Smart solar panel monitoring system

Using image processing

Turkish Online Journal of Qualitative Inquiry (TOJQI)

Volume 12, Issue 5, May 2021: 2181-2191

Smart solar panel monitoring system Using image processing

Ms. M. Perarasia, Dr. Geetha Ramadasb, Sherwin Joseph [Link], Somasundaram [Link], Vasanth
Kumar .Ke, Venkata Vamsi Krishna Somaf

a,b,c,d,e,f,
Rmk Engineering College, Electrical And Electronics Department

Corresponding author: a [Link]@[Link], b [Link]@[Link], c

[Link]@[Link], [Link]@[Link], [Link]@[Link],
f
[Link]@[Link]

Abstract

In the present study, the dust is being deposited on the Panel where the efficiency and the output power
decreases. So in order to increase the efficiency, we are proposing a system based on Image Processing. Image
processing uses a camera and senses the dust which triggers the motor to clean the Solar Panel using Wiper. In
addition to this, a webpage is also developed in order to check the Power moni]toring. In case if the automated
cleaning process isn't responding, a operating button will also be present in the webpage to clean the Solar Panel
manually.

Keywords: Image processing, Solar Power Monitoring, Webpage

1. Introduction
Solar energy is the cleanest and the most important source of Renewable energy. It generates Electrical
energy from Light energy. At Present, the Solar Panel accounts for the efficiency ratings as 23% while the other
panels show between 15% and 17%. In the United Arab Emirates region, dust on the Solar Panel is one of the
main issues that reduce efficiency. It is classified into two types as Active and Passive solar. Photovoltaics (PV)
is one of the examples for active systems where it generates electricity from sunlight directly via the electronic
process known as photovoltaic effect Solar Panels. Dust accumulation is the major drawback in the functions of
the PV module. UAE witnesses a large measure of residue storms consistently. Soiling is used to show the
amount of dust and other pollutants accumulated on the PV module and lead
to form a thin coating and prevent the high power of light from falling on the module. Dust is a compact
substance that is less than 250μm in radius. The dust settlement depends on factors such as location, weather
conditions, dust properties, wind speed, and slant angle of a Solar Panel. Dirtying can become lasting when
moistness condensates and makes dust stick on the outside of the Solar Panel. Residue deposition on the edge
can cause concealing on the cells which will harm the covering also. Soiling in the PV framework can cause
power misfortune up to 17% to 18% each year. Several methods are available for cleaning which depends on the
weather, location, and type of mounting.
2. Literature Survey
DC Jordan (1) reviews that Degradation rates are needed to know to predict power delivery. The degradation
rates measured in the last 40 years on individual modules. The rate is 0.5% per year.
K V Vidyanandan (2) assesses the impact of soiling

2181
Ms. M. Perarasi, Dr. Geetha Ramadas, Sherwin Joseph N.A, Somasundaram V.L, Vasanth Kumar .K,
Venkata Vamsi Krishna Soma

