Energy Reports 4 (2018) 400–406

Contents lists available at ScienceDirect

Energy Reports
journal homepage: www.elsevier.com/locate/egyr

Research paper

Performance evaluation of a solar photovoltaic system

Wael Charfi a , Monia Chaabane b, *, Hatem Mhiri b , Philippe Bournot c
a
Institut supérieure des études technologiques de Tozeur, BP 150 Tozeur 2200, Tunisie
b
Unité de thermique et thermodynamique des procédés industriels, Ecole Nationale d’Ingénieurs de Monastir, route de Ouardanine, 5000 Monastir, Tunisie
c
IUSTI, UMR CNRS 6595, 5 Rue Enrico Fermi, Technopôle de Château- Gombert, 13013 Marseille, France

article info a b s t r a c t
Article history: The solar energy conversion into electricity is a very promising technique, knowing that the source is
Received 20 December 2017 free, clean and abundant in several countries. However, the effect of the solar cells temperature on the
Received in revised form 23 April 2018 photovoltaic panel performance and lifespan remains one of the major disadvantages of this technology.
Accepted 27 June 2018
In this work, we present an experimental study of a particular photovoltaic panel. It is self-cooled due
to its open design which facilitates natural ventilation helping to improve its performance mainly in hot
hours of the day and to avoid dust accumulation on its surface. This solar system is tested for two soil
Keywords:
Solar photovoltaic system natures, white and gray, and for two inclination angles, 0◦ and 30◦ . Results show that the photovoltaic
Soil nature and inclination angle effect panel performs better when it is inclined and placed on a white soil. A 3D CFD model describing the
Experimental characterization performance of this solar system is then developed and a good agreement between the numerical results
CFD validation and experimental data is found. Similarly, this CFD model was used to compare the thermal performance
Comparison of the CIGS PV and flat panel of this solar system to that of the flat PV system and to show that its lower temperature allows better
electrical production.
© 2018 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Gomez-Gila et al., 2012; Kelly and Gibson, 2009) and concentra-
tor based in stationary linear Fresnel lenses and secondary CPC
The increasing political and environmental problems related systems (Chemisana et al., 2009). Similarly, the effect of some
to the fossil fuel are the main drawbacks of this energy source parameters affecting the PV systems performance like the angle of
exploitation. A way to overcome these difficulties and to satisfy inclination (Wilson and Paul, 2011; Gajbert et al., 2007), the heat
the growing electricity demand around the world is the use of transfer mode (Kumar et al., 2012) and the Thomson effect (Ari
photovoltaic systems which allow converting solar energy into and Kribus, 2011) was numerically discussed using different simu-
electricity from sunlight. This clean technology inspired many lation tools in order to find the optimum values of these param-
researchers who studied the performance of different systems eters and to evaluate the optimum configuration of these solar
aiming to maximize the PV production with the least cost mod- systems. Skoplaki and Palyvos (2009) were interested to another
ifications. Ramadhan and Nasseb (2011) presented an economic
parameter which is the operating temperature of solar cells and
study of the viability of the PV solar energy implementation in the
modules, they presented different correlations concerning its ef-
state of Kuwait. Other technical and economical analyses of con-
fect on the electrical performance of photovoltaic installations.
centrating solar thermal systems, non tracked and tracked photo-
Other papers (Razykov et al., 2011; Parida et al., 2011; El Chaar
voltaic systems were performed by Quasching (2004) for different
et al., 2011; Si et al., 2017) were focused on the progress made
sites in Africa and Europe while Grasso et al. (2012) evaluated the
economical competitiveness of a stationary low concentration PV in the different photovoltaic technologies, and particularly on the
system and discussed the different parameters influencing its opti- effect of the solar cells materials. In other experimental investiga-
cal power ratio. Other research and experimental studies have fo- tions (Wu et al., 2012; Ryu et al., 2006), different configurations
cused on the comparison of concentrating and non-concentrating of solar photovoltaic concentrating systems using Fresnel lenses
photovoltaic systems (Mallick et al., 2006, 2007; Matsushima et al., were proposed and tested under different operating parameters
2003), fixed and different tracking systems (Koussa et al., 2011; such as the solar radiation intensity, the ambient air temperature
and the natural and forced convection. Another new technology, in
which PV cells of high flux levels were used for different designs of
* Corresponding author.
E-mail addresses: [email protected] (W. Charfi), [email protected] solar cell concentrators, was proposed by Feuermann and Gordon
(M. Chaabane), [email protected] (H. Mhiri). (2001). These authors proposed a new approach for concentrating

