Kapitel

• FrontmatterWerner, Lienhart; Markus, Krüger; 10.3217/978-3-99161-057-1-000
• Digitalizing Infrastructure: Advancing Structural Health Monitoring for Smarter Asset ManagementKeßler, Sylvia; 10.3217/978-3-99161-057-1-001
• SHM for bridges – the work flowSodeikat, Christian; 10.3217/978-3-99161-057-1-002
• Bridge in service structural monitoring: the SCSHM benchmarkLimongelli, Maria Pina; 10.3217/978-3-99161-057-1-003
• Distributed fiber optic sensing in civil structural health monitoring at the next level – Realization of a comprehensive sensing network along the Brenner Base TunnelMonsberger, Christoph M.; Buchmayer, Fabian; Winkler, Madeleine; Cordes, Tobias; 10.3217/978-3-99161-057-1-004The Brenner Base Tunnel (BBT) is one of the key infrastructure projects currently under construction and will be one of the longest underground railway connection with a total length of approximately 64 km once completed. Its service life of 200 years implies essential requirements on the tunnel design. One important focus is to increase the availability of the tunnel, e.g. by enabling optimized maintenance work based on appropriate monitoring. The tunnel owner BBT SE has therefore initiated an enhanced Distributed Fiber Optic Sensing (DFOS) network inside the segmental lining for structural health monitoring without human access. The technology has significantly evolved in recent years to monitor large scale infrastructure, especially for in-situ tunnel monitoring as the distributed sensing feature can provide a complete picture of the strain distribution without blind spots. This contribution introduces the designed DFOS network, consisting of more than 35 km sensing cable along numerous tunnel cross-sections, spread over more than 30 km tunnel drive and two different construction lots. The monitoring data is autonomously evaluated and transferred to the online dashboard in real time. Analysis of the strain distribution provides fundamental information about the actual loading state of the segmental lining. The results together with experiences gained from practical implementations demonstrate the technology’s high potential for innovative civil structural health monitoring.
• Stevenson Creek Experimental Dam Monitoring Centenary: Overview and Perspectives of Strain Sensing and Strain-Based Monitoring of Civil StructuresGlisic, Branco; 10.3217/978-3-99161-057-1-005
• The Power of Optical and SAR Imaging for Remote Monitoring of Land and InfrastructureMazzanti, Paolo; 10.3217/978-3-99161-057-1-006
• Monitoring of fatigue crack propagation by means of distributed fiber optic sensingDohnalík, Petr; Lachinger, Stefan; Kwapisz, Maciej; Vorwagner, Alois; 10.3217/978-3-99161-057-1-007Distributed Fiber Optic Sensing (DFOS) is an innovative technique for Structural Health Monitoring (SHM). Taking advantage of the fact that DFOS can conveniently measure mechanical strain continuously along an optical fiber, it is increasingly used in monitoring of concrete bridges and tunnels. However, DFOS still needs research in new application areas, such as monitoring of steel bridges. In the present study, DFOS is used to investigate the potential to monitor fatigue crack initiation and propagation by experiments. In a full-scale test, a steel railway bridge was dynamically excited into resonance, generating fatigue-effective vibration amplitudes. The fiber was glued to the flange of the main beam in several loops to cover a larger area for crack detection. The measured strain signal was compared with results obtained from Finite Element Method (FEM) simulations supported by data acquired from conventional strain gauges and extensometers. The strain measurement with DFOS showed excellent agreement with the simulated strain. In this context, additional information about crack initiation, propagation, opening and length can be obtained indirectly from the DFOS measurement. However, when the crack is crossing the fiber, nonlinear effects come into play. To consider the nonlinear effects, a hysteresis model taking steel-fiber interaction into account was applied. The results of the study are presented and the applicability and potential of DFOS for fatigue crack monitoring in railway bridges is discussed.
• Updating prediction of fatigue reliability index of railway bridges using structural monitoring data and updated load historiesRalbovsky, Marian; Lachinger, Stefan; 10.3217/978-3-99161-057-1-008The assessment of fatigue consumption and the remaining lifetime of structural components is affected by considerable uncertainties on the side of the traffic loads, fatigue resistance and structural response. The purpose of the presented work was to develop methods for dealing with these uncertainties, as well as methods for improving the accuracy of assessment with the use of additional data. Within the research project Assets4Rail, a structural monitoring system was installed on a railway bridge located on a local track in Austria. The system consisted of strain sensors, acceleration sensors and inclinometers. It was used to measure the bridge response during train passages with known axle loads in course of a test with controlled conditions. This data was used to calibrate the structural model and develop probabilistic methods for fatigue assessment. Influence lines at fatigue-critical locations were evaluated from measured bridge strain response including their uncertainty. Further uncertainties considered in the assessment include the load histories and the fatigue resistance. The results showed the largest contribution by evaluation of model uncertainties from monitoring data. The effect of model updating was also considerable, but less significant. Further increase of estimation accuracy is achieved using section-specific traffic data. Whereas wayside monitoring data represent the reference scenario, the use of traffic management data provides a usable alternative.
• Monitoring of wind induced vibration on a tied-arch railway bridge.Castillo Ruano, Pablo; Martinez, Cesar; Dallinger, Sonja; 10.3217/978-3-99161-057-1-009The ÖBB Rheinbrücke, situated in the vicinity of Lustenau, represents a novel approach to steel-concrete composite arch structure engineering, boasting a span of 102 meters. The primary supporting structure is a tied-arch comprising 12 round steel hangers, each with a diameter of 100 mm, situated on either side of the bridge. The maximum length of the hangers is 18.9 meters. In arch bridges of this type with steel hangers, wind-induced vibrations in the hangers can result in high-frequency, high-amplitude fluctuations in stress levels, particularly in the hangers and their connections. This can be problematic from the perspective of fatigue, particularly given that the hanger connections often have notch-sensitive details. Following the completion of the bridge, comprehensive monitoring was conducted in accordance with the original plan. This was done with the objective of acquiring data regarding the vibrations experienced by the hangers and the subsequent damage to the material. This data was then used to determine whether vibration-reducing measures were necessary. During the course of monitoring and subsequent evaluation, it was observed that wind-induced vibrations in the hangers could result in the occurrence of fatigue-relevant stress ranges. This article serves to emphasise the importance of structural health monitoring in confirming the efficacy of vibration reduction measures, which have the potential to extend the service life of railway bridges.
• Acoustic emission monitoring of fatigue cracks for railway steel bridge inspectionProkofyev, Mikhail; Marihart, Heribert; Lackner, Gerald; 10.3217/978-3-99161-057-1-010Railway steel bridges are often affected by material fatigue, i.e. crack formation and crack growth at areas of high stress concentrations under millions of load cycles due to the traffic load. To support the continued safe operation of these structures a flexible and robust system solution for structural health monitoring is needed. It provides the responsible bridge inspector with useful information regarding the current condition and the future development of the monitored area. This offers the infrastructure operator an opportunity for optimised planning of inspection intervals and maintenance measures and helps to extend the service life of these structures. For this purpose, the RISE system was developed by TÜV AUSTRIA in close cooperation with the Austrian Federal Railways ÖBB. The RISE system monitors fatigue cracks and/or highly stressed areas using acoustic emission (AE). The system records the AE response from the monitored area while the material is stressed by the usual day-to-day railway operations. The analysis of the change of this material response over the monitoring period is used to predict the future development of the crack. In this paper the application of the RISE system for bridge inspection is presented for steel bridges in the railway network of ÖBB. The system solution is presented as a whole, from the installation on-site until the evaluation of the monitoring data and the obtained results supporting the responsible bridge inspectors.
• Rail track subsurface imaging from train vibrations recorded at dark fiber networksBehm, Michael; Brauner, Michael; 10.3217/978-3-99161-057-1-011We demonstrate the practical feasibility of assessing geotechnical parameters of rail infrastructure based on ‘dark fiber’ distributed acoustic sensing (DAS) recordings using trains as sources. A workflow to image the shallow (3 m – 20 m depth) subsurface in terms of shear wave velocity has been established. The shear wave velocity distribution is obtained from inversion of seismic surface waves excited by the trains and recorded on the fiber optic cables, and is used as a proxy for the geotechnical strength (e.g., shear modulus). Our results allow for the interpretation of potential geologic hazards and other features relevant for assessing the geotechnical integrity of rail infrastructure. This approach does not require dedicated field measurements or interruption of the train schedule, and therefore represents a cost-effective and robust method for different application scenarios.
• Monitoring of Concrete Infrastructure with Active Ultrasound Coda Wave InterferometryEpple, Niklas; Sanchez Trujillo, Camila; Fontoura Barroso, Daniel; Hau, Julia; Niederleithinger, Ernst; 10.3217/978-3-99161-057-1-012Coda Wave Interferometry has been used in Geophysics to detect weak changes in scattering media. Past research in Structural Health Monitoring has shown that this methodology can be applied to concrete structures to detect material changes by calculation of relative velocity changes. Successive measurements with embedded ultrasonic transducers provide a repeatable signal for reliable long-term monitoring of concrete. To research the application in real-world structures, we have embedded ultrasonic transducers in a bridge in Ulm and a Metro station in Munich, Germany. This study gives an overview of the monitoring of these two structures. The results show the potential and challenges of the method. Data evaluation can be largely automated to gain insights into material changes and other influences on the structure, such as traffic-induced load and temperature variations. The experiments demonstrate the ease of installation, longevity of the sensor installation, and sensitivity of the measurement technique, but highlight problems with the application, especially if electromagnetic noise affects data quality. As no confirmed substantial damage was recorded during the monitoring period on both structures, we evaluate load tests to investigate the effect of static load on the structures and the coda monitoring results. The experiments show that the influence of load can be detected, even if the temperature influence is not removed from the data. This indicates that online damage detection with coda monitoring is possible, but further research on damage detection in real-world structures has to be conducted to confirm laboratory findings.
• Finite element mesh construction for seismic analysis using drone imageryDelaney, Evan; Marty, Patrick; Gebraad, Lars; Zunino, Andrea; Fichtner, Andreas; 10.3217/978-3-99161-057-1-013Computational meshes serving as input to wave simulations are often crafted manually and bear a significant cost to construct. The aim of this work is to minimize that overhead and apply it to structures lacking models and meshes, particularly in earthquake-prone regions, to image and monitor their structural health in response to ground movement. To address this challenge, we have developed a workflow to facilitate the creation of 3D finite element meshes, starting with 2D photos acquired by an inexpensive consumer-level unmanned aerial system (i.e., a drone). After photo acquisition, the process proceeds by utilizing computer graphics and vision software to transform these photos into a 3D surface composed of triangles. Surface meshes are generally sufficient products for other workflows that likewise create 3D assets via reconstruction methods (e.g., for topographic mapping, archiving, and entertainment). However, to simulate waves through complex structures with high fidelity, we employ a spectral element wave solver, which requires a 3D volume composed of hexahedra. The steps from a 3D triangular surface to a 3D hexahedral volume include enclosing the surface, conditioning, and remeshing it with appropriate element geometry. We apply a first version of this workflow to the Contra (Verzasca) dam in Switzerland, from which we discuss key stages, challenges, and learnings in developing the pipeline – showcasing elastic wave simulation through the constructed mesh.
