Decentralized Algorithm for Cell-PSC Detection on Cooperative Neighboring PV Panels Based on Real-Time Voltage Measurements

This article addresses the current limitations in fault detection and diagnosis (FDD) for photovoltaic (PV) systems, which often overlook partial-shading conditions (PSCs) at the individual cell level. It highlights the need for innovation in Cell-PSC detection and identifies gaps in the Internet of Things (IoT)-based systems for distributed architectures and local processing algorithms. In response, this article presents a novel decentralized edge-computing algorithm for Cell-PSC detection at the PV panel level, utilizing real-time voltage comparisons from neighboring PV panels within the same string. Validated through simulation and real-world experimentation, the algorithm operates in all weather conditions. The proposed plug-and-play solution, requiring no specific training, shifts from traditional artificial intelligence (AI) and machine learning (ML) methods to a more efficient edge-computing paradigm, reducing computational costs while enhancing performance through decentralized processing.