Structural health monitoring plays a crucial role in ensuring the safety and resilience of engineering structures. Detecting structural anomalies is essential for maintaining the safety of citizens and the normal operation of civil infrastructures. In this study, a novel anomaly detection framework is proposed based on deep autoencoders (DAEs), information fusion, and active sensing. The framework involves exciting the structure at specific locations and collecting acceleration data. The data collected from multiple excitation and sensor locations are analyzed and fused to enhance anomaly detection performance. More specifically, an unsupervised anomaly detection framework using DAEs has been developed. Continuous wavelet transforms (CWTs) of acceleration signals are utilized to train DAEs. Information fusion strategies are proposed to enhance the robustness of the approach to both structural uncertainty and measurement noise. A comprehensive evaluation is performed to compare the performance of fully connected AEs, convolutional AEs (CAEs), variational AEs, transformer AEs, and one-class support vector machines, each trained separately on either CWTs or raw acceleration signals, to investigate the effect of input representation on anomaly detection performance. The results show that CAEs outperform other DAE-based approaches in detecting structural anomalies, achieving higher F1 scores and lower computational costs. Additionally, training DAEs with CWTs yields better performance than using acceleration time series. Numerical studies based on the ASCE benchmark structure and experimental studies based on a geodesic dome structure have been carried out to study the capabilities as well as limitations of the proposed approach. The effects of sensor and damage locations on anomaly detection performance are analyzed through damage identifiability. Solutions are proposed for the practical cases of limited sensors and insufficient data. The framework’s ability to extract information from multiple sources allows it to identify anomalies that traditional detection methods might have missed.
BaoYLiH.Machine learning paradigm for structural health monitoring. Struct Health Monit2021; 20: 1353–1372.
2.
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.
3.
HouRXiaY.Review on the new development of vibration-based damage identification for civil engineering structures: 2010–2019. J Sound Vib2021; 491: 115741.
4.
FlahMNunezIBen ChaabeneW, et al. Machine learning algorithms in civil structural health monitoring: a systematic review. Arch Comput Methods Eng2021; 28: 2621–2643.
5.
GomesGFMendezYADda Silva Lopes AlexandrinoP, et al. A review of vibration based inverse methods for damage detection and identification in mechanical structures using optimization algorithms and ANN. Arch Comput Methods Eng2019; 26: 883–897.
6.
RittoTGRochinhaFA.Digital twin, physics-based model, and machine learning applied to damage detection in structures. Mech Syst Signal Process2021; 155: 107614.
7.
PaganiAEneaMCarreraE.Component-wise damage detection by neural networks and refined FEs training. J Sound Vib2021; 509: 116255.
8.
NickHAziminejadAHosseiniMH, et al. Damage identification in steel girder bridges using modal strain energy-based damage index method and artificial neural network. Eng Fail Anal2021; 119: 105010.
9.
PadilKHBakharyNAbdulkareemM, et al. Non-probabilistic method to consider uncertainties in frequency response function for vibration-based damage detection using artificial neural network. J Sound Vib2020; 467: 115069.
10.
PadilKHBakharyNHaoH.The use of a non-probabilistic artificial neural network to consider uncertainties in vibration-based-damage detection. Mech Syst Signal Process2017; 83: 194–209.
11.
ShuJZhangZGonzalezI, et al. The application of a damage detection method using artificial neural network and train-induced vibrations on a simplified railway bridge model. Eng Struct2013; 52: 408–421.
12.
LiMJiaDWuZ, et al. Structural damage identification using strain mode differences by the iFEM based on the convolutional neural network (CNN). Mech Syst Signal Process2022; 165: 108289.
13.
AlazzawiOWangD.A novel structural damage identification method based on the acceleration responses under ambient vibration and an optimized deep residual algorithm. Struct Health Monit2022; 21: 2587–2617.
14.
SeventekidisPGiagopoulosD.Model-based damage identification with simulated transmittance deviations and deep learning classification. Struct Health Monit2022; 21: 2206–2230.
15.
ParzialeMLomazziLGiglioM, et al. Vibration-based structural health monitoring exploiting a combination of convolutional neural networks and autoencoders for temperature effects neutralization. Struct Control Health Monit2022; 29: e3076.
16.
LinMGuoSHeS, et al. Structure health monitoring of a composite wing based on flight load and strain data using deep learning method. Compos Struct2022; 286: 115305.
17.
SajediSOLiangX.Vibration-based semantic damage segmentation for large-scale structural health monitoring. Comput Aided Civ Infrastruct Eng2020; 35: 579–596.
18.
GulgecNSTakáčMPakzadSN.Convolutional neural network approach for robust structural damage detection and localization. J Comput Civ Eng2019; 33: 04019005.
