and also after receiving the
heat of the PV module returns to the reservoir (2 Lit capacity), and
until the steady conditions, the process continues. To compare the
experiments, fluid’s inlet temperature must always be constant. To
achieve this aim, a plate heat exchanger was employed to transfer
the heat of the fluid to the coolant fluid from an 1800 W constant
temperature bath. To obtain a constant inlet temperature, the trial
and error method was employed to realise the required thermal
bath temperature. The real and a schematic view of the utilized
experimental setup are illustrated in Fig. 2(b) and (c), respectively.
K-type thermocouples were employed in 20 points to measure both

Table 1
Specifications of PV module at standard test conditions.

PV cell type Mono-crystalline silicon

Nominal power output (W) 10

Total nos. of PV cell 72
Dimensions 330mm  180mm  20mm
Maximum efficiency 16.0%
Weight 0.82 kg
Fill factor 0.726
Short circuit current 0.69 A
Open circuit voltage 20.6 V
Temperature coefficient of power 0.47%/
C
Temperature range 40 to 80
C
Tempered glass thickness 3.2 mm

Fig. 1. Schematics of the proposed serpentine cooling tubes in the back of PV-T
module.
4 A. Shahsavar et al. / Energy 208 (2020) 118409

Fig. 2. (a) A real photograph of the utilized PV module; (b) A real photograph of the experimental setup and (c) A schematic sketch of the experimental setup.
 A. Shahsavar et al. / Energy 208 (2020) 118409 5

Table 2
Detailed specifications of the instruments used in the investigation.

Instrument Model Operating range Accuracy

Thermometer K-type 200
Ce1260
C ±0.75%

Temperature sensor TES-1327 K 150
Ce1260
C ±0.1% 1
C
Luxmeter TROTEC-BF06 0e40 klx ±5% ± 10 lx
Multimeter DC current LUTURON-DW6060 0e10 A ±1%
Multimeter DC voltage LUTURON-DW6060 0e600 V ±0.8%
Rotary flow meter Malema-M10000 0.1 to 225 Lit/min ±2%
Pressure transmitter Rosemount 3051CD 0-300 psi ±1%
Liquid density gravity meter KEN-DA130N 0e2 g/cm3 ±0.001 g/cm3

Fig. 3. Water/magnetite NF preparation through various steps [32].

the inlet and outlet temperatures of the coolant fluid and the PV 4. 50 mL of 0.7 M ammonia hydroxide under vigorous stirring is
front surface’s temperature. The thermocouples were connected to added dropwise to the solution achieved in step 3 for 10 min at
a JUMO data logger. The voltage and current of the PV module were 700 rpm.
determined by a LUTRUN multi-meter. Table 2 tabulates more de- 5. The solution received in step 4 contains magnetic nanoparticles
tails about the used equipment. which are then collected at the bottom of a container by placing
it over a permanent magnet.
6. 8 mL of 25% tetramethylammonium hydroxide, as well as
magnetite particles, then underwent the centrifugation process
for 1 min at 4000 rpm.
4. NF preparation, characterization and thermophysical
properties Fig. 3 depicts the aforementioned discussed steps. The details of
the characterization of NF provided by this method are presented in
In this experimental investigation, water/magnetite is consid- Refs. [32,33], and for the sake of brevity, we have avoided repeating
ered as the NF. To make NF, the co-precipitation method is them here. To evaluate the stability of the prepared NF samples, we
employed [32,33]. In the preparation of the water/magnetite NF, measured the density of a certain amount of NFs during various
some distinct steps involved and detailed as follows: experiments. For this purpose, a liquid density gravity meter (DA-
130N, KEN, Japan) is employed. The results showed that the density
1. 67.58 g of FeCL3.6H2O mixed with 100 mL of 2M HCL (Hydro- changes were negligible. The thermophysical aspects of the
chloric acid). considered NF samples are reported in Table 3. Thermal conduc-
2. 39.76 g of FeCL2.4H2O mixed with 100 mL of 2M HCL. tivity of NF was measured using a KD2-Pro thermal properties
3. The solutions collected during steps 1 and 2 are mixed in a analyzer (Decagon devices, USA) that operates based on the tran-
proportion of 4 to 1 for 2 min at 100 rpm with the help of a sient hot-wire technique. This device has a probe (KS-1) that must
magnetic stirrer.

Table 3
Thermophysical properties of the utilized NF samples (aqueous magnetite).

NF concentration (%) Particle diameter (nm) Heat capacity (J/kgK) Density (kg/m3) Thermal conductivity (W/mK)

