A Novel Graph-Based Spatiotemporal Two-Stream Network Method for Power Transformer Fault Diagnosis With Limited Labeled Data
IEEE Internet of Things Journal · 15.06.2026 · 13 (12) · 27943-27958
DOI: 10.1109/jiot.2026.3680149 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 15.06.2026. Verfasst von Yazhong Zhou und 7 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Gefördert von: Smart Grid-National Science and Technology Major Project, Open Fund Project of the National Key Laboratory of Power Grid Operation Risk Defense Technology and Equipment.
Abstract (Originalsprache)
The development of Internet of Things (IoT) technology promotes the application of a fault diagnosis method based on data-driven in power transformers. However, collecting a large number of labeled fault samples in real-world engineering scenarios is often costly and labor-intensive. To address this challenge, a novel graph-based spatiotemporal two-stream network (GSTSN) is proposed to accurately identify transformer fault conditions under limited labeled data. Specifically, the designed two-stream graph module constructs a spatiotemporal feature graph, incorporating both real label and pseudolabel via a label propagation strategy (LPS). Subsequently, a spatiotemporal feature augmentation graph convolutional network (SFA-GCN) is designed to capture both specific and shared information embedded in the spatiotemporal graph. Finally, an adaptive feature fusion module based on the minimum redundancy maximum relevance (mRMR) principle is developed to dynamically and selectively integrate temporal, spatial, and common feature embeddings, achieving reliable diagnostic results with limited supervision. Real vibration data from a 693/400-V test transformer are employed to assess the performance of the GSTSN framework. Experimental results show that under a 5% labeling rate, an average diagnostic accuracy of 96.14% is attained by the proposed method, highlighting its significant applicability in real-world engineering scenarios.
Förderung
- Smart Grid-National Science and Technology Major Project
- Open Fund Project of the National Key Laboratory of Power Grid Operation Risk Defense Technology and Equipment
Zitieren
APA
Zhou, Y., He, Y., Zhang, C., Li, Z., Wang, L., Shao, K., Li, W., & Liu, X. (2026). A Novel Graph-Based Spatiotemporal Two-Stream Network Method for Power Transformer Fault Diagnosis With Limited Labeled Data. IEEE Internet of Things Journal, 13(12), 27943-27958. https://doi.org/10.1109/jiot.2026.3680149
Vancouver
Zhou Y, He Y, Zhang C, Li Z, Wang L, Shao K, et al. A Novel Graph-Based Spatiotemporal Two-Stream Network Method for Power Transformer Fault Diagnosis With Limited Labeled Data. IEEE Internet of Things Journal. 2026;13(12):27943-27958. doi:10.1109/jiot.2026.3680149
BibTeX
@article{Zhou2026,
author = {Zhou, Yazhong and He, Yigang and Zhang, Chenran and Li, Zihao and Wang, Lei and Shao, Kaixuan and Li, Wenyue and Liu, Xiaoyan},
title = {A Novel Graph-Based Spatiotemporal Two-Stream Network Method for Power Transformer Fault Diagnosis With Limited Labeled Data},
journal = {IEEE Internet of Things Journal},
year = {2026},
volume = {13},
number = {12},
pages = {27943--27958},
doi = {10.1109/jiot.2026.3680149}
}
EUVIMED