Open AccessVisible defects detection based on UAV‐based inspection in large‐scale photovoltaic systems
Author(s) -
Li Xiaoxia,
Yang Qiang,
Chen Zhebo,
Luo Xuejing,
Yan Wenjun
Publication year - 2017
Publication title -
iet renewable power generation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.005
H-Index - 76
ISSN - 1752-1424
DOI - 10.1049/iet-rpg.2017.0001
Subject(s) - photovoltaic system , computer science , fault detection and isolation , artificial intelligence , computer vision , scale (ratio) , matching (statistics) , real time computing , feature (linguistics) , engineering , linguistics , statistics , physics , philosophy , mathematics , quantum mechanics , electrical engineering , actuator
The asset assessment and condition monitoring of large‐scale photovoltaic (PV) systems spanning over a large geographical area has imposed urgent challenges and demands for novel and efficient inspection paradigm. In this study, an automatic UAV‐based inspection system is presented and implemented for asset assessment and defect detection for large‐scale PV systems. Two typical visible defects of PV modules, snail trails and dust shading, are characterised and the defect detection through image processing algorithms based on first order derivative of Gaussian function and feature matching is carried out for the aerial PV module images captured by visible light cameras. The functionality of the developed unmanned aerial vehicle (UAV)‐based inspection system can be easily extended with more advanced fault detection algorithms and different forms of sensing devices (e.g. infrared thermal camera) for specialised inspection tasks. Such UAV‐based imaging can carry out a variety of inspection and condition monitoring tasks in PV systems spanning over a large geographical area in an autonomous or supervised fashion with significantly promoted efficiency in comparison with conventional methods.