SOLAR PANEL CLEANING SYSTEM

Project ID. 2023: 111

Session: BSc. Fall 2019
Project Supervisor: Lec. Abdul Basit Taj

Submitted By

Muhammad Arslan Janjua

Fahad Ahmad
Muhammad Saad Ashfaq

Electrical Engineering Department

National University of Technology (NUTECH)
Islamabad, Pakistan
2023

i
 Certification

This is to certify that Muhammad Arslan Janjua, [Link]. F19603006, Muhammad Saad
Ashfaq, [Link]. F19603025 and Fahad Ahmad, [Link]. F19603035, have successfully
completed the final project Solar Panel Cleaning System, at the National University of
Technology, to fulfill the partial requirement of the degree BS of Electrical Engineering.

External Examiner Project Supervisor

[Name of Examiner] Lec. Abdul Basit Taj

[Designation] Lecturer

Dean of University

Electrical Engineering Department, National University of Technology

ii
 Solar Panel Cleaning System

Sustainable Development Goals

SDG No Description of SDG SDG No Description of SDG

SDG 1 No Poverty SDG 9 ✓ Industry, Innovation, and Infrastructure

SDG 2 Zero Hunger SDG 10 Reduced Inequalities

SDG 3 Good Health and Well Being SDG 11 Sustainable Cities and Communities

SDG 4 Quality Education SDG 12 Responsible Consumption and Production

SDG 5 Gender Equality SDG 13 Climate Change

SDG 6 Clean Water and Sanitation SDG 14 Life Below Water

SDG 7 Affordable and Clean Energy SDG 15 Life on Land

SDG 8 ✓ Decent Work and Economic Growth SDG 16 Peace, Justice and Strong Institutions

SDG 17 Partnerships for the Goals

iii
 Range of Complex Problem Solving
Attribute Complex Problem
Range of conflicting Involve wide-ranging or conflicting technical, engineering and other
1
requirements issues.
Have no obvious solution and require abstract thinking, originality in
2 Depth of analysis required
analysis to formulate suitable models.
Requires research-based knowledge much of which is at, or informed
Depth of knowledge
3 by, the forefront of the professional discipline and which allows a
required
fundamentals-based, first principles analytical approach.
4 Familiarity of issues Involve infrequently encountered issues
Are outside problems encompassed by standards and codes of practice
5 Extent of applicable codes
for professional engineering.
Extent of stakeholder
involvement
6 and level of Involve diverse groups of stakeholders with widely varying needs.
conflicting requirements
7 Consequences Have significant consequences in a range of contexts.
Are high level problems including many component parts or sub -
8 Interdependence
problems
Range of Complex Problem Activities
Attribute Complex Activities
Involve the use of diverse resources (and for this purpose, resources
1 Range of resources
include people, money, equipment, materials, information and technologies).
Require resolution of significant problems arising from interactions
2 Level of interaction
between wide ranging and conflicting technical, engineering or other issues.
Involve creative use of engineering
3 Innovation
principles and research-based knowledge in novel ways.
Consequences to society Have significant consequences in a range of contexts, characterized by
4
and the environment difficulty of prediction and mitigation.
Can extend beyond previous experiences by applying principles -based
5 Familiarity
approaches.

iv
 Abstract

During the last decade, the usage of solar energy technology has increased
tremendously due to the increase in electricity demand all around the world. Solar energy
is harnessed using photovoltaic (PV) Panels which are arranged in the form of arrays. Due
to the accumulation of dust, debris, and grime (moisture & dust compound), the efficie nc y
of the solar panel decreases in terms of power generation. Loss in power generation will
result in fewer electricity units being produced. This emphasizes the significance of keeping
solar panel surfaces as clean as possible.

The goal of this project is to create an automated solar panel cleaning system that employs
water, wiper, and brush mechanism to clean the panel. We autonomously control the
mechanical mechanism using environmental data where the solar system is installed. The
data is monitored through an online cloud-based server and is interconnected with an
Android app.

Our designed mechanism works and performs its functions adequately on L1-array type
solar placement and shed-type solar placement. To clean solar panels arranged in L2 and
L3-array type solar placement, the system design must be extended.

Furthermore, the designed system is economical on a larger scale as compared to the cost
of solar system installation which produces power in 50KW and above. The cleaning
accuracy achieved by our system is about 70%, which is based on the analysis during
cleaning.

v
 Undertaking

I certify that the project Solar Panel Cleaning System is our own work. The work
has not, in whole or in part, been presented elsewhere for assessment. Where material has
been used from other sources it has been properly acknowledged/ referred.

[Muhammad Arslan Janjua]

[F-19603006]

[Muhammad Saad Ashfaq]

[F-19603025]

[Fahad Ahmad]

[F-19603035]

vi
 Acknowledgement

All the acclamation and appreciations are for Almighty ALLAH who created the universe
and bestowed mankind with knowledge and wisdom to search for its secrets.
We would like to express our deep gratitude to Professor Dr. Yasir Awais Butt for his
extensive knowledge and expertise in the field of Electrical Engineering. His valuable
insights and advice played a crucial role in shaping our project and pushing us toward
excellence. His unwavering support and encouragement have been instrumental in our
success.
We would also like to extend our sincere appreciation to Lecturer Abdul Basit Taj for his
continuous support and dedication throughout the project. His practical expertise and
willingness to share his knowledge have been invaluable to our understanding of the subject
matter. His guidance and feedback were instrumental in refining our ideas and ensuring the
successful implementation of our project. Their availability and willingness to answer our
questions and address our concerns were immensely helpful in overcoming the challenges
faced during the project.
Finally, we would like to express our gratitude to the entire faculty of the Electrica l
Engineering Department for providing an environment conducive to learning and
innovation. Their dedication to education and commitment to fostering the growth of their
students have been truly inspiring.
We are thankful to our parents for their support and cooperation during this project. It
was their sincerest prayers and blessings which directed our efforts towards a
meaningful cause, that is, the completion of this project.

vii
 TABLE OF CONTENTS

SOLAR PANEL CLEANING SYSTEM............................................................................ i

LIST OF FIGURES ..................................................................................................... xi

LIST OF ABBREVIATIONS ...................................................................................... xiv

ABSTRACT .............................................................................................................. xv

Chapter 1 ................................................................................................................. 1

INTRODUCTION ....................................................................................................... 1
1.1 Background ...................................................................................................... 1
1.1.1 Worldwide Solar Energy Consumption ..................................................... 1

1.2 Problem Statement .......................................................................................... 2

1.2.1 Soiling issue ............................................................................................... 3

1.3 Aims and Objectives ......................................................................................... 3

1.4 Applications of the Project ............................................................................... 4

1.5 Project Plan ...................................................................................................... 5

1.6 Report Organization......................................................................................... 6

Chapter 2 ................................................................................................................. 7

LITERATURE REVIEW ............................................................................................... 7

2.1 Solar Panel Cleaning Current technologies and Practices ............................... 7
2.1.1 Current Technologies ................................................................................ 7
2.1.2 Current Practices ....................................................................................... 9

2.2 Projects Related to Solar Panel Cleaning System........................................... 10

2.2.1 Automatic Solar Cleaning System For Dust Removal [26] ...................... 10
2.2.2 Novel Dry Cleaning Machine[27] ............................................................ 11
2.2.3 Cleaning Robot For Solar Panels[28] ....................................................... 13
2.2.4 Smart Solar Photovoltaic Panel Cleaning System ................................... 13

2.3 Limitations...................................................................................................... 14
2.3.1 Current Technologies and Practices Limitations..................................... 14
viii
 2.3.2 Related Projects Limitations.................................................................... 14

2.4 Summary ........................................................................................................ 15

Chapter 3 ............................................................................................................... 16

DESIGN AND SOFTWARE SIMULATIONS............................................................... 16

3.1 Methodology .................................................................................................. 16

3.2 Procedures for Design .................................................................................... 16

3.3 Frame Mechanism Design.............................................................................. 17

3.3.1 Weight Calculation .................................................................................. 17
3.3.2 Torque And Speed Of Motor ................................................................... 22

3.4 Software Simulation ....................................................................................... 23

3.4.1 Motor Driver Design................................................................................ 23
3.4.2 Limit Switch ............................................................................................. 25
3.4.3 Pump Motor ............................................................................................ 26
3.4.4 Ultrasonic Sensor .................................................................................... 27

3.5 Summary ........................................................................................................ 28

Chapter 4 ............................................................................................................... 29

Hardware Development and Specification .......................................................... 29

4.1 System Specification .......................................................................................... 29

4.2 System Outline ................................................................................................... 29

4.3 Component Explanation .................................................................................... 30

4.3.1 ESP32 Wi-fi Module..................................................................................... 30
4.3.2 DHT11 Humidity Sensor .............................................................................. 31
4.3.3 BMP 280 Atmospheric Pressure Sensor...................................................... 32
4.3.4 LM2596 Buck Converter .............................................................................. 32
4.3.5 Limit Switch ................................................................................................. 33
4.3.6 12V Relay..................................................................................................... 33
4.3.7 12V Water Pump ......................................................................................... 33
4.3.8 Battery Selection ......................................................................................... 34

ix
 4.3.9 Solar Panel Parameters ............................................................................... 35

