Computer-Aided Civil and Infrastructure Engineering最新文献

{"title":"Data-efficient flood depth prediction through domain-aware coreset selection and tabular foundation models","authors":"Lipai Huang, Adithi Srinath, Manas Singh, Junwei Ma, Ali Mostafavi","doi":"10.1016/j.cacaie.2026.100268","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100268","url":null,"abstract":"Near-real-time flood depth prediction demands surrogate models that are accurate, fast, and transferable across watersheds. Supervised surrogates can match physics-based simulators in accuracy but need millions of training rows per watershed and cannot extrapolate beyond their original mesh. We propose a domain-aware coreset construction pipeline that conditions a tabular foundation model at inference time. The pipeline stratifies storms by return period and most-affected watershed, then samples hexagons with a target-aware spatial selector. With <ce:math altimg=\"si124.svg\"></ce:math> of the per-watershed training pool, the model attains a mean <ce:math altimg=\"si5.svg\"></ce:math> of <ce:math altimg=\"si126.svg\"></ce:math> across nine Houston-area watersheds, statistically indistinguishable from the supervised reference (<ce:math altimg=\"si127.svg\"></ce:math>). It transfers to held-out watersheds within the same metropolitan region without task-specific retraining, staying ahead of a coreset-trained supervised baseline. On real storms it exceeds the supervised reference on a far out-of-distribution case and trails it on a mixed case that is only moderately out of distribution. Domain-aware coreset construction lets tabular foundation models deliver data-efficient flood predictions that transfer across watersheds within a region without per-watershed training.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"14 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2026-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149374264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Temperature-Dependent Fracture Characterization of Semi-Flexible Pavement Using Windowed Acoustic Emission Features and Interpretable Random Forests","authors":"Weimin Song, Hao Wu, Yuanqi Liang, Xiaoyang Zhang","doi":"10.1016/j.cacaie.2026.100266","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100266","url":null,"abstract":"Temperature-dependent fracture behavior of semi-flexible pavement (SFP) materials can be characterized using acoustic emission (AE) monitoring, although reliable identification and interpretation of AE patterns remain challenging. This study develops an interpretable random forest (RF) framework using loading-stage-dependent AE features obtained from semi-circular bending (SCB) tests. Eighteen independent specimens of unmodified (Control) and KH550-modified SFP materials were tested at −10, 0, and 25°C. The loading history of each specimen was divided into five windows based on normalized load levels, and 120 statistical features were extracted from 15 AE-related variables. Four RF configurations were evaluated using leave-one-specimen-out cross-validation (LOSO-CV), with feature preprocessing and selection performed within the training folds. For binary classification of LowTemp (−10 and 0°C) and RoomTemp (25°C), the Combined and 15-feature Optimized models achieved window-level accuracies of 90.00% and 91.11%, respectively, with identical specimen-level accuracy of 94.44%. An exploratory three-temperature model achieved 61.11% window-level accuracy, with most misclassifications occurring between −10 and 0°C. Alternative loading-window definitions produced accuracies of 88.89%–93.33%, indicating reasonable predictive robustness. Feature importance and partial dependence analyses highlighted the roles of AE duration, cumulative energy, rise-angle, and frequency-related descriptors. Event-level RA–AF analysis further revealed temperature-dependent waveform-pattern distributions, although the estimated crack-mode fractions were sensitive to the classification boundary. These findings demonstrate the potential of windowed AE features for interpretable, specimen-level-validated characterization of temperature-dependent SFP fracture responses, while emphasizing the need for further validation before real-time engineering applications.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"35 1","pages":""},"PeriodicalIF":11.775,"publicationDate":"2026-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149358907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bridge damage diagnosis using robust InSAR-derived displacement features of virtual sensors by graph-informed unsupervised learning and hierarchical Bayesian thresholding","authors":"Alireza Entezami, Hassan Sarmadi, Zheng-Han Chen, Bahareh Behkamal, Ali Nadir Arslan, Yazhou Xie","doi":"10.1016/j.cacaie.2026.100264","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100264","url":null,"abstract":"Early detection and localization of structural anomalies are critical components of damage diagnosis for bridge health monitoring within intelligent transportation networks. Implementation of these tasks requires dense and spatially distributed sensor networks