https://doi.org/10.1016/j.egyr.2018.06.004
2352-4847/© 2018 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 W. Charfi et al. / Energy Reports 4 (2018) 400–406 401

Table 1
Nomenclature Photovoltaic panel characteristics.
Parameter Value
Ac Collector area (m2 ) Maximum power at STC (P max) 200 W
G Solar irradiance (W m−2 ) Optimum operating voltage (U mp) 78.3 V
I Current (A) Optimum operating current (I mp) 2.55 A
P Power (W) Short-circuit current (I sc) 2.78 A
Open circuit voltage (U oc) 99.7 V
T Temperature (K) Electrical efficiency 10.18%
U Voltage (V) Dimensions of PV panel 1820∗1080∗50 mm
V Velocity (m s−1 ) Temperature coefficient of power −0.38%/K
η Efficiency Temperature coefficient of voltage −0.289 V/K

Subscripts
panel for a gray soil. All the measurements were continuously
a Ambient
monitored each 1/2 hour, from 7 am to 6 pm.
el Electrical
The characteristics and dimensions of the photovoltaic panel
max Maximum
used in this experimental study are specified in Table 1:
mp Optimum operating
oc Open-circuit
2.2. Analyzed parameters and measuring instruments
sc Short-circuit
w Wind
For this PV system electrical performance evaluation, the cur-
rent I and voltage U were continuously measured. The meteoro-
logical parameters defined by the ambient temperature Ta, the
PV systems that can easily attain the maximum flux level commen- wind speed Vw and the incoming solar irradiance G were also
surate with solar cell technology. Chaabane et al. (2013) are among experimentally determined using specific data acquisition devices.
the authors who are interested in the exploitation of solar energy in The characteristics of the measuring equipments used in this ex-
Tunisia using photovoltaic systems. They proposed an experimen- perimental study are listed in Table 2.
tal study in which they discussed the effect of the cooling system
that they designed on the global performance of their photovoltaic 2.3. Experimental results
device.
In this paper, we present an experimental study of a particular The ambient temperature, incident solar irradiance and wind
photovoltaic panel. The open design of this solar system, which velocity were experimentally measured and respectively pre-
allows solar cells cooling by natural ventilation, is its main charac- sented in Figs. 2 and 3. During these three days, the climatic
teristic. Measurements are taken for two different soil natures and conditions were characterized by clear sky conditions. The solar
radiation intensity and therefore the ambient air temperature were
inclination angles. Computational fluid dynamics (CFD) are also
slightly higher for the gray soil test days, they achieved respec-
used to model the performance of this photovoltaic system and to
tively 861 W m−2 and 30.5 ◦ C. Concerning the wind velocity, its
compare it to that of a Flat one.
maximal value has not exceed 5 m/s, so the climate corresponding
to our experiments was stable and without sand storms which can
2. Experimental study cause disturbances during measurements.
The current I and the voltage U delivered by the PV panel were
2.1. Experimental setup measured, the electrical power generated by these PV systems,
which is defined as their product, was calculated and its temporal
In this study, the tested PV panel consists of 40 cylindrical solar evolution is presented in Fig. 4. The analysis of this figure shows
cells made of CIGS (Fig. 1). Due to cylindrical shape of the tube that the electrical power increases during the day up to noon,
and its concentrating effect, the PV panel is collecting light over then decreases with the solar radiation intensity decrease. It is also
360◦ and thus operating with direct, diffuse and reflected solar ra- noted that the produced electrical power is higher for the inclined
diation. This experimental investigation has been conducted under PV paned due to the difference in the incoming solar radiation
a Tunisian Saharan climate, in the city of Tozeur. The experiments relatively to the horizontal panel. Concerning the effect of the soil
were undertaken during three consecutive spring days, 19, 20 and nature, it can be seen that the highest production corresponds to
21 April. These tests correspond respectively to an horizontal panel the white soil (181 W), and this is due to its reflected part of the
for a white soil, an horizontal panel for a gray soil and a sloped solar radiation which is negligible in the case of the gray soil.