0 e 4180.00 997.00 0.610

0.5 13 3446.61 1194.65 0.695
1 13 2921.44 1392.30 0.873
1.5 13 2526.83 1589.95 0.917
2 13 2219.49 1787.60 0.961
6 A. Shahsavar et al. / Energy 208 (2020) 118409

be placed in the solution for 90 s during the test. To be more precise, electrical energy is called high-grade energy. The overall energy
the thermal conductivity measurement was repeated 5 times at 20- efficiency is calculated as [15]:
min intervals (to allow thermal stability), and the value reported in
helI
Table 3 is actually the average of the values obtained from the 5 hovI ¼ hthI þ (6)
experiments. To calibrate the device, the thermal conductivity of 0:36
glycerol and water was measured at 20
C and the obtained values
where 0.36 is the electrical-power generation efficiency related to a
(0.274 W/m.K for glycerol and 0.586 W/m.K for water) were
thermal power plant [39].
compared with the values in a valid source (0.285 W/m.K for
Exergy efficiency is calculated according to the maximum
glycerol and 0.598 W/m.K for water) [34]. The difference between
theoretically-available and actual desired output [40] As mentioned
the experimental results and the values reported in ref. [34] was
recently, electrical energy is high-grade energy; thus, electrical
less than 3.9%. In addition, the density (rnf ) and specific heat (cp;nf )
exergy is equal to electrical energy. The ratio of the electrical power
of NF were computed using the following equations [35]: output to the solar radiation exergy is defined as the electrical
exergy efficiency of the PV/T units [41]:
rnf ¼ ð1  fÞrw þ frp (1)
VI  E_ p
helII ¼ "    4 # (7)
ð1  fÞrw cp;w þ frp cp;p
cp;nf ¼ (2) 1  3 Tsun þ 3 TTsun
4 Ta 1
rnf Ir Apv a

where f is the volume concentration of NF and the subscripts nf ,

where Ta and Tsun (y5800 K) are the ambient and sun tempera-
wand p stand for the NF, water and nanoparticles. The density and
tures, respectively.
specific heat of magnetite nanoparticles are respectively 5200 kg/
The ratio of thermal exergy output to the solar radiation exergy
m3 and 670 J/kg.K and the specific heat of water is 4179 J/kg.K [35].
is the thermal exergy efficiency given as [42]:
It is noteworthy that, in addition to the advantages of nanofluid
on increasing the thermal conductivity of nanofluid, there are some " #
drawbacks including poor long-term stability of nanoparticle in _ p ðTout  Tin Þ 1  TTa
mc f ;m
suspension, which is the main disadvantage of using nanofluids "
hthII ¼    4 # (8)
[36]. Furthermore, higher pressure drop and pumping power, lower 4 Ta 1 Ta
specific heat compared with the base fluid (as presented in Table 3) Ir Apv 1  3 Tsun þ 3 Tsun
and high cost of nanoparticles are among other disadvantages of
employing nanoparticles [36,37].
where Tf ;m is the mean fluid temperature (¼Tin þT
2
out
).
The overall exergy efficiency is then given as:
5. Energy and exergy analysis
hovII ¼ hthII þ helII (9)
Since PV/T units are capable of producing electricity and heat at
the same time, there are two different categories for analysing their
energies which are electrical and thermal performances. Electrical
energy efficiency is used to measure the electrical performance, 6. Uncertainty analysis
which is the ratio of the electrical power output of the system to the
rate of incoming solar energy [16]: The accuracy and reliability of the results are one of the most

VI  E_ p
helI ¼ (3)
Ir Apv

where Ir is the intensity of solar radiation, Apv is the PV area, V is the

PV output voltage, I is the PV output current and E_ p is the electric
power needed for pumping the fluid defined as [38]:

m_ Dp
E_ p ¼ (4)
rnf hp

where Dpis the pressure drop of fluid flowing inside the PV/T unit
and hp is the pump efficiency (¼75%).
The thermal performance is measured with the use of thermal
efficiency of the system, which is the ratio of the heat transferred to
the fluid passing through the system to the rate of incoming solar
energy [12]:

_ p ðTout  Tin Þ
mc
hthI ¼ (5)
Ir Apv

where cp is the specific heat capacity, Tin is the inlet temperature and
Fig. 4. Effect of adding coolant tube under the PV unit on the (a) electrical efficiency
Tout is the outlet temperature related to the fluid. and (b) temperature of PV module for the PV/T unit with a sheet-and-circular
Thermal energy can easily be achieved from electrical energy of serpentine tube collector (case 1) compared with the PV module without a cooling
almost 100%, while the opposite is not possible. Therefore, system.
 A. Shahsavar et al. / Energy 208 (2020) 118409 7

Fig. 7. Electrical and thermal exergy efficiencies of the PV/T unit with a sheet-and-
serpentine tube collector with circular (case 1), triangular (case 2), and rectangular
(case 3) cross-sections.

Fig. 5. The average PV module temperature for the PV/T unit with a sheet-and-
serpentine tube collector with circular (case 1), triangular (case 2), and rectangular 7. Result and discussion
(case 3) cross-sections.
In the following, the performance features of the PV module
without cooling and the PV/T unit with a sheet-and-circular
serpentine tube collector is presented comparatively. Then, the
effects of changing the cross-section of the serpentine tube are
carried out. In these sections, the flow rate is considered to be equal
to 20 kg/h, and pure water is used as the coolant. Afterwards, the
influences of the volume concentration of magnetite nanoparticles
are examined. Finally, sensitivity analysis on mass flow rate for the
best case is carried out to better understand its influence on the
system’s performance.

