Real Time Sustainable Dust Detection Using AI Driven Image Processing Techniques for Rooftop Solar Installations

View at publisher Summarise with VIRAT
Authors: Y.A. Munasinghe, Isuru Lakmal
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Dust accumulation on solar panels mounted on rooftops is a dangerous phenomenon lowering the energy production level and increasing the operation cost of solar installations. Traditional detection and washing are inefficient, require a lot of work, and are not cost-effective in the long run. The current research proposes an online sustainable dust detection and management system using computer vision and image processing using Artificial Intelligence (AI) by Convolutional Neural Networks (CNNs) and RGB imaging using CCTV. A mini-pilot solar Installation was constructed for image capture of the dust under various conditions. The system has a detection rate of 91% and shows a general average increment of 15% in energy generation through optimal utilization through scheduling of cleaning. The findings support cost-effective production of energy, cost savings on maintenance, and deployable scalability in the case of urban rooftop solar deployments.

Fields of Study & Subjects

Solar Thermal and Photovoltaic Systems Photovoltaic System Optimization Techniques Solar Radiation and Photovoltaics

Publication details

Journal / source
IEEE International Conference on Industrial and Information Systems (ICIIS)
Pages
517–522
Publication date
16 January 2026
Language
EN