4.4 Hardware Prototype .......................................................................................... 36

4.4.1 Motor Mount............................................................................................... 36
4.4.2 Dc Gear Motor............................................................................................. 37
4.4.3 Rods Shaft With Frame Structure ............................................................... 37
4.4.4 Working Prototype ...................................................................................... 38

4.5 Summary............................................................................................................ 40

Chapter 5 ............................................................................................................... 41

RESULT AND ANALYSIS.......................................................................................... 41

5.1 Code Explanation ............................................................................................... 41

5.2 Cleaning System Testing and Analysis ............................................................... 42

5.2.1 Analysis Before Cleaning ............................................................................. 43
3.5.2 Analysis After cleaning ............................................................................ 43

5.3 Data Analysis ..................................................................................................... 43

5.3.1Humidity vs Temperature At daytime and Nighttime ................................. 43
5.3.2 Voltage and Current Data Comparison ....................................................... 46

5.4 Summary............................................................................................................ 47

Chapter 6 ............................................................................................................... 48

CONCLUSION ......................................................................................................... 48
6.1 Recommendations ............................................................................................. 48

REFERENCES ........................................................................................................... 50

APPENDICES ........................................................................................................... 53

APPENDIX A ........................................................................................................... 53
CODE ........................................................................................................................ 53

x
 LIST OF FIGURES
FIGURE 1.1 INSTALLED CAPACITY OF SO LAR ENERGY WORLDWIDE[10] ................................................................2
FIGURE 2 PROJECT PHASE 1: SEMESTER 7 SCHEDULE................................................................................................5
FIGURE 3 PROJECT PHASE 2: SEMESTER 8 SCHEDULE................................................................................................6
FIGURE 4 REPORT ORGANIZATION FLOW CHART .......................................................................................................6
FIGURE 5 MANUAL HAND CLEANING SYSTEM ............................................................................................................7
FIGURE 6 LINEAR PIEZOELECTRIC SYSTEM ...................................................................................................................8
FIGURE 7 SMART SO LAR PANEL CLEANING SYSTEM[22] ...........................................................................................9
FIGURE 8 GECKO SOLAR ROBOT[23] .............................................................................................................................9
FIGURE 9 BOSO N ROBOTIC PANEL CLEANING SYSTEM[25] ................................................................................... 10
FIGURE 10 DIFFERENT TYPES OF SAND USED FOR EXPERIMENTAL TEST [26] .................................................... 11
FIGURE 11 FLOW CHART FOR FO UR STAGES AUTO MATIC DRY-CLEANING SYSTEM ......................................... 12
FIGURE 12 DRY AUTOMATIC CLEANING SYSTEM OVERVIEW[27]......................................................................... 13
FIGURE 13 CLEANING ROBOT PROTOTYPE AND HARDW ARE[28]......................................................................... 13
FIGURE 14 ROBOTIC UNIT: FABRICATED PRO TOTYPE[29] ...................................................................................... 14
FIGURE 15 3D MODEL FO R WATER SPRINKLING SHAFT ......................................................................................... 18
FIGURE 16 3D MODEL NYLO N BRUSH SHAFT ........................................................................................................... 18
FIGURE 17 3D MODEL RECTANGULAR FRAME ......................................................................................................... 19
FIGURE 18 3D MODEL WINDSCREEN WIPER 24 INCH ............................................................................................. 20
FIGURE 19 DIAGRAM FOR MO TOR MO UNT ............................................................................................................. 20
FIGURE 20 3D MODEL FO R FIXED SHAFTS................................................................................................................ 21
FIGURE 21 MO TOR DRIVER USING AT89C51 AND IRF840...................................................................................... 24
FIGURE 22 RELAY-BASED MOTOR DIRECTION CHANGE ......................................................................................... 25
FIGURE 23 PROTEUS SYMMETRIC FOR LIMIT SWITCH USING SPDT .................................................................... 25
FIGURE 24 ARDUINO CODE FOR LIMIT SWITCH ....................................................................................................... 26
FIGURE 25 PUMP MOTOR PROTEUS SIMULATION .................................................................................................. 27
FIGURE 26 ULTRASONIC SENSOR PROTEUS SIMULATION..................................................................................... 27
FIGURE 27 SYSTEM OUTLINE FLOW DIAGRAM ....................................................................................................... 30
FIGURE 28 ESP32 WI-FI AND BLUETOOTH MODULE[30] ........................................................................................ 31
FIGURE 29 DHT11 HUMIDITY & TEMPERATURE S ENSOR[31] ................................................................................ 31
FIGURE 30 BMP 280 TEMPERATURE AND ATMOSP HERIC PRESSURE SENSO R[32] ........................................... 32
FIGURE 31 LM2596 BUCK CONVERTER ADJUSTABLE[33] ....................................................................................... 33
FIGURE 32 LIMIT SWITCH ............................................................................................................................................. 33
FIGURE 33 12V RELAY[34] ............................................................................................................................................ 33
FIGURE 34 WATER PUMP 12V [35] ............................................................................................................................. 34
FIGURE 35 12V 7AH BATTERY ...................................................................................................................................... 35
FIGURE 36 SOLAR PANEL CURRENT & VO LTAGE CIRCUIT ...................................................................................... 36

xi
FIGURE 37 MO TOR MOUNT SIDE AND TOP VIEW ................................................................................................... 37
FIGURE 38 DIFFERENT VIEW OF DC GEARED MOTOR ............................................................................................. 37
FIGURE 39 SHAFT ROD BETW EEN TWO MO TOR MO UNTS .................................................................................... 38
FIGURE 40 SOLAR MOUNTING FRAME STRUCTURE ................................................................................................ 38
FIGURE 41 WORKING PRO TOTYPE FOR SO LAR PANEL CLEANING SYSTEM......................................................... 39
FIGURE 42 REAL-TIME DATABASE ON FIREBASE CLOUD SERVER.......................................................................... 40
FIGURE 43 GETTING DATA FOR SENSORS.................................................................................................................. 41
FIGURE 44 CODE FOR CHECKING THE DATA AND MAKING CLEANING DECISION ............................................. 41
FIGURE 45 SOLAR PANEL WITH DUST AND SAND PARTICLES ................................................................................ 42
FIGURE 46 SOLAR PANEL AFTER CLEANING WITH MECHANISM........................................................................... 43
FIGURE 47 GRAPH BETWEEN HUMIDITY VS TEMPERATURE AT DAYTIME .......................................................... 45
FIGURE 48 TEMPERATURES VS HUMIDITY DURING NIGHTTIME .......................................................................... 45
FIGURE 49 VO LTAGE VS CURRENT DURING DAY TIME ........................................................................................... 46
FIGURE 50 VO LTAGE VS CURRENT DURING NIGHTTIME ........................................................................................ 47

xii
LIST OF TABLES

TABLE 1 POWER LOSSES IN TERM OF FINANCIAL COST ANALYSIS...........................................................................3

TABLE 2 SYSTEM SPECIFICATION SHEET .................................................................................................................... 29
TABLE 3 HUMIDITY VS TEMPERATURE DATASET ..................................................................................................... 44

xiii
 LIST OF ABBREVIATIONS
IRENA INTERNATIONAL RENEWABLE ENERGY AGENCY

xiv
 ABSTRACT
During the last decade, the usage of solar energy technology has increased tremendous ly
due to the increase in electricity demand all around the world. Solar energy is harnessed
using photovoltaic (PV) Panels which are arranged in the form of arrays. Due to the
accumulation of dust, debris, and grime (moisture & dust compound), the efficiency of the
solar panel decreases in terms of power generation. Loss in power generation will result in
fewer electricity units being produced. This emphasizes the significance of keeping solar
panel surfaces as clean as possible.

The goal of this project is to create an automated solar panel cleaning system that employs
water, wiper, and brush mechanism to clean the panel. We autonomously control the
mechanical mechanism using environmental data where the solar system is installed. The
data is monitored through an online cloud-based server and is interconnected with an
Android app.

Our designed mechanism works and performs its functions adequately on L1-array type
solar placement and shed-type solar placement. To clean solar panels arranged in L2 and
L3-array type solar placement, the system design must be extended.

Furthermore, the designed system is economical on a larger scale as compared to the cost
of solar system installation which produces power in 50KW and above. The cleaning
accuracy achieved by our system is about 70%, which is based on the analysis during
cleaning.

xv
Chapter 1
INTRODUCTION
1.1 Background
The generation of energy in Pakistan is heavily reliant on fossil fuels. Due to the large
population and industry energy demands, these upwellings are not meeting the demand for
the energy needed by Pakistan. Pakistan is in its worst energy crisis.[1][2] Energy demand
all around the world has been increasing tremendously in the last decade. Energy demand
in Pakistan is increasing annually at a rate of 9%. It is expected that by 2050, energy
demand will increase to about 20% annually.[3]
Geographically Pakistan lies on the equator and throughout the year receives a large amount
of sunlight. To overcome the energy crises, present in the country, it is necessary to exploit
solar energy and other renewable resources available. Public and private sector investme nt
is important for harnessing the true potential of solar energy.[4]
Shaikh et al.[5] stressed the importance of solar energy as a natural asset for Pakistan.
Shakeel et al.[6] has reported that the country has an overall solar power potential of
1600,000 MW. Many researchers have found out about the challenges related to solar
energy faced in Pakistan.[7][8][3]
1.1.1 Worldwide Solar Energy Consumption

According to the renewable global status report and the International renewable energy
agency, the installed capacity of solar systems has increased from 40,000 MW in 2011 to
850,000 MW in 2020 in electricity generation. It is expected to cross the 1 million GWh
mark by the end of 2022. solar PV has emerged as a key component in the low-carbon
sustainable energy system required to provide access to affordable and dependable
electricity, assisting in fulfilling the Paris climate agreement and in achieving the 2030
SDG targets[9]. Figure.1 represents the installation of solar energy worldwide till 2020.