Fig. 1. Photographic picture of the horizontal (a) and titled (b) PV panel.
402 W. Charfi et al. / Energy Reports 4 (2018) 400–406

Table 2
Equipments used for the experimental measurements.
Parameter Unit Measuring instrument Model
◦
temperature T C Infrared thermometer with K thermocouple Fluke 63
Wind speed Vw m/s Handheld Rotating Vane Anemometer HHF141
Solar irradiance G W/m2 Solarimeter CR 100
Current I A Compact digital multimeter Amprobe AM-240
Voltage V V Compact digital multimeter Amprobe AM-240

Fig. 2. Variation of the solar irradiance and ambient temperature with time.

Fig. 3. Daily temporal evolution of the wind velocity.

Regarding the energy conversion efficiency, it was evaluated ac- temperature was measured and its average variation is shown in
cording to the following relation: ηel = GAP ; where Ac is the panel Fig. 6. From this figure, it can be seen that the PV tubes temperature
c

surface, P is its electrical production and G is the incoming solar increases at the midday, achieves its maximum value which is of
irradiation. The temporal evolution of this solar system electrical 54 ◦ C for the case of a sloped panel on a gray soil, and decreases
efficiency is described in Fig. 5. It can be seen that the electrical next with the decrease of the solar radiation. Similarly, by ana-
efficiency is decreasing up to noon due to the negative effect of lyzing the measured values of the temperature, we can notice the
the solar cells temperature during this period of the day and it advantage of this open design of the PV panel which allows the
is increasing for the rest of the day. It is also noticed that the PV solar cells cooling by natural ventilation.
panel electrical efficiency is higher for a white soil due to its higher In this study, the experimental performance analysis of a pho-
electrical production. tovoltaic system was presented. It has been observed that the
In addition to these electrical parameters, a thermal character- electrical production and efficiency of the solar panel are much
ization of this solar system is presented. Indeed, the solar tubes higher when it is tilted and above a white soil.
 W. Charfi et al. / Energy Reports 4 (2018) 400–406 403

Fig. 4. Electrical power of the PV system.

Fig. 5. Electrical efficiency of the PV system.

3. Numerical results the Boussinesq approximation expressed as follows: (ρ − ρ0 ) =

−ρ0 β (T − T0 ).
3.1. Geometry description and meshing The turbulence model used to simulate the flow in this solar
system is the standard K-ε turbulence model which presents the
A 3D CFD models interpreting the photovoltaic system is devel- advantage of its accuracy in addition to its fast calculation time.
oped. The same dimensions, physical characteristics and specifica- Concerning the radiative term, we considered a non-gray model,
tions as the experimentally tested system are considered. A grid the Discrete Ordinates DO which is based on the radiative transfer
dependency test was carried out and we chosen a mesh size of equation resolution. High resolution schemes are selected for the
pressure, velocity and temperature terms discretization and a con-
1067040 cells. The generated mesh, which consists of hexahedral
vergence criteria of 10−4 is used for all the equations resolutions,
cells clamped at the tubes and looser in the rest of the field, is
except for the energy residual where a value of 10−6 is given.
shown in Fig. 7.
3.3. Simulation results
3.2. Numerical procedure
In order to prove the credibility of the developed CFD model,
Numerical simulations are carried out under the same climatic the thermal behavior of the PV tubes in this solar system is char-
and operating conditions of the experimental study. The simplify- acterized and the PV panel electrical performance is then evalu-
ing assumptions concern mainly the fluid properties and partic- ated. We chosen to present results corresponding to an horizontal
ularly its density which changes with temperature according to panel placed on a white soil. Numerical results of the solar tubes
404 W. Charfi et al. / Energy Reports 4 (2018) 400–406

Fig. 6. Measured solar cells temperature in the PV system.

