Structural Control & Health Monitoring最新文献

{"title":"A Multistage Interpretable Framework for High-Speed Train Bearing Fault Diagnosis: Adaptive Domain Transfer With SE-ResMLP and Prototype Reasoning","authors":"Xiaoyan Shen,&nbsp;Hongkui Zhong,&nbsp;Yujie Gu,&nbsp;Yuqi Zhang,&nbsp;Ruiqing Han","doi":"10.1155/stc/1707014","DOIUrl":"https://doi.org/10.1155/stc/1707014","url":null,"abstract":"<p>An interpretable adaptive domain transfer learning framework is proposed to address key challenges in high-speed train bearing fault diagnosis, including strong signal-noise interference, significant variations in operating conditions, scarce field fault data, and limited model interpretability. The framework aims to transfer diagnostic knowledge from laboratory bearing data to field-measured operational data. First, raw vibration signals are preprocessed using wavelet-based denoising, linear interpolation for missing values, and Pauta criterion–based outlier detection to improve signal quality. To enhance transfer reliability, source-domain samples that are highly compatible with the target-domain distribution are selected through a feature-similarity scoring mechanism. A residual multilayer perceptron with squeeze-and-excitation attention (SE-ResMLP) is then constructed for fault diagnosis. During feature alignment, covariance alignment (CORAL), maximum mean discrepancy (MMD), and adversarial domain alignment are jointly introduced to reduce distributional discrepancies between source-domain and target-domain features. By integrating parameter transfer, unsupervised adversarial training, and temperature-scaling-based confidence calibration, the adaptability of the model to target-domain operating conditions is further improved. In addition, a multistage interpretability mechanism is established to increase diagnostic transparency. In the pretraining phase, physics-prior gating guides the model to focus on bearing fault–related features. During the transfer process, CORAL provides covariance-level alignment evidence. In the decision-making phase, calibrated Softmax probabilities are used for window-level prediction and file-level weighted voting, while a prototype-based Mahalanobis distance mechanism is applied to the learned feature embeddings to provide supplementary geometric evidence between samples and class prototypes. In the post-transfer stage, gated Grad-CAM and prototype-distance heatmaps are used to verify the consistency between model attention and physical fault mechanisms. Case study results show that the proposed framework achieves good cross-domain diagnostic performance, including an accuracy of 0.930 ± 0.012, Macro-F1 of 0.943 ± 0.013, and PR-AUC of 0.981 ± 0.008, while maintaining interpretable diagnostic evidence across pretraining, transfer, decision-making, and post-transfer stages under complex field conditions, within a transductive unsupervised domain adaptation setting.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/1707014","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Field-Based Characterization of Stay Cables: Traffic-Induced Fatigue Demand and Wind-Vibration Susceptibility","authors":"Miguel Anaya-Díaz,&nbsp;Saúl Enrique Crespo-Sánchez,&nbsp;Francisco Javier Carrión-Viramontes,&nbsp;Juan Antonio Quintana-Rodríguez,&nbsp;Andrés Antonio Torres-Acosta","doi":"10.1155/stc/4195193","DOIUrl":"https://doi.org/10.1155/stc/4195193","url":null,"abstract":"<p>Stay cables are critical components of cable-stayed bridges and are subjected to repeated stress fluctuations induced by traffic and wind, which may contribute to fatigue damage and vibration-related deterioration. This study presents a field-based experimental investigation of two stay cables of the Ing. Antonio Dovalí Jaime Bridge using simultaneous strain and acceleration measurements obtained during a controlled static load test and under normal operating conditions. Natural frequencies and modal damping ratios were identified from acceleration records, while effective flexural stiffness and mass per unit length were estimated using a mechanical cable model. Effective axial stiffness was determined from measured strain increments and corresponding tension-force variations. The identified properties were then used to estimate live-load-induced tension histories during bridge operation. Stress histories in the prestressing strands were processed using rainflow counting to obtain stress-range distributions and equivalent stress ranges, which were compared with an infinite-life fatigue threshold. Wind-vibration susceptibility was additionally assessed using the identified modal damping ratios and a Scruton-number-based reference criterion. The results indicate low traffic-induced fatigue demand during the monitored period, while the low modal damping ratios suggest potential susceptibility to wind-induced vibrations. Overall, the field measurements provide a direct characterization of stay-cable mechanical properties and support a screening-level evaluation of traffic-induced fatigue demand and wind-vibration susceptibility.