Generalized bearing fault diagnosis using comparative time-frequency scalogram features and machine learning with advanced feature ranking
DOI:
https://doi.org/10.15282/ijame.23.1.2026.20.1018Keywords:
Bearing faults, Time-frequency analysis, Wavelet transform, KNN, Bagged TreeAbstract
Bearings are critical components of rotating machinery, ensuring smooth operation by supporting shafts and loads. Even minor defects can severely degrade performance, causing unplanned downtime and economic loss. Accurate and timely fault diagnosis is therefore essential for condition-based maintenance. Conventional time- and frequency-domain methods often fail to capture the non-stationary and transient nature of bearing vibration signals. This work proposes a time–frequency diagnostic framework using Continuous Wavelet Transform (CWT)–based scalograms and machine learning. Coiflet and Morlet wavelets are selected using the Maximum Relative Wavelet Energy (MRWE) criterion, yielding peak MRWE values of 0.1347 for the CWRU dataset and 0.0584 for the Machinery Fault Simulator (MFS) dataset, ensuring optimal fault localization. Scalograms are converted into RGB images, from which 21 texture features are extracted and ranked using the Fisher score. The method is validated on two independent datasets: the benchmark Case Western Reserve University (CWRU) data and an experimental MFS dataset using SKF 6004 bearings with induced inner-race, outer-race, and ball defects. Logistic Regression, SVM, KNN, and Bagged Tree classifiers are evaluated using tenfold cross-validation. On the CWRU dataset, KNN and Bagged Tree achieve accuracies of 98.3% with eight ranked features, while the Bagged Tree reaches 100% accuracy using all 21 features.
References
[1] Z. Zhai, L. Luo, Y. Chen, X. Zhang, “Rolling bearing fault diagnosis based on a synchrosqueezing wavelet transform and a transfer residual convolutional neural network”. Sensors, 25(2), p.325, 2025.
[2] H. Lv, Z. Dong, X. Li, “A fault diagnosis method for gas turbine rolling bearings with variable speed based on dynamic time-varying response and joint attention mechanism”. Sensors, 25(21), 6617, 2025.
[3] Y. Li, C.Liu, W. Wang, “A bearing fault diagnosis method based on M-SSCNN and multi-layer attention mechanism”. Structural Health Monitoring, 24(1), 2025.
[4] P.K. Sahu R.N. Rai, "Fault diagnosis of rolling bearing based on an improved denoising technique using complete ensemble empirical mode decomposition and adaptive thresholding method," Journal of Vibration Engineering and Technologies, vol. 11, no. 2, pp. 513–535, 2023.
[5] J. Du, X. Li, Y. Gao, L. Gao, “Integrated Gradient-Based Continuous Wavelet Transform for Bearing Fault Diagnosis,” Sensors, vol. 22, no. 22, p. 8760, 2022.
[6] S. Li, Y. Peng, Y. Shen, S. Zhao, H. Shao, G. Bin, et al., “Rolling‐bearing fault diagnosis under data imbalance and variable speed based on adaptive clustering weighted oversampling,” Reliability Engineering and System Safety, vol. 242, p. 109010, 2023
[7] Y. Ma, J. Cheng, P. Wang, J. Wang, Y. Yang, “Rotating Machinery Fault Diagnosis Based on Multivariate Multiscale Fuzzy Distribution Entropy and Fisher Score,” Measurement, vol. 2021, p. 109495, 2021.
[8] J. Zhou, M. Xiao, Y. Niu, G. Ji, "Rolling bearing fault diagnosis based on WGWOA-VMD-SVM," Sensors, vol. 22, p. 6281, 2022.
[9] A. Orhan, N. Yordanov, M. Ertargin, M. Zhilevski, M. Mikhov, “A Comparative Study of Time–Frequency Representations for Bearing and Rotating Fault Diagnosis Using Vision Transformer”, Machines, vol. 13, no. 8, p. 737, 2025.
[10] J. Liang, X, Mao, "Rectifier fault diagnosis based on euclidean norm fusion multi-frequency bands and multi-scale permutation entropy." Electronics., vol. 14.3, p. 612, 2025.
[11] Y. Zhang, S. Cheng, J. Luo, "Rolling bearing fault diagnosis using continuous wavelet transform and convolutional neural network," Mechanical Systems and Signal Processing, vol. 172, p. 109034, 2022.
[12] V. Suthar, V. Vakharia, V.K. Patel, M. Shah, “Detection of Compound Faults in Ball Bearings Using Multiscale-SinGAN, Heat Transfer Search Optimization, and Extreme Learning Machine,” Machines, vol. 11, no. 1, p. 29, 2023.
[13] J. Du, X. Li, Y. Gao, L. Gao, “Integrated Gradient-Based Continuous Wavelet Transform for Bearing Fault Diagnosis,” Sensors, vol. 22, p. 8760, 2022.