19.
AbdeljaberOAvciOKiranyazS, et al. Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks. J Sound Vib2017; 388: 154–170.
20.
LinY-zNieZ-hMaH-w.Structural damage detection with automatic feature-extraction through deep learning. Comput Aided Civ Infrastruct Eng2017; 32: 1025–1046.
21.
BishehHBAmiriGGNekooeiM, et al. Damage detection of bridges based on combining efficient cepstral coefficients. J Vib Control2021; 27: 2279–2290.
22.
SajediSOLiangX.A data-driven framework for near real-time and robust damage diagnosis of building structures. Struct Control Health Monit2020; 27: e2488.
23.
CremonaCCuryAOrcesiA. Supervised learning algorithms for damage detection and long term bridge monitoring. In: IABMAS 2012 – international conference on bridge maintenance, safety and management, Stresa, Lake Maggiore, Italy, 8–12 July 2012, pp. 2144–2151. Boca Raton: CRC Press.
24.
AlvesVCuryARoitmanN, et al. Structural modification assessment using supervised learning methods applied to vibration data. Eng Struct2015; 99: 439–448.
25.
ChegeniMHSharbatdarMKMahjoubR, et al. New supervised learning classifiers for structural damage diagnosis using time series features from a new feature extraction technique. Earthq Eng Eng Vibrat2022; 21: 169–191.
26.
SvendsenBTØisethOFrøsethGT, et al. A hybrid structural health monitoring approach for damage detection in steel bridges under simulated environmental conditions using numerical and experimental data. Struct Health Monit2023; 22: 540–561.
27.
Soleimani-BabakamaliMHSoleimani-BabakamaliRSarloR.A general framework for supervised structural health monitoring and sensor output validation mitigating data imbalance with generative adversarial networks-generated high-dimensional features. Struct Health Monit2022; 21: 1167–1182.
28.
LiLMorgantiniMBettiR.Structural damage assessment through a new generalized autoencoder with features in the quefrency domain. Mech Syst Signal Process2023; 184: 109713.
29.
GiglioniVVenanziIPoggioniV, et al. Autoencoders for unsupervised real-time bridge health assessment. Comput Aided Civ Infrastruct Eng2023; 38: 959–974.
30.
HesserDFAltunKMarkertB.Monitoring and tracking of a suspension railway based on data-driven methods applied to inertial measurements. Mech Syst Signal Process2022; 164: 108298.
31.
Fernandez-NavamuelAMagalhãesFZamora-SánchezD, et al. Deep learning enhanced principal component analysis for structural health monitoring. Struct Health Monit2022; 21: 1710–1722.
32.
JiangKHanQDuX, et al. A decentralized unsupervised structural condition diagnosis approach using deep auto-encoders. Comput Aided Civ Infrastruct Eng2021; 36: 711–732.
33.
MousaviZVarahramSEttefaghMM, et al. Deep neural networks–based damage detection using vibration signals of finite element model and real intact state: an evaluation via a lab-scale offshore jacket structure. Struct Health Monit2021; 20: 379–405.
34.
ShangZSunLXiaY, et al. Vibration-based damage detection for bridges by deep convolutional denoising autoencoder. Struct Health Monit2021; 20: 1880–1903.
35.
SilvaMFSantosASantosR, et al. Damage-sensitive feature extraction with stacked autoencoders for unsupervised damage detection. Struct Control Health Monit2021; 28: e2714.
36.
WangZChaY-J.Unsupervised deep learning approach using a deep auto-encoder with a one-class support vector machine to detect damage. Struct Health Monit2021; 20: 406–425.
37.
WangZJahanshahiMRAzimiM, et al. Sensor fault detection in smart extraterrestrial habitats using unsupervised learning. AIAA J2024; 62: 3225–3243.
38.
WanH-PZhuY-KLuoY, et al. Unsupervised deep learning approach for structural anomaly detection using probabilistic features. Struct Health Monit2025; 24: 3–33.
39.
ShuXBaoTZhouY, et al. Unsupervised dam anomaly detection with spatial–temporal variational autoencoder. Struct Health Monit2023; 22: 39–55.
40.
KimMSongJ.Near-real-time identification of seismic damage using unsupervised deep neural network. J Eng Mech2022; 148: 04022006.
41.
SharmaSSenS.Bridge damage detection in presence of varying temperature using two-step neural network approach. J Bridge Eng2021; 26: 04021027.
42.
GuJGulMWuX.Damage detection under varying temperature using artificial neural networks. Struct Control Health Monit2017; 24: e1998.
43.
ZhouHFNiYQKoJM.Eliminating temperature effect in vibration-based structural damage detection. J Eng Mech2011; 137: 785–796.
44.
