Bridge damage identification based on convolutional autoencoders and extreme gradient boosting trees

Abstract

To enhance the accuracy and efficiency of bridge damage identification, a novel data-driven damage identification method was proposed. First, convolutional autoencoder(CAE)was used to extract key features from the acceleration signal of the bridge structure through data reconstruction. The extreme gradient boosting tree(XGBoost)was then used to perform analysis on the feature data to achieve damage detection with high accuracy and high performance. The proposed method was applied in a numerical simulation study on a three-span continuous girder and further validated experimentally on a scaled model of a cable-stayed bridge. The numerical simulation results show that the identification errors remain within 2.9% for six single-damage cases and within 3.1% for four double-damage cases. The experimental validation results demonstrate that when the tension in a single cable of the cable-stayed bridge decreases by 20%, the method accurately identifies damage at different cable locations using only sensors installed on the main girder, achieving identification accuracies above 95.8% in all cases. The proposed method shows high identification accuracy and generalization ability across various damage scenarios.

Keywords

structural health monitoring / damage identification / convolutional autoencoder(CAE) / extreme gradient boosting tree(XGBoost) / machine learning

References

[1] Zhu H P, Shen Z H, Weng S. Damage identification for vertical stiffness of joints of periodic continuous beams based on spectral element method[J]. Journal of Southeast University(English Edition), 2023, 39(4): 323-332. DOI: 10.3969/j.issn.1003-7985.2023.04.001.

[2] Jin W L, Yu Y F, Bai Y L. Life prediction and sensitivity analysis of reinforced concrete beams after corrosion and fatigue damage[J]. Journal of Southeast University(Natural Science Edition), 2024, 54(2): 260-267. DOI:10.3969/j.issn.1001-0505.2024.02.002. (in Chinese)

[3] Tu Y M, Lu S L, Wang C. Damage identification of steel truss bridges based on deep belief network[J]. Journal of Southeast University(English Edition), 2022, 38(4): 392-400. DOI: 10.3969/j.issn.1003-7985.2022.04.008.

[4] Sun L M, Shang Z Q, Xia Y, et al. Review of bridge structural health monitoring aided by big data and artificial intelligence: From condition assessment to damage detection[J]. Journal of Structural Engineering, 2020, 146(5): 04020073. DOI: 10.1061/(asce)st.1943-541x.0002535.

[5] An Y H, Chatzi E, Sim S H, et al. Recent progress and future trends on damage identification methods for bridge structures[J]. Structural Control and Health Monitoring, 2019, 26(10): e2416. DOI: 10.1002/stc.2416.

[6] Karimi S, Mirza O. Damage identification in bridge structures: Review of available methods and case studies[J]. Australian Journal of Structural Engineering, 2023, 24(2): 89-119. DOI: 10.1080/13287982.2022.2120239.

[7] Yang D H, Sun J Z, Yi T H, et al. Early warning technology of long-span bridge bearing deterioration considering time lag effects of thermal-induced displacement[J]. Journal of Southeast University(Natural Science Edition), 2024, 54(2): 268-274. DOI:10.3969/j.issn.1001-0505.2024.02.003. (in Chinese)

[8] Sun H S, Song L, Yu Z W. A deep learning-based bridge damage detection and localization method[J]. Mechanical Systems and Signal Processing, 2023, 193: 110277. DOI: 10.1016/j.ymssp.2023.110277.

[9] Wang Z L, Cha Y J. Unsupervised deep learning approach using a deep auto-encoder with a one-class support vector machine to detect damage[J]. Structural Health Monitoring, 2021, 20(1): 406-425. DOI: 10.1177/1475921720934051.

[10] Ma X R, Lin Y Z, Nie Z H, et al. Structural damage identification based on unsupervised feature-extraction via Variational Auto-encoder[J]. Measurement, 2020, 160: 107811. DOI: 10.1016/j.measurement.2020.107811.

[11] Ni Y H, Lu H, Ji C, et al. Comparative analysis on bridge corrosion damage detection based on semantic segmentation[J]. Journal of Southeast University(Natural Science Edition), 2023, 53(2): 201-209. DOI:10.3969/j.issn.1001-0505.2023.02.003. (in Chinese)

[12] Duan Y Y, Chen Q Y, Zhang H M, et al. CNN-based damage identification method of tied-arch bridge using spatial-spectral information[J]. Smart Structures and Systems, 2019, 23: 507-520. DOI: 10.12989/SSS.2019.23.5.507.

[13] Masci J, Meier U, Ciresan D, et al. Stacked convolutional auto-encoders for hierarchical feature extraction[C]// International Conference on Artificial Neural Networks. Espoo, Finland, 2011: 52-59.

[14] Friedman J H. Greedy function approximation: A gradient boosting machine[J]. The Annals of Statistics, 2001, 29(5): 1189-1232. DOI: 10.1214/aos/1013203451.

[15] Chen T Q, Guestrin C. XGBoost: A scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, California, USA, 2016: 785-794.

[16] Song Y Y, Lu Y. Decision tree methods: Applications for classification and prediction[J]. Shanghai Archives of Psychiatry, 2015, 27(2): 130-135. DOI: 10.11919/j.issn.1002-0829.215044.

[17] Sagi O, Rokach L. Approximating XGBoost with an interpretable decision tree[J]. Information Sciences, 2021, 572: 522-542. DOI: 10.1016/j.ins.2021.05.055.

[18] Asselman A, Khaldi M, Aammou S. Enhancing the prediction of student performance based on the machine learning XGBoost algorithm[J]. Interactive Learning Environments, 2023, 31(6): 3360-3379. DOI: 10.1080/10494820.2021.1928235.

[19] Lei X J, Sun L M, Xia Y. Lost data reconstruction for structural health monitoring using deep convolutional generative adversarial networks[J]. Structural Health Monitoring, 2020, 20: 2069-2087. DOI: 10.1177/1475921720959226.

[20] Li Y X, Ni P, Sun L M, et al. A convolutional neural network-based full-field response reconstruction framework with multitype inputs and outputs[J]. Structural Control and Health Monitoring, 2022, 29(7): e2961. DOI: 10.1002/stc.2961.

Funding

The National Natural Science Foundation of China(52361165658); The National Natural Science Foundation of China(52378318); The National Natural Science Foundation of China(52078459)