Efficient fault diagnosis in rolling bearings lightweight hybrid model

被引:0
作者
Yang, Peng [1 ]
Zhang, Bozheng [1 ]
Zhao, Jianda [1 ]
机构
[1] Tianjin Univ Technol, Comp Sci & Technol, Tianjin 300384, Peoples R China
来源
SCIENTIFIC REPORTS | 2025年 / 15卷 / 01期
关键词
Bearing fault detection; LSTM; Transformer; Multi-head attention mechanism;
D O I
10.1038/s41598-025-96285-z
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
To address the issue of low efficiency in feature extraction and model training when traditional deep learning methods handle long time-series data, this paper proposes a Time-Series Lightweight Transformer (TSL-Transformer) model. According to the data characteristics of bearing fault diagnosis tasks, the model makes lightweight improvements to the traditional Transformer model, and focuses on adjusting the encoder module (core feature extraction module), introducing multi-head attention mechanism and feedforward neural network to efficiently extract complex features of vibration signals. Considering the rich temporal features present in vibration signals, a Long Short-Term Memory (LSTM) module is introduced in parallel to the encoder module of the improved lightweight Transformer model. This enhancement further strengthens the model's ability to capture temporal features, thereby improving diagnostic accuracy. Experimental results demonstrate that the proposed TSL-Transformer model achieves a fault diagnosis accuracy of 99.2% on the CWRU dataset. Through dimensionality reduction and visualization analysis using the t-SNE method, the effectiveness of different network structures within the proposed TSL-Transformer model is elucidated.
引用
收藏
页数:14
相关论文
共 50 条
  • [31] Real-Time Diagnosis of Abrupt and Incipient Faults in IMU Using a Lightweight CNN-Transformer Hybrid Model
    Song, Jia
    Chen, Zhipeng
    Li, Wenling
    IEEE SENSORS JOURNAL, 2025, 25 (07) : 12496 - 12510
  • [32] Power Transformer Fault Diagnosis Based On Hybrid Intelligent Algorithm
    Xu, Yong
    Lu, Xiaojuan
    JOURNAL OF APPLIED SCIENCE AND ENGINEERING, 2023, 27 (01): : 1859 - 1866
  • [33] Transformer fault diagnosis method based on the three-stage lightweight residual neural network
    Liu, Hang
    Niu, Ben
    Liu, Zhijian
    Li, Ming
    Shi, Zhiyu
    ELECTRIC POWER SYSTEMS RESEARCH, 2025, 238
  • [34] A lightweight and rapidly converging transformer based on separable linear self-attention for fault diagnosis
    Yin, Kexin
    Chen, Chunjun
    Shen, Qi
    Deng, Ji
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2025, 36 (01)
  • [35] A novel transformer fault diagnosis model based on integration of fault tree and fuzzy set
    Zhang, Kefei
    Guo, Jiang
    Yuan, Fang
    2015 11TH INTERNATIONAL CONFERENCE ON SEMANTICS, KNOWLEDGE AND GRIDS (SKG), 2015, : 112 - 118
  • [36] Neural-transformer: A brain-inspired lightweight mechanical fault diagnosis method under noise
    Wang, Changdong
    Tian, Bowen
    Yang, Jingli
    Jie, Huamin
    Chang, Yongqi
    Zhao, Zhenyu
    RELIABILITY ENGINEERING & SYSTEM SAFETY, 2024, 251
  • [37] Fault Diagnosis of Motor Bearings Based on a Convolutional Long Short-Term Memory Network of Bayesian Optimization
    Li, Zhen
    Wang, Yang
    Ma, Jianeng
    IEEE ACCESS, 2021, 9 : 97546 - 97556
  • [38] A data enhanced algorithm for fault diagnosis of slewing bearings based on times-series generative adversarial networks
    Sun, Li
    Wu, Jun
    Li, Guochao
    Ren, Xiaodie
    Wang, Jinjun
    Wen, Sizhao
    SIGNAL IMAGE AND VIDEO PROCESSING, 2025, 19 (04)
  • [39] Research on Fault Diagnosis Method of Bearings in the Spindle System for CNC Machine Tools Based on DRSN-Transformer
    Li, Xiaoxu
    Chen, Jiaming
    Wang, Jiahao
    Wang, Jianqiang
    Li, Xiaotao
    Kan, Yingnan
    IEEE ACCESS, 2024, 12 : 74586 - 74595
  • [40] An efficient bearing fault detection strategy based on a hybrid machine learning technique
    Khalid Alqunun
    Mohammed Bachir Bechiri
    Mohamed Naoui
    Abderrahmane Khechekhouche
    Ismail Marouani
    Tawfik Guesmi
    Badr M. Alshammari
    Amer AlGhadhban
    Abderrahim Allal
    Scientific Reports, 15 (1)