2

Table of Contents

Abstract ................................................................................................................................................................ 2
1. Introduction ...................................................................................................................................................... 4
1.1 Project Definition ................................................................................................................................... 4
1.2 Project Objectives ................................................................................................................................... 4
1.3 Project Specifications ............................................................................................................................. 4
1.4 Product Architecture and Components ................................................................................................... 5
1.5 Applications ............................................................................................................................................ 5
2. Literature Review ............................................................................................................................................. 6
2.1 Project background ................................................................................................................................. 6
2.2 Previous Work ........................................................................................................................................ 7
2.3 Comparative Study ............................................................................................................................... 10
3. System Design ................................................................................................................................................ 11
3.1 Design Constraints................................................................................................................................ 11
3.2 Design Methodology ............................................................................................................................ 12
3.3 Product Subsystems and Components .................................................................................................. 16
3.3.1Product Subsystem1: Communication … ........................................................................................ 16
3.3.2Product Subsystem2: Automatic and Control … ............................................................................. 16
3.3.3Product Subsystem3: Power … ........................................................................................................ 16
3.4 Implementation ..................................................................................................................................... 16
4. System Testing and Analysis ......................................................................................................................... 20
4.1 Subsystem 1: Mechanical Subsystem ................................................................................................... 20
4.2 Subsystem 2: DC Motors Control ........................................................................................................ 20
4.3 Subsystem 3. Water Pump Control ...................................................................................................... 20
5. Future Work and Expected Final Prototype ................................................................................................... 21
5.1 DC Motors Speed Control ................................................................................................................. 21
5.2 Victron Solar Charge Controller Complete Termination and Setup .................................................... 22
5.3 Raspberry Pi Complete Interference and Data Mining ........................................................................ 23
6. Limitations and challenges ............................................................................................................................. 25
7. Conclusion...................................................................................................................................................... 25
8. References ...................................................................................................................................................... 26
Appendix A: Progress Reports ............................................................................................................................... 27
Appendix B: Bill of Materials ................................................................................................................................ 29
Appendix C: Program Codes ................................................................................................................................. 30

3
 1. Introduction

1.1 Project Definition

There is more than enough solar radiation available around the world to satisfy the demand for
solar power systems. The proportion of the sun’s rays that reach the earth’s surface is enough to
provide for global energy consumption 10,000 times over. On average, each square meter of land
is exposed to enough sunlight to produce 1,700 kWh of power every year. Solar Panel has a huge
effect on our world. It can helps our environment to be better without using other power
generation plants that can harm the environment, but solar power plant needs to be cleaned at
least every 3 days. It generally depends on the country for example in the Middle East, it needs
to be cleaned every day so it will cost so much. There are a lot of techniques for cleaning the
solar panels; our idea is to design a smart solar panel that cleans itself automatically and
remotely in order to maintain a high level of efficiency of the solar panel.

1.2 Project Objectives

1. Design a solar panel cleaning system which can increase the efficiency of solar panels.
2. Increase the use of solar panels.
3. Make the cleaning of solar panels simple and automated.
4. Minimize human intervention.
5. A cleaning system that does not affect the quality of the original solar panel.
6. An environmentally friendly cleaning system.

1.3 Project Specifications

1. The solar panel cleaning system operates automatically and remotely.

2. Increase the efficiency at least by 10%.
3. Recycle the cleaning water.
4. An autonomous mechanism brush to clean the 100 W solar panel.

4
1.4 Product Architecture and Components

Figure 1

Figure 1 shows the subsystems that we are going to implement in our project. We have two
main subsystems which are Mechanical subsystem that is in the left of the block diagram, and
the Control subsystem that is shown in the middle and the right of the block diagram. The
Mechanical subsystem contains the cleaner along with the DC motors and the stand that holds
the solar panel along with the rechargeable battery and the water pump. The Control subsystem
contains the relay, solar charger, Bluetooth module and the microcontroller.

1.5 Applications

• In Solar Power Plants.

• In residential houses that use solar power.

