A Two-Stage Machine Learning Approach for Quantitative Gear Crack Detection Using Vibration Signal Analysis
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Early detection of gear tooth cracks is essential for preventing catastrophic failures in rotating machinery, yet existing approaches struggle to accurately detect small incipient cracks due to their distinct vibration characteristics compared to larger cracks. Current machine learning methods optimize for overall performance, sacrificing sensitivity to early-stage damage where preventive maintenance is most effective. This study presents a regime-aware feature-driven two-stage modeling framework employing separate polynomial Ridge regression models for small cracks and large cracks, selected through systematic correlation analysis and exhaustive grid search optimization using vibration features extracted from residual signals. The small crack model utilizes wavelet
detail coefficients, while the large crack model employs clearance factor, envelope peak, and wavelet d1 std features, both validated using Leave-One-Out cross-validation with limited training data. Simulation results demonstrate that the proposed approach achieves R2 = 0.9982 under simulated, data-scarce conditions, with performance evaluated using leave-one-out cross-validation on 19 samples.
How to Cite
##plugins.themes.bootstrap3.article.details##
Gear fault diagnosis, vibration analysis, two- stage modeling, Ridge regression, limited data, predictive maintenance
Cerrada, M., Sánchez, R. V., Cabrera, D., Zurita, G., & Li, C. (2015). Multi-stage feature selection by using genetic algorithms for fault diagnosis in gearboxes based on vibration signal. Sensors, 15(9), 23903–23926.
Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794).
Elyousfi, B., Soualhi, A., Medjaher, K., & Guillet, F. (2020). A model-based analysis of crack fault in a two-stage spur gear system. In Proceedings of the 2020 Prognostics and Health Management Conference (PHM-Besançon) (pp. 4–7). Besançon, France.
Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of Statistics, 1189–1232.
Goel, V., & Kumar, N. (2021). Vibration response analysis of healthy and cracked gears through different signal processing techniques. Vibroengineering Procedia, 39, 43–47.
Guo, J.-L., Guo, Y., Sun, S.-B., & Gao, Y. (2014). Simulation study of phase demodulation in resonance band for gear tooth cracking fault detection. In 2015 International Conference on Material Science and Applications (ICMSA-15) (pp. 795–798).
Ho, D., & Randall, R. (2000). Optimisation of bearing diagnostic techniques using simulated and actual bearing fault signals. Mechanical Systems and Signal Processing, 14(5), 763–788.
Hoerl, A. E., & Kennard, R. W. (1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1), 55–67.
Li, Z., Ma, Z., Liu, Y., Teng, W., & Jiang, R. (2015). Crack fault detection for a gearbox using discrete wavelet transform and an adaptive resonance theory neural network. Journal of Mechanical Engineering, 61(1), 63–73.
Mohammed, O. D. (2023). Crack detection of a modified high contact ratio gear. Engineering Failure Analysis, 152, 107493.
Mohammed, O. D., & Rantatalo, M. (2015). Residual signal techniques used for gear fault detection. In Current trends in reliability, availability, maintainability and safety: An industry perspective (pp. 157–163). Springer.
Mohammed, O. D., Rantatalo, M., & Aidanpää, J.-O. (2015). Dynamic modelling of gear system with gyroscopic effect and crack detection analysis. In Proceedings of the 9th IFToMM International Conference on Rotor Dynamics (pp. 1303–1314).
Naveen, G., SampathKumaran, P., Arvind, P., & Seetharamu, S. (2024). Analyzing gear blank failure: A comprehensive industrial case study. Journal of Failure Analysis and Prevention, 24(1), 83–96.
Nguyen, P.-D., Pham, T.-D., Do, D.-T.-B., Liang, J.-W., & Nguyen, T.-D. (2025). Strain energy-based gear mesh stiffness modeling and synthetic data generation for AI-driven fault diagnosis in smart manufacturing. Frontiers in Mechanical Engineering, 11, 1682102.
Omar, I., Khan, M., & Starr, A. (2022). Compatibility and challenges in machine learning approach for structural crack assessment. Structural Health Monitoring, 21(5), 2481–2502.
Rezaei, M., Poursina, M., & Rezaei, E. (2021). Experimental investigation of helical gear tooth crack location and depth detection using moving average method on transmission error. Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications, 235(10), 2266–2275.
Smola, A. J., & Schölkopf, B. (2004). A tutorial on support vector regression. Statistics and Computing, 14(3), 199–222.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.