Applied Thermal Engineering 185 (2021) 116381

Contents lists available at ScienceDirect

Applied Thermal Engineering

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

A numerical investigation on optimization of PV/T systems with the field

synergy theory
Chao Shen *, Yingbo Zhang **, Chunxiao Zhang, Jihong Pu, Shen Wei, Yu Dong
School of Architecture, Harbin Institute of Technology, Key Laboratory of Cold Region Urban and Rural Human Settlement Environment Science and Technology, Ministry
of Industry and Information Technology, Harbin 150090, China

A R T I C L E I N F O A B S T R A C T

Keywords: High cell temperature of PV modules would reduce electrical efficiency and hinder the development of photo­
PV modules voltaic application. Many potential cooling channels have been proposed for PV modules cooling but perform
Heat regulation inefficiently, due to the insufficient heat transfer between the cooling water and PV modules. In current research,
Field synergy theory
a PV/T system with a special cooling channel based on field synergy theory, was designed to optimize the
Cell temperature
performance of PV/T systems. The performance of the new PV/T system was studied numerically, and mathe­
matical models were validated with the experimental data. Results indicated that electrical efficiency of PV
modules increased by 0.5% (from 11.4% to 11.9%) when the solar radiation decreased from 1200 to 300 W/m2.
The electrical efficiency increased 1.11% (from 10.8% to 11.9%) as the inlet mass flowrate of cooling water
increased from 0.0018 to 0.018 kg/s at the inlet temperature of cooling water of 20 ◦ C. Additionally, the special
channel with shark dorsal fin type sawtooth or regular type sawtooth, had the optimal parameter values, with the
channel depth of 0.05 m, the sawtooth peak of 0.035 m and the sawtooth width of 0.01 m. The average cell
temperature with the optimized channel was 6.05 ◦ C lower than the conventional smooth channel, and the new
cooling channel performed better with a high solar radiation.

Also, it was pointed out that in the field of PV/T cooling, it was necessary
1. Introduction to study the heat transfer enhancement. In addition to air/water cooling,
a structure optimization of the cooling channels on the PV modules,
The development and utilization of renewable and environmental- targeting a larger heat transfer area and a higher Reynolds number, is
friendly energy, have been concerned in recent years due to the aggra­ another effective way to enhance the cooling performance and improve
vation of fossil energy depletion and global warming diffusion [1]. Solar the electrical efficiency [9].
energy belongs to an excellent renewable energy with the advantages of Many researchers tried to optimize structural configurations of
pollution-free, inexhaustibility, and huge amount. In order to utilize cooling channels [10], including adding fins with different shapes,
solar energy, PV modules are commonly used as a kind of equipment to multi-channel forms, channel shapes and arrangement, to enhance the
convert solar radiation into electrical energy. However, a challenge, the overall efficiency of PV/T systems [11]. Some literatures with optimized
limited performance of solar cells caused by a high cell temperature, was structural configurations were summarized as following: (1) Adding fins
raised in the application of PV modules. When the cell temperature in­ with different shapes [12]. Hussain et al. [13] carried out an experi­
creases by 1 ◦ C, the electrical efficiency of PV modules decreases by mental study on a PV/T thermal cooling channel with hexagonal shape
0.3–0.5% [2,3]. Therefore, some feasible solutions have been proposed fins. They reported that the electrical efficiency increased by about 0.2%
to regulate the heat accumulated on the surface of PV modules [4]. and the thermal efficiency was improved by 20–70%. Özakin [14] per­
Existing cooling technologies, including air cooling [5], water cooling formed a comparative study on PV efficiency of three kinds of cooling
[6], cooling with PCM [7] and special structure designs [8], have been channels including the empty, the frequent and the sparse fins status.
discussed to reduce cell temperature. In a review study [8], it was re­ Results indicated that the exergy efficiency with sparse fins was
ported that the thermal efficiency of PV modules with water cooling approximately increased by 70%, and increased by 30% with frequent
could reach up to 50–70%, while that of air cooling was only 30–50%. fins, compared with the empty status. (2) Multichannel forms. Sopian

* Corresponding author.
** Corresponding author.
E-mail addresses: [email protected] (C. Shen), [email protected] (Y. Zhang).

https://doi.org/10.1016/j.applthermaleng.2020.116381
Received 7 July 2020; Received in revised form 22 October 2020; Accepted 23 November 2020
Available online 27 November 2020
1359-4311/© 2020 Elsevier Ltd. All rights reserved.
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Nomenclature ρ Density (kg/m3)

η Efficiency (dimensionless)
G Solar radiation (W/m2) τ Transmissivity (dimensionless)
E Rate of energy transfer (W) α Absorptivity (dimensionless)
T Temperature (◦ C) ε Emissivity (dimensionless)
P Pressure (Pa) ∇ Vector differential operator (dimensionless)
A Area (m2) γ Temperature coefficient (◦ C− 1)
Q Heat source (W/m3) σ Stefan-Boltzmann value (W⋅m− 2⋅K− 4)
p Perimeter (m) μ dynamic viscosity of the fluid (m2 s− 1)
cp Specific heat at constant pressure (J/kg⋅◦ C)
k Thermal conductivity of material (W/m⋅◦ C) Subscripts
h Heat transfer coefficient (W/m2⋅K) a Ambient
Re Reynolds number (dimensionless) g Glass
u Inlet water flow velocity (m/s) c Cell
Dh Hydraulic diameter (m) f Fluid
q Heat flux vector (dimensionless) w Wall
n Surface normal (dimensionless) in Inlet
u Velocity field (m/s) out Outlet
F Volume force (N) el Electrical
I Moment of inertia (kg⋅m2) th Thermal
ref Reference
Greek symbols
γf Kinematic viscosity (m2/s)

