2014 IEEE International Conference Power & Energy (PECON)
Impact of Dust on Solar Energy Generation based on Actual Performance
Mohammadreza Maghami1,2,*, Hashim Hizam1,2, Chandima Gomes1,
1. Department of Electrical and Electronic Engineering, Universiti Putra Malaysia, Serdang, 43400, Selangor, Malaysia
2. Center of Advanced Power and Energy Research, Universiti Putra Malaysia, 43400, Selangor, Malaysia
*Corresponding author: Mohammadreza Maghami Fax: +60389466312; Tel:+60389576967; HP: 006012345-3694
Abstract Energy and Environment are two important issues
in this decade. Using the clean energy such as solar energy
generation is growing around word, but with low efficiency.
Many parameters from the environment affect the solar
photovoltaic panel such as shadow, air pollution and dust. In
some experimental setup, there is a small layer that accumulates
on top of the photovoltaic (PV). In order to evaluate the effect of
dust on Photovoltaic, using two Fix Flat Photovoltaic (FFP) was
installed at the Universiti Putra Malaysia. One array is the
Clean array that was cleaned regularly and other is the
Dusty array that was not cleaned during this research. Data
was collected at interval each of 30mins from 1st April to 5th
December 2013 for both arrays. Power output and Energy yield
for both PV arrays are considered. Findings from this research
shows that overly power and energy decrease due to the dust,
which accumulates on the surface. The contribution this work
with other research was done is we are evaluate the power loss in
tropical climate which force with many rainy days.
Keywordscomponent; Fix flat array; photovoltaic generation;
PV generation; p (key words)
I. INTRODUCTION
Nowadays energy and environmental issues coexist. The
development of new energy sources is forced by the crisis of
energy and environment. As a source of clean energy, solar
energy should be more effectively and rationally used. In
recent years, solar PV technology has made great progress.
However, most part of the research at home and abroad
focused mainly on theoretical research, design and
construction, such as design of photovoltaic panels towards
the inclination, impact of dust shelter on the performance and
temperature of the battery and so on. In the domestic
"Technical Specifications about Civil Application of Solar
Photovoltaic Systems", there are many relevant provisions
about construction site, geography, climate, solar resources,
shelter, impact of temperature on the power generation and so
on. But research on the impact of dust, air cleanliness, rainfall
and other factors on PV projects are also very little.
In some PV engineering practices, there is a thick layer of
dust accumulation on the surface of PV panels [1-7]. In a test
on a 300kWp solar PV demonstration projects in Shenzhen,
Guangdong province, China, we discovered that under the
same solar radiation intensity and outside temperature, the
efficiency of PV panels power generation with surface soling
978-1-4799-7297-5/14/$31.00 2014 IEEE
is 7.8%, while with the surface cleaned is 9.3% [13], and 1.5%
difference between the two efficiencies. Thus, there is more
significant impact on the efficiency of PV power generation
with dust accumulation on the surface. Therefore, it is
necessary to study the fouling effect on PV power generation
projects.
Hottel and Woertz [8] first studied the effect of dust on
solar panel performance by investigating the dust
accumulation on such panels. A three-month test was
performed in an industrial area near a four-track railroad 90 m
away from Boston, Massachusetts. They found an average of
1% loss of incident solar radiation was caused by dust that
accumulated on the surface of the solar panel with a tilt angle
of 30. The maximum degradation reported during the test
period was 4.7%. The researchers deduced a correction factor,
defined as the ratio of the transmittance from an unclean or
exposed glass plate to a clean one, of 0.99, with a 45 tilt
angle; this value was adopted and accepted in the design of
flat-plate collectors until 1970. In 2001,
Kimber et al. investigated the effects of soiling on large
grid-connected PV panels in California, United States. The
objective of the study was to provide a better model to
accurately predict soiling losses throughout the year rather
than assuming a constant annual value. Another objective was
to characterize the effect of soiling on PV systems for an
entire region rather than for a specific location. For this study,
a linear regression model was used to characterize soiling
losses over the dry season. Data from 250 sites were gathered
and later filtered to 46 system data sets after excluding sites
with non-linear soiling behavior and sites with significant
rainfall. Soiling rates and rainfall data were later implemented
in a simulation program and energy yield predictions were
compared with a model using only a constant annual soiling
rate. The authors concluded that results from the study
indicated an average daily efficiency decline of 0.2% in days
without rainfall in dry climates. This translated to annual
losses due to soiling ranging from 1.5% to 6.2% depending on
the location of the PV plant. Furthermore, the authors
concluded that the monthly yield simulation with the soiling
data provided better yield predictions than the simulation
model with an assumed constant annual soiling loss. There
were some very interesting observations from the study. For
instance, some systems exhibited a sharp loss in system
performance even after some light rain while others improved.
The authors mentioned variability in the amount of rainfall
388
 2014 IEEE International Conference Power & Energy (PECON)