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/4195193","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SE-CNN-Based Damage Detection in Welded Joints of Spatial Grid Structures","authors":"Hui Liu,&nbsp;Wenkai Sheng,&nbsp;Jianwei Huang,&nbsp;Yi Zou,&nbsp;Jing Wang","doi":"10.1155/stc/7052218","DOIUrl":"https://doi.org/10.1155/stc/7052218","url":null,"abstract":"<p>Owing to welding process-related factors and environmental influences, stress concentration is prone to occur in the welded regions of grid structure joints, rendering these joints the most damage-susceptible locations within the entire structural system. In this paper, an intelligent damage diagnosis framework that integrates deep learning techniques with joint stress field analysis is proposed to enable damage diagnosis of structural joints. A quasidistributed fiber Bragg grating (FBG) sensor network deployed at the joints facilitated the construction of a dual-dimensional damage index, characterized by high sensitivity to damage location and severity. The proposed index enables precise identification of joint damage by capturing the abrupt changes in the local stress field caused by joint failure. By leveraging the spatial distribution patterns of the damage index, a convolutional neural network (CNN) model incorporating the squeeze-and-excitation (SE) attention mechanism was developed, with Bayesian optimization algorithms employed for hyperparameter tuning to perform simultaneous localization and quantification of joint damage. Moreover, the model exhibits robust noise tolerance, retaining 93.8% identification accuracy at a signal-to-noise ratio (SNR) above 10 dB without requiring preprocessing denoising. To validate the feasibility of the proposed SE-CNN model, a spatial grid structure model with adjustable damage joints and controllable damage levels was designed and fabricated, and identification tests were conducted under various joint damage conditions. The damage diagnosis accuracy for joint damage in the spatial grid structure reached 97.2% using the proposed method with damage indices derived from stress data measured by quasidistributed FBG sensors. Therefore, the SE-CNN model established based on the characteristics of the damage index proves to be an effective method for joint damage diagnosis in spatial grid structures.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/7052218","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Composite Active Disturbance Rejection Vibration Isolation Control Based on Compensation for Unknown Disturbance","authors":"Zhijun Zhu,&nbsp;Naoxi Dou,&nbsp;Kun Zheng,&nbsp;Yongli Lou,&nbsp;Bin Li,&nbsp;Yong Xiao","doi":"10.1155/stc/6814922","DOIUrl":"https://doi.org/10.1155/stc/6814922","url":null,"abstract":"<p>Active vibration isolation is widely recognized as an effective approach for suppressing low-frequency vibrations. Linear active disturbance rejection control (LADRC) serves as a powerful strategy for active vibration isolation. However, classic LADRC depends on displacement sensors, which are often bulky and prone to measurement instability. Substituting accelerometers for displacement sensors offers a promising alternative, as the former are generally more compact and provide superior measurement stability. Nevertheless, the integration of accelerometers into classic LADRC requires double integration to derive displacement, leading to low-frequency drift that may drive the system out of control. To overcome such limitation, this paper proposes a composite ADRC algorithm with a collaborative disturbance observer (CDO). The CDO integrates reduced-order LESO (RLESO) and Kalman filter (KF) in a complementary manner, which directly uses acceleration as input to avoid the risk of integral drift and realizes the functional complementarity between KF in state reconstruction and RLESO in disturbance estimation. Moreover, a composite scheme combining known-disturbance feedforward, unknown-disturbance feedback, and state errors feedback is adopted. Theory and experiments show that this method can achieve effective state reconstruction and disturbance suppression with only accelerometers while exhibiting good tolerance to small control-loop time delays, providing a streamlined and efficient solution for low-frequency broadband vibration isolation.