5
 2. Literature Review

2.1 Project background

One of the major issues that people face with the installation and the use of solar panels is the
cost that is involved in it. But the cost can be drastically reduced by increasing the efficiency of
each solar panel and hence reducing the number of solar panels that needs to be installed. Using
less number of solar panels in order to get the required electricity will not only be cost efficient
but will also help in having a positive impact on the environment. In order to improve the
efficiency of the solar panels, there are two main aspects that need to be considered; the first
aspect is the amount of light that falls directly on the solar, and the second aspect is how much of
this light energy is capable of using effectively in order to generate power. The issue that is
faced with the use of solar panels is the dust that forms over it. Dust on the cells of the solar
panels reduces the efficiency of the solar panels to a large extent especially in Saudi Arabia
where dust and sand storms are very common. Hence there will be an automated system that will
periodically clean the solar panels in order to make sure that they perform at peak performance
level. The use of right fabrication and controllers will help in making this project possible.

6
2.2 Previous Work

• Previous Projects 1:

The electrical energy produced by photovoltaic cells depends upon solar irradiance falling on the
top of the panel. To maximize the exposure of photovoltaic cells to solar irradiance, solar voltaic
cells are installed outdoor. In outdoor environment, dust starts to accumulate on the surface of
photovoltaic panels and as a result reduces the effective solar irradiance falling on the surface of
photovoltaic panels. Similarly, other bird droppings, ice, and salt layer on photovoltaic panels
also obstruct solar irradiances falling on solar panel. To tackle these issues, cleaning systems for
photovoltaic cells are used. Surajit et al. (2018) have discussed in detail different types of
cleaning systems that used for cleaning solar system. Among different types of technologies used
to clean solar panels, autonomous cleaning robots have emerged as a leading technology.
Autonomous cleaning robots slide on the surface of the solar panel and clean the solar panel
using soft brushes which don’t scratch the surface of the panels. The head is attached to
motorized trolleys which move it horizontally. The vertical motion of the head is usually
controlled by the belt-driven system. Grando et al. (2017) have also surveyed different types of
solar cleaning projects and technologies. In their analysis, they have discussed the working of
solar brush UAV robot, Ecoppia E4, wash panels, and Nomad cleaning systems. Solar brush
robot is a drone that flies very close to the surface of solar panel and cleans the panel through air
pressure and by gently rubbing its tail on the surface of the panel. The main advantage of using
this type of robot is that very little area of the robot is in direct contact with the panel. This
significantly reduces the chances of damage to the panel. The solar brush UAV robot is shown in
Figure 2.

Figure 2: Solar Brush UAV Panel.

7
 • Previous Projects 2:

Ecopia solar cleaning robot uses a slightly different technique to clean the solar panel. In this
system, a guiding railing is attached over the solar panels as shown in Figure 3. The guiding
railing frame work can move horizontally over the surface of the array of panel. For top to down
movement, the robot moves on the railing frame work. By sweeping its microfiber brushes
connected to the head, Ecopia solar panel cleaning project cleans the surface of the solar array.
The system has its own battery which is charged through its own solar cell. This energy storage
features allows the system to clean the panels in the night. Furthermore, the system can also be
controlled through internet.

Figure 3: Ecopia Solar Cleaning System.

• Previous Projects 3:

Washpanel solar cleaning systems use water and horizontal brush to clean the solar panel. As
shown in Figure 4, a horizontal brush is attached to the moveable assembly of the system. This
brush remains in contact with the panel. When the assembly of the system moves, water
sprinkles through water on panels and brush rub the surface of the panel by removing any mud
on the surface of the panel. Nomad cleaning system is very similar to Ecopia system. The
primary difference is between the types of brush that revolves on the surface of the panel. In
Ecopia system, a small vehicle containing brushes moves on the array. Several pulleys and
motors control the movement of solar cleaning robot. Unlike Ecopia system, nomad system
doesn’t contain many motors, belts, and pulleys. It contains only two motors which are used to
move the brush horizontally. The railing on which the brush rolls usually lies behind the solar
panels as show in Figure 4. When the assembly moves horizontally, the brush revolves around its
axis in a circular motion. By moving in circular motion, the brush sweeps across the surface of
the panels, removing any dust and other blockings. The system is fully automated and
intelligence system that can be controlled remotely and configured for a variety of cleaning
schedules.

