Document Type : Review Paper
Authors
1 Mathematics, Informatics and Applications Team, National School of Applied Sciences of Tangier, Abdelmalek Essaâdi University, Tetouan, Morocco
2 Intelligent Systems and Applications Laboratory, Moroccan School of Engineering Sciences, Tangier, Morocco.
Abstract
The growing emphasis on renewable energy has led to a notable rise in photovoltaic (PV) installations, highlighting the need for advanced diagnostic methods to enable efficient maintenance and performance optimisation. This paper offers a comprehensive overview of recent developments in the use of artificial intelligence (AI) for image-based fault detection and diagnosis in PV systems with a particular focus on the influence of modality and task design on the performance and real-world deployability. First, a bibliometric analysis was performed to trace the field’s evolution and structural trends, followed by an in-depth technical synthesis organized by fault type, learning paradigm, and application goal. The bibliometric data were sourced from Scopus and Web of Science (WoS). Using Bibliometrix/Biblioshiny, we observe rapid growth in publication activity, increasing collaboration, and a thematic evolution from early classification-oriented studies toward more deployment-relevant pipelines. Consequently, this review highlights persistent technical challenges as well as emerging research directions providing a structured reference for both researchers and practitioners and outlines actionable priorities for advancing reliable field-deployable inspection systems.
Keywords
Main Subjects
)