A Data-Driven IoT Framework for Intelligent Soiling Diagnosis and Automated Cleaning of Photovoltaic Systems

Abstract

Photovoltaic (PV) systems operating in dusty and high-pollution environments experience significant energy losses due to surface soiling, leading to reduced efficiency and increased maintenance requirements. This paper proposes an intelligent IoT-based PV cleaning and monitoring framework designed to automatically detect and mitigate soiling effects while enabling real-time performance supervision. The system integrates voltage and current sensing modules, environmental monitoring, an embedded microcontroller unit, and cloud-based data logging to compute key performance indicators, including the Performance Ratio (PR). A comparative PR analysis algorithm is implemented to identify abnormal efficiency drops and trigger an automated cleaning mechanism. The developed prototype was experimentally evaluated under controlled and outdoor conditions to assess detection accuracy, power recovery capability, system responsiveness, and remote monitoring reliability. Results demonstrate measurable improvement in energy yield and enhanced operational transparency through continuous cloud connectivity. The proposed solution offers a scalable, low-cost, and energy-efficient approach for improving PV system sustainability, particularly in regions prone to heavy dust accumulation.

Introduction

The growing demand for reliable and affordable electricity, along with rising energy costs and environmental concerns, has increased the adoption of solar photovoltaic (PV) systems, especially in developing countries. Solar energy is a clean and sustainable alternative to fossil fuels, but the efficiency of PV panels is significantly affected by environmental factors. One major issue is soiling, where dust, dirt, and other particles accumulate on the panel surface, blocking sunlight and reducing energy output. Studies show that this can reduce efficiency by 10% to over 40%, particularly in dry and dusty regions. Traditional cleaning methods using water and brushes are labor-intensive, costly, and unsafe for large installations, highlighting the need for automated solutions.

To address this problem, the study proposes an IoT-based solar panel maintenance system that monitors panel performance, detects efficiency losses caused by dirt accumulation, and automatically initiates cleaning while allowing remote monitoring. The system integrates sensors, embedded controllers, and internet connectivity to improve energy production, reduce maintenance costs, and enhance system reliability.

Previous research has explored the impact of soiling and proposed solutions such as robotic cleaners, automated water-based systems, and intelligent monitoring techniques using machine learning and fuzzy logic. Some studies also introduced IoT-based monitoring platforms. However, many of these solutions focus either on monitoring or cleaning separately, and few provide a complete integrated system that combines monitoring, soiling detection, automated cleaning, and remote data access.

The proposed system consists of four main subsystems:

1. Power Supply Subsystem – Uses a 200W solar panel, MPPT charge controller, lithium-ion battery, and DC–DC buck converter to provide stable power to the system.

2. Sensing and Measurement Subsystem – Measures panel voltage and sunlight intensity using a voltage divider circuit and an LDR sensor to assess panel performance.

3. Control and Processing Subsystem – Uses an ESP32 microcontroller to process sensor data, evaluate performance using a performance ratio algorithm, decide when cleaning is needed, and transmit data via Wi-Fi for IoT monitoring.

4. Cleaning and Actuation Subsystem – Uses a 12V geared DC motor, motor driver, rail mechanism, and soft brush to physically clean the panel when efficiency drops beyond a defined threshold.

The cleaning process is automated and controlled using limit switches and safety algorithms to prevent mechanical faults. Cleaning is triggered only when sunlight is sufficient and the system detects a performance loss of about 25%, ensuring energy-efficient operation and avoiding unnecessary cleaning cycles.

System testing under the climatic conditions of Kaduna Polytechnic, characterized by high sunlight and dusty Harmattan seasons, demonstrated effective system operation. The sensing subsystem successfully monitored environmental light conditions, and the LDR sensor showed consistent daily variations in light intensity, confirming its reliability for real-time monitoring.

