EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Power transformer online monitoring and state assessment method based on 2DCNN-BiLSTM multi-source feature fusion

Wenbo Zhang iD, Li Liu, Chen Chen, Songze Wei, Ying Zhang, Zhiyuan Huo, Dongxing Li, Hong Yin

PLOS One · 20.04.2026 · 21 (4) · e0345949

DOI: 10.1371/journal.pone.0345949 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 20.04.2026. Verfasst von Wenbo Zhang und 7 weiteren Autor:innen. Beteiligte Einrichtung: State Grid Jibei Electric Power Co., Ltd, Luanping County Power Supply Branch Company, Chengde, China. Hauptthemen laut MEDLINE: Electric Power Supplies. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Technology Project of Research and Application of Transformer Panoramic Monitoring and Operation Status Evaluation for State Grid Chengde Luanping County Power….

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As key power system equipment, real-time monitoring and accurate assessment of power transformers’ operating status are critical for grid safety. To address deficiencies in existing methods regarding multi-source heterogeneous data fusion, unstructured information processing, and small-sample fault recognition, this paper proposes a multi-source deep feature extraction and fusion method combining two-dimensional convolutional neural networks (2D-CNN) and bidirectional long short-term memory networks (BiLSTM) for online transformer status monitoring. A multi-channel sensor platform collects key variables (e.g., partial discharge acoustic signals, voltage, current, oil-dissolved gas concentration) to acquire and standardize spatial and temporal features. A dual-channel model is designed: 2D-CNN extracts spatial features from images, BiLSTM captures temporal dependencies, with an attention mechanism weighting and fusing these features, followed by a fully connected layer. A Softmax classifier with ensemble learning performs state discrimination to enhance stability and generalization. Experiments using real data from east China power grid Dongwu Station transformers for classifying five typical states show the method outperforms traditional and single-modal deep models in accuracy, achieving effective transformer status monitoring.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Electric Power SuppliesConvolutional Neural NetworksLong Short Term MemoryAlgorithms

Autor:innen und Institutionen

  • State Grid Jibei Electric Power Co., Ltd, Luanping County Power Supply Branch Company, Chengde, China.
  • State Grid Chengde Power Supply Company, Chengde, China.
  • Chengde Haoyuan Electric Power Installation Group Co., Ltd, Chengde, China.

ORCID-Profile: Wenbo Zhang

Förderung

  • Technology Project of Research and Application of Transformer Panoramic Monitoring and Operation Status Evaluation for State Grid Chengde Luanping County Power…

Zitieren

APA

Zhang, W., Liu, L., Chen, C., Wei, S., Zhang, Y., Huo, Z., Li, D., & Yin, H. (2026). Power transformer online monitoring and state assessment method based on 2DCNN-BiLSTM multi-source feature fusion. PLOS One, 21(4), e0345949. https://doi.org/10.1371/journal.pone.0345949

Vancouver

Zhang W, Liu L, Chen C, Wei S, Zhang Y, Huo Z, et al. Power transformer online monitoring and state assessment method based on 2DCNN-BiLSTM multi-source feature fusion. PLOS One. 2026;21(4):e0345949. doi:10.1371/journal.pone.0345949

BibTeX
@article{Zhang2026,
  author = {Zhang, Wenbo and Liu, Li and Chen, Chen and Wei, Songze and Zhang, Ying and Huo, Zhiyuan and Li, Dongxing and Yin, Hong},
  title = {Power transformer online monitoring and state assessment method based on 2DCNN-BiLSTM multi-source feature fusion},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {4},
  pages = {e0345949},
  doi = {10.1371/journal.pone.0345949}
}
Kostenlos starten →