Research Progress in Electric Power Vision Technology for Photovoltaic Module Fault Identification
Solar RRL · 29.03.2026 · 10 (6)
DOI: 10.1002/solr.202500846 ↗Kurz gefasst
Fachartikel in der Zeitschrift Solar RRL, online erschienen am 20.03.2026. Verfasst von Yuan Jing und 6 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 78 Literaturangaben.
Abstract (Originalsprache)
As a critical element of clean energy systems, fault detection in photovoltaic modules plays a pivotal role in ensuring both energy efficiency and safety. Electric Power Vision Technology integrates computer vision and artificial intelligence to deliver an efficient, real‐time solution for identifying photovoltaic faults. This article first provides an overview of Electric Power Vision Technology and the primary types of faults in photovoltaic modules. It then outlines the two key processes involved in fault detection using this technology: image processing and fault identification. The article subsequently offers a comprehensive review of fault detection methods utilizing visible light (RGB color space), infrared (IR), and electroluminescence (EL) images, as well as multimodal fusion. It analyzes the advantages and limitations of each approach, concluding with a forward‐looking assessment of these technologies.
Zitieren
APA
Jing, Y., Yang, L., Zhang, R., Wang, J., Liao, J., Lv, X., & Li, H. (2026). Research Progress in Electric Power Vision Technology for Photovoltaic Module Fault Identification. Solar RRL, 10(6). https://doi.org/10.1002/solr.202500846
Vancouver
Jing Y, Yang L, Zhang R, Wang J, Liao J, Lv X, et al. Research Progress in Electric Power Vision Technology for Photovoltaic Module Fault Identification. Solar RRL. 2026;10(6). doi:10.1002/solr.202500846
BibTeX
@article{Jing2026,
author = {Jing, Yuan and Yang, Liru and Zhang, Ruijia and Wang, Jing and Liao, Jia and Lv, Xiangxue and Li, Haimin},
title = {Research Progress in Electric Power Vision Technology for Photovoltaic Module Fault Identification},
journal = {Solar RRL},
year = {2026},
volume = {10},
number = {6},
doi = {10.1002/solr.202500846}
}
EUVIMED