TY - GEN
T1 - Mechanical Cross-Domain Diagnosis Method Based on Multi-source Weighted Domain Adaptation
AU - Zhang, Jie
AU - Chen, Ke
AU - Zhao, Kangkang
AU - Lv, Yufan
AU - Zhang, Chuntao
AU - Kong, Yun
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - Source-target domains with large distributional differences may lead to the performance attenuation of multi-source domain adaptation, which in turn affects the accuracy of intelligent transfer fault diagnosis. To address this issue, this paper proposes a mechanical cross-domain diagnosis method based on multi-source weighted domain adaptation. First, a multi-source domain weighting method based on orthogonal bases in principal component space is proposed to assign weights to different source domains by measuring the similarity of each pair of source-target domains in order to guide the domain adaptation process and favor the high-weighted source domains. Then, a multi-attention network is designed to enhance the domain-invariant representation of critical fault features by fusing multi-level features. Finally, the health state identification on the target domain is realized by the trained feature extractor and classifier. Validations are performed on planetary gearbox datasets, and experimental results show that the proposed method exhibits superior domain adaptation and cross-domain diagnosis performance over other methods, obtaining the highest diagnosis accuracies of 98.48%.
AB - Source-target domains with large distributional differences may lead to the performance attenuation of multi-source domain adaptation, which in turn affects the accuracy of intelligent transfer fault diagnosis. To address this issue, this paper proposes a mechanical cross-domain diagnosis method based on multi-source weighted domain adaptation. First, a multi-source domain weighting method based on orthogonal bases in principal component space is proposed to assign weights to different source domains by measuring the similarity of each pair of source-target domains in order to guide the domain adaptation process and favor the high-weighted source domains. Then, a multi-attention network is designed to enhance the domain-invariant representation of critical fault features by fusing multi-level features. Finally, the health state identification on the target domain is realized by the trained feature extractor and classifier. Validations are performed on planetary gearbox datasets, and experimental results show that the proposed method exhibits superior domain adaptation and cross-domain diagnosis performance over other methods, obtaining the highest diagnosis accuracies of 98.48%.
KW - Attention mechanism
KW - Fault diagnosis
KW - Multi-source domain adaptation
KW - Source domain weighted
UR - https://www.scopus.com/pages/publications/105028284471
U2 - 10.1007/978-3-032-00968-5_63
DO - 10.1007/978-3-032-00968-5_63
M3 - Conference contribution
AN - SCOPUS:105028284471
SN - 9783032009678
T3 - Mechanisms and Machine Science
SP - 761
EP - 774
BT - Proceedings of the UNIfied Conference of DAMAS, IncoME and TEPEN Conferences, UNIfied 2025 - Volume 1
A2 - Wei, Kexiang
A2 - Yang, Wenxian
A2 - Chen, Bingyan
A2 - Dai, Juchuan
PB - Springer Science and Business Media B.V.
T2 - UNIfied Conference of International Conference on Damage Assessment of Structures, DAMAS 2025, International Conference on Maintenance Engineering, IncoME 2025 and The Efficiency and Performance Engineering, TEPEN 2025
Y2 - 16 May 2025 through 19 May 2025
ER -