et al. [15] studied comparatively steady-state models of single-channel channel on the performance of PV/T systems were discussed. The result
and double-channel PV/T based on air cooling respectively, and provides a view for the cooling technology of PV modules based on the
discovered that the combined efficiency of double-channel type was field synergy theory, which is helpful to further boost electricity yield of
10–15% higher than that of the single-channel one. Ooshaksaraei et al. PV modules.
[16] found that the PV/T system with double-path parallel flow of the
cooling water, had the highest total energy efficiency (51–67%), fol­ 2. Field synergy theory and special design of cooling channels
lowed by double-path counter flow (47–62%), single-path returning
flow (42–56%), and single-path (28–49%). (3) Channel arrangement. 2.1. Field synergy theory
Fudholi et al. [17] investigated the performance of PV/T systems with
three different arrangements of cooling water flow channels including As known, the performance of convective heat transfer is superior to
web flow absorber, direct flow absorber and spiral flow absorber. They that of conductive heat transfer because the heat is carried by macro
found that the spiral flow absorber exhibited the highest performance, moving fluid for the former kind [25]. Based on the principle of uni­
with a total efficiency of 68.4%, an electrical efficiency of 13.8% and a formity of temperature difference field in heat exchangers, Guo [26]
thermal efficiency of 54.6% at a solar radiation of 800 W/m2 and a reexamined the physical mechanism of convective heat transfer, and put
cooling water flowrate of 0.041 kg/s. Nahar et al. [18] simulated the forward the field synergy principle to enhance heat transfer. The prin­
performance of PV modules with a cooling flow pancake-shaped chan­ ciple shows that the heat transfer coefficient could be improved by
nel. Results demonstrated that the cell temperature was cooled down to improving the coupling between velocity field and temperature gradient
an average cell temperature of 42 ◦ C, and the electrical efficiency field. The dimensionless quantity, the field synergy number (Fc ), was
increased by 2% with the inlet velocity of cooling water from 0.0009 to defined to characterize the synergy between the velocity field and the
0.05 m/s. (4) Channel with different shapes. Baloch et al. [19] investi­ heat flow field in the whole flow and heat transfer region. The expres­
gated the performance of PV modules with a cooling technique, called a sion of Fc is as follows.
converging channel heat exchanger, and found that the cell temperature ∫ 1( )
Nu
reduced significantly to 45.1 ◦ C in June and to 36.4 ◦ C in December by
⇀
Fc = = U ⋅∇T dy (1)
using the converging cooling technique. The maximum percentage in RePr 0

improvement of power output was 35.5%, whereas the maximum per­ ⃒⇀⃒
⇀ ⃒ ⃒
centage of increase in the conversion efficiency was 36.1%, compared to U ⋅∇T = ⃒U ⃒ × |∇T|cosβ (2)
the performance of an uncooled PV module. Fudholi et al. [20] per­
formed an energy analysis on a PV/T system with ∇-groove shape and where βis the angle between the velocity vector and the temperature
found that the average electrical efficiency was 12.66%. gradient vector of the fluid.
As analyzed above, many potential channel structures and theoret­ Equation (1) indicates that the heat transfer enhancement could be
ical analysis [21] have been proposed to cool down PV modules, but carried out from three aspects: (1) The Reynolds number can be
there is still a room to improve [22]. As well known, the field synergy increased by increasing the flow rate and reducing the diameter of tubes,
between velocity and temperature gradient has a great influence on the so as to improve the heat transfer performance; (2) The Prandtl number
heat transfer performance [23,24]. Therefore, a special design of cooling of the fluid could be raised through increasing the viscosity or specific
channel based on the field synergy theory was proposed, which had a heat capacity of the fluid; (3) The integral value of dimensionless term
smaller synergy angle between the velocity gradient of cooling water (
∫1 ⇀
)
and the temperature gradient on the surface of PV modules, to promote 0 U
⋅∇T dy could be improved based on the field synergy theory. The
the overall efficiency of the PV/T system. Meanwhile, the effects of solar
radiation, inlet flow rate of cooling water and structural parameters of

2
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 1. Special designs of cooling channels.

Fig. 2. Parameter structure diagram.

( )
∫1 ⇀ analysis, it can be seen that reducing the angle between the velocity
item of 0 U⋅∇T dy means the sum of dimensionless heat source in­ vector and temperature gradient is a good way to enhance heat transfer
tensity in the entire thermal boundary layer with thickness of x. The heat for PV/T systems.
⇀ Different methods can be used to make the velocity field and heat
transfer performance can be improved by increasing the value of U⋅∇T. flow field achieve a better field synergy effect to enhance heat transfer
As shown in Equation (2), its value is related to not only the absolute ⇀ ∫1 ⇀
value of velocity and temperature gradient, but also their included [27]. Since the item of U⋅∇T in the integral formula 0 (U⋅∇T)dy is the dot
angle. When <90β◦ , the smaller the angle between the directions of product of two vectors, in order to get a larger value, the component of
velocity and temperature gradients, the better the field synergy. When one vector in the direction of other vector should be as large as possible.
⃒⇀⃒
>90β◦ , the value of β close to 180◦ is favorable to a better collaboration ⃒⃒
⃒U⃒cosβ represents the component of velocity vector in the direction of
between the velocity field and temperature field and a better convective
heat transfer performance between the fluid and wall. From the above the temperature gradient. As for wall convective heat transfer, the

Fig. 3. Streamline of flow field (holistic, local).

3
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 4. Profile of flow streamline.

Fig. 5. Layer configuration of the PV/T system.

direction of temperature gradient near the wall surface is approximately

⃒⇀⃒ Table 1
⃒⃒
perpendicular to the wall, thus ⃒U⃒cosβ represents the component of ve­ The materials and their physical properties.
locity in the direction perpendicular to the wall. For the convective heat Materials Depth Conductivity Density Specific heat
⃒⇀⃒ [mm] [kg/m3] [kg/m3] [J/(kg K)]
⃒⃒
transfer on a plate, the main flow direction is x direction, thus ⃒U⃒cosβ
Glass 3 1.8 2500 500
represents the component of fluid velocity in Y direction. EVA 0.3 0.311 950 2090
Silicon 0.5 148 2329 700
EVA 0.3 0.311 950 2090
2.2. Special design of cooling channels Tedlar 0.1 0.15 1200 1250
Thermal paste 0.3 1.9 2600 700
In the traditional PV/T system, the synergy angle is close to 90◦ Aluminum 0.1 237 2700 903
mostly, with a poor cooling performance. From the view of field synergy
theory, two geometries of the cooling channel surface that far from the
been done with various parameters, such as inlet mass flow rate ranged
PV module surface are proposed, targeting disturbing the flow field of
from 0.0018 to 0.018 kg/s and solar radiation ranged from 300 to 1200
cooling water in the cooling channel, as shown in Fig. 1. When the
W/m2.
cooling fluid passes through the special cooling channel, a special ve­
The layer configuration of the PV/T system is shown in Fig. 5,
locity field with a certain angle to the temperature gradient would
including the glass cover plate, two layers of transparent ethylene vinyl
appear under the induction of the channel surface. When the angle is
acetate (EVA) material, PV cells (polysilicon), one layer of tedlar. The
close to 0◦ or 180◦ , the cooling channel reached the best cooling per­
cooling channel made in aluminum metal with various structures was
formance. The structure “A” was designed similarly with shark dorsal
attached to the PV module bottom surface using thermal paste. Under
fin, as shown in Fig. 1. The structure “B” is regular sawtooth channel
the PV module, there was a thermal insulation material to prevent heat
which looks like a washboard. Fig. 2 depicts the schematics of these two
loss to the ambient air. In order to simplify the simulation process, the
special cooling channels of PV modules, where “p” is the depth of the
peak of the sawtooth and “w” is the pitch of the sawtooth. Also, the
curvature for shark dorsal type is shown in Fig. 2. Table 2
Fig. 3 and Fig. 4 illustrate the streamline of the special channel, both Parameters of the PV/T system [28]
of which indicated there was a better synergism between velocity field Parameters Values
and temperature gradient field near the hot surface of the PV module. Channel length 1m
Heat on the surface of PV modules was collected to the cooling water, Channel width 0.6 m
with vertical temperature gradient field to the surface. While, the Channel depth 0.01–0.1 m
Glass transmissivity (τg ) 0.96
cooling fluid is introduced by the special geometry, so as to go forward
Emissivity of glass (εg ) 0.04
up and down continuously, forming a better angle between the velocity
Solar cell absorptivity (αc ) 0.9
and temperature gradient. Based on the field synergy, the heat transfer
8
performance should be enhanced as a result. Stefan-Boltzmann value (σ) 5.670368*10−
Reference temperature (Tref ) 25 ◦ C

