Solar Energy 230 (2021) 943–953

Contents lists available at ScienceDirect

Solar Energy
journal homepage: www.elsevier.com/locate/solener

Cooling channel effect on photovoltaic panel energy generation

Zeynep Özcan, Miray Gülgün, Ecem Şen, Nezir Yağız Çam, Levent Bilir *
Energy Systems Engineering Department, Yaşar University, Üniversite Caddesi No: 37-39 Bornova, 35100 İzmir, Turkey

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

Keywords: It is a well-known fact that even though the electricity generation is higher when the solar radiation is high on a
Photovoltaic panel photovoltaic panel, its efficiency drops as its temperature increases. In this study, it is intended to achieve cooling
Efficiency effect using an air duct placed under a photovoltaic panel, thereby increase its efficiency. Hourly electricity
Electricity generation
generation, PV efficiency and cell temperature values over a year are calculated using annual temperature and
Cooling channel
CFD analysis
radiation data by using MATLAB and PV Sol software. Maximum cell temperature for the uncooled case is
determined as 57.91 ◦ C on July 21st at 1p.m. as a result of hourly calculations. The incident solar radiation is
976 W/m2 when the panel reached its maximum temperature. The PV panel and cooling channel are modelled in
ANSYS Fluent software and cooling effect was investigated for different air velocities and air-cooling channel
geometries for the hour when maximum cell temperature is reached. Environmental analyses are also made. It is
observed that with finned cooling channel, it is possible to cool PV temperature more than with the flat cooling
channel. Cooling the PV panel from its maximum cell temperature to 39.82 ◦ C with 5 m/s air velocity and 82 fins
cooling channel is achieved and new PV panel efficiency is recorded as 18.92 %. Environmentally considerations
show that the use of solar energy provides the reduction of coal and natural gas-based CO2 emissions as 15 and 8
tons, respectively.

is converted into heat, which increases the temperature of the cells and
reduces the performance of the PV module. There are many ways to cool
1. Introduction a PV module for increasing efficiency. The cooling of PV cells is divided
into five main topics; passive cooling techniques, heat pipe cooling,
In today’s society, the effective use of renewable energy is becoming active cooling methods, nano-fluid cooling and thermoelectric cooling.
mandatory with the increase of human population and environmental Erkan et al. (2018) mentioned that cooling with a channel method can
issues. The most important issue in today’s world is the efficient usage of be active or passive. Active systems use energy consuming equipment
energy. Erkan et al. (2018) indicated that decreasing the initial costs, such as pump or fan, but passive systems do not require any energy
reducing and recovering waste energy and efficient usage of energy are consuming equipment.
the main research topics. The consumption of fossil fuels causes envi­ There had been many researches related to cooling processes of PV
ronmental pollution and global warming, which are the most importants panels, such as Mittelman et al. (2009) who aimed cooling the PV panel
problem that emerged in our age. However, solar energy is an important with radiation and free convection. Similarly, Palumbo (2013) devel­
renewable energy source, with its abundance, simple technology to use, oped the cooling channel that worked with natural convection by using
and environment friendly nature. Because of these reasons, the use of air as cooling material and he tested his system with different air ve­
solar energy, which is a clean, endless and low-cost source of energy, has locities using fans. Kaiser et al. (2014) aimed to provide an open-air
increased in the past years. Photovoltaic (PV) panel studies have also channel for PV panel to achieve cost effective system. Moreover, they
been increased day by day. Researchers often emphasize that one of the used a ventilation system to record temperature differences by sending
improvement studies for the PV panels is to increase the efficiency of PV different air velocities. Baloch et al. (2014) stated the working principle
cells by cooling. of a cooling for photovoltaic by using jet impingement. Irwan et al.
There are two types of energy that can be produced from solar en­ (2015) developed a solar simulator with halogen lamps to analyse the
ergy: electrical energy and thermal energy. The electrical energy can be performance of a PV panel with and without air cooling mechanism.
produced by using PV cells which directly convert a part of the incident Baloch et al. (2015a) observed the behavior of thermal and electrical
solar irradiance to electricity. The remaining part of the solar irradiation

* Corresponding author.
E-mail address: [email protected] (L. Bilir).

https://doi.org/10.1016/j.solener.2021.10.086
Received 24 March 2021; Received in revised form 3 September 2021; Accepted 31 October 2021
Available online 12 November 2021
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Nomenclature Ta Ambient Temperature (◦ C)

Tc Cell Temperature (◦ C)
A Cross Sectional Area (m2) Tref Reference Temperature (◦ C)
APV PV Panel Area (m2) v̇ Volumetric Flow Rate (m3/s)
Ei Hourly PV Electricity Generation (W) w Hour Angle (◦ )
Gref Ground Reflection ws Sunset Hour Angle (◦ )
Id Diffuse Radiation (W/m2)
IG Global Radiation (W/m2) Greek Letters
IT Hourly Radiation Incident of the Tilted PV Panel (W/m2) ß Slope angel of PV (◦ )
k Ventilation coefficient δ Declination Angle, Solar Radiation Coefficient (◦ )
n Day Number ƞ Efficiency (%)
npv Number of PV Panel ƞmp,ref PV Panel Efficiency(%)
Rb The ratio of beam radiation on the inclined surface to a μmp Temperature Coefficient of Maximum Power Point
horizontal surface ρ Density (kg/m3)
T Temperature (◦ C) ρg Ground Reflectivity

