RESEARCH PAPER
A Physics-Guided Transfer Learning Framework with Consistency Verification for Cross-Domain Bearing Fault Diagnosis
1
Xinjiang University, China
Submission date: 2025-07-14
Final revision date: 2025-08-01
Acceptance date: 2025-10-07
Online publication date: 2025-10-19
Publication date: 2025-10-19
Eksploatacja i Niezawodność – Maintenance and Reliability 2026;28(2):211797
HIGHLIGHTS
- A Physics-Constrained Transfer Learning (PCTL) framework is proposed, exploiting invariant physical fault patterns to move beyond purely statistical alignment for trustworthy diagnosis.
- A 'diagnosis-verification-feedback' loop uses a rule-based physics validator to supervise a confidence predictor, deeply coupling model confidence with the physical plausibility of its decisions.
- The framework achieves superior cross-domain accuracy and quantifiable trustworthiness, enabling the model to reliably self-assess and signal physically implausible diagnoses with low confidence.
KEYWORDS
Transfer Learning Explainable AIFault DiagnosisRolling BearingsDomain AdaptationPhysical Consistency
TOPICS
ABSTRACT
Deep learning models face a trust deficit due to poor generalization and 'black-box' interpretability. Conventional transfer learning, reliant on statistical alignment, fails to guarantee physical plausibility. We propose a Physics-Constrained Transfer Learning (PCTL) framework based on the core insight that while raw signals vary, intrinsic physical fault patterns—like harmonic structures in the envelope spectrum—remain domain-invariant. Its key innovation is a 'diagnosis-verification-feedback' loop where an external, rule-based PCV module quantifies the consistency between a diagnosis and its physical evidence. This consistency score guides a confidence predictor, compelling the model's confidence to align with physical rationality. Extensive experiments show PCTL achieves superior accuracy and embeds reliable self-assessment, demonstrated by a strong correlation between predicted confidence and physical consistency. This research offers a new paradigm for developing intelligent diagnostic systems that are accurate, physically interpretable, and trustworthy.
REFERENCES (29)
1.
Huang M, Yin J, Yan S, Xue P. A fault diagnosis method of bearings based on deep transfer learning. Simulation Modelling Practice and Theory 2023;122:102659.
2.
Hou W, Zhang C, Jiang Y, Cai K, Wang Y, Li N. A new bearing fault diagnosis method via simulation data driving transfer learning without target fault data. Measurement 2023;215:112879.
3.
Guo Y, Cheng Z, Zhang J, Sun B, Wang Y. A review on adversarial–based deep transfer learning mechanical fault diagnosis. Journal of Big Data 2024;11:151.
4.
Yang B, Lei Y, Jia F, Xing S. An intelligent fault diagnosis approach based on transfer learning from laboratory bearings to locomotive bearings. Mechanical Systems and Signal Processing 2019;122:692–706.
5.
Huo C, Jiang Q, Shen Y, Zhu Q, Zhang Q. Enhanced transfer learning method for rolling bearing fault diagnosis based on linear superposition network. Engineering Applications of Artificial Intelligence 2023;121:105970.
6.
Li Z, Li Z, Gu F. Intelligent diagnosis method for machine faults based on federated transfer learning. Applied Soft Computing 2024;163:111922.
7.
Davoodabadi A, Behzad M, Arghand HA, Mohammadi S, Gelman L. Intelligent Diagnosis of Rolling Element Bearings Under Various Operating Conditions Using an Enhanced Envelope Technique and Transfer Learning. Machines 2025;13:351.
8.
Huo C, Jiang Q, Shen Y, Qian C, Zhang Q. New transfer learning fault diagnosis method of rolling bearing based on ADC-CNN and LATL under variable conditions. Measurement 2022;188:110587.
9.
Liao M, Liu C, Wang C, Yang J. Research on a rolling bearing fault detection method with wavelet convolution deep transfer learning. IEEE Access 2021;9:45175–88.
10.
Zheng Z, Fu J, Lu C, Zhu Y. Research on rolling bearing fault diagnosis of small dataset based on a new optimal transfer learning network. Measurement 2021;177:109285.
11.
Wang H, Wang P, Wang S, Li D. Rolling bearing fault diagnosis method based on dynamic simulated model from source domain to target domain with improved alternating transfer learning. Nonlinear Dynamics 2025;113:4485–510.