Conclusion

The proposed IoT-based PV cleaning system effectively mitigates performance losses due to panel soiling through a Performance Ratio (PR)-based comparative detection approach that accurately identifies dust accumulation and activates cleaning only when required. Experimental evaluation demonstrated a restoration of approximately 20–40% of lost output power, confirming the technical effectiveness of the intelligent cleaning mechanism under varying environmental conditions. The system operates autonomously with integrated safety controls to prevent overcurrent, mechanical obstruction, and unnecessary actuation, thereby ensuring reliable and secure operation. In addition, real-time cloud-based monitoring provides continuous tracking of electrical parameters and PR values, enabling remote diagnostics and data-driven maintenance. Overall, the integration of IoT-enabled monitoring with adaptive automated cleaning significantly enhances energy yield, operational reliability, and long-term cost efficiency for residential and commercial photovoltaic installations.

References

[1] M. Suleiman, D. K. Joshi, B. Muhammad, and I. Bello, “Performance improvement of PV modules under the effect of partial shading using bil-bot wiping mechanism,” Int. J. Recent Eng. Res. Dev., vol. 9, no. 4, pp. 108–116, 2024. [2] A. O. Maka and J. M. Alabid, “Solar energy technology and its roles in sustainable development,” Clean Energy, vol. 6, no. 3, pp. 476–483, 2022. [3] S. Shaik et al., “Experimental analysis on the impacts of soil deposition and bird droppings on the thermal performance of photovoltaic panels,” Case Stud. Therm. Eng., vol. 48, 103128, 2023. [4] M. J. Alshareef, “A comprehensive review of the soiling effects on PV module performance,” IEEE Access, vol. 11, pp. 134623–134651, 2023. [5] T. Pivem et al., “Application of a computer vision method for soiling recognition in photovoltaic modules for autonomous cleaning robots,” Signal Image Process. Int. J., vol. 10, no. 3, pp. 43–59, 2019. [6] J. Sment and A. Zolan, “Status quo and gap analysis of heliostat field deployment processes,” J. Solar Energy Eng., vol. 146, no. 6, 061004, 2024. [7] H. Abdulla, A. Sleptchenko, and A. Nayfeh, “Optimizing cleaning schedules for spatially distributed photovoltaic installations,” Renew. Energy, vol. 248, 122971, 2025. [8] M. S. Yahya et al., “Implementation of a real-time IoT based energy management system,” J. Eng. Res. Rep., vol. 25, no. 10, pp. 19–29, 2023. [9] H. A. Kazem et al., “A review of dust accumulation and cleaning methods for solar photovoltaic systems,” J. Cleaner Prod., vol. 276, 123187, 2020. [10] S. Gochhait et al., “Application of IoT: A study on automated solar panel cleaning system,” in Proc. 4th Int. Conf. Electrical, Control and Instrumentation Eng. (ICECIE), 2022, pp. 1–4. [11] M. Anderson et al., “Robotic device for cleaning photovoltaic panel arrays,” in Mobile Robotics: Solutions and Challenges, pp. 367–377, 2010. [12] M. T. Grando et al., “Robots for cleaning photovoltaic panels: State of the art and future prospects,” Rev. Tecnología y Ciencia, no. 35, pp. 137–150, 2019. [13] L. Zhou, Y. Jiang, X. Zhang, and Y. Zhai, “An AI-based solar panel cleaning system,” J. Renewable Energy, vol. 177, pp. 275–281, 2021. [14] I. J. Khan, M. Akhlaq, and F. Qasim, “Removal of dust particles from solar panels using fuzzy logic,” Pak. J. Emerg. Sci. Technol., vol. 3, no. 1, pp. 1–12, 2022. [15] V. Tiwari, S. Kumari, and P. P. Sahoo, “Monitoring and control of PV microgrid using IoT,” in Proc. 7th Int. Conf. Electrical Energy Systems (ICEES), 2021, pp. 292–298.

Copyright

Copyright © 2026 Ahmad Muhammad Bello, Hamza Musa, Bilyamin Muhammad, Isah Bello. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.