8
 Figure 4: Nomad Solar Cleaning System.

• Previous Projects 4:

Sweezey et al. (2009) has produced another kind of robotic solar panel cleaning system which
utilizes a support vehicle in addition to the cleaning head and moving trolleys, as shown in
Figure 5. Similar to NOMAD and Ecopia systems, cleaning head sweeps the brushes on the
surface of the panel. To move the drive system, DC motors of 12 volts are used. The upper and
lower trolleys are capable to move in different positions and direction. This feature allows the
system to clean the panels in square wave pattern.

Figure 5: Support Vehicle Based Solar Panel Cleaning System.

9
2.3 Comparative Study

Looking at the three previous projects, a comparative study was made to determine and quantify the
similarities and the differences with our project by observing communication, attached water system
and water recycling. As it can be seen in Table 1, our project differs from the three previous projects
by using Bluetooth as a communication protocol instead of Wi-Fi. Furthermore, our project has an
nd rd
attached water system in which both the 2 and the 3 previous projects have. In fact, the only
nd
previous project that uses water recycling system similar to ours is the 2 previous project.
However, our project is distinguished with having all of the three
categories.

Table 1: Comparasion Between Three Previous Projects and our Solar Panel Cleaning System.

10
 3. System Design

3.1 Design Constraints

Design constraints are conditions that need to happen for a project to be successful. In this
project, we tried to relate our project to the Engineering Standards as much as we could; and that
was through reading and searching about previous research papers that were conducted by expert
people, and they have the same idea of our project in general. If we start with the body structure
of the system, we can see the materials that we used are suitable for each part. We designed the
cleaner with a body that is made from stainless tail to avoid the iron corrosion reactions because
the corrosion will affect the cleaning process and the movement of the cleaner itself. Also, we
used stainless tail for the water pan for the same reason. We used the aluminum as a stand of the
solar panel and at the same time as a rail of the cleaner to keep the wheels running smoothly.
The second important constraint is the environmental constraint. The water recollecting process
was added in the design for environmental purposes, to save and not waste the water, to reuse the
water in the cleaning process or other such as plant irrigations. The water recollecting depends
on the engineering design like the angle of the panel and the slope of the water. Based on that,
we designed the water pan. In order to increase the reliability of our design, we had to think
about something that makes our system more practical. When we brought the idea of the trolley,
the trolley is carrying the whole system and it has four wheels with its stoppers just to make sure
that it will not move by any condition. We designed the trolley to organize the structure,
especially the circuits' wirings and the other mechanical parts in the system.

11
3.2 Design Methodology

In order to meet all the requirements of the project, it is divided into several stages and phases.
The overall aim of the project is to design a smart solar panel that cleans itself automatically
and remotely.

Phase 1 (Term 1, Design M.):

● The first stage of the project was to do a primary research in order to check if the
project is possible to be made technically.
● The second step that we took is to look into the various sensors, controllers and motors.
● In the next step, we divided the project into categories based on its function. It can be
seen that there are two main subsystems on the project and along with this; the project
requires some amount of manufacturing and design of mechanical parts to hold the
solar panel and the cleaner of the solar panel.
● We started by an AutoCAD sketch of our mechanical structure, so we can build it
easily in real life as shown in Figure 6.

Figure 6: AutoCAD Sketch.

12
Phase 2 (Term 2, ASSE III):

• We calculated the needed power for our automation system to determine the size of the
needed battery for our off-grid system, as shown in Table 2 below:
Table 2: The Daily Energy Usage Calculations of the System.

DC Water Raspberry Charge

APPILANCE
Motor Pump Pi Controller
QUANTITY 4 1 1 1
OPERATION
0.017 0.017 24 24
(Hours/Day)
VOTAGE
12 12 5 12 Total Watt Hours
(V)
per Day
CURRENT
0.5 1.6 1.8 0.02
(A)
Watts
24 19.2 9 0.24
(V*A)
Watt-Hrs
0.4 0.32 216 5.76
(V*A*Hours/Day) 222.48

After calculating the wattage hours per day for the system, we found that it
consumes 222.5 W-Hrs per day. Thus, its wattage consumption equals to 9.27 W as
shown in Equation 1.
54.48 W-Hrs / 24 Hrs = 9.27 W Equation 1

However, according to the performance characteristics shown in Figure 7, the battery

(Solar Gel Acid Battery 12 V 65 AH / 10 HR, 1.8 V / Cell) supplies 10.8 V. So, as it can
be seen in Equation 2, the automation system consumes 0.86 A. Therefore, the battery
will supply the system with maximum daily consumption for 75.73 Hrs long as shown in
Equation 3.