12.
Wang Z, He X, Yang B, Li N. Subdomain adaptation transfer learning network for fault diagnosis of roller bearings. IEEE Transactions on Industrial Electronics 2021;69:8430–9.
13.
Chen H, Luo H, Huang B, Jiang B, Kaynak O. Transfer Learning-motivated Intelligent Fault Diagnosis Designs: A Survey, Insights, and Perspectives 2022. https://doi.org/10.36227/techr....
14.
Fang C, Chang Q, Hu X, Zhou W, Meng X. A digital twin-enabled domain adaptation network for cross-space fault diagnosis of roller bearings. Mechanical Systems and Signal Processing 2025;236:113053.
15.
Fu W, Li S, Wen B, Zheng B, Liao W, Tan C. A review of rolling bearing fault diagnosis: data preprocessing and model optimization. Measurement Science and Technology 2025.
16.
Zhang P, Liu D. A rolling bearing fault diagnosis method under insufficient samples condition based on MSLSTM transfer learning. Journal of Vibroengineering 2025;27:93–107.
17.
Yang B, Wang T, Xie J, Yang J, Pan T, Wang Y. A Source Domain Adaptive Method for Transfer-Learning-Based Fault Diagnosis of Train Bogie Bearings at Higher Speeds. IEEE Transactions on Instrumentation and Measurement 2025.
18.
Tang Z, Bo L, Liu X, Wei D. An autoencoder with adaptive transfer learning for intelligent fault diagnosis of rotating machinery. Measurement Science and Technology 2021;32:055110.
19.
Zhi S, Su K, Yu J, Li X, Shen H. An unsupervised transfer learning bearing fault diagnosis method based on multi-channel calibrated Transformer with shiftable window. Structural Health Monitoring 2025:14759217251324671.
20.
Wu M, Zhang J, Xu P, Liang Y, Dai Y, Gao T, et al. Bearing fault diagnosis for cross-condition scenarios under data scarcity based on transformer transfer learning network. Electronics 2025;14:515.
21.
Qin Y, Qian Q, Wang Y, Zhou J. Intermediate distribution alignment and its application into mechanical fault transfer diagnosis. IEEE Transactions on Industrial Informatics 2022;18:7305–15.
22.
Xie S, Li Y, Tan H, Liu R, Zhang F. Multi-scale and multi-layer perceptron hybrid method for bearings fault diagnosis. International Journal of Mechanical Sciences 2022;235:107708.
23.
Kang X, Cheng J, Yang Y, Liu F. Repetitive transient impact detection and its application in cross-machine fault detection of rolling bearings. Mechanical Systems and Signal Processing 2025;228:112422.
24.
Li Z, Zhong Z, Zhang Z, Mao W, Zhang W. Rolling Bearing Dynamics Simulation Information-Assisted Fault Diagnosis with Multi-Adversarial Domain Transfer Learning. Lubricants 2025;13:116.
25.
Yao X, Wang B, Shen Y, Fu G, Zhang D, Jiang Q. Semi-Supervised Class Adaptive Prototype Network for Cross-Working Rolling Bearing Fault Diagnosis Under Limited Samples. IEEE Internet of Things Journal 2025.
26.
Sobie C, Freitas C, Nicolai M. Simulation-driven machine learning: Bearing fault classification. Mechanical Systems and Signal Processing 2018;99:403–19.
27.
Yang B, Lei Y, Li X, Li N. Targeted transfer learning through distribution barycenter medium for intelligent fault diagnosis of machines with data decentralization. Expert Systems with Applications 2024;244:122997.
28.
Kibrete F, Woldemichael DE, Gebremedhen HS. Transfer Learning‐Based Domain‐Adaptive One‐Dimensional Convolutional Neural Network for Fault Diagnosis of Rotating Machines. Engineering Reports 2025;7:e70258.
29.
Matania O, Cohen R, Bechhoefer E, Bortman J. Zero-fault-shot learning for bearing spall type classification by hybrid approach. Mechanical Systems and Signal Processing 2025;224:112117.
Share
RELATED ARTICLE
| eISSN: | 2956-3860 |
| ISSN: | 1507-2711 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
