Journal of Infrastructure Intelligence and Resilience最新文献_第3页

Strain signal denoising in bridge SHM: A comparative analysis of MODWT and other techniques 桥梁SHM应变信号去噪：MODWT与其它方法的比较分析

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-04-29 DOI: 10.1016/j.iintel.2025.100155

Yun-Xia Xia , Ru-Kai Xu , Yi-Qing Ni , Zu-Quan Jin

Accurate denoising of strain signals is critical for early damage detection in bridge structural health monitoring (SHM). However, signals denoising methods often struggle with the non-stationary and broadband noise encountered in real-world environments. This study provides the first comprehensive comparison of various denoising techniques specifically tailored for bridge strain signals, emphasizing the maximal overlapping discrete wavelet transform (MODWT) for its capacity to handle complex noise profiles. We rigorously compare MODWT with time-domain (moving average filter, finite impulse response filter, empirical mode decomposition), frequency-domain (bandpass filter, Fourier mode decomposition), and other wavelet-based (discrete wavelet transform) approaches. Uniquely, this study employs three datasets from two distinct bridge types (masonry arch and steel bowstring) and evaluates performance using both expert assessments and quantitative metrics (signal-to-noise ratio, peak signal-to-noise ratio, root mean square error, and correlation coefficient). Our findings demonstrate that MODWT exhibits a distinct advantage in high-intensity white noise environments, a common scenario in real-world bridge monitoring, offering valuable guidance for engineers in selecting appropriate denoising strategies. The results not only validate MODWT as a promising preprocessing technique but also offer critical insights into the limitations of existing methods, paving the way for the development of more adaptive and robust denoising solutions in bridge SHM.

应变信号的准确去噪是桥梁结构健康监测中早期损伤检测的关键。然而，信号去噪方法经常与现实环境中遇到的非平稳和宽带噪声作斗争。本研究首次对各种专门针对桥梁应变信号的去噪技术进行了全面比较，强调了最大重叠离散小波变换（MODWT）处理复杂噪声剖面的能力。我们严格比较了MODWT与时域（移动平均滤波器、有限脉冲响应滤波器、经验模态分解）、频域（带通滤波器、傅立叶模态分解）和其他基于小波的（离散小波变换）方法。独特的是，本研究采用了来自两种不同桥梁类型（砌体拱桥和钢弓弦桥）的三个数据集，并使用专家评估和定量指标（信噪比、峰值信噪比、均方根误差和相关系数）来评估性能。我们的研究结果表明，MODWT在高强度白噪声环境中表现出明显的优势，这是现实世界桥梁监测中常见的场景，为工程师选择适当的去噪策略提供了有价值的指导。研究结果不仅验证了MODWT作为一种有前途的预处理技术，而且对现有方法的局限性提供了重要的见解，为开发更具适应性和鲁棒性的桥梁SHM去噪解决方案铺平了道路。

{"title":"Strain signal denoising in bridge SHM: A comparative analysis of MODWT and other techniques","authors":"Yun-Xia Xia , Ru-Kai Xu , Yi-Qing Ni , Zu-Quan Jin","doi":"10.1016/j.iintel.2025.100155","DOIUrl":"10.1016/j.iintel.2025.100155","url":null,"abstract":"<div><div>Accurate denoising of strain signals is critical for early damage detection in bridge structural health monitoring (SHM). However, signals denoising methods often struggle with the non-stationary and broadband noise encountered in real-world environments. This study provides the first comprehensive comparison of various denoising techniques specifically tailored for bridge strain signals, emphasizing the maximal overlapping discrete wavelet transform (MODWT) for its capacity to handle complex noise profiles. We rigorously compare MODWT with time-domain (moving average filter, finite impulse response filter, empirical mode decomposition), frequency-domain (bandpass filter, Fourier mode decomposition), and other wavelet-based (discrete wavelet transform) approaches. Uniquely, this study employs three datasets from two distinct bridge types (masonry arch and steel bowstring) and evaluates performance using both expert assessments and quantitative metrics (signal-to-noise ratio, peak signal-to-noise ratio, root mean square error, and correlation coefficient). Our findings demonstrate that MODWT exhibits a distinct advantage in high-intensity white noise environments, a common scenario in real-world bridge monitoring, offering valuable guidance for engineers in selecting appropriate denoising strategies. The results not only validate MODWT as a promising preprocessing technique but also offer critical insights into the limitations of existing methods, paving the way for the development of more adaptive and robust denoising solutions in bridge SHM.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 3","pages":"Article 100155"},"PeriodicalIF":0.0,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144115297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bridge post-disaster rapid inspection using 3D point cloud: a case study on vehicle-bridge collision 基于三维点云的桥梁灾后快速检测——以车桥碰撞为例

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-04-01 DOI: 10.1016/j.iintel.2025.100153

Tianyu Ma , Yanjie Zhu , Wen Xiong , Beiyang Zhang , Kaiwen Hu

With the increase in traffic volume, vehicle-bridge collision accidents have been more frequent, creating significant threats to the safe operation of bridges. In the face of sudden vehicle collision accidents, bridge management agencies urgently require fast and accurate damage inspection methods to assess the service performance of the damaged bridge and provide support for post-disaster recovery. However, the service performance of a bridge is related to its overall structure and localized damage morphology. It is challenging for traditional measurement methods to obtain the three-dimensional (3D) morphology of the bridge and damaged areas. They can only obtain limited data points, which cannot provide adequate data for bridge damage assessment. Recently developed 3D laser scanning technology has guaranteed an accurate and timely 3D morphology inspection for the damaged bridge. Based on 3D laser scanning technology, this research proposed a post-disaster emergency inspection solution using a vehicle-bridge collision accident as a practical case, which provides a basis for emergency response decisions. This study focused on the rapid acquisition of the bridge digital model, spatial morphology identification of bridge components, and refined assessment of collision damage. The inspecting results revealed anomalies in the elevation of the damaged main girder and main cable, which necessitated urgent reinforcement measures. Additionally, the damaged hanger was found to have exhibited a lateral deflection angle of 17.12°, with a maximum cable clamp damage depth of 33.06 mm, requiring immediate replacement.