7.1. Effect of employing cooling fluid

The electrical efficiency and temperature of the conventional PV

module compared with those of the PV/T unit with a sheet-and-
circular serpentine tube collector (case 1) are represented in

Fig. 6. Electrical and thermal efficiencies of the PV/T unit with a sheet-and-serpentine
tube collector with circular (case 1), triangular (case 2), and rectangular (case 3) cross-
sections.

significant problems which to be evaluated in any experimental

research. In this work, the well-known and widely used Moffat
method [43] is used to study the uncertainty of results. According
to this method, the uncertainty of a dependent variable (R) is a
function of the uncertainty of the independent variables (v1 , v2 , …,
vn ) and is calculated using the following equation:

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2  2  2
vR vR vR
dR ¼ dv1 þ dv2 þ … þ dvn (10)
vv1 vv2 vvn

where dshows a parameter’s uncertainty. The results illustrated

that the uncertainty of helI , hthI , helII and hthII were less than Fig. 8. Influence of changing geometrical specifications on the electrical and thermal
5.16%, 4%, 5.54% and 4.43%, respectively, which are acceptable. The energy outputs of the PV/T unit with a sheet-and-serpentine tube collector with cir-
details of the uncertainty calculations are given in “Appendix A”. cular (case 1), triangular (case 2), and rectangular (case 3) cross-sections.
8 A. Shahsavar et al. / Energy 208 (2020) 118409

Fig. 4-a and 4-b, respectively. As seen from the figure, the electrical
efficiency of the PV/T unit is higher than that of the without the
cooled case. The higher electrical efficiency of the PV module will
be obtained when the surface temperature is lower. Besides, the
electrical efficiency enhances approximately 12% by using the
cooling tube since the PV module temperature in the PV/T unit
declines by 14 K compare with the conventional system. Thus,
employing the cooling tube at the bottom of the PV unit is highly
beneficial.

7.2. Effect of geometrical specifications on the performance of the

PV/T unit

In this part, the influence of two different configurations of the

serpentine tube (i.e. triangular and rectangular) on the system’s
performance is examined and the outcomes are compared with
those of the conventional circular one. The average PV module
temperature of the PV/T unit with the circular serpentine tube (case
1) in comparison with the triangular (case 2) and rectangular (case
3) tubes is illustrated in Fig. 5. As shown, by using triangular tube
instead of cylindrical one, the average temperature of the PV Fig. 10. Effect of volume concentration of magnetite nanoparticles on the electrical
module increases by 3 K; however, it drops by almost 2 K when the and thermal efficiencies of the PV/T unit with a sheet-and-rectangular serpentine tube
collector (case 3).
rectangular tube is replaced with the cylindrical one. This can be
because of the fact that the heat transfer surface area of the case 2
and case 3 compared to case 1 decline and rises by 27.3% and 4.5%,
respectively.
Fig. 6 presents the electrical and thermal efficiencies for cases
1e3. As is shown, both the electrical and thermal efficiencies
decline by roughly 3% and 12% from case 1 to case 2, respectively.
However, by changing the tube from the circular (case 1) to the
rectangular configuration (case 3), both the electrical and thermal
efficiencies improve. According to Florschuetz equation [8], the
lower the temperature of the PV module, the higher its efficiency.
For all the cases examined in the current contribution, the findings
reveal that the pumping power required to flow fluid in the system
is in the range of 0.01e0.05 W and is negligible compared to the
electric output of the system. Therefore, it can be said that the
similarity of the pattern of electrical efficiency changes of the
studied PV/T units and PV module temperature is reasonable and
acceptable. On the other hand, the results confirm that the
maximum temperature of the fluid leaving the PV/T unit respec-
tively belongs to case 1, case 3 and case 2 which makes the thermal

Fig. 11. Variation of the thermal and electrical energy outputs of the PV/T unit with a
sheet-and-rectangular serpentine tube collector (case 3) in terms of NF concentration.

efficiency of case 1 more than other systems under study.

The electrical and thermal exergy efficiencies are illustrated in
Fig. 7 for cases under investigation. As is displayed, both the elec-
trical and thermal efficiencies improve by using rectangular
serpentine tube compared with that for the circular and triangular
ones due to the rise of the heat transfer surface. The electrical and
thermal exergy efficiencies, respectively, increase by 2% and 6% for
case 3 compared to case 1.
The generated net electrical and thermal powers for various
proposed PV/T units are represented in Fig. 8. The outcomes show
that the net electrical power increases by almost 1.1% and 3.5% from
case 1 to 3 and from case 2 to 3, respectively. This figure also depicts
that the PV/T unit with rectangular tube has the highest thermal
power compared to other ones, which is due to a higher decrease in
Fig. 9. Effect of volume concentration of magnetite nanoparticles on the PV module
the PV temperature.
temperature for the PV/T unit with a sheet-and-rectangular serpentine tube collector
(case 3).
 A. Shahsavar et al. / Energy 208 (2020) 118409 9

Fig. 14. Influence of mass flow rate on the PV temperature for the PV/T unit with a
sheet-and-serpentine tube collector with circular (case 1), triangular (case 2), and
rectangular (case 3) cross-sections.
Fig. 12. Overall energy and exergy efficiencies of the PV/T unit with a sheet-and-
rectangular serpentine tube collector (case 3) versus volume concentration of NF.