3. Methodology Ambient temperature (Ta ) 20 ◦ C

Inlet temperature (Tin ) 20 ◦ C

3.1. Numerical investigation Solar radiation (G) 300–1200 W/m2

Mass flowrate of cooling fluid (m) 0.0018–0.018 kg/s
PV cell efficiency at standard test conditions (ηref ) 12%
Numerical simulations were conducted to investigate the perfor­ Heat transfer coefficient from glass cover to ambient (hg− a ) 5.84 W/m2 K
mance of PV/T systems with the new cooling channel. The analysis has

4
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Structure “A” Structure “B”

Fig. 6. Finite element meshing of the PV/T system.

following assumptions were proposed considering the actual situation

Table 3
and the characteristics of components.
Grid test at a radiation of 800 W/m2 and a mass flow rate of 0.009 kg/s.

(1) The influence of the dust on the glass cover on solar radiation was Type of meshing Elements Cell temperature (◦ C) Time of solution (s)
ignored. Extra Coarse 17,620 30.08 9
(2) The thermal physical properties of the cooling channel were Super Coarse 50,022 30.619 153
Coarser 91,292 30.782 175
assumed to be constant as temperature varied.
Coarse 248,229 30.858 200
(3) EVA was considered transparent and had a 100% transmission Normal 336,766 30.847 414
efficiency. Fine 1,186,885 30.912 1457
(4) The fluid in the channel was completely uniform and Finer 4,123,650 30.905 5515
incompressible.
(5) Rate of heat dissipation through the upper and the back surface
was equal.
(6) The sky temperature and the ambient temperature were assumed
to be equal.

Table 1 and Table 2 depicted the detailed information of the PV/T

system, and the main size of these two special PV/T systems are similar
to that reported in Ref. [28], except the back surface of special cooling
channels.
Numerical simulations were conducted using COMSOL Multiphysics
software based on the finite element technique [29]. The finite element
technique was applied to solve Equations (3)–(6) in the PV/T system.
The current simulation ran in a steady state.

3.1.1. Governing equations

In all domains in the PV/T system, the heat transfer mechanism
included heat conduction and convection. The energy balance equation
is given in Equation (3). Equation (4) is Fourier heat conduction
equation.
Fig. 7. Hourly variation of the cell temperature, glass temperature and outlet
ρcP u⋅∇T + ∇⋅q = Q (3) temperatures versus local time.

− k∇T = q (4) c. At the channel walls: no-slip condition, u = 0

Given the cooling fluid was Newtonian and incompressible fluid, the d. At the channel inlet: V = uin , T = Tin
continuity and momentum equations for the cooling fluid were e. At the channel outlet: P = 0
expressed in Equations (5) and (6). f. At the side boundaries of the PV module: insulation, − n⋅q = 0
g. At the solid-fluid interface boundary: (∂∂TnS )fluid = ks ∂TS
kw ( ∂n )solid
ρ∇⋅u = 0 (5)
[ ] 3.1.3. Meshing and grid test
ρ(u⋅∇)u = ∇⋅ − pI + μ(∇u + ∇uT ) + F (6)
Fig. 6 depicts the mesh generation of the PV/T system, where the
subdomain boundary was free tetrahedral, and boundary elements were
3.1.2. Boundary condition free triangular. Grid independent test of a numerical model determined
In order to solve partial differential equations (PDEs), the boundary
the accuracy of simulation results. Therefore, the performance of seven
conditions were set in the type of mathematical equations that listed as types of non-uniform grid systems was compared in this mesh conver­
follows. gence test, and the mesh convergence test was conducted with an inlet
mass flow rate of 0.009 kg/s, an ambient temperature of 20 ◦ C, and a
∂T
a. At the top of surface of the PV module: Inward heat flux: − kg ∂zg = G; solar radiation of 800 W/m2. The cell temperature was taken as the
Heat loss condition: − n⋅q = hg− a (Ta − Tg ); − n⋅q = εg σ(Tg4 − Ta4 ) control parameter. The results of grid test were shown in Table 3, and
b. At the bottom of heat exchanger: Convective heat loss con­ there were negligible changes of the cell temperature between the mesh
dition: − n⋅q = hg− a (Ta − Tcha ) size of 1,186,885 and 4,123,650. However, the calculation with mesh

5
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

size of 4,123,650 used more computing time and space. Thus, the nu­
merical model with 1,186,885 domain elements was employed for the
numerical investigation.

3.2. Model validation

In current study, to validate the mathematical model of the PV/T

system, experimental results, as reported by Tiwari et al. [30], were
used. The simulation was set for the PV module with the area of one
module of 0.61 m2, a packing factor of 83%, a standard electrical effi­
ciency of 12%. The size of the heat exchanger duct was 1.2 * 0.45 * 0.1
m. In the validation process, the following factors were set: solar radi­
ation ranged from 167 to 704 W/m2, ambient air temperature ranged
from 31 to 36 ◦ C, inlet velocity ranged from 2.83 to 4.83 m/s. As
demonstrated in Fig. 7, there was a good agreement between numerical
results and experimental values, with a correlation coefficient of 0.99
and a standard percentage deviation error of 2.9–6.9%. Fig. 8. The effect of solar radiation on the electrical efficiency of PV modules
(simulated results).