• A review of methods and challenges for monitoring of differential settlement in railway transition zonesNasrollahi, Kourosh; Nielsen, Jens; Dijkstra, Jelke; 10.3217/978-3-99161-057-1-014Differential settlement in ballasted railway tracks, particularly in transition zones between two track forms, poses a critical challenge for railway infrastructure. Such settlement, often exacerbated by a stiffness gradient due to changes in track superstructure and substructure, typically causes a local dip in the longitudinal track level a few metres from the transition, leading to higher dynamic traffic loading and reduced passenger comfort. Regular monitoring of transition zones is essential for safe operations and cost-effective maintenance. This paper reviews methods for monitoring differential settlement in railway tracks. To measure the properties and loading of the superstructure, potential methods include fibre Bragg grating (FBG) sensors, point receptance measurements, track geometry (and track stiffness) recording cars, and wheel load impact detectors (WILD). Characterisation of the subgrade can be carried out via a multichannel analysis of surface waves (MASW), dynamic cone penetration tests (CPT), interferometric synthetic aperture radar (InSAR), frost sticks for temperature monitoring, and total stations. Lessons learned from an in-situ measurement involving an extensive FBG-based system deployed in northern Sweden to monitor a transition zone in harsh weather conditions are presented. Integrating a combination of monitoring methods with a simulation model to verify and support the accurate prediction of differential settlement is a useful approach to addressing challenges associated with track stiffness gradients and guiding the improvement of transition zone designs.
• Advanced Structural Health Monitoring and Predictive Maintenance of the Parchi Viaduct Using Distributed Fiber Optic Sensors and Digital Twin TechnologyWeissenbach, Nils; Penasa, Massimo; 10.3217/978-3-99161-057-1-015Aging infrastructure poses significant challenges in ensuring safety, reliability, and long-term serviceability. The Parchi Viaduct, a 3-km multi-span structure on Milan’s A51 Eastern Ring Road, experienced critical degradation in its Gerber saddles, necessitating temporary closure for safety assessments. In response, Milano Serravalle Milano Tangenziali S.p.A. and CAEmate S.R.L. deployed an advanced Structural Health Monitoring (SHM) system, integrating distributed fiber optic sensing (DOFS) and a physics-informed digital twin (PINN) to enable real-time load-bearing capacity evaluation and predictive maintenance. This paper presents the implementation of the WeStatiX SHM platform, utilizing DOFS to capture strain, temperature, and vibration data while dynamically updating a finite element model (FEM) through inverse analysis and multi-objective optimization. By continuously refining modal parameters such as natural frequencies, mode shapes, and damping ratios, the system enables early detection of structural anomalies and degradation trends. The validated digital twin successfully predicted real-world structural behavior, confirming residual load-bearing capacity despite saddle deterioration and supporting the safe reopening of the viaduct under real-time monitoring per Italian NTC standards. Load test results and FEM simulations demonstrated excellent agreement, with taller piers exhibiting ~20% greater deflection, emphasizing pier height's impact on load distribution and deformation patterns. These findings enhance predictive maintenance planning, improve stress redistribution modeling, and contribute to prolonging the structural lifespan of aging infrastructure assets.
• A digital twin based integrated sustainability and quality assurance concept for subway constructionsGrosse, Christian U.; Wurzer, Otto; 10.3217/978-3-99161-057-1-016In regard to subway structures, non-destructive testing and structural health monitoring techniques are beneficial for construction and operation, which require an integrated quality control and sustainability concept. Such an integrated concept is presented, focusing on two main tasks. Inspection during construction will lead to a better quality of the components and structures. Proper data can be integrated into a building information model (BIM). The conceptual design should, however, anticipate later impacts and possible deteriorations at critical parts. The building information model could then be continued (updated) in the form of structural health monitoring (SHM) to make (visual) maintenance of subway structures more efficient, resulting in fewer disruptions (fewer closures, less downtime) and lower costs. It can also contain sensors at non-visible or non-assessible locations. Recording impacts on the structure (e.g. loads, vibrations, chlorides) enables a digital model as a so-called digital twin and the calculation of the remaining service life. Such a concept is presented for a new subway station in Munich.
• Evolving reliability-based condition indicators for structural health monitoring into a digital twin of a cable-stayed bridgeHerbrand, Martin; Wenner, Marc; Lazoglu, Alex; Ullerich, Christof; Zehetmaier, Gerhard; Marx, Steffen; 10.3217/978-3-99161-057-1-017In Germany, the bad condition of many older bridges and changes in the code provisions often result in deficits after assessment and recalculation. In case the necessary structural safety is not provided, structural health monitoring can be employed to gain knowledge about the time variant actions, the progression of structural damage and the overall condition of structures. To allow for effective use of the usually dense monitoring raw data, the derivation of condition indicators is key, since they indicate a need for action for the owners and the engineers. At the same time, real-time data as well as comprehensive condition indicators are key elements for creating a Digital Twin of a structure, as a Digital Twin requires a bidirectional flow of data, which affects the physical entity of the twin. In this paper, a method for deriving condition indicators from monitoring data is described which was developed for a large cable-stayed bridge, the Köhlbrand Bridge in Hamburg, Germany. The method allows for the calculation of a reliability index as a time variant condition indicator based on dynamic monitoring data, which is then implemented into a Digital Twin of the structure.
• Advancements in Distributed Fiber-Optic Sensing: Comparing Brillouin and Rayleigh Technologies for Geotechnical and Structural MonitoringNöther, Nils; Facchini, Massimo; Aguilar-López, Juan Pablo; 10.3217/978-3-99161-057-1-018We report on recent developments in distributed fiber-optic strain and temperature sensing (DTSS) technologies. In recent years, both Brillouin- and Rayleigh-based fiber-optic sensing systems have found an increasing number of applications measuring static and dynamic displacement and deformation events in geotechnical and structural health monitoring. The focus of this contribution is on Brillouin-based DTSS systems, for which we present recent advancements in spatial resolution and signal-to-noise ratio under harsh real-world conditions. The state-of-the-art Brillouin DTSS technology is considered also in relation to Rayleigh-based technologies like c-OFDR and DAS systems that also play an increasing role in geotechnical and structural monitoring, in order to illuminate the technology-specific strengths and challenges within the DFOS family. Recent insights from industrial projects and research activities in embankment monitoring are presented.
• Hybrid monitoring systems: synergising distributed fibre optic sensing with spot measurementsSieńko, Rafał; Howiacki, Tomasz; Bednarski, Łukasz; Zuziak, Katarzyna; 10.3217/978-3-99161-057-1-019The diagnosis and maintenance of both new and ageing infrastructure are among the main challenges facing the civil engineering and geotechnical industries today. The effectiveness of monitoring systems depends on several factors, including the choice of measurement techniques. Conventional point-based methods (e.g., vibrating wire sensors, electrical strain gauges, or accelerometers) are inherently limited by their locality, as they cannot directly capture what occurs between discrete measurement points. In contrast, distributed fibre optic sensing (DFOS) introduces new capabilities for structural condition assessment by enabling continuous measurement of various physical quantities along the entire length of the sensor. This eliminates the risk of missing localized extreme events or damages, such as cracks, leakages, or stress concentrations. However, the widespread adoption of DFOS is hindered by the high costs of optical interrogators, which often restrict its use to periodic measurements rather than fully automated monitoring. A practical solution to this challenge is the synergistic combination of point-based and distributed technologies within hybrid monitoring systems. Such systems leverage the strengths of both approaches, offering a more comprehensive understanding of structural behavior. This paper explores the concept of hybrid systems, illustrating their potential and real-world applications through selected case studies.
• Water distribution pipeline anomaly detection using distributed acoustic sensing (DAS)Jasiak, Maksymilian; Chiu, Shih-Hung; Saw, Jaewon; Hubbard, Peter; Katzev, David; Soga, Kenichi; 10.3217/978-3-99161-057-1-020Leak detection for water pipelines, and anomaly detection more broadly, is vital to ensuring reliable access to drinking water. Monitoring transmission and distribution pipelines supports proactive fault detection to reduce water loss amid deteriorating infrastructure and depleting water resources. Distributed acoustic sensing (DAS) in the form of phase-sensitive optical time-domain reflectometry (φ-OTDR) can quantify vibrations and sound along fiber optic cables over long distances with high spatial resolution and frequency. In this study, DAS was deployed on a new fiber optic cable-instrumented pipeline to investigate DAS sensitivity to pipe water leakage noise. A reproducible workflow for system deployment and signal processing aimed at pipe water leak detection in field conditions is presented. The influence of fiber optic cable type (tight-buffered vs. loose tube) and installation condition (pipe-mounted vs. trench-lain) on DAS sensitivity was assessed during pipe water filling and simulated leakage. Findings demonstrate relatively high sensitivity to water leak noise detection when DAS is deployed on fiber optic cables near the pipeline. This informs best practices for data-driven pipeline monitoring by presenting a reproducible procedure to operationalize water pipeline leak detection using DAS.
• Integrating Distributed Acoustic Sensing for Damage Detection in Old Pre-Stressed Concrete Girders: Preliminary Experimental ResultsStrasser, Lisa; Lienhart, Werner; Moser, Thomas; Anžlin, Andrej; Kosič, Mirko; Kreslin, Maja; Hekič, Doron; 10.3217/978-3-99161-057-1-021In this study, we investigate the load-bearing capacity of pre-stressed concrete girders under various damage levels. We employed Distributed Acoustic Sensing (DAS) technology to monitor and quantify changes in the girder response as damage levels were incrementally introduced. This approach enabled the real-time measurement of dynamic behavior over the entire length of the girder, allowing for a detailed characterization of damage-induced structural changes. To complement the DAS-based approach, we also applied classical acceleration-based damage detection techniques. By integrating these methods, we aimed to cross-validate the results and provide a more comprehensive understanding of damage progression and its impact on structural performance. The experimental campaign, conducted in Ljubljana, ZAG, involved full-scale testing of pre-stressed concrete girders subjected to controlled damage scenarios. This setup ensured a realistic assessment of the girders’ residual capacity and failure mechanisms. The paper presents preliminary results from this experimental study, emphasizing the capability of DAS measurements to detect and characterize damage, while also comparing its performance against traditional methods. By combining advanced sensing technologies with established techniques, this research highlights the potential of DAS as a transformative tool in structural health monitoring.
• Structural performance monitoring for concrete girder bridges with distributed fiber optic sensorsFabbricatore, Francesco; Bertola, Numa; 10.3217/978-3-99161-057-1-022The alarming frequency of bridge collapses in recent years underscores the critical need for advanced monitoring strategies tailored to existing infrastructure. Many concrete bridges, built decades ago, now face increasing traffic demands and environmental stressors that threaten their structural integrity. This study investigates the use of distributed fiber optic sensors (DFOSs) with high spatial resolution (independent strain measurements every 2.6 mm) during static load tests to assess the structural performance of concrete girder bridges. The goal is to gain a deeper understanding of their condition using data-driven approaches. The fiber optic technology provides detailed strain profile information that gives insights into global bridge behavior, such as stress distributions, support conditions and static responses. It also allows the detection of cracks along the fiber path and other localized effects that may remain undetected without a calibrated numerical model. This method of structural performance monitoring is applied to a prestressed concrete bridge in Switzerland. Static load tests have been performed on a full-scale bridge in Switzerland and the resulting distributed strain datasets allow the accurate understanding of bridge behavior, including deflection extrapolation and crack detection. The results underline the potential of DFOS to develop novel data-driven solutions for extending the service life of structures.