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/6814922","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Interpretable Multisensor Data Fusion Framework for Intelligent Monitoring of Historical Bridges","authors":"Quanjun Hua,&nbsp;Qing Chun,&nbsp;Yu Yuan,&nbsp;Xiaoyang Sun","doi":"10.1155/stc/5573568","DOIUrl":"https://doi.org/10.1155/stc/5573568","url":null,"abstract":"<p>Historical bridges are often subjected to long-term environmental actions and gradual material degradation, while practical constraints on inspection and intervention make reliable condition assessment highly challenging. In this context, intelligent monitoring systems generate large volumes of heterogeneous sensor data, yet transforming such data into interpretable and actionable information for engineering decision-making remains a key problem. This study proposes an interpretable multisensor data fusion framework for intelligent monitoring of historical bridges, integrating temporal prediction and dependency modelling to achieve system-level anomaly assessment under strong environmental influence. Long short–term memory (LSTM) networks are employed to model the temporal evolution of individual sensor signals and to extract prediction residuals that represent deviations from expected behaviour. To capture nonlinear and non-Gaussian dependencies among heterogeneous sensors, a Copula-based dependency model is constructed on the residual space, enabling coordinated anomaly identification across structural and environmental measurements. A hierarchical fusion strategy is adopted to balance local sensor-group interactions and global system-level behaviour. Furthermore, Shapley value analysis is introduced to quantify the contribution of different sensor groups to detected anomalies, enhancing interpretability and supporting engineering judgement. The framework is validated using 34 months of continuous monitoring data from Lugou Bridge, a historical masonry arch bridge. The results demonstrate that system-level fusion effectively reduces scattered single-sensor alarms and highlights temporally concentrated anomalous periods consistent with documented extreme environmental events. The proposed framework is designed as a generalisable and interpretable solution for long-term multisensor monitoring, with the historical bridge serving as a representative case study.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/5573568","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Multilevel Ensemble Intelligent Framework for Rapid Seismic Damage-State Classification of Mega-Sub Controlled Structural Systems Under Data-Scarce Conditions","authors":"Xinwei Wang,&nbsp;Muhammad Moman Shahzad,&nbsp;Hanmin Wang,&nbsp;Shuaiqiang Wei,&nbsp;Haifeng Wang,&nbsp;Shixuan Yang,&nbsp;Tianlong Wang,&nbsp;Zhihao Wang","doi":"10.1155/stc/2554157","DOIUrl":"https://doi.org/10.1155/stc/2554157","url":null,"abstract":"<p>To address the difficulties in rapid postearthquake damage-state identification of complex high-rise structures, the high computational cost of nonlinear time-history analysis, and the limited availability of postearthquake samples, this study proposes an adaptive multilevel ensemble intelligent framework for mega frame structure (MFS) and mega-sub controlled structural system (MSCSS), aiming to achieve rapid and stable seismic damage assessment under data-scarce conditions. The proposed framework adopts regularized kernel extreme learning machine (RKELM) as the core classifier, where kernel mapping is used to characterize the complex nonlinear mapping relationship between ground motion intensity measures and structural damage states, and regularization is introduced to reduce the risk of overfitting under small-sample conditions. On this basis, a bidirectional beetle swarm adaptive optimization (BBSAO) algorithm is developed to achieve stable optimization of the kernel parameter and regularization coefficient of RKELM through bidirectional subpopulation cooperative search and adaptive step-size adjustment. Furthermore, a confusion-focused multiclass boosting (CFMBoost) mechanism is constructed to dynamically adjust sample weights according to class confidence and classification difficulty, enabling the base learners to focus more on easily confused and damage-sensitive states. The proposed framework is validated using the nonlinear responses and damage states of MFS and MSCSS under different ground motion inputs. The results show that the proposed framework achieves high accuracy and stability in small-sample damage-state classification, providing an efficient and reliable intelligent approach for rapid postearthquake damage assessment and emergency decision-making of complex high-rise structures.