1
 Figure 1.1 Installed Capacity of Solar Energy Worldwide[10]

1.2 Problem Statement

Solar energy has huge potential if people utilize it for sustainable, reliable, and lasting
energy production and consumption. Many problems arise with solar systems after they are
installed. It is a fact that solar panels available in markets all around the world have an
efficiency of 16-22% that converts sunlight into usable power. Now this efficiency is
affected by different factors ranging from dust deposition and grimes due to moisture, bird
droppings, and partial shading[11]. Other factors that can affect the transplant glass are
corrosive gases[12] or natural disasters such as strong wind[13] can also have signific a nt
effects on solar panels. The main factor that our project will be dealing with is the dust,
grime, and bird droppings. According to a study, the efficiency of solar panels decreases
by 15-30% due to dust depending upon the location where the solar system is being
installed such as deserts and coastal areas.[14] Depending upon the location where dust
content is large, efficiency will be high and vice versa. The below table shows how much
losses occur in different residential solar systems across Pakistan if they are placed in an
environment having high dust content. The data below is calculated keeping in mind that

2
the solar system uses a 540W solar panel or any equivalent. One unit of energy in Pakistan
cost about 30 rupees according to WAPDA and this price may vary over time.
Type Of Total Losses due Remaining Units Units Financial
Solar Wattage to Dust at Wattage produced produced losses
Systems (W) 30% after daily at daily after (Rupees)
(KW) (W) Loss(W) maximum losses
capacity (KWh)
(KWh)

5 5940 1780 4160 20 16 3600

10 10260 3078 7182 40 30 90000

15 15120 4536 10464 60 40 18000

Table 1 Power Losses in Term of Financial Cost Analysis

The data shown in the above table may not seem to be creating a tremendous impact on
small residential systems where electricity is used on a small daily basis because, in the
long term, such losses are elusive and could financially have provided a person with greater
reliability and sustainability on their instilled system. In large-scale solar systems, huge
financial losses in terms of power generation could have made utility companies corpus the
amount of cash that is lost due to no regular cleaning of Solar panels. So, cleaning Solar
panels is necessary to maintain efficiency in terms of power generation.

1.2.1 Soiling issue

Atmospheric dust and soiling significantly reduce the performance of solar modules.
Soiling is an issue caused by the deposition of dust on solar modules.[15] Atmospheric dust
may absorb solar radiation, but soiling can degrade solar modules. The power generated by
solar panels depends upon the amount of irradiance. Soiling reduces the irradiance absorbed
by a solar panel. Researchers have concluded that soiling plays a tremendous role in
reducing the energy generation of solar power plants.[16]

1.3 Aims and Objectives

This Project aims to develop a sustainable, reliable, and easy-to-use system dedicated

3
to cleaning solar panels across any type of solar arrangement. The designed system should
meet the following objectives:
 System should be able to move across the panels in both forward and reverse
directions.
 The system would also have a wiper, brush, and water sprinkling mechanism
installed in it.
 The water sprinkling Mechanism should be able to control the flow of water.

1.4 Applications of the Project

This Project is aimed at developing a cleaning solution dedicated to solar panel cleaning.
The scope of this research is:
 Maximize Energy Production: Solar panels operate most efficiently when they are
clean and free from dirt, dust, bird droppings, and other debris. Regular cleaning
using automated cleaning systems ensures optimal sunlight absorption and
maximizes energy production.
 Improve System Performance: Accumulated dirt or dust on solar panels can reduce
their overall performance. By removing these contaminants, cleaning systems help
maintain the panels' efficiency and output capacity, ensuring the system operates at
its full potential.
 Cost Reduction: Dirty or partially obstructed solar panels lead to decreased power
generation, which can result in reduced financial returns. Cleaning systems help
mitigate this issue by preventing energy loss and reducing the need for additiona l
maintenance or panel replacements, resulting in long-term cost savings.
 Maintenance Efficiency: Manual cleaning of large-scale solar installations can be
time-consuming and labor-intensive. Automated cleaning systems, such as robotic
or waterless cleaning systems, enable efficient and faster cleaning of multip le
panels simultaneously, reducing the maintenance workload and associated costs.
 Longevity and Durability: Regular cleaning helps extend the lifespan of solar panels
by preventing corrosion and damage caused by environmental factors like dirt, salt,
or pollutants. By maintaining the panels in a clean state, the cleaning systems
contribute to their long-term durability.
 Environmental Benefits: Clean solar panels have a positive impact on the
environment. By optimizing energy production, cleaning systems facilitate the
generation of clean, renewable energy, reducing the reliance on fossil fuels and

4
 greenhouse gas emissions.
 Remote Monitoring: Some advanced cleaning systems are equipped with
monitoring capabilities that allow operators to remotely track the performance and
cleaning needs of solar panels. This enables proactive maintenance and ensures
timely cleaning, enhancing overall system efficiency.

1.5 Project Plan

The Project is divided into two phases. Each project phase is stretched over a period of 8
months for completion of various tasks about the solution proposed. Below is the project
plans for Phase 1 and Phase 2 respectively.

Figure 2 Project Phase 1: Semester 7 Schedule

5
 Figure 3 Project Phase 2: Semester 8 Schedule

1.6 Report Organization

This research is organized into six chapters as given in detail below:

Figure 4 Report Organization Flow Chart

6
Chapter 2
LITERATURE REVIEW
The idea for a solar panel cleaning system was conceived through the increasing energy
demand in Pakistan. For such an idea to be given a completely applicable shape, we need
to understand how different cleaning technologies work and people around the world have
made solar panel cleaning systems and different technologies employed to clean solar
panels.
2.1 Solar Panel Cleaning Current technologies and Practices
2.1.1 Current Technologies
Large-scale solar farms or projects are deployed in deserts or places where sunlight remains
for 8-10 hours during the day. In such environments where issues like soiling, moisture ,
and dust build-up on solar panel surfaces cause a huge rift in the power generation capacity
of solar plants. To avoid such issues, both traditional and modern cleaning solutions are
being used in industry to clean solar panels. Due to the advancement of materials and
robotics, traditional methods such as manual cleaning of solar panels are going to be
obsolete. Modern solutions such as robotics, cleaning kits, etc. are available to clean a panel
effectively. Rainwater
1. Manual cleaning
Different tools and techniques are used to clean solar panels manually with the help
of an operator. The most common type of manual cleaning tools is cleaning kits
which include extension poles, carrying bags, brushes, clothes, hose connections ,
and more[17]. Below picture operator working manually to clean the solar panels.

Figure 5 Manual hand cleaning System

2. Piezoelectric Systems
Piezoelectric actuators are being used in various optical adjustments, biomedical
manipulation, space exploration, and other areas due to their high torque to
volume ratio, flexible structure, and high positioning precision. Such systems are

7
 used to clean solar panels by using water as a cleaning agent and spreading it to a
depth of 0.1 to 1 mm during the rarefaction cycle of compression waves. A
vacuum is created in liquid during rarefaction called ultrasonic cavity and this
cavity sucks the dust from the solar panels.[18] [19] Below figure shows a
piezoelectric system used to clean solar panels.

Figure 6 Linear Piezoelectric System

3. Self- Cleaning Mechanism
A translucent self-cleaning nano-film can be coated onto the surface of solar
panels to prevent dust deposition. This nano-film is composed of either super-
hydrophilic or super-hydrophobic materials. The super-hydrophilicity method
relies on rainwater to disperse across the solar module and cleanse the dust
particles. However, this approach is not widely adopted due to the limited
availability of rainfall in solar farm locations. Alternatively, super-hydrophobic
materials facilitate the rapid removal of water droplets, carrying away dust
particles like the lotus leaf effect. However, careful examination is needed to
assess the practicality of implementing these materials on solar panels, as solar
farms are typically situated in regions with infrequent rainfall.[20] [21]
4. Robotic systems
Robotic systems consist of actuators, drivers, and gears having some movement
above the solar panel surface, and it is controlled through a virtual operator. The
rise of the Internet of Things and fuzzy logic has allowed engineers to develop
robotic solar cleaning systems capable of cleaning dust and other excrement of
panel surfaces. Robotic units act as slave units and perform the cleaning as
commanded by the autonomous unit.

