EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fault Diagnosis Method for Centrifugal Pumps in Nuclear Power Plants Based on a Multi-Scale Convolutional Self-Attention Network

Chen Li, Xinkai Liu, Hang Wang, Minjun Peng

Sensors · 05.03.2025 · 25 (5) · 1589

DOI: 10.3390/s25051589 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 05.03.2025. Verfasst von Chen Li und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Nuclear Science and Technology, Harbin Engineering University, Harbin 150001, China. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Abnormal Operating Condition Identification and Auxiliary Emergency Intervention System for Nuclear Power Plants Foundation Project, China national nuclear corporation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The health status of rotating machinery equipment in nuclear power plants is of paramount importance for ensuring the overall normal operation of the power plant system. In particular, significant failures in large rotating machinery equipment, such as main pumps, pose critical safety hazards to the system. Therefore, this paper takes pump equipment as a representative of rotating machinery in nuclear power plants and proposes a fault diagnosis method based on a multi-scale convolutional self-attention network for three types of faults: outer ring fracture, inner ring fracture, and rolling element pitting corrosion. Within the multi-scale convolutional self-attention network, a multi-scale hybrid feature complementarity mechanism is introduced. This mechanism leverages an adaptive encoder to capture deep feature information from the acoustic signals of rolling bearings and constructs a hybrid-scale feature set based on deep features and original signal characteristics in the time–frequency domain. This approach enriches the fault information present in the feature set and establishes a nonlinear mapping relationship between fault features and rolling bearing faults. The results demonstrate that, without significantly increasing model complexity or the volume of feature data, this method achieves a substantial increase in fault diagnosis accuracy, exceeding 99.5% under both vibration signal and acoustic signal conditions.

Schlagwörter

Stichwörter der Autor:innen

acoustics and vibration signalscentrifugal pumpdeep learningfault diagnosismulti-scale hybrid feature

Autor:innen und Institutionen

  • College of Nuclear Science and Technology, Harbin Engineering University, Harbin 150001, China.
  • Fundamental Science on Nuclear Safety and Simulation Technology Laboratory, Harbin Engineering University, Harbin 150001, China.

Förderung

  • Abnormal Operating Condition Identification and Auxiliary Emergency Intervention System for Nuclear Power Plants Foundation Project
  • China national nuclear corporation

Zitieren

APA

Li, C., Liu, X., Wang, H., & Peng, M. (2025). Fault Diagnosis Method for Centrifugal Pumps in Nuclear Power Plants Based on a Multi-Scale Convolutional Self-Attention Network. Sensors, 25(5), 1589. https://doi.org/10.3390/s25051589

Vancouver

Li C, Liu X, Wang H, Peng M. Fault Diagnosis Method for Centrifugal Pumps in Nuclear Power Plants Based on a Multi-Scale Convolutional Self-Attention Network. Sensors. 2025;25(5):1589. doi:10.3390/s25051589

BibTeX
@article{Li2025,
  author = {Li, Chen and Liu, Xinkai and Wang, Hang and Peng, Minjun},
  title = {Fault Diagnosis Method for Centrifugal Pumps in Nuclear Power Plants Based on a Multi-Scale Convolutional Self-Attention Network},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {5},
  pages = {1589},
  doi = {10.3390/s25051589}
}
Kostenlos starten →