This article proposes a data-driven method for bridge damage detection utilizing a Transformer architecture with a patching mechanism to extract damage-sensitive features from vehicle-induced vibrations (VIVs). The core concept is the implementation of the advanced bridge weigh-in-motion (BWIM) technique to enable automated data collection and processing of VIV signals from continuous monitoring data. First, the BWIM data (including force magnitude, velocity, and wheelbase) is used to selectively sample VIV events, creating the sampled VIV (SVIV) dataset, which is often induced by specific vehicle types. Subsequently, a patching-based time series Transformer model, called PTAD and trained on the SVIV dataset from intact bridge structure, is specifically designed for the unsupervised structural anomaly detection task. Finally, two damage indices (DIs) corresponding to the long-term and short-term patterns are defined based on the reconstruction error of VIV samples by the trained PTAD model. These DIs can enable continuous monitoring and real-time detection of the health status of the target bridge. Both numerical examples and laboratory experiments on a scaled steel bridge demonstrate that the proposed method can accurately and efficiently detect the presence and extent of damage in bridge structures, even with sparse sensor deployment.
EltounyKGomaaMLiangX.Unsupervised learning methods for data-driven vibration-based structural health monitoring: a review. Sensors2023; 23: 3290.
4.
WanH-PZhuY-KLuoY, et al. Unsupervised deep learning approach for structural anomaly detection using probabilistic features. Struct Health Monit2025; 24: 3–33.
5.
CorballyRMalekjafarianA.A deep-learning framework for classifying the type, location, and severity of bridge damage using drive-by measurements. Comput Aided Civ Infrastruct Eng2023; 39: 852–871.
6.
LuoKKongXZhangJ, et al. Computer vision-based bridge inspection and monitoring: a review. Sensors2023; 23: 7863.
7.
SunLShangZXiaY, et al. Review of bridge structural health monitoring aided by big data and artificial intelligence: from condition assessment to damage detection. J Struct Eng2020; 146: 04020073.
8.
OuJLiH.Structural health monitoring in mainland China: review and future trends. Structural Health Monitoring2010; 9: 219–231.
9.
ZhangFLKimCWGoiY.Efficient Bayesian FFT method for damage detection using ambient vibration data with consideration of uncertainty. Struct Control Health Monit2020; 28: e2659.
10.
HouRXiaY.Review on the new development of vibration-based damage identification for civil engineering structures: 2010–2019. J Sound Vib2021; 491: 115741.
11.
SunHSongLYuZ.A deep learning-based bridge damage detection and localization method. Mech Syst Signal Process2023; 193: 110277.
12.
LeeKJeongSSimS-H, et al. Field experiment on a PSC-I bridge for convolutional autoencoder-based damage detection. Struct Health Monit2020; 20: 1627–1643.
13.
FengDFengMQ.Output-only damage detection using vehicle-induced displacement response and mode shape curvature index. Struct Control Health Monit2016; 23: 1088–1107.
14.
MustafaSMatsumotoYYamaguchiH.Vibration-based health monitoring of an existing truss bridge using energy-based damping evaluation. J Bridge Eng2018; 23: 04017114.
15.
Abdel WahabMMde RoeckG. Damage detection in bridges using modal curvatures: application to a real damage scenario. J Sound Vib1999; 226(2): 217–235.
16.
KimJ-TRyuY-SChoH-M, et al. Damage identification in beam-type structures: frequency-based method vs mode-shape-based method. Eng Struct2003; 57–67.
17.
FigueiredoERaduLWordenK, et al. A Bayesian approach based on a Markov-chain Monte Carlo method for damage detection under unknown sources of variability. Eng Struct2014; 80: 1–10.
18.
YangDYiTLiH, et al. Correlation-based estimation method for cable-stayed bridge girder deflection variability under thermal action. J Perform Constr Facil2018; 32: 4018070.
19.
YangDYiTLiH, et al. Monitoring and analysis of thermal effect on tower displacement in cable-stayed bridge. Measurement2018; 115: 249–257.
20.
AiedHGonzálezACanteroD.Identification of sudden stiffness changes in the acceleration response of a bridge to moving loads using ensemble empirical mode decomposition. Mech Syst Signal Process2016; 66–67: 314–338.
21.
HesterDGonzálezA.A wavelet-based damage detection algorithm based on bridge acceleration response to a vehicle. Mech Syst Signal Process2012; 28: 145–166.
22.
RoveriNCarcaterraA.Damage detection in structures under traveling loads by Hilbert–Huang transform. Mech Syst Signal Process2012; 28: 128–144.
23.
SilvaMSantosASantosR, et al. Deep principal component analysis: an enhanced approach for structural damage identification. Struct Health Monit2018; 18: 1444–1463.
24.
CasasJRMoughtyJJ.Bridge damage detection based on vibration data: Past and new developments. Front Built Environ2017; 3: 1–12.
25.
CavadasFSmithIFCFigueirasJ.Damage detection using data-driven methods applied to moving-load responses. Mech Syst Signal Process2013; 39: 409–425.
26.
SanayeiMKhalooAGulM, et al. Automated finite element model updating of a scale bridge model using measured static and modal test data. Eng Struct2015; 102: 66–79.
27.
SippleJDSanayeiM.Finite element model updating using frequency response functions and numerical sensitivities. Struct Control Health Monit2013; 21: 784–802.
28.
ChangK-CKimC-W.Modal-parameter identification and vibration-based damage detection of a damaged steel truss bridge. Eng Struct2016; 122: 156–173.
29.
LuYGaoF.A novel time-domain auto-regressive model for structural damage diagnosis. J Sound Vib2005; 283: 1031–1049.
30.