3.3. Performance evaluation

3.3.1. Electrical model

The total solar energy (W/m2), Ein , absorbed by the PV/T system was
given by Equation (7). The electrical energy (W/m2),Eel , produced by
the PV cell was expressed by Equation (8). The electrical efficiency of the
PV module (ηel ) was a function of the cell temperature and given by
Equation (9) [17].
Ein = τg αc G (7)

Eel = ηel Ein (8)

ηel = ηref (1 − γ(Tc − Tref )) (9)

where ηref is the reference efficiency of the PV module (ηref = 0.12); γ is a

temperature coefficient (γ = 0.0045 ◦ C− 1); Tc is the cell temperature;
and Tref is the reference temperature (25 ◦ C, optimum operating tem­
perature of PV modules at the rated power). Fig. 9. The effect of solar radiation on the thermal efficiency of PV/T systems
(simulated results).
3.3.2. Thermal model
The output thermal energy of the PV/T system was calculated by where, Dh was the hydraulic diameter defined as follows:
Equation (10). The mass flow rate of the cooling fluid was calculated by
4Af
Equation (11). The thermal efficiency of the PV/T system was calculated Dh = (14)
by Equation (12). Pf

Eth = mcf (Tout − Tin )/A (10) For the specific cases in the simulation, where the depth of the
channel is 0.1 m, the width of the cross section is 0.6 m, and the inlet
where Eth is the thermal energy collected by the PV/T system; cf is the velocity selected as maximum velocity of 0.01 m/s. Calculated by
specific heat of the cooling fluid; Tout andTin are the outlet temperature Equation (13), the maximum Reynolds number is 1697 less than 2300.
and inlet temperature of the cooling channel; A is the area of the PV Therefore, the fluid is considered to be laminar.
module.
4. Results and discussion
m = ρf Af uf (11)
In this section, the performance of the PV/T system, including the
where m is the inlet mass flow rate of cooling fluid; ρf , Af , uf are density, electrical efficiency and thermal efficiency, were investigated with
cross-sectional flow area, inlet velocity of cooling fluid; following factors: (1) solar radiation ranged from 300 W/m2 to 1200 W/
Eth m2 and the mass flow rate of cooling water ranged from 0.0018 to 0.018
ηth = (12) kg/s in Section 4.1; (2) the sawtooth peak height, the sawtooth width,
Ein
and the channel depth in Section 4.2; (3) a comparison between the new
where ηth is thermal efficiency of the PV/T system; Ein is the energy cooling channel and the conventional cooling channel in Section 4.3.
received energy by solar cell.
The Reynolds number of the cooling fluid in the cooling channel was 4.1. The performance of PV/T systems with different operation conditions
calculated by Equation (13), which was used to determine whether the
fluid was laminar flow or turbulent flow. 4.1.1. The effect of solar radiation on the performance of PV/T systems
uin Dh Fig. 8 shows the effect of solar radiation on the electrical efficiency of
Re = (13) PV modules with an inlet mass flow rate of 0.009 kg/s, a width of
γf
sawtooth of 0.05 m, a peak of sawtooth of 0.02 m, a channel depth of
0.03 m. As the solar radiation increased from 300 to 1200 W/m2, the

6
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 10. The surface temperature for various radiations at a flowrate of 0.009 kg/s.

would be accumulated on the surface of PV modules, leading to a higher

Table 4
cell temperature and a lower electrical efficiency.
Comparison of cell temperatures obtained from various references.
Fig. 9 depicts the effect of solar radiation on thermal efficiency of
References Cooling Solar radiation Ambient cell PV/T systems. The outlet temperature went up with the increase of solar
channel (W/m2) temperature temperature
radiation, because more heat produced by PV modules was transferred
form (◦ C) increment
At At
per 100 W/ to the cooling water. Even though, the thermal efficiency in percentage
start peak
m2 increase of PV/T systems still went down, because more heat of the PV module
period period
of solar was released into ambient air by convective heat transfer and radiation
radiation (◦ C)
heat transfer. Results indicated that thermal efficiency of structure “A”
Chandrasekar Cotton 600 1300 37 1.4 and structure “B” decreased by 1.09% and 1.34% respectively when
et al. [31] wick solar radiation rose from 300 to 1200 W/m2.
structures
The effect of solar radiation on the temperature distribution of the
Bahaidarah Cooling 240 979 21 1.9
et al. (2013) panel PV/T system was demonstrated with 3D simulation plots, and the results
[32] are shown in Fig. 10. As solar radiation increased, the surface temper­
Teo et al. A parallel 550 1050 – 1.4 ature of PV modules increased obviously. Meanwhile, there were ripples
(2012) [33] array of
for the temperature distribution on the surface of PV modules, due to the
finned
ducts
induction of the special geometry of the bottom surface of the cooling
Rahman et al. Baffles in 1000 5000 35 2.71 channel. The cooling water went forward up and down continuously and
(2017) [34] channel the upper surface of the cooling channel was impacted. It can be found
Nasrin et al. A box- 1000 3000 32 1.85 that structure “A” caused a greater impact of cooling fluid on the dis­
(2017) [35] shaped
tribution of upper surface. This was due to the fact that the shark dorsal
channel
Current paper shark 300 1200 20 1.29 fin made a greater disturbance on the velocity field of the fluid so that a
dorsal fin better synergy with the temperature field was made.
type Table 4 shows the comparison on the averaged increment of cell
channel
temperature with per 100 W/m2 increase of solar radiation. Researchers
have reported that cell temperature increased averagely by 1.4 ◦ C
average cell temperature increased from 23.90 to 35.04 ◦ C, and the [31,33], 2.71 ◦ C [34], 1.85 ◦ C [35] and 1.9 ◦ C [32] with per 100 W/m2
electrical efficiency of structure “A” decreased from 12.06% to 11.46%. increment of solar radiation. While the average increase of cell tem­
For the PV module with structure “B”, the average cell temperature perature with structure “A” in this study was 1.29 ◦ C with per 100 W/m2
increased from 24.13 to 35.93 ◦ C and the electrical efficiency decreased increment of solar radiation, which was lower than conventional PV/T
from 12.05% to 11.40%. With the increase of solar radiation, more heat systems. These increments in temperature meant the ability for PV
modules with different cooling methods to resist temperature growth for

