v

TABLE OF CONTENTS

CHAPTER TITLE PAGE NO.

NO.

ABSTRACT iii

ACKNOWLEDGEMENT iv

LIST OF FIGURES ix

LIST OF ABBREVIATIONS xi

LIST OF TABLES xii

1 INTRODUCTION 1

1.1 GENERAL STUDY 3

1.2 PROBLEM STATEMENT 4

1.3 MOTIVATION 5

1.4 OBJECTIVE 6

1.4.1 Improving Solar Energy Efficiency 6

1.4.2 Enhancing Panel Maintenance with Automation 6

1.4.3 Reducing Manual Intervention 6

1.4.4 Increasing Energy Output through SmartTracking 7

1.4.5 AI-Based Solar Optimization 7

2 LITERATURE SURVEY

2.1 ADVANCEMENTS IN SOLAR TRACKING SYSTEMS 8

USING AI AND GPS-BASED ALGORITHMS

 vi

CHAPTER TITLE PAGE NO.

NO.

2.2 STUDYING THE EFFECTS OF DUST ACCUMULATION. 9

ON PV PERFORMANCE AND CLEANING AUTOMATION

2.3 INTEGRATING AI AND MACHINE LEARNING FOR 10

. OPTIMIZATION OF SOLAR POWER SYSTEMS

2.4 SMART CLEANING TECHNOLOGIES FOR 10

MAINTAINING SOLAR PANEL EFFICIENCY

2.5 SUMMARY 11

3 METHODOLOGY 13

3.1 EXISTING METHODS 13

3.2 PROPOSED METHODS 13

3.3 HARDWARE DESIGN 15

3.3.1 ESP32 Microcontroller 16

3.3.2 Dust Sensor (GP2Y1010AU0F) 17

3.3.3 Driver Circuit (L293D) 17

3.3.4 LDR (LM393) 18

3.3.5 LCD Display (16 * 2 inches) 19

3.3.6 Power Supply 20

3.3.7 Voltage Sensor 20

3.3.8 Solar Panel 21

3.4 SOFTWARE DESIGN 22

3.4.1 Python 22
 vii

CHAPTER TITLE PAGE NO.

NO.
3.4.2 Machine Learning 23

3.4.3 Machine Learning Operations 23

3.4.4 Python Libraries 24

3.4.5 Pandas 24

3.4.6 Scikit-learn 25

3.4.7 Matplotlib 25

3.4.8 Seaborn 25

3.4.9 Gradio 26

3.4.10 Algorithm 26

3.4.10.1 Dust Regression plot 26

3.4.10.2 Voltage Regression Plot 28

3.4.10.3 Correlation Heatmap 29

3.4.11 Supervised and Unsupervised Learning 32

3.4.11.1 Classification 32

3.4.11.2 Regression 32

3.4.11.3 Clustering 33

4 IMPLEMENTATION AND WORKING 34

4.1 IMPLEMENTATION - HARDWARE 34

4.1.1 ESP32 Microcontroller Integration 35

4.1.2 Sunlight & Performance Detection System 36

4.1.3 Dust Detection & Analog Signal Conversion 37

 viii

CHAPTER TITLE PAGE NO.

NO.
4.1.4 Power and Motor Control System 38

4.1.5 Structural Support and Mounting System 39

4.1.6 Gear Mechanism and Cleaning Brush System 40

4.1.7 Motor Assembly for Tracking and Cleaning 41

4.2 IMPLEMENTATION - SOFTWARE 42

4.2.1 Sensor Data Acquisition and Processing 42

4.2.2 Solar Panel Tracking and Position Control 43

4.2.3 Dust Detection and Automated Cleaning Activation 43

4.2.4 Data Transmission to Web Dashboard 44

4.2.5 AI-Based Optimization Using Linear Regression 44

4.2.6 Web Dashboard Features and User Interaction 45

4.2.7 Continuous Optimization and Scalability 45

5 RESULTS AND DISCUSSION 46

5.1 GENERATED DATASET 49

6 CONCLUSION AND FUTURE WORK 50

6.1 CONCLUSION 50

6.2 FUTURE WORK 51

REFERENCES 52
 ix

LIST OF FIGURES

Figure Title Page No.

No.

3.1 Block Diagram for Smart Solar Panel Cleaning System 14

3.2 Circuit Diagram for Smart Solar Panel Cleaning System 15

3.3 ESP32 Microcontroller 16

3.4 Dust Sensor 17

3.5 Driver Circuit 18

3.6 LDR 18

3.7 LCD Display 19

3.8 Power Supply 20

3.9 Voltage Sensor 21

3.10 Solar Panel 21

3.11 Dust Regression Plot 27

3.12 Voltage Regression Plot 28

3.13 Correlation Heatmap 30

4.1 Electrical Setup 34

4.2 Mechanical Setup 34

4.3 ESP32 Microcontroller Integration 36

4.4 LDR with Voltage Sensor 37

4.5 Dust Sensor & ADC Converter 38

4.6 Low & High Voltage Driver Circuit 39

4.7 Mounting Frame with Pipe Structure 40

 x

4.8 Yellow Pinion Gear with Microfibre Cleaning Brush 41

4.9 12V Geared Motor and Wiper Motor 42

5.1 AI-powered solar panel dashboard 46

 xi

ABBREVIATIONS

ADC Analog to Digital Converter

API Application Program Interface
CSV Comma Separated Values
DS Dust Sensor
IOT Internet of Things
LCD Liquid Crystal Display
LDR Light Dependent Resistor
MCB Microfibre Cleaning Brush
MCU Microcontroller Unit
ML Machine Learning
MPPT Maximum Power Point Tracking
PWM Pulse Width Modulation
SP Solar Panel
VS Voltage Sensor
 xii

LIST OF TABLES

TABLE NO. TITLE PAGE NO.

5.1 IOT Dashboard for Realtime 46

webserver readings

5.2 AI-powered solar panel analysis dashboard 48

Displaying voltage & dust predictions

5.3 Generated Dataset 49