随着交通运输量的增加，车桥碰撞事故日益频繁，对桥梁的安全运行造成了重大威胁。面对突发性车辆碰撞事故，桥梁管理机构迫切需要快速准确的损伤检测方法，以评估受损桥梁的使用性能，为灾后恢复提供支持。然而，桥梁的使用性能与桥梁的整体结构和局部损伤形态有关。传统的测量方法难以获得桥梁和损伤区域的三维形貌。只能获得有限的数据点，不能为桥梁损伤评估提供充分的数据。近年来发展起来的三维激光扫描技术，保证了对受损桥梁进行准确、及时的三维形态检测。本研究基于三维激光扫描技术，以某车桥碰撞事故为实际案例，提出灾后应急检测方案，为应急响应决策提供依据。研究重点是桥梁数字模型的快速获取、桥梁构件的空间形态识别以及碰撞损伤的精细化评估。检查结果显示主梁和主缆受损标高异常，需要紧急采取加固措施。此外，发现受损的悬挂器横向偏转角度为17.12°，电缆夹的最大损坏深度为33.06 mm，需要立即更换。

{"title":"Bridge post-disaster rapid inspection using 3D point cloud: a case study on vehicle-bridge collision","authors":"Tianyu Ma , Yanjie Zhu , Wen Xiong , Beiyang Zhang , Kaiwen Hu","doi":"10.1016/j.iintel.2025.100153","DOIUrl":"10.1016/j.iintel.2025.100153","url":null,"abstract":"<div><div>With the increase in traffic volume, vehicle-bridge collision accidents have been more frequent, creating significant threats to the safe operation of bridges. In the face of sudden vehicle collision accidents, bridge management agencies urgently require fast and accurate damage inspection methods to assess the service performance of the damaged bridge and provide support for post-disaster recovery. However, the service performance of a bridge is related to its overall structure and localized damage morphology. It is challenging for traditional measurement methods to obtain the three-dimensional (3D) morphology of the bridge and damaged areas. They can only obtain limited data points, which cannot provide adequate data for bridge damage assessment. Recently developed 3D laser scanning technology has guaranteed an accurate and timely 3D morphology inspection for the damaged bridge. Based on 3D laser scanning technology, this research proposed a post-disaster emergency inspection solution using a vehicle-bridge collision accident as a practical case, which provides a basis for emergency response decisions. This study focused on the rapid acquisition of the bridge digital model, spatial morphology identification of bridge components, and refined assessment of collision damage. The inspecting results revealed anomalies in the elevation of the damaged main girder and main cable, which necessitated urgent reinforcement measures. Additionally, the damaged hanger was found to have exhibited a lateral deflection angle of 17.12°, with a maximum cable clamp damage depth of 33.06 mm, requiring immediate replacement.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 3","pages":"Article 100153"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143879314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Large multimodal model assisted underground tunnel damage inspection and human-machine interaction 大型多模态模型辅助地下隧道损伤检测和人机交互

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-04-01 DOI: 10.1016/j.iintel.2025.100154

Yanzhi Qi , Zhi Ding , Yaozhi Luo

Artificial Intelligence is playing an increasingly important role in tunnel inspection as a core driver of the new generation of engineering. Traditional methods are difficult to directly generate human linguistic information and lack valid messages extracted from different modalities. This paper proposes Damage LMM, a multimodal damage detection model that can handle images or videos as well as text inputs, to realize fast damage identification and human-computer interaction. The visual instruction database is first created from real damage data collected using different visual sensors and captions extracted by a regional convolutional neural network. The basic language model is then fine-tuned into a specialised Damage LMM, which enhances user instructions by integrating virtual prompt injection and system messages. Finally, the enhanced prompts are processed through the tuned multimodal model to generate a detailed visual description of the damage. The performance of the method is evaluated using a real tunnel dataset, and the results show that it has better robustness and accuracy than other models in multimodal data, with an accuracy of 0.93 for the in-domain image data and a contextual correlation of 0.94. The proposed method can effectively identify tunnel defects and realize multimodal user interaction functions with a moderate number of markers and a short delay time, which will greatly help engineers to quickly obtain effective information and assess the degree of damage at the tunnel inspection site.