7
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

temperature difference between the solar cell and the cooling medium
(water), leading to more heat being transferred into the cooling water, as
a result, a lower cell temperature appeared. However, being restricted
by the convective heat transfer coefficient of the cooling water, the
contribution of increasing inlet mass flow rate to reducing the cell
temperature was limited.
The effect of the inlet mass flow rate on the thermal efficiency of PV/
T systems was investigated, as illustrated in Fig. 12. As the inlet mass
flow rate increased from 0.0018 kg/s to 0.018 kg/s, the outlet temper­
ature of the cooling water with structure “A” decreased from 59.67 ◦ C to
25.73 ◦ C and the thermal efficiency was improved from 37.31% to
53.90%, while the outlet temperature of the cooling water with struc­
ture “B” decreased from 59.17 ◦ C to 25.66 ◦ C and thermal efficiency
increased from 36.84% to 53.26%. Being similar to the electrical effi­
ciency, the change rate of the outlet temperature of the cooling water
and the thermal efficiency was decreasing gradually. The increase of the
Fig. 11. The effect of the mass flowrate on the electrical efficiency of PV
mass flow rate of cooling water reduced the cell temperature, thus the
modules (simulated results).
heat loss from PV modules to the environment was reduced, and more
heat was taken away by the cooling water, leading to an increased
thermal efficiency.
As shown in Fig. 13, as the mass flow rate of the cooling water
increased, the averaged surface temperature decreased by 33.70 ◦ C for
structure “A” and 34.70 ◦ C for structure “B”. A higher surface temper­
ature of structure “B” caused more heat being lost to the environment
and a lower thermal efficiency. Meanwhile, due to the induction of the
sawtooth and shark dorsal fin, the synergy between the velocity field
and temperature gradient field at the peak, was greater, leading to a
better heat transfer and more obvious ribbon patterns of the surface
temperature. With the increase of the mass flow rate of the cooling
water, the Reynolds number increased correspondingly (the inlet ve­
locity is in positive proportion to the mass flow rate when other condi­
tions are given), resulting in a better heat transfer effect and more
remarkable ribbon patterns of the surface temperature. Compared to
structure “B”, the PV module with structure “A” of shark dorsal fin had a
more distinct ribbon distribution of the surface temperature, particu­
Fig. 12. The effect of the mass flowrate on the thermal efficiency of PV/T larly at an inlet flow rate of 0.018 kg/s.
systems (simulated results).
4.2. The effect of the cooling channel on the performance of PV/T systems
every 100 W/m2 increase in solar radiation. Considering the electrical
efficiency of PV modules, the special cooling channel was advantageous 4.2.1. The effect of the sawtooth peak height
to resist more temperature growth because a low cell temperature kept Fig. 14 and Fig. 15 indicate the effect of the sawtooth peak height on
the electrical efficiency high. Special channels in this study created a the performance of PV/T systems, including the cell temperature, the
flow angle by the diversion structure, leading to the angle between ve­ outlet temperature of the cooling water, the electrical efficiency and
locity vector and temperature gradient vector closer to 0◦ or 180◦ , so as thermal efficiency at a solar radiation of 800 W/m2 and an inlet mass
to enhance convective heat-transfer coefficient and take more heat from flow rate of 0.009 kg/s. As sawtooth peak height increased from 0.002 to
PV modules to the cooling water. 0.02 m, the average cell temperature decreased by 3.59 ◦ C for structure
“A” and 2.82 ◦ C for structure “B”, while the electrical efficiency was
4.1.2. The effect of the inlet mass flow rate on the performance of PV/T improved by 0.2% for structure “A” and 2.33% for structure “B”. The
systems thermal efficiency was enhanced by 0.15% for structure “A” and 1.74%
Fig. 11 depicts the effect of the inlet mass flow rate of cooling water for structure “B”. An increased Reynolds number caused by a suddenly
on the performance of PV/T systems with a solar radiation of 800 W/m2, smaller channel at the peaks of the sawtooth and shark dorsal fin,
a width of sawtooth of 0.05 m, a peak of sawtooth of 0.02 m, a channel boosted the convective heat transfer coefficient and enhanced the
depth of 0.03 m. As the inlet mass flow rate increased from 0.0018 kg/s cooling efficiency, resulting in an increased electrical efficiency and
to 0.018 kg/s, the electrical efficiency of PV modules with structure “A” thermal efficiency.
was improved from 10.81% to 11.92% and the average cell temperature The effect of the sawtooth peak height on the temperature distribu­
decreased from 47.22 ◦ C to 26.61 ◦ C. While the electrical efficiency with tion of PV modules and the streamline distribution in the cooling
structure “B” increased from 10.81% to 11.85% with the average cell channel were investigated, as demonstrated in Fig. 16 (structure “A”)
temperature decreasing from 47.22 to 26.61 ◦ C. As seen from Fig. 11, and Fig. 17 (structure “B”). The surface temperature of PV modules with
both change rates of electrical efficiency and average cell temperature 0.02 m sawtooth peak was significantly lower than that of 0.01 m. The
were decreasing gradually. Averagely, the increase of the inlet mass flow main reason was that a high sawtooth peak would produce a strong
rate at per 0.001 kg/s would lead to a decrease of average cell tem­ impact and disturbance near the surface of PV modules. Meanwhile, for
perature of 1.167 ◦ C, and similar results have been reported by Nahar a higher sawtooth peak, the cooling water would continuously flow up
et al. [28] (An increase in the inlet mass flow rate of 0.001 kg/s resulted and down more fiercely, which would destroy the heat transfer
in the average cell temperature decreased by 1.35 ◦ C). A higher inlet boundary layer of the cooling water near the surface and enhance the
mass flow rate created a lower water temperature and a bigger heat transfer performance in the cooling channel [36].
As illustrated in Fig. 18, with the same conditions, the field synergy

8
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 13. The surface temperature with different inlet mass flow rates at a solar radiation of 800 W/m2.