• AI-driven Smart-Liner System with DFOS for Digital Twin-Based Real-Time Monitoring of Oil and Gas InfrastructureDuan, Junyi; Wang, Xingyu; Yan, Huaixiao; Wang, Sike; Huang, Ying; Tao, Chengcheng; 10.3217/978-3-99161-057-1-023This study presents an innovative AI-powered smart-liner system designed to enhance the safety and efficiency of oil and gas transportation and storage infrastructure. By integrating polymer composite liners with distributed fiber optic sensors (DFOS), the system enables continuous monitoring of mechanical deformations and damage formation, providing real-time insights into the infrastructure’s condition throughout its lifespan. Finite element analysis (FEA) is employed to simulate the mechanical responses of the smart-liner-protected specimen over time. Machine learning (ML) algorithms are applied to analyze images generated from collected DFOS data, enabling the identification and assessment of risk variations across different locations and time steps. This approach demonstrates the high accuracy and effectiveness of ML in automatically detecting deformations and crack formation under buckling loading conditions. The methods enable comprehensive structural health monitoring, allowing for precise localization, visualization, and quantification of mechanical changes and damage within the infrastructure. With the above approaches, the smart-liner system facilitates continuous data collection across the entire protected surface, supporting the development of a dynamic digital twin model that evolves alongside the infrastructure. The findings provide critical insights for the oil and gas industry, offering an advanced and efficient solution for monitoring and mitigating risks associated with transportation and storage infrastructure.
• Middle range, rapid strain sensing based on PNC-OFDR and its application to bridge monitoringYoshimura, Yuichi; Fujiwara, Kotaro; Taira, Yohei; Imai, Michio; Zhang, Chao; Ito, Fumihiko; 10.3217/978-3-99161-057-1-024Distributed fiber optic sensing is a suitable method for long-term, wide-area monitoring of civil engineering structures such as the ground, tunnels, dams, and bridges. In recent years, distributed strain sensing technologies such as distributed acoustic sensing (DAS) and optical frequency domain reflectometry (OFDR), which can realize real-time monitoring, have made remarkable progress. In particular, OFDR, which performs strain sensing with high spatial resolution, can quantitatively evaluate the strain distribution of civil engineering structures with an accuracy comparable to conventional strain gauges. This method has been limited in its application to structural health monitoring due to its short measurement range. However, by extending the sensing distance, it is evolving into a practical technology for on-site testing. This paper introduces middle-range, rapid strain sensing based on Phase-noise-compensated OFDR (PNC-OFDR) and its application to bridge monitoring. Optical fiber sensors were installed on bridge girders, and the change in strain distribution when the moving load was applied by vehicles was measured using the PNC-OFDR sensing system.
• Advanced Structural Monitoring and Predictive Maintenance for Railway Bridges Using Distributed Fiber-Optic SensorsMunoz, Felipe; Eguidazu, Iván; Rodriguez, Julio; Gaston-Beraza, Diego; Basarte, Fernando; Urricelqui, Javier; Perez-Casas, José María; Jimenez-Rodriguez, Marco; 10.3217/978-3-99161-057-1-025This submission presents a structural monitoring solution for railway bridges and viaducts that leverages distributed fibre optic sensors (distributed temperature and strain sensing, DTSS, and distributed acoustic sensing, DAS) to capture both long-term static trends and dynamic behaviour under train loads. The long-term monitoring uses hourly DTSS strain measurements, accounting for day/night and seasonal variations, while the dynamic monitoring system records real-time strain and vibration data during train passages. By integrating these measurements with structural calculation services, the system can detect anomalies (e.g., stiffness changes, potential cracking) and inform predictive maintenance. Lastly, the results are displayed via a digital twin, providing an intuitive, web-based platform for analysing historical data and forecasting future conditions.
• Structural health monitoring in underground mining using fiber-optic sensing and 3D laser scanning for digital twin developmentMartin, Michael Dieter; Nöther, Nils; Paffenholz, Jens-André; 10.3217/978-3-99161-057-1-026This study aims to evaluate the use of distributed fiber-optic strain and temperature sensing for structural health monitoring in underground mining drifts and chambers including 3D mobile laser scanning. This method seeks to create a digital twin to improve safety and efficiency through better digital planning. Temperature and deformation data from distributed fiber-optic sensing (DFOS) cables will serve as boundary conditions of the combined ventilation and geomechanical models of the drift and chambers. Initially, a 60-meter-long drift will be monitored using fiber-optic cables. Next, deformations of a flexible arch support, induced by hydraulic cylinders, will be observed. A hydraulic cylinder will then apply load orthogonally to the rock. Fiber-optic cables will be inserted and cemented into the rock, along rock bolts, and in boreholes around each bolt to measure deformations from rock bolt pull-out tests. Preliminary examinations identified the best adhesive bonding method for DFOS cables, considering the specific ambient conditions. A 3D point cloud will be used to plan and validate the cable installation. The meshed 3D cloud will serve as the foundation for the combined ventilation and geomechanical models, creating a virtual reality-capable digital twin enhanced with live DFOS measurements.
• Fibradike sensor: validation through full-scale field testingHöttges, Alessio; Rabaiotti, Carlo; Rosso, Alessandro; 10.3217/978-3-99161-057-1-027Earthen geohydraulic structures, such as dams and river embankments, are vital for water resource management and flood control, especially as climate change and urbanization increase hydrological risks. Internal erosion, often triggered by seepage, remains a major failure mechanism and can cause sudden, catastrophic collapses. Traditional monitoring systems lack the spatial and temporal resolution needed for effective early detection. To address this gap, a novel Distributed Pressure Sensor (DPS) based on distributed fiber optic (DFO) technology has been developed by the University of Applied Sciences of Eastern Switzerland (OST). The DPS offers high spatial resolution and extended range, enabling precise measurement of distributed pore water pressure - key for early detection of internal erosion processes. Following successful laboratory validation, the DPS was deployed in a full-scale test embankment (84 m long, 39 m wide, 4 m high) at the AIPo Research and Technical Centre in Boretto, Italy. Preliminary results show that the DPS accurately captured pore pressure evolution, matching conventional piezometer readings while detecting localized variations and two-dimensional flow effects that point sensors could not resolve. These findings highlight the DPS system’s strong potential for improving early warning capabilities in geohydraulic structure monitoring.
• Identification and quantification of concrete cracks using various distributed fiber optic sensing techniquesMonsberger, Christoph M.; Winkler, Madeleine; Kornberger, Anna Theresa; Schlicke, Dirk; 10.3217/978-3-99161-057-1-028Distributed fiber optic sensors (DFOS) are extensively used for concrete crack monitoring in recent years, especially in scientific-related applications and laboratory testing. These mainly focus on Rayleigh scattering due to its high spatial resolution and strain resolution, but with significant limitations in the sensing range. This contribution introduces an enhanced laboratory test series, in which five individual test specimens were equipped with multiple installation setups and tested under well-known conditions. The sensing network was interrogated using four different sensing units based on high-resolution Rayleigh as well as Brillouin scattering. The resulting strain sensing profiles do not only allow an identification of the crack location itself, but also a quantification of the crack width. It can be demonstrated that Brillouin sensors are definitely capable of capturing reliable crack widths over long distances, despite their limitation in the spatial resolution. The outcomes are significantly important in practice as civil infrastructures often require monitoring over several kilometers.
• DFOS-Based Monitoring of Prestressed Concrete Bridge GirdersLila, Kleo; Herbers, Max; Richter, Bertram; Agreiter, Andrea; Kreslin, Maja; Triller, Petra; Anžlin, Andrej; Lienhart, Werner; Marx, Steffen; 10.3217/978-3-99161-057-1-029Due to bridges’ critical role in transportation networks, the assessment and maintenance of existing bridges have become a priority. Prestressed concrete bridges constitute a significant portion of Europe’s transportation network, yet many no longer meet today’s technical requirements. This is primarily due to two factors: (i) the unforeseen increase in heavy goods traffic, and (ii) insufficient experience with early reinforced and prestressed concrete construction methods, coupled with inadequate regulations, which resulted in design weaknesses and structural deficiencies. One critical failure mechanism, identified when recalculating existing bridges based on updated guidelines, is insufficient shear load-bearing capacity, which has prompted the premature demolition of numerous bridges. A thorough understanding and rigorous monitoring of shear behavior is essential since neglecting this problem could lead to notable consequences, especially for aging infrastructure. In this paper, a distributed fiber optic sensor (DFOS) based monitoring system, inspired by shear detection concepts, is tested. A decommissioned prestressed concrete bridge girder was equipped with a DFOS grid, allowing for detailed monitoring of crack width, location, and shape. Preliminary test results confirm the successful installation and early detection of cracks, highlighting the system’s potential to identify microcrack formation, monitor crack growth, and support maintenance strategies.
• Proposed approach for direct rail state monitoring with distributed acoustic sensing DASDługosz, Szymon; Howiacki, Tomasz; Sieńko, Rafał; Bednarski, Łukasz; 10.3217/978-3-99161-057-1-030Railways are one of the fundamental modes of transportation, dating back centuries. They allow for the movement of people and goods across hundreds and thousands of kilometres. Such a large system relies on precise timing and excellent organization. Any incident or failure can result in losses amounting to millions of euros and cause unacceptable delays. Monitoring the condition of the railway is necessary to ensure safety and system effectiveness, but it is challenging due to the long distances that need to be monitored. Conventional sensors can provide high-quality data, but they do not offer a complete picture of the railway’s state, and local defects can be overlooked. A great solution for railway monitoring is DAS. A fibre optic sensor integrated with the structure can be used to obtain information about strain and vibration, with a fine resolution of even down to 1 metre, over tens of kilometres of track. Installing the sensor in the railway substructures can be challenging and exposes the sensor to potential damage. Another approach discussed in the article is to attach the sensor directly to the rail. Long sections of track can be covered with monitoring within a few hours using automated machine, enabling direct measurement of the rail’s condition. This paper presents the results of such installation, showing the potential of synergizing monolithic distributed fibre optic sensors with DAS technology to increase the safety and reliability of rail transport.
• Monitoring Timber Structures with Fiber Optics Sensors: State of the Art and Application to a Timber BeamMansilla-Ruiz, Roberto; Paya-Zaforteza, Ignacio; Garcia-Castillo, Ester; Calderon, Pedro A.; 10.3217/978-3-99161-057-1-031Fiber optic sensors (FOS) offer compelling advantages for Structural Health Monitoring (SHM). However, their application in timber structures remains underexplored. This article reviews the state-of-the-art use of FOS in timber structures and presents an experimental study conducted at the Universitat Politècnica de València. A 3-meter-span timber beam was subjected to a four-point bending test and instrumented with long-gauge strain FOS. The measured strains were used to derive stresses, which were then compared to theoretical values. The results highlight the potential of FOS for accurate stress monitoring in timber elements and contribute valuable insights to the advancement of SHM in sustainable construction.