7.3. Effect of volume concentration of nanoparticles

In this section, first, the effect of adding nanoparticles to the PV/

T unit is examined for case 3 with rectangular serpentine cooling
tubes as the best case comprehensively. Then, the influences of
nanoparticles addition for different proposed geometries are
assessed comparatively. As mentioned, four various concentrations
of magnetite nanoparticles (0.5%, 1%, 1.5%, and 2%) are studied.
The effect of NF concentration on the PV module temperature
for case 3 is presented in Fig. 9 for the mass flow rate of 20 kg/h. It is
seen that increasing the NF concentration leads to a lower module
temperature, which is due to higher conductivity of the NF by
increasing the concentration of the nanoparticles (please see
Table 3). As shown, for the NF with 2% nanoparticles, the PV module
temperature reduces by 0.54
C compared with pure water. It has
been also reported in the literature that the use of NF modifies the
cooling performance of the PV modules. Sardarabadi et al. [44]
reported that the use of different NFs reduces the PV temperature
in the range of 0.03e0.85
C compared with pure water using
Fig. 13. Influence of NF concentration on the PV temperature for the PV/T unit with a different nanoparticles when the mass flow rate of the cooling fluid
sheet-and-serpentine tube collector with circular (case 1), triangular (case 2), and is 30 kg/h. Note that the panel temperature is 11
C and the
rectangular (case 3) cross-sections.

Table 4
Effect of NF concentration on the performance of PV/T unit with a sheet-and-serpentine tube collector with circular (case 1), triangular (case 2), and rectangular (case 3) cross-
sections.

NF concentration (%) Electrical efficiency (%) Thermal efficiency (%) Electrical exergy efficiency (%) Thermal exergy efficiency (%)

Case 1 0 12.06 32.88 12.68 0.85

0.5 12.07 33.59 12.70 0.86
1 12.08 35.08 12.71 0.88
1.5 12.1 36.6 12.73 0.89
2 12.12 38.03 12.76 0.9
Case 2 0 11.79 29.2 12.4 0.78
0.5 11.8 29.9 12.42 0.79
1 11.82 31.7 12.43 0.82
1.5 11.84 32.6 12.45 0.83
2 11.86 34.2 12.48 0.85
Case 3 0 12.19 35.4 12.83 0.90
0.5 12.2 36.1 12.84 0.904
1 12.22 37.2 12.85 0.91
1.5 12.24 39.3 12.87 0.93
2 12.26 40.5 12.9 0.94
10 A. Shahsavar et al. / Energy 208 (2020) 118409

electrical efficiency is almost 13.4% in the case of pure water in the

study of Sardarabadi et al. [44] where the electrical efficiency
enhanced to 13.58% for the best NF which was ZnO/water.
In Fig. 10, the electrical and thermal efficiencies are depicted for
different concentrations of nanoparticles for case 3. The trends of
both the electrical and thermal efficiencies are similar and by
enhancing the concentration of nanoparticles, both efficiencies
improve. As shown, by adding magnetite nanoparticles up to 2%,
the electrical and thermal efficiencies enhance from 12.1% to 12.26%
and 35.4%e40.5%, respectively.
The variation of electrical and thermal energy outputs of case 3
with the volume concentration of magnetite nanoparticles is
illustrated in Fig. 11. By employing magnetite nanoparticles, the
electrical energy power enhances from 8.3 W for the water to
8.4 W, 8.41 W, 8.43 W, and 8.44 W for the NF with 0.5%, 1%, 1.5%, and
2% volume concentration of magnetite nanoparticles, respectively.
This figure also depicts that the thermal energy power increases
from 24.36 W for pure water to 28.87 W by adding nanoparticles up
to 2%.
Fig. 12 shows the overall energy and exergy outputs of case 3
versus NF concentration. It is seen that both the efficiencies
augment with the rise of NF concentration. According to the find-
ings, the overall energy and exergy efficiencies of case 3 at a volume
concentration of 2% are respectively 6.6% and 0.7% compared to the
water-based PV/T unit.
The influence of volume concentration of NF on the PV module
temperature is represented in Fig. 13 for all the studied geometries
to show the effects of the nanoparticles when the configuration of
the cooling tube is varied. Adding nanoparticles and increasing its
amount result in decreasing temperature of the PV module in all
cases, since the thermal conductivity of the carrier fluid enhances b.
By adding nanoparticle from 0 to 2%, the PV module temperature
decreases approximately 0.2% for all proposed cases.
The impacts of NF concentration on the performance metrics of
the studied NF-based PV/T units are presented in Table 4. The effect
of NF concentration on all the parameters features of the studied
PV/T units is desirable. As shown, when nanoparticles are added to
the base fluid (water), the PV/T unit performance with various
layouts improves in relation to energy and exergy efficiencies. The Fig. 15. Variation of the electrical and thermal output powers of PV/T unit with a
sheet-and-serpentine tube collector with circular (case 1), triangular (case 2), and
electrical efficiency enhances by roughly 0.07% for all proposed
rectangular (case 3) cross-sections in terms of mass flow rate.
cases. Thus, adding nanoparticles have an almost similar effect on
different configuration according to the electrical efficiency. The
thermal efficiency for case 1, case 2 and case 3 increases by almost 7.4. Effect of mass flow rate
5.15%, 5 and 5.1%, respectively, by using higher NF concentration
from 0 to 2% which is almost similar for all the cases. The behaviour To understand the effect of mass flow rate on different proposed
of increasing NF concentration is also similar to the exergy point of cases, four flow rates of 20, 40, 60, and 80 kg/s are experimentally
view. tested for different NF concentrations. Note that since the effect of
NF concentration was studied in the previous section, the results for

Table 5
Effect of mass flow rate of NF with 2% volume concentration on the performance of PV/T unit with a sheet-and-serpentine tube collector with circular (case 1), triangular (case
2), and rectangular (case 3) cross-sections.