Fig. 14. The effect of the sawtooth peak height on the electrical efficiency of Fig. 15. The effect of the sawtooth peak height on the thermal efficiency of PV/
PV modules (simulated results). T systems (simulated results).

angle of the temperature gradient filed and velocity filed with the 4.2.2. The effect of the width of tooth
structure “A” was closer to 0◦ or 180◦ , which was better than that of Fig. 19 depicts the effect of the sawtooth width on the electrical ef­
structure B, resulting in a better heat transfer process, thus a better ficiency of PV/T systems at a solar radiation of 800 W/m2 and an inlet
performance of PV/T systems with structure “A” was found in this study. mass flow rate of 0.009 kg/s. As the sawtooth width increased from 0.01
to 0.07 m, the average cell temperature increased by 2.28 ◦ C for

9
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 16. The temperature distribution and streamline distribution of structure “A”.

Fig. 17. The temperature distribution and streamline distribution of structure “B”.

Fig. 18. A comparison of the streamline angle between structure “A” and structure “B”.

structure “A” and 1.90 ◦ C for structure “B”, while the electrical effi­ because the disturbance of the cooling water induced by the sawtooth
ciency decreased by 0.13% for structure “A” and 0.10% for structure and shark dorsal fin was weaker at a bigger sawtooth width. As a
“B”. In Fig. 20, both outlet temperatures of the cooling water and the consequence, less heat was transferred to the cooling water, leading to a
thermal efficiency decreased with the increase of the sawtooth width, higher cell temperature and less thermal energy being collected. The

10
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

thermal efficiency and electrical efficiency with a smaller sawtooth

width, was higher than that of a greater sawtooth width.
Fig. 21 and Fig. 22 illustrate the temperature distribution and
streamline distribution with a sawtooth width of 0.02 m and 0.07 m,
respectively. The surface temperature of PV modules with a sawtooth
width of 0.02 m was lower than that of 0.07 m, which was attributed to
the fact that the cooling water in the cooling channel with a smaller
sawtooth width had a higher disturbance and a better cooling perfor­
mance. As a result, the PV/T system with a small sawtooth had a lower
averaged cell temperature and a higher electrical efficiency.

4.2.3. The effect of the channel depth

Apart from the sawtooth peak height and the sawtooth width, the
channel depth was also an important geometry parameter to optimize
the cooling efficiency of PV modules. The channel depths ranging from
Fig. 19. The effect of the sawtooth width on the electrical efficiency of PV 0.01 to 0.1 m were selected to investigate its effect on the performance
modules (simulated results). of PV/T systems with a fixed inlet velocity of 0.005 m/s (the inlet mass
flowrate was changed with the channel depth).
As depicted in Fig. 23, because the range of 0.01 to 0.02 m was small
for the channel depth, there is a big room for improvement of the cooling
capacity of the cooling channel. Thus, as the channel depth increased
from 0.01 to 0.02 m, the cell temperature decreased rapidly, resulting in
a remarkable increase of electrical efficiency. Whilst, when the channel
depth increased from 0.02 to 0.06 m, the cell temperature went down
slowly and the electrical efficiency rose at a slow growth speed, in which
process the thermal efficiency increased gradually. While the channel
depth was greater than 0.06 m, the cell temperature decreased slowly
and tended to be constant, the increase of the channel depth had no
effect of the cooling capacity.
As illustrated in Fig. 24, with the channel depth increased, the outlet
temperature of the cooling water decreased gradually and presented a
slowing-down trend, which was similar to the change profile of the cell
temperature. However, the thermal efficiency increased with the in­
crease of the channel depth and reached the maximum value at a
channel depth of 0.08 m, and then decreased. According to Equations
Fig. 20. The effect of the sawtooth width on the thermal efficiency of PV/T (10) and (11), the thermal efficiency was not only related to the tem­
systems (simulated results). perature difference between the inlet and outlet temperatures of the
cooling water, but also determined by the mass flow rate. With the inlet
temperature of the cooling water at a fixed value of 20 ◦ C, the thermal

Fig. 21. Temperature distribution and streamline distribution of structure “A”.

11
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

Fig. 22. The temperature distribution and streamline distribution of structure “B”.

Fig. 25. A comparison of PV modules with two types of cooling channels

Fig. 23. The effect of channel depth on the electrical efficiency of PV modules against solar radiation (simulated results).
(simulated results).

efficiency was related to the mass flow rate of the cooling water and the
outlet temperature. In the simulation of this study, the inlet velocity of
the cooling water was set at 0.005 m/s. With the changing channel
depth, the outlet temperature and the inlet mass flow rate changed,
which affected the electrical efficiency and the thermal efficiency, and
this result complied with the data reported by Nahar et al. [37].
In general, the PV/T system with a channel depth of 0.08 m, had the
best cooling efficiency and the highest overall efficiency. Note that, a
bigger channel depth would not only increase the initial cost, but also
cause a heavy weight of the PV/T system. Therefore, the channel depth
of 0.03 m to 0.05 m was recommended.

4.3. Comparison between the special channel and the conventional

channel

4.3.1. Performance comparison under fixed operating conditions

Fig. 25 shows the comparison of the efficiency of PV/T systems
Fig. 24. The effect of channel depth on the thermal efficiency of PV/T systems
among the uncooled PV module, the PV module with the new channel
(simulated results). and the PV module with the conventional channel under different solar
radiations at an inlet mass flow rate of 0.01 kg/s. When solar radiation
ranged from 300 to 1200 W/m2, the electrical and thermal efficiency the
PV module with the new channel was always higher than that of the
conventional channel. As solar radiation increased, the thermal

12
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

m2), ambient temperature (10–20 ◦ C). The inlet temperature of cooling

water was set at 20 ◦ C and the inlet mass flowrate was set at 0.0036 kg/s.
The results were shown in Fig. 28. The two systems were in the same
state at the starting point of 8o’clock. Then systems started to run. It is
clear that the total efficiency of the new PV/T system was consistently
higher than that of the conventional PV/T system over a period of 8 h. In
addition, the average cell temperature of the new PV/T system was
lower than that of the conventional system. In conclusion, the new PV/T
system still showed excellent performance in real conditions.