• Pi-bracket fatigue sensor for crack detection monitoring near stiffeners in bridge girdersTelehanic, Boris; Mufti, Aftab; Thomson, Douglas; Bakht, Baidar; Murison, Evangeline; 10.3217/978-3-99161-057-1-032This study investigates an innovative pi-bracket sensor system integrating distributed fiber optic sensing with Brillouin Optical Time Domain Analysis to detect cracks in bridge girders near stiffeners. The system is designed to overcome challenges in crack detection at these critical locations. Experimental validation was conducted on a 3-meter steel beam featuring a welded stiffener positioned 25mm from a simulated crack. An aluminum pi-bracket served as a mounting device for the fiber optic sensor. Comparative analysis between experimental measurements and finite element simulations demonstrated the system's ability to detect crack openings as small as 0.2mm. Abaqus Finite Element Analysis predicted strain values of 145μɛ, while laboratory experiments recorded 129μɛ, a discrepancy of approximately 11%. Strain concentrations were localized to the regions where the pi-bracket was in direct contact with the beam. The strong correlation between computational models and empirical data substantiates the efficacy of the proposed sensing system. These findings highlight the system's potential for structural health monitoring of bridge infrastructure, particularly for detecting and quantifying cracks near stiffeners.
• Monitoring of civil engineering structures – current and future use casesGeorge, Joyal K.; von Wangenheim, Kristian; Kaplan, Felix; Schneider, Ronald; Hindersmann, Iris; 10.3217/978-3-99161-057-1-033Monitoring represents an effective approach for addressing the diverse challenges associated with the maintenance of civil engineering structures. It contributes to improving both the availability and safety of these structures. By increasing the amount of information available about the structure, monitoring supports better-informed decisions regarding its preservation. Due to the complexity of monitoring applications, specific use cases are outlined. A key advantage of these use cases is that new technologies can be tested within well-defined and limited scopes. The use cases “monitoring of known, localized damage,” “monitoring of known deficits identified through reassessment or resulting from outdated design procedures” and “monitoring aimed at assessing traffic loads and their effects” currently account for the majority of implemented monitoring measures. Their practical implementation is demonstrated through case studies from the Brandenburg State Road Authority. Additional use cases, such as “monitoring to support structural inspections” - for example through the use of imaging techniques - and “monitoring of major structures,” such as large viaducts, are gaining importance, with initial practical examples already present in Europe. Future applications reveal potential for expanded use, particularly in the context of “monitoring to support predictive lifecycle management.” This will become increasingly important in the implementation of digital twins, as announced in the national BIM master plan. Furthermore the concept of a “Birth Certificate” is intended to establish a reference state of the structure prior to commissioning, which can then be used for comparison with future measurements over time. The integration and interaction of these individual use cases pave the way for the implementation of digital twins.
• A Structural Health Monitoring Framework For Intelligent and Sustainable Infrastructure: A Conceptual PerspectivePedram, Masoud; Taylor, S; Hamill, Gerard; 10.3217/978-3-99161-057-1-034This paper presents a vision for next-generation Artificial Intelligence (AI) based structural health monitoring (SHM) systems through the lens of DREAM-SHM: a framework comprising Dynamic, Real-time, Evaluative, Adaptive (AI-based), Modular, Self-diagnostic, Holistic, and Multi-sensory principles. The aim is to enable smart infrastructure that can sense, and evolve corresponding to structural behaviour, material degradation, environmental effects, and changing operational or economic constraints. The paper reviews current SHM technologies, highlighting the strengths and limitations of contact-based sensors, such as accelerometers, strain gauges, fibre optic sensors, and non-contact approaches including vision-based systems, infrared thermography, radar, and ultrasonic techniques. Emphasis is placed on their integration with wireless sensor networks, Internet of Things (IoT) platforms, and Artificial Intelligence (AI) for advanced data fusion, anomaly detection, and predictive analytics. The computational aspects underpinning SHM systems, such as cloud-edge processing, machine learning, and multi-modal sensor data integration, are described to enable timely and informed decision-making. In addition, the paper situates DREAM-SHM within the context of sustainability, demonstrating how adaptive and intelligent SHM systems support the goals of circular economy and net-zero carbon by prolonging asset life, reducing maintenance burdens, and improving environmental responsiveness. This work outlines a pathway toward structurally intelligent and resource-efficient infrastructure.
• Best Practices for Data Acquisition System Design: Practical Wisdom for Engineers and PractitionersSimmonds, Tony; Randall, Brent; 10.3217/978-3-99161-057-1-035As global infrastructure ages and demands on new and existing structures increase, effective monitoring programs are essential for managing risk and public safety. This paper provides a practical guide for practitioners to design and implement structural health monitoring (SHM) systems, leveraging the combined expertise of the authors, who have extensive experience with leading equipment manufacturers. Building on the 10 Steps of Data Acquisition System Design, the paper outlines best practices for developing robust monitoring systems tailored to bridges, dams, and other critical infrastructure. These steps include defining objectives, selecting appropriate sensors, communications design, data acquisition (DAQ) system design, power system considerations, civil works and mounting structures, installation, and managing data effectively. A significant focus is placed on sensor and DAQ selection, exploring their critical roles in SHM system performance. The paper covers practical techniques for selecting, installing, maintaining, calibrating, and verifying sensors across traditional analog, frequency, and digital technologies. Examples from large channel count wired systems and distributed wireless monitoring systems are shared to illustrate diverse applications. This paper aims to deliver actionable insights and practical wisdom, equipping attendees with the tools to overcome real-world challenges and achieve reliable, scalable, and long-lasting SHM implementations.
• From Insight to Action: Deploying SHM for a suspension bridgeLilja, Heikki; Mikkonen, Atte; Ukkonen, Eero; 10.3217/978-3-99161-057-1-036This paper presents the implementation of a structural health monitoring (SHM) system for a suspension bridge approaching the end of its service life. Constructed in the 1960s with a 220-meter main span, the bridge is the sole vital link to a region hosting key socioeconomic industries. Over time, heavy traffic—with vehicles often exceeding 80 tons—has led to pronounced fatigue issues. Throughout its operational life, extensive repairs—such as the addition of a cantilevered pedestrian lane, modifications to the bearing system, and hanger replacements—have been undertaken. More than a decade of monitoring via over 50 sensors has yielded comprehensive data that both ensures safety and informs maintenance strategies. A sophisticated finite element (FE) model, developed in 2023 and calibrated using real-time data, improved the correlation between simulated and actual performance. Following this, a streamlined near-real-time monitoring framework was established in early 2024 to promptly identify structural anomalies. Designed for adaptability, the SHM system can be implemented on various structures. This study highlights the importance of clear data presentation in enabling informed decisions that optimize infrastructure management and enhance operational safety.
• Use of Monitoring for Highway Bridges on Federal Highways in Germany – Current Status and Future DevelopmentHindesmann, Iris; 10.3217/978-3-99161-057-1-037The use of monitoring for bridges on federal highways is currently not widespread. Monitoring is primarily used when damage is already present or when deficits resulting from the recalculation or which occur due to the design. The potential of monitoring to support maintenance towards a predictive lifecycle management approach is not being fully utilized. As part of the exchange with the structure managers in Germany and a literature search, it has been shown that the challenges lie in the lack of standardization, insufficient expertise, and missing fundamental principles. To address these challenges, various research projects have been planned and carried out. These include the “Documentation on Monitoring of Bridge Structures,” the “Guideline – Strategic Use of Monitoring for Civil Engineering Structures,” the “Birth Certificate for Bridge Structures,” and the project on “Standardized Data Models.” This article aims to present the challenges and initial solution approaches from these projects. The goal is to illustrate support options for the increased and targeted use of monitoring in bridge structures on Germany’s federal highways. Additionally, the article provides a classification of these challenges within the European context.
• Study on the suitable sensor locations for tilt monitoring of power transmission towerKurihara, Tatsuya; Saeki, Masayuki; 10.3217/978-3-99161-057-1-038Due to recent extreme weather conditions, there have been many reports of damage to infrastructure. For example, two power transmission towers collapsed due to landslides in 2022. The landslides may not only cause the tower collapse but also cause the base displacement. The base displacement of only a several millimeters can generate secondary stress, resulting in member deformation and insufficient strength of the steel tower members. Therefore, the towers that are at risk of landslides are surveyed once a year to investigate the progress of base displacement. However, the on-site investigation creates other risks, such as delays in detection and accidents during the travel to the site. So, the authors have been developing the tilt monitoring system of the power transmission towers. In the tilt monitoring system, one tilt sensor is installed on each of the four main members of the tower. The progress of base displacement is monitored by checking whether the observed tilt change exceeds a set threshold. In the current system, the threshold value is tentatively set to be 0.05 degrees. This system has already been installed to about one hundred towers in the field. In this study, a full-scale experiment is newly conducted to examine the optimal installation location of the tilt sensors to monitor the base displacement. In this experiment, ten tilt sensors are placed on each of the four main members, and the sensitivities to the base displacement are examined in detail.
• On potentials and challenges of physics-informed structural health monitoring for civil engineering structuresBaeßler, Matthias; Ebell, Gino; Herrmann, Ralf; Falk, Hille; Schneider, Ronald; 10.3217/978-3-99161-057-1-039Physics-informed structural health monitoring, which integrates realistic physical models of material behavior, structural response, damage mechanisms, and aging processes, offers a promising approach to improve monitoring capabilities and inform operation and maintenance planning. However, the associated technical challenges and model requirements are context-specific and vary widely across applications. To illustrate the relevance and potential of the topic, two application examples are presented. The first focuses on monitoring the modal characteristics of a prestressed road bridge, where strong sensitivity to temperature variations limits the diagnostic capabilities of conventional vibration-based global monitoring. The discussion highlights how environmental influences can obscure structural changes, and emphasizes that purely data-based approaches are inherently limited to detecting anomalies and do not enable comprehensive condition diagnostics. The second example explores a physics-informed monitoring approach for prestressed concrete bridges affected by hydrogen-induced stress corrosion cracking. By combining acoustic emission data with a calibrated acoustic model of the structure, it is possible to detect and localize wire failures. As an outlook, the integration of mechano-electro-chemical models for stress corrosion cracking is discussed, enabling predictive assessments of the strand condition.
• Retrofitting load measurement devices on existing anchored structuresRebhan, Matthias J.; Daxer, Hans-Peter; Schuch, Markus A.; Klass, Clemens; Scharinger, Florian; Reiterer, Michael; 10.3217/978-3-99161-057-1-040Anchored structures are an essential part of infrastructure corridors, as they enable high cuts and the reinforcement of existing structures. As pre-stressed elements, anchors allow for economical construction with low-deformation, making them critical components for ensuring stability. Consequently, comprehensive monitoring is often required alongside inspections, especially when the reliability of the load-bearing elements is in question - e.g., due to creep or corrosion of metallic components. Based on research activities due to the ageing of such structures, several options have been investigated to retrofit load measurement devices to already installed pre-stressed anchors. In addition to different types of coupling methods, this paper presents first results from a trial in which, through the retrofitting of an external thread to strands, a new method for load monitoring is investigated. These methods are particularly useful when information about the current anchor load is lacking, as they provide better insight into the condition of the tension elements and, furthermore, the overall behavior of the structure. The paper shows some approaches already in use and the results of an initial test series, in which the potential of a new approach to amend a load measuring device to an already installed pre-stressed anchor is investigated and validated.
