The use of AIoT- integrated approaches to detect accumulated dust on solar panels utilizing color recognition-based techniques

This paper presents a novel approach to solar panel cleaning by detecting dust levels using an adaptive neuro-fuzzy inference system (ANFIS) technique interfaced with myRIO microcontroller and integrated with Internet of Things technology (IoT). The accumulation of dust and debris on the surface of solar panels has a significant impact on their performance, resulting in lower efficiency. When the concentration of dust on the surfaces of photovoltaic panels exceeds a specific threshold (10 g/m2), the new proposed technique detects it and cleans it automatically. The detection relies on color recognition technology using the RGB color sensor (TCS34725). The neuro-fuzzy inference model trained 80 % of the data, achieving a significant efficiency of more than 90 %. Furthermore, the integration of ANFIS and (IoT) technologies helped to estimate the possibility of dust accumulation on the surface of the solar panel, allowing for optimized cleaning process. This improves the functionality of the proposed system and allows for greater scalability in future applications. The remote monitoring approach significantly reduces the need for on-site personnel intervention. As a result, cleaning costs are reduced, and system performance is improved.