8
 Figure 7 Smart solar panel cleaning System[22]
2.1.2 Current Practices
Mostly cleaning of solar panels is done in two ways: either using water solvent cleaning
agents or without water using special brushes. Nowadays most solar panel practices involve
a robotic system that incorporates above mentioned practices. The selection of practices
may change depending upon the location, and seasonal variations in the area where solar
plants are deployed.
1. Water-based robotic cleaning system
These systems use water as a primary component for cleaning solar panels. Water-
based systems are suited in urban and tropical climates where is abundantly
available. The Serbot Swiss innovations developed a cleaning mechanism for
mobile deployment onto SPV panels. This system cleans through a rotating brush
and demineralized water. The designed system is controlled via a radio and is easy
to move from one to place to another.[23] below shows the figure for the gecko
solar panel cleaning system.

Figure 8 Gecko solar robot[23]

2. Waterless Robotic Cleaning System
Though water-based is a more effective cleaning method in dry places such as the
desert where there is no availability of water. In waterless cleaning technology, a

9
 brush is used as a dust repellent to wipe the dust from the solar panels. Aerial
Power’s Solar Brush drone is targeting desert regions of the world[24]. Similarly,
the Boson solar panel cleaning robot shown in Figure 8, is a fully automated robot
that can clean the surface of solar panels without the use of water.

Figure 9 Boson Robotic Panel Cleaning System[25]

2.2 Projects Related to Solar Panel Cleaning System.

2.2.1 Automatic Solar Cleaning System For Dust Removal [26]
The project under discussion uses two-step cleaning techniques. Firstly, an exhaust fan
removes dust from solar panels quickly. And the second step involves a wiper with a soft
cloth to wipe the panel surface. The main components used in the project are
Microcontroller Arduino Uno, a Dc gear motor, Buck-Boost Converter, and Motor Drive
Module. In this project, researchers have used four different types of dust to demonstrate
the cleaning process. Such a system is applicable in areas having high environmental dust
such as deserts or semi-arid areas.

10
 Figure 10 Different Types of Sand Used for Experimental Test [26]
2.2.2 Novel Dry Cleaning Machine[27]
The solution proposed in this paper consists of three separate cleaning mechanisms which
are used one after the other in the cleaning process. In the first stage, compressed air is
sprayed on the panel with the help of a nozzle to clean dust and remove sand. Limit switches
are used at both ends of the panel to signal the completion of a specific stage. In the second
stage, the machine moves back to its initial starting point. For this low-density flexib le
polyurethane foam roller is used to wipe the surface of the panel. In the third stage, the
compressor sprays air to remove any dust that remains after the roller moves over its
surface. Figure 10 shows how the system makes the algorithm work and what functions are
performed by the system in a specific order. Figure 11 shows the prototype for a Novel dry-
cleaning Machine for solar panels.

11
Figure 11 Flow Chart for Four Stages Automatic Dry-Cleaning System

12
 Figure 12 Dry Automatic Cleaning System Overview[27]
2.2.3 Cleaning Robot For Solar Panels[28]
The cleaning Robot uses 5 dc motors from which 2 motor acts as horizontal drives along
the row level of solar panels, while 2 motor acts as vertical drives (up and down movement)
on the solar panel. The last dc motor is used to rotate the microfiber component that cleans
and maintains stability in the vertical motion of the robotic system. The ultrasonic sensor
is used to measure the distance traveled by the robot across the panel surface. Figure 12
shows the Prototype design and view of hardware moving on the solar panel.

Figure 13 Cleaning Robot Prototype and Hardware[28]

2.2.4 Smart Solar Photovoltaic Panel Cleaning System
The proposed model in this paper utilizes two different controlling units: A robotic Unit
and Autonomous Unit that communicate through an internet Cloud-based platform known
as UBIDOTS. The robotic unit acts as a slave of the autonomous unit and UBIDOTS
interface directly under the operator’s command. The robotic unit moves the frame across
the solar panel. Two limit switches are used to detect the start and end of solar panels.
The autonomous unit commands the robotic unit to perform the cleaning on the solar panel.
It also contains different sensors to sense humidity, temperature and light intensity, dust
density, and output power to automate the cleaning process. Figure 13 shows the robotic
unit for the above-mentioned system in running condition.

13
 Figure 14 Robotic Unit: Fabricated Prototype[29]

2.3 Limitations
2.3.1 Current Technologies and Practices Limitations
Both current technologies and Practices related to solar cleaning have numerous advantages
and disadvantages. Since many solar cleaning systems are designed according to the
location of solar placements. Some of the limitations are as follows:
1. Most water-based systems are utilizing large amounts of water as a cleaning
solvent. Clean water availability all around the world is becoming a major issue.
2. Traditional cleaning method that requires manual labor and technicians for
cleaning purposes. It is a time-consuming process.
3. Piezoelectric cleaning systems are not viable for large-scale solar farms.
4. Self-cleaning of solar panels is dependent upon rainfall. In areas with less rainfall,
self-cleaning is not viable.
5. Water-Based Robotic systems are moveable on the panel surface, but the length of
the water pipe can be expensive in large solar plants.
6. Waterless robotic systems are expensive to make, and they need to be outfitted
with various components for control.
7. Waterless robotic systems are mounted on solar surfaces and need to be modified
for use in other arrangements of solar panels. (Refer to Figure 8)

2.3.2 Related Projects Limitations

Many engineers and researchers have designed different types of solar cleaning systems.
They have incorporated a variety of solutions and components to design such cleaning
systems. Depending upon different solutions, limitations related to the above projects are
as follows:
1. Using only brushes to clean solar panels is less effective as compared to
water-based systems.
2. Initial investment to deploy such projects on large solar farms or plants is
very high.

14
 3. The robotic technology used in above mentioned projects is new and a lot of
research and development is required.
4. Most of the mentioned projects have no scope for customization. The
designed systems are difficult to change in size as per the type of solar panels.
5. One of the main drawbacks is the reliability and efficiency of solar cleaning
systems as to how many solar panels they can clean each time.
6. When robotic-based cleaning systems are deployed on solar plates, one of the
main factors is the weight of the mechanism.
7. If the weight of the mechanism is high, it will put an unnecessary burden on
solar plates.
8. Fitting of fixed robotic solar cleaning system on the solar plates on a large
scale is a cumbersome process.

2.4 Summary
This chapter discusses the various types of technologies and practices related to the
cleaning of solar panels. It also sheds light on different projects done by engineers and
researchers in the field of solar cleaning systems for R&D. This chapter promulgates the
various solutions and ideas, opinions, and their usage, and advantages in cleaning solar
panels. The chapter comprises the limitations of different technologies and projects as
discussed above. Now the question arises of how we designed our system keeping in view
the drawbacks and limitations of solar cleaning technologies.

15
Chapter 3
DESIGN AND SOFTWARE SIMULATIONS
3.1 Methodology
Every system made since the dawn of the industrial age comprises both mechanical and
electrical components. The systems being designed nowadays are a mixture of power,
Communication, and sensory technology. We also utilized the same technology such as
power modules, controllers, and sensors for measuring data. The system designed for
cleaning of solar panels by us consists of a variety of things ranging from Weight
calculations, Motors torque and speed, type of motor Driver used, and sensor used to
measure data ranging from environmental factors such humidity, Temperature,
atmospheric pressure, etc. and solar panel measurements such as power, Voltage, and
Current. The idea behind getting such data is utilized to decide when to clean the solar
panel. Mechanical actuators such as limit switches will be used for safety purposes.
3.2 Procedures for Design
To start development of the hardware prototype we need some calculations and simulatio ns
of electrical components which are based on the data collected through different research
papers and projects that have already been done. We design our system we decided that a
540W solar panel will be used as a reference for our hardware development. The procedure
followed throughout the project is given below in steps:
1. We start by comparing different systems and designs previously derived by
engineers and researchers.
2. Data taken from literature reviews, research papers, and project reports is
used as a base to form our design.
3. Our system consists of two main sections: one of the parts is electrical which
controls sensors, motors, and water pump. The other part is mechanical which
will consist of a frame structure for the solar panel, a shaft for the wiper,
brush, and pump mechanism.
4. Firstly, we need to design the mechanical parts or systems, for this purpose,
we need calculations based on the type of material used to design the
mechanism.
5. Next step will be an adjustment of the motor with the frame and test runs will
be taken to ensure that the frame stays in balance while moving across the
solar panel.
6. Limit switches will be attached to the system to start and stop the moving
mechanism while it’s moving in the forward and reverse direction.
7. The system will be controlled by an ESP 32 Wi-fi Module with a built-in
microcontroller.
8. Decision of cleaning will be based on the data which will be collected by the
sensor.
9. The data will be monitored in real-time through the internet-based data
server.

16
 10. Data will be shown in an app that will have the cloud base server API
integrated into it for real-time data monitoring.

3.3 Frame Mechanism Design

The first step to designing our system mechanism will be the calculation of various
parameters such as the weight of the frame and how much speed and torque of the motor is
needed to move the system across the panel. To calculate the total weight of the moving
mechanism we need to know the mechanical parts we will need to be fabricated. The
Mechanism includes the following mechanical Parts:
1. Windscreen Wiper weight
2. Roller Brush Shaft
3. Rectangular frame
4. Trolley Tires
5. Water pump weight
6. Battery weight
7. Mechanism Movement Holding Shaft
8. Coupling for tightening Purpose
9. Miscellaneous weight

3.3.1 Weight Calculation

 Water sprinkling Shaft.