necessary to clean some PV systems. For eexample, some 100 cm2 (0.01 m2) surface area would
w produce 2.0 W.
systems werent fully cleaned with a 2mm rain fall and
continued to experience performance loss untiil a rainfall of
20mm managed to clean the system.
Soleimani et al. [9] studied the effect of aiir pollution on B. Array Discription
PV performance. The influence of air pollution iis considerable The array is made with 12 modu ules and each module is made
for a large city such as Tehran. The researchers found that the from monocrystalline silicon CSUN modules .Table 1 shows
power output of PV decreased by more than 600% because of
the data sheet of a Photovoltaic module. These Photovoltaic
air pollution that covered the surface of the PVV panel which
obstructed the sunlight. array modules are made of 108 monocrystalline
m silicon (9 pcs
12 pcs) and 125 mm 62.5 5 mm. The standard testing
In this work, two photovoltaic arrays are coompared which
one array cleaned weekly and the other one ddo not cleaned condition (STC) is IEE-61215 1000 W/M, 25C. The two
during this research in Universiti Putar Malysiaa. The data was solar arrays employed in thiss study have been ground
collected to evolutes the impact of dust includinng total power mounted. These two PV arrays consist of 2*1 kWp units of
generation and energy generation among mouunt of the two fixed flat photovoltaic (FFP) array
ys, which are shown in figure
arrays. 1. According to Table 1, the PV V arrays are made up of CEEG
95 W monocrystalline PV modu ules. In addition, this requires
the use of PV module parameterss, such as short circuit current
II. MATERIALS AND METHODOLOGY
at STC, open circuit voltage att STC, voltage, current and
power temperature coefficients. The
T PV plates are connected
A. Calculation of Electrical Energy to the distribution grid.
ments were of
1. In this study, only two electrical measurem
interest:

Power (measured in Watts or W) is the instanttaneous rate of

using or producing electrical energy.
Energy (in kilowatt-hours, or kWh) is the tootal amount of
electricity used/produced over a period of time.
power is the rate at which energy is transferrred from one
point to another or from one form to another. Thhe formula that
relates energy and power is:

Power = voltage x Current (1)

Energy = Power x Time. (22)

The unit of energy is the joule, the unit of pow

wer is the watt,
and the unit of time is the second.
Fig 1 UPM
M site
If the power in watts of an appliance and tthe number of
seconds it uses is known we can calculate the nuumber of joules Table1 Characteristicss of CSUN module
of electrical energy that has been converted.
Characteristics Pmax P Nominal
By convention, solar cell efficiencies are m measured under Maximum power(w) 95 92
standard test conditions (STC) unless stated ootherwise. STC Voltage (v) 18.3 17.9
specifies a temperature of 25 C and an irradiance of Current (I) 5.21 5
1000 W/m2 with an air mass 1.5 (AM1.5) speectrums. These Open circuit voltage(V) 22.5 20
conditions correspond to a clear day with sunnlight incident Short circuit current(I) 5.56 5.4
upon a sun-facing 37-tilted surface with the sunn at an angle of Weight (KG) 8 8
41.81 above the horizon. This represents solarr noon near the
spring and autumn equinoxes in the continentall United States
with surface of the cell aimed directly at the suun. Under these C. Data Collection
test conditions a solar cell of 20% efficiency with a To compare energy and pow wer output generation among
the months, two types of data weere collected from the arrays,
current and voltage. The data ou
utput was connected to a data
Identify applicable sponsor/s here. If no sponsors, deelete this text box logger and was monitored onlinee every minute. In this study,
(sponsors). data was selected every 30 min n to obtain samples from 1st
th
April to 5 December 2013.

