EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Graph Neural Networks for Fault Diagnosis in Photovoltaic-Integrated Distribution Networks with Weak Features

Junhao Liu iD, Yuteng Huang, Ke Chen, Guojin Liu iD, Jiaxiang Yan, Shan Chen, Yuqing Xie, Yantao Yu iD, Tiancong Huang iD

Sensors · 12.09.2025 · 25 (18) · 5691

DOI: 10.3390/s25185691 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 12.09.2025. Verfasst von Junhao Liu und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, China. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: State Grid Zhejiang Electric Power Co., LTD..

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Effective diagnosis of distribution network faults is crucial to ensuring the reliability of power systems. However, the bidirectional power flow caused by the integration of new energy limits the effectiveness of traditional detection methods. Although data-driven approaches are not restricted by power flow direction, their performance is heavily dependent on the quantity and quality of training samples. In addition, factors such as measurement noise, variable fault impedance, and volatile photovoltaic output complicate fault information. To address this, we present a new fault diagnosis model named the dynamic, adaptive, and coupled dual-field-encoding graph neural network (DACDFE-GNN), which introduces a dynamic aggregation module to assign different weights to reduce noise interference and fully integrates information from observable nodes. On this basis, the coupled dual-field-encoding module is proposed, which encodes topological information and physical–electrical domain information as part of the initial features, thereby capturing fault features and learning the law of feature propagation. The experimental results for the IEEE 34- and IEEE 123-node feeder systems indicate that the proposed model surpasses recent fault diagnosis methods in detection performance, particularly regarding its low training sample rate.

Schlagwörter

Stichwörter der Autor:innen

distribution network fault diagnosisfault informationgraph neural networkslow training sample ratephotovoltaic generation

Autor:innen und Institutionen

  • School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, China.
  • Information and Communication Branch, State Grid Zhejiang Electric Power Co., Ltd., Hangzhou 310008, China.

ORCID-Profile: Junhao Liu, Guojin Liu, Yantao Yu, Tiancong Huang

Förderung

  • State Grid Zhejiang Electric Power Co., LTD.

Zitieren

APA

Liu, J., Huang, Y., Chen, K., Liu, G., Yan, J., Chen, S., Xie, Y., Yu, Y., & Huang, T. (2025). Graph Neural Networks for Fault Diagnosis in Photovoltaic-Integrated Distribution Networks with Weak Features. Sensors, 25(18), 5691. https://doi.org/10.3390/s25185691

Vancouver

Liu J, Huang Y, Chen K, Liu G, Yan J, Chen S, et al. Graph Neural Networks for Fault Diagnosis in Photovoltaic-Integrated Distribution Networks with Weak Features. Sensors. 2025;25(18):5691. doi:10.3390/s25185691

BibTeX
@article{Liu2025,
  author = {Liu, Junhao and Huang, Yuteng and Chen, Ke and Liu, Guojin and Yan, Jiaxiang and Chen, Shan and Xie, Yuqing and Yu, Yantao and Huang, Tiancong},
  title = {Graph Neural Networks for Fault Diagnosis in Photovoltaic-Integrated Distribution Networks with Weak Features},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {18},
  pages = {5691},
  doi = {10.3390/s25185691}
}
Kostenlos starten →