properties for a cold PV system and a combined duct cooling PV system the air blown by the help of a fan. The inclined PV panel is located in
in Dharan climate. Baloch et al. (2015b) aimed to obtain low and uni­ İzmir. In summer season, due to high radiation, cell temperature in­
form temperature on the PV panel using a convergent water channel creases rapidly. Cell temperature calculations are performed with
cooling. According to their study, water consumption was found to be MATLAB. According to temperature rising, efficiency of PV module is
dependent on the application of PV systems. Zeyad et al. (2018) focused decreasing. Therefore, in order to prevent this, PV panel must be cooled.
on evaporative cooling using water. In their system, water was supplied For this reason, cooling channel is installed under the PV panel. Analyses
from a tank by gravity to the back of the PV panel. Wu et al. (2018) built with ANSYS Fluent software are performed using the solar irradiation at
a three-dimensional numerical model of the water-cooled PV/T system the time when maximum PV temperature is reached and considering
with cooling channel on the PV panel. Erkan et al. (2018) investigated cooling channels with different air velocities and different geometries.
the cooling of a single crystalline photovoltaic cell using computational Studies in literature are primarily focused on theoretical and numerical
fluid dynamics method by taking water as a cooling material. Bayrak assessments of PV cooling systems using various cooling materials such
et al. (2019) experimentally investigated the performance of a poly­ as air, water and ethanol–water mixture, as can be seen on the above
crystalline 75 W PV panel with a cell structure in Elazığ, Turkey climatic literature survey. Even though there are systematic studies available in
conditions. They analyzed system performances such as temperature, the literature that compare various channel types such as flat plate and
power and efficiency by applying different fin parameters to PV panels. finned plate using air, Maleki et al. (2020) stated that other forms of
Wu et al. (2019) worked to investigate the effect of cooling channel channels, such as zig-zag and wavy-shaped one, would be advantageous.
position on heat transfer and thermoelectric behaving of air-cooled PV/ Additionally, according to the same authors, optimization studies on the
T systems. Zheng et al. (2019) focused on the passive air-cooling system channel’s geometrical qualities, as well as other operating conditions,
for PV, they showed the effect of channel geometry on the cooling would be advantageous in order to reach thermal solutions with envi­
performance of the system. Han et al. (2019) investigated two- ronmental analysis. As far as the authors’ best knowledge, the investi­
dimensional numerical analysis of the fluid flow and heat transfer be­ gation of curved fins was not been performed in detail in any other
haviors of the natural convection driven by the buoyancy force in the study. Additionally, the main objective of this study is based on the lack
passive cooling air duct formed by the two vertical parallel walls. Pio­ of optimization studies and making performance analyses with channel
trowskia et al. (2020) presented a technical and economic analysis of the design since they are new and open in development. Moreover, this
efficiency and durability of PV panels with and without a cooling system study focuses on improving PV panel performance by lowering panel
and they observed an improvement of 3% in the cell efficiency of cooled surface temperatures using various air channels such as flat plate, finned
down panels. Amber et al. (2020) evaluated the performance of the plate and curved plate geometries and to determine the best design.
passive cooling technique by using circular and rectangular fins applied More specifically, in this study, detailed cell temperature, electricity
to the rear surface of the PV panels. The efficiency obtained by applying generation and PV efficiency are investigated, the optimum channel is
rectangular and circular fins increased by 14.5% and 13.2% compared to selected and environmental analyses are made. In the determination of
the reference module, respectively. As a result, the rectangular fins are the optimum channel, net electricity gain (increase in PV electricity
decided for PV installations. Egab et al. (2020) investigated the reduc­ generation minus fan electricity use to provide cooling) value was taken
tion of the panel temperature using an air-cooled heat sink. The system into consideration, which greatly makes the results more reasonable.
consists of rectangular fins and rectangular fins with holes. According to Hence, it can be said that the present study provides a very detailed
the results, the temperature of the PV panel with fins decreased by 50% numerical analysis about the effect of the use of flat and especially
compared to the PV panel without fins. Arifin et al. (2020) designed an curved finned cooling channels on the electricity generation perfor­
aluminum plate with perforated fins mounted to the back of the panel in mance of PVs. The evaluation of the impact of curved finned cooling
order to reduce the operating temperature of the PV panel. According to channel can be considered as an addition to the literature. Moreover,
CFD analysis, the average temperature of the PV panel decreased from considering the net electricity generation increase (electricity genera­
85.3◦ C to 72.8◦ C when 1.5 m / s of air was given to the heat sink and the tion increase due to the cooling minus electricity consumption of the fan
electrical efficiency increased by 2.6%. Consequently, the effects of the in order to provide cooling), the best cooling channel option with the
heat sink on the heat transfer area and heat transfer performance were best air velocity value is determined as the result of the study, which can
understood. We develop our cooling design and pursue performance also be considered as a novelty. Finally, the environmental analyses
analysis on a specific PV panel, selected as Panasonic N330. The PV show the benefit of the use of solar energy in terms of sustainability and
panel cells are monocrystalline silicon and have 19.4% efficiency under environmental friendliness of the renewable energy.
standard test conditions (air mass = 1.5, Radiation = 1000 W/m2, cell
temperature = 25 ◦ C). A cooling channel is used to cool the PV cell with

944
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Fig. 1. Schematic description of PV panel (a) Ansys-Fluent model, (b) Dimensions of the PV panel (Panasonic Module HIT Datasheet, 2021).