9. According to these figures, the maximum difference between

the two results does not exceed 10%, which shows a satisfactory
agreement proving the validity of the CFD model and justifying its
used for further improvements in this PV panel conception.
To confirm the previous findings, and based on our validated
CFD model, a flat PV panel having the same peak power than
the studied CIGS panel was numerically modeled and results of
these two systems solar cells temperature and electrical power are
respectively compared in Figs. 10 and 11. The analysis of these
figures show the advantage of the naturally ventilated system
which is the CIGS one relatively to the flat PV panel, and that its
lower temperature results in higher electrical power.
In Fig. 12, we present the temperature distribution for a cross
section of these two photovoltaic panels, for a solar irradiation
of 800 W m−2 . These temperature contours show that the tubes
Fig. 7. Geometry meshing. spacing in the CIGS panel allows the air circulation in these zones
and thus their cooling by natural ventilation, which results in
the PV panel performance improvement relatively to the flat one.
average temperature and electrical power are so compared to Similarly, it is noted that this arrangement of the tubes gives an
the experimental data and respectively presented in Figs. 8 and uniform temperature profile, without any overheating zone.

Fig. 8. Comparison of the solar cells temperature numerical and experimental results.
 W. Charfi et al. / Energy Reports 4 (2018) 400–406 405

Fig. 9. Comparison of the PV system electrical power numerical and experimental results.

Fig. 10. Comparison of the cells temperature in the CIGS PV panel to that in the flat one.

Fig. 11. Comparison of the electrical power in the CIGS PV panel to that in the flat one.

4. Conclusion showed the advantage of the inclined system which collects higher
intensity of the solar radiation, resulting in higher electrical pro-
The electrical performance of a particular photovoltaic system duction of the PV panel. Concerning the effect of the nature of the
was experimentally evaluated under a spring day in the Tunisian soil on which the panel is placed, a higher electrical production
Saharan city Tozeur. This solar system consists of 40 spaced solar was obtained for the white soil which, unlike the gray one, allows
cells made of CIGS and of cylindrical shapes. Due to its open the panel to operate with an additional part of the solar radiation
design, this solar system allows the solar cells cooling by natural which is that reflected by this soil. A 3D CFD model interpreting
ventilation. The measurements were taken for an horizontal and this solar system’s operating was then developed and simulation
inclined panel, and for a gray and white soil. Experimental results results showed a good agreement with the experimental data.
406 W. Charfi et al. / Energy Reports 4 (2018) 400–406

Fig. 12. Temperature contours for a cross section of the CIGS (a) and the flat (b) PV panels.