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/2554157","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Approach for Bending Stiffness Inversion and Rapid Evaluation of Continuous Beam Bridges Based on a Deflection Influence Line","authors":"Yu Zhou,&nbsp;Zhe Feng,&nbsp;Sheng-Kui Di,&nbsp;Zhi-Jun Jia","doi":"10.1155/stc/1361157","DOIUrl":"https://doi.org/10.1155/stc/1361157","url":null,"abstract":"<p>To address the issues of traffic interruption and safety risks associated with high loading efficiency in bridge static load tests, this study proposes a rapid assessment method for bridge bending stiffness by inverting deflection influence lines. This method identifies quasistatic deflection influence lines of beam bridges under idle-speed loading of single heavy vehicles through variational mode decomposition (VMD) and vehicle‒bridge coupling matrix analysis. A conversion function between deflection influence lines and sectional bending stiffness is established to invert the bending stiffness of each bridge section. The accuracy of the conversion function is verified by solving the bending stiffness through modal eigenvalue functions of uniform cross-section continuous beam bridge models. A bending stiffness solution method for variable cross-section continuous beam bridges is developed based on the Gauss‒Legendre formula. Experimental validation is conducted on a variable cross-section continuous beam bridge, where the full-span sectional bending stiffness is inverted using measured deflection influence lines and compared with the finite element model-derived stiffness. Four bending stiffness evaluation indicators are proposed considering both the sectional characteristics and full-span contributions of the stiffness distribution. This work demonstrates that the proposed stiffness inversion method can rapidly and intuitively reflect the distribution characteristics of the full-span sectional stiffness. Among these, two redundancy coefficient indicators effectively characterize the bending stiffness redundancy status in both the measurement span and adjacent spans. A comparative analysis with static load tests reveals that the reciprocals of the two verification coefficient indicators yield more conservative results than do the static load test verification coefficients. The sectional stiffness verification coefficient reciprocal is in better agreement with the static load test results, whereas the full-span stiffness verification coefficient considering full-span contributions aligns more closely with practical conditions. This study establishes a novel rapid assessment methodology for bridge bending stiffness, providing theoretical support and methodological references for rapid safety evaluation of bridge structures.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/1361157","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145777","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Dual-Branch Neural Network for High-Load Strain Extrapolation Using Finite Element and Experimental Data","authors":"Haotian Chu,&nbsp;Qiliang Zhao,&nbsp;Haoming Xu,&nbsp;Fangqiang Yu,&nbsp;Chao Yang,&nbsp;Yaozhi Luo,&nbsp;Nikolai Ivanovich Vatin,&nbsp;Bochao Sun","doi":"10.1155/stc/3970814","DOIUrl":"https://doi.org/10.1155/stc/3970814","url":null,"abstract":"<p>Under high-load conditions, accurate strain prediction remains challenging due to nonlinear responses, noise, and incomplete measurements. This study proposes a pointwise dual-branch neural regression model using multisource data. A 1:8 scale-reinforced concrete frame was modeled in ABAQUS, and based on measurements up to 1.35 times the design load, three loading conditions were simulated to generate finite element (FE) strain data. The model combines an encoder-style nonlinear branch implemented using a Transformer encoder block with an MLP branch representing the dominant load–strain trend, while a dynamically weighted loss increases emphasis on large-strain samples. The best tested configurations achieve validation <i>R</i>² values of 0.9975 and 0.9997, with high-load MAPE values of 2.56% for experimental S7 extrapolation and 0.39% for held F5-at-1.0× cross-condition FE prediction. Under high-noise conditions, multisource data fusion increases validation <i>R</i>² from 0.9621 to 0.9965 while reducing high-load MAPE from 11.15% to 3.09%. These results demonstrate stable prediction performance within the tested benchmark.