Mass flow rate (kg/h) Electrical efficiency (%) Thermal efficiency (%) Electrical exergy efficiency (%) Thermal exergy efficiency (%)

Case 1 20 12.129 38.03 12.762 0.9

40 12.124 41.43 12.757 1.19
60 12.125 44.37 12.758 1.35
80 12.131 47.37 12.764 1.48
Case 2 20 11.863 34.2 12.482 0.854
40 11.858 36.5 12.477 1.075
60 11.859 38.8 12.478 1.2
80 11.865 41.4 12.485 1.311
Case 3 20 12.263 40.5 12.903 0.94
40 12.258 44.9 12.898 1.26
60 12.259 48.4 12.896 1.45
80 12.265 51.7 12.905 1.6
 A. Shahsavar et al. / Energy 208 (2020) 118409 11

NF concentration of 2% according to thermal and electrical effi-

ciencies. Increasing the mass flow rate, on the one hand, leads to
the intensified cooling of PV module and, as a result, improves the
electrical performance, but on the other hand, the pumping power
also rises for a higher mass flow rate. Fig. 16 shows that up to mass
flow rate of 40 kg/h, the impact of increasing mass flow rate on the
system’s electrical output is less than its effect on the pumping
power, while the opposite is true for higher mass flow rates. Ac-
cording to the results, increasing the mass flow rate from 20 to
40 kg/h decreases the electrical efficiency by only 0.04%. However,
electrical efficiency enhances just 0.01% by rising the mass flow rate
from 40 to 80 kg/h. In addition, Fig. 16 displays that the higher mass
flow rate entails 27.65% increases in the thermal efficiency of case 3.

8. Conclusion

Four distinct solar photovoltaic systems including traditional PV

module, PV/T unit with circular, triangular and rectangular
serpentine tubes were analysed experimentally based on the first
and second law of thermodynamics. Furthermore, the influences of
Fig. 16. Effect of mass flow rate of NF on the (a) electrical and (b) thermal efficiencies adding magnetite nanoparticles to the base carrier fluid (water)
of PV/T unit with a sheet-and-serpentine tube collector with circular (case 1), trian- with different concentrations as well as increasing the inlet mass
gular (case 2), and rectangular (case 3) cross-sections.
flow rate of the cooling fluid on the performance of the system were
studied. The main conclusions of the current study which can be
the NF concentration of 2% is presented in the following. derived from the results are made as follows:
The influence of increasing inlet mass flow rate of NF with 2%
volume concentration of NF from 20 kg/h to 80 kg/h on the systems,  The comparison of traditional PV and PV/T units reveals that by
performance is illustrated in Fig. 14. The responses of different employing coolant fluid in PV/T unit, 14 K temperature drop is
types of PV/T units to the rise in the mass flow rate is almost similar obtainable for PV-T system with circular serpentine tube
and as displayed, the relation between the PV module temperature compared with the PV module. Consequently, the measured
and the inlet mass flow rate is similar. The higher mass flow rate electrical efficiency of the system is promoted from 10.94% for
leads to lower PV module temperature. For cases 1 to 3, the PV the PV to 12.21% for the PV/T unit.
module temperature reduces by nearly 0.67 K using a higher mass  By substituting the conventional circular with the triangular and
flow rate from 20 kg/h to 80 kg/h. It has been also reported in the rectangular cross-sections of the serpentine coolant tube, higher
literature that increasing the mass flow rate of the cooling fluid and lower PV module temperature are achieved, respectively,
results in a lower temperature of the PV module. In the study of which result in 3% and 2% decrease and increase, respectively, in
Fudholy et al. [45], the effect of flow pattern under the PV module electrical efficiency. Moreover, exergy analysis shows that the
was studied and it was reported that for the spiral flow, the PV overall exergy efficiencies of the PV/T units with circular,
temperature reduces from 50.86
C to 46.24
C when the mass flow triangular and rectangular are 13.54%, 13.19% and 13.73%,
rate increases from 0.011 kg/s to 0.041 kg/s using water as the respectively, for water as the coolant fluid. Thus, it is concluded
working fluid. that the utilization of the rectangular tubes leads to more
Table 5 presents the electrical and thermal efficiencies from the effective cooling of the PV/T unit.
energy and exergy points of view for different mass flow rates for  Water/magnetite NF with 2% volume concentration of NF pos-
the proposed systems. From 20 kg/h to 40 kg/h, the electrical and sesses the highest ability for being employed as a cooling fluid
electrical exergy efficiencies have downward trends while for the compared with the NF volume concentration of 0% (pure water),
further increments, they show upward trends. On balance, the 0.5%, 1% and 1.5%. The highest electrical, thermal, electrical
electrical and electrical exergy efficiencies for all cases are pro- exergy and thermal exergy efficiencies belong to case 3 which
moted roughly by 0.002% using the flow rate of 80 kg/h compared are 12.24%, 39.3%, 12.87% and 0.93%, respectively.
with 20 kg/h. Similar to the PV module temperature, the thermal  It is concluded that the effect of increasing the inlet mass flow
energy and exergy efficiencies have consistent behaviours when rate using Water/magnetite NF with 2% volume concentration is
the mass flow rate rises. Note that the amount of enhancement constructive. For the best configuration, by increasing the mass
varies in the three studied cases. Quantitatively, for cases 1, 2 and 3, flow rate from 20 kg/h to 80 kg/h the PV module temperature
the thermal efficiency increases by 9.34%, 7.2% and 11.2%, respec- reduces 0.68 K and the electrical, thermal, electrical exergy and
tively, and the thermal exergy efficiency promotes by 0.58%, 0.457% thermal exergy efficiencies increase 0.002%, 11.2%, 0.002% and
and 0.66%, respectively. 0.66%, respectively.
Effect of increasing inlet mass flow rate on the electrical and
thermal power of the proposed systems is illustrated in Fig. 15. Both Declaration of competing interest
electrical and thermal powers improve using a higher amount of
mass flow rate due to the higher PV temperature reduction. The The authors declare that they have no known competing
electrical power of the system enhances by almost 0.03% from 20 to financial interests or personal relationships that could have
80 kg/h for all proposed cases. However, the thermal power in cases appeared to influence the work reported in this paper.
1, 2, and 3 almost develop by 25%, 22%, and 28%, respectively.
To study the influence of the mass flow rate on case 3 as the best Appendix A. Details of uncertainty analysis
configuration, Fig. 16 shows the effect of the mass flow rate for the
- Pumping power:
12 A. Shahsavar et al. / Energy 208 (2020) 118409

