A Multi-Task-Based Deep Multi-Scale Information Fusion Method for Intelligent Diagnosis of Bearing Faults

被引:4
|
作者
Xin, Ruihao [1 ]
Feng, Xin [2 ]
Wang, Tiantian [1 ]
Miao, Fengbo [1 ]
Yu, Cuinan [3 ]
机构
[1] Jilin Inst Chem Technol, Sch Informat & Control Engn, Jilin 132022, Peoples R China
[2] Jilin Inst Chem Technol, Sch Sci, Jilin 132022, Peoples R China
[3] Jilin Univ, Sch Software, Changchun 130015, Peoples R China
基金
中国国家自然科学基金;
关键词
fault diagnosis; multi-task; information fusion; multi-layer attention; MODE DECOMPOSITION; TRANSFORM; ATTENTION;
D O I
10.3390/machines11020198
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The use of deep learning for fault diagnosis is already a common approach. However, integrating discriminative information of fault types and scales into deep learning models for rich multitask fault feature diagnosis still deserves attention. In this study, a deep multitask-based multiscale feature fusion network model (MEAT) is proposed to address the limitations and poor adaptability of traditional convolutional neural network models for complex jobs. The model performed multidimensional feature extraction through convolution at different scales to obtain different levels of fault information, used a hierarchical attention mechanism to weight the fusion of features to achieve an accuracy of 99.95% for the total task of fault six classification, and considered two subtasks in fault classification to discriminate fault size and fault type through multi-task mapping decomposition. Of these, the highest accuracy of fault size classification reached 100%. In addition, Precision, ReCall, and Sacore F1 all reached the index of 1, which achieved the accurate diagnosis of bearing faults.
引用
收藏
页数:20
相关论文
共 50 条
  • [31] Bearing fault diagnosis method based on multi-source heterogeneous information fusion
    Zhang, Ke
    Gao, Tianhao
    Shi, Huaitao
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2022, 33 (07)
  • [32] Roller bearing fault diagnosis based on LMD and multi-scale symbolic dynamic information entropy
    Minghong Han
    Yaman Wu
    Yumin Wang
    Wei Liu
    Journal of Mechanical Science and Technology, 2021, 35 : 1993 - 2005
  • [33] Roller bearing fault diagnosis based on LMD and multi-scale symbolic dynamic information entropy
    Han, Minghong
    Wu, Yaman
    Wang, Yumin
    Liu, Wei
    JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY, 2021, 35 (05) : 1993 - 2005
  • [34] Deep multi-scale adversarial network with attention: A novel domain adaptation method for intelligent fault diagnosis*
    Zhao, Bo
    Zhang, Xianmin
    Zhan, Zhenhui
    Wu, Qiqiang
    JOURNAL OF MANUFACTURING SYSTEMS, 2021, 59 : 565 - 576
  • [35] Intelligent bearing fault diagnosis technology based on deep learning and multi-domain decision fusion
    Lin, Shiqi
    Chen, Zhili
    Li, Yupeng
    Meng, Weiying
    Jisuanji Jicheng Zhizao Xitong/Computer Integrated Manufacturing Systems, CIMS, 2024, 30 (10): : 3708 - 3718
  • [36] Intelligent fault diagnosis of robot bearing based on multi-information
    Wu, Ziyue
    Liu, Yang
    Li, Xinxing
    Du, Xingxing
    Wang, Bing
    UPB Scientific Bulletin, Series D: Mechanical Engineering, 2020, 82 (03): : 13 - 28
  • [37] A multi-scale collaborative fusion residual neural network-based approach for bearing fault diagnosis
    Qian, Chen
    Gao, Jun
    Shao, Xing
    Wang, Cuixiang
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2024, 35 (04)
  • [38] Rolling bearing fault diagnosis based on dilated convolution and enhanced multi-scale feature adaptive fusion
    Han K.
    Zhan H.
    Yu J.
    Wang R.
    Zhejiang Daxue Xuebao (Gongxue Ban)/Journal of Zhejiang University (Engineering Science), 2024, 58 (06): : 1285 - 1295
  • [39] Image Dehazing Method Based on Multi-scale Feature Fusion
    Yao, Minghai
    Miao, Qi
    Hao, Qiaohong
    PROCEEDINGS OF THE 2017 3RD INTERNATIONAL CONFERENCE ON ECONOMICS, SOCIAL SCIENCE, ARTS, EDUCATION AND MANAGEMENT ENGINEERING (ESSAEME 2017), 2017, 119 : 2163 - 2166
  • [40] An intelligent fault diagnosis for machine maintenance using weighted soft-voting rule based multi-attention module with multi-scale information fusion
    Xu, Zifei
    Bashir, Musa
    Zhang, Wanfu
    Yang, Yang
    Wang, Xinyu
    Li, Chun
    INFORMATION FUSION, 2022, 86-87 : 17 - 29