Innovations
Vol. 13 (2025), Issue 1, pg(s) 30-33

INNOVATIVE SOLUTIONS

Comparative Study of Bayesian-Optimized 1-D CNN, Bi-LSTM and MLP for Bearing Fault Classification from Raw Vibration Signals

  • 1 AGH University of Krakow, Poland

Abstract

This study evaluates the performance of newly designed deep-learning model—bidirectional long short-term memory network (Bi-LSTM), with baseline to a conventional multilayer perceptron (MLP)—for classification faults of rolling-element bearings from raw vibration signals. The models are benchmarked against a previously optimised one-dimensional convolutional neural network (1-D CNN), originally obtained via Bayesian hyperparameter search. A carefully selected dataset of 3600 one-second segments was captured under varying speed conditions and dynamically enhanced with Gaussian noise during processing. On the test set, the Bi-LSTM achieves 100 % accuracy, the 1-D CNN 97.9 %, and the MLP 53.3 %. Training dynamics, confusion patterns, and model complexity were thoroughly analysed, highlighting the trade-offs between accuracy, latency and deployment cost in edge-computing scenarios.

Keywords

References

  1. R. B. Randall and J. Antoni, “Rolling element bearing diagnostics—a tutorial,” Mechanical Systems and Signal Processing, vol. 25, no. 2, pp. 485–520, [2011]. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0888327010002 530
  2. R. B. Randall, “Vibration-based condition monitoring: industrial, automotive and aerospace applications”, John Wiley & Sons, [2021].
  3. A. Jabłoński, “Condition Monitoring Algorithms in MATLAB”. Kraków: Springer, [2021].
  4. U. Jachymczyk, P. Knap, and K. Lalik, “Improved intelligent condition monitoring with diagnostic indicator selection,” Sensors, vol. 25, no. 1, p. 137, [2024].
  5. U. Jachymczyk and P. Knap, “Review of feature selection methods for predictive maintenance systems,” Industry 4.0, vol. 9, no. 3, pp. 97–100, [2024].
  6. M. Gharavian, F. Almas Ganj, A. Ohadi, and H. Heidari Bafroui, “Comparison of fda-based and pca-based features in fault diagnosis of automobile gearboxes,” Neurocomputing, vol. 121, pp. 150–159, [2013], advances in Artificial Neural Networks and Machine Learning.
  7. Y. Ran, X. Zhou, P. Lin, Y. Wen, and R. Deng, “A survey of predictive maintenance: Systems, purposes and approaches,” arXiv preprint arXiv:1912.07383, pp. 1–36, [2019].
  8. T. P. Carvalho, F. A. Soares, R. Vita, R. d. P. Francisco, J. P. Basto, and S. G. Alcalá, “A systematic literature review of machine learning methods applied to predictive maintenance,” Computers & Industrial Engineering, vol. 137, p. 106024, [2019].
  9. G. Singh and S. Ahmed Saleh Al Kazzaz, “Induction machine drive condition monitoring and diagnostic research—a survey,” Electric Power Systems Research, vol. 64, no. 2, pp. 145–158, [2003]. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0378779602001 724
  10. Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A survey of convolutional neural networks: analysis, applications, and prospects,” IEEE transactions on neural networks and learning systems, vol. 33, no. 12, pp. 6999–7019, [2021].
  11. C. Gianoglio, E. Ragusa, P. Gastaldo, F. Gallesi, and F. Guastavino, “Online predictive maintenance monitoring adopting convolutional neural networks,” Energies, vol. 14, no. 15, p. 4711, [2021].
  12. W. Silva and M. Capretz, “Assets predictive maintenance using convolutional neural networks,” in 2019 20th IEEE/ACIS International conference on software engineering, artificial intelligence, networking and parallel/distributed computing (SNPD). IEEE, [2019], pp. 59–66.
  13. P. Knap, K. Lalik, and P. Bałazy, “Boosted convolutional neural network algorithm for the classification of the bearing fault form 1-d raw sensor data,” Sensors, vol. 23, no. 9, p. 4295, [2023].
  14. P. Knap and U. Jachymczyk, “Bayesian-tuned convolutional neural networks for precise bearing fault classification,” in 2024 25th International Carpathian Control Conference (ICCC). IEEE, [2024], pp. 01–05.

Article full text

Download PDF