 
E_ p ¼ f m;
_ Dp; rnf ; hp

 
uE_p ¼ f um_ ; uDp ; urnf ; uhp
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 2  _ 2 !2  2
u dE_ dEp dE_ p dE_ p
¼ t p
um_ þ uD p þ urnf þ uhp
dm_ dDp drnf dhp

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u !2  
uE_puu 2 u 2 urnf uhp 2
¼t m þ
_ Dp
þ þ ¼ 3:0%
_
Ep m _ Dp rnf hp

- Electrical energy efficiency:

 
helI ¼ f V; I; E_ p ; Ir ; Apv

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  u 2 dh 2 dh 2  2  2
u dh dhelI dhelI
uhelI ¼ f uV ; uI ; uE_p ; uIr ; uApv ¼t elI
uV þ elI
uI þ elI
uE_p þ u Ir þ uApv
dV dI dE_ p dIr dApv

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ffi
uhelI 100  2 E_  VI 2 E_  VI
2 2 p p
¼ ðIuV Þ þ ðVuI Þ þ  uE_p þ uIr þ uApv ¼ 5:16%
helI Ir Apv helI Ir Apv

- Electrical exergy efficiency:

 
- Thermal energy efficiency: helII ¼ f V; I; E_ p ; Ir ; Apv ; Ta

 
 
_ cp ; Tout ; Tin ; I r ; Apv
hthI ¼ f m; uhelII ¼ f uV ; uI ; uE_p ; uIr ; uApv ; uTa

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  dh 2 dh 2 
dhthI
2 
dhthI
2 
dhthI
2 
dhthI
2
thI thI
uhthI ¼ f um_ ; ucp ; uTout ; uTin ; uIr ; uApv ¼ um_ þ ucp þ uTout þ uTin þ u Ir þ uApv
dm_ dcp dTout dTin dIr dApv

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uhthI  u  2 u 2  u 2 
uTin
2  2 
uIr uApv 2

m_ cp Tout
¼ þ þ þ þ þ ¼ 4%
hthI m_ cp Tout  Tin Tout  Tin Ir Apv
 A. Shahsavar et al. / Energy 208 (2020) 118409 13

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 2 dh 2 dh 2  2  2  2
u dh dhelII dhelII dhelII
¼t elII
uV þ elII
uI þ elII
uE_p þ u Ir þ uApv þ uTa
dV dI dE_ p dIr dApv dTa

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u
u   _ 2  _ 2 " ! #2
uhelII 100 tðIu Þ2 þðVu Þ2 þ u 2 þ Ep VIu Ep VI 4T 3sun þ12T 3a
¼ "    4 # V I E_p Ir þ u þ uTa ¼5:54%
helII Ir Apv Apv 3T 4sun
Ta Ta
Ir Apv helI 1 43 Tsun þ 13 Tsun

- Thermal exergy efficiency: multicriteria optimization of different arrangements of integrated photovol-

 
_ cp ; Tout ; Tin ; I r ; Apv ; Ta
hthII ¼ f m;

