Improving Mechanical Fault Diagnosis Using Graph Neural Networks with Dynamic and Multiscale Features
Received: 19 April 2025 | Revised: 14 May 2025 | Accepted: 31 May 2025 | Online: 2 August 2025
Corresponding author: Suchetha Sheka
Abstract
Mechanical systems face a major drawback to fault diagnosis, as class imbalance greatly undermines it since minority class instances (critical faults) are underrepresented, resulting in biased predictions. This paper introduces a new Multiscale Receptive Fields and Dynamic Edge Weighting (MRS-GNN) framework, which fuses MRF and Dynamic Edge Weighting (DEW) on a GNN to improve classification performance in imbalanced datasets. Graph edge strengths are dynamically weighted with the learned node embeddings during training according to the DEW mechanism, and MRF allows the model to aggregate information from different neighborhood scopes for robust feature representation. In addition, a graph-specific oversampling algorithm, MR-SMOTE, was used to generate synthetic minority class nodes respecting and preserving the topology of the graph. The proposed model was evaluated through experiments on the 2009 PHM gearbox dataset and was found to have an accuracy of 92.1% and an AUC-ROC score of 0.95, better than traditional oversampling methods such as SMOTE, LR-SMOTE, and Graph-SMOTE. The results of an ablation study indicate that 3.7% and 2.4% accuracy drops occur in DEW and MRF removals, respectively, highlighting their importance. This study proposes a scalable and topology-preserving solution to the imbalanced fault diagnosis problem and makes substantial improvements compared to existing GNN-based methods.
Keywords:
fault diagnosis, class imbalance, dynamic edge weighting, graph neural networks, predictive maintenanceDownloads
References
J. Zhu, Q. Jiang, Y. Shen, C. Qian, F. Xu, and Q. Zhu, "Application of recurrent neural network to mechanical fault diagnosis: a review," Journal of Mechanical Science and Technology, vol. 36, no. 2, pp. 527–542, Feb. 2022. DOI: https://doi.org/10.1007/s12206-022-0102-1
Z. Li, Y. Wang, and K. S. Wang, "Intelligent predictive maintenance for fault diagnosis and prognosis in machine centers: Industry 4.0 scenario," Advances in Manufacturing, vol. 5, no. 4, pp. 377–387, Dec. 2017. DOI: https://doi.org/10.1007/s40436-017-0203-8
L. Tang, H. Tian, H. Huang, S. Shi, and Q. Ji, "A survey of mechanical fault diagnosis based on audio signal analysis," Measurement, vol. 220, Oct. 2023, Art. no. 113294. DOI: https://doi.org/10.1016/j.measurement.2023.113294
S. Kumar, M. Lokesha, K. Kumar, and K. R. Srinivas, "Vibration based Fault Diagnosis Techniques for Rotating Mechanical Components: Review Paper," IOP Conference Series: Materials Science and Engineering, vol. 376, Jun. 2018, Art. no. 012109. DOI: https://doi.org/10.1088/1757-899X/376/1/012109
H. Henao et al., "Trends in Fault Diagnosis for Electrical Machines: A Review of Diagnostic Techniques," IEEE Industrial Electronics Magazine, vol. 8, no. 2, pp. 31–42, Jun. 2014. DOI: https://doi.org/10.1109/MIE.2013.2287651
W. Lu, W. Wang, X. Qin, and Z. Cai, "Enhancing Fault Diagnosis in Mechanical Systems with Graph Neural Networks Addressing Class Imbalance," Mathematics, vol. 12, no. 13, Jul. 2024, Art. no. 2064. DOI: https://doi.org/10.3390/math12132064
T. Zhang et al., "Intelligent fault diagnosis of machines with small & imbalanced data: A state-of-the-art review and possible extensions," ISA Transactions, vol. 119, pp. 152–171, Jan. 2022. DOI: https://doi.org/10.1016/j.isatra.2021.02.042
X. Dong, H. Gao, L. Guo, K. Li, and A. Duan, "Deep Cost Adaptive Convolutional Network: A Classification Method for Imbalanced Mechanical Data," IEEE Access, vol. 8, pp. 71486–71496, 2020. DOI: https://doi.org/10.1109/ACCESS.2020.2986419
L. Xiao, X. Yang, and X. Yang, "A graph neural network-based bearing fault detection method," Scientific Reports, vol. 13, no. 1, Mar. 2023, Art. no. 5286. DOI: https://doi.org/10.1038/s41598-023-32369-y
J. Li, X. Cao, R. Chen, X. Zhang, X. Huang, and Y. Qu, "Graph neural network architecture search for rotating machinery fault diagnosis based on reinforcement learning," Mechanical Systems and Signal Processing, vol. 202, Nov. 2023, Art. no. 110701. DOI: https://doi.org/10.1016/j.ymssp.2023.110701
R. Liu, Q. Zhang, D. Lin, W. Zhang, and S. X. Ding, "Causal intervention graph neural network for fault diagnosis of complex industrial processes," Reliability Engineering & System Safety, vol. 251, Nov. 2024, Art. no. 110328. DOI: https://doi.org/10.1016/j.ress.2024.110328
M. Zareapoor, P. Shamsolmoali, and J. Yang, "Oversampling adversarial network for class-imbalanced fault diagnosis," Mechanical Systems and Signal Processing, vol. 149, Feb. 2021, Art. no. 107175. DOI: https://doi.org/10.1016/j.ymssp.2020.107175