This solar system’s performance was then compared to that of the Kumar, Natarajan Sendhil, Matty, Katz, Rita, Ebner, Simon, Weingaertner, Ortrun,
flat PV panel and numerical results showed the effect of the CIGS Aßländer, Alex, Cole, Roland, Wertz, Tim, Giesen, Kumar, Mallick Tapas, 2012.
Experimental validation of heat transfer model for concentrating photovoltaic
system ventilation and its lower temperature in its performance
system. Appl. Therm. Eng. 34, 175–182.
improvement. The developed and validated CFD model can be later Mallick, Tapas K., Eames, Philip C., Norton, Brian, 2006. Non-concentrating and
used for the CIGS PV panel’s performance optimization, especially asymmetric parabolic concentrating building façade integrated photovoltaic:
for the evaluation of the optimum tubes spacing in order to reduce An experimental validation. Sol. Energy 80, 834–849.
the PV panel surface. Mallick, Tapas K., Eames, Philip C., Norton, Brian, 2007. Power lodes in an asymmet-
ric compound parabolic photovoltaic concentrator. Sol. Energy Mater. Sol. Cells
91, 1137–1146.
References Matsushima, Toshio, Setaka, Tatsuyuki, Muroyama, Seiichi, 2003. Concentrating
solar module with horizontal reflectors. Sol. Energy Mater. Sol. Cells 75, 306–
Ari, Nimrod, Kribus, Abraham, 2011. Impact of the Thomson effect on concentrating 312.
photovoltaic cells. Sol. Energy Mater. Sol. Cells 94, 1421–1425. Parida, Bhubaneswari, Iniyan, S., Goic, Ranko, 2011. A review of solar photovoltaic
Chaabane, M., Charfi, W., Mhiri, H., Bournot, P., 2013. Performance evaluation of technologies. Renewable Sustainable Energy Rev. 15, 1625–1636.
concentration photovoltaic and photovoltaic/thermal systems. Sol. Energy 98, Quasching, Volker, 2004. Technical and economical system comparison of photo-
315–321. voltaic and concentrating solar thermal power systems depending on annual
Chemisana, Daniel, Ibanez, Manuel, Barrau, Jerome, 2009. Comparison of Fresnel irradiation. Sol. Energy 77, 171–178.
concentrators for building integrated photovoltaics. Energy Convers. Manage. Ramadhan, Mohammad, Nasseb, Adel, 2011. The cost benefits analysis of imple-
50, 1079–1084. menting photovoltaic solar system in the state of Kuwait. Renew. Energy 36,
El Chaar, L., Iamont, L.A., El Zein, N., 2011. Review of photovoltaic technologies. 1272–1276.
Renewable Sustainable Energy Rev. 15, 2165–2175. Razykov, T.M., Ferekides, C.S., Morel, D., Stefanakos, E., Ullal, H.S., Upadhyaya, H.M.,
Feuermann, Daniel, Gordon, Jeffrey M., 2001. High-concentration photovoltaic de- 2011. Solar photovoltaic electricity: current status and future prospects. Sol.
signs based on miniature parabolic dishes. Sol. Energy 70, 423–430. Energy 85, 1580–1608.
Gajbert, Helena, Hall, Maria, Karlsson, Bjorn, 2007. Optimization of reflector and Ryu, Kwangsun, Rhee, Jin-Geun, Park, Kang-Min, Kim, Jeong, 2006. Concept and
module geometries for stationary, low-concentrating, façade-integrate photo- design of modular Fresnel lenses for concentration solar PV system. Sol. Energy
voltaic systems. Sol. Energy Mater. Sol. Cells 91, 1788–1799. 80, 1580–1587.
Gomez-Gila, Francisco Javier, Wang, Xiaoting, Barnett, Allen, 2012. Energy pro- Si, Fai Tong, Isabella, Olindo, Zeman, Miro, 2017. Thin-film amorphous silicon
duction of photovoltaic systems: fixed, tracking, and concentrating. Renewable germanium solar cells with p- and n-type hydrogenated silicon oxide layers.
Sustainable Energy Rev. 16, 306–313. Sol. Energy Mater. Sol. Cells 163, 430–440.
Grasso, Giorgio, Righetti, Aldo, Ubaldi, Maria Chiara, Morichetti, Francesco, Skoplaki, E., Palyvos, J.A., 2009. On the temperature dependence of photovoltaic
Pietralunga, Silvia Maria, 2012. Competitiveness of stationary planar low con- module electrical performance: A review of efficiency/power correlations. Sol.
centration photovoltaic modules using silicon cells. Sol. Energy 86, 1725–1732. Energy 83, 614–624.
Kelly, Nelson A., Gibson, Thomas L., 2009. Improved photovoltaic energy output for Wilson, Michael J., Paul, Manosh C., 2011. Effect of mounting geometry on convec-
cloudy conditions with a solar tracking system. Sol. Energy 83, 2092–2102. tion under a photovoltaic panel and the corresponding efficiency using CFD. Sol.
Koussa, M., Cheknane, A., Hadji, S., Haddadi, M., Noureddine, S., 2011. Measured and Energy 85, 2540–2550.
modelled improvement in solar energy yield from plate photovoltaic systems Wu, Yupeng, Eames, Philip, Mallick, Tapas, Sabry, Mohamed, 2012. Experimental
utilizing different tracking systems and under a range of environmental condi- characterization of a Fresnel lens photovoltaic concentrating system. Sol. En-
tions. Appl. Energy 88, 1756–1771. ergy 86, 430–440.