2. Description of the system cooling channel continues along the channel and takes heat from the PV
panel. At the end, heated air leaves the channel and the PV panel has
The inclined Panasonic N330 PV panel is located with 30◦ angle in been cooled. The geometrical dimensions of the evaluated PV and its
İzmir, Turkey. Panasonic N330 has monocrystalline silicon cells and a Ansys-Fluent model are presented in Fig. 1.
19.7% efficiency. PV panel has 1053 mm width, 1590 mm length and 45
mm thickness. The PV panel with cooling channel is constituted of layers 3. Analysis
of glass, airgap, PV cells, aluminum plate and cooling channel. The
thicknesses of these layers are 10 mm, 20 mm, 1 mm, 4 mm and 10 mm, In the first part of the research, a comprehensive numerical analysis
respectively. For all surfaces, except the upper surface, convective heat of the PV panel is performed using MATLAB software to simulate time-
transfer with the ambient air at 25 ◦ C is assumed. Cooling channel dependent changes. Hourly diffuse, global radiation and ambient tem­
material is aluminum and current cooling fluid is air. The outlet pressure perature data for one year were used to calculate hourly electricity
of the cooling channel is accepted as atmospheric pressure. generation and panel temperature values. For the design, the reference
When photons from the sun arrives the PV panel, it activates elec­ temperature was taken as 25 0C and the slope angle of the PV was taken
trons, causing them to separate from their atoms. If the conductors are as 30◦ . It is aimed to calculate the monthly electricity generation of this
connected to the solar cells, an electrical circuit is formed, and elec­ PV panel and to compare them with the values obtained in the PV Sol
tricity is generated by the flow of electrons in the circuit. Electricity software. Consequently, the change in the cooling amount, PV electricity
generation reduces the efficiency of the system by allowing cells to heat generation and efficiency depending on the velocity of the air supplied
up over time. While generating electricity, air flows simultaneously to the evaluated cooling channels are observed.
through the cooling channel for the cooling process. The air sent into the In the second part of the study, analyses are made using ANSYS

Fig. 2. Steps of the analyses in the present study.

945
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Fluent software. Cooling channel designs are investigated according to Table 1

the parameters calculated in the previous section. The designs are Evaluated data used in calculations.
created with different geometrical shapes and different air velocities. Ground Reflection 0.2
The considered geometrical shapes of the cooling channel are flat Slope Angle of PV (β) 30◦
channel, channel with flat fins and channel with curved fins. After Latitude of Izmir (ϕ) 38.4
cooling analyses in ANSYS Fluent software, air pressure drops encoun­
( )
PV efficiency ηmp,ref 19.7%
tered during the air flow through the cooling channel and new cell Reference Temperature 25 ◦ C
temperatures of the PV panel are recorded. New efficiencies of the PV Global Radiation 1000 W/m2
panel and new electricity generations with cooling channels are calcu­ Number of PV 1
PV Area 1.67 m2
lated in MATLAB software. Fan powers required for the system are
found. The optimal design is chosen by calculating net production by
subtracting fan power required for air flow inside the cooling channel ⎛ ⎛ ⎞⎞
from PV electricity generation. The number of panels for a typical house ( ) IT ⎠ ⎠
roof installation is taken into consideration and environmental analyses ηi = ηmp,ref ⎝
× 1 − μmp Tc − Tref + δ.ln ⎝ (7)
Gref
are also performed. The procedure followed in the study is summarized
in Fig. 2. where, ηmp,ref is PV panel efficiency, μmp is the temperature coefficient
of maximum power point efficiency which is 0.00258 and Tref is the
3.1. Electricity generation analyses reference temperature.

The hourly electricity generation formula and steps of the calcula­

• Hourly PV electricity generation (Ei ):
tions are performed according to Duffie and Beckman (2013) listed by
Eqs. 1–9 as follows; Ei = ηi × nPV × APV × I T (8)
where nPV is the number of PV and APV is PV area.
• Declination Angle (δ):
( )
284 + n • Some amount of solar radiation has been absorbed by glass and PV
δ = 23.45 × sin 360 × (1)
365 cells. Therefore, net heat flux (W/m2) is:

where, n is the day number in the year from 1 to 365. Heat Flux = Glass Transmissivity × Incoming Radiation (9)
Data used in PV electricity calculations are presented in Table 1.
• Hour Angle (w):
w = (Solar Time − 12) × 15o (2) 3.2. System and software analysis
where, solar time is hours of the day from 1 to 24.
3.2.1. PV simulation using MATLAB
• Sunset Hour Angle (ws ): Using hourly diffuse, global radiation and ambient temperature data
[ ] for one year, declination angle (δ), hour angle (w), sunset hour angle
ws = min
arccos( − tanϕtanδ)
(3) (ws ), beam radiation (Rb ), hourly radiation incident (IT ), cell tempera­
arccos( − tan(ϕ − β)tanδ )
ture (Tc ), hourly PV efficiency (ηi ), hourly PV electricity generation (Ei )
where, ϕ is latitude of Izmir and β is slope angel of PV. and maximum cell temperature (Tc,max ) are calculated in MATLAB. The
maximum cell temperature is encountered as 57.91 ◦ C. These values are
• The ratio of beam radiation on the inclined surface to a horizontal also compared with the results obtained from PV Sol software to make
surface for the collector directed south in the northern hemisphere the validation of the results of the code written in MATLAB software.
(Rb ): Moreover, the panel efficiency and electricity generation values, which
depend on the new cell temperature (resulting from the created cooling
cos(ϕ − β)cosδcosw + sin(ϕ − β)sinδ
Rb = (4) effect of the channels), are calculated using the same formula.
cosϕcosδcosw + sinϕsinδ
3.2.2. PV simulation using PV Sol software
• Hourly radiation incident of the tilted PV panel (IT ): In PV Sol software, the system is designed with two PV panels. Set up
( ) ( ) ( ) type was 3D, grid-connected PV system and Panasonic N330 PV panels
I T = I G − I d Rb + I d
1 + cosβ
+ I G ρg
1 − cosβ
(5) are selected. Climate data are chosen for İzmir/Çiğli covering the period
2 2 of January to December 2018. The system is located on a roof with 30
◦
angle. A single inverter is used in the system. Cable losses are taken as
where, IG is the global radiation, Id is the diffuse radiation and ρg is
1% and there is no energy transfer from the grid to the system.
ground reflection.
If |w|>|ws| => IT = 0
3.2.3. Numerical model on ANSYS Fluent
Software analyses are made based on the maximum cell temperature,
• Cell temperature (Tc ):
which is 57.91 ◦ C, because as it was mentioned earlier that the main aim
Tc = Ta + k × I T (6) of this study is to provide cooling which compensate the temperature
rising of the PV. ANSYS Fluent analyses for the case with flat cooling
where, Ta is the ambient temperature for each hour and k is the channel are made for different air velocities, which are 5, 7 and 10 m/s.
ventilation coefficient, taken as 0.2 (Erkan et al., 2018) The air velocity value is taken as 3 and 5 m/s for the cases with flat or
curved fins. Furthermore, after the determination of the cooling channel
• Hourly PV efficiency (ηi ): that provides the best PV panel efficiency and electricity generation, an
environmental analysis is performed taking the number of panels as 30
Panasonic N330 330 W panels (the total capacity is 9.9 kW). The se­
lection of the PV number is determined considering the installation of

946
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Fig. 3. View of PV panel and cooling channel with 82 flat fins in ANSYS Fluent (a) Front view of half PV and cooling channel, (b) Bottom view of half PV and
cooling channel.

unlicensed roof-top on grid solar power plant limit of 10 kW in Turkey. 3.3. Environmental analysis
In the ANSYS Fluent software, firstly, flat plate cooling channel is taken
into consideration with different air velocity values. Then, flat finned There are many reasons for using renewable energy types in the
cooling channel is evaluated with different air velocities. Finally, curved world and preferring solar energy systems is one of the best ways to
finned cooling channel is taken into account with the same air velocity contribute to a healthier environment. Solar energy is the most abun­
values. Fan power consumption values in order to supply the required dant and ready energy source and can be used unlimitedly. Also, solar
air volumetric flow rate and pressure drop values are also calculated and systems do not necessarily need water to work, thus, it does not play a
considered. As a result, the curved finned cooling channel is found to role in the consumption of water resources. Moreover, an important
provide the best cooling amount. In order to obtain better analysis re­ advantage for the environment is that it does not release chemicals such
sults, as many as possible mesh elements are used and 1,717,100 and as carbon dioxide, methane and nitrous oxide to the atmosphere, and
1,816,400 and 5,281,744 meshes were generated for cooling channel does not cause air pollution and health problems. The calculations of
with 71 flat fins, 82 flat fins and curved fins with 30 mm fin distance, carbon dioxide emissions from coal and natural gas that are prevented as
respectively. Half of the system is modelled and one side surface of the a result of PV panel installation are shown in the results part by using
system is defined as symmetry in order to be able to use finer meshes, data given in Table 3.
which result in more accurate outcomes. Moreover, the heat transfer Total amount of CO2 emission from source;
coefficient on all outer surfaces (defined as wall) is 5.8 W/m2K, the heat
Electricity Generation × CO2 Emission Value (10)
flux on the top surface is defined as 597 W/m2. This heat flux value is
determined with the help of inclined radiation that is coming to the where, Electricity Generation in GWh and CO2 Emission Value in tons/
surface of the panel (975.97 W/m2), the slope angle of the panel area of GWh.
the PV panel (30◦ ), absorptivity of the PV cell and transmissivity of the
glass. 4. Validation of the model
The model with a flat plate channel is consisted of the glass layer on
the top, the air layer between the glass and the absorber, the cooling Baloch et al. (2015a,b) analyzed a convergent water channel cooling
channel layer and the cooling air layer. The model view of the PV with which aimed to obtain low and uniform temperature on the PV panel
flat fins, which includes only the half of the PV and the channel due to surface. They made evaluations for an uncooled PV system and a
symmetrical conditions, is illustrated in Fig. 3. Similarly, the model with converged channel-cooled PV system due to the climate of Saudi Arabia
curved finned cooling channel, which includes only the half of the PV during June and December. Before cooling the PV string, the maximum
and the channel, can be seen in Fig. 4. temperature was recorded as 71.2 ◦ C and 48.2 ◦ C in June and December,
The main geometrical dimensions of the models with cooling chan­ respectively.
nels are summarized in Table 2. The mesh methods applied in this study are similarly applied to the
study of the Baloch et al. (2015a,b) and the obtained results are
compared with their outcomes.
As it is seen in the Table 4, when water with a temperature of 27 ◦ C is

947
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Fig. 4. View of PV panel and cooling channel with curved fins (with 30 mm fin distance) in ANSYS Fluent (a) Front view of half PV and cooling channel, (b) Bottom
view of half PV and cooling channel.

