4-санға «Автоматтандыру және ақпараттық технологиялар», «Азық-түлік инженериясы және биотехнология», сондай-ақ «Химиялық технология» бағыттары бойынша мақалалар қабылдау жабылды.

Приём статей в четвёртый номер по направлениям «Автоматизация и информационные технологии», «Пищевая инженерия и биотехнология», а также «Химическая технология» закрыт.

Submissions for the fourth issue in the fields of Automation and Information Technology, Food Engineering and Biotechnology, as well as Chemical Technology, are now closed.

Preview

Bulletin of Shakarim University. Technical Sciences

Bulletin of Shakarim University. Technical Sciences

Advanced search

AUTOMATED SYSTEM FOR PREDICTING FAILURES OF LOW-VOLTAGE ELECTRIC MOTORS IN ELECTRIC POWER PLANTS

https://doi.org/10.53360/2788-7995-2026-3(23)-35

Abstract

Induction motors are available everywhere in industrial systems, used to drive pumps, compressors and fans, but they are prone to electrical failures (broken rotor bars, stator Inter-turn short circuiting, voltage anomalies) and mechanical failures (bearing damage, misalignment, rotor eccentricity). These flaws lower productivity, increase vibration and temperature, and lead to unplanned closures. Predictive maintenance is thus keen on accurate early detection. The current paper introduces an automated fault diagnosis system that has been tested on the University of Ottawa Electric Motor Dataset (UOEMD-VAFCVS) which offers synchronized vibration (three accelerometers), acoustic, and temperature measurements of eight motors at constant speed and ramp-speed conditions. The diagnostic pipeline is based on multi-sensor fusion with signal processing (FFT/Welch PSD, STFT, discrete wavelet transforms, Hilbert envelope analysis) and a hierarchical machine learning model (Random Forest (RF), Convolutional Neural Network (CNN), and a hybrid CNN Transformer model). CNN-Transformer is the most accurate (91.4) and macro F1-score (0.91) model and is superior to RF (81.2) and CNN alone (87.6), with all models satisfying real-time inference requirements (less than 10ms per window on CPU). These findings confirm the usefulness of multi-sensor fusion and attention-based temporal modeling to Industry 4.0 predictive maintenance.

About the Authors

O. N. Sadu
Almaty University of Power Engineering and Telecommunications
Kazakhstan

Orazbek Nurzhanuly Sadu – 2nd year doctoral student

050013, Almaty, Kazakhstan, st. Baitursynuly, 126/1



G. D. Basil
Almaty University of Power Engineering and Telecommunications
Kazakhstan

Gulmira Duysenbekkyzy Bazil – PhD, professor

050013, Almaty, Kazakhstan, st. Baitursynuly, 126/1



A. Akmar
Universiti Teknologi PETRONAS
Malaysia

Ainul Akmar – PhD, Professor

32610 Seri Iskandar, Perak Darul Ridzuam, Malaysia



References

1. Advancements in induction motor diagnosis and condition monitoring: A comprehensive review / S. Kirschner et al // Sensors. – 2025. – Vol. 25, № 19. – Art. 5942. https://doi.org/10.3390/s25195942.

2. Integrating AI and IoT for predictive maintenance in Industry 4.0 manufacturing environments: A practical approach / R. Rakholia et al // Information. – 2025. – Vol. 16, № 9. – Art. 737. https://doi.org/10.3390/info16090737.

3. Bo L. Enhanced CNN for induction motor fault diagnosis via multi-source data fusion / L. Bo, M. Li, H. Tan // PLoS ONE. – 2025. – Vol. 20, № 8. – Art. e0330761. https://doi.org/10.1371/journal.pone.0330761.

4. One-dimensional time-frequency dual channel visual transformer for fault diagnosis under strong noise and limited data conditions / S. Cai et al // Scientific Reports. – 2025. – Vol. 15, № 1. – Art. 26361. https://doi.org/10.1038/s41598-025-12533-2.

5. Ren S. Rolling bearing fault diagnosis method based on SWT and improved vision transformer / S. Ren, X. Lou // Sensors. – 2025. – Vol. 25, № 7. – Art. 2090. https://doi.org/10.3390/s25072090.

6. Allal A. Diagnosis of induction motor faults using the motor current normalized residual harmonic analysis method / A. Allal, A. Khechekhouche // International Journal of Electrical Power and Energy Systems. – 2022. – Vol. 141. – Art. 108219. https://doi.org/10.1016/j.ijepes.2022.108219.

