Journal article
Active federated transfer algorithm based on broad learning for fault diagnosis
Measurement : journal of the International Measurement Confederation, Vol.208, 112452
01/2023
DOI: 10.1016/j.measurement.2023.112452
Abstract
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.
Details
- Title: Subtitle
- Active federated transfer algorithm based on broad learning for fault diagnosis
- Creators
- Guokai LiuWeiming ShenLiang GaoAndrew Kusiak
- Resource Type
- Journal article
- Publication Details
- Measurement : journal of the International Measurement Confederation, Vol.208, 112452
- DOI
- 10.1016/j.measurement.2023.112452
- ISSN
- 0263-2241
- eISSN
- 1873-412X
- Grant note
- DOI: 10.13039/501100002855, name: Ministry of Science and Technology of the People's Republic of China; DOI: 10.13039/501100004543, name: China Scholarship Council, award: 201906160078; DOI: 10.13039/501100012166, name: National Key Research and Development Program of China, award: 2022YFE0114200; DOI: 10.13039/501100013804, name: Fundamental Research Funds for the Central Universities; DOI: 10.13039/501100001809, name: National Natural Science Foundation of China, award: 52188102
- Language
- English
- Date published
- 01/2023
- Academic Unit
- Nursing; Industrial and Systems Engineering
- Record Identifier
- 9984358158802771
Metrics
5 Record Views