389
 2014 IEEE International Conference Power & Energy (PECON)
D. Location
Four panels of 1 kW-rated PV arrays, namely, FFP and TFP,
were installed in Serdang, Malaysia, with ccoordinates of
259'20"N: 10143'30"E and under tropical--based ground
conditions. Figure 8 shows the geographical loocations of the
two PV plates built at Universiti Putra Malaysiaa. Each array is
south exposed and the tilt angle is 15 (FFT).. As shown in
Figure 8, two highways with vegetation are llocated 500 m
away, and a new Graduate School of Manageement building
was constructed near the site.
Fig 1. Environment around
III. RESULT AND DISCUSSION
A. Power output generation
From the Table 2, it can be seen that the totaal power output
generation from the clean array is significantly more than the
dusty panel. The power generated from the clean array is
3533289.77 with standard deviation 275.799, which was
collected in 8 months (from 1st April 2013 to 5th December
2013), For the dusty array the total generation is 3440211.73
with standard deviation 270.72. It can be concluuded from this
table that the overly Power generation from thee clean panel is
2.71 % more than the dusty panel. The nnominal power
generation for each PV array (1 kW) is estimatted to be 1 kW
based on module performance under the STC.
TABLE 2. OVERALL POWER OUTPUT FOR BOT
TH ARRAYS
Panel N
Sum S.D
Clean
9544 3533289.77 275.79
Dusty 9544 3440211.73 270.72
N: Number of sample, SD: Standard Deviaation
Total monthly generation was considered inn this research
during the 8 months. In this period two scennario effect on
solar generation can be seen in Figure 2. In JJune 2013, the
amount of generation between the clean array aand dusty array
390
was lower than other months. In June 2013, the total average
power output was 352 (w) for thee clean array but for the dusty
array during the same period wass 326(w). It can be seen that,
in this month the average mon nthly power output from the
clean array was 26 (w) more than t the dusty array. In the
month of July 2013 the total daily average power output
monthly for the clean array and dusty array was 347.55 (w)
n November 2013, the power
and 343.21 (w), respectively. In
output decreased. The total averaage monthly power output for
the clean array and dusty array was
w lower than June and July
2013, by 332(w) and 318 (w), reespectively. The highest daily
power output for May and au ugust 2013, is 406 (w) and
397(w).
Fig. 2 Total power outtput among month
B. Energy generation from both arrray
Total energy production was con nsidered for both clean array
and dusty array. Figure three showss the total energy yield
generation from April 2013 until November 2013 for the both
arrays. From this figure, it can beb seen that high-energy is
generated in October 2103(clean array with 168.92 KWh and
dusty array with 165.82), Followed by September with 119.75
KWh and 117.63 KWh for dusty array. Behind these two
mount July and August are the highest in this research. In
April, May, June and November has lowest energy production
in this research. In May and April, the system became faculty
for more than ten days; this is lead to reduce energy in this two
month. Southeast Asia haze pollution 2013. Is cause for
decrease energy output in June 20 013 and increase cloudy days
which is reduce irradiation on th he top of PV is the important
parameter which is reduce energ gy in November 2013. In the
Table 3, shows the Total energy y generation, Average Energy
and maximum energy generation between the Dusty array and
clean array. One of the interestin ng part in this table, the last
Row of this table, which is showss total energy during the eight
month between clean and dusty array.
a Total energy generated
from clean array is 854.41 KWh and for dusty array is 842.80
KWh. Dust was reduce energy ou utput around 12 KWh. In the
next column daily energy generaation shows that again clean
array 3.95 KWh each day but dussty array around 3.90 KWh.
 2014 IEEE International Conference Power & Energy (PECON)