WangZChaYJ.Unsupervised machine and deep learning methods for structural damage detection: a comparative study. Eng Rep2022; 7(1): e12551.
31.
KaewnuratchadasornCWangJKimCW.Neural operator for structural simulation and bridge health monitoring. Comput Aided Civ Infrastruct Eng2023; 39: 872–890.
32.
KongQXiongQXiongH, et al. Semi-supervised networks integrated with autoencoder and pseudo-labels propagation for structural condition assessment. Measurement2023; 214: 112779.
33.
LiZHeWRenW, et al. Damage detection of bridges subjected to moving load based on domain-adversarial neural network considering measurement and model error. Eng Struct2023; 293: 116601.
34.
ZhangYZhuJ.Damage detection of bridge structures under vehicle loads using moving-window correlation coefficient of signals with limited sensors. Int J Struct Stab Dyn2022; 22: 2250105.
35.
MaXLinYNieZ, et al. Structural damage identification based on unsupervised feature-extraction via variational auto-encoder. Measurement2020; 160: 107811.
36.
WangBZengYFengD.Deep learning-based damage assessment of hinge joints for multi-girder bridges utilizing vehicle-induced bridge responses. Eng Struct2025; 333: 120148.
37.
SarwarMZCanteroD.Probabilistic autoencoder-based bridge damage assessment using train-induced responses. Mech Syst Signal Process2024; 208: 111046.
38.
AvciOAbdeljaberOKiranyazS, et al. A review of vibration-based damage detection in civil structures: from traditional methods to Machine Learning and Deep Learning applications. Mech Syst Signal Process2021; 147: 107077.
39.
SarwarMZCanteroD.Deep autoencoder architecture for bridge damage assessment using responses from several vehicles. Eng Struct2021; 246: 113064.
40.
FengDSunHFengMQ.Simultaneous identification of bridge structural parameters and vehicle loads. Comput Struct2015; 157: 76–88.
41.
BSI. Mechanical vibration-road surface profiles reporting of measured data: ISO 8608. London: BSI Standards Publication, 2016.
42.
ZhangLFengDWuG.Simultaneous identification of bridge damage and vehicle parameters based on bridge strain responses. Struct Control Health Monit2022; 29: e2945.
43.
LiJ-AFengD.A comparative study of vehicle-bridge interaction dynamics with 2D and 3D vehicle models. Eng Struct2023; 292: 116493.
44.
YauJDYangJPYangYB.Wave number-based technique for detecting slope discontinuity in simple beams using moving test vehicle. Int J Struct Stab Dyn2017; 17: 1750060.
45.
MahmoudMAAbou ZaidMA.Dynamic response of a beam with a crack subject to a moving mass. J Sound Vib2002; 256: 591–603.
46.
LuoDWangX. ModernTCN: a modern pure convolution structure for general time series analysis. In: Proceedings of the international conference on learning representations (ICLR), Vienna, Austria, 7–11 May 2024. Vienna, Austria: International Conference on Learning Representations (ICLR).
47.
NieYNguyenNHSinthongP, et al. A time series is worth 64 words: long-term forecasting with transformers. In: Proceedings of the international conference on learning representations (ICLR), Kigali, Rwanda, 1–5 May, 2023. Kigali, Rwanda: International Conference on Learning Representations (ICLR).
48.
VaswaniAShazeerNParmarN, et al. Attention is all you need. In: Proceedings of the conference on neural information processing systems (NIPS), Long Beach, CA, USA, 4–9 December 2017.
49.
BaiSKolterJZKoltunV.An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. 2018.
50.
RangapuramSSSeegarMWGasthausJ, et al. Deep state space models for time series forecasting. In: Proceedings of the conference on neural information processing systems (NIPS), Montréal, Canada, 3–8 December 2018. Montréal, Canada: Conference on Neural Information Processing System (NIPS).
51.
ZengAChenMZhangL, et al. Are transformers effective for time series forecasting? In: Proceedings of the association for the advancement of artificial intelligence (AAAI), Washington, D.C., 7–14 February 2023. Washington, D.C.: AAAI Conference on Artificial Intelligence (AAAI).
52.
LuYTangLChenC, et al. Unsupervised structural damage assessment method based on response correlations. Eng Struct2024; 302: 117413.
53.
ZhangXZhouXLinM, et al. Shufflenet: an extremely efficient convolutional neural network for mobile devices In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, 18–23 June 2018. New York: IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR).
54.
WangCZhuangZQiQ, et al. Drift doesn’t matter: dynamic decomposition with diffusion reconstruction for unstable multivariate time series anomaly detection. In: Proceedings of the conference on neural information processing systems (NIPS), New Orleans, United States, 10–16 December 2023. New Orleans: Conference on Neural Information Processing System (NIPS).
55.
ZengYFengDLiJ-A, et al. Deep learning-based identification of vehicular moving forces for bridges without axle configurations. Eng Struct2024; 304: 117646.
56.
ÁsgrímssonDSGonzálezISalviG, et al. Bayesian deep learning for vibration-based bridge damage detection. In: CuryARibeiroDUbertiniF, et al. (eds) Structural health monitoring based on data science techniques. Cham: Springer, 2022, pp. 27–43.
57.
GalYG, Z. Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In Proceedings of the international conference on machine learning (ICML), Montréal, Canada, 7–12 December 2015; 2016. Montréal, Canada: Conference on Neural Information Processing System (NIPS).
58.
GonzalezIKaroumiR.BWIM aided damage detection in bridges using machine learning. J Civ Struct Health Monit2015; 5: 715–725.
59.
Calderon HurtadoAKaurKMakki AlamdariM, et al. Unsupervised learning-based framework for indirect structural health monitoring using adversarial autoencoder. J Sound Vib2023; 550, 117598.