Table 2
Geometrical dimensions of the models with cooling channels.
Fin Type
Dimensions Flat Curved

71 fins 82 fins

Length 1590 mm
Thickness 5 mm 3 mm 1.5 mm
Installation spacing 10 mm 30 mm
Curvature radius – – 26.5 mm

Table 3
CO2 emission values (Enerji Atlası, 2020).
Source Approximately CO2 emission (tons-CO2/ /GWh)

Coal 888
Natural gas 499

Fig. 5. Geometry of PV panel with cooling channel in the study of Wu

Table 4 et al. (2019).
Comparison of water outlet temperatures and cell temperatures between the
applied mesh methods into Baloch et al. (2015a,b)
given to the cooling channel at a velocity of 0.012 m/s, the water outlet
Water velocity Water outlet Cell temperature
temperature is found to be 29.77 ◦ C, while the temperature of the cell
(m/s) temperature (◦ C) (◦ C)
before cooling (which is 71.2 ◦ C) is found to drop to 42.79 ◦ C after
Results of Baloch 0.012 30.50 45.10 cooling in the applied mesh methods in Baloch et al. (2015a,b).
et al. (2015a,b)
In the study of Baloch et al. (2015a,b), the temperature of the water
Present study 0.012 29.77 42.79
results entering the cooling channel at a velocity of 0.012 m/s increased from
27 ◦ C to 30.5 ◦ C. The cell temperature decreased from 71.2 ◦ C to 45.1 ◦ C.
Looking at the results obtained in the present study and the values given

948
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Table 5 presented (Fig. 7a). Secondly, comparison of electricity generation data

Comparison of air temperatures at different locations with the results given by calculated with MATLAB and PV Sol is made (Fig. 7b and Table 6). Next,
Wu et al. (2019) heat transfer characteristics and temperature differences occurred dur­
Location x=0m x = 0.5 x=1m x = 1.5 x=2m x = 2.4 ing heat transfer between PV and cooling channel are examined using
m m m the calculated data from ANSYS Fluent software. Finally, all results and
Results of Wu 298.15 298.81 299.19 299.52 299.83 300.03 graphs from Fluent software are discussed.
et al. (2019) As it is seen in Table 6, monthly electricity generation values, which
Present study 298.15 298.50 298.90 299.33 299.76 300.10 are calculated in MATLAB, are almost the same as the values found by
results
PV Sol. The maximum electricity generation is seen in July with 56 kWh.
The PV reaches its maximum temperature at July 21st at 1p.m. The
by Baloch et al. (2015a,b), it can be seen that both the cell temperature values about the PV and solar radiation at this time are summarized in
and water outlet temperature values are near in both studies. Table 7.
Another validation is made with the study of Wu et al. (2019). The When the highest cell temperature which is 57.91 ◦ C is reached, PV
investigated geometry view is presented in Fig. 5. panel efficiency is found to be 18%, solar radiation is seen as 975.97 W/
They numerically studied the effect of cooling channel location on m2 and electricity generation is calculated as 293 W.
the heat transfer properties and thermoelectric performance of PV/T The effect of cooling with a flat plate cooling channel is investigated
systems. Air was used as the cooling fluid and air inlet temperature was firstly using Ansys-Fluent software. The cell temperature, the PV tem­
25 ◦ C, while the ambient temperature was taken as 20 ◦ C. Air inlet ve­ perature drop due to cooling, the electricity generation value as a result
locity to the channel was taken as 0.93 m/s and wind velocity was of the cooling effect and the relevant PV efficiency are summarized for
considered as 1 m/s. Similar mesh applications were applied to the different air velocities in Table 8. In the numerical calculations, residual
model of Wu et al. (2019) study and 428,400 volume elements are values smaller than 1 × 10− 3 for continuity, 1 × 10− 6 for momentum
created. The mesh number in Wu et. al (2019) was 421,200. The insu­ equations and 1 × 10− 8 were considered for the converged solution.
lation layer is not modeled in Fluent, because the bottom surface is As it is seen in Table 8, the highest amount of cooling is provided
defined as insulation, so there is no need to draw the insulation as a with 10 m/s air velocity and the electricity generation is increased
volume. The wind effect is considered by defining convection to the side significantly. For the 5 m/s air velocity case, electricity generation is
walls of the model. The convection coefficient is calculated as 9.5 (W/ increased from 293 W to 302.95 W so that 3.4% increase has been
m2K) using the formula given in Wu et al. (2019), which considers the happened. With 7 m/s air velocity, electricity generation is risen from
wind velocity and the free steam temperature is taken as 20 ◦ C as this is 293 W to 307.01 W, which means an increase of 4.78%. Finally, with 10
the ambient temperature. m/s air velocity, electricity generation is increased from 293 W to 310.4
As a result, the air temperature at different locations (namely at the W so that 5.94% increase has occurred. As an expected result, when the
inlet, at x = 0.5, 1, 1.5, 2 and 2.4 m from the inlet) were compared with air velocity increases, the cooling effect and hence the electricity gen­
the values given in Wu et al. (2019) in Table 5. It can be seen that very eration increases for flat plate cooling channel. After completing flat
similar results were obtained with the ones presented in Wu et al. plate analysis, different geometrical shapes for cooling channel are
(2019). considered. Flat finned and curved finned cooling channel effect are
The air temperature contours at the defined locations are presented investigated.
in Fig. 6. The heating of air is clearly observed in this figure. In Table 9, it is clearly seen that, the channel with flat fins provides
As a result of two different comparisons, it can be clearly seen that much more cooling than the flat plate channel. The cooling effect
the using of the current modeling and meshing strategy gives reliable observed with 82 fins and 3 m/s air velocity is close to the case with flat
outcomes. Therefore, it can be said that the results obtained in the plate cooling channel with 5 m/s air velocity. Additionally, a similar
present investigation are reliable. cooling is achieved for the case with 71 flat fins, 5 m/s air velocity and
for the case with flat plate cooling channel, 7 m/s air velocity. In the case
5. Results with 82 flat fins and 5 m/s air velocity, almost the same cooling value is
observed as the case with flat plate cooling channel and 10 m/s. The air
In this section firstly, monthly electricity generation of the PV is velocity is limited with 5 m/s in this section considering that the flat fins