Z. Ren, T. Lin, K. Feng, Y. Zhu, Z. Liu, and K. Yan, "A Systematic Review on Imbalanced Learning Methods in Intelligent Fault Diagnosis," IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1–35, 2023. DOI: https://doi.org/10.1109/TIM.2023.3246470
X. W. Liang, A. P. Jiang, T. Li, Y. Y. Xue, and G. T. Wang, "LR-SMOTE — An improved unbalanced data set oversampling based on K-means and SVM," Knowledge-Based Systems, vol. 196, May 2020, Art. no. 105845. DOI: https://doi.org/10.1016/j.knosys.2020.105845
M. F. Muzakki, R. D. Prayogo, and M. A. Rizky A, "Handling Imbalanced Data for Acute Coronary Syndrome Classification Based on Ensemble and K-Means SMOTE Method," JOIV : International Journal on Informatics Visualization, vol. 7, no. 3–2, Nov. 2023, Art. no. 1989. DOI: https://doi.org/10.30630/joiv.7.3-2.1429
Y. Zhang, L. Deng, and B. Wei, "Imbalanced Data Classification Based on Improved Random-SMOTE and Feature Standard Deviation," Mathematics, vol. 12, no. 11, Jan. 2024, Art. no. 1709. DOI: https://doi.org/10.3390/math12111709
T. Li, Z. Zhao, C. Sun, R. Yan, and X. Chen, "Multireceptive Field Graph Convolutional Networks for Machine Fault Diagnosis," IEEE Transactions on Industrial Electronics, vol. 68, no. 12, pp. 12739–12749, Sep. 2021. DOI: https://doi.org/10.1109/TIE.2020.3040669
Z. Chen, J. Xu, T. Peng, and C. Yang, "Graph Convolutional Network-Based Method for Fault Diagnosis Using a Hybrid of Measurement and Prior Knowledge," IEEE Transactions on Cybernetics, vol. 52, no. 9, pp. 9157–9169, Sep. 2022. DOI: https://doi.org/10.1109/TCYB.2021.3059002
X. Yu, B. Tang, and K. Zhang, "Fault Diagnosis of Wind Turbine Gearbox Using a Novel Method of Fast Deep Graph Convolutional Networks," IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–14, 2021. DOI: https://doi.org/10.1109/TIM.2020.3048799
T. Li, Z. Zhao, C. Sun, R. Yan, and X. Chen, "Domain Adversarial Graph Convolutional Network for Fault Diagnosis Under Variable Working Conditions," IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1–10, 2021. DOI: https://doi.org/10.1109/TIM.2021.3075016
X. Zhang, Y. Hu, J. Liu, X. Zhang, and B. Wu, "Robust rotating machinery diagnosis using a dynamic-weighted graph updating strategy," Measurement, vol. 202, Oct. 2022, Art. no. 111895. DOI: https://doi.org/10.1016/j.measurement.2022.111895
D. Zuo, T. Tang, and M. Chen, "Rolling bearing fault diagnosis based on multi-scale weighted visibility graph and multi-channel graph convolution network," Measurement Science and Technology, vol. 34, no. 11, Dec. 2023, Art. no. 115019. DOI: https://doi.org/10.1088/1361-6501/ace7e5
N. Heidari and A. Iosifidis, "Progressive Graph Convolutional Networks for Semi-Supervised Node Classification," IEEE Access, vol. 9, pp. 81957–81968, 2021. DOI: https://doi.org/10.1109/ACCESS.2021.3085905
D. Chen, R. Liu, Q. Hu, and S. X. Ding, "Interaction-Aware Graph Neural Networks for Fault Diagnosis of Complex Industrial Processes," IEEE Transactions on Neural Networks and Learning Systems, vol. 34, no. 9, pp. 6015–6028, Sep. 2023. DOI: https://doi.org/10.1109/TNNLS.2021.3132376
M. Shi, C. Ding, R. Wang, C. Shen, W. Huang, and Z. Zhu, "Graph embedding deep broad learning system for data imbalance fault diagnosis of rotating machinery," Reliability Engineering & System Safety, vol. 240, Dec. 2023, Art. no. 109601. DOI: https://doi.org/10.1016/j.ress.2023.109601
M. A. Abbas et al., "Residual Attention Augmentation Graph Neural Network for Improved Node Classification," Engineering, Technology & Applied Science Research, vol. 14, no. 2, pp. 13238–13242, Apr. 2024. DOI: https://doi.org/10.48084/etasr.6844
M. Sreenatha and P. B. Mallikarjuna, "A Fault Diagnosis Technique for Wind Turbine Gearbox: An Approach using Optimized BLSTM Neural Network with Undercomplete Autoencoder," Engineering, Technology & Applied Science Research, vol. 13, no. 1, pp. 10170–10174, Feb. 2023. DOI: https://doi.org/10.48084/etasr.5595
S. Mogal, S. Deshmukh, and S. Talekar, "Fault Diagnosis of Rotating Machinery based on the Minutiae Algorithm," Engineering, Technology & Applied Science Research, vol. 13, no. 5, pp. 11649–11654, Oct. 2023. DOI: https://doi.org/10.48084/etasr.6175
"2009 PHM Challenge Competition Data Set." PHM Society, [Online]. Available: https://phmsociety.org/public-data-sets/.
P. Boškoski and A. Urevc, "Bearing fault detection with application to PHM Data Challenge," International Journal of Prognostics and Health Management, vol. 2, no. 1, Jan. 2011. DOI: https://doi.org/10.36001/ijphm.2011.v2i1.1340
Downloads
How to Cite
License
Copyright (c) 2025 Suchetha Sheka, Asha B. Saraswathi
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
