Photovoltaic module fault detection based on an improved residual-bidirectional connection network

Read full text Summarise with VIRAT
Authors: Daosheng Liu, Fei Xie, Zijian Zhang, Huanjing Qiu
Cited by: 0
References: 4

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Abstract To address the issues of high sensitivity to environmental interference and low accuracy in small-sample scenarios caused by insufficient feature representation in photovoltaic (PV) module fault detection, this paper proposes a PV fault diagnosis model based on a residual bidirectional connection network. This model introduces a bidirectional connection mechanism between multiple layers to enhance the information interaction of features at different scales. Finally, a semi-supervised training mechanism is adopted to enhance the generalization ability in scenarios with small samples. It takes the I-V curve, maximum power point, open-circuit voltage, short-circuit current, and fill factor as input features and employs Z-normalization to achieve feature normalization. Experimental results indicate that this method still exhibits high accuracy and good stability under small-sample conditions.

Fields of Study & Subjects

Photovoltaic System Optimization Techniques Electrical Fault Detection and Protection Fault Detection and Control Systems

Publication details

Journal / source
Journal of Physics Conference Series
Volume
3198
Issue
1
Pages
012033–012033
Publication date
1 March 2026
Language
EN