It should be noted that among the experiments performed in the

 
uhthII ¼ f um_ ; ucp ; uTout ; uTin ; uIr ; uApv ; uTa
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
dh 2 dh 2 
dhthII
2 
dhthII
2 
dhthII
2 
dhthII
2 
dhthII
2
thII thII
¼ um_ þ ucp þ uTout þ uTin þ uIr þ uApv þ uTa
dm_ dcp dTout dTin dIr dApv dTa
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uhthII  u  2 u c 2  u 2 
uTin
2  2 
uIr uApv 2
 
dhthII uTa 2

m_ Tout
¼ þ p
þ þ þ þ þ ¼ 4:43%
hthII m_ cp Tout  Tin Tout  Tin Ir Apv dTa hthII

present study, the highest amount of uncertainty belongs to the taic thermal and heat recovery wheel system. Int J Energy Res 2020;44:
1488e505.
following case:
[9] Nkurikiyimfura I, Wang Y, Safari B, Nshingabigwi E. Electrical and thermal
rnf ¼ 997 kg/m3, cp ¼ 4180 J/kg.K, m_ ¼ 20 kg/h, V ¼ 4.632 V, performances of photovoltaic/thermal systems with magnetic nanofluids: a
I ¼ 1.811 A, Tout ¼ 295.05 K, helI ¼ 12.193, hthI ¼ 35.4, helII ¼ review. Particuology 2020. https://doi.org/10.1016/j.partic.2020.04.004. In
Press.
12.83, hthII ¼ 0.9. [10] He W, Barzinjy AA, Khanmohammadi S, Shahsavar A, Moghimi MA, Afrand M.
Multi-objective optimization of a photovoltaic thermal-compound sensible
rotary heat exchanger system using exergo-economic and enviro-economic
References approaches. J Environ Manag 2020;254:109767.
[11] Afrand M, Shahsavar A, Sardari PT, Sopian K, Salehipour H. Energy and exergy
[1] Maina N, Murray J, McKenzie M. Climate change and the fossil fuel divestment analysis of two novel hybrid solar photovoltaic geothermal energy systems
movement in Canadian higher education: the mobilities of actions, actors, and incorporating a building integrated photovoltaic thermal system and an earth
tactics. J Clean Prod 2020;253:119874. air heat exchanger system. Sol Energy 2019;188:83e95.
[2] Shakouri HG. The share of cooling electricity in global warming: estimation of [12] Shahsavar A, Salmanzadeh M, Ameri M, Talebizadeh P. Energy saving in
the loop gain for the positive feedback. Energy 2019;179:747e61. buildings by using the exhaust and ventilation air for cooling of photovoltaic
[3] Shahsavar A, Moayedi H, Al-Waeli AHA, Sopian K, Chelvanathan P. Machine panels. Energy Build 2011;43(9):2219e26.
learning predictive models for optimal design of building-integrated photo- [13] Farshchimonfared M, Bilbao J, Sproul A. Channel depth, air mass flow rate and
voltaic-thermal collectors. Int J Energy Res 2020;44:5675e95. air distribution duct diameter optimization of photovoltaic thermal (PV/T) air
[4] Pan A, Lu H, Zhang LZ. Experimental investigation of dust deposition reduction collectors linked to residential buildings. Renew Energy 2015;76:27e35.
on solar cell covering glass by different self-cleaning coatings. Energy [14] Shan F, Tang F, Cao L, Fang G. Comparative simulation analyses on dynamic
2019;181:645e53. performances of photovoltaicethermal solar collectors with different config-
[5] Chen J, Pan G, Ouyang J, Ma J, Fu L, Zhang L. Study on impacts of dust accu- urations. Energy Convers Manag 2014;87:778e86.
mulation and rainfall on PV power reduction in East China. Energy 2020;194: [15] Shahsavar A, Ameri M. Experimental investigation and modeling of a direct-
116915. coupled PV/T air collector. Sol Energy 2010;84(11):1938e58.
[6] Al-Waeli A, Chaichan M, Sopian K, Kazem H. Comparison study of indoor/ [16] Sardarabadi M, Hosseinzadeh M, Kazemian A, Passandideh-Fard M. Experi-
outdoor experiments of SiC nanofluid as a base-fluid for a photovoltaic mental investigation of the effects of using metal-oxides/water nanofluids on
thermal PV/T system enhancement. Energy 2018;151:33e44. a photovoltaic thermal system (PVT) from energy and exergy viewpoints.
[7] Hu H, Yuan D, Wang T, Jiang Y. Dynamic performance of high concentration Energy 2017;138:682e95.
photovoltaic/thermal system with air temperature and humidity regulation [17] Qiu Z, Ma X, Zhao X, Li P, Ali S. Experimental investigation of the energy
system (HCPVTH). Appl Therm Eng 2019;146:577e87. performance of a novel Micro-encapsulated Phase Change Material (MPCM)
[8] Shahsavar A, Khanmohammadi S. Energy and economic evaluation and slurry based PV/T system. Appl Energy 2016;165:260e71.
14 A. Shahsavar et al. / Energy 208 (2020) 118409