on PV module. Soiling is a large amount of dust and pollutants accumulated on solar panels which prevent
the falling of large light on the panel. So efficiency reduces up to 17% - 18% per annum so efficiency can be
increased by cleaning dust depends on location, wind speed, and weather conditions.
Javed (3) talks about regular cleaning of the panel and there will be a high difference between estimated
yield and actual field.
Hussain (4) stated that power generated is reduced to 850 W/m², 750 w/m², 650 w/m² in different radiations.
The output is reduced because solar radiation is blocked by dust particles.
J Antonanzas (5) stated various methods of forecasting solar power. The first method is obtaining accurate
irradiance forecasts. The second method is predicting output using machine learning and statistical method and a
hybrid model using first and second methods.
Brabec (6) reviews the advantages and performance of the statistical method since the output can arrive
without previous data. Thus temporary and pixel resolutions cause errors.
Lorenz (7) found a new way to remove errors by the model yield statistics. This method is helpful for
weather forecasting and improving resolution by inserting values.
Almeida (8) has scheduled the training sets for empirical distribution based on the same clearness index for
each day with 30days’ data collected.
B. M. A Mohandes (9) discussed about PV modules exposing to different climate conditions and more
chance of gathering dust particles, industrial residue, bird droppings on the solar panel. Some cleaning products
with chemical and other forms will attract unwanted dirt particles and prevents PV cells of a panel to enter solar
irradiance.
M J Adinoyi (10) stated that various factors like PV module direction, coating, rough surface, slope
influence dust gathering. It depends upon the size, weight, shape, and chemical properties of dust.
3. Effects Of Dust
A.A. Hachicha (21) audits the impact of residue on the force yield proficiency of sunlight-based PV
framework under UAE environment conditions. The dust particles were little in size in which a large number of
the particles were underneath 25mm greatly affecting the proficiency just as the electrical attributes of the
photovoltaic framework. During high residue thickness, ISC steadily diminished. The testimony of the residue
relies upon the direction and slant point of the photovoltaic board. Expulsion of residue is elevated because of
the impact of gravity for example with expansion in the slanted point of the PV surface, dust statement
increments. Precipitation is arbitrary in UAE yet it can help increment the photovoltaic proficiency and decrease
the thickness of the amassed dust. The activity and support of PV modules can't rely upon rare precipitation and
ought to be planned oftentimes relying upon the gathered residue thickness.
4. Dust Cleaning Agents
There are several processes for cleaning solar panels such as natural cleaning, automatic cleaning, and
manual cleaning. In the process of natural cleaning by rain, wind, melting snow When it is raining, the dust
particles lying on the panel are washed out by rainwater. During the wind, dust particles will be blown off the
solar panel. On to the next in the process of automatic cleaning, there are many types like microcontroller-based
sweeper which is used to control the brush or sweeper. Another type is automatic water spray is used to spray
the water on the panel to clean the surface which can be controlled by a set of timer-based boards. At last, in
automatic cleaning there are cleaning robots which will be placed on the panel which is used to detect the dust
and clean the panel hence it costs high. In the manual cleaning process, it depends upon the person to clean the
panel. The person may use a soft water cloth or water jet sweeper to clean the panel.

2182
 Smart solar panel monitoring system
Using image processing

Fig.1
Hence we consider the overall methods to increase the efficiency. Here we explore the exclusive method for
cleaning panels using a single line wiper with the help of a motor.
5. Components
A. Hardware Components

[Link] COMPONENTS SPECIFICATIONS

1 SOLAR PANEL  Pmax 5W
Fig. 2  Vmp 19.00V
 Voc 22.76V
 Isc 0.39A
2 RASPBERRY PI  1GB RAM
Fig. 3  HDMI PORT (Full Size)
 Micro SD card port (16
GB)
 Camera port for connecting in
Raspberry Pi camera

3 CAMERA  2MP
Fig. 4
4 POWER SUPPLY CABLE  230V
Fig. 5
5 MOTOR WITH WIPER  Step Motor
Fig. 6  6.0V
6 ANALOG TO  MCB 3008
DIGITAL CONVERTER
(ADC) Fig. 7
7 BATTERY  6V
Fig. 8
8 ULN2003 STEPPER  A-B-C-D LED is used to
MOTOR DRIVER indicate the 4- Phase working
Fig. 9 condition.
9 CURRENT SENSOR  5V
Fig. 10
10 VOLTAGE SENSOR  0-25V
Fig. 10
11 ARDUINO NANO  Operating Voltage : 5 V
Fig. 11  Input Voltage : 7-12 V

2183
Ms. M. Perarasi, Dr. Geetha Ramadas, Sherwin Joseph N.A, Somasundaram V.L, Vasanth Kumar .K,
Venkata Vamsi Krishna Soma

[Link] Requirements

 RASPBERRY PI OS
 PYTHON 3
 WEB SERVER

Solar Panel Fig. 2 Raspberry Pi Fig. 3

Camera Fig. 4 Power Supply Cables Fig. 5

Fig. 6 Motor with wiper ADC Fig. 7

Fig. 8 Battery Fig. 9 ULN2003 Stepper motor

Fig. 10 Voltage & Current Sensor Fig. 11 Arduino Nano

2184
 Smart solar panel monitoring system
Using image processing

6. Existing System
In the existing system Arduino board, relay, DC motor with a motor drive is used to clean the solar panel.
The panel is connected to the Arduino board. A relay and L293D are connected to Arduino. L293D motor
module can drive the motor in both directions by varying the voltage. The power supply is given to the circuit.
the motor driver drives the motor and a motor starts rotating to forward and backward directions. The
microcontroller is programmed as per desired time interval. Data is seen on the LCD screen. Also in the existing
system, we cannot monitor the status of the panel when it is required. So, to overcome such difficulties, we are
proposing a new method where we can monitor the panel lively with a camera and can also see the required data
on a webpage.

