Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems

Amar Amouri; Walid Ayadi; Saeed Althabahi

doi:10.35377/saucis...1892264

Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems

Abstract

Anomaly detection in Direct Current (DC) motors plays a vital role in various industries. However, it remains a challenge to develop systems that are intrinsically robust to unpredicted speed variations and environmental noise. This paper proposes the use of a Multi-Scale Temporal Convolutional Network (MSTCN) enhanced with a Convolutional Block Attention Module (CBAM) for feature extraction. The proposed scheme was validated using a proprietary dataset collected with an Arduino Nano 33 BLE IMU. A zero-shot multi-variate stress test involving Fourier-domain resampling and additive Gaussian noise with speed factors and standard deviation levels ranging from 0.5 to 2.0 and 1.5 to 2.0, respectively, was used to perturb the dataset. Under these conditions, it was observed that while the vanilla CNN achieved the lowest latency of 207 microseconds, it failed under high interference, scoring an F1-score of nearly 60 percent at noise level 2.0. The LSTM model reached high accuracy but experienced a computational bottleneck with a latency of 1620 microseconds. In contrast, our MSTCN-CBAM model achieved an F1-score of nearly 90 percent at noise level 2.0 and up to 98 percent under moderate noise conditions while providing a 2.6-fold inference speedup (617 microseconds) and a 55 percent reduction in training time compared with the LSTM baseline. These encouraging results highlight the advantages of our MSTCN-CBAM model for robust anomaly detection in real-world conditions.

Keywords

Ethical Statement

It is declared that during the preparation process of this study, scientific and ethical principles were followed, and all the studies benefited from are stated in the bibliography.

References

Y. R. Thota, M. Afshar, S. Boden, B. Dunlap, B. Akin, and T. Nikoubin, “TinyML enabled real-time bearing fault classification in motors using vibration signals,” in Proc. Great Lakes Symp. VLSI (GLSVLSI), pp. 911–915, 2025. doi: 10.1145/3716368.3735272. [Online]. Available: https://doi.org/10.1145/3716368.3735272
A. Mostafavi, A. Alizadeh, and A. Sadighi, “Edge-computing-based anomaly detection of rotating machines using artificial neural networks,” Sharif J. Mech. Eng., vol. 41, no. 2, pp. 33–47, 2025, doi: 10.24200/j40.2025.65863.1730. [Online]. Available: https://doi.org/10.24200/j40.2025.65863.1730
W. Ayadi, A. Saidi, and I. Channoufi, “Exploring human activity patterns: Investigating feature extraction techniques for improved recognition with ANN,” in Proc. IEEE 7th Int. Conf. Adv. Technol., Signal Image Process. (ATSIP), vol. 1, pp. 188–193, 2024. doi: 10.1109/ATSIP62566.2024.10639004. [Online]. Available: https://doi.org/10.1109/ATSIP62566.2024.10639004
W. Ayadi, A. Amouri, S. Ali, N. Alzahmi, and M. N. Khalid, “Leveraging lightweight AI for anomaly detection in mechanical power transmission at the edge,” EPJ Web Conf., vol. 343, Art. no. 05014, 2025, doi: 10.1051/epjconf/202534305014. [Online]. Available: https://doi.org/10.1051/epjconf/202534305014
Z. Dengfeng et al., “Multi-scale convolutional neural network combining BiLSTM and attention mechanism for bearing fault diagnosis under multiple working conditions,” Sci. Rep., vol. 15, Art. no. 13035, 2025, doi: 10.1038/s41598-025-96137-w. [Online]. Available: https://doi.org/10.1038/s41598-025-96137-w
E. T. Samuel, T. Polonelli, M. Magno, and A. Esan, “TinyML anomaly detection and fault prediction for industrial applications,” Research Square, 2025, doi: 10.21203/rs.3.rs-8379288/v1. [Online]. Available: https://doi.org/10.21203/rs.3.rs-8379288/v1
Z. Gao, Z. Jiang, Z. Dong, X. Fu, and Y. Chen, “An edge-deployable TinyML approach enhanced by transfer learning for efficient bearing fault diagnosis,” Sci. China Technol. Sci., 2025, doi: 10.1007/s11431-025-3072-9. [Online]. Available: https://doi.org/10.1007/s11431-025-3072-9
S. Gawde, S. Patil, S. Kumar, and P. Kamat, “An explainable predictive maintenance strategy for multi-fault diagnosis of rotating machines using multi-sensor data fusion,” Decis. Analytics J., vol. 10, Art. no. 100425, 2024, doi: 10.1016/j.dajour.2024.100425. [Online]. Available: https://doi.org/10.1016/j.dajour.2024.100425