Fig. 6. Air temperature contours at the defined locations.

949
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Fig. 7. Comparison of monthly electricity generation based on MATLAB and PV Sol (a) Monthly electricity generation graph in PV Sol, (b) Monthly electricity
generation graph in MATLAB and PV Sol.

Table 6 Table 7
The comparisons of electricity generation values between MATLAB and PV Sol MATLAB result for the day in which maximum PV temperature is encountered.
Months MATLAB PV Sol At Maximum Cell Temperature: 57.91 (◦ C)

January 26.61 kWh 28.3 kWh PV Efficiency (%) 18

February 31.42 kWh 28.5 kWh Solar Radiation (W/m2) 975.97
March 45.73 kWh 42 kWh Electricity Generation (W) 293
April 46.18 kWh 48 kWh
May 59.51 kWh 55 kWh
June 64.63 kWh 56 kWh will create high pressure drop if the air velocity is increased more.
July 66.60 kWh 59.2 kWh
Additionally, as nearly the same amount of cooling with the flat plate is
August 65.07 kWh 57.8 kWh
September 58.50 kWh 52 kWh
achieved with a slower air speed when the fins are used, higher air ve­
October 43.91 kWh 42.9 kWh locity values are considered unnecessary as they can dramatically in­
November 32.30 kWh 32 kWh crease the needed fan power consumption.
December 20.73 kWh 25.1 kWh In Fig. 8, the temperature distribution on the PV module is given. As,
Total 561.19 kWh 526.8 kWh
symmetry condition is taken in the model, the right surface represent the
middle area of the PV.

950
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Table 8 transfer even higher compared to flat fins. In this part of the study, the
Comparison of the results for air velocities of 5 m/s, 7 m/s and 10 m/s for the same air velocity values are used as the ones used in flat fins part of the
case with flat plate cooling channel. study. At the first look, it can be said that the use of curved fins is more
Inlet Cell PV Electricity PV advantageous as they provide higher cooling effect and thus higher
Velocity Temperature Temperature Generation Efficiency electricity generation. The values calculated for the flat plate, flat fins
(m/s) (◦ C) Drop (◦ C) (W) (%) and curved fins are summarized in Table 11.
5 46.40 11.51 302.95 18.59 For each cooling channel design, a decrease in cell temperature is
7 41.49 16.42 307.01 18.84 observed when the air velocity increases. Consequently, electricity
10 37.40 20.51 310.40 19.04
generation and panel efficiency are increased. The highest panel effi­
ciency, which is 19.08%, is reached with curved finned cooling channel.
As it is seen in Table 10, if curved finned cooling channel is mounted Even though the increasing air velocity increases the electricity
to PV module assembly, higher cooling effect is reached. It can be seen generation of the PV, it also increases the air pressure drop through the
that the PV efficiency takes a higher value than 19%, when the air ve­ cooling channel. The net electricity generation increase should also be
locity is 5 m/s. The curved shape of the fins provides a longer way for the taken into account by considering the electricity consumption of the fan
air inside the cooling channel and the disturbance effect makes the heat to supply the required air volumetric rate with the required pressure

Table 9
Comparison of the results with 3 m/s and 5 m/s air velocities for flat finned cooling channel.
Inlet Air Velocity (m/s) Cell Temperature (◦ C) Temperature Difference (◦ C) Electricity Generation (W) PV Efficiency (%)

71 Fins 3 49.74 8.17 300.18 18.42

5 41.36 16.55 307.12 18.84
82 Fins 3 47.48 10.43 302.05 18.53
5 39.82 18.02 308.40 18.92

Fig. 8. Temperature contour of the PV module with 82 fins for 5 m/s in ANSYS Fluent.

Table 10
The results with 3 m/s and 5 m/s air velocities for curved fined cooling channel.
Fin Distance (mm) Inlet Velocity (m/s) Cell Temperature (◦ C) Temperature Drop (◦ C) Electricity Generation (W) PV Efficiency (%)

30 3 42.61 15.30 306.09 18.78

5 36.66 21.25 311.01 19.08

Table 11
The results of all types geometry.
Inlet Velocity (m/s) Cell Temperature (◦ C) Temperature Drop (◦ C) Electricity Generation (W) PV Efficiency (%)

Flat Plate
5 46.4 11.51 302.95 18.59
7 41.49 16.42 307.01 18.84
10 37.40 20.51 310.40 19.04
Flat Fins
71 Fins 3 49.74 8.17 300.18 18.42
5 41.36 16.55 307.12 18.84
82 Fins 3 47.48 10.43 302.05 18.53
5 39.82 18.02 308.40 18.92
Curved Fins
30 mm 3 42.61 15.30 306.09 18.78
5 36.66 21.25 311.01 19.08