9.27 W / 10.8 V = 0.86 A Equation 2

65 AH / 0.86 A = 75.73 Hrs Equation 3

So, as it can be noticed in Equation 4, the total power consumed by the

automation system of the total battery capacity per day equals to 48%.

[(24+24+0.017+0.017) Hrs / 75.73 Hrs] 100 = 48% Equation 4

13
 Figure 7: The Performance Characteristics of the Solar Gel Acid Battery.

• The control system for our project was divided into three subsystems: DC motors
control, water pump control and charge tracker control in which each subsystem will be
controlled and interfere with each other by Raspberry Pi.
• To control the speed and the direction of the +12V DC motors, Raspberry Pi 3 B+ was
coded and then connected to a motor driver. However, we found that the required speed
and direction control could be achieved by replacing the motor driver with two relays
and connecting the motors in parallel. In fact, this replacement helped us to reduce our
project’s cost.
• The water pump control was designed by using only one relay. However, the amount
of water needed to clean the panel was reached by trial-and-error method. In our
control code, we have tried different times until we found that 5 seconds is applicable.
• We have designed a control circuit for both the DC motor and the Water pump by using
SRD-05VDC-SL-C 4-channel relay. As shown in Figure 8, two relays were controlled for
the four DC motors (forward and backward) and one for the water pump (on or off).

Figure 8: Schematic Diagram of the Dc Motors and Water Pump Control.

14
 • We chose Victron MPPT charge controller for making the cleaning decisions. By mining
wattage data from the charge controller to the Raspberry Pi, we will write a python code that
will detect the dirt on the panel based on three factors: the determined efficient wattage,
time and weather as shown in the following algorithm diagram in Figure 9.

Figure 9: The Cleaning System Algorithm Diagram.

The efficient wattage was founded by calculating an approximate loss then comparing it to our
solar specifications. Therefore, for an off-grid solar system, there is 25% loss caused by global
incident below threshold, irradiance loss, IAM factor on global, soiling loss factor,
temperature, quality and mismatch loss, and ohmic wiring. Thus, by knowing that our solar
panel provides 100 W peak power, a clean solar panel will grant 75 W during the useful hours.
• The charge tracker will be connected and tested.
• All the system will be integrated and tested if they function well.
• Add data acquisition system to take measurements.
• Check the system efficiency.
• The final report will be written.

15
3.3 Product Subsystems and Components

3.3.1 Communication Subsystem: Bluetooth and WIFI to initiate the cleaner manually.

We have chosen Bluetooth over the WIFI to initiate the cleaner manually. Thus, for the
connection between the charge controller and the raspberry pi, Bluetooth analog will be used
so data can be read from the MPPT controller.

3.3.3 Power Subsystem: 100 W Solar Panel.

The 100W solar panel embedded with the Solar Gel Acid Battery (12V 65AH/10HR, 1.8V/Cell)
is sufficient for our project. In fact, the automation system will consume approximately %50 of
the battery if continually operated.

3.4 Implementation

We have implemented the AutoCAD sketch that was shown in Figure 6. So at the end of the first
semester, we could say that we have finished the Mechanical structure, and we have fitted the
Panel on it and we have tested the initial mechanism of the cleaner as shown in Figures 10-12.

Figure 10: Installing the Cleaner that has Four DC Motors and Four Wheels on the Structure.

16
 Figure 11: Fitting the Solar Panel on Board.

Figure 12: Final Sketch.

At the beginning of the second semester, the solar panel position was adjusted on the frame
structure to avoid any sunlight blocking that was faced in the initial fabrications. Also, a
steel gear track was added for the DC motors to smoothen the dynamics of the wheels as
shown in Figures 13 and 14.