• Reliability Assessment of Structural Health Monitoring Systems using Model – Assisted Probability of Detection and Bayesian Model UpdatingJaelani, Yogi; Marsili, Francesca; Grashorn, Jan; Knoth, Sven; Keßler, Sylvia; 10.3217/978-3-99161-057-1-041Structural health monitoring (SHM) is a key method for assessing the condition of civil infrastructure, detecting and localizing damage through continuous data acquisition. Damage detection methods are divided into physically based approaches, using finite element (FE) models, and data-driven approaches, relying on signal processing. A key challenge in SHM is the lack of data from the damaged state, which complicates the validation of the technique. However, the successful deployment of SHM systems on real civil infrastructure depends mainly on their reliability. For non-destructive testing (NDT) systems, the Probability of Detection (POD) is an accepted approach for quantifying reliability. In contrast to NDT, there is no generally applicable procedure to assess the reliability of SHM systems. This study addresses this gap by evaluating SHM reliability with POD models and data generated from calibrated FE models. These FE models are calibrated through Bayesian inverse methods. To manage computational challenges, generalized Polynomial Chaos Expansion (gPCE) surrogate models are employed. These methods are tested using vibration-based measurements on a laboratory-scale four-degree-of-freedom (4-DOF) wood frame. The results highlight the use of MAPOD and limitations of the method, emphasizing their potential to enhance SHM reliability and enable smarter infrastructure systems.
• Advanced Monitoring Systems for Infrastructures: Integrating 6D Sensors and Low-Cost High-Precision GNSSWindl, Roman; Weitensfelder, Herbert; Stempfhuber, Werner; 10.3217/978-3-99161-057-1-042Structural Health Monitoring (SHM) is vital for ensuring the safety and longevity of infrastructures, utilizing various sensor-based techniques to detect damage, assess performance, and monitor long-term deterioration. Traditional methods, such as visual inspections, lack precision and are prone to human error, whereas more advanced techniques like vibration-based monitoring, acoustic emission, strain gauges, and GNSS offer real-time damage detection and millimeter-level precision but often require complex planning and high costs. The presented 6D sensor, developed for infrastructure monitoring, accurately measures complex displacements and rotations, offering enhanced precision through a combination of machine learning and mathematical algorithms. When paired with low-cost, high-precision GNSS systems, it provides comprehensive real-time data on both localized and large-scale structural movements, improving insights into infrastructure behavior under various environmental conditions and loads.This paper explores the integration of 6D sensors with GNSS technology, discussing the advantages of real-time monitoring for predictive maintenance and presenting insights from ongoing project results.
• Structural damage detection, localization, quantification for high-rise buildings under earthquake excitations based on machine learning and sub-structuring approachAbdelbarr, Mohamed H.; Ikeda, Yoshiki; Masri, Sami F.; 10.3217/978-3-99161-057-1-043This work presents an evaluation of promising sub-structuring and machine learning SHM approaches suitable for high-rise buildings, based on real data from an 18-story steel-moment resisting framing building, tested at an E-Defense facility in Japan. This building is instrumented with a relatively dense set of sensor arrays and is subjected to different excitation levels until full collapse. The main contribution of this study is to demonstrate the practical feasibility of the proposed sub-structuring approach in conjunction with machine learning when relying on different levels of response measurements. The study assesses the accuracy and reliability of the estimates of the dominant modal features of the structure and can subsequently provide a probabilistic measure of confidence in the extent and location of changes/damage if an anomaly is detected, as well as the propagation of damage throughout the structure's life span. Due to the minimal computational resources needed to implement the sub-structuring approach, it is shown to be quite efficient for near-real-time applications where important structures need to be continuously monitored for sustainability as well as resiliency requirements.
• ‘Machine Learning – Based Data Interpretation and Visualization for Tunnel Monitoring: A Case Study of Changshui Airport Tunnel’Sourov, Asif Ahmed; Xue, Yadong; Guo, Yongfa; Ding, Wenyun; Yang, Jinjing; 10.3217/978-3-99161-057-1-044Specially, when it comes to such risky construction projects as the Changshui Airport Railway Tunnel, the ground settlements surrounding the structure will be observed and evaluated. This case study is a combination of advanced machine learning algorithms, which are augmented with MATLAB to reinterpret, visualize and analyze settlement trends based on real – world tunnel monitoring data. The research starts with some of the time – series data that had been registered in the tunnel (settlement, factors that impact settlement, temperature, etc.). K – means clustering and hierarchical clustering are used to classify the settlement patterns and the clustering result indicates the difference in settlement monitoring points. Finally, we have used the feature importance analysis to explore the most significant factors that affect settlement decisions and to know more about the settlement processes in tunnels. The discussion of the Random Forests, Gradient Boosting and Artificial Neural Networks regression models is provided to predict settlement patterns to enable predictive risk. Heatmaps, time – series graphs and scatter plots are some of the comprehensive visualizations constructed to convey the discovery and help in decision making. The indicators to assess the model performance are R2, RMSE, MAE and the findings show how to forecast settlements in the most optimal manner. Besides presenting the utility of machine learning tunnel surveillance, the case study also provides data driven decision making framework in underground engineering projects.
• Developing physics-informed neural networks for structural parameters identification of beam with moving loadsAl-Adly, Anmar Ibrahim Fadhil; Kripakaran, Prakash; 10.3217/978-3-99161-057-1-045Physics-Informed Neural Networks (PINNs) seamlessly integrate the predictive capabilities of neural networks with established physical principles. By integrating constraints such as displacement and force boundary conditions alongside governing equations, PINNs can generate digital twins of physical systems and processes. This fusion allows for more accurate modelling and simulation of complex physical phenomena, bridging the gap between data-driven approaches and traditional physics-based methods. Nevertheless, the practical implementation of PINNs remains challenging, primarily due to numerous influential hyperparameters and the complex nature of modelling the governing physics through partial differential equations (PDEs). This challenge becomes especially critical in the context of dynamic loads, where higher-order PDEs encompassing both spatial and temporal domains, alongside relevant structural parameters and generalised (distributed) load’s function, must be carefully optimised during the PINNs training process. This study presents a novel application of PINNs model, developed, trained, and validated using real-world bridge monitoring data, for the inverse problem of predicting structural parameters of a girder subjected to moving loads. Two case studies are considered. In the first, PINNs model is utilised to estimate the structural parameters of a bridge girder under varying levels of noise in the data. In the second, the model is trained with actual field monitoring measurements to estimate structural parameters while predicting girder deflection and other internal forces. The findings advance the existing body of knowledge in structural health monitoring (SHM) by demonstrating a practical PINNs-based solution for bridge girders under moving loads.
• Structural Health Monitoring of a suspended steel infrastructure: A statistical approachMolon, Nicola; Casarin, Filippo; Targa, Alessandro; Codato, Renzo; da Porto, Francesca; 10.3217/978-3-99161-057-1-046The inspection, maintenance and monitoring of existing infrastructure are critical aspects for ensuring a proper structural performance during their lifespan, also guaranteeing their capacity vs. the ultimate limit state. The use of structural health monitoring systems has become increasingly important for managing infrastructural assets, not only to detect structural damages and degradation phenomena but also to evaluate the performance of structures subjected to retrofit interventions. This is achieved using signal processing techniques that integrate statistical methods and machine learning algorithms within the framework of statistical pattern recognition. The proposed framework introduces a novel statistical analysis framework aimed at characterizing the normal behaviour of structures, detecting potential damage development. The method is applied to a suspended steel truss healthcare facility, demonstrating its effectiveness in characterizing its typical structural behaviour, detecting any onset of possible structural decay. While the method is demonstrated on a specific case study, it is designed to be adaptable to a wide range of structural systems. The ultimate objective is to develop a reliable analysis tool for the early detection of damage, thereby enhancing the efficiency of maintenance strategies and ensuring long-term structural safety.
• Is it possible that AI can help us detect all damage in structural assets? A discussion on the scope of applicability of DL methods for diagnosis of the construction assets’s technical conditionTomaszkiewicz, Karolina; Owerko, Tomasz; 10.3217/978-3-99161-057-1-047The use of computer vision supported by artificial intelligence methods is growing in popularity for solving problems concerning the assessment of building and civil engineering structures. At the same time, SHM-class systems allow for the collection of large amounts of data. Despite the rapid development of machine learning and the increasing number of solutions supporting the process of technical condition diagnosis of objects, the amount of damage that can be detected in images using these algorithms is significantly limited. At the same time, due to the lack of publicly available datasets that can be used to train AI algorithms, the actual support of the civil engineer's work with these algorithms is limited to a few of the most common problems. This paper presents the current applicability of artificial intelligence methods for damage detection of buildings and engineering structures based on images. At the same time, the authors focus on showing the limitations for the development of artificial intelligence algorithms due to the lack of publicly available datasets. The paper identifies a research gap related to the lack of datasets for damage, pointing out the types of damage, types of damaged materials and solution classes not covered in research on the application of deep learning to the diagnosis of the technical condition of buildings and civil engineering structures.
• Perspectives on vision-based bridge vibrational monitoring by dronesPanigati, Tommaso; Giordano, Pier Francesco; Tonelli, Daniel; Limongelli, Maria Pina; Zonta, Daniele; 10.3217/978-3-99161-057-1-048Vision-based vibrational monitoring aims to extract the modal parameters of civil structures—such as natural frequencies—from recorded video data for Structural Health Monitoring (SHM) purposes. The use of drones for vision-based vibrational monitoring is particularly promising, as drones can access vantage points for video recording that may otherwise be difficult to reach. However, certain drawbacks exist, including potential limitations in resolution, stability, and environmental sensitivity. This paper explores the capabilities, opportunities, and limitations of using drones for vision-based vibrational monitoring. To evaluate technological limits, a target with controlled displacement is used to test various combinations of target distances, displacement amplitudes, and displacement frequencies. Additionally, factors such as environmental conditions and drone hardware are considered. The study defines the practical limits of this approach, aiming to determine the minimum displacement of a vibrating bridge that can be detected by drones. Case studies from the literature are used as benchmarks to identify the dynamic properties of different types of bridges.