to obtain sufficient structural responses, which may be impractical and economically infeasible. To address this limitation, this study introduces a virtual-sensor construction strategy based on region-wise partitioning and point-wise selection of screened measurement points obtained from synthetic aperture radar (SAR) images along with a robust baseline-normalized displacement departure feature. Despite the possibility of capturing spatially-distributed structural displacements through interferometric SAR processing, damage diagnosis is often constrained by limited temporal sampling and spatial measurement coverage, leading to small-sample learning and reduced structural observability. To address these issues, this article proposes a graph-informed unsupervised learning method that comprises graph-aware feature extraction, a graph denoising contractive autoencoder (GDCAE) for learning normal structural behavior and detecting structural anomalies, a graph-weighted residual variation (GWRV) indicator for sensor-wise damage localization, and hierarchical Bayesian thresholding for uncertainty-aware decision-boundary estimation using limited anomaly scores. The major contributions of this research include constructing virtual sensors from satellite remote sensing, extracting graph-based features for bridge damage diagnosis, designing a graph-enhanced unsupervised neural network, and developing a probabilistic threshold estimator. Two historical river-crossing arch bridges monitored through TerraSAR-X and Sentinel-1 SAR images are utilized to validate the proposed method under complementary temporal-sampling and spatial-coverage conditions. Results demonstrate the detection and localization of candidate pre-collapse anomalies in one bridge, and no threshold exceedances in another bridge without documented damage.","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"7 1","pages":"100264"},"PeriodicalIF":11.775,"publicationDate":"2026-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149367881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Wet and reflective pavement crack segmentation based on a pseudo-multimodal collaborative network","authors":"Jiao Jin,Hao Xu,Jiaxin Li,Ling Li,Nan Zhang,Yao Deng,Kunfeng Ma,Mingzeng Zhang,Hassane Mimoun,Jianlong Zheng","doi":"10.1016/j.cacaie.2026.100234","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100234","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"5 1","pages":"100234"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SpecU-Former: A degradation-aware method for underwater concrete crack segmentation and mask-based width quantification","authors":"Chengbo Fan,Yangtao Li,Bo Chen,Jiaxin Yu,Mengfan Zhao,Yunlin Ma,Yu Tian","doi":"10.1016/j.cacaie.2026.100233","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100233","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"25 1","pages":"100233"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An explicit physical-state neural network method for constitutive modeling with consistent integration into nonlinear finite element frameworks","authors":"Quan Gu,Xiujia Yang,Zhe Lin","doi":"10.1016/j.cacaie.2026.100227","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100227","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"18 1","pages":"100227"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A real-time vision-based framework for non-contact full-field monitoring of traffic-induced bridge responses","authors":"Jiachang Gu,Zida Chen,Kang Gao,Yi You,Zhen Yang,Bin Chen","doi":"10.1016/j.cacaie.2026.100240","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100240","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"25 1","pages":"100240"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning-based video motion magnification for cable force estimation under micro-amplitude vibration","authors":"Weiming Zeng,Naiwei Lu,Mohammed El Ganaoui,Wen Chen,Chunjun Hu,Yang Liu","doi":"10.1016/j.cacaie.2026.100225","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100225","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"82 1","pages":"100225"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A CNN-enhanced probability density evolution framework for reliability analysis of offshore wind turbines","authors":"Jia-Xuan Li,Da-Gang Lu,Yinghui Tian","doi":"10.1016/j.cacaie.2026.100238","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100238","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"27 1","pages":"100238"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149090944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recovering large-scale Urban traffic flow data from sparse observations via analytical optimization","authors":"Sicheng Fu,Haotian Shi,Shixiao Liang,Wei Ma,Bin Ran","doi":"10.1016/j.cacaie.2026.100180","DOIUrl":"https://doi.org/10.1016/j.cacaie.2026.100180","url":null,"abstract":"","PeriodicalId":156,"journal":{"name":"Computer-Aided Civil and Infrastructure Engineering","volume":"5 1","pages":"100180"},"PeriodicalIF":11.775,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149092244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}