Gear Damage Estimation Based on Impulse Accumulation Model and Feedback-Guided Domain Adaptation
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
This paper presents a gear damage estimation method developed through our participation in the PHM 2026 Data Challenge.
The main challenges in this competition are described below.
(1) Lifetime damage estimation: Participants are required not only to detect the occurrence of an anomaly but also to quantify damage progression over the entire lifecycle. Various condition indicators have been developed to detect changes in gears (Sharma & Parey, 2016). However, no established indicator has been shown to accurately capture damage progression throughout an entire lifecycle.
(2) Lack of ground-truth labels: The datasets provided for this competition do not include ground-truth labels for the target metric. Participants must create ground-truth labels based on the image data included in the training dataset, train a model, and then apply the model to the test and validation datasets. Their estimation accuracy is then evaluated by the organizers. Since accuracy is evaluated using ground-truth labels withheld by the organizers, participants are required to train a model that predicts the target metric without access to the actual labels. In real-world applications, models developed using data from one set of experiments often need to be adapted to data from other experiments conducted under different conditions, despite having few or no ground-truth labels for the target data. This competition provides a realistic simulation of such a challenging situation.
How to Cite
##plugins.themes.bootstrap3.article.details##
Data Challenge
French, G., Mackiewicz, M., & Fisher, M. (2018). Self-Ensembling for Visual Domain Adaptation. ICLR, 2018.
Hines, J. A., & Mark, W. D. (2014). Bending-fatigue damage-detection on notched-tooth spiral-bevel gears using the average-log-ratio, ALR, algorithm. Mechanical Systems and Signal Processing, vol. 43, no. 1-2, pp. 44-56. doi:10.1016/j.ymssp.2013.09.013
Kumar, A., Ma, T., & Liang, P. (2020). Understanding self-training for gradual domain adaptation. In International conference on machine learning (pp. 5468-5479). PMLR.
Lee, D.-H. (2013). Pseudo-Label: The Simple and Efficient Semi-Supervised Learning Method for Deep Neural Networks. ICML Workshop: Challenges in Representation Learning, 2013.
Lee, J., & Su, H. (2025). Rethinking industrial artificial intelligence: A unified foundation framework. International Journal of AI for Materials and Design, 2(2), 56–68.
Li, H., Haiqi, Z., & Liwei, T. (2010). Gear Fault Detection Based on Teager-Huang Transform. International Journal of Rotating Machinery. 2010. doi:10.1155/2010/502064.
Luo, Y., Baddour, N., & Liang, M. (2019). Dynamical modeling and experimental validation for tooth pitting and spalling in spur gears. Mechanical Systems and Signal Processing, vol. 119, pp. 155-181. doi:10.1016/j.ymssp.2018.09.027
Mark, W. D., Lee, H., Patrick, R., & Coker, J. D. (2010). A simple frequency-domain algorithm for early detection of damaged gear teeth. Mechanical Systems and Signal Processing, vol. 24, no. 8, pp. 2807-2823. doi:10.1016/j.ymssp.2010.04.004
Mark, W. D., Wagner, M. E., & Isaacson, A. C. (2023). Detection and diagnosis of gear-tooth pitting damage by the Average-Log-Ratio ALR algorithm. Mechanical Systems and Signal Processing, vol. 186, Article 109911. doi:10.1016/j.ymssp.2022.109911.
Sharma, V., & Parey, A. (2016). A Review of Gear Fault Diagnosis Using Various Condition Indicators. Procedia Engineering, 144, 253–263.
Su, H., & Lee, J. (2024a). Machine learning approaches for diagnostics and prognostics of industrial systems using open source data from phm data challenges: A review. International Journal of Prognostics and Health Management, 15(2).
Su, H., & Lee, J. (2024b). An advanced diagnostic model for gearbox degradation prediction under various operating conditions and degradation levels. Annual Conference of the PHM Society, vol. 16, no. 1.
Su, H., & Lee, J. (2025). Transfer learning and ensemble learning for fault diagnosis using vibration signals. In 2025 IEEE International Conference on Prognostics and Health Management (ICPHM) (pp. 1–9).
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.