Material Used = light aluminum alloy

Density of the material=2650-2800 Kg/m^3

Weight of the Shaft = Density * Volume

Volume of hollow shaft =pi*(R^2 – r^2) *h

= 3.14* ((15mm) ^2 – (30) ^2) * 1034mm
= 4.058 * 10^34 m^3

Weight = density * Volume

= 2650 Kg/m^3 *4.058 * 10^-4 m^3

 Weight of the water sprinkling shaft = 1.075 Kg

17
 Figure 15 3D Model for Water Sprinkling Shaft
 Nylon Brush Shaft

Density is the same as of aluminum alloy= 2650 Kg/m^3

Volume = pi*r^2*h
= 3.14 * (0.01 m)*2 * 1.034 m
= 3.246 * 10^-4 m^3
Weight = Density ^ Volume
= 0.86019 Kg
Approximately Nylon weigh = .600 Kg
 Weight of the Nylon shaft = 0.86019 + 0.600 Kg = 1.5 Kg
Weight may vary if the Nylon material exceeds in quantity.

Figure 16 3D Model Nylon Brush Shaft

 Rectangular Frame
Length = 1134 mm = 1.134 m
Width = 282 mm = .282 m
Depth = height = 40 mm = 0.04 m
Thickness =50 mm = 0.05 m
Material Used For fabrication = Mild steel/ Aluminum alloy

18
 The density of material = 7.850 / 2650 Kg/m^3
Volume of Outer Section V1 = L * W * H
= 1.134 * .282 *0.04
V1 = .0128 m^3
The Volume of the hollow frame = L-2t * W-2t *H-2t
Volume V2= (1.134 – 2*0.05) * (0.282 -2*0.05) * (0.04 – 2*0.05)
Volume V2= .0112 m^3
Total Volume = V1 – V2
= 0.0128 -0.0112
= 0.0016 m^3
Weight of the rectangular Frame = 7850 Kg/m^3 * 0.0016 m^3
= 12.56 Kg (not Suitable for cleaning system)
For aluminum alloy
 Weight of the rectangular frame= 2650 Kg/m^3 * .0016 m^3
= 4.096 Kg

Figure 17 3D Model Rectangular Frame

Due to the large weight which will cause stress upon the edges of the solar panel which
can lead to breakage of the transparent tempered glass coating so we replace the rectangular
frame with a fixed shaft that will be coupled with a Motor mount in which Dc gear motors
will be adjusted for moving the frame across the length of the panel.
 Trolley Tyers weight
These tires are used to move the frame structure on which the solar panel
is placed. So
 Weight of the trolley tires = 4* 0.2 Kg = 0.8 kg

19
 Figure 18 3D Model Trolley Tyers

 Wiper Mechanism Weight

Since we are using two windscreen wipers as one of the media to remove dust
from the panel, so the approximate weight of both 16-inch and 24-inch
windscreen wiper is as follow:
 Weight of the Wiper Mechanism = .6 Kg

Figure 19 3D Model Windscreen Wiper 24 inch

 Motor Mount Weight
Density of Acrylic=1190Kg/m^3

254
mm

254 mm

Figure 20 Diagram for Motor Mount

Volume of Motor Mount = 254 mm *254 mm

20
 = 0.254 *0.254 * 0.006
=3.87096 *10^-4 m^3
Weight of single side Motor Mount = Volume * Density
= 3.87096 *10^-4 m^3 * 1190 Kg/m^3
= 0.5 Kg approx.
For both sides of the solar panel, we will have two Motor Mount of the Same
Weight so
 Total Weight of Motor Mount =1 Kg
 Fixed Shaft Coupled with Motor Mount
The material used = Iron alloy
The density of Iron alloy
Approximate weight of a single Hollow rod= 0.8 kg
For 4 Shaft,
 Total weight of Rods= 4 *0.8 Kg =3.2 Kg

Figure 21 3D Model for Fixed Shafts

 Water Pump Weight
Weight= 250-300 grams
 Battery 12V 14 Ah – 21 Ah
Weight of battery = 2.5 kg
 Miscellaneous Components Weight
These components may include a water pipe, metal coupler and nuts bolts,
and other small accessories attached to the system.
Miscellaneous Weight= 2 Kg

[Link] Total Weight of the Mechanism

Due to stress upon the solar panel surface, we can only put a maximum weight of amount
20 Kg on the solar panel to avoid breakage of tempered glass, so we neglected the battery
and water pump weight. The main reason is that both components are not viable to be
attached to the mechanism in the long term due to the charging of the battery and placement
of the water pump in the water can become an issue.
Total Weight of the Mechanism = weight of Nylon brush Shaft + Weight of Motor
Mount + Weight of fixed Shafts + Weight of Wiper mechanism + Weight of Miscellaneo us

21
parts
Total Weight = 1kg + 0.600Kg + 1.5Kg + 2 Kg
 Total Weight = 8.3 Kg

Initially, when we were designing our system, we neglected some factors such as how
will the system move across the panel. To overcome this problem, we thought to utilize a
chain-Sprocket mechanism. But the problem with such a system was that we could not
incorporate flexibility meaning we could pick and place our system on another panel
without overcoming a large amount of weight due to the rectangular frame, so we shifted
towards a lightweight approach for the development of the hardware prototype.
3.3.2 Torque And Speed Of Motor
Mass m= 8.3 Kg
Assuming no energy loss occurs due to friction we can say that
Force = mass * acceleration
Since the weight is being displaced horizontally, so there will be friction between the
Tyers so the force of friction is given by:
Ff= Coefficient u * Normal Force N
Normal Force N f =w= mg
Nf=8.3Kg * 9.81 m/s^2
Nf=81.53 N
Since rolling friction between rubber and steel is u=0.002
Since we are using 4 rubber tyers to move the mechanism so
Ff = u *N
Ff = 0.002 * 81.53N
Ff = 0.16506 N
Ff = 4 * 0.16506 = .66025 N (due to usage of Four Tyers)
 Now the force that is required to move the weight.

F= m*a +Ff
F= 8.3 * a + 0.66025 N
Suppose the weight travels up to 3 m.
Distance d = vi *t + 1/2*a8t^2
Assume that vi is zero at the initial starting position of the mechanism.
D=1/2*a*t^2
a=2*3m/t^2
if it takes about 5 sec to reach 3m
a=6m/5^2 sec^2
a=0.24 m/s^2
For the Torque of the Motor
Torque=F*r Suppose r=0.3 m
= (8.3 *.24 + 0.66025 N) * 0.3m
 Torque of Motor = 0.8 N.m

22
 For Motor Speed,
Speed = Torque /moment of Inertia I *angular acceleration
I=0. 002Kg.m^2 (Motor Specification)
w=T/(I*alpha)
angular acceleration alpha = a / r
alpha = 0.24 m/s^2 /0.3m
alpha= 0.8 rad/s^2
speed w=0.8 N.m / (0.002 Kg/m^2 * 0.8 rad/s^2)
 speed of motor = 500 rpm

Rpm is divided into four motors that are fitted in motor mounts.
3.4 Software Simulation
In this section, we deal with the simulation work done for various electrical components
such as the Motor driver, limit switches and water pump, etc.
The Simulation tool used for this purpose is Proteus 8.9 Professional which provides an
integrated easy-to-use electronic design suite environment capable of handling small to
large power electronic circuits. It provides a specific analysis of a circuit and gives an
overview of how the circuit will behave in real-life scenarios.
3.4.1 Motor Driver Design
The purpose of this motor driver is to control the speed as well as the direction of the motor
that is attached to the motor mount coupled with fixed shafts. The main idea behind a motor
driver was to control the direction of four motors at the same time meaning that all motor
should change their polarity at the same time frame.

23
 Figure 22 Motor Driver using AT89C51 and IRF840 MOSFET
The motor driver is designed using a h-bridge built using IRF 840 MOSFET whose
switching is controlled IR2112 Voltage regulator. The IR2112 is controlled by the
microcontroller AT89C51. The problem with building the motor driver from scratch was
that all four motors would not change their direction at the same time causing the Moving
mechanism to get stuck. To remove this problem, we opted for a relay-based switching of
polarity so that all motors can synchronize with each other and switch their direction easily.

24
 Figure 23 Relay-Based Motor Direction Change
3.4.2 Limit Switch
The limit was used in our system to detect and stop the moving mechanism at the start and
end points of the solar panel. Limit switches work when external pressure is applied to the
switch state, the moving mechanism moves across the panel and when external pressure is
removed, the moving mechanism stops at either end of the solar panel.

Figure 24 Proteus Symmetric for Limit Switch Using SPDT

25
 Figure 25 Arduino Code for Limit Switch
Figure 23 shows the working of a limit switch when controlled by a microcontroller such
as Arduino Nano or ESP32 Wi-fi Module.
3.4.3 Pump Motor
A pump motor is used to suck water from the water tank and using a pipe mechanis m,
sprinkles the water on the solar panels. The Nylon brush and Wiper mechanism swipe all
the extra grime from the panel surface achieving maximum cleaning accuracy.
Figure 24 Shows how the simulations of the Pump motor work and what type of component
is used to create a controlled water pump.