Fig. 12
7. Proposed System
In Proposed System, Image processing is used to detect the dust present in the Panel. Camera is installed on
the Solar Panel to capture the image . If there is any dust found in the Panel, the Motor starts cleaning
automatically. The condition of the Solar Panel such as Cracks, Moisture, Obstacles if any, can be seen in
monitor in live. Also the Current and Voltage ratings can be monitored through Webpage remotely.
8. Block Diagram

Fig. 13

2185
Ms. M. Perarasi, Dr. Geetha Ramadas, Sherwin Joseph N.A, Somasundaram V.L, Vasanth Kumar .K,
Venkata Vamsi Krishna Soma

9 Working
In this Project, the Image Processing Technology is used to detect the dust on the Solar Panel using
Raspberry Pi. A Raspberry Pi 3 Model B+ is used in which a Camera is attached. The Camera captures the
image (condition) of the Solar Panel and indicates 'Normal' or 'Abnormal'. If there is no dust present on the
Solar Panel, it indicates as 'Normal', whereas if there is dust present on the Solar Panel, it indicates and
'Abnormal'. This process is done with the help of Arduino Nano which has a program and connected to
Raspberry Pi. If the dust is identified, the ULN2003 Step Motor Drive triggers the motor to clean the Solar
Panel with the help of a Wiper. A Battery is connected to the Solar Panel along with the current sensor and the
voltage sensor. The Current sensor senses the current rating while the Voltage sensor senses the voltage rating.
The output will be of an Analog form, so an ADC is used to display the output digitally and it is connected to
Raspberry Pi. The output can be monitored through a webpage.
10 Flowchart

Fig. 14

2186
 Smart solar panel monitoring system
Using image processing

11 Algorithm
STEP 1: Start the system with the help of a Power supply.
STEP 2: Solar Panel will obtain energy from the sun and that
energy is stored in the battery.
STEP 3: The Voltage Sensor senses the voltage rating and
Current Sensor senses the current rating and
connected to the battery.
STEP 4: Output of the sensors will be of an analog form and
it is converted to digital form with the help of ADC.
STEP 5: To display the digital data, ADC is connected to
Raspberry Pi.
STEP 6: Motor is mounted to the panel and it is connected
motor driver (ULN2003) which works with the
help of Arduino Nano.
STEP 7: Python program is fed to the raspberry pi which
will operate the motor using Arduino Nano.
STEP 8: Switch on the Monitor.
STEP 9: Open the Command Prompt.
STEP 10: Enter the following commands as given below
Cd rpi-vision/ ( Press Enter )
[Link]/bin/activate (Press Enter )
Python3 [Link] (Press Enter )
The above code should be given to the
Raspberry Pi Command Prompt to activate the
Code.
STEP 11: Using a Camera, the dust particles are checked.
11.1: If Yes (Abnormal), Motor triggers and
the wiper starts cleaning.
11.2: If No (Normal), Motor will not run.
STEP 12: Webpage can be used to check the ratings of
Current and Voltage. (Fig. 17 & 18).
STEP 13: Motor stops running.
12 Conclusion
In this paper, a camera is utilized to catch a picture of the Solar Panel and reads the amount of the dust. If
the amount of the dust exceeds certain limit, it triggers the motor which cleans the Solar Panel with the help of a
wiper. Additionally, the parameters like battery capacity, voltage and current via sensors (Figure 8) can be
monitored through Webpage from anywhere and at anytime. So through this paper, we can analyze condition of
the solar panel effectively. The proposal of the project has been applied and we have taken some images in
environment and result have been shown through images.

2187
Ms. M. Perarasi, Dr. Geetha Ramadas, Sherwin Joseph N.A, Somasundaram V.L, Vasanth Kumar .K,
Venkata Vamsi Krishna Soma

13. Result
Here, we have shown the output of the Solar Panel where the webpage shows the Voltage and Current
rating. The Condition of Solar Panel will also be mentioned as ‘Normal’ or ‘Abnormal’. When there is no dust
on the Solar Panel, the camera senses and will mention as ‘Normal’ in the webpage, whereas when the dust is
present on the Solar Panel, the camera captures and mentions it as ‘Abnormal’ and it triggers the motor to clean
the panel.