人工智能作为新一代工程的核心驱动力，在隧道检测中发挥着越来越重要的作用。传统方法难以直接生成人类语言信息，缺乏从不同模态中提取的有效信息。为了实现快速的损伤识别和人机交互，本文提出了一种可以处理图像或视频以及文本输入的多模态损伤检测模型——损伤LMM。视觉指令数据库首先由不同视觉传感器收集的真实损伤数据和由区域卷积神经网络提取的字幕创建。然后将基本语言模型微调为专门的损害LMM，该LMM通过集成虚拟提示注入和系统消息来增强用户指令。最后，通过调整后的多模态模型对增强的提示进行处理，以生成损坏的详细视觉描述。使用真实隧道数据集对该方法进行了性能评估，结果表明，该方法在多模态数据中具有更好的鲁棒性和精度，对域内图像数据的精度为0.93，上下文相关性为0.94。该方法可以有效识别隧道缺陷，实现多模态用户交互功能，标记数量适中，延迟时间短，将极大地帮助工程师在隧道检测现场快速获取有效信息和评估损伤程度。

{"title":"Large multimodal model assisted underground tunnel damage inspection and human-machine interaction","authors":"Yanzhi Qi , Zhi Ding , Yaozhi Luo","doi":"10.1016/j.iintel.2025.100154","DOIUrl":"10.1016/j.iintel.2025.100154","url":null,"abstract":"<div><div>Artificial Intelligence is playing an increasingly important role in tunnel inspection as a core driver of the new generation of engineering. Traditional methods are difficult to directly generate human linguistic information and lack valid messages extracted from different modalities. This paper proposes Damage LMM, a multimodal damage detection model that can handle images or videos as well as text inputs, to realize fast damage identification and human-computer interaction. The visual instruction database is first created from real damage data collected using different visual sensors and captions extracted by a regional convolutional neural network. The basic language model is then fine-tuned into a specialised Damage LMM, which enhances user instructions by integrating virtual prompt injection and system messages. Finally, the enhanced prompts are processed through the tuned multimodal model to generate a detailed visual description of the damage. The performance of the method is evaluated using a real tunnel dataset, and the results show that it has better robustness and accuracy than other models in multimodal data, with an accuracy of 0.93 for the in-domain image data and a contextual correlation of 0.94. The proposed method can effectively identify tunnel defects and realize multimodal user interaction functions with a moderate number of markers and a short delay time, which will greatly help engineers to quickly obtain effective information and assess the degree of damage at the tunnel inspection site.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 3","pages":"Article 100154"},"PeriodicalIF":0.0,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automatic settlement assessment of urban road from 3D terrestrial laser scan data 利用三维地面激光扫描数据自动评估城市道路沉降状况

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-03-01 DOI: 10.1016/j.iintel.2025.100142

Xinchen Zhang , Qian Wang , Hai Fang , Guogang Ying

Tunnel construction in urban environments often requires passing beneath existing roads, where excessive soil excavation can lead to road cracking, settlement, or heaving, posing risks to road safety. Traditional road settlement monitoring methods rely on manual measurements, which are time-consuming, labor-intensive, and costly. Some existing approaches also require extensive sensor deployment, complicating installation and maintenance. To address these challenges, this study introduces a LiDAR-based method for efficient and accurate road settlement assessment. The impact of various LiDAR measurement parameters on assessment accuracy and efficiency was analyzed under typical urban road conditions. A comprehensive workflow was developed, incorporating both rough and fine alignment processes. Key steps in the workflow, such as automated identification of matching planes between point clouds, directional alignment, and angle fine-tuning, were automated using advanced algorithms. The proposed method was applied and validated in a region undergoing tunneling works in Singapore. Results demonstrated that the partially automated LiDAR-based approach achieved comparable accuracy to manual point cloud alignment methods while significantly improving efficiency and reducing labor costs. Furthermore, when compared to traditional total station methods, the LiDAR-based technique maintained errors within acceptable limits and enabled broader spatial coverage. Overall, this study highlights the feasibility and potential of LiDAR technology to enhance road settlement monitoring in engineering practice, offering a cost-effective and scalable alternative to traditional methods.

在城市环境中建造隧道往往需要从现有道路的下方通过，过度挖掘泥土会导致道路开裂、沉降或隆起，对道路安全构成威胁。传统的道路沉降监测方法依赖于人工测量，耗时长，劳动强度大，成本高。一些现有的方法还需要大量的传感器部署，使安装和维护变得复杂。为了应对这些挑战，本研究引入了一种基于激光雷达的方法，用于高效准确的道路沉降评估。在典型城市道路条件下，分析了各种激光雷达测量参数对评估精度和效率的影响。开发了一个综合的工作流程，结合了粗糙和精细的校准过程。工作流程中的关键步骤，如点云之间匹配平面的自动识别、方向对齐和角度微调，都使用先进的算法实现了自动化。该方法在新加坡某隧道施工区域进行了应用和验证。结果表明，基于激光雷达的部分自动化方法在显著提高效率和降低人工成本的同时，取得了与人工点云对齐方法相当的精度。此外，与传统的全站仪方法相比，基于激光雷达的技术将误差保持在可接受的范围内，并实现了更广泛的空间覆盖。总体而言，本研究强调了激光雷达技术在工程实践中增强道路沉降监测的可行性和潜力，为传统方法提供了一种具有成本效益和可扩展性的替代方案。