FigueiredoEParkGFarrarCR, et al. Machine learning algorithms for damage detection under operational and environmental variability. Struct Health Monit2011; 10: 559–572.
45.
MovsessianACavaDGTcherniakD.An artificial neural network methodology for damage detection: demonstration on an operating wind turbine blade. Mech Syst Signal Process2021; 159: 107766.
46.
NevesACGonzálezIKaroumiR, et al. The influence of frequency content on the performance of artificial neural network–based damage detection systems tested on numerical and experimental bridge data. Struct Health Monit2021; 20: 1331–1347.
47.
UmarSVafaeiMAlihSC.Sensor clustering-based approach for structural damage identification under ambient vibration. Autom Constr2021; 121: 103433.
48.
RafieiMHAdeliH.A novel unsupervised deep learning model for global and local health condition assessment of structures. Eng Struct2018; 156: 598–607.
49.
Soleimani-BabakamaliMHSoleimani-BabakamaliRSarloR, et al. On the effectiveness of dimensionality reduction for unsupervised structural health monitoring anomaly detection. Mech Syst Signal Process2023; 187: 109910.
50.
EltounyKALiangX.Large-scale structural health monitoring using composite recurrent neural networks and grid environments. Comput Aided Civ Infrastruct Eng2023; 38: 271–287.
51.
VaswaniAShazeerNParmarN, et al. Attention is all you need. arXiv:1706.03762, 2023.
52.
XuJWuHWangJ, et al. Anomaly transformer: time series anomaly detection with association discrepancy. arXiv:2110.02642, 2022.
53.
ChenLYouZZhangN, et al. UTRAD: anomaly detection and localization with U-Transformer. Neural Netw2022; 147: 53–62.
54.
WangXPiDZhangX, et al. Variational transformer-based anomaly detection approach for multivariate time series. Measurement2022; 191: 110791.
55.
MaMHanLZhouC.Research and application of Transformer based anomaly detection model: a literature review. arXiv:2402.08975, 2024.
56.
WangRShaoYLiQ, et al. A novel transformer-based semantic segmentation framework for structural condition assessment. Struct Health Monit2024; 23: 1170–1183.
57.
BenfenatiLPagliariDJZanattaL, et al. Foundation models for structural health monitoring. arXiv:2404.02944, 2025.
58.
BommasaniRHudsonDAAdeliE, et al. On the opportunities and risks of foundation models. arXiv:2108.07258, 2022.
59.
ZhuQWuQYueY, et al. Vision transformer–based anomaly detection method for offshore platform monitoring data. Struct Control Health Monit2024; 2024: 1887212.
60.
HuTMaKXiaoJ.Graph feature refinement and fusion in transformer for structural damage detection. Sensors2024; 24: 4415.
61.
OzdagliAIKoutsoukosX.Machine learning based novelty detection using modal analysis. Comput Aided Civ Infrastruct Eng2019; 34: 1119–1140.
62.
LeeKJeongSSimS-H, et al. Field experiment on a PSC-I bridge for convolutional autoencoder-based damage detection. Struct Health Monit2021; 20: 1627–1643.
63.
SarwarMZCanteroD.Deep autoencoder architecture for bridge damage assessment using responses from several vehicles. Eng Struct2021; 246: 113064.
64.
LillyJMOlhedeSC.Generalized Morse wavelets as a superfamily of analytic wavelets. IEEE Trans Signal Process2012; 60: 6036–6041.
65.
SilikANooriMAltabeyWA, et al. Comparative analysis of wavelet transform for time-frequency analysis and transient localization in structural health monitoring. Struct Durab Health Monit2021; 15: 1–22.
66.
RautelaMSenthilnathJMonacoE, et al. Delamination prediction in composite panels using unsupervised-feature learning methods with wavelet-enhanced guided wave representations. Compos Struct2022; 291: 115579.
67.
TaherSALiJJeongJ-H, et al. Structural health monitoring of fatigue cracks for steel bridges with wireless large-area strain sensors. Sensors2022; 22: 5076.
68.
ModirACastermanATanselI.Detection of anomalies in additively manufactured metal parts using CNN and LSTM networks. Recent Prog Mater2023; 5: 028.
69.
AndòBGrecoDNavarraG.A Low-cost solution and continuous wavelet transform analysis for structural health monitoring. Proceedings2024; 97: 38.
70.
ParkHSYooSHOhBK.A dynamic strain prediction method for malfunction of sensors in buildings subjected to seismic loads using CWT and CNN. Sci Rep2024; 14: 28156.
71.
NagarajaiahSBasuB.Output only modal identification and structural damage detection using time frequency & wavelet techniques. Earthq Eng Eng Vibrat2009; 8: 583–605.