[14] B. Kim, J. Lee, “Fault Diagnosis and Noise Robustness Comparison of Rotating Machinery using CWT and CNN,” Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 1279-1285, 2021.
[15] K. Gao, Z. Wu, C. Yu, "Composite fault diagnosis of rolling bearings: A feature selection approach based on the causal feature network," Appl. Sci., vol. 13, no. 16, p. 9089, 2023.
[16] M. Wang, Z. Zhang, X. Li, "Bearing fault detection by using graph autoencoder and ensemble learning," Sensors, vol. 24, no. 21, p. 7589, 2024.
[17] M.F. Siddique, M.A. Ganaie, S.S.S.R. Depuru, "Advanced bearing-fault diagnosis and classification using Mel-transformed scalograms obtained from vibrational signals," Journal of Vibration Engineering and Technologies, vol. 12, no. 1, pp. 1931–1942, 2024.
[18] M. Altaf, M. A. Ganaie, and S. S. S. R. Depuru, "A new statistical features based approach for bearing fault diagnosis using vibration signals," Applied Sciences, vol. 13, no. 16, p. 9089, 2023.
[19] R.K. Mishra, A. Choudhary, S. Fatima, A.R. Mohanty, B.K. Panigrahi, "A self-adaptive multiple-fault diagnosis system for rolling element bearings." Measurement Science and Technology., vol. 33.12, p. 125018, 2022.
[20] J. Yang, C. Zhou, X. Li, "A novel feature extraction method based on symbol-scale entropy for rolling bearing fault diagnosis," Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, vol. 237, no. 15, pp. 3582–3595, 2023.
[21] Y. Gao, Z. Ahmad, J.-M. Kim, “Fault Diagnosis of Rotating Machinery Using an Optimal Blind Deconvolution Method and Hybrid Invertible Neural Network,” Sensors, vol. 24, no. 1, p. 256, 2024.
[22] H.P. Raturi, S. Kushari, P.K. Karsh, S. Dey, “Evaluating stochastic fundamental natural frequencies of porous functionally graded material plate with even porosity effect: a multi-machine learning approach,” Journal of Vibration Engineering & Technologies, vol. 12, no. 2, pp. 1931-1942, 2024.
[23] Vaishali, P.K. Karsh, S. Kushari, R.R. Kumar, S. Dey, “Stochastic free vibration and impact responses of functionally graded plates: a support vector machine learning model approach,” Journal of Vibration Engineering and Technologies, vol. 11, no. 7, pp. 2927–2943, 2023.
[24] P. K. Karsh, S. Dey, “Comparison of multiple surrogate models probing uncertainty in natural frequency of hybrid functionally graded sandwich cylindrical shells,” Journal of Mechanics of Materials and Structures, vol. 17, no. 2, pp. 97–121, 2022.
[25] N. Lei, F. Huang, C. Li, "Rolling bearing fault diagnosis based on variational mode decomposition and weighted multidimensional feature entropy fusion." Journal of Vibroengineering, vol. 26.3, p. 590-614, 2024.
[26] H. Luo, Y. Zhang, X. Liu, "A lightweight and small sample bearing fault diagnosis algorithm based on probabilistic decoupling knowledge distillation and meta-learning," Journal of Vibration Engineering and Technologies, vol. 12, no. 2, pp. 1931–1942, 2024
[27] W. Li, Y. Zhang, Z. Zhang, "An orthogonal wavelet transform-based K-nearest neighbour classifier for bearing fault diagnosis," Computers and Electrical Engineering, vol. 98, p. 107585, 2022.
[28] J. Lu, X. Zhang, Y. Zhang, "Enhanced K-nearest neighbour for intelligent fault diagnosis of rolling bearings," Electronics, vol. 11, no. 3, p. 919, 2021.
[29] A. Alhams, A. Abdelhadi, Y. Badri, S. Sassi, J. Renno, "Enhanced bearing fault diagnosis through trees ensemble method and feature importance analysis," Journal of Vibration Engineering and Technologies, vol. 12, no. 1, pp. 1931–1942, 2024.
[30] Z. Mian, Y. Zhang, H. Zhang, "A literature review of fault diagnosis based on ensemble learning," Journal of Intelligent Manufacturing, vol. 35, no. 5, pp. 1245–1263, 2024.
[31] H. Yang, Y. Zhang, X. Li, "Deep ensemble learning with non-equivalent costs of fault diagnosis," Mechanical Systems and Signal Processing, vol. 146, p. 107019, 2021.
[32] S. Gao, J. Liu, Z. Zhang, "Fault diagnosis of rolling bearings based on improved energy entropy and vibration amplitude attenuation," Mechanical Systems and Signal Processing, vol. 157, p. 107661, 2021.
[33] K.A. Loparo, M.L. Adams, W. Lin, M.F. Abdel-Magied, N. Afshari, "Fault detection and diagnosis of rotating machinery," IEEE Transactions on Industrial Electronics, vol. 47, no. 5, pp. 1005–1014, 2000.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 The Author(s)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