26
 Figure 26 Pump Motor Proteus Simulation
3.4.4 Ultrasonic Sensor
This sensor was selected for the specific reason that depending upon the distance from the
solar panel and between the solar panel edges we can control the moving mechanism for
greater mobility for cross panel cleaning.

Figure 27 Ultrasonic Sensor Proteus simulation

27
 3.5 Summary
In this chapter, the emphasis was on the design and simulation part of the project. We
calculated the total weight of the Moving mechanism that will clean the solar panel surface.
From the total weight of the system, we calculated the speed and torque of the motor. In
the Software stage, we simulate different electrical components that can be used in our
system to help mitigate control of the device.

28
Chapter 4
Hardware Development and Specification
Hardware development will include the Fabrication of different mechanical parts such as
Motor mount, Solar frame structure, and Cleaning Mechanism. We also discussed the
working of different electrical Components and how they are being used in our cleaning
system.

4.1 System Specification

Overall system specifications are given in Table 2
Parameter Max Value Min value Modality/Platform
Operating Voltage 14V 12V Li-Po battery
Current 10A 7A -
Torque (One Motor) 1.2 N.m 0.8N.m Dc gear Motor
Mass 12 Kg 9 Kg -
Power 140W 84W Electrical Power
Communication - - Wi-fi
Processor Speed 2.4Ghz - 32-bit Quad Core
(Main CPU)
Clock Speed (MHz) 8MHz - ESP 32

Table 2 System Specification Sheet

4.2 System Outline

The system outline gives an overview of the whole system and how it is interconnec ted
with each other. Figure 26 shows the flow diagram of the system, how data is monitored
and how the motor changes polarity due to the motor drives.

29
 Figure 28 System Outline Flow Diagram

4.3 Component Explanation

4.3.1 ESP32 Wi-fi Module

ESP32 is a single 2.4 GHz Wi-Fi and Bluetooth combo chip designed with ultra-low-power
40 nm technology. ESP32 comes with 448 KB ROM and 520Kb SRAM. It also provides
an internal 8MHz clock and an external clock capability of 2-60 MHz for Wi-fi and
Bluetooth Functionality. The built-in microcontroller provides 34 programmable I/O’s, 3
UART pins, and 1 data transmission channel. ESP32 has a wide variety of applicatio ns
ranging from home automation, audio, health care, and wearable electronics.

[Link] ESP32 Usage

ESP32 is used to control the overall performance of the cleaning system. It helps to monitor
sensory data and decide when to clean the solar panel. ESP32 Wi-fi module provides a
gateway between the data and the cloud-based database.

30
 Figure 29 ESP32 Wi-fi and Bluetooth Module[30]

4.3.2 DHT11 Humidity Sensor

It is a low-cost digital sensor for sensing temperature and humidity. This sensor can be
easily interfaced with any microcontroller such as ESP32, Arduino, Raspberry Pi, etc. to
measure humidity and temperature instantaneously. DHT11 is a relative humidity sensor.
To measure the surrounding air, the sensor uses a thermistor and capacitive humid ity
sensor.

[Link] DHT11 Usage

The sensor is used to measure both temperature and humidity and data is transmitted via
ESP32 to Real-time data monitoring (firebase server and Android app).

Figure 30 DHT11 Humidity & Temperature Sensor[31]

31
4.3.3 BMP 280 Atmospheric Pressure Sensor

The BMP 280 is an absolute barometric pressure sensor that provides high accuracy and
linearity as well as long-term stability and high robustness. BMP 280 Sensor has
applications in indoor & outdoor navigation, Weather forecast, etc. BMP 290 has an
operation range of 300- 1100 hectopascal for pressure sensing and -40-85 degree Celsius
for temperature sensing.
We are using the sensor for measuring both temperature and Atmospheric pressure in a
real-time cloud-based data server.

Figure 31 BMP 280 Temperature and Atmospheric Pressure Sensor[32]

4.3.4 LM2596 Buck Converter

LM2596 is a step-down switching regulator with all required active functions. It operates
at 12 V and can provide an output voltage of 3.3 V, 5V, and is adjustable depending on the
version you are using in the circuit. LM2596 can easily drive loads up to 3A with excelle nt
line and load regulation. Applications of LM2596 Buck converter are in an area where there
is a need for a high-efficiency Step-down regulator and positive-to-negative converter.
Our system utilizes the adjustable LM2596 by converting 12 V into 5V which is used for
powering up ESP32 and other sensors.

32
 Figure 32 LM2596 Buck Converter Adjustable[33]

4.3.5 Limit Switch

Limit switches are pressure-based mechanical actuators that are used for starting or
stopping a particular task in a system. It can be used for controlling machinery as part of
the control system, for safety purposes, and as a counter on conveyor belt type systems.
We use a limit in our system to start and stop the cleaning mechanism at either end of
the solar plates.

Figure 33 Limit Switch

4.3.6 12V Relay

A combination of two 12V relays is used to change the polarity of the motor. This is done
to operate all system motors in both reverse and forward modes.

Figure 34 12V Relay[34]

4.3.7 12V Water Pump

It is a 12V 1.5A operated water pump generally used for water coolers and in aquariums
etc. We are using this pump to sprinkle water on the solar panel surface.

33
 Figure 35 Water Pump 12V [35]

4.3.8 Battery Selection

• At Full Load Condition:

• ESP 32 Wi-Fi module
IfL1 = 250mA
• BMP 280 Humidity Sensor
IfL2 = 0.001mA
• DH II Atmospheric Pressure Sensor
IfL3 = 3mA
• Limit Switches
Since we are using 2 switches
IfL4 = 2 x 25mA= 50mA
• Water Pump
5V operating Voltage; IfL5 = 1.5A
• 12V Dc Gear Motors with Helical rotor
IfL6 = 1.2A x 4= 4.8A
• 12V 3A Dc gear roller motor
IfL7 = 3A
Total Current= 9.55A≈10A
So, we will select the battery of 12V 14AH.

34
 Figure 36 12V 7Ah Battery

Since the maximum current is 14 Ah so we used two batteries of 7Ah in parallel to get
the desired current output.

4.3.9 Solar Panel Parameters

To measure the panel voltage and current, we use direct current from the solar panel and
pass it through a dc load to get Current Isc and voltage open circuit voc. For specifica tio n
of solar panels refer to [36]

35
 Figure 37 Solar Panel Current & Voltage Circuit

4.4 Hardware Prototype

The hardware prototype consists of Motor Mount, DC gear motors, a Frame structure for
holding a Solar panel, and a cleaning mechanism fitted on the panel.

4.4.1 Motor Mount

They are fabricated through a laser engraving machine by using acrylic as a base material.
The mount is made in such a way that it can hold two dc gear motors with them at each end
of the solar cleaning mechanism. Figure 36 shows the fabricated model for the Motor mount
top and side view.

36
 Figure 38 Motor Mount Side and Top View

4.4.2 Dc Gear Motor

The Dc gear motor that we used operates at 12V with a maximum full load current of 1.2A.
The cleaning system used four motors of the same specification to control the moveme nt
of the cleaning mechanism in both forward and reverse directions. Figure 37 shows a
different view of the Dc Gear Motor.

Figure 39 Different View of Dc Gear Motor

4.4.3 Rods Shaft With Frame Structure

Rod shafts are used to make a grip between the two motor mounts. The rods are fixed in

37
place with the help of metal couplers. The solar panel is mounted on an iron frame structure
which holds the panel in place at an angle of 26 degrees from the ground.

Figure 40 Shaft rod between Two Motor Mounts

Figure 41 Solar Mounting Frame Structure

4.4.4 Working Prototype

The working prototype consists of a wiper, brush, and water-spraying mechanism attached
to the fixed shafts. The fixed shafts are coupled with the motor mount at both ends of the
solar panel. Water is pumped from the water tank that is attached to the solar frame structure
on which the solar panel is placed at an angle of 26 degrees. The brush shaft is rotatable
with the help of a 12V 3A Dc gear Motor. The wiper mechanism is freely mounted with
the cleaning mechanism in such a way that it is automatically attached its rubbers to the
panel surface. The data is monitored with the help of the ESP32 Wi-fi module which sends
data to an online database server called Firebase. The decision when to clean the panel is
based upon the sensory parameter such as solar panel voltage, current and environme nta l

38
data such as humidity and Temperature, etc.

Figure 42 Working Prototype for Solar Panel Cleaning System

[Link] Firebase Real-Time Database

Firebase real-time database is a cloud-hosted database. Data is stored as JSON and

Synchronized in Realtime to every connected client. On Firebase we have created an
account named Solar panel cleaning System in which we monitor real-time solar panel and
environmental factors such as humidity, temperature, and atmospheric pressure. Figure 42
shows how and where data is being monitored in real-time.

39
 Figure 43 Real-time Database on Firebase Cloud Server

4.5 Summary
In this chapter, we covered the component we used and what are their benefits and usage.
In industries and markets. How we used the component in our system to achieve reliability
and sustainability in the environment. The chapter also covers how Fabricated hardware
parts are made and used in the prototyping of the system. The next chapter will be based
on the results achieved and testing of the prototype and how our results compared to other
people’s work.