Fig. 15

Fig. 16
Fig. 15 shows the output of the Solar Panel where it shows the status as ‘Normal or ‘Abnormal’ along with
the Voltage and Current rating.
Fig. 16 shows that the camera capturing the image of the Solar Panel along with the output.

2188
 Smart solar panel monitoring system
Using image processing

Fig. 17

Fig. 18
Fig. 17 and Fig. 18 shows the Current and Voltage rating of the Solar Panel with the date and time
mentioned in the Webpage

References

[1] D.C. Jordan and Kurt , “Photovoltaic Degradation Rates – An Analytical Review”, Progress in
Photovoltaic’s: Researchs and Applications, vol. 19-22, no. 1, pp. 11-30, Jan. 2013.
[2] Vidyanandan, K.V, “ An Overview of Factors Affecting the Performance of Solar PV Systems”, NTPC
Lt d, vol. 26-28, pp. 1-9, February 2017.
[3] Javed, Wubulikasimu & Wasim & Yiming, Figgis,, Bing. Characteristic of dust accumulated on PV
panels in Qatar, Doha, Solar Energy, vol. 141, pp.121-137, January 2017.

2189
Ms. M. Perarasi, Dr. Geetha Ramadas, Sherwin Joseph N.A, Somasundaram V.L, Vasanth Kumar .K,
Venkata Vamsi Krishna Soma

[4] Hussain Ankit, Pachauri Athar, Batra Rupendra, “ An experiment al study on effect of dust on power
loss in solar photovoltaic module ”,Renewable : Water Solar and Wind. vol. 4-5, December 2017
[5] Antonanzas, , N., Escobar, J., Osorio R., Urraca, R., Martinez-de-Pison., & Antonanzas-Torres,“ Review
of PV power forecasting”, Solar Energy, vol.135-136, pp. 77 –112, June 2016.
[6] Brabec, E., Krc, P . M., Pelikán, Eben, K., Juruš, P, Maly´,“ A coupled model for energy production in
the method of forecasting from PV farms”, In: ES1003, Workshop March 22nd–23rd, 2011.
[7] Lorenz, D. Wickramarathne, E. Heinemann, Bofinger S,H. Beyer “ Forecasting of ensemble power
product ion by grid-PV connected systems”, In: 20th European PV Conference, Milano,Italy, 09. –
08.09.2007.
[8] Almeida, O., Narvart , M.P, Perpiñán, “ PV power forecast using a non-parametric PV model”, Solar
Energy, vol.114-116, pp. 352–370, 2015.
[9] L. El-Chaar, L.A. Lamont , B.M.A. Mohandes “Application study of 500 W photovoltaic (PV) system
in the UAE”, Applied Solar Energy,vol. 46, pp. 243-247, December 2010.
[10] S.A. Adinoyi, Said, “Effect of dust gathering on the power outputs of solar PV modules”, Renewable
Energy, vol. 61, pp.632-637, December 2013.
[11] Corkish, C. B. Honsberg, and Bremner, “A New Generalized Detailed Balance Formulation to Calculate
Solar Cell Efficiency Limits”, 17th European Photovoltaic Solar Energy Conference, pp. 22- 26, 2001.
[12] Venkateswari, R. & Sreejith S, “ Factors influencing the efficiency of photovoltaic system”, Renewable
and Sustainable Energy Reviews, vol. 101, pp. 376-394, March 2019.
[13] G. Boyle, “Renewable energy poher for a sustainable future”, 2nd edition, Oxford Press, 2004.
[14] Hachicha, Al-Sawafta, Ahmed Amine Said, "Impact of dust on the performance of solar photovoltaic
(PV) systems under United Arab Emirates weather conditions”, Renewable Energy, Elsevier,vol. 140-
141(C), pp. 286-299, April 2019
[15] R. Pillai, M. Mani, “Impact of dust on solar photovoltaic (PV)performance: research status, challenges
and recommendations”,Renewable and Sustainable Energy Reviews, vol.14, pp. 3123-3231, December
2010.
[16] , R. Heijnen, VerzijlberghPS. Los, A Jonker H., . de Roode, “ Improved model output statistics of
numerical weather prediction-based irradiance forecasts for solar power applications”, Solar Energy,
vol. 119-120, pp. 633–646, 2015.
[17] Gradit i, o, S. Adinolfi, G., G. Ferlit “ Comparison of photovoltaic plant power production prediction
methods using a large measured dataset ”,Renewable Energy, vol. 90-91, pp. 512–517, 2016.
[18] Mellit , A Lughi, Massi Pavan,, “ Short-term forecasting of power production in a large-scale
photovoltaic plant ”, Solar Energy, vol. 104-106,pp. 400–414, 2014.
[19] W.J., Liu, Y., Shi, J., Lee, Yang, Y., Wang, P., “ Forecasting power output of photovoltaic system based
on weather classification and support vector machine” IEEE Industry Applications Society Annual
Meeting.
[20] Hachicha, Ahmed Amine & Al-Sawafta, Israa & Said, Zafar, "Impact of dust on the performance of
solar photovoltaic (PV) systems under United Arab Emirates weather conditions”, Renewable Energy.
[21] Ramos, E. Z., Ramos, M., Moutafis, K., & Yfantis, E. A. (2016). Robot-server architecture for
optimizing solar panel power output.
[22] Trupti Rajendra Ingale. (2017) “A Review paper on Biometrics Implementation Based on Internet of
things using Raspberry Pi.”, International Journal of Scientific Research in Computer Science,
Engineering and Information Technology: Volume 2,Issue 2,ISSN:2456-3307.
[23] Neelam Sharma,Rohini Awsare, Rasika Patil, and Pawan Kumar. (2017), “Review on Smart Mirror
Using Raspberry PI 3 Based On Iot”, International Journal of Research in Science Engineering: e-ISSN:
2294-8299, Volume-2, Issue-6.
[24] Shruti G. Suryawanshi, and Suresh A. Annadate. (2016) “Raspberry Pi based Interactive Smart Home
Automation System through E-mail using Sensors.”,International Journal of Advanced Research in
Computer and Communication Engineering: Vol-5,Issue-2,February.