951
Z. Özcan et al. Solar Energy 230 (2021) 943–953

Table 12 For 30 PV panels without cooling channel, annual electricity gen­

Pressure Drops, Volumetric Flow Rate and Required Fan Power according to the eration is calculated as 16,835 kWh. A 3.37% yearly loss assumption is
all types of geometry. made. The annual electricity generation for 30 PV panels with cooling
Channel Geometry Pressure Drop Volumetric Flow Rate Fan Power channel (82 fins and 5 m/s air velocity) is calculated as 17,400 kWh.
(Pa) (m3/s) (W) 888 tons of carbon dioxide emerges as a result of burning coal for 1
Without fins (5 m/ 56.83 0.05265 2.99 GWh electricity generation in coal-based power generation plants. By
s) putting a cooling channel, 506.2 kg coal based, or 270 kg natural gas-
Without fins (7 m/ 96.97 0.07371 7.41 based carbon dioxide emissions are prevented compared to the case
s)
without cooling channel. Considering the use of 30 PVs with a cooling
Without fins (10 175.05 0.10530 18.43
m/s) channel with flat fins, 15,451.2 kg coal based and 8,682 kg natural gas-
With 71 fins (3 m/ 58.49 0.01156 0.67 based CO2 emissions are prevented, respectively, while 17,400 kWh
s) electricity is generated for an unlicensed PV panel station. In the case of
With 71 fins (5 m/ 126.41 0.01970 2.49 PV use without cooling channel, 14,945 kg coal based and 8,390 kg
s)
With 82 fins (3 m/ 56.15 0.01284 0.72
natural gas-based CO2 emissions can be prevented, respectively, with
s) 16,835 kWh electricity generation, as can be seen in Fig. 9.
With 82 fins (5 m/ 121.25 0.02140 2.59
s) 6. Conclusion
30 mm (3 m/s) 415.78 0.01525 6.34
30 mm (5 m/s) 1124.55 0.02542 28.58
In this study, the monthly electricity generation, panel efficiency,
solar radiation and cell temperatures are found in MATLAB software by
value. In Table 12, the required fan power for different cases are given. using the hourly solar and temperature data of 2018 for İzmir, Turkey.
Pressure drop values are calculated with ANSYS Fluent Software by The cooling channel is designed by adhering to the highest cell tem­
taking difference of inlet and outlet pressure values of air into consid­ perature. It is observed how much electricity generation and panel ef­
eration. ficiency increase can be achieved depending on the drop in cell
temperature by sending air at different velocities to this channel by
ΔP(Pa) = Poutlet − Pinlet (11)
performing simulations in ANSYS Fluent software. Different geometrical
The calculations of the required fan power and net electricity gen­ shapes for the cooling channel are tried such as the plate channel, the
eration are calculated as; channel with flat fins and the channel with curved fins. At the end,
higher cooling effect is observed on the PV with these new channels.
Fan power(W) = ΔP(Pa) × V̇(m3 /s) (12) Pressure drop values are also discussed.
As a conclusion, it is observed that, it is possible to cool the PV with
Net generation = Electricity generation(W) − Fan power(W) (13)
curved finned cooling channel more than with the flat finned cooling
From Table 12, it can be clearly seen that even though the increase in channel. It is also observed that the maximum cooling of the PV panel is
heat transfer provides a high electricity generation efficiency for the encountered with the cooling channel design with curved fins and 5 m/s
case with curved fins, the increasing air pressure drop may surpass this air velocity. In this case the temperature of the PV is decreased to
increase as there will be a high fan electricity consumption. For this 36.66 ◦ C from its maximum cell temperature, which is recorded as
reason, the net electricity generation increase for each case is investi­
gated in order to see which case is optimum. The net electricity gener­
ation for all investigated cases are summarized in Table 13.
The results presented in Table 13 show an interesting point.
Although the use of curved fins provides higher PV electricity generation
values, the maximum net electricity is accomplished with the use of
cooling channel with 82 flat fins. The reason of this is that fan power
depends on pressure drops and volumetric flow rate in the air channel.
Moreover, pressure drops increase with increasing air velocity. In the
study, the required fan power increased when the number of fins is
increased for the same air velocities and when the curved fin air channel
geometry is used. The highest fan power was observed when curved fin
geometry is used and 5 m/s air velocity is given. Because the pressure
drop increases depending on the air velocity and therefore more fan
power is needed. It is concluded that the most efficient cooling channel
design is the one with 82 flat fins with 5 m/s air velocity. This design
provides a net electricity generation of 305.81 W, which represents a Fig. 9. Prevented emission values using PV panel with and without cool­
ing channel.
4.37% increase compared to 293 W with the PV with no cooling channel.

