Comprehensive overview of photovoltaic plants operations and maintenance using UAV-based IR imaging and advanced image analysis techniques

• An in-depth review of thermographic PV plants using UAVs is conducted. • A comparison of the characteristics of UAVs and thermal cameras is presented. • Several computing methods used in the literature are discussed. • A description of the failure types and their classification is provided. • An absence of standardized evaluation criteria is identified in this area. With the rapid expansion of photovoltaic (PV) energy production and the consequent growth of large-scale solar plants, traditional manual operation and maintenance (O&M) practices for managing solar plants will become impractical due to their high cost and complexity. The effective management of these solar plants now demands the integration of advanced technologies to ensure enhanced efficiency, flexibility, and safety. In this paper, we review the integration of Unmanned Aerial Vehicles (UAVs), commonly referred to as drones, equipped with thermal and infrared (IR) cameras, alongside sophisticated algorithms for data analysis. Individually, the domains of PV, UAVs, machine learning, and fault analysis exhibit significant growth potential. This study investigates how their convergence can revolutionize O&M practices within the PV sector. By leveraging infrared cameras mounted on UAVs programmed to follow predefined flightpaths allowing for automated image recognition and failure detection, the complexity and resource requirements of maintaining large-scale solar plants can be substantially reduced. Our work encompasses a comprehensive overview of this interdisciplinary field, with a specific focus on UAVs and IR hardware, advanced algorithms, and the identification of common defects. Furthermore, we conduct a meticulous analysis of the existing literature to elucidate the untapped potential and synergies within this specialized domain.