Maximum Energy yield for both arrays is in October with 7.12 10th with 3.17 KWh for Dusty array. Maximum energy in this
kWh, followed by August with 6.68 KWh and 6.61 KWh for month generated at 1st of April with 6.03 From clean array.
the clean array and Dusty array. Maximum Energy in July was
less than other months. Energy in May
In May 2013, Data loss due to system faulty for 12 days from
Average daily energy generation was 5.45 KWh for the Clean 19th May to 1st June. According to the Figure 5, the total
array and the Dusty array with 5.35 KWh had the highest energy in this month was the same as April reduced. The total
daily average energy in October 2013, followed by May with energy yield for the clean and dusty array was 74.35KWh and
4.33 KWh and 4.29 KWh for the Clean array and Dusty array, 73.55KWh, respectively. The minimum energy generation in
respectively. April and September with 4.11 KWh and 3.99 14th of May was produce with 3.01 KWh from Dusty array.
KWh respectively. Both arrays had low energy yield with Maximum daily energy yield in 10th May with 5.97 KWh from
2.89 KWh for the clean array and 2.86 KWh for the Dusty clean array and daily average energy generation 4.33 KWh
array due to more cloudy days in this month. The daily which is highest among other month.
average energy for all the month is shown, with the clean
array 3.95KWh and 3.90 KWh for the Dusty array.
Energy in June
Table 3 Energy yield (kWh) in 2013
Southeast Asia haze pollution is the issues, which is decrease
N Total Total energy yield in June 2013, (start from 10th June Until 20th
Energy energy Ave Ave Max Max June). Figure 6, shows that daily energy generation in June
Month Clean Dusty Clean Dust Clean Dusty 2013, the maximum daily energy yield generate at 23rd June
Apr 18 72.12 71.50 4.11 4.08 6.03 5.98 with 6.11 KWh from clean array. Minimum energy generation
May 15 74.35 73.55 4.33 4.29 5.97 5.92
at 20th June with 3.22 KWh and 3.12 KWh for the clean and
Jun 27 104.23 102.59 3.60 3.54 6.11 6.07
dusty array, respectively. The Daily average generation in this
Jul 30 117.84 116.44 3.80 3.76 5.56 5.50
month is 3.60 KWh for clean array and 3.54 for dusty array.
Aug 24 106.66 105.41 3.44 3.40 6.68 6.61
27
Total energy produced from the clean array was 104 KWh and
Sep 119.75 117.63 3.99 3.92 6.32 6.26
Oct 30 168.92 165.82 5.45 5.35 7.12 7.05
102 KWh for the Dusty array. Number of date, which data
Now 25 80.93 80.19 2.89 2.86 5.85 5.78 collected from the both PV arrays, is 27 days.
Total 203 854.41 842.80 3.95 3.90 7.12 7.05
N: Number of day data collected each month
Energy in July
In July 2013, the total energy produced from the clean array
was 117 KWh and 116 KWh for the dusty array, the minimum
energy was generated on 26th July with 2.20 KWh and
maximum generation with 5.56 KWh at 7th of July. In this
month the average daily energy was 3.80 KWh and 3.76 KWh
for the clean and dusty array, respectively. Data collected for
this month was for 30 days. Figure 7, show the daily energy
yield in July 2013.