2190
 Smart solar panel monitoring system
Using image processing

[25] Jagdish A. Patel, Aringale Shubhangi, Shweta Joshi, Aarti Pawar, and Namrata Bari. (2016) “Raspberry
PI Based Smart Home.”, IJESC:DOI
[26] Pratibha Jha, Prashant Jha, Mufeed Khan,and Kajol Mittal. (2019) “Smart Mirror: A Journey to the new
world.”, International Journal of Computer Sciences and Engineering: Vol.-7, Issue-1, January 2019 E-
ISSN: 2347-2693.
[27] Saima Shaikh, Dipali Gadakh, Tarulata Patil, Divya Borse, and [Link]. (2019) “Smart Mirror For
Vehicular System Using Raspberry Pi.”, IJIRT: Volume 5 Issue 10, March 2019 ISSN: 2340-61.
[28] Yao, Y. Y., & Hu, Y. (2017). Recognition and Location of Solar Panels Based on Machine Vision. In
2nd Asia-Pacific Conference on Intelligent Robot Systems (ACIRS). [7–12].
[29] Yap, W.K., Galet, R., & Yeo, K.C. (2015). Quantitative Analysis of Dust and Soiling on Solar PV
Panels in the Tropics Utilizing Image-Processing [Link] Asia Pacific Solar Research Conference.
[30] Yfantis, E., & Fayed, A. (2014). A camera system for detecting dust and other deposits on solar panels.
Advances in Image and Video Processing,
[31] Bacher, P., Madsen, H., Nielsen, H., “Online short-term solar power forecasting” ,Solar Energy, vol. 83,
pp. 1772–1783, 2009.
[32] Shu, L., “An ARMAX model for forecasting the power output of a grid connect ed photovoltaic
system”, Renewable Energy, vol. 66, pp. 78–89, 2014.
[33] Pedro, H.T.C., Coimbra, C.F.M., “Assessment of forecasting techniques for solar power production with
no exogenous input s”, Solar Energy, vol. 86, pp. 2017–2028, 2012.
[34] Guan, H., Xiao, B., Zhou, Z., & Yan, X. (2017). In-situ investigation of the effect of dust deposition on
the performance of polycrystalline silicon photovoltaic modules. Renewable Energy, 101, 1273–1284.

2191