A. R. D. Shirly and S. Prathiba, “Edge-based TinyML for fault detection using sound vibration analysis,” in Proc. Int. Conf. Intell. Comput., 2025, doi: 10.1109/itechsecom64750.2025.11307279. [Online]. Available: https://doi.org/10.1109/itechsecom64750.2025.11307279
H. Al-Libawy, “Efficient implementation of a tiny deep learning classifier based on vibration-related fault detection using limited resources hardware,” J. Univ. Babylon Eng. Sci., pp. 88–105, 2025, doi: 10.29196/770r2493. [Online]. Available: https://doi.org/10.29196/770r2493
X. Zhao and P. Wang, “A deployable edge computing solution for machine condition monitoring,” in Proc. IEEE Int. Instrum. Meas. Technol. Conf. (I2MTC), 2024, doi: 10.1109/I2MTC60896.2024.10560795. [Online]. Available: https://doi.org/10.1109/I2MTC60896.2024.10560795
H. Çalış, “Vibration and motor current analysis of induction motors to diagnose mechanical faults,” J. Meas. Eng., vol. 2, no. 4, pp. 190–198, 2014, doi: 10.21595/jme.2014.15126. [Online]. Available: https://doi.org/10.21595/jme.2014.15126
P. Girdhar and C. Scheffer, “Machinery fault diagnosis using vibration analysis,” in Practical Machinery Vibration Analysis and Predictive Maintenance. Oxford, U.K.: Newnes, ch. 5, pp. 89–133, 2004. doi: 10.1016/B978-075066275-8/50005-9. [Online]. Available: https://doi.org/10.1016/B978-075066275-8/50005-9
Ö. Yılmaz, M. Aksoy, and Z. Kesilmiş, “Misalignment fault detection by wavelet analysis of vibration signals,” Int. Adv. Res. Eng. J., vol. 3, no. 3, pp. 156–163, 2019, doi: 10.35860/iarej.451528. [Online]. Available: https://doi.org/10.35860/iarej.451528
G. Wang, S. Zhao, Z. Zhong, Y. Lv, C. Shao, and H. Zhang, “Anomaly detection method of traction motor bearing based on multi-scale sub-band fuzzy entropy manifold fusion index,” J. Vibroeng., 2025, doi: 10.21595/jve.2024.24282. [Online]. Available: https://doi.org/10.21595/jve.2024.24282
B. Battulga, M. F. Shaikh, T. Göktaş, M. Arkan, and S. B. Lee, “Vibration-based identification of mechanical defects in induction motor-driven systems during the starting transient,” in Proc. IEEE Int. Conf. Elect. Mach. (ICEM), 2024, doi: 10.1109/ICEM60801.2024.10700439. [Online]. Available: https://doi.org/10.1109/ICEM60801.2024.10700439
L. Qu, J. Lin, J. Lin, Y. Liao, and M. Zhao, “Changes in rotor response characteristics based diagnostic method and its application to identification of misalignment,” Measurement, 2019, doi: 10.1016/j.measurement.2019.01.075. [Online]. Available: https://doi.org/10.1016/j.measurement.2019.01.075
P. Jayaswal and A. Aherwar, “Fault detection and diagnosis of gear transmission system via vibration analysis,” IUP J. Mech. Eng., vol. 4, no. 3, pp. 26–43, 2011. [Online]. Available: https://ssrn.com/abstract=2097703
E. H. Demircioğlu and E. Yılmaz, “A method based on an autoencoder for anomaly detection in DC motor body temperature,” Appl. Sci., vol. 13, no. 15, Art. no. 8701, 2023, doi: 10.3390/app13158701. [Online]. Available: https://doi.org/10.3390/app13158701
Z. Yang, Y. Wang, and J. Li, “Research on CNN-LSTM DC power system fault diagnosis and differential protection strategy based on reinforcement learning,” Front. Energy Res., vol. 11, Art. no. 1258549, 2023, doi: 10.3389/fenrg.2023.1258549. [Online]. Available: https://doi.org/10.3389/fenrg.2023.1258549
F. Soresini, D. Barri, I. Cazzaniga, F. M. Ballo, G. Mastinu, and M. Gobbi, “Artificial intelligence for fault detection of automotive electric motors,” Machines, vol. 13, no. 6, Art. no. 457, 2025, doi: 10.3390/machines13060457. [Online]. Available: https://doi.org/10.3390/machines13060457
G. Fu, Q. Wei, and Y. Yang, “Bearing fault diagnosis with parallel CNN and LSTM,” Math. Biosci. Eng., vol. 21, no. 2, pp. 2385–2406, 2024, doi: 10.3934/mbe.2024105. [Online]. Available: https://doi.org/10.3934/mbe.2024105
F. Xie, Q. Fan, G. Li, Y. Wang, E. Sun, and S. Zhou, “Motor fault diagnosis based on convolutional block attention module-Xception lightweight neural network,” Entropy, vol. 26, no. 9, Art. no. 810, 2024, doi: 10.3390/e26090810. [Online]. Available: https://doi.org/10.3390/e26090810
W. Liu, Z. Zou, F. Gu, and G. Shen, “Machine learning for fault diagnosis of electric motors in actuator systems,” Actuators, vol. 14, no. 12, Art. no. 596, 2025, doi: 10.3390/act14120596. [Online]. Available: https://doi.org/10.3390/act14120596
Z. Liu et al., “Research on fault diagnosis of marine diesel engines based on CNN-TCN-ATTENTION,” Appl. Sci., vol. 15, no. 3, Art. no. 1651, 2025, doi: 10.3390/app15031651. [Online]. Available: https://doi.org/10.3390/app15031651
I. Rosyadi, Y. Y. Nazaruddin, and P. I. Siregar, “Enhancing fault diagnosis accuracy in electric motors: A digital twin approach with Transformer model,” in Proc. 14th Asian Control Conf. (ASCC), pp. 1591–1596, 2024.
S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, “CBAM: Convolutional block attention module,” in Proc. Eur. Conf. Comput. Vis. (ECCV). Cham, Switzerland: Springer, 2018, doi: 10.1007/978-3-030-01234-2 1. [Online]. Available: https://doi.org/10.1007/978-3-030-01234-2_1