5. Conclusion

In this paper, PV/T systems with special cooling channels were

proposed based on the theory of field synergy, targeting the enhance­
ment of heat transfer between the cooling water and the surface of PV
modules. Mathematical models were established and validated to
Fig. 26. A comparison of PV modules with two types of cooling channels investigate the performance of PV/T systems using the finite software
against the mass flow rate (simulated results). COMSOL Multiphysics. The effects of solar radiation, the inlet mass flow
rate of the cooling water and structural parameters of special channels
efficiency difference between the new cooling channel and the con­ on the cooling efficiency of PV modules were discussed. Some conclu­
ventional cooling channel rose. Meanwhile, the new cooling channel sions were drawn as follows:
performed better at a higher solar radiation. In general, the new cooling
channel had a better cooing effect than the conventional channel (1) The structure “A” in shark dorsal fin type had better heat transfer
because of a higher total efficiency, including a higher electrical and performance compared to structure “B” in regular type, due to a
thermal efficiency. better synergy angle (closer to 0◦ or 180◦ ) between the velocity
Fig. 26 demonstrates the comparison of the efficiency of PV/T sys­ filed and temperature gradient field induced by structure “A”.
tems among the uncooled PV module, the PV module with the new (2) As solar radiation ranged from 300 W/m2 to 1200 W/m2, the
channel and the PV module with the conventional channel when the electrical efficiency and thermal efficiency with structure “A” of
inlet mass flow rate of cooling water increased at a solar radiation of shark dorsal fin decreased by 0.067% and 1.1% respectively per
1000 W/m2. With the same mass flow rate, the PV module with the new 100 W/m2 increase of solar radiation in average.
channel, had a higher cooling efficiency than that of the conventional (3) As the inlet mass flow rate of cooling water ranged from 0.0018 to
cooling channel. 0.006 kg/s, the cooling efficiency was improved rapidly.
The performance comparison between PV/T systems with the con­
ventional smooth channel and the special shark dorsal fin type channel
(the optimized structure with a depth of 0.03 m, a sawtooth peak of 0.02
m, a sawtooth width of 0.01 m), was analyzed in this section. Fig. 27
shows the surface temperature distribution of PV/T systems with a solar
radiation of 1200 W/m2, an inlet mass flowrate of 0.009 kg/s and an
inlet temperature of 20 ◦ C. The average cell temperature of the PV/T
system with the conventional smooth cooling channel (35.303 ◦ C) was
6.05 ◦ C higher than that of the shark dorsal fin type cooling channel
(29.253 ◦ C). In comparison with the conventional smooth channel, the
electric efficiency and thermal efficiency with special shark dorsal fin
type channel increased by 0.3% and 3.9% respectively, indicating that
the special shark dorsal fin type channel was preferable.

4.3.2. Performance comparison under actual operating conditions

In order to see the performance of the new PV/T system in real
conditions, this section focused on the efficiency of the new PV/T system
under actual operating conditions. The data of a typical day were
selected from the literature [30], including solar radiation (23–546 W/ Fig. 28. Efficiency comparison chart under actual operating conditions.

Fig. 27. The surface temperature of the smooth channel and the shark dorsal fin type channel.