• Universal unsupervised image segmentation model of multi-type component and damage for vision-based autonomous UAV inspection of bridgesYang, Guangshuo; Zhang, Chuao; Xu, Yang; 10.3217/978-3-99161-057-1-049Although recent advances have been widely gained in UAV-based visual inspection for bridges, the accuracy and generalization ability of recognition model highly rely on sufficient, complete, and high-quality annotations. Current damage segmentation models are often trained in a fragmented manner based on substantial pixel-level labels for specific structural components and damage types, lacking universality and robustness under real-world open scenarios. This study establishes a universal unsupervised image segmentation model of multi-type component and damage for vision-based autonomous UAV inspection of bridges using a teacher-student network architecture. The inputs are unlabeled image pairs after data augmentation including random clipping, rotation, illumination transformation, and color transformation. The pre-trained backbone of original DINO is adopted as frozen image feature extractor to obtain high-level feature representations, and a CNN-based segmentation head with learnable parameters is designed to generate dense segmentation maps with strong point-wise correlations. A synthetic loss function, comprising a correlation loss and a contrastive loss, is proposed for model training. The proposed method is validated on a unified multi-scale imageset including various structural components and surface damage for cable-supported bridges and concrete bridges. The recognition accuracy, generalization ability, and robustness under complex background are demonstrated.
• Large-Scale Structural Anomaly Detection During Seismic Events Using Optical Flow and Transfer Learning from Video DataWang, Sifan; Saida, Taisei; Nishio, Mayuko; 10.3217/978-3-99161-057-1-050Civil structures inevitably experience anomalies and damage, especially during disasters like earthquakes, tsunamis, and hurricanes, causing performance degradation or even collapse. Identifying such anomalies plays an extremely critical role in the maintenance and life extension of civil structures. This study proposes a novel approach based on video data due to its accessibility and rich temporal-spatial information for anomaly detection in large-scale civil structures by integrating transfer learning (TL) techniques with optical flow. Given the low importance of structural Region-of-Uninterest (RoU) like windows and doors, TL with BEIT+UPerNet pre-trained models identifies them. The extended node strength network then leverages video data to focus on structural components and detect disturbances in the nonlinearity vector field. The approach was validated using open video data from E-Defense, capturing two large-scale structural shaking-table tests that featured both pronounced shear cracks and tiny cracks. The detection and quantitative analysis results confirmed the method’s effectiveness in detecting structural anomalies and improved computational efficiency by approximately 10%, with a positive correlation observed between this efficiency gain and the proportion of structural RoUs in the video. This study advances anomaly detection in large-scale structures, offering a promising approach to enhancing safety and maintenance practices for critical infrastructure.
• Development of a Wireless Stereo Vision System for 3D Displacement Online Long-Term Monitoring of Tall StructuresWang, Miaomin; Koo, Ki-Young; 10.3217/978-3-99161-057-1-051This paper presents a novel wireless stereo vision system for 3D displacement monitoring of tall structures. The system uses two GNSS wireless camera nodes to capture images of a target and calculate its 2D displacement independently. Each node uploads its data to the cloud, where 3D displacement is reconstructed using a triangulation technique. This system has several advantages over traditional cable-based stereo vision systems for tall structures: 1) The distance between the two camera nodes is not limited, allowing flexible deployment to optimise measurement accuracy for structures of varying heights. 2) It avoids transmitting large volumes of image data between the cameras and the 3D image-processing computer, reducing the need for high network bandwidth. 3) It simplifies the stereo calibration process by eliminating the need for a checkerboard, which is often impractical to be positioned in the field of view of the cameras in tall structure applications. An outdoor test was conducted to validate the system, with the cameras placed about 200 m from the target and 100 m apart. The results showed a measurement accuracy of approximately 1 mm within the horizontal plane.
• Overview and Challenges of Computer Vision-Based Visual Inspection for the Assessment of Bridge DefectsKhan, Rizwan Ullah; Kromanis, Roland; 10.3217/978-3-99161-057-1-052Visual inspection remains the most fundamental and widely used method for assessing the condition of bridges. This process involves observation of structural surfaces at a close distance to identify visible signs of deterioration such as cracking, spalling, corrosion, and delamination. Traditionally, human inspectors perform visual inspections manually. This labour-intensive process is associated with many limitations, for example, subjectivity to an inspector’s interpretation, difficulty accessing structural components, management of large volumes of unstructured data and the lack of consistent historical records. Recent advancements in computer vision and artificial intelligence have enabled considerable progress toward automating visual inspections. However, the full automation of visual inspections in practical, real-world scenarios remains constrained by several challenges: (i) the continued need for human intervention, (ii) the limited availability of high-quality labelled datasets, (iii) the generalizability of existing models, and (vi) the lack of standardized inspection protocols. In this positioning paper, we present an overview of the current state of automated visual inspection for defects identification in bridges. It reviews key open-source datasets of defects and state-of-the-art deep learning models. We give our forward-looking perspective on fully automated defects identification systems that align with standardized visual inspection guidelines.
• Integrating Mixed Reality Technology, Deep Learning and Domain Knowledge for bridge inspectionLiu, Zhe Yu; Sun, Zhen; Vanova, Patricia; 10.3217/978-3-99161-057-1-053This study presents an integrated intelligent framework for bridge inspection that synergizes Mixed Reality (MR) technology, deep learning-based object detection, and domain-specific engineering knowledge. Utilizing Microsoft HoloLens 2 as the hardware platform, the system captures real-time 3D bridge surface imagery and deploys the optimized YOLOv11n-ZY model—enhanced with a ZZ convolutional module, YY attention mechanism, and SPPF-LSKA fusion module—to automatically detect and classify multi-category defects including cracks, corrosion, and spalling. Detection results are visualized within an MR interface and dynamically assessed through embedded expert knowledge. Validated on a custom dataset containing 4,176 images of 12 defect types under complex backgrounds, the proposed model achieves 40.3% mAP50 at 60 FPS with only 2.87 million parameters, outperforming existing YOLO variants. Implementation at the case study bridge demonstrates real-time defect localization, 3D model updating, and closed-loop maintenance functionality. The framework advances intelligent infrastructure management by establishing a scalable pipeline for accurate defect assessment and lifecycle-oriented bridge maintenance.
• 3D projection of AI-derived concrete cracks on a Hydro Dam outlet towerEmgård, Ludvig; 10.3217/978-3-99161-057-1-054In a Structural Health Monitoring (SHM) project, a hydro-dam's outlet tower in Northern Sweden required inspection and assessment. This was due to the lack of digital records regarding existing damage and the client's concerns about its progression. The tower's location in a lake made traditional inspection methods like scaffolding or skylifts extremely difficult. The concrete engineering expert hired for the evaluation was uncomfortable with rope access, making data collection an expensive and cumbersome task. The structure is significantly affected by fluctuating water levels in the lake, leading to suspicions of certain types of damage (e.g., freeze-thaw damage, erosion damage). For this project, a Phase One P3 camera was used to capture thousands of 100-megapixel images from every angle of the tower. Every inch of the concrete surface was covered by at least five different images. From these images, a 3D reconstruction was created using Spotscale's proprietary software pipeline1, which is specifically designed for high-resolution processing. Subsequently, each image was analyzed by Spotscale's machine learning algorithm to detect cracks, spalling, and visible rebar. For every pixel on the 3D model, all images that observed that pixel were analyzed, and this information was used to project the best possible representation of the damage onto the model, creating a 3D texture of the damage. The results revealed a distinct crack pattern, identified with a 98.7% confidence level when compared to human assessment of the same cracks. This provided the dam owner with a comprehensive understanding of the overall damage and an overview of the most severely affected areas.
• Computer vision-based recognition of random traffic flow for live load performance analysis of existing bridgesYu, Weilei; Nishio, Mayuko; 10.3217/978-3-99161-057-1-055The dynamics and complexity of stochastic traffic flows play a crucial role in the management of infrastructure such as bridges. This study presents a computer vision-based method for random traffic flow identification and load estimation that integrates the YOLOv8 object detection model and the DeepSORT multi-target tracking algorithm. By utilizing high-resolution bridge surveillance video, the method can accurately identify vehicle type, axle count, and traffic flow. A case study conducted on an actual bridge validates the effectiveness of the method. The results show that the accuracy of vehicle identification based on 24-hour video data is more than 93%, the statistical error is less than 10%, and the temporal distribution of traffic flow matches well with the actual situation. This study provides a new technical reference for AI-based bridge traffic management and low cost structural health monitoring solutions.
• A novel approach to bridge repair using photogrammetry and additive manufacturingTaha, Raguez; Ozevin, Didem; 10.3217/978-3-99161-057-1-056Traditional inspections of aged steel bridges rely primarily on visual assessment, often depending on qualitative analysis and requiring expert engineering judgement. Additionally, these methods are often labor-intensive, time-consuming, and costly, limiting their feasibility for widespread implementation. Accurate condition assessment, due to corrosion loss, remains a challenging factor in structural inspection, complicating the evaluation of its impact on bridge performance. This paper presents a novel workflow that integrates smartphone-based photogrammetry and metal additive manufacturing (AM) to improve condition assessment and enable data-informed repairs. High-resolution 3D models of a corroded decommissioned steel beam retrieved from a Chicago Transit Authority bridge were generated using a photogrammetry pipeline optimized for image quality and overlap. These models allowed for precise quantification of section loss and the design of custom-fit repair parts. A proof-of-concept repair component was fabricated using metal 3D printing and designed to restore the original geometry of a corroded flange section. While mechanical validation of the repair part is ongoing, this workflow demonstrates the potential for scalable, low-cost integration of digital imaging and AM in bridge maintenance.
• Machine Vision-Based Super-Resolution Reconstruction for High-Precision Displacement Monitoring of Hydraulic StructuresYang, You; Chen, Bo; Liu, Weiqi; Ma, Zekai; 10.3217/978-3-99161-057-1-057In response to the issues of high cost, limited monitoring accuracy, and susceptibility to environmental factors in traditional hydraulic structure displacement automation monitoring methods, a non-contact intelligent monitoring method based on machine vision image super-resolution reconstruction is proposed. This method uses artificial targets as markers and combines a high-order image degradation model with a camera to analyze real monitoring scenarios, carry out image data collection, and perform displacement calculation. It innovatively introduces a feature fusion attention mechanism to improve the Real-ESRGAN network and generator, enabling the reconstruction of image contours and fine details to enhance displacement calculation accuracy. Laboratory and field test results show that this method can effectively improve image resolution and clarity, achieving sub-pixel and millimeter-level precise monitoring of hydraulic structure surface displacement. Compared with traditional super-resolution algorithms and target tracking methods, the improved Real-ESRGAN algorithm performs the best, with a coefficient of determination (R²) of up to 0.9975, an average absolute error (MAE) as low as 0.5552, and residual errors controlled within 5mm. The edge contours and details in the images are successfully reconstructed, effectively improving the displacement monitoring accuracy of hydraulic structures based on machine vision.
