EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A CNN Feature Extraction and BKA-Optimized LSSVM Classification Method for Small-Sample Rolling Bearing Fault Diagnosis

Shiyan Sun iD, Yujun Shi, Quan Li, Jiwei Wang iD, Haifeng Lu

Sensors · 14.08.2026 · 26 (16) · 5148

DOI: 10.3390/s26165148 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 14.08.2026. Verfasst von Shiyan Sun und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Electrical Engineering, Xinjiang University, Urumqi 830017, China. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Xinjiang Uygur Autonomous Region, Natural Science Foundation of Xinjiang Uygur Autonomous Region of China, Tianchi Talent Project of Xinjiang Uygur Autonomous Region of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Rolling bearings are indispensable elements in mechanical equipment, and their condition is closely related to system reliability and operational safety. To enhance diagnostic performance with limited samples and reduce the dependence on manual parameter selection, this study proposes a fault diagnosis approach that combines Continuous Wavelet Transform (CWT), Convolutional Neural Network (CNN), Black-winged Kite Algorithm (BKA), and Least Squares Support Vector Machine (LSSVM). The original 1-D vibration signals are first processed by CWT to obtain 2-D time–frequency representations. CNN is then employed to learn deep fault-sensitive features, which are subsequently fed into LSSVM for state classification. To further improve classification performance, BKA is used to automatically search for the optimal LSSVM parameters, with validation accuracy adopted as the fitness criterion. Experiments conducted on three public bearing datasets, namely CWRU, JNU, and SEU, indicate that the proposed method outperforms CNN, CNN-SVM, and CNN-BiGRU under small-sample conditions. In addition, t-SNE results show more distinct feature clusters, while BKA exhibits faster convergence and better global search capability than Particle Swarm Optimization (PSO) and Genetic Algorithm (GA).

Schlagwörter

Stichwörter der Autor:innen

bearing fault diagnosisblack-winged kite algorithmcontinuous wavelet transformconvolutional neural networkleast squares support vector machinesmall-sample diagnosis

Autor:innen und Institutionen

  • School of Electrical Engineering, Xinjiang University, Urumqi 830017, China.
  • Xinjiang E&T Research Center for Operation and Control of New Power System with High Penetration of Wind and Solar Energy, Xinjiang University, Urumqi 830017, China.

ORCID-Profile: Shiyan Sun, Jiwei Wang

Förderung

  • Natural Science Foundation of Xinjiang Uygur Autonomous Region
  • Natural Science Foundation of Xinjiang Uygur Autonomous Region of China
  • Tianchi Talent Project of Xinjiang Uygur Autonomous Region of China
  • Research Startup Fund for Excellent Talents at Xinjiang University

Zitieren

APA

Sun, S., Shi, Y., Li, Q., Wang, J., & Lu, H. (2026). A CNN Feature Extraction and BKA-Optimized LSSVM Classification Method for Small-Sample Rolling Bearing Fault Diagnosis. Sensors, 26(16), 5148. https://doi.org/10.3390/s26165148

Vancouver

Sun S, Shi Y, Li Q, Wang J, Lu H. A CNN Feature Extraction and BKA-Optimized LSSVM Classification Method for Small-Sample Rolling Bearing Fault Diagnosis. Sensors. 2026;26(16):5148. doi:10.3390/s26165148

BibTeX
@article{Sun2026,
  author = {Sun, Shiyan and Shi, Yujun and Li, Quan and Wang, Jiwei and Lu, Haifeng},
  title = {A CNN Feature Extraction and BKA-Optimized LSSVM Classification Method for Small-Sample Rolling Bearing Fault Diagnosis},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {16},
  pages = {5148},
  doi = {10.3390/s26165148}
}
Kostenlos starten →