RESEARCH PAPER
Electromagnetic Relay Contact Resistance Prediction Based on TimeGAN with CNN-LSTM-Attention
More details
Hide details
1
College of Automation Engineering, Jiangsu University of Science and Technology, China
2
School of Electronic and Electrical Engineering, University of Leeds, United Kingdom
These authors had equal contribution to this work
Submission date: 2025-02-13
Final revision date: 2025-04-08
Acceptance date: 2025-05-03
Online publication date: 2025-05-22
Publication date: 2025-05-22
Corresponding author
Zhaobin Wang
College of Automation Engineering, Jiangsu University of Science and Technology, 212100, Zhenjiang, China
Eksploatacja i Niezawodność – Maintenance and Reliability 2025;27(4):204610
HIGHLIGHTS
- Proposes TimeGAN-CNN-LSTM-Attention model for precise contact resistance prediction.
- TimeGAN-generated synthetic data validated via PCA/KDE to resolve degradation data.
- Combines adversarial data generation and deep learning in resource-limited systems.
- Achieves 30.22% R² boost , demonstrating superior prediction accuracy.
KEYWORDS
TOPICS
ABSTRACT
Electromagnetic relays are critical in aerospace and military systems, affecting the safety and stability of applications like aircraft control, satellite communication, and missile launchers. However, the scarcity of degradation data and complex variations in contact resistance pose challenges. Traditional methods often struggle with small samples. To address these issues, we propose a novel framework integrating TimeGAN with a CNN-LSTM-Attention model. TimeGAN generates synthetic degradation data that aligns with the statistical distribution of the original dataset, mitigating data scarcity. Data quality is evaluated using PCA and KDE. The CNN-LSTM model captures multi-scale temporal features, while the attention mechanism highlights critical features to improve contact resistance prediction accuracy. Experimental results show that the proposed framework outperforms traditional methods, demonstrating robust performance even without data augmentation. These findings offer a valuable foundation for health monitoring and fault prediction in high-reliability systems.
ACKNOWLEDGEMENTS
The authors are grateful to the anonymous reviewers, and the editor for their critical and constructive review of the manuscript. This study was co-supported by the National Natural Science Foundation of China (51507074), Postgraduate Research &Practice Innovation Program of Jiangsu Province (KYCX25_4368).
REFERENCES (21)
1.
Wang, Z., Li, W., Shang, S., Wang, Z., Han, C. (2020). Performance degradation comparisons and failure mechanism of silver metal oxide contact materials in relays application by simulation. Eksploatacja i Niezawodność – Maintenance and Reliability, 22(1), 86-93.
https://doi.org/10.17531/ein.2....
2.
Wang X., Wang BX., Wu W., Hong Y. Reliability analysis for accelerated degradation data based on the Wiener process with random effects. Quality & Reliability Eng 2020;36:1969–81.
https://doi.org/10.1002/qre.26....
3.
Liu, Y., Wang, Y., Fan, Z., Hou, Z., Zhang, S., Chen, X. (2020). Lifetime prediction method for MEMS gyroscope based on accelerated degradation test and acceleration factor model. Eksploatacja i Niezawodność – Maintenance and Reliability, 22(2), 221-231.
https://doi.org/10.17531/ein.2....
4.
Su, Z., Hua, Z., Hu, D., Zhao, M. (2025). Research on Equipment Reliability Modeling and Periodic Maintenance Strategies in Dynamic Environment. Eksploatacja i Niezawodność – Maintenance and Reliability, 27(1).
https://doi.org/10.17531/ein/1....
5.
Wang, Z., Shang, S., Wang, J., Huang, Z., & Sai, F. (2017). Accelerated storage degradation testing and failure mechanisms of aerospace electromagnetic relay. Eksploatacja I Niezawodnosc-maintenance and Reliability, 19, 530-541.
https://doi.org/ 10.17531/ein.2017.4.6.
6.
Zhan, X., Liu, Z., Yan, H., Wu, Z., Guo, C., Jia, X. (2023). A novel method of health indicator construction and remaining useful life prediction based on deep learning. Eksploatacja i Niezawodność – Maintenance and Reliability, 25(4).
https://doi.org/10.17531/ein/1....
7.