40
Chapter 5
RESULT AND ANALYSIS
5.1 Code Explanation
The code given in Appendix A shows how data is being monitored and how the data is used
to make the decision when to clean the solar panel.

Figure 44 Getting Data for Sensors

The solar panel data measuring circuit is connected to analog pins 34 and 35 of ESP32. The
Analog read command converts the analog voltage and current data into digital values and
sends them to the online cloud database. The data is compared with the assigned value to
decide to clean the solar panel.

Figure 45 Code for Checking the data and making Cleaning Decision

41
After getting both environmental type and solar data we assigned a check for one of the
parameters to decide the cleaning process. In the above screenshot, we assigned a humid ity
check for a specific value condition if the condition becomes true then the cleaning
mechanism will start the cleaning process and once the mechanism reaches one end of the
solar panel then it waits for some, and the motor moves the mechanism in the reverse
direction.

5.2 Cleaning System Testing and Analysis

To test whether the system cleans effectively and analyze the accuracy of the cleaning
system. Before starting the test, we need to create a situation by pouring dust and sand over
the panel surface. We also need to check whether the system cleans the panel accurately
with neatness and robustness.

Figure 46 Solar Panel with Dust and Sand particles

42
 Figure 47 Solar panel after cleaning with Mechanism

5.2.1 Analysis Before Cleaning

It is observed that when the solar panel is not cleaned, the solar panel does not produce
the maximum output current due to less absorption of sunlight. Before cleaning the solar
panel, data shows that the main effect will be on the current which produces less energy as
compared to when it is clean.

3.5.2 Analysis After cleaning

It is observed that after cleaning the solar panel, the power generated by the solar panel
has seen a significant increase. The increase in power corresponds to an increase in the
efficiency of the system. Since our solar panel is damaged the system may show
irregularities in data. These are due to the breakage of solar panels.

5.3 Data Analysis

We are comparing different data from the sensors to decide on the best value for each
parameter so the system may know at what instance should cleaning mechanism should be
operated. Manual operation is also a viable solution if data is not being monitored or there
is a crash in internet service.

5.3.1Humidity vs Temperature At daytime and Nighttime

The table below shows the comparison between humidity and temperature during the

43
 daytime Over a period of one month.
Humidity vs Temperature Sensor Data

Dated Temp during daytime (Deg C) Temp during nighttime (deg C) Humidity during Daytime Humidity during nighttime
1-Jun-23 26 17 43 50
3-Jun-23 33 19 41 58
5-Jun-23 36 22 44 62
7-Jun-23 34 23 43 51
9-Jun-23 39 25 67 56
11-Jun-
23 36 24 49 57
13-Jun-
23 39 26 63 58
15-Jun-
23 34 23 47 62
17-Jun-
23 38 28 59 65
19-Jun-
23 39 23 64 65
21-Jun-
23 41 31 68 67
23-Jun-
23 40 27 64 62
25-Jun-
23 42 26 70 65
27-Jun-
23 36 25 48 57
29-Jun-
23 35 24 46 58
Table 3 Humidity Vs Temperature Dataset

44
 Figure 48 Graph Between Humidity Vs temperature at Daytime

Since it is summer season in June, due to high temperatures, humidity varies from 40
% to 70 % during the daytime.

Figure 49 Temperatures vs Humidity during Nighttime

At nighttime due to temperature drops, the humidity stills vary from 40 to 60 percent so we
can say that in normal temperatures day, we can tell our system to decide according to
weather forecast data. It is a huge drawback that during the daytime and nighttime, there is
no huge change in humidity content in the air. So other parameters can make decisions
based on their data.

45
 5.3.2 Voltage and Current Data Comparison

Figure 50 Voltage Vs Current During Daytime

From the graph, the solar panel voltage and current directly vary with each other. When the
solar panel is not cleaned, the straight line will curve downward showing a decrease in
power generation. The solar system will show the maximum power drops when it is in
dusty conditions. Since the amount of dust varies from location to location, the voltage
during peak sunlight will be high. At nighttime solar system goes on standby producing
little to no power so voltage will also drop accordingly as shown in Figure 50.

46
 Figure 51 Voltage Vs Current during Nighttime

5.4 Summary
This chapter gives in-depth details of Arduino code written to control the cleaning
mechanism and monitor sensory data in real-time database on cloud server called Firebase.
The chapter analyses the data that we monitored during a single month. The section
analyses the temperature vs Humidity and Voltage and Current at Daytime and Nighttime.

47
Chapter 6
CONCLUSION
Solar panel cleaning systems play a crucial role in ensuring the optimal performance and
efficiency of solar panels. Regular maintenance and cleaning of solar panels can keep them
free of dust, debris, dirt, and other atmospheric pollutants present in the air. The cleaner the
panel, the more power will be generated by that panel.
The designed system provides more accurate cleaning of solar panels when water is used
as a solvent but when the water pump is turned off, the cleaning mechanism is roughly
wiping the dust towards the edges of the solar panels. To reduce such a drawback, we need
a spongy type of rubber that can easily drag excess dust and dirt away from the panel
surface when operating in forward mode.
The system designed will face water availability issues whenever the water tank reaches a
dead level. The methodology applied to design the system is feasible if the system is
automated and easy to move from panel to panel while cleaning.
The decision-making process is dependent upon real-time cloud-based database. The data
value if they exceed a certain point then the cleaning mechanism will start its operations.
The result achieved during the testing phase of the cleaning mechanism shows that the
system is working without interruptions and may need a tweak to the mechanical model to
further improve cleaning accuracy and optimal speed.

6.1 Recommendations
The system designed has many discrepancies which need to be addressed in the future.
Some Recommendations that future Engineers and Researchers must keep in mind while
building such type of cleaning systems. Some of these recommendations are as followed:
1. Water availability is one of the major issues for water-based robotic systems so,
design a system with less or no water usage.
2. Try to start the new project from scratch and design every part of the mechanical
mechanism. It will give you better understanding of the working of Mechanical
machinery.
3. Introduce new Sensor such as Light intensity, Irradiance Meter, and Power loss
meters etc. to monitor solar system parameters for decision making process of the
cleaning system.

48
4. The system that needs to be designed in the future must need to be cloudy and
fully automated. The system designed has some limitations in its aspect of
automation.
5. Apply new controlling technologies to have better system reliability and
sustainability towards achieving maximum cleaning of dust, debris and dirt from
the solar.

49
REFERENCES
[1] N. F. Zainuddin, M. N. Mohammed, S. Al-Zubaidi, and S. I. Khogali, “Design and
Development of Smart Self-Cleaning Solar Panel System,” 2019 IEEE Int. Conf.
Autom. Control Intell. Syst. I2CACIS 2019 - Proc., no. June, pp. 40–43, 2019, doi:
10.1109/I2CACIS.2019.8825056.
[2] I. Naseem and J. Khan, “Impact of Energy Crisis on Economic Growth of Pakistan,”
Int. J. African Asian Stud. J., vol. 7, pp. 33–43, 2015.
[3] H. Saulat, “Renewable energy resources current status and barriers in their
adaptation for Pakistan Design and Construct Two-Dimensional Super Thin
Materials View project wastewater treatment View project,” no. January 2015, 2017,
[Online]. Available: [Link]
[4] and A. A. Bakhtiar, Farid, “‘A review of solar energy in Pakistan: Current status
and prospects.,’” Sci. Technol. Dev., vol. 36.3 189–1, 2017.
[5] P. H. Shaikh, F. Shaikh, and M. Mirani, “Solar energy: Topographical asset for
Pakistan,” Appl. Sol. Energy (English Transl. Geliotekhnika), vol. 49, no. 1, pp. 49–
53, 2013, doi: 10.3103/S0003701X1301012X.
[6] S. R. Shakeel, J. Takala, and W. Shakeel, “Renewable energy sources in power
generation in Pakistan,” Renew. Sustain. Energy Rev., vol. 64, pp. 421–434, 2016,
doi: 10.1016/[Link].2016.06.016.
[7] A. W. Bhutto, A. A. Bazmi, and G. Zahedi, “Greener energy: Issues and challenges
for Pakistan-Solar energy prospective,” Renew. Sustain. Energy Rev., vol. 16, no. 5,
pp. 2762–2780, 2012, doi: 10.1016/[Link].2012.02.043.
[8] “No TitleHassan, M., Afridi, M. K., & Khan, M. I. (2018). An overview of
alternative and renewable energy governance, barriers, and opportunities in
Pakistan. Energy & Environment, 29(2), 184–203.”, doi: doi.
[Link]
[9] “Trading into a Bright Energy Future,” Trading into a Bright Energy Futur., 2021,
doi: 10.30875/b2e01618-en.
[10] “IRENA. Solar Energy—International Renewable Energy Agency. 2021.”.
[11] R. J. Mustafa, M. R. Gomaa, M. Al-Dhaifallah, and H. Rezk, “Environme nta l
impacts on the performance of solar photovoltaic systems,” Sustain., vol. 12, no. 2,
pp. 1–17, 2020, doi: 10.3390/su12020608.
[12] A. A. S. A. El-gharabawy, “Review on Corrosion in Solar Panels,” Int. J. Smart grid,