[18] Eisapour M, Eisapour AH, Hosseini M, Talebizadehsardari P. Exergy and en- [32] Shahsavar A, Khanmohammadi S, Afrand M, Shahsavar A, Rostami S. On
ergy analysis of wavy tubes photovoltaic-thermal systems using micro- evaluation of magnetic field effect on the formation of nanoparticles clusters
encapsulated PCM nano-slurry coolant fluid. Appl Energy 2020;266:114849. inside aqueous magnetite nanofluid: an experimental study and compre-
[19] Shahsavar A, Ameri M, Gholampour M. Energy and exergy analysis of a hensive modeling. J Mol Liq 2020;312:113378.
photovoltaic-thermal collector with natural air flow. J Sol Energy Eng [33] Shahsavar A, Salimpour MR, Saghafian M, Shafii MB. Experimental investi-
2011;134(1). gation on laminar forced convective heat transfer of ferrofluid loaded with
[20] Tomar V, Norton B, Tiwari G. A novel approach towards investigating the carbon nanotubes under constant and alternating magnetic field. Exp Therm
performance of different PVT configurations integrated on test cells: an Fluid Sci 2016;76:1e11.
experimental study. Renew Energy 2019;137:93e108. [34] https://webbook.nist.gov/.
[21] Aste N, Leonforte F, Del Pero C. Design, modeling and performance monitoring [35] Sheikholeslami M, Ganji DD. Ferrohydrodynamic and magnetohydrodynamic
of a photovoltaicethermal (PVT) water collector. Sol Energy 2015;112:85e99. effects on ferrofluid flow and convective heat transfer. Energy 2014;75:
[22] Siddiqui MU, Arif AF, Kelley L, Dubowsky S. Three-dimensional thermal 400e10.
modeling of a photovoltaic module under varying conditions. Sol Energy [36] Dharmalingam R, Sivagnanaprabhu KK, Senthilkumar B, Thirumalai R. Nano
2012;86(9):2620e31. materials and nanofluids: an innovative technology study for new paradigms
[23] Fu H, Li G, Li F. Performance comparison of photovoltaic/thermal solar water for technology enhancement. Procedia Engineering 2014;97:1434e41.
heating systems with direct-coupled photovoltaic pump, traditional pump [37] Solangi KH, Kazi SN, Luhur MR, Badarudin A, Amiri A, Sadri R, Zubir MNM,
and natural circulation. Renew Energy 2019;136:463e72. Gharehkhani S, Teng KH. A comprehensive review of thermo-physical prop-
[24] Huo Y, Lv J, Li X, Fang L, Ma X, Shi Q. Experimental study on the tube plate PV/ erties and convective heat transfer to nanofluids. Energy 2015;89:1065e86.
T system with iron filings filled. Sol Energy 2019;185:189e98. [38] Yazdanpanahi J, Sarhaddi F, Mahdavi Adeli M. Experimental investigation of
[25] Huaxu L, Fuqiang W, Dong Z, Ziming C, Chuanxin Z, Bo L, Huijin X. Experi- exergy efficiency of a solar photovoltaic thermal (PVT) water collector based
mental investigation of cost-effective ZnO nanofluid based spectral splitting on exergy losses. Sol Energy 2015;118:197e208.
CPV/T system. Energy 2020;194:116913. [39] Joshi AS, Tiwari A, Tiwari GN, Dincer I, Reddy BV. Performance evaluation of a
[26] Mahian O, Kolsi L, Amani M, Estelle P, Ahmadi G, Kleinstreuer C, Marshall JS, hybrid photovoltaic thermal (PV/T)(glass-to-glass) system. Int J Therm Sci
Siavashi M, Taylor RA, Niazmand H, Wongwises S, Hayat T, Kolanjiyil A, 2009;48:154e64.
Kasaeian A, Pop I. Recent advances in modeling and simulation of nanofluid [40] Aberoumand S, Ghamari S, Shabani B. Energy and exergy analysis of a
flows-part I: fundamentals and theory. Phys Rep 2019;790:1e48. photovoltaic thermal (PV/T) system using nanofluids: an experimental study.
[27] Al-Waeli AH, Chaichan MT, Sopian K, Kazem HA. Influence of the base fluid on Sol Energy 2018;165:167e77.
the thermo-physical properties of PV/T nanofluids with surfactant. Case [41] Petela R. Exergy of heat radiation. ASME Journal of Heat Transfer 1964;86:
Studies in Thermal Engineering 2019;13:100340. 187e92.
[28] Abdallah SR, Saidani-Scott H, Abdellatif OE. Performance analysis for hybrid [42] Bellos E, Tzivanidis C. A detailed exergetic analysis of parabolic trough col-
PV/T system using low concentration MWCNT (water-based) nanofluid. Sol lectors. Energy Convers Manag 2017;149:275e92.
Energy 2019;181:108e15. [43] Moffat R. Describing the uncertainties in experimental results. Exp Therm
[29] Han X, Chen X, Wang Q, Alelyani SM, Qu J. Investigation of CoSO4-based Ag Fluid Sci 1988;1:3e17.
nanofluids as spectral beam splitters for hybrid PV/T applications. Sol Energy [44] Sardarabadi M, Hosseinzadeh M, Kazemian A, Passandideh-Fard M. Experi-
2019;177:387e94. mental investigation of the effects of using metal-oxides/water nanofluids on
[30] Khanjari Y, Pourfayaz F, Kasaeian A. Numerical investigation on using of a photovoltaic thermal system (PVT) from energy and exergy viewpoints.
nanofluid in a water-cooled photovoltaic thermal system. Energy Convers Energy 2017;138:682e95.
Manag 2016;122:263e78. [45] Fudholi A, Sopian K, Yazdi MH, Ruslan MH, Ibrahim A, Kazem HA. Performance
[31] Al-Waeli AH, Chaichan MT, Sopian K, Kazem HA, Mahood HB, Khadom AA. analysis of photovoltaic thermal (PVT) water collectors. Energy Convers
Modeling and experimental validation of a PVT system using nanofluid Manag 2014;78:641e51.
coolant and nano-PCM. Sol Energy 2019;177:178e91.