17
 Figure 13: Adjusted Frame.

Figure 14: Motor Steel Track.

As it can be seen below in Figure 15, we implemented the control circuit we have designed for the
motors and the pump. Figure 16 shows the Raspberry Pi wiring connection of our system.

19
 4. System Testing and Analysis

4.1 Subsystem 1: Mechanical Subsystem

The mechanical system was tested and we found that it's applicable to our project and strong
enough to hold the solar panel along with the battery and the water pump. Also, the structure is
equipped with wheels, and that gives it more flexibility since anyone can move it from one
place to another.

4.2 Subsystem 2: DC Motors Control

The DC motors control system was tested and verified, this subsystem is very important since the
cleaner will be driven by the motors along the solar panel to clean it. Along with the testings, we
found that the four DC motors are enough to drive the cleaner forward to the bottom end of the
solar panel then backwards to the top end of the solar panel.

4.3 Subsystem 3: Water Pump Control

The water pump control system was accomplished and verified. The water pump can pump the
water on the solar panel through some holes installed in the cleaner, and it can deliver a perfect
amount of water to the panel in such a short time. Also, after the cleaning process, the water that
we used to clean the panel will fall into a path that was adjusted on the structure to collect the
water. From there, recycling techniques will take a place.

20
 5. Future Work and Expected Final Prototype

Our Future work for this project will have three progress stages to reach the final prototype:
1. DC motors speed control
2. Victron Solar Charge Controller complete termination and setup
3. Raspberry Pi complete interference and data mining

5.1 DC Motors’ Speed Control

To control the motors speed, we will use Adafruit DC & stepper Motor HAT as shown in
Figure 17.

Figure 17: Adafruit DC Motor Driver/HAT.

The Adafruit HAT can drive up to 4 DC or 2 stepper motors with full PWM speed control. The
motors are controlled by 4 H-Bridges TB6612 MOSFET drivers with 1.2 A per channel and 3 A
peak current capability with thermal shutdown protection, internal kickback protection diodes.
The driver can run motors on 4.5 VDC to 13.5 VDC Up to 4 bi-directional DC motors with
individual 8-bit speed selection (so, about 0.5% resolution). Therefore, after connecting the 4 DC
motors to this driver, a code was written to control the speed and the direction. The code controls
the motors. We were able to test it and prove it, by changing duty cycle of the enables to have
three different speeds and two directions as will be shown in appendix C.

21
5.2 Victron Solar Charge Controller

Figure 18: Victron Solar Charge Controller.

The Victron Solar Charge Controller will be used to control the solar panel and the battery while
being the source of the data. It features highlights flexible charge algorithm Over-temperature
protection and power de-rating when temperature is high. It provides the user with the voltage
and the efficiency of the solar panel in the time domain.

Figure 19: Victron Solar Charge Controller Physical Interference.

Also, by adding VE.Direct Bluetooth Dongle, the user can display programming, real-time data
and history on phone or tablet. However, these data will be monitored and extracted by the
Raspberry Pi.

22
 6. Working
As climate change and global warming threaten the future of our planet, it is becoming
increasingly crucial to find sustainable ways to fulfill our energy requirements. One of the
most efficient ways of moving towards renewable and non-polluting energy sources is to
generate electricity using solar panels to harness the sun’s energy.

Since they have no moving parts, solar panels are one of the most cost-effective and low-
maintenance ways of generating. Despite all their benefits, the efficiency of solar panels can
plummet if dust, dirt, and grime are allowed to accumulate. If maximum efficiency in power
generation is to be maintained, solar panels need to be cleaned timely. However, manually
cleaning solar panels is hazardous and time-consuming.

This Solar Panel Cleaning Robot aims to maintain the efficiency of Solar power production
by making sure the Solar panels are kept clean without putting humans at risk.

This robot comes equipped with a roller brush and a water sprayer to clean all dirt and
grime from the surface of the panels. The sprayer gets its supply of water through an
onboard tank. The rubber caterpillar tracks ensure that this robot can adhere to the slick
surface of solar panels. This robot operates remotely and wirelessly.