• Monitoring Slow and Dynamic Deformations of High-Rise Buildings Using Low-Cost GNSS ReceiversJayamanne, Jayamanne Mudalige Oshadee; Psimoulis, Panagiotis; Owen, John; Penna, Nigel; Xue, Chenyu; 10.3217/978-3-99161-057-1-058The structural integrity and safety of high-rise buildings rely heavily on effective deformation monitoring. Global Navigation Satellite System (GNSS) techniques are frequently utilized to monitor these deformations; yet, despite their great accuracy, they possess possible limitations due to high costs. Thus, this research investigates the potential of low-cost GNSS receivers for monitoring deformations in high-rise structures. The study focuses on incorporating low-cost GNSS receivers to capture slow motion movements caused by factors such as solar radiation and temperature fluctuations as well as dynamic movements induced by forces including wind loads and seismic forces. The performance of low-cost GNSS receivers is assessed against high-precision geodetic-grade GNSS receivers through a series of experiments conducted on a high-rise building under both slow-motion and dynamic conditions. The study primarily investigates the U-blox F9P dual-frequency GNSS receiver with Leica AS10, Tallysman TWI, and U-blox patch antennas. Results indicate that low-cost GNSS receivers demonstrate significant potential for capturing accurate and precise deformation measurements. The selection of GNSS antenna is found to significantly influence the overall quality of the GNSS data. However, the results indicate that with proper configuration, these low-cost receivers can be successfully integrated to develop an efficient and sustainable deformation monitoring system for high-rise buildings.
• Potential of profile laser scanning (PLS) for the application in load testsSchill, Florian; Schacht, Gregor; Harke, Torsten; 10.3217/978-3-99161-057-1-059The transport infrastructure is reaching in many cases the end of its effective life cycle. The present condition is attributable to a combination of ageing and progressive deterioration as traffic volumes continue to increase. However, it is possible that even recently constructed bridges may already have significant structural damage. Consequently, the maintenance management of existing bridges is becoming increasingly important. However, there is often a lack of up-to-date information on the actual condition of the structures. This is because measuring infrastructure is merely the final stage of the monitoring process, given that tactile sensors are extremely time-consuming and labour-intensive to use. Additionally, due to these principles, measurements can only be taken at a few selected points. The combination of these limitations offers great potential for the use of non-contact profile laser scanning (PLS) in the context of load testing of bridge structures. The structure is scanned with a high-frequency laser beam without the necessity of entering the structure. The spatially distributed displacement measurements obtained in this manner provide a significantly higher density of spatial information about the structure than was previously feasible. Until now, dynamic investigations have been primarily conducted in the domain of profile scanning. This study primarily focuses on static load tests, where spatial resolution and measurement precision can be further enhanced. Two case studies are presented, illustrating non-contact PLS measurements for load testing: one example is on a 160 m arched railway bridge, and the other example is on a steel-concrete composite motorway bridge. It has been demonstrated that a precision of a few tenths of a millimetre can be attained with a spatial resolution in the centimetre range.
• Advancing Infrastructure Health Monitoring with Multi-Sensor Systems and Geospatial TechnologiesGeutebrueck, Ernst; Elsaid, Ahmed; Merten, Bastian; Glueck, Georg; 10.3217/978-3-99161-057-1-060This paper presents an integrated structural health monitoring (SHM) system combining multi-sensor networks with geospatial GNSS technologies to enhance infrastructure resilience. Developed by the authors, the system unites millimeter-accurate TEXtant® GNSS monitoring with MSS® leak detection and environmental parameter sensing, all synchronized via the TEX-Sky Monitoring Cloud. Validation efforts included comparative field testing against International GNSS Service (IGS) standards at GFZ Potsdam, demonstrating millimeter-level positioning precision. Real-world deployments—including the Aquitaine Bridge (France), slope monitoring for mining operations (Germany), tank basin and catch basin monitoring at a chemical facility (Germany), and the first station of the cross-border Asia Minor GNSS Network—confirmed the system's operational robustness. Time-synchronized multi-sensor datasets enable predictive maintenance, early anomaly detection, and asset life extension. The modular and scalable system architecture adapts flexibly to diverse infrastructure contexts. Future integration with machine learning technologies is anticipated to enhance pattern recognition and anomaly detection. This work highlights the role of synchronized, multi-parameter monitoring as a foundation for next-generation infrastructure management and public safety.
• LiDAR for vibration monitoring of infrastructure: stretching limits by spatio-temporal time domain frequency analysisGeißendörfer, Oliver L.; Holst, Christoph; 10.3217/978-3-99161-057-1-061Structural health monitoring (SHM) is crucial for ensuring the integrity and safety of infrastructure. Traditional vibration analysis techniques rely on sensors such as Inertial Measurement Units (IMU) and Global Navigation Satellite Systems (GNSS), total stations (TS) and fibre-optics (FO), which require a physical attachment to structures in the observation process. However, Light Detection and Ranging (LiDAR) offers a contactless alternative, enabling high-resolution, time-synchronized observations that capture spatially continuous deformation information. This paper presents an innovative framework for SHM that leverages LiDAR-based time-domain frequency analysis to monitor dynamic structural behavior effectively. By integrating spatio-temporal modeling techniques, we establish a robust methodology for detecting oscillations and deformations in infrastructure. Our approach enhances current SHM practices by providing a scalable solution that does not require physical sensor deployment. Thus, this methodology provides information in much higher spatial resolution compared to the aforementioned approaches. The proposed methodology is evaluated by controlled experiments, demonstrating its applicability to real-world SHM scenarios and its potential for continuous, non-invasive structural assessment.
• Application of LiDAR technology in geodetic monitoring of reclaimed landfillsPasternak, Grzegorz; Zaczek-Peplinska, Janina; 10.3217/978-3-99161-057-1-062Geodetic monitoring of reclaimed landfills is essential in ensuring the geotechnical safety of slopes and for monitoring the process of landfill settlement caused by biological and physico-chemical decomposition of the deposited waste. Insufficient recognition of the size and directions of displacements may lead to severe damage to the landfill body (landslides, sinkholes) and endanger the environment and the life and health of people living near the landfill. Classic geodetic monitoring of such facilities is based on measurements of single control points (benchmarks) located on the landfill body, the displacements of which often do not represent actual changes occurring in the area of the entire facility. The solution to this problem is to use Light Detection and Ranging (LiDAR) technology, which allows surface measurement of the entire studied area, making it possible to obtain a complete image of changes in the geometry of the landfill body. This paper presents a case study of a reclaimed municipal solid waste landfill located in Poland for which monitoring was applied using Terrestrial Laser Scanning (TLS) and Airborne Laser Scanning (ALS) from an Unmanned Aerial Vehicle (UAV). The acquired 3D data made it possible to obtain reliable information on the deformation processes on the landfill's surface and to decide on the direction of development of the post-remediation landfill as a Renewable Energy Sources (RES) station with solar panels and a biogas plant.
• Global Perspectives on Structural Monitoring in Civil EngineeringLehmann, Frank A.; Hille, Falk; Glišić, Branko; 10.3217/978-3-99161-057-1-063Structural Monitoring (SM) is crucial in civil engineering for ensuring the safety, functionality, and longevity of civil infrastructure, especially bridges. As its importance grows, SM practices are guided mainly by national standards, leading to fragmented approaches and limited global integration. This paper examines SM guidelines, focusing on contributions from Germany, while exploring the broader international framework. In Germany, key guidelines such as the DGZfP Merkblatt B09 and others offer structured methods and practice examples for long-term monitoring and performance assessment. Internationally, countries have developed their own SM frameworks. Amongst others, Austria’s RVS Richtlinie 13.03.01, France’s COFREND Livre Blanc, Canada’s ISIS Guidelines, the ACI Report 444.2-21 from the USA, the TRB Circular E-C246 and the CIRIA Guideline from the UK contribute to a global understanding of SM. These guidelines address common technical, theoretical, and economic challenges across regions. This paper highlights the need for international collaboration, identifying synergies and gaps to promote a unified approach to SM. It offers insights into global standards and how successful strategies can foster innovation and cohesion in SM practices worldwide.
• Structural health monitoring guidelines for bridges in GermanyHille, Falk; Wedel, Frederik; Lehmann, Frank; Pirskawetz, Stephan; 10.3217/978-3-99161-057-1-064With the advancement of digitalization and related technological developments, Structural Health Monitoring (SHM) has become a useful and increasingly widespread tool to assist in the maintenance management of bridges and other engineering structures. The process of implementing monitoring requires expertise in many fields such as civil engineering, bridge operation and maintenance, monitoring technology, and data analysis. In recent years, monitoring has moved from method and technology development to standard practice. However, the implementation of monitoring as a standardized process can be an obstacle, especially for bridge operators, due to a lack of practical experience combined with the various expertise required. This can affect several areas, such as determining the cost-effectiveness of a monitoring measure, proper tendering and contracting, quality control, analysis and evaluation of measurement data, and last but not least, data management. In order to support the introduction of monitoring technologies into the practice of infrastructure operators, several guidelines have been developed in Germany in recent years by different interest groups, each with a different focus and essentially complementing each other. This paper aims to provide an overview of four different recently published guidelines and to highlight their strengths and advantages.
• Structural Health Monitoring in the Italian Guidelines for BridgesCosta, Giancarlo; Morleo, Eleonora; Giordano, Pier Francesco; Limongelli, Maria Pina; 10.3217/978-3-99161-057-1-065
• ANYTWIN - Characterization and standardization of monitoring dataBasamad, Khadijah; Mosig, Oliver; Koca, Matthias; Radhakrishnan, Lakshanadevi; Walker, Maria; Marx, Steffen; 10.3217/978-3-99161-057-1-066Aging bridges were not designed for today’s higher traffic loads and often fail to meet current requirements. However, complete demolition or reconstruction is rarely feasible due to resource limitations, sustainability concerns and economic factors. A key issue lies in conservative assumptions regarding loads and resistance. Structural health monitoring (SHM) addresses this by providing real measurement data for a more accurate assessment. Monitoring produces large volumes of data that must be well-structured and stored for reliable assessments. This requires collaboration between civil engineers, measurement specialists, IT experts, and data analysts. As Building Information Modeling (BIM) adoption grows, standardized monitoring methods must ensure consistency and comply with the Single Source of Truth (SSoT) principle, enabling an integration of monitoring data in a BIM environment. The ANYTWIN research project aims to develop a framework for structured data storage and processing. It examines how measurement data relates to time and location, defines metadata and information for evaluation criteria and assigns responsibilities for data provision. A processing method ensures data preparation, analysis, and data mining, while quality indicators enhance reliability. These findings contribute to a tendering template, helping to structure monitoring tasks and improve maintenance strategies.
• Inspection as a basis for structural health monitoringGrubinger, Stefan; Burtscher, Stefan L.; Huber, Peter; Tutschku, Morris; Rebhan, Matthias; 10.3217/978-3-99161-057-1-067To effectively monitor engineering structures such as bridges, tunnels, and retaining walls, comprehensive knowledge of load-bearing behavior and load transmission mechanisms is essential. This knowledge allows for the assessment of its behavior and the identification of damage mechanisms. A thorough and clear documentation of regular inspection forms the foundation for this. It is important not only to capture damage patterns and visible defects but also to determine their origin and precise location. This information aids in the development of monitoring processes, the selection of measurement variables, and the implementation of monitoring technologies. The paper addresses the progress on how a documentation process of inspections is carried out to serve as a basis for a valid monitoring of such structures. Therefore, the use of digital models and a standardized description of damage in combination with a precise localization on the structure is described in the beginning. Such documentation provides valuable insights into load-bearing behavior and possible underlying damage mechanisms. Additionally, the development of suitable monitoring systems that serve as key parameters for structural inspection is demonstrated. The interaction and exchange of information between inspection and monitoring are emphasized, including clear visualizations and meaningful data overlays, to offer valuable benefits for building owners and inspection personnel.
