Rolling element bearings are important to rotating machinery in industries, failure of which results in significant economic damage and safety risks. The traditional vibration analysis relies on manual feature extraction and human judgment, which collapse on non-stationary signals and noise. The proposed paper presents a 1D Convolutional Neural Network (1D-CNN) and a Bidirectional Long Short-Term Memory (BiLSTM) network that is an end-to-end automated diagnostic system. The 1D- CNN generates strong spatial characteristics to raw vibration data, the BiLSTM encodes long-range temporal relationships to periodic fault impulses, and a self-attention module gives precedence to important frequency bands. The hybrid model was tested on Case Western Reserve University (CWRU) dataset to provide 99.3% classification accuracy, which is higher than traditional SVM and standard CNN. It is resistant to changes in load and provides a reliable instrument of predictive maintenance in real-time in Industry 4.0.

The implementation of spatiotemporal deep learning models brings a tremendous change in the industrial maintenance. The proposed system provides a solution to bearer health monitoring that is resilient to industrial noise and operational variability through a combination of the automated feature extraction of CNNs and the temporality of BiLSTMs.

[1]R. Liu, B. Yang, E. Zio, and X. Chen, “Artificial intelligence for fault diagnosis of rotating machinery: A review,” Mechanical Systems and Signal Processing, vol. 108, pp. 33-58, 2018. [2]W. Zhang, C. Li, G. Peng, Y. Chen, and Z. Zhang, “A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment,” IEEE Access, vol. 5, pp. 20401-20412, 2017. [3]X. Li, W. Zhang, and Q. Ding, “Deep learning-based remaining useful life estimation of bearings using multi- scale feature extraction,” Reliability Engineering & System Safety, vol. 185, pp. 212-222, 2019. [4]S. Albright, “Case Western Reserve University Bearing Data Center Website,” [Online]. Available: engineering.case.edu. [Accessed: March 2026]. [5]Y. Lei, B. Yang, X. Jiang, and F. Jia, “Applications of machine learning to machine fault diagnosis: A review and comparative study,” Mechanical Systems and Signal Processing, vol. 133, p. 106290, 2020. [6]J. Wang, P. Fu, and R. X. Gao, “Hybrid deep learning for fault diagnosis of rolling element bearings,” IEEE Transactions on Instrumentation and Measurement, vol. 69, no. 9, pp. 6408-6417, 2020. [7]H. Shao, H. Jiang, Y. Lin, and X. Li, “A novel deep convolutional neural network for fault diagnosis of rotating machinery,” IEEE Access, vol. 6, pp. 18721- 18733, 2018.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close