72.
MaoJWangHSpencerBF.Toward data anomaly detection for automated structural health monitoring: exploiting generative adversarial nets and autoencoders. Struct Health Monit2021; 20: 1609–1626.
73.
DosovitskiyABeyerLKolesnikovA, et al. An image is worth 16x16 words: transformers for image recognition at scale. arXiv:2010.11929, 2021.
74.
HuangP-YXuHLiJ, et al. Masked autoencoders that listen. arXiv:2207.06405, 2023.
75.
HeKChenXXieS, et al. Masked autoencoders are scalable vision learners. arXiv:2111.06377, 2021.
76.
SchwartzEArbelleAKarlinskyL, et al. MAEDAY: MAE for few and zero shot AnomalY-Detection. arXiv:2211.14307, 2024.
77.
EntezamiAShariatmadarH.An unsupervised learning approach by novel damage indices in structural health monitoring for damage localization and quantification. Struct Health Monit2018; 17: 325–345.
78.
EntezamiAShariatmadarHMarianiS.Fast unsupervised learning methods for structural health monitoring with large vibration data from dense sensor networks. Struct Health Monit2020; 19: 1685–1710.
79.
SchölkopfBWilliamsonRSmolaA, et al. Support vector method for novelty detection. In: Proceedings of the 13th international conference on neural information processing systems, Denver, Colorado, USA, 29 November–4 December 1999, pp. 582–588. Denver, CO: MIT Press.
80.
WuR-TJahanshahiMR.Data fusion approaches for structural health monitoring and system identification: past, present, and future. Struct Health Monit2018; 19: 552–586.
81.
GravinaRAliniaPGhasemzadehH, et al. Multi-sensor fusion in body sensor networks: state-of-the-art and research challenges. Inf Fusion2017; 35: 68–80.
82.
KannanVDaoDVLiH.An information fusion approach for increased reliability of condition monitoring with homogeneous and heterogeneous sensor systems. Struct Health Monit2023; 22: 1601–1612.
83.
LiXLinSLawS, et al. Fusion of structural damage identification results from different test scenarios and evaluation indices in structural health monitoring. Struct Health Monit2021; 20: 2540–2565.
84.
WangXFolienteGSuZ, et al. Multilevel decision fusion in a distributed active sensor network for structural damage detection. Struct Health Monit2006; 5: 45–58.
85.
LiaoXYanQZhongH, et al. Integrating PZT-enabled active sensing with deep learning techniques for automatic monitoring and assessment of early-age concrete strength. Measurement2023; 211: 112657.
86.
ZengYWangHDengK, et al. Detection of rupture inside rubber bearings using active sensing method. Eng Struct2022; 271: 114950.
87.
GaoWZhangCChenL.Monitoring mechanical behaviors of CLT connections under reciprocating loading based on PZT-enabled active sensing and machine learning algorithms. Smart Mater Struct2023; 32: 024001.
88.
HanFJiangJXuK, et al. Damage detection of common timber connections using piezoceramic transducers and active sensing. Sensors2019; 19: 2486.
89.
NaitoHBolanderJE.Damage detection method for RC members using local vibration testing. Eng Struct2019; 178: 361–374.
90.
ZhuDYiXWangY.A local excitation and measurement approach for decentralized damage detection using transmissibility functions. Struct Control Health Monit2016; 23: 487–502.
91.
HuZXiangZLuQ.Passive tap-scan damage detection method for beam structures. Struct Control Health Monit2020; 27: e2510.
92.
IkedaTYasuiSMinamiyamaS, et al. Stable impact and contact force control by UAV for inspection of floor slab of bridge. Adv Robot2018; 32: 1061–1076.
93.
JohnsonEALamHFKatafygiotisLS, et al. Phase I IASC-ASCE structural health monitoring benchmark problem using simulated data. J Eng Mech2004; 130: 3–15.
94.
ZerveasGJayaramanSPatelD, et al. A transformer-based framework for multivariate time series representation learning. arXiv:2010.02803, 2020.
95.
LimBArikSOLoeffN, et al. Temporal fusion transformers for interpretable multi-horizon time series forecasting. Int J Forecasting2021; 37(4): 1748–1764.
96.
ErfaniSMRajasegararSKarunasekeraS, et al. High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning. Pattern Recognit2016; 58: 121–134.
97.
FalchiFGirardiMGurioliG, et al. Deep learning and structural health monitoring: temporal fusion transformers for anomaly detection in masonry towers. Mech Syst Signal Process2024; 215: 111382.
98.
SubediAQianCSadeghianR, et al. CorFormer: a hybrid transformer-CNN architecture for corrosion segmentation on metallic surfaces. Mach Vis Appl2025; 36: 45.