Fig. 3 Total energy Yield each month in 2013
Energy in April
Figure 4 shows daily energy yield in April 2013, for both
arrays. The total energy produced in this month was 72.12
KWh for the clean array and 71.50 KWh for the dusty array.
In This month both arrays became faulty more that 15 days,
for this reason the total generation was less than other months.
However, daily average energy was 4.11 KWh, which is more
than other months. The lowest energy generation was on the
Energy in August
According to the figure 8, in August 2013, the both arrays
became faulty for 6 days and data was collected for 24 days;
the total energy yield in this month for the clean and dusty
array was 106 KWh and 105 KWh, respectively. Maximum
energy was 6.68 KWh on 14th August and the minimum
energy produced was 1.82 KWh on 19th August. Average daily
energy yield for Dust array is around 3.40 KWh and 3.44
KWh for clean array.
Energy in September
Total data collected in September for 27 days, the report of
energy yield in September brought the figure to 9. Total
energy generation for the clean array was more than the dusty

391
 2014 IEEE International Conference Power & Energy (PECON)

array with 119 KWh and 117 KWh. The maxximum energy Fig 5: May 2013, Total Energy Yieeld for Clean and Dusty Array is
produced on the 21st was 6.32KWh and minnimum energy 74.35 kWh and 73.55 KWh, respectiv
vely
generated from the clean array on 26th was 1.699 KWh. Daily
average energy for this month was 3.99 KWh aand 3.92 KWh
for the clean and dusty array, respectively.
Daily Energy Yeild (KW
Wh) in June 2013
Energy in October 9
The highest amount of energy generation was pproduce in this Clean Array Dusty Array
month with 168.92 KWh and 165.82 KWh for Clean and 7

Power (w)
dusty array, respectively. Maximum energy yieeld in October 5
produce in 7th October, with 7.12 from cleean array and
minimum energy generation comes from 19th October with 3
3.20 KWh form dusty array. The daily avverage energy 1
generation for clean array is 5.45 KWh and 55.35 KWh for
1 3 5 7 9 11 13
3 15 17 19 21 23 25 27
dusty array, which is highest energy yield amonng other month. D
Day
In this month, Number of rainy days more thaan other month
and both PV get more irradiation from the sun. Fig 6: June 2013, Total Energy Yieeld for Clean and Dusty Array is
104.22 kWh and 102.58 KWh, respecctively.
Energy in November
According to the figure 11, in November 2013, tthe both arrays
became faulty for 5 days and data was collecteed for 25 days;
the total energy yield in this month for the cllean and dusty
Daily Energy Yeild
d (KWh) in July
array was 80.93 KWh and 80.19 KWh, respectively.
Maximum energy was 5.85 KWh on 1st Novem mber. Average 7
Clean Array Dusty Array
daily energy yield for Dust array is around 2.86 KWh and 2.89
Power (w)

KWh for clean array. In this month, number off cloudy day is 5
more than others month.
3

3
Daily Energy Yield (KWh) in April 2013
1
8
Clean Array Dusty Array 1 3 5 7 9 11 13 15
1 17 19 21 23 25 27 29 31
Day
Power (w)

4 Fig 7: July 2013, Total Energy Yieeld for Clean and Dusty Array is
117.83 kWh and 116.43 KWh, respecctively.
2
1 3 5 7 9 11 13 15 17 19 21 23 25 227 29
Day
Daily Energy Yield (KW
Wh) in August 2013

Fig 4: April 2013, Total Energy Yield for Clean andd Dusty Array is 7
Clean Array Dusty Array
72.12 kWh and 71.50 KWh, respectively
Power (w)

3
Daily Energy Yield (KWh) in May 20113
8
Clean array Dusty array 1
Power (w)

6 1 3 5 7 9 11 13 1y5 17 19 21 23 25 27 29 31
Day
4

2 Fig 8: August 2013, Total Energy Yield

Y for Clean and Dusty Array
is 106.66 kWh and 105.41 KWh, resp
pectively.
1 3 5 7 9 11 13 15 17 19 21 23 225 27 29
Day

392
 2014 IEEE International Conference Power & Energy (PECON)

