This paper presents an automated multi-class industrial fault diagnostic methodology based on the Continuous Wavelet Transform (CWT) and a fine-tuned ResNet18 deep residual network. Normalised one-dimensional time-series signals for eight fault scenarios were transformed into 64 x 64 time-frequency scalograms using the Morlet wavelet. These single-channel scalograms were used to fine-tune a ResNet18 model modified to accept grayscale input and classify eight categories. The proposed approach was tested on a 1,000-sample dataset divided into three sets training, validation, and test. The experimental results showed an overall test accuracy of 92%, with a weighted precision of 0.93, a recall of 0.92, and an F1-score of 0.92. The model performed well in classification for the Bearing, Blocking, Bearing & Blocking, and No-Fault conditions, but had somewhat worse recall for compound fault classes. These results show the power and reliability of combining CWT-based time-frequency representation with deep residual learning that can be applied in automated industrial condition monitoring.
X. Ye, X. Ma, and C. Wen, “Rotating machinery fault diagnosis method by combining time-frequency domain features and CNN knowledge transfer,” Sensors, vol. 21, no. 24, p. 8168, 2021, doi: 10.3390/s21248168.
G. Xu, M. Liu, Z. Zhuofu, D. Söffker, and W. Shen, “Bearing fault diagnosis method based on deep convolutional neural network and random forest ensemble learning,” Sensors, vol. 19, no. 5, p. 1088, 2019, doi: 10.3390/s19051088.
M. F. Siddique, F. Saleem, M. Umar, C. H. Kim, and J.-M. Kim, “A hybrid deep learning approach for bearing fault diagnosis using continuous wavelet transform and attention-enhanced spatiotemporal feature extraction,” Sensors, vol. 25, no. 9, p. 2712, 2025, doi: 10.3390/s25092712.
M. F. Siddique, S. Ullah, and J. M. Kim, “A Deep Learning Approach for Fault Diagnosis in Centrifugal Pumps through Wavelet Coherent Analysis and S-Transform Scalograms with CNN-KAN,” Computers, Materials & Continua, vol. 84, no. 2, 2025.
Y. Yan, T. Liu, X. Ye, Q. Jing, and Y. Dai, “Rotating Machinery Fault Diagnosis Based on a Novel Lightweight Convolutional Neural Network,” PLoS ONE, vol. 16, no. 8, p. e0256287, 2021, doi: 10.1371/journal.pone.0256287.
Q. Zhang et al., “Intelligent fault diagnosis of rotating machinery using time-frequency images and hybrid CNN architecture,” Scientific Reports, vol. 15, p. 12344, 2025, doi: 10.1038/s41598-025-95895-x.
J. Yuan et al., “Image representation of vibration signals and its fault diagnosis using CNN,” Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2025, doi: 10.1177/09544062241284465.
Q. Verstraete et al., “Deep learning enabled fault diagnosis using time-frequency representations,” Mathematical Problems in Engineering, vol. 2017, 2017, doi: 10.1155/2017/5067651.
X. Zhu et al., “Rotor fault diagnosis using a convolutional neural network,” Mechanical Systems and Signal Processing, vol. 125, 2019, doi: 10.1016/j.ymssp.2019.01.039.
L. Pan, X. Zhang, Z. Jiang, et al., “Real-time intelligent diagnosis of co-frequency vibration faults in rotating machinery,” Chinese Journal of Mechanical Engineering, vol. 37, p. 41, 2024, doi: 10.1186/s10033-024-01021-9.
H. Wang, H. Wang, and X. Tang, “A review of deep learning in rotating machinery fault diagnosis and its prospects for port applications,” Applied Sciences, vol. 15, no. 21, p. 11303, 2025, doi: 10.3390/app152111303.
L. Zhang et al., “An imbalanced fault diagnosis method based on time-frequency feature oversampling and CNN,” International Journal of Machine Learning, vol. 13, 2022.
S. Liu et al., “Intelligent fault diagnosis of rotating machine via hybrid deep learning,” Knowledge-Based Systems, 2025.
X. Li, W. Zhang, Q. Ding, and J. Q. Sun, “Intelligent rotating machinery fault diagnosis based on deep learning using data augmentation,” Journal of Intelligent Manufacturing, vol. 31, no. 2, pp. 433-452, 2020.
S. Tang, S. Yuan, and Y. Zhu, “Deep learning-based intelligent fault diagnosis methods toward rotating machinery,” IEEE Access, vol. 8, pp. 9335-9346, 2019.
T. Luo, M. Qiu, Z. Wu, Z. Zhao, and D. Zhang, “Bearing Fault Diagnosis Based on Multi-scale Spectral Images and Convolutional Neural Network,” Journal of Vibroengineering, vol. 27, no. 6, pp. 1012-1027, 2025, doi: 10.21595/jve.2025.24934.
H. Lee and J. Yu, “A fault detection framework for rotating machinery with a spectrogram and convolutional autoencoder,” Applied Sciences, vol. 15, no. 14, p. 7698, 2025.
J. Zhang et al., “Fault diagnosis for rotating machinery based on deep learning integrating multi-domain features,” Measurement Science and Technology, 2024, doi: 10.1088/1361-6501/adc774d.
T. E. Alam, M. M. Ahsan, and S. Raman, “Multimodal bearing fault classification under variable conditions: A 1D CNN with transfer learning,” Machine Learning with Applications, vol. 21, p. 100682, 2025.
R. M. Silva, A. L. de Oliveira Vitor, M. F. Castoldi, A. Goedtel, and W. A. Souza, “Optimized Time-Frequency Analysis for Induction Motor Fault Detection Using Hybrid Differential Evolution and Deep Learning Techniques,” International Journal of Adaptive Control and Signal Processing, 2025, doi: 10.1002/acs.4075.
C. Wang, “Federated Learning with ResNet-18 for Medical Image Diagnosis,” in Proceedings of the 2023 8th International Conference on Multimedia Systems and Signal Processing (ICMSSP 2023), Shenzhen, China, pp. 1-6, ACM, New York, NY, USA, 2023, [Online]. Available: https://doi.org/10.1145/3613917.3613922.
K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778, 2016.