7. An autoregressive-based motor current signature analysis approach for fault diagnosis of electric motor-driven mechanisms / R. Diversi et al // Sensors. – 2025. – Vol. 25, № 4. – Art. 1130. https://doi.org/10.3390/s25041130.

8. Chisedzi L.P. Broken rotor bar detection in cage induction motors using machine learning methods / L.P. Chisedzi, M. Muteba // Sensors. – 2023. – Vol. 23, № 22. – Art. 9079. https://doi.org/10.3390/s23229079.

9. Elhaija W.A. A novel dataset and lightweight detection system for broken bars in induction motors using optimizable neural networks / W.A. Elhaija, Q.A. Al-Haija // Intelligent Systems with Applications. – 2023. – Vol. 17. – Art. 200167. https://doi.org/10.1016/j.iswa.2022.200167.

10. Antonello R. Machine learning-based detection of broken bars in dual three-phase induction motors: methodology and experimental validation / R. Antonello, M. Pastura, M. Zigliotto // IET Electric Power Applications. – 2025. – Vol. 19, № 1. – Art. e70069. https://doi.org/10.1049/elp2.70069.

11. Chen L. et al. Inter-turn short circuit fault detection in induction motor stator windings using startup current envelope energy / L. Chen et al. // Sensors. – 2023. – Vol. 23, № 20. – Art. 8581. https://doi.org/10.3390/s23208581

12. AI-driven fault diagnosis approach for stator turn-to-turn faults in induction motors / T.A. Kawady et al. // Scientific Reports. – 2025. – Vol. 15, № 1. – Art. 20125. https://doi.org/10.1038/s41598-025-04462-x.

13. Comprehensive fault diagnosis of three-phase induction motors using synchronized multisensor data collection / K. Thomas et al // Scientific Data. – 2025. – Vol. 12, № 1. – Art. 1468. https://doi.org/10.1038/s41597-025-05437-3.

14. Toma R.N. Bearing fault diagnosis of induction motors using a genetic algorithm and machine learning classifiers / R.N. Toma, A.E. Prosvirin, J.M. Kim // Sensors. – 2020. – Vol. 20, № 7. – Art. 1884. https://doi.org/10.3390/s20071884.

15. Lee J.H. Fault diagnosis of induction motor using convolutional neural network / J.H. Lee, J.H. Pack, I.S. Lee // Applied Sciences. – 2019. – Vol. 9, № 15. – Art. 2950. https://doi.org/10.3390/app9152950.

16. Fault diagnosis of electric motors based on multi-modal network of time series and ensemble transformers / B. Xu et al // Scientific Reports. – 2025. – Vol. 15. – Art. 7834. https://doi.org/10.1038/s41598-025-89695-6.

17. Chen X. Cross-domain fault diagnosis of induction motor based on GAN and fine-tuning with limited labeled data / X. Chen, Z. Chen // Measurement. – 2025. – Vol. 249. – Art. 116988. https://doi.org/10.1016/j.measurement.2025.116988.

18. Rapid diagnosis of induction motor electrical faults using convolutional autoencoder feature extraction / A.B. Husebo et al // PHM Society European Conference. – 2020. – Vol. 5, № 1. – P. 27-31. https://doi.org/10.36001/phme.2020.v5i1.1247.

19. Unsupervised electric motor fault detection by using deep autoencoders / E. Principi et al // IEEE/CAA Journal of Automatica Sinica. – 2019. – Vol. 6, № 2. – P. 441-451. https://doi.org/10.1109/JAS.2019.1911393.

20. Stator fault diagnosis of induction motor based on discrete wavelet analysis and neural network technique / A. Almounajjed et al // Chinese Journal of Electrical Engineering. – 2023. – Vol. 9, № 1. – P. 142-157. https://doi.org/10.23919/CJEE.2023.000003.


Review

For citations:


Sadu O.N., Basil G.D., Akmar A. AUTOMATED SYSTEM FOR PREDICTING FAILURES OF LOW-VOLTAGE ELECTRIC MOTORS IN ELECTRIC POWER PLANTS. Bulletin of Shakarim University. Technical Sciences. 2026;(3(23)):369-378. https://doi.org/10.53360/2788-7995-2026-3(23)-35

Views: 16

JATS XML


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


ISSN 2788-7995 (Print)
ISSN 3006-0524 (Online)
X