13
C. Shen et al. Applied Thermal Engineering 185 (2021) 116381

However, when the inlet mass flow rate increased from 0.006 to [14] A. Özakin, F. Kaya, Effect on the exergy of the PVT system of fins added to an air-
cooled channel: A study on temperature and air velocity with ANSYS Fluent, Sol.
0.018 kg/s, the cooling efficiency rose slowly. The recommended
Energy 184 (2019) 561–569.
mass flow rate cooling water for the test device was 0.006 kg/s. [15] K. Sopian, K.S. Yigit, H.T. Liu, S. Kakaç, T.N. Veziroglu, Performance analysis of
(4) The optimal structure “A” (shark dorsal fin type) and structure photovoltaic thermal air heaters, Energy Convers. Manage. 37 (1996) 1657–1670.
“B” (regular type), had a channel depth of 0.03–0.05 m, a [16] P. Ooshaksaraei, K. Sopian, S.H. Zaidi, R. Zulkifli, Performance of four air-based
photovoltaic thermal collectors configurations with bifacial solar cells, Renewable
sawtooth peak of 0.02–0.035 m and a sawtooth width of 0.01 m. Energy 102 (2017) 279–293.
(5) Because the new channel has better performance in a high solar [17] A. Fudholi, K. Sopian, M.H. Yazdi, M.H. Ruslan, A. Ibrahim, H.A. Kazem,
radiation, it is more suitable for concentrating PV/T systems from Performance analysis of photovoltaic thermal (PVT) water collectors, Energy
Convers. Manage. 78 (2014) 641–651.
the application point of view. As for installation, there is no [18] A. Nahar, M. Hasanuzzaman, N.A. Rahim, A three-dimensional comprehensive
obvious difference with the conventional PV/T system. numerical investigation of different operating parameters on the performance of a
Regarding operation, there might be some dirt depositing in the photovoltaic thermal system with pancake collector, J. Sol. Energy Eng. 139
(2017).
new channel if the PV/T system operates for a long time because [19] A.A.B. Baloch, H.M.S. Bahaidarah, P. Gandhidasan, F.A. Al-Sulaiman,
of the corrugated structure [38]. Therefore, in the actual opera­ Experimental and numerical performance analysis of a converging channel heat
tion process, it is necessary to clean the cooling channel regularly. exchanger for PV cooling, Energy Convers. Manage. 103 (2015) 14–27.
[20] A. Fudholi, M. Zohri, G. Jin, A. Ibrahim, H.-Y. Chan, M. Othman, M.H. Ruslan,
K. Sopian, Energy and exergy analyses of photovoltaic thermal collector with
Declaration of Competing Interest ∇-groove, Sol. Energy 159 (2018) 742–750.
[21] B. Boumaaraf, H. Boumaaraf, E.A. Slimani, S. Tchoketch-Kebir, M.S. Ait-Cheikh, K.
J.M. Touafek, c.i. simulation, Performance evaluation of a locally modified PV
The authors declare that they have no known competing financial
module to a PV/T solar collector under climatic conditions of semi-arid region, 167
interests or personal relationships that could have appeared to influence (2020) 135–154.
the work reported in this paper. [22] C. Zhang, C. Shen, S. Wei, Y. Wang, G. Lv, C. Sun, A review on recent development
of cooling technologies for photovoltaic modules, J. Therm. Sci. (2020), https://
doi.org/10.1007/s11630-020-1350-y.
Acknowledgement [23] Y. Li, Z. Wang, J. Yang, H. Liu, Thermal and hydraulic characteristics of
microchannel heat sinks with cavities and fins based on field synergy and
The authors gratefully acknowledge the funding support from the thermodynamic analysis, Appl. Therm. Eng. 175 (2020) 115348.
[24] B. Mehra, J.V. Simo Tala, C. Habchi, J.L. Harion, Local field synergy analysis of
Natural Science Foundation of Heilongjiang Province (Project#: conjugate heat transfer for different plane fin configurations, Appl. Therm. Eng.
YQ2020E019) 130 (2018) 1105–1120.
[25] C. Hu, M. Sun, Z. Xie, L. Yang, Y. Song, D. Tang, J. Zhao, Numerical simulation on
the forced convection heat transfer of porous medium for turbine engine heat
References exchanger applications, Appl. Therm. Eng. 180 (2020) 115845.
[26] Z.Y. Guo, D.Y. Li, B.X. Wang, A novel concept for convective heat transfer
[1] N. Franzese, I. Dincer, M. Sorrentino, A new multigenerational solar-energy based enhancement, Int. J. Heat Mass Transf. 41 (1998) 2221–2225.
system for electricity, heat and hydrogen production, Appl. Therm. Eng. 171 [27] J.M. Wu, W.Q. Tao, Investigation on laminar convection heat transfer in fin-and-
(2020) 115085. tube heat exchanger in aligned arrangement with longitudinal vortex generator
[2] A. Bianchini, A. Guzzini, M. Pellegrini, C. Saccani, Photovoltaic/thermal (PV/T) from the viewpoint of field synergy principle, Appl. Therm. Eng. 27 (2007)
solar system: experimental measurements, performance analysis and economic 2609–2617.
assessment, Renewable Energy 111 (2017) 543–555. [28] A. Nahar, M. Hasanuzzaman, N.A. Rahim, Numerical and experimental
[3] S.A. Kalogirou, Y. Tripanagnostopoulos, Hybrid PV/T solar systems for domestic investigation on the performance of a photovoltaic thermal collector with parallel
hot water and electricity production, Energy Convers. Manage. 47 (2006) plate flow channel under different operating conditions in Malaysia, Sol. Energy
3368–3382. 144 (2017) 517–528.
[4] A. Zarei, M. Liravi, M. Babaie Rabiee, M. Ghodrat, A Novel, eco-friendly combined [29] C. Taylor, P. Hood, A numerical solution of the Navier-Stokes equations using the
solar cooling and heating system, powered by hybrid Photovoltaic thermal (PVT) finite element technique, Comput. Fluids 1 (1973) 73–100.
collector for domestic application, Energy Convers. Manage. 222 (2020) 113198. [30] A. Tiwari, M.S. Sodha, A. Chandra, J.C. Joshi, Performance evaluation of
[5] R. Sellami, M. Amirat, A. Mahrane, M.-E.-A. Slimani, A. Arbane, R. Chekrouni, photovoltaic thermal solar air collector for composite climate of India, Sol. Energy
Experimental and numerical study of a PV/Thermal collector equipped with a PV- Mater. Sol. Cells 90 (2006) 175–189.
assisted air circulation system: configuration suitable for building integration, [31] M. Chandrasekar, S. Suresh, T. Senthilkumar, M. Ganesh karthikeyan, Passive
Energy Build. 190 (2019) 216–234. cooling of standalone flat PV module with cotton wick structures, Energy Convers.
[6] J. Yazdanpanahi, F. Sarhaddi, M. Mahdavi Adeli, Experimental investigation of Manage. 71 (2013) 43–50.
exergy efficiency of a solar photovoltaic thermal (PVT) water collector based on [32] H. Bahaidarah, A. Subhan, P. Gandhidasan, S. Rehman, Performance evaluation of
exergy losses, Sol. Energy 118 (2015) 197–208. a PV (photovoltaic) module by back surface water cooling for hot climatic
[7] A.H.A. Al-Waeli, H.A. Kazem, M.T. Chaichan, K. Sopian, Experimental conditions, Energy 59 (2013) 445–453.
investigation of using nano-PCM/nanofluid on a photovoltaic thermal system [33] H.G. Teo, P.S. Lee, M.N.A. Hawlader, An active cooling system for photovoltaic
(PVT): technical and economic study, Therm. Sci. Eng. Prog. 11 (2019) 213–230. modules, Appl. Energy 90 (2012) 309–315.
[8] A. Al-Waeli, K. Sopian, H.A. Kazem, M. Chaichan, Photovoltaic/Thermal (PV/T) [34] R. Nasrin, M. Hasanuzzaman, N.A. Rahim, Effect of high irradiation and cooling on
systems: status and future prospects, Renew. Sustain. Energy Rev. 77 (2017) power, energy and performance of a PVT system, Renewable Energy 116 (2018)
109–130. 552–569.
[9] S. García, A. Trueba, Influence of the Reynolds number on the thermal [35] R. Nasrin, M. Hasanuzzaman, N.A. Rahim, Effect of high irradiation on
effectiveness of tubular heat exchanger subjected to electromagnetic field-based photovoltaic power and energy, Int. J. Energy Res. 42 (2018) 1115–1131.
antifouling treatment in an open once-through seawater cooling system, Appl. [36] A. Aziz, A similarity solution for laminar thermal boundary layer over a flat plate
Therm. Eng. 140 (2018) 531–541. with a convective surface boundary condition, Commun. Nonlinear Sci. Numer.
[10] M.E.-A. Slimani, R. Sellami, A. Mahrane, M. Amirat, Study of Hybrid Photovoltaic/ Simul. 14 (2009) 1064–1068.
Thermal Collector Provided With Finned Metal Plates: A Numerical Investigation [37] A. Nahar, M. Hasanuzzaman, N.A. Rahim, S. Parvin, Numerical investigation on
under Real Operating Conditions, 2019. the effect of different parameters in enhancing heat transfer performance of
[11] A. Shukla, K. Kant, A. Sharma, P.H. Biwole, Cooling methodologies of photovoltaic photovoltaic thermal systems, Renewable Energy 132 (2019) 284–295.
module for enhancing electrical efficiency: a review, Sol. Energy Mater. Sol. Cells [38] C. Shen, Y. Wang, Z. Tang, Y. Yao, Y. Huang, X.J.I.J.o.H. Wang, M. Transfer,
160 (2017) 275–286. Experimental study on the interaction between particulate fouling and
[12] M.E.-A. Slimani, R. Sellami, M. Said, B. Boumaaraf, Numerical Simulation of a precipitation fouling in the fouling process on heat transfer tubes, 138 (2019)
Combined Solar PV/Thermal Device with two Mechanical Pre-Control Systems, 1238–1250.
2020.
[13] F. Hussain, M.Y.H. Othman, B. Yatim, H. Ruslan, K. Sopian, Z. Anuar,
S. Khairuddin, An improved design of photovoltaic/thermal solar collector, Sol.
Energy 122 (2015) 885–891.

14