Detection of rotating looseness faults in electric motors using triaxial vibration data

• Thiago Carvalho Bittencourt UFSC
• Bruno Forte de Lunardi Pinto UFSC
• João Paulo Vieira UFSC
• Victor Afonso Bauler UFSC
• Diego Marques do Carmo UFSC
• Karoline Da Rocha UFSC
• Danilo Silva UFSC

DOI: https://doi.org/10.5753/eniac.2026.31232

Resumo

This study investigates the detection of rotating looseness in industrial electric motors using a private dataset of triaxial vibration signals. Following a step of data cleaning and preprocessing, four classical models utilizing expert-engineered features are benchmarked against a deep learning (DL) architecture. To ensure robust evaluation and prevent data leakage, we implement a strict machine-based data splitting strategy combined with Double Cross-Validation (CVM-CV). Experimental results demonstrate that incorporating triaxial data consistently yields superior predictive performance compared to single-axis inputs. Furthermore, the classical models outperformed the DL approach, with LightGBM achieving the best overall results.

Referências

Aleinik, S. and Matveev, Y. (2014). Detection of clipped fragments in speech signals. Int. J. Electr. Electron. Sci. Eng, 8(2):74–80.

Ali, A. and Abdelhadi, A. (2022). Condition-based monitoring and maintenance: state of the art review. Applied Sciences, 12(2):688.

Arshad, M. A., Shahriar, S., and Anjum, K. (2023). The power of simplicity: Why simple linear models outperform complex machine learning techniques – case of breast cancer diagnosis.

Ayas, G. N., Goktas, T., Arkan, M., and Lee, S. B. (2026). Multi-axis accelerometer based evaluation and classification of misalignment, soft foot, and looseness faults in induction motors. IEEE Sensors Journal.

Barreto, L. d. T., Rosa, R. K., Silva, D., and Braga, D. (2023). Fault detection for rotating machinery based on vibration data using machine learning. In Encontro Nacional de Inteligência Artificial e Computacional (ENIAC), pages 242–256. SBC.

Battulga, B., Shaikh, M. F., Chun, J. W., Park, S. B., Shim, S., and Lee, S. B. (2025a). Wireless sensor-embedded leakage flux and vibration measurement for diagnosis of electrical and mechanical faults in induction motors. In 2025 IEEE Symposium on Diagnostics for Electric Machines, Power Electronics and Drives (SDEMPED), pages 01–07. IEEE.

Battulga, B., Shaikh, M. F., Goktas, T., Arkan, M., and Lee, S. B. (2025b). Vibration-based detection and classification of mechanical defects in induction motor-driven systems during the starting transient. IEEE Transactions on Industry Applications.

Borges, V. K. (2023). Análise de transformações bidimensionais de sinais de vibração para classificação de defeitos e falhas em rolamentos usando redes neurais convolu cionais. Trabalho de conclusão de curso, Universidade Federal de Santa Catarina, Florianópolis, SC. Orientador: Danilo Silva.

Guillame-Bert, M., Bruch, S., Stotz, R., and Pfeifer, J. (2023). Yggdrasil decision forests: A fast and extensible decision forests library. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 4068–4077.

Junior, R. F. R., dos Santos Areias, I. A., Campos, M. M., Teixeira, C. E., da Silva, L. E. B., and Gomes, G. F. (2022). Fault detection and diagnosis in electric motors using 1d convolutional neural networks with multi-channel vibration signals. Measurement, 190:110759.

Kaufman, S., Rosset, S., Perlich, C., and Stitelman, O. (2012). Leakage in data mining: Formulation, detection, and avoidance. ACM Transactions on Knowledge Discovery from Data (TKDD), 6(4):1–21.

Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30.

Ribeiro Junior, R. F., Areias, I. A. d. S., and Gomes, G. F. (2021). Fault detection and diagnosis using vibration signal analysis in frequency domain for electric motors considering different real fault types. Sensor Review, 41(3):311–319.

Saeed, S. and Reddy, B. (2025). Diagnostics of vibrations due to looseness fault and unbalance in rotating machinery with neural network. Diagnostyka, 26.

Sehri, M. and Dumond, P. (2024). University of ottawa constant and variable speed electric motor vibration and acoustic fault signature dataset. Data in brief, 53:110144.

Semenova, L., Rudin, C., and Parr, R. (2022). On the existence of simpler machine learning models. In 2022 ACM Conference on Fairness Accountability and Transparency, FAccT ’22, page 1827–1858. ACM.

Suechoey, B., Siriporananon, S., Chupun, P., Boonkhun, C., and Chompooinwai, C. (2021). Performance analysis and fault classification in a large electric motor using vibration assessment technique. International Journal of Intelligent Engineering and Systems, 14(1):124–133.

Tsamardinos, I., Rakhshani, A., and Lagani, V. (2015). Performance-estimation properties of cross-validation-based protocols with simultaneous hyper-parameter optimization. International Journal on Artificial Intelligence Tools, 24(05):1540023.

Vieira, J. P., Bauler, V. A., Rosa, R. K., and Silva, D. (2026). Towards a more realistic evaluation of machine learning models for bearing fault diagnosis. Mechanical Systems and Signal Processing, 258:114640.

Wheat, L., Mohrenschildt, M. V., Habibi, S., and Al-Ani, D. (2024). Impact of data leakage in vibration signals used for bearing fault diagnosis. IEEE Access, 12:169879–169895.

Zhang, W., Peng, G., Li, C., Chen, Y., and Zhang, Z. (2017). A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals. Sensors, 17(2):425.