• Dynamic monitoring using hidden markov regression model for predicting remaining useful lifeIke, Vincent Ifeanyi; Jesus, Andre; Shaheen, Mohamed; 10.3217/978-3-99161-057-1-068This paper presents the remaining useful life (RUL) prediction problem in civil engineering applications using a hidden Markov regression model (HMRM), as a promising approach for model-based degradation. Unlike self-transition hidden Markov models for mass-produced components, where prior lifetime signals are available to estimate state information, the proposed HMRM formulates the conditional probability of RUL in terms of the estimated regressor parameters, after temporally fitting the damage model. The discrete property of state in HMRM makes it possible to handle heterogeneities in the degradation process. The HMRM can also synthesise multiple signals by adopting a decision-level fusion. An adaptive closed-form solution for RUL prediction is presented. The performance of HMRM is demonstrated on synthetic measurements and compared with a Bayesian extended Kalman filter (EKF) updating technique.
• Identification of Damping Coefficients of Multi-degree of Freedom SystemMiah, Mohammad Shamim; Lienhart, Werner; 10.3217/978-3-99161-057-1-069Structural dynamical properties are vulnerable to the dynamic loads because such loads can change those parameters significantly. It is not possible to halt the aforementioned issue as dynamic loads are entirely unpredictable. The changes in stiffness, mass, and damping can lead to minor to serious damage scenarios depending on the level of changes of those parameters. Typically, the displacements trajectories of any systems are unknown, and if any other physical parameters e.g. damping is unknown that will form a nonlinear problem. Herein, to deal with the early mentioned problem a nonlinear observer namely the unscented Kalman filter (UKF) is employed. In conventional practice, the partial or full stiffness matrix are identified but identifying damping matrix is rare due to inherent complicacy. Hence, this study has focused on the identification of the entire damping matrix by adopting the UKF. The outcome of study shows that UKF is capable of identifying damping coefficients quite accurately. This outcome can play a vital role in the area of structural health monitoring and control applications.
• Integrated Motion Measurement – a Tool for Structural Health Monitoring?Kohl, Michael; Wagner; Friedrich, Jörg; 10.3217/978-3-99161-057-1-070Integrated Motion Measurement Systems (IMMSs) are multi-sensor systems, based on the principle of integrated navigation with inertial sensors as central components and an aiding by, e.g., GNSS receivers. IMMSs can be used to track elastic motions as additional degrees of freedom (DOFs) that capture the deformations of the object. To support the estimation of elastic properties, additional internal aiding measurements like strain gauges can be implemented. In addition to the raw sensor data, the elastic DOFs in the time and frequency domain are possible indicators to be used for Structural Health Monitoring (SHM). With the increasing availability of low-cost micro-electro-mechanical systems (MEMSs), combined with their ease of implementation, applications in large sensor quantities become feasible. To validate and experimentally test such an integrated motion measurement, a test rig with a movable, flexible pendulum beam was designed, to represent an idealization of a mast, rotor blade, or aircraft wing. In this study a short categorization of possible SHM applications for IMMSs is given, based on previous work and state-of-the-art SHM approaches. In this context, the principle of IMMS is explained with the experimental realization, validation, and the resulting modal characteristics of the elastic DOFs as potential indicators for SHM. Furthermore, the importance of strain gauges is investigated with methods to reduce their number by redundant sensors and restricted aiding.
• Solving structural dynamics with uncertainty quantification via evidential neural operatorsLi, Pei-Lin; Ni, Yi-Qing; Ling, Jian-Ming; Liu, Shi-Fu; Wang, You-Wu; 10.3217/978-3-99161-057-1-071Solving structural dynamic equations is crucial for evaluating the reliability and safety of civil infrastructures such as bridges, airport runways and railways under various loads. Currently, the Neural Operator (NO) shows great potential in solving structural dynamic equations under various excitations and boundary conditions without retraining and are capable of zero-shot learning. However, there has been a dearth of research into providing prediction errors and explicit uncertainty quantification of the operator-learned model for computing structural responses in different data regimes. This research aims to approximate the solution operator of structural dynamic equations with uncertainty quantification. Deep evidential learning is introduced to establish the Evidential Neural Operator (ENO) and the epistemic uncertainty of structural responses can be obtained. An illustrative example is given in this paper, which shows that the E-NO model can effectively identify the well-prediction condition and the worse-prediction condition. This work can provide an end-to-end framework for building surrogate models of real-world structures, which can rapidly compute structural responses with uncertainty.
• Nonparametric identification of structural nonlinear behavior based on extended Kalman particle filter and Chebyshev polynomial modelZhao, Ye; Xu, Bin; Yuan, Yikai; 10.3217/978-3-99161-057-1-072Describing the damage initiation and development of engineering structures during strong dynamic loadings such as earthquake is one of the most important topics in structural condition monitoring and identification. Structural nonlinear restoring force (NRF) can not only directly describe the initiation and development process of nonlinear behavior of the structure during strong dynamic loadings but also can be used to evaluate the energy dissipation of structural members or substructures. However, it is hard to measure structural dynamic responses at all degree of freedoms (DOFs) of a structure in practice, and to model the NRF with an accurate parametric mathematical model in advance due to the variability and individuality of structural materials and types. In this study, a Chebyshev polynomial model as a nonparametric model is employed to model the NRF of a structure and structural stiffness, damping, mass and NRF are identified based on the extended Kalman particle filter (EKPF) algorithm by using acceleration measurements at limited DOFs during the known external excitation. Then, two multi-degree-of-freedom (MDOF) numerical models equipped with different types of magnetorheological (MR) dampers are used as numerical examples to validate the performance of the proposed approach. Identified results show that the proposed method is effective for identifying the nonlinear MDOF structures with different nonlinearity with limited noise-polluted acceleration measurements.
• Insights into Rail Track Buckling from Distributed Fibre Optic Sensing DataHoult, Neil; Sun, Fuzheng; Butler, Liam; Zhang, Merrina; 10.3217/978-3-99161-057-1-073As climate change leads to increasing temperatures around the globe, rail track buckling has become an increasing concern for rail operators. This paper provides an overview of the key outcomes from a four-year research program that sought to explore the use of distributed fibre optic sensors (DFOS), analytical modeling, and artificial intelligence techniques to aid in track buckling assessment and detection. Lab tests and field monitoring data were used to develop and evaluate two DFOS systems, one for short length dynamic buckling assessment due to train passage and the other for long length thermal buckling assessment. The data from each system was used to develop models for the detection of buckling using different techniques depending on the quality of the initial data and the required output. Finite element model (FEM) updating and statistical FEMs were explored to predict buckling response based on measurements at service loads. Beam on elastic spring models were used to estimate the influence of train passage on buckling capacity while Gaussian process regression (GPR) techniques provided insights into buckling indicators at the field sites.
• Distributed Acoustic Sensing for Civil and Geotechnical Infrastructure Monitoring ApplicationsZhang, Cheng-Cheng; Shi, Bin; Xie, Tao; Zhang, Taiyin; Chen, Zhuo; Wang, Zheng; Xu, Qi-Yu; 10.3217/978-3-99161-057-1-074Distributed acoustic sensing (DAS) has emerged as a powerful technology for monitoring the health and integrity of civil and geotechnical infrastructure. This technology leverages existing fiber-optic cables as dense arrays of vibration sensors, enabling continuous, real-time monitoring over long distances. This abstract summarizes recent research advancements in applying DAS to various infrastructure monitoring challenges. We first provide a brief overview of the DAS technology and its working principles. Subsequently, we present several case studies demonstrating the versatility of DAS. These include: (1) monitoring and identifying geohazards, such as landslides and rockfalls, that threaten the stability of linear infrastructure; (2) detecting disturbance events, including drilling and excavation activities, near a high-speed railway tunnel; (3) identifying wire breaks in prestressed concrete cylinder pipes for early warning of potential failures; (4) measuring flow rates and detecting illicit flows in urban underground pipelines for improved water management; and (5) integrating DAS with deep learning for traffic monitoring, providing insights into traffic dynamics and patterns, particularly during the COVID-19 pandemic. These examples highlight the potential of DAS as a cost-effective and comprehensive solution for enhancing the safety, resilience, and operational efficiency of critical infrastructure.
• DFOS solutions covering full monitoring needs of an enlarged concrete deck viaductVillar, Miguel; Ferrario, M.; Morosi, J.; 10.3217/978-3-99161-057-1-07560s economic boom led to spread construction of large transport infrastructures. Many of these steel reinforced concrete structures attain their end of lifespans on this and next decade. With no major renewing plan, repairing and retrofitting are explored alternatives. A good example is Milano's ring-road viaduct; while already repaired and its concrete deck enlarged, SHM begins nowadays. Monitoring of thermal and mechanical induced strain, static and dynamic, brings access to permanent strain, thermal expansion, eigenmodes of each single road span and better understanding of the whole structure dynamic behavior. Three trucks moving at 30 km/h load dynamically the enlarged deck, while real traffic is used for modal analysis. Often, this kind of comprehensive monitoring requires combining various measurement technologies, making their installation time-consuming and expensive. Thus, the number of sensors may be undercut, and measurement campaigns duration reduced, which may result in poorer monitoring results and mismatching between experimental results and model's ones. FEBUS SHM solutions based on DAS (Distributed Acoustic Sensing), DSS (Distributed Strain Sensing) and DTS (Distributed Temperature Sensing) provide quick instrumentation and easy monitoring. With long-range devices to address tens of km of infrastructure instrumented in a row, up to 400 kHz continuous monitoring, state-of-the-art DAS repeatability threshold of only 2 picoStrain/SquareRoot(Freq), FEBUS DFOS (Distributed Fiber Optics Sensing) solutions brings values for every node of the structure, remote monitoring and mastered opex and capex.
• Experimental study on two tunnel micro-leakage monitoring methods based on distributed fiber optic sensing technologyGuo, Junyi; Shi, Bin; Fang, Jinhui; Jiang, Hongtao; Sun, Menya; 10.3217/978-3-99161-057-1-076Real-time monitoring and accurate localization of pipeline leaks are crucial for pipeline safety. Conventional methods often fail to detect micro-leaks and face issues such as low positioning accuracy, high false alarm rates, and inability to monitor long distances. This study proposes two innovative leakage monitoring methods based on distributed temperature sensing (DTS) and distributed strain sensing (DSS). The first method utilizes evaporation-induced relative humidity measurement. By measuring the temperature of DTS fiber optic cables wrapped in gauze, real-time detection and localization of leaks in pipelines carrying ambient-temperature liquids can be achieved. A significant temperature drop at the leakage location allows precise identification of micro-leaks. The second method employs fiber optic cables with water-swelling blocking yarns for localized strain sensing, combined with Optical Frequency Domain Reflectometry (OFDR). At the leakage location, the strain of the sensing fiber significantly decreases, providing a distinct signal for detection. Indoor experiments confirmed the feasibility of both methods, demonstrating their ability to achieve real-time monitoring and precise