Wang, Z., Shangguan, W., Peng, C., Cai, B. (2024). Similarity Based Remaining Useful Life Prediction for Lithium-ion Battery under Small Sample Situation Based on Data Augmentation. Eksploatacja i Niezawodność – Maintenance and Reliability, 26(1).
https://doi.org/10.17531/ein/1....
8.
Lu, Y., Shen, M., Wang, H., Wang, X., van Rechem C., Fu T., et al. Machine Learning for Synthetic Data Generation: A Review 2024.
9.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative adversarial nets. In Advances in neural information processing systems (pp. 2672-2680).
https://doi.org/10.1109/ICCCNT....
10.
Yoon, J., Jarrett, D., & Schaar, M.V. (2019). Time-series Generative Adversarial Networks. Neural Information Processing Systems.
11.
Yang, L., He, P., & Chou, X. (2024). TimeGAN data augmentation-based fault classification method for complex processes. Journal of Zhejiang University (Engineering Science), 59(9), 1768–1780.
https://doi.org/10.3785/j.issn....
12.
Ma, Z. , Sun, Y. , Ji, H. , Li, S. , Nie, S. , & Yin, F. (2024). A CNN-BiLSTM-Attention approach for EHA degradation prediction based on time-series generative adversarial network. Mechanical Systems & Signal Processing(Jun.), 215.
https://doi.org/10.1016/j.ymss....
13.
Chen, Y., Wang, X., Wang, Y., Gao, Y., Qu, J., Dai, H., & Xu, C. (2024). Multi-combination fault data augmentation method of aero-engine gas path based on Extraction TimeGAN. Measurement, 225, 115778.
https://doi.org/10.1016/j.meas....
14.
Z. G. Liu, T. Y. Ji, J. W. Chen, L. J. Zhang, L. L. Zhang and Q. H. Wu, "Conditional-TimeGAN for Realistic and High-Quality Appliance Trajectories Generation and Data Augmentation in Nonintrusive Load Monitoring," in IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1-15, 2024, Art no. 2514415.
https://doi.org/10.1109/TIM.20....
15.
W. Deng et al., "Short-Term Wind Power Prediction Based on Wind Speed Interval Division and TimeGAN for Gale Weather," 2023 International Conference on Power Energy Systems and Applications (ICoPESA), Nanjing, China, 2023, pp. 352-357.
https://doi.org/10.1109/ICoPES....
16.
Z. Zhi., L. Liu, D.. Liu and C. Hu, "Fault Detection of the Harmonic Reducer Based on CNN-LSTM With a Novel Denoising Algorithm," in IEEE Sensors Journal, vol. 22, no. 3, pp. 2572-2581, 1 Feb.1, 2022.
https://doi.org/10.1109/JSEN.2....
17.
Xu Y., Xiang Y., &Ma T.(2022). Short-term Power Load Forecasting Method Based on EMD-CNN-LSTM Hybrid Model. Journal of North China Electric Power University(02), 81-89.
https://doi.org/10.3969/j.ISSN....
18.
Hou H., Wu W., Wei R., HEe H., Wang L., Li Z & Lin X.(2024). Attention Mechanism Based CNN-LSTM-XGBoost Electric Power Meteorological Hybrid Forecasting Model of Typhoon Rainstorm. Smart Power & Energy Security (10), 96-102.
https://doi.org/10.20204/j.sp.....
19.
Chu Y., Chen Y & Mi Y.(2022). Lithium battery health state estimation based on attention mechanism CNN-LSTM. Chinese Journal of Power Sources (06), 634-637+651.
https://doi.org/10.3969/j.issn....
20.
J. Liu, H. Gao, Q. Yan, B. Bai and X. Wang, "Short-term Load Forecasting Method Based on LSTM-Attention Mechanism," 2021 IEEE Sustainable Power and Energy Conference (iSPEC), Nanjing, China, 2021, pp. 3837-3842 .
https://doi.org/10.1109/iSPEC5....
21.
Zhongshan, H., Ling, T., Dong, X., & Yaozhong, W. (2018). Prediction of oil temperature variations in a wind turbine gearbox based on PCA and an SPC-dynamic neural network hybrid. Journal of Tsinghua University, 58, 539-546.
https://doi.org/10.16511/j.cnk....