50
 vol. 2, no. 4, pp. 3–5, 2018, doi: 10.20508/ijsmartgrid.v2i4.31.g29.
[13] R. Abdollahi, “Impact of wind on strength and deformation of solar photovolta ic
modules,” Environ. Sci. Pollut. Res., vol. 28, no. 17, pp. 21589–21598, 2021, doi:
10.1007/s11356-020-12111-1.
[14] “Sreedath Panat Kripa K. Varanasi ,Electrostatic dust removal using adsorbed
moisture–assisted charge induction for sustainable operation of solar [Link].
Adv.8,eabm0078(2022).DOI:10.1126/sciadv.abm0078”.
[15] M. R. Maghami, H. Hizam, C. Gomes, M. A. Radzi, M. I. Rezadad, and S.
Hajighorbani, “Power loss due to soiling on solar panel: A review,” Renew. Sustain.
Energy Rev., vol. 59, pp. 1307–1316, 2016, doi: 10.1016/[Link].2016.01.044.
[16] “M. R. Maghami, H. Hizam, C. Gomes, M. A. Radzi, M. I. Rezadad, and S.
Hajighorbani, ‘Power loss due to soiling on solar panel: A review,’ Renewable and
Sustainable Energy Reviews, vol. 59, pp. 1307-1316, 2016”.
[17] Https://[Link]/shop-all/solar-panel-cleaning-kit-for-water-fed-
systems/, “No Title”.
[18] “X. Lu, Q. Zhang, and J. Hu, ‘A linear piezoelectric actuator based solar panel
cleaning system,’ Energy, vol. 60, pp. 401-406, 2013.”.
[19] “]P. Vasiljev, S. Borodinas, R. Bareikis, and A. Struckas, ‘Ultrasonic system for
solar panel cleaning,’ Sensors and Actuators A: Physical, vol. 200, pp. 74-78,
2013.”.
[20] “]J. Son et al., ‘A practical superhydrophilic self cleaning and antireflective surface
for outdoor photovoltaic applications,’ Solar Energy Materials and Solar Cells, vol.
98, pp. 46-51, 2012”.
[21] “G. He, C. Zhou, and Z. Li, ‘Review of self-cleaning method for solar cell array,’
Procedia Engineering, vol. 16, pp. 640-645, 2011.”.
[22] “N. Khadka, A. Bista, B. Adhikari, A. Shrestha, and D. Bista, ‘Smart solar
photovoltaic panel cleaning system,’ IOP Conference Series: Earth and
Environmental Science, vol. 463, no. 012121, 2020”.
[23] “[Link]
[24] “Solar Brush. Available: [Link]
[25] “(2020, April 15). Intellegent cleaning robots. Availab le :
[Link]
[26] M. R. Habib et al., “Automatic Solar Panel Cleaning System Based on Arduino for
Dust Removal,” Proc. - Int. Conf. Artif. Intell. Smart Syst. ICAIS 2021, pp. 1555–

51
 1558, 2021, doi: 10.1109/ICAIS50930.2021.9395937.
[27] S. P. Aly, P. Gandhidasan, N. Barth, and S. Ahzi, “Novel dry cleaning machine for
photovoltaic and solar panels,” Proc. 2015 IEEE Int. Renew. Sustain. Energy Conf.
IRSEC 2015, 2016, doi: 10.1109/IRSEC.2015.7455112.
[28] N. Ronnaronglit and N. Maneerat, “A cleaning robot for solar panels,” Proceeding
- 5th Int. Conf. Eng. Appl. Sci. Technol. ICEAST 2019, pp. 1–4, 2019, doi:
10.1109/ICEAST.2019.8802521.
[29] N. Khadka, A. Bista, B. Adhikari, A. Shrestha, and D. Bista, “Smart solar
photovoltaic panel cleaning system,” IOP Conf. Ser. Earth Environ. Sci., vol. 463,
no. 1, 2020, doi: 10.1088/1755-1315/463/1/012121.
[30] “ESP32.”
[31] “DHT11.”
[32] “[Link]
sensors/bmp280/”.
[33] “LM2596.”
[34] W. Y. W. Com, “5Volt Relay datasheet,” pp. 4–5.
[35] “[Link] for-air-cooler-shafi-12-volt-
water-pump-12-volt-dc-water-pump-12v-dc-water-pump- [Link]”.
[36] L. Jinko Solar Co., “Tiger Pro 72HC 550 Watt MONO-FACIAL MODULE,” pp. 3–
4, 2020, [Online]. Available: [Link]

52
APPENDICES
APPENDIX A
CODE
#include <SimpleDHT.h>
int pinDHT11 = 19;
#include <Wire.h>
#include <Adafruit_Sensor.h>
#include <Adafruit_BMP280.h>
SimpleDHT11 dht11(pinDHT11);
#define SEALEVELPRESSURE_HPA (1013.25)
Adafruit_BMP280 bmp; // I2C
//Adafruit_BME280 bme(BME_CS); // hardware SPI
//Adafruit_BME280 bme(BME_CS, BME_MOSI, BME_MISO, BME_SCK); //
software SPI
unsigned long delay time;
#include <WiFi.h>
#include <FirebaseESP32.h>
#include "WiFi.h"
#define WIFI_SSID "project1"
#define WIFI_PASSWORD "11223344"
#define FIREBASE_HOST "[Link]
[Link]"
#define FIREBASE_AUTH
"QVecb3HgQwLomoOioO0aHTaAGdogNrPeng2ws8Jz"
FirebaseData fbdo;
FirebaseData getdata1;
float v=0;
float i=0;
byte temperature = 0;
byte humidity = 0;
void setup() {
[Link](115200);
pinMode(15,OUTPUT);
pinMode(2,OUTPUT);
pinMode(4,INPUT_PULLUP);
digitalWrite(15,1);digitalWrite(2,0);
[Link]("Satart");
[Link](WIFI_SSID, WIFI_PASSWORD);
[Link]("Connecting to Wi-Fi");
while ([Link]() != WL_CONNECTED)
{

53
 [Link](".");
delay(300);
}
[Link]();
[Link]("Connected with IP: ");
[Link]([Link]());
[Link]();
[Link](FIREBASE_HOST, FIREBASE_AUTH);
[Link](true);
[Link](F("BME280 test"));
bool status;
// default settings
// (you can also pass in a Wire library object like &Wire2)
status = [Link](0x76);
if (!status) {
[Link]("Could not find a valid BME280 sensor, check wiring!");
// while (1);
}
[Link]("-- Default Test --");
delayTime = 1000;
[Link]();
}
void loop() {
v=0;
i=0;
for(int a=0;a<1000;a++)
{
v+=analogRead(35);
i+=analogRead(34);
}
v=v/52500.0;
[Link]("V:");[Link](v);
[Link](" I:");[Link](i);
[Link](fbdo,"Volt",v);
[Link](fbdo,"Cur",i);
[Link](fbdo,"Power",(v*i));
DHT11();
printValues();
delay(1000);
if((int)humidity>=55){
[Link]("run");
digitalWrite(15,0);digitalWrite(2,1); for(int a=0;a<=35;a++){delay(1000);}
digitalWrite(15,1);digitalWrite(2,0); for(int a=0;a<=35;a++){delay(1000);}

54
 }

if(digitalRead(4)==0){
[Link]("Manual");

digitalWrite(15,0);digitalWrite(2,1); for(int a=0;a<=35;a++){delay(1000 );

[Link]("Left");}
digitalWrite(15,1);digitalWrite(2,0); for(int
a=0;a<=35;a++){delay(1000);[Link]("Rite");}
}
}
void DHT11(void) {
// start working...
[Link]("=================================");
[Link]("Sample DHT11...");
// read without samples.
int err = SimpleDHTErrSuccess;
if ((err = [Link](&temperature, &humidity, NULL)) != SimpleDHTErrSucce ss)
{
[Link]("Read DHT11 failed, err="); [Link](SimpleDHTErrCode(err));
[Link](","); [Link](SimpleDHTErrDuration(err)); delay(1000);
return;
}
[Link]("Sample OK: ");
[Link]((int)temperature); [Link](" *C, ");
[Link]((int)humidity); [Link](" H");
[Link](fbdo,"Temp",temperature);
[Link](fbdo,"Humi",humidity);
// DHT11 sampling rate is 1HZ.
}
void printValues() {
[Link]("Temperature = ");
[Link]([Link]());
[Link](" *C");

// Convert temperature to Fahrenheit

/*[Link]("Temperature = ");
Serial. Print(1.8 * [Link]() + 32);
[Link](" *F");*/
[Link]("Pressure = ");
[Link]([Link]() / 100.0F);
[Link](" hPa");
[Link]("Approx. Altitude = ");
[Link]([Link](SEALEVELPRESSURE_HPA));

55
 [Link](" m");
[Link](fbdo,"Temp1",[Link]());
[Link](fbdo,"Pressure",[Link]() / 100.0F);
// [Link]("Humidity = ");
// [Link]([Link]());
// [Link](" %");
[Link]();
}

56