Active federated transfer algorithm based on broad learning for fault diagnosis

被引:12
作者
Liu, Guokai [1 ]
Shen, Weiming [1 ]
Gao, Liang [1 ]
Kusiak, Andrew [2 ]
机构
[1] Huazhong Univ Sci & Technol, Natl Ctr Technol Innovat Intelligent Design & Nume, Sch Mech Sci & Engn, State Key Lab Digital Mfg Equipment & Technol, Wuhan 430074, Peoples R China
[2] Univ Iowa, Ind & Syst Engn, 4627 Seamans Ctr, Iowa City, IA 52242 USA
基金
中国国家自然科学基金;
关键词
Active learning; Broad learning; Cross-domain; Fault diagnosis; Federated learning; SYSTEM;
D O I
10.1016/j.measurement.2023.112452
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Federated learning (FL) guaranteeing data privacy is of great interest in decentralized fault diagnosis. However, limited research attention has been paid to the dynamic domain-shift issue due to varying working conditions. This paper proposes an active federated transfer algorithm based on broad learning to address the domain shift issue in FL. First, a central server dispatches a global model to the source clients for collaborative modeling. Subsequently, the global model is initialized with a federated averaging strategy. Next, the initialized global model is used to annotate emerging signals from the target clients based on an active sampling strategy proposed. Finally, an asynchronous update scheme is designed to adapt the global model to the target domain. The performance of the AFTBL algorithm is validated with three datasets, including 24 centralized-and decentralized-modeling tasks. The computational results indicate that the proposed algorithm is more accurate and efficient than the prevalent algorithms.
引用
收藏
页数:10
相关论文
共 34 条
  • [1] Distributed optimization and statistical learning via the alternating direction method of multipliers
    Boyd S.
    Parikh N.
    Chu E.
    Peleato B.
    Eckstein J.
    [J]. Foundations and Trends in Machine Learning, 2010, 3 (01): : 1 - 122
  • [2] Universal Approximation Capability of Broad Learning System and Its Structural Variations
    Chen, C. L. Philip
    Liu, Zhulin
    Feng, Shuang
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (04) : 1191 - 1204
  • [3] Broad Learning System: An Effective and Efficient Incremental Learning System Without the Need for Deep Architecture
    Chen, C. L. Philip
    Liu, Zhulin
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2018, 29 (01) : 10 - 24
  • [4] Federated Transfer Learning for Bearing Fault Diagnosis With Discrepancy-Based Weighted Federated Averaging
    Chen, Junbin
    Li, Jipu
    Huang, Ruyi
    Yue, Ke
    Chen, Zhuyun
    Li, Weihua
    [J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2022, 71
  • [5] Novel Efficient RNN and LSTM-Like Architectures: Recurrent and Gated Broad Learning Systems and Their Applications for Text Classification
    Du, Jie
    Vong, Chi-Man
    Chen, C. L. Philip
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2021, 51 (03) : 1586 - 1597
  • [6] Bearing fault diagnosis based on improved federated learning algorithm
    Geng, DaoQu
    He, HanWen
    Lan, XingChuan
    Liu, Chang
    [J]. COMPUTING, 2022, 104 (01) : 1 - 19
  • [7] Research Review for Broad Learning System: Algorithms, Theory, and Applications
    Gong, Xinrong
    Zhang, Tong
    Chen, C. L. Philip
    Liu, Zhulin
    [J]. IEEE TRANSACTIONS ON CYBERNETICS, 2022, 52 (09) : 8922 - 8950
  • [8] Cross-machine intelligent fault diagnosis of gearbox based on deep learning and parameter transfer
    Han, Te
    Zhou, Taotao
    Xiang, Yongyong
    Jiang, Dongxiang
    [J]. STRUCTURAL CONTROL & HEALTH MONITORING, 2022, 29 (03)
  • [9] DeepFedWT: A federated deep learning framework for fault detection of wind turbines
    Jiang, Guoqian
    Fan, WeiPeng
    Li, Wenyue
    Wang, Lijin
    He, Qun
    Xie, Ping
    Li, Xiaoli
    [J]. MEASUREMENT, 2022, 199
  • [10] Smart manufacturing must embrace big data
    Kusiak, Andrew
    [J]. NATURE, 2017, 544 (7648) : 23 - 25