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/3970814","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AE Criteria for the Loss of Load-Bearing Capacity of Prestressing Bars After Thermal Exposure–Microstructural Validation","authors":"Anna Adamczak-Bugno,&nbsp;Sebastian Lipiec,&nbsp;Piotr Matusiewcz,&nbsp;Jakub Adamczak,&nbsp;Peter Koteš,&nbsp;František Bahleda","doi":"10.1155/stc/8228957","DOIUrl":"https://doi.org/10.1155/stc/8228957","url":null,"abstract":"<p>The paper presents the results of experimental studies on the effect of thermal exposure (500°C and 700°C) on the microstructure, mechanical properties and acoustic emission (AE) signatures of cold-drawn prestressing bars. Specimens in the as-received state (RT) and after annealing at 500°C and 700°C were subjected to uniaxial tensile tests to failure, with simultaneous AE signal acquisition. The analysis was based on a three-parameter diagnostic approach employing the Ib-value index, the number of counts and the amplitude–duration relationship, enabling quantitative detection of the moment of load-bearing capacity loss. It was shown that for RT specimens, the failure process is ductile with late crack localisation; for S_500 specimens, it is abrupt, while for S_700 specimens, it is dispersed and microplastic. Changes in AE parameters exhibit a clear correlation with microstructural evolution: from lamellar pearlite in the RT state, through partial spheroidisation and lamella fragmentation at 500°C, to complete spheroidisation and cementite coarsening at 700°C. A decision criterion was developed, based on a permanent exceedance of the stability threshold in at least two out of the three analysed AE parameters, which enabled consistent identification of the failure moment within the investigated specimens and constitutes a potential diagnostic tool for assessing the integrity of steel components under and after fire exposure.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/8228957","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid Eigensensitivity-Machine Learning Framework for Model Updating of a Generally Damped Base-Isolated Bridge","authors":"Muhammad Tariq Chaudhary","doi":"10.1155/stc/1738114","DOIUrl":"https://doi.org/10.1155/stc/1738114","url":null,"abstract":"<p>Structural model updating of generally damped systems often requires prior assumptions on structural parameters to constrain the solution space when classical eigensensitivity (ES)-based methods are employed. This study proposes a hybrid ES-machine learning (ML) framework that integrates physics-based ES analysis with supervised learning for estimating physically admissible structural parameters in generally damped base-isolated systems while accounting for variability in known parameters. An iterative ES-based procedure was implemented to quantify relationships between structural and modal parameters and to generate physics-informed training data under controlled parameter variability. These data were then used to train ML models to estimate structural parameters corresponding to experimentally identified modal properties. The proposed methodology was implemented on field-recorded seismic data from a multispan base-isolated bridge and evaluated using four ML models. During training, Gaussian process regression (GPR), support vector machine (SVM), and bagged tree ensemble (BTE) models captured nonlinear relationships between modal and structural parameters better than the multivariable linear regression (MVLR) model. Structural parameters estimated by the baseline ES and ML models satisfied the physical bounds and were validated using the structural response metrics of the modal residuals (<i>δ</i><i>z</i>), measure of fit in acceleration response (<i>E</i>) and modal assurance criterion (MAC). Modal residuals (<i>δ</i><i>z</i>) were within the tolerance limits for the baseline ES and 2 ML models (GPR and SVM), and MAC values were greater than 0.97, while corresponding <i>E</i> values ranged between 0.164 and 0.301. This indicated excellent modal correlation and good structural response accuracy with the recorded parameters. The best-performing GPR ML model achieved average <i>δ</i><i>z</i> and <i>E</i> values that were 29.1% and 58.1% better than the baseline ES model, demonstrating the usefulness of adding ML to the ES based model updating process.</p>","PeriodicalId":49471,"journal":{"name":"Structural Control & Health Monitoring","volume":"2026 1","pages":""},"PeriodicalIF":5.0,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/stc/1738114","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}