Time–Frequency-Domain Fusion Cross-Attention Fault Diagnosis Method Based on Dynamic Modeling of Bearing Rotor System
Source abstract
Abstract
Deep learning (DL) and machine learning (ML) have advanced rapidly. This has driven significant progress in intelligent fault diagnosis (IFD) of bearings. However, methods like self-attention have limitations. They only capture features within a single sequence. They fail to effectively extract and fuse time- and frequency-domain characteristics from raw signals. This is a critical bottleneck. To tackle this, a dual-channel cross-attention dynamic fault diagnosis network for time–frequency signals is proposed. This model’s intrinsic correlations between time-domain and frequency-domain features, which overcomes single-sequence limitations. The simulation and experimental data validate the method. It achieves over 95% diagnostic accuracy. It effectively captures complex fault patterns. This work provides a theoretical basis for better fault identification in bearing–rotor systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Fault Diagnosis Techniques · Fault Detection and Control Systems · Engineering Diagnostics and Reliability
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mechanical Engineering, Shenyang Jianzhu University, Shenyang 110168, China
North Huajin Chemical Industries Group Corporation, Panjin 124021, China
Funding
National Natural Science Foundation of China · Department of Science and Technology of Liaoning province · Education Department of Liaoning Provincial Program · Shenyang bureau of science and technology
References (36)
Show source-provided references
1. Shang, L., Zhang, Z., Tang, F., Cao, Q., Pan, H., and Lin, Z. (2023). CNN-LSTM hybrid model to promote signal processing of ultrasonic guided lamb waves for damage detection in metallic pipelines. Sensors, 23.10.20944/preprints202307.0929.v1 ↗
2. Bearing fault diagnosis based on mel frequency cepstrum coefficient and deformable space-frequency attention network10.1109/ACCESS.2023.3264276 ↗
3. Zegai, M.L., Bendjebbar, M., Belhadri, K., Doumbia, M.L., Hamane, B., and Koumba, P.M. (2015, January 26–28). Direct torque control of Induction Motor based on artificial neural networks speed control using MRAS and neural PID controller. Proceedings of the 2015 IEEE Electrical Power and Energy Conference (EPEC), London, ON, Canada.10.1109/EPEC.2015.7379970 ↗
4. Douiri, M.R., Cherkaoui, M., and Douiri, S.M. (2012, January 10–12). Rotor resistance and speed identification using extended kalman filter and fuzzy logic controller for induction machine drive. Proceedings of the 2012 International Conference on Multimedia Computing and Systems, Tangiers, Morocco.10.1109/ICMCS.2012.6320298 ↗
5. Douiri, M.R., Cherkaoui, M., Nasser, T., and Essadki, A. (2011, January 7–9). A neuro fuzzy PI controller used for speed control of a direct torque to twelve sectors controlled induction machine drive. Proceedings of the 2011 International Conference on Multimedia Computing and Systems, Ouarzazate, Morocco.10.1109/ICMCS.2011.5945686 ↗
6. Feature extraction based on hierarchical improved envelope spectrum entropy for rolling bearing fault diagnosis
7. Dynamic analysis of deep groove ball bearing with localized defects and misalignment10.1016/j.jsv.2023.118071 ↗
8. Vibration characteristics investigation of a single/dual rotor-bearing-casing system with local bearing defects10.1016/j.ymssp.2024.112227 ↗
9. Experimental and numerical study of an angular contact ball bearing vibration response with spall defect on the outer race10.1007/s00419-020-01733-z ↗
10. Modeling and validation of reaction wheel micro-vibrations considering imbalances and bearing disturbances10.1016/j.jsv.2020.115766 ↗
11. Theoretical and experimental vibration studies of lubricated deep groove ball bearings having surface waviness on its races10.1016/j.measurement.2018.07.031 ↗
12. A nonlinear dynamic vibration model of a defective bearing: The importance of modelling the angle of the leading and trailing edges of a defect10.1177/1475921720963950 ↗
13. Theoretical and experimental analyses of vibration impulses and their influence on accurate diagnosis of ball bearing with localized outer race defect10.1016/j.jsv.2021.116407 ↗
14. Vibration analysis-based fault diagnosis of a dynamically loaded bearing with distributed defect10.1007/s13369-021-05862-7 ↗
15. Amplitudes of components in vibration spectra of rolling bearings with localized defects under harmonic loads10.1177/1077546320945459 ↗
16. Analysis of contact behaviours and vibrations in a defective deep groove ball bearing10.1016/j.jsv.2023.118104 ↗
17. Nonlinear vibration characteristics of a rotor bearing system with irregular raceway defect10.1007/s11071-024-10689-6 ↗
18. Song, X., Li, Z., and Liu, Y. (2025). MVB fault diagnosis based on time-frequency analysis and convolutional neural networks. Sci. Rep., 15.10.1038/s41598-025-89793-5 ↗
19. Recent progress in digital twin-driven fault diagnosis of rotating machinery: A comprehensive review10.1016/j.neucom.2025.129914 ↗
20. Rolling bearing fault diagnosis with combined convolutional neural networks and support vector machine10.1016/j.measurement.2021.109022 ↗
21. Graph-based semi-supervised random forest for rotating machinery gearbox fault diagnosis10.1016/j.conengprac.2021.104952 ↗
22. Deep convolutional tree-inspired network: A decision-tree-structured neural network for hierarchical fault diagnosis of bearings10.1007/s11465-021-0650-6 ↗
23. Lee, D., Choo, H., and Jeong, J. (2024). Gcn-based lstm autoencoder with self-attention for bearing fault diagnosis. Sensors, 24.10.3390/s24154855 ↗
24. Borré, A., Seman, L.O., Camponogara, E., Stefenon, S.F., Mariani, V.C., and Coelho, L.d.S. (2023). Machine fault detection using a hybrid CNN-LSTM attention-based mode. Sensors, 23.10.3390/s23094512 ↗
25. Djaballah, S., Saidi, L., Meftah, K., Hechifa, A., Bajaj, M., and Zaitsev, I. (2024). A hybrid LSTM random forest model with grey wolf optimization for enhanced detection of multiple bearing faults. Sci. Rep., 14.10.1038/s41598-024-75174-x ↗
26. A novel bearing fault diagnosis method for compound defects via zero-shot learning10.1007/s12206-024-0801-x ↗
27. Adaptive wavelet-informed physics-based CNN for bearing fault diagnosis
28. Bearing fault diagnosis method based on attention mechanism and multi-channel feature fusion10.1109/ACCESS.2024.3381618 ↗
29. A bearing fault diagnosis method based on M-SSCNN and M-LR attention mechanism10.1177/14759217241244477 ↗
30. Self-supervised feature learning for motor fault diagnosis under various torque conditions10.1016/j.knosys.2024.111465 ↗
31. Physics-informed time-frequency fusion network with attention for noise-robust bearing fault diagnosis10.1109/ACCESS.2024.3355268 ↗
32. Integrating self-attention mechanisms in deep learning: A novel dual-head ensemble transformer with its application to bearing fault diagnosis10.1016/j.sigpro.2024.109683 ↗
33. Simulation data driven time-frequency fusion 1D convolutional neural network with multiscale attention for bearing fault diagnosis10.1088/1361-6501/adb329 ↗
34. A new bearing fault diagnosis method via simulation data driving transfer learning without target fault data10.1016/j.measurement.2023.112879 ↗
35. A small sample diagnosis method driven by simulation and test data: Applied to axle box bearings of high-speed train10.1088/1361-6501/acf67d ↗
36. Analysis electroencephalogram signals using ANFIS and periodogram techniques