Table 13
Net Electricity Generation of all types of geometry.
Channel Without Without fins Without fins Without fins With 71 With 71 With 82 With 82 With With
Geometry cooling (Flat channel) (Flat channel) (Flat channel) flat fins (3 flat fins (5 flat fins (3 flat fins (5 Curved fins Curved fins
(5 m/s) (7 m/s) (10 m/s) m/s) m/s) m/s) m/s) (3 m/s) (5 m/s)

Electricity 293 302.95 307.01 310.40 300.18 307.12 302.05 308.40 306.09 311.01
Generation
(W)
Fan Power (W) – 2.99 7.41 18.43 0.67 2.49 0.72 2.59 6.34 28.58
Net Generation – 299.96 299.60 291.97 299.51 304.63 301.33 305.81 299.75 282.43
(W)

952
Z. Özcan et al. Solar Energy 230 (2021) 943–953

57.91 ◦ C. However, the maximum net electricity generation is found for Bayrak, Fatih, Oztop, Hakan F., Selimefendigil, Fatih, 2019. Effects of different fin
parameters on temperature and efficiency for cooling of photovoltaic panels under
the case with the cooling channel with 82 flat fins and 5 m/s air velocity,
natural convection. Solar Energy 188, 484–494.
when the fan power requirement is taken into account. The higher Duffie, J.A, & Beckman, W.A. 2013. Solar Engineering of Thermal Process. New Jersey.
pressure drop values created by the curved fins make this design less Egab, K., Okab, A., Dywan, H.S., Oudah, S.K., 2020. Enhancing a solar panel cooling
efficient. system using an air heat sink with different fin configurations. IOP Conference
Series: Materials Science and Engineering 671. Article ID 012133.
Lastly, the use of solar energy decreased CO2 emissions significantly. Elektrik Üretiminde Karbon Salınımı, 2020. Enerji Atlası. www.enerjiatlasi.com/haber
The PV with the cooling channel with 82 fins can prevent 15,451.2 kg /elektrik-uretiminde-karbon-salinimi (retrieved May 27, 2020 (in Turkish)).
coal based and 8,682 kg natural gas-based CO2 emissions, respectively. Erkan, O., Özkan, M., Arslan, O., 2018. Mini-Channel Cooling of a Photovoltaic Cell
Mini-Channel Cooling of a Photovoltaic Cell. International Journal of
Multidisciplinary Studies and Innovative Technologies 2, 34–38.
Declaration of Competing Interest Han, J., Lu, L., Yang, H., Cheng, Y., 2019. Thermal regulation of PV façade integrated
with thin-film solar cells through a naturally ventilated open air channel. Energy
Procedia 158, 1208–1214.
The authors declare that they have no known competing financial Irwan, Y.M., Leow, W.Z., Irwanto, M., Fareq, M., Amelia, A.R., Gomesh, N., Safwati, I.,
interests or personal relationships that could have appeared to influence 2015. Analysis air cooling mechanism for photovoltaic panel by solar simulator.
the work reported in this paper. International Journal of Electrical and Computer Engineering 5, 636–643.
Maleki, A., Haghighi, A., El Haj Assad, M., Mahariq, I., Alhuyi Nazari, M., 2020. A review
on the approaches employed for cooling PV cells. Solar Energy 209, 170–185.
References Mittelman, G., Alshare, A., Davidson, J.H., 2009. A model and heat transfer correlation
for rooftop integrated photovoltaics with a passive air-cooling channel. Solar Energy
Amber, K.P., Akram, W., Bashir, M.A., Khan, M.S., Kousar, A., 2021. Experimental 83, 1150–1160.
performance analysis of two different passive cooling techniques for solar Palumbo, A.M., 2013. Design and Analysis of Cooling Methods for Solar Panels. Thesis,
photovoltaic installations. Journal of Thermal Analysis and Calorimetry 143, Youngstown State University, MSc.
2355–2366. Panasonic Module HIT Datasheet, https://panasonic.net/lifesolutions/solar/download/
Arifin, Z., Tjahjana, D.D.D.P., Hadii, S., Rachmanto, R.A., Setyohandoko, G., Sutanto, B., pdf/N330_325_320SJ47Datasheet_190226.pdf. Retrieved September 2, 2021.
2020. Numerical and experimental investigation of air cooling for photovoltaic Piotrowski, L.J., Simões, M.G., Farret, F.A., 2020. Feasibility of water-cooled
panels using aluminum heat sinks. International Journal of Photoenergy 2020. photovoltaic panels under the efficiency and durability aspects. Solar Energy 207,
Article ID 1574274. 103–109.
Baloch, A. A. B., Bahaidarah, H. M. S., Gandhidasan, P. 2014. “Modeling and Wu, S.Y., Chen, C., Xiao, L., 2018. Heat transfer characteristics and performance
Comparative Analysis of Jet Impingement Cooling and Conventional Channel evaluation of water-cooled PV/T system with cooling channel above PV panel.
Cooling for Photovoltaic Strings”. IEEE 40th Photovoltaic Specialist Conference Renewable Energy 125, 936–946.
(PVSC), 8-13 June 2014, Denver, CO, USA. Wu, S.Y., Wang, T., Xiao, L., Shen, Z.G., 2019. Effect of cooling channel position on heat
Baloch, A.A.B., Bahaidarah, H.M.S., Gandhidasan, P., 2015a. “An Experimental Study of transfer characteristics and thermoelectric performance of air-cooled PV/T system.
the Effect of Converging Channel Heat Exchanger on PV System” IEEE 42nd Solar Energy 180, 489–500.
Photovoltaic Specialist Conference (PVSC), 14–19 June 2015. LA, USA, New Zheng, Z., Yan, W., Gong, H., Wang, Y., Shao, J., 2019. Quantifying the performance
Orleans. advantage of the novel passive air cooling system for PV array and system structure
Baloch, A.A.B., Bahaidarah, H.M.S., Gandhidasan, P., Sulaiman, F.A., 2015b. optimization. Applied Thermal Engineering 149, 899–908.
Experimental and numerical performance analysis of a converging channel heat
exchanger for PV cooling. Energy Conversion and Management 103, 14–27.

953