(clean array and dusty array). Power output and energy

er
Daily Energy Yield (KWh) in Septembe
generation were two parameterss, which were considered in
7 this work. The result showed that the total energy generated
Clean Array Dusty Array
from the clean array was 854 4.41KWh more than energy
Power (w)

5 generated from the dusty array y with 842.80 KWh. Daily

energy generation showed that thet average energy produced
3 from the clean and dusty array was
w 3.95 KWh and 3.90 KWh,
respectively. From this researchh, it can be said that dust one
1 of the important parameters that reduced
r energy yield.
1 3 5 7 9 11 13 15 17 19 21 223 25 27 29
Day
V. ACKNOWLEDGMENT
Fig 9, September 2013 Total Energy Yield ffor Clean and
Dusty Array is 119.75 kWh and 117.62 KWh, reespectively. The authors are grateful for th
he assistance and financial
support provided by the Univeersiti Putra Malaysia
Daily Energy Yield (KWh) in October 22013
VI. REFER
RENCES
8
Clean Array Dusty A
Array
Power (w)

6 [1] A. Tiwari and M. Sodhaa, "Performance evaluation of

solar PV/T system: an a experimental validation,"
4 Solar Energy, vol. 80, pp p. 751-759, 2006.
[2] M. Mani and R. Pillaii, "Impact of dust on solar
2 photovoltaic (PV) perrformance: research status,
challenges and recomm mendations," Renewable and
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 Sustainable Energy Reviews, vol. 14, pp. 3124-3131,
Day
2010.
[3] M. S. El-Shobokshy an nd F. M. Hussein, "Effect of
Fig 10, October 2013, Total Energy Yield for C
Clean and Dusty dust with different physical
p properties on the
Array is 168.92 kWh and 165.82 KWh, respectively. performance of photov voltaic cells," Solar Energy,
vol. 51, pp. 505-511, 199 93.
Daily Energy Yeild (KWh) in Nowember [4] S. Mekhilef, R. Saidurr, and M. Kamalisarvestani,
"Effect of dust, hum midity and air velocity on
7 efficiency of photovoltaic cells," Renewable and
Clean array Dusty array Sustainable Energy Reviews, vol. 16, pp. 2920-2925,
2012.
Power (w)

5
[5] H. Jiang, L. Lu, an nd K. Sun, "Experimental
investigation of the impact of airborne dust
3 deposition on the perforrmance of solar photovoltaic
(PV) modules," Atmosp pheric Environment, vol. 45,
1 pp. 4299-4304, 2011.
1 3 5 7 9 11 13 15 17 19 21 23 25 [6] M. E. Yaaacob, H. Hizaam, T. Khatib, and M. A. M.
DAY Radzi, "A comparative study of three types of grid
connected photovoltaicc systems based on actual
performance," Energy Conversion
C and Management,
Fig 11. November 2013, Total Energy Yield for C
Clean and Dusty vol. 78, pp. 8-13.
Array is 80.93 kWh and 80.19KWh, respectively [7] M. Maghami, H. Hizaam, C. Gomes, and I. AG,
"Characterization of Dusst Materials on the Surface of
Solar Panel," Life Sciencce Journal, vol. 11.
[8] H. Hottel and B. Woerttz, "Performance of flat-plate
IV. CONCLUSION solar-heat collectors," Trans.
Tr ASME (Am. Soc. Mech.
Eng.);(United States), vool. 64, 1942.
In order to study the dust on the surface of thhe solar arrays, [9] E. Asl-Soleimani, S. Faarhangi, and M. Zabihi, "The
two Fix flat photovoltaic (FFP) arrays werre installed in effect of tilt angle, air pollution on performance of
Unversiti Putra Malaysia. One of these arraays was Clean photovoltaic systems in Tehran," Renewable Energy,
regularly and the other one did not cleaneed during this vol. 24, pp. 459-468, 200 01.
research. Data was collected from the PV arraay like current
and voltage and environment so as to evolutees both arrays

393