Details

Primary Language

English

Subjects

Applied Computing (Other), Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Amar Amouri
0000-0002-5409-5311
United Arab Emirates

Walid Ayadi ^*
0000-0001-9264-5743
United Arab Emirates

Saeed Althabahi
0009-0005-9121-1122
United Arab Emirates

Publication Date

September 30, 2026

Submission Date

February 21, 2026

Acceptance Date

April 2, 2026

Published in Issue

Year 2026 Volume: 9 Number: 4

DOI

https://doi.org/10.35377/saucis...1892264

IZ

https://izlik.org/JA36WP68MM

Cite

RIS / Bibtex

APA

Amouri, A., Ayadi, W., & Althabahi, S. (2026). Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems. Sakarya University Journal of Computer and Information Sciences, 9(4), 1141-1149. https://doi.org/10.35377/saucis...1892264

AMA

1.Amouri A, Ayadi W, Althabahi S. Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems. SAUCIS. 2026;9(4):1141-1149. doi:10.35377/saucis.1892264

Chicago

Amouri, Amar, Walid Ayadi, and Saeed Althabahi. 2026. “Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems”. Sakarya University Journal of Computer and Information Sciences 9 (4): 1141-49. https://doi.org/10.35377/saucis. 1892264.

EndNote

Amouri A, Ayadi W, Althabahi S (September 1, 2026) Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems. Sakarya University Journal of Computer and Information Sciences 9 4 1141–1149.

IEEE

[1]A. Amouri, W. Ayadi, and S. Althabahi, “Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems”, SAUCIS, vol. 9, no. 4, pp. 1141–1149, Sept. 2026, doi: 10.35377/saucis...1892264.

ISNAD

Amouri, Amar - Ayadi, Walid - Althabahi, Saeed. “Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems”. Sakarya University Journal of Computer and Information Sciences 9/4 (September 1, 2026): 1141-1149. https://doi.org/10.35377/saucis. 1892264.

JAMA

1.Amouri A, Ayadi W, Althabahi S. Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems. SAUCIS. 2026;9:1141–1149.

MLA

Amouri, Amar, et al. “Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems”. Sakarya University Journal of Computer and Information Sciences, vol. 9, no. 4, Sept. 2026, pp. 1141-9, doi:10.35377/saucis. 1892264.

Vancouver

1.Amar Amouri, Walid Ayadi, Saeed Althabahi. Multi-Scale Temporal Convolutional Networks for Robust Variable-Speed Fault Diagnosis in DC Motor Transmission Systems. SAUCIS. 2026 Sep. 1;9(4):1141-9. doi:10.35377/saucis. 1892264