Along with large-scale industrial applications such as dedicated solar power plants, this
robot can also help boost the efficiency of solar panels in smaller applications such as
rooftop solar panels in homes and offices.
Features of Solar Panel Cleaning Robot:

 Maintains efficiency of solar panels by keeping them clean.

 Remote and wireless operation ensures workers aren’t put in danger
 Roller brush cleans all dust, dirt, grime, and debris.
 Equipped with water sprayer supplied with onboard water tank.
 Compact, portable, and user-friendly design.
The solar panel cleaner robot makes use of a water tank with motorized pump along with 4x
DC motors to achieve vehicle motion using caterpillar wheel motion. The robotic vehicle is
built over a metal chassis with a controller circuitry operated over RF wireless remote.

A remote controller is used to wirelessly transmit control movement data to the robotic
vehicle. The controller receives the data and operates the wheel motors in desired directions
to achieve the desired movement. The front brush is fixed to the main chassis front and
operated by a geared DC motor. The front panel also has an integrated water pipe that is
used to drive water for cleaning using a dc pump to the front of the brush. The system thus
allows for easy solar panel cleaning using wireless control.
 7. Conclusion

This project highlights the effect of dust, dirt, pollen, sea salt, and bird droppings on the PV
systems’ efficiency. Dust has a major impact on the efficiency and performance of the solar
panels. The reduction in the peak power generation can be up to 10 to 30%. Power reduction was
observed due to dust accumulation on the panels and this can be improved by using robotic
cleaning method. It has increased Power generation capacity of the solar panels. Easy
maintenance, low cost and less power usage are few advantages of this process. Finally, the
reduction in the peak power generation can also be overcome by using this cleaning system. The
device is lightweight because most of its material is made of aluminium. Comparing the costs of
cleaning by Manual operation and Automatic operation, the cost for automatic cleaning is proved
to be more economic and significantly less cumbersome, particularly, in systems with large
number of solar panels. Frequent and periodical cleaning ensures that the solar panels work
consistently with a good transmittance at all times [

25
 8. References

1. Adinoyi, M. J., & Said, S. A. (2013). Effect of dust accumulation on the power outputs
of solar photovoltaic modules. Renewable energy, 60, 633-636.

2. Shajan K .Thomas ,Shelvin Joseph ,Sarrop T S ,and Sahad Bin Haris" ,Solar Panel
Automated Cleaning (SPAC) System ",in 2018 International Conference on Emerging
Trends and Innovations In Engineering And Technological Research (ICETIETR), 2018.

3. Surajit Mondal ,Amit Kumar Mondal ,Abhishek Sharma, and Vindhya Devalla" ,An
overview of cleaning and prevention processes for enhancing efficiency of solar
photovoltaic panels ",Current Science, vol. 115, no. 6, pp .1065-1077 ,September 2018.

4. Marcel Grando ,Elias Renã Maletz ,Daniel Martins ,and Henrique Simas" ,Robots for
Celaning Photo Voltaic Panels: State of the Art and Future Prospects ",in IX
JORNADAS ARGENTINAS DE ROBÓTICA, Cordoba, Argentina, 2017, pp. 29-34.

5. Andrew Sweezey ,Mark Anderson, Ashton Grandy ,and Jeremy Hastie" ,Robotic Device
for Cleaning Photovoltaic Arrays ",in International Conference on Climbing and
Walking Robots , 2009.

6. Victronenergy.com. 2020. VE.Direct Protocol FAQ [Victron Energy]. [online]

Available at: <https://www.victronenergy.com/live/vedirect_protocol:faq>

7. 2020. CP12650F-X 12V 65Ah(10Hr). 2nd ed. [ebook] Shenzhen Center Power Technology Co.,
Ltd, p.2. Available at: <https://meibat.com/wp-content/uploads/2016/10/CP12650F-X.pdf>

8. 2015. Grid-Connected System: Loss Diagram. 1st ed. [ebook] Seneca: Canadian Solar, p.07.
Available at: <https://ww3.arb.ca.gov/fuels/lcfs/crude-oil/innovative-crude/2016-0804_seneca-
attachments.pdf>

26
 • Bill of Materials:
Table 3: Budget Estimation of the Project.

29
34