Mohammad Reza Maghami
Dust accumulation on photovoltaic (PV) modules is a major factor contributing to reduced power output, lower efficiency, and accelerated material degradation, particularly in arid and industrialized regions. This study presents a comprehensive review and analysis of the influence of dust deposition on PV performance, covering its optical, thermal, and electrical impacts. Findings from global literature indicate that dust-induced efficiency losses typically range from 10% to 70%, depending on particle characteristics, environmental conditions, and surface orientation. Experimental and modeled I–V and P–V characteristics further reveal significant declines in current and power output as soiling levels increase. Through an extensive literature assessment, this paper identifies Machine Learning (ML)-based approaches as emerging and highly effective techniques for dust detection and mitigation. Recent studies demonstrate the integration of image processing, drone-assisted monitoring, and convolutional neural networks (CNNs) to enable automated, real-time soiling assessment. These intelligent methods outperform conventional manual and time-based cleaning strategies in accuracy, scalability, and cost efficiency. By synthesizing current research trends, this review highlights the growing role of ML and data-driven technologies in enhancing PV system reliability, informing predictive maintenance, and supporting sustainable solar energy generation.