{"title":"Automatic settlement assessment of urban road from 3D terrestrial laser scan data","authors":"Xinchen Zhang , Qian Wang , Hai Fang , Guogang Ying","doi":"10.1016/j.iintel.2025.100142","DOIUrl":"10.1016/j.iintel.2025.100142","url":null,"abstract":"<div><div>Tunnel construction in urban environments often requires passing beneath existing roads, where excessive soil excavation can lead to road cracking, settlement, or heaving, posing risks to road safety. Traditional road settlement monitoring methods rely on manual measurements, which are time-consuming, labor-intensive, and costly. Some existing approaches also require extensive sensor deployment, complicating installation and maintenance. To address these challenges, this study introduces a LiDAR-based method for efficient and accurate road settlement assessment. The impact of various LiDAR measurement parameters on assessment accuracy and efficiency was analyzed under typical urban road conditions. A comprehensive workflow was developed, incorporating both rough and fine alignment processes. Key steps in the workflow, such as automated identification of matching planes between point clouds, directional alignment, and angle fine-tuning, were automated using advanced algorithms. The proposed method was applied and validated in a region undergoing tunneling works in Singapore. Results demonstrated that the partially automated LiDAR-based approach achieved comparable accuracy to manual point cloud alignment methods while significantly improving efficiency and reducing labor costs. Furthermore, when compared to traditional total station methods, the LiDAR-based technique maintained errors within acceptable limits and enabled broader spatial coverage. Overall, this study highlights the feasibility and potential of LiDAR technology to enhance road settlement monitoring in engineering practice, offering a cost-effective and scalable alternative to traditional methods.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 1","pages":"Article 100142"},"PeriodicalIF":0.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Automated Operational Modal Analysis of a steel truss railway bridge employing free decay response 采用自由衰减响应对钢桁梁铁路桥进行自动化运行模态分析

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-03-01 DOI: 10.1016/j.iintel.2025.100145

Francesco Morgan Bono, Antonio Argentino, Lorenzo Bernardini, Lorenzo Benedetti, Gabriele Cazzulani, Claudio Somaschini, Marco Belloli

The efficiency and resilience of transportation networks depend significantly on the integrity of bridges, which are increasingly threatened by ageing, traffic, and extreme climate events. Traditional visual inspections have notable limitations, necessitating the adoption of more objective methods like Structural Health Monitoring (SHM). This study explores the application of Operational Modal Analysis (OMA) to estimate the modal parameters of railway bridges, specifically using the Covariance-based Stochastic Subspace Identification (SSI-COV) algorithm. The case study involves a steel Warren truss bridge monitored over 20 months. The research demonstrates that SSI-COV, typically requiring stationary random input, can effectively utilise the bridge’s free decay responses following train passages. This approach strongly improves signal-to-noise ratio, which is vice-versa critical for railway bridges ambient vibrations due to the very low input energy, enabling precise modal parameter estimation with shorter time windows and lower-performance sensors. Results were validated against the Peak-Picking (PP) and the Enhanced Frequency Domain Decomposition (EFDD) methods, with SSI-COV identifying three additional natural frequencies and exhibiting lower dispersion in frequency estimates throughout the monitored period. Statistical analysis further indicated that using multiple free decays enhances the accuracy and reduces variability for challenging modes, while dominant modes are reliably estimated with minimal decay data. These findings endorse the combination of SSI-COV and free decays as a robust tool for detailed and long-term bridge monitoring, offering a valuable and potentially low-cost alternative to ambient vibration-based OMA techniques.

交通网络的效率和弹性在很大程度上取决于桥梁的完整性，而桥梁正日益受到老化、交通和极端气候事件的威胁。传统的目视检查有明显的局限性，需要采用更客观的方法，如结构健康监测（SHM）。本研究探讨了运行模态分析（OMA）在铁路桥梁模态参数估计中的应用，特别是基于协方差的随机子空间识别（SSI-COV）算法。该案例研究涉及一座钢制沃伦桁架桥，监测时间超过20个月。研究表明，SSI-COV通常需要固定随机输入，可以有效地利用列车通过后桥梁的自由衰减响应。这种方法极大地提高了信噪比，这对铁路桥梁环境振动至关重要，因为输入能量非常低，可以用更短的时间窗口和更低性能的传感器进行精确的模态参数估计。结果与拾峰（PP）和增强频域分解（EFDD）方法进行了验证，SSI-COV识别了三个额外的固有频率，并且在整个监测期间的频率估计中表现出较低的色散。统计分析进一步表明，使用多个自由衰变可以提高具有挑战性模态的精度并减少变异，而优势模态则可以用最小的衰变数据可靠地估计。这些研究结果表明，SSI-COV和自由衰减的结合是一种可靠的工具，可用于详细和长期的桥梁监测，为基于环境振动的OMA技术提供了一种有价值且潜在低成本的替代方案。

