Abstract:
This study investigates the impact of soiling on photovoltaic (PV) systems in semi-arid
urban and coastal environments, focusing on the influence of environmental factors such as particulate matter, air quality indices, relative humidity, and atmospheric haze. A comprehensive field analysis revealed that soiling losses exhibit a linear correlation with elevated deciviews, AQI PM2.5, AQI PM10, and particulate matter ≥1 μm, while relative humidity showed an inverse relationship. However, prolonged exposure to humidity caused cementation and caking of dust particles, leading to increased soiling when cleaning was delayed. Monthly soiling trends for September, October, and November demonstrated the highest losses in October due to intensified particulate activity. Coastal sites consistently outperformed urban locations in power output, with optimal tilt angles of 24° and 30° significantly enhancing efficiency.
To address the challenges of soiling, an automated cleaning robot was developed, featuring a high-pressure rolling brush, lightweight aluminium structure, and automated scheduling. The robot effectively removes contaminants while optimizing water and energy usage Economic analysis showed the robot's affordability at USD 226, making it a cost-effective solution for utility scale and commercial PV systems.
This research also developed a predictive model with an R-squared value of 56%,
providing a reliable tool for estimating soiling losses. The findings contribute to improving the performance and maintenance of PV systems in challenging environments, offering practical insights for renewable energy optimization in semi-arid regions.