{"title":"Automated Operational Modal Analysis of a steel truss railway bridge employing free decay response","authors":"Francesco Morgan Bono, Antonio Argentino, Lorenzo Bernardini, Lorenzo Benedetti, Gabriele Cazzulani, Claudio Somaschini, Marco Belloli","doi":"10.1016/j.iintel.2025.100145","DOIUrl":"10.1016/j.iintel.2025.100145","url":null,"abstract":"<div><div>The efficiency and resilience of transportation networks depend significantly on the integrity of bridges, which are increasingly threatened by ageing, traffic, and extreme climate events. Traditional visual inspections have notable limitations, necessitating the adoption of more objective methods like Structural Health Monitoring (SHM). This study explores the application of Operational Modal Analysis (OMA) to estimate the modal parameters of railway bridges, specifically using the Covariance-based Stochastic Subspace Identification (SSI-COV) algorithm. The case study involves a steel Warren truss bridge monitored over 20 months. The research demonstrates that SSI-COV, typically requiring stationary random input, can effectively utilise the bridge’s free decay responses following train passages. This approach strongly improves signal-to-noise ratio, which is vice-versa critical for railway bridges ambient vibrations due to the very low input energy, enabling precise modal parameter estimation with shorter time windows and lower-performance sensors. Results were validated against the Peak-Picking (PP) and the Enhanced Frequency Domain Decomposition (EFDD) methods, with SSI-COV identifying three additional natural frequencies and exhibiting lower dispersion in frequency estimates throughout the monitored period. Statistical analysis further indicated that using multiple free decays enhances the accuracy and reduces variability for challenging modes, while dominant modes are reliably estimated with minimal decay data. These findings endorse the combination of SSI-COV and free decays as a robust tool for detailed and long-term bridge monitoring, offering a valuable and potentially low-cost alternative to ambient vibration-based OMA techniques.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 1","pages":"Article 100145"},"PeriodicalIF":0.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep learning in crack detection: A comprehensive scientometric review 裂纹检测中的深度学习：科学计量学综述

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-02-23 DOI: 10.1016/j.iintel.2025.100144

Yingjie Wu , Shaoqi Li , Jingqiu Li , Yanping Yu , Jianchun Li , Yancheng Li

Cracks represent one of the common forms of damage in concrete structures and pavements, leading to safety issues and increased maintenance costs. Therefore, timely crack detection is crucial for preventing further damage and ensuring the safety of these structures. Traditional manual inspection methods are limited by factors such as time consumption, subjectivity, and labor intensity. To address these challenges, deep learning-based crack detection technologies have emerged as promising solutions, demonstrating satisfactory performance and accuracy. However, the field still lacks comprehensive scientometric analyses and critical surveys of existing works, which are vital for identifying research gaps and guiding future studies. This paper conducts a bibliometric and critical analysis of the collected literature, providing novel insights into current research trends and identifying potential areas for future investigation. Analytical tools, including VOSviewer and CiteSpace, were employed for in-depth analysis and visualization. This study identifies key research gaps and proposes future directions, focusing on advancements in model generalization, computational efficiency, dataset standardization, and the practical application of crack detection methods.

裂缝是混凝土结构和路面的常见损坏形式之一，会导致安全问题并增加维护成本。因此，及时检测裂缝对防止结构进一步破坏，保证结构安全至关重要。传统的人工检测方法受时间、主观性、劳动强度等因素的限制。为了应对这些挑战，基于深度学习的裂缝检测技术已经成为有前途的解决方案，表现出令人满意的性能和准确性。然而，该领域仍然缺乏对现有工作的全面科学计量分析和批判性调查，这对于确定研究差距和指导未来的研究至关重要。本文对收集到的文献进行了文献计量学和批判性分析，为当前的研究趋势提供了新的见解，并确定了未来研究的潜在领域。使用VOSviewer和CiteSpace等分析工具进行深入分析和可视化。本研究确定了关键的研究差距并提出了未来的研究方向，重点关注模型泛化、计算效率、数据集标准化和裂纹检测方法的实际应用方面的进展。

{"title":"Deep learning in crack detection: A comprehensive scientometric review","authors":"Yingjie Wu , Shaoqi Li , Jingqiu Li , Yanping Yu , Jianchun Li , Yancheng Li","doi":"10.1016/j.iintel.2025.100144","DOIUrl":"10.1016/j.iintel.2025.100144","url":null,"abstract":"<div><div>Cracks represent one of the common forms of damage in concrete structures and pavements, leading to safety issues and increased maintenance costs. Therefore, timely crack detection is crucial for preventing further damage and ensuring the safety of these structures. Traditional manual inspection methods are limited by factors such as time consumption, subjectivity, and labor intensity. To address these challenges, deep learning-based crack detection technologies have emerged as promising solutions, demonstrating satisfactory performance and accuracy. However, the field still lacks comprehensive scientometric analyses and critical surveys of existing works, which are vital for identifying research gaps and guiding future studies. This paper conducts a bibliometric and critical analysis of the collected literature, providing novel insights into current research trends and identifying potential areas for future investigation. Analytical tools, including VOSviewer and CiteSpace, were employed for in-depth analysis and visualization. This study identifies key research gaps and proposes future directions, focusing on advancements in model generalization, computational efficiency, dataset standardization, and the practical application of crack detection methods.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 3","pages":"Article 100144"},"PeriodicalIF":0.0,"publicationDate":"2025-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modular approach to model order reduction for offshore wind turbines supported by multi-bucket jacket foundation 多筒导管基础支撑海上风力机模型降阶的模块化方法

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-02-20 DOI: 10.1016/j.iintel.2025.100143

Zhaofeng Shen , Yue Chen , Pengfei Li , Jun Liang , Ying Wang , Jinping Ou

Offshore wind turbines (OWTs) supported by multi-bucket jacket foundations (MBJF) provide a cost-effective solution for offshore wind energy production when water depth exceeds 50 m. However, numerical simulation of their dynamic behaviors towards high accuracy and efficiency becomes challenging due to the intricate structural configuration. To tackle it, this paper introduces a model order reduction framework for OWTs with MBJF. The framework strategically decomposes the structure into five substructures, whose reduced-order models (ROMs) are individually constructed and then assembled into a ROM for the entire OWT structure with fixed boundary conditions. The parameters of the assembled ROM on soil are subsequently calibrated through a model updating process, to ensure the alignment of modal parameters and structural displacements between ROM and full-order model (FOM). The results show that Young's moduli of both tower and jacket dominate the frequencies of global bending modes while Young's modulus of the blade dominates the frequencies of blade bending modes. Among the support parameters, the combined T-Z soil spring stiffness plays a critical role, affecting the frequencies of global motion and bending modes. The proposed model order reduction framework provides a robust methodology towards accurate and efficient simulation of structural dynamics for OWTs supported by MBJF.

多桶套基础（MBJF）支撑的海上风力涡轮机（owt）为水深超过50米的海上风力发电提供了一种经济有效的解决方案。然而，由于其复杂的结构结构，对其动态特性进行高精度和高效率的数值模拟具有挑战性。为了解决这一问题，本文引入了基于MBJF的owt模型降阶框架。该框架将结构战略性地分解为五个子结构，每个子结构的降阶模型（ROM）被单独构建，然后组装成具有固定边界条件的整个OWT结构的ROM。随后，通过模型更新过程对土壤上组装的ROM参数进行校准，以确保ROM与全阶模型（FOM）之间的模态参数和结构位移对齐。结果表明，塔架和导管套的杨氏模量主导了整体弯曲模态的频率，而叶片的杨氏模量主导了叶片弯曲模态的频率。在支护参数中，T-Z组合土弹簧刚度对整体运动模态和弯曲模态的频率起着关键作用。所提出的模型降阶框架为MBJF支持的owt结构动力学的准确和有效模拟提供了一种强大的方法。

{"title":"Modular approach to model order reduction for offshore wind turbines supported by multi-bucket jacket foundation","authors":"Zhaofeng Shen , Yue Chen , Pengfei Li , Jun Liang , Ying Wang , Jinping Ou","doi":"10.1016/j.iintel.2025.100143","DOIUrl":"10.1016/j.iintel.2025.100143","url":null,"abstract":"<div><div>Offshore wind turbines (OWTs) supported by multi-bucket jacket foundations (MBJF) provide a cost-effective solution for offshore wind energy production when water depth exceeds 50 m. However, numerical simulation of their dynamic behaviors towards high accuracy and efficiency becomes challenging due to the intricate structural configuration. To tackle it, this paper introduces a model order reduction framework for OWTs with MBJF. The framework strategically decomposes the structure into five substructures, whose reduced-order models (ROMs) are individually constructed and then assembled into a ROM for the entire OWT structure with fixed boundary conditions. The parameters of the assembled ROM on soil are subsequently calibrated through a model updating process, to ensure the alignment of modal parameters and structural displacements between ROM and full-order model (FOM). The results show that Young's moduli of both tower and jacket dominate the frequencies of global bending modes while Young's modulus of the blade dominates the frequencies of blade bending modes. Among the support parameters, the combined T-Z soil spring stiffness plays a critical role, affecting the frequencies of global motion and bending modes. The proposed model order reduction framework provides a robust methodology towards accurate and efficient simulation of structural dynamics for OWTs supported by MBJF.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 2","pages":"Article 100143"},"PeriodicalIF":0.0,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143681861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Positional inaccuracy investigation and innovative connection solution for robotic construction of load carrying structures 承载结构机器人施工位置误差研究及创新连接解决方案

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-01-16 DOI: 10.1016/j.iintel.2025.100141

Cheav Por Chea, Yu Bai, Yihai Fang

Robotic construction of load carrying structures in civil engineering becomes promising with the supports from robotics, computer-vision, and design for manufacturing and assembly. A multi-robot system was developed to demonstrate an automated construction of reciprocal frame structures where mobile robots were used to facilitate the access of robotic arms and a series of programming packages were developed to automate the construction. Furthermore, the AprilTag fiducial marker system was applied as a positioning system to align the mobile robots during construction tasks and to target the structural components. In this context, the key challenges are centred on the understanding of the accuracy and tolerance of the robotic system in positioning and navigation. To this end, experimental methods were developed in this study to understand the observed distances and the accuracy of the positioning system. The optimal observation distance for the positioning system in the robotic system was then determined considering the positional and orientational accuracies of the AprilTag fiducial marker system using a red, green, blue-depth (RGB-D) camera. Moreover, experiments were conducted to study the impact of the barycentre of robotic arms on the precision of the mobile robots and to determine the offset of the mobile robot during the manoeuvre. In consideration of the positional inaccuracies, the magnetic connection approach was creatively implemented using their inherent self-aligning property. The corresponding effective range was also firstly determined, within which the structural components could be installed successfully.

在机器人技术、计算机视觉以及制造和装配设计的支持下，土木工程中承载结构的机器人施工前景广阔。开发了一个多机器人系统来演示互惠框架结构的自动化构建，其中移动机器人用于方便机械臂的访问，并开发了一系列编程包来实现自动化构建。此外，应用AprilTag基准标记系统作为定位系统，在施工任务中对移动机器人进行对齐，并对结构部件进行定位。在这种情况下，关键的挑战集中在机器人系统的定位和导航的精度和公差的理解。为此，本研究开发了实验方法来了解定位系统的观测距离和精度。考虑到AprilTag基准标记系统采用红绿蓝深（RGB-D）相机的定位精度和方位精度，确定了机器人系统中定位系统的最佳观测距离。此外，通过实验研究了机械臂重心对移动机器人精度的影响，确定了移动机器人在机动过程中的偏移量。考虑到定位误差，创造性地利用磁连接的自对准特性实现磁连接。首先确定了相应的有效范围，在该有效范围内结构构件可以顺利安装。

{"title":"Positional inaccuracy investigation and innovative connection solution for robotic construction of load carrying structures","authors":"Cheav Por Chea, Yu Bai, Yihai Fang","doi":"10.1016/j.iintel.2025.100141","DOIUrl":"10.1016/j.iintel.2025.100141","url":null,"abstract":"<div><div>Robotic construction of load carrying structures in civil engineering becomes promising with the supports from robotics, computer-vision, and design for manufacturing and assembly. A multi-robot system was developed to demonstrate an automated construction of reciprocal frame structures where mobile robots were used to facilitate the access of robotic arms and a series of programming packages were developed to automate the construction. Furthermore, the AprilTag fiducial marker system was applied as a positioning system to align the mobile robots during construction tasks and to target the structural components. In this context, the key challenges are centred on the understanding of the accuracy and tolerance of the robotic system in positioning and navigation. To this end, experimental methods were developed in this study to understand the observed distances and the accuracy of the positioning system. The optimal observation distance for the positioning system in the robotic system was then determined considering the positional and orientational accuracies of the AprilTag fiducial marker system using a red, green, blue-depth (RGB-D) camera. Moreover, experiments were conducted to study the impact of the barycentre of robotic arms on the precision of the mobile robots and to determine the offset of the mobile robot during the manoeuvre. In consideration of the positional inaccuracies, the magnetic connection approach was creatively implemented using their inherent self-aligning property. The corresponding effective range was also firstly determined, within which the structural components could be installed successfully.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 2","pages":"Article 100141"},"PeriodicalIF":0.0,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A survey of generative models for image-based structural health monitoring in civil infrastructure 基于图像的民用基础设施结构健康监测生成模型综述

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-01-10 DOI: 10.1016/j.iintel.2025.100138

Gi-Hun Gwon, Hyung-Jo Jung

Accurately assessing and monitoring the condition of structures is essential for ensuring the safety and integrity of civil infrastructure. Over the past decade, image-based structural health monitoring technologies have emerged as powerful tools to enhance efficiency and improve the objectivity of structural evaluations. The integration of deep learning technologies with these monitoring systems has significantly improved the efficiency and reliability of structural condition diagnostics. Of particular interest are specifically Variational Autoencoders, Generative Adversarial Networks, and Diffusion Models, which have gained increasing attention due to their versatility in data generation and ability to address fundamental challenges in structural monitoring. While image-based structural health monitoring encompasses both damage detection and structural response measurements, this review primarily focuses on local-level monitoring applications such as damage detection, where generative models have demonstrated particular effectiveness in addressing challenges like limited data availability and environmental variations. This paper provides a comprehensive analysis of these generative models, examining their underlying concepts, mechanisms, and applications in image-based structural health monitoring. Key applications are reviewed, including structural damage detection, data augmentation for training, and emerging areas such as image quality enhancement and domain generalization. Our analysis presents the current state of generative models in structural monitoring, identifying critical challenges and promising future research directions. This systematic review serves as a foundational resource for researchers and practitioners in the field, offering insights into current achievements and potential advancements.

准确评估和监测结构状况对于确保民用基础设施的安全性和完整性至关重要。在过去十年中，基于图像的结构健康监测技术已成为提高效率和改善结构评估客观性的有力工具。深度学习技术与这些监测系统的整合极大地提高了结构状态诊断的效率和可靠性。特别值得关注的是变异自动编码器、生成式对抗网络和扩散模型，由于它们在数据生成方面的多功能性和应对结构监测中基本挑战的能力，这些技术越来越受到关注。虽然基于图像的结构健康监测包括损伤检测和结构响应测量，但本综述主要关注损伤检测等局部级监测应用，在这些应用中，生成模型在应对有限数据可用性和环境变化等挑战方面表现出了特殊的有效性。本文全面分析了这些生成模型，研究了它们的基本概念、机制以及在基于图像的结构健康监测中的应用。本文回顾了关键应用，包括结构损伤检测、用于训练的数据增强，以及图像质量增强和领域泛化等新兴领域。我们的分析介绍了结构监测中生成模型的现状，确定了关键挑战和有前景的未来研究方向。这篇系统综述为该领域的研究人员和从业人员提供了基础资源，让他们深入了解当前的成就和潜在的进步。

{"title":"A survey of generative models for image-based structural health monitoring in civil infrastructure","authors":"Gi-Hun Gwon, Hyung-Jo Jung","doi":"10.1016/j.iintel.2025.100138","DOIUrl":"10.1016/j.iintel.2025.100138","url":null,"abstract":"<div><div>Accurately assessing and monitoring the condition of structures is essential for ensuring the safety and integrity of civil infrastructure. Over the past decade, image-based structural health monitoring technologies have emerged as powerful tools to enhance efficiency and improve the objectivity of structural evaluations. The integration of deep learning technologies with these monitoring systems has significantly improved the efficiency and reliability of structural condition diagnostics. Of particular interest are specifically Variational Autoencoders, Generative Adversarial Networks, and Diffusion Models, which have gained increasing attention due to their versatility in data generation and ability to address fundamental challenges in structural monitoring. While image-based structural health monitoring encompasses both damage detection and structural response measurements, this review primarily focuses on local-level monitoring applications such as damage detection, where generative models have demonstrated particular effectiveness in addressing challenges like limited data availability and environmental variations. This paper provides a comprehensive analysis of these generative models, examining their underlying concepts, mechanisms, and applications in image-based structural health monitoring. Key applications are reviewed, including structural damage detection, data augmentation for training, and emerging areas such as image quality enhancement and domain generalization. Our analysis presents the current state of generative models in structural monitoring, identifying critical challenges and promising future research directions. This systematic review serves as a foundational resource for researchers and practitioners in the field, offering insights into current achievements and potential advancements.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 2","pages":"Article 100138"},"PeriodicalIF":0.0,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143776268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-trained artificial intelligence framework to detect chloride induced degradation in concrete 物理训练的人工智能框架，以检测氯化物引起的混凝土降解

Journal of Infrastructure Intelligence and Resilience

Pub Date : 2025-01-10 DOI: 10.1016/j.iintel.2025.100139

Parth Patel , Abhinav Gupta , Saran Srikanth Bodda , Harleen Kaur Sandhu

Numerous critical infrastructures in the United States, including bridges, dams, and nuclear plants, are aging and prone to concrete degradation, compromising their performance and structural integrity. One of the leading causes of degradation is chloride-induced corrosion, where chloride ions diffuse into the concrete, leading to reinforcement corrosion, spalling, and cracking. Detecting chloride degradation at an early stage is crucial for ensuring the safety of these vital structures. However, the visible signs of degradation, such as spalling and cracking, often appear only after significant damage has occurred. Degradation occurs gradually over many years, making it impractical to collect real-time non-destructive testing (NDT) data over extended periods while allowing the structure to continue deteriorating. To overcome this challenge, an integrated structural health monitoring framework is proposed that combines advanced finite element modeling, sensor data, and deep learning techniques. This framework follows a multi-step approach to simulate chloride degradation over the service life of the structure. Subsequently, finite element analyses are performed to numerically simulate non-destructive testing at various stages of degradation to generate corresponding sensor data. By leveraging these simulated data and insights, a physics-driven artificial intelligence framework is developed. The proposed framework offers a state-of-the-art solution to mitigate the challenges associated with long-term degradation monitoring by utilizing high-fidelity simulations and data-driven techniques to achieve detection of chloride-induced concrete damage.

美国的许多关键基础设施，包括桥梁、水坝和核电站，都在老化，容易出现混凝土退化，影响了它们的性能和结构完整性。腐蚀的主要原因之一是氯化物引起的腐蚀，氯离子扩散到混凝土中，导致钢筋腐蚀、剥落和开裂。在早期阶段检测氯化物降解对于确保这些重要结构的安全至关重要。然而，可见的退化迹象，如剥落和开裂，往往只出现在重大损害发生后。在许多年的时间里，老化是逐渐发生的，这使得在长时间内收集实时无损检测（NDT）数据变得不切实际，同时允许结构继续恶化。为了克服这一挑战，提出了一种综合结构健康监测框架，该框架结合了先进的有限元建模、传感器数据和深度学习技术。该框架遵循多步骤方法来模拟结构在使用寿命期间的氯化物降解。随后，进行有限元分析，数值模拟不同退化阶段的无损检测，生成相应的传感器数据。通过利用这些模拟数据和见解，开发了一个物理驱动的人工智能框架。提出的框架提供了一种最先进的解决方案，通过利用高保真模拟和数据驱动技术来实现氯化物引起的混凝土损伤检测，缓解了与长期退化监测相关的挑战。

{"title":"Physics-trained artificial intelligence framework to detect chloride induced degradation in concrete","authors":"Parth Patel , Abhinav Gupta , Saran Srikanth Bodda , Harleen Kaur Sandhu","doi":"10.1016/j.iintel.2025.100139","DOIUrl":"10.1016/j.iintel.2025.100139","url":null,"abstract":"<div><div>Numerous critical infrastructures in the United States, including bridges, dams, and nuclear plants, are aging and prone to concrete degradation, compromising their performance and structural integrity. One of the leading causes of degradation is chloride-induced corrosion, where chloride ions diffuse into the concrete, leading to reinforcement corrosion, spalling, and cracking. Detecting chloride degradation at an early stage is crucial for ensuring the safety of these vital structures. However, the visible signs of degradation, such as spalling and cracking, often appear only after significant damage has occurred. Degradation occurs gradually over many years, making it impractical to collect real-time non-destructive testing (NDT) data over extended periods while allowing the structure to continue deteriorating. To overcome this challenge, an integrated structural health monitoring framework is proposed that combines advanced finite element modeling, sensor data, and deep learning techniques. This framework follows a multi-step approach to simulate chloride degradation over the service life of the structure. Subsequently, finite element analyses are performed to numerically simulate non-destructive testing at various stages of degradation to generate corresponding sensor data. By leveraging these simulated data and insights, a physics-driven artificial intelligence framework is developed. The proposed framework offers a state-of-the-art solution to mitigate the challenges associated with long-term degradation monitoring by utilizing high-fidelity simulations and data-driven techniques to achieve detection of chloride-induced concrete damage.</div></div>","PeriodicalId":100791,"journal":{"name":"Journal of Infrastructure Intelligence and Resilience","volume":"4 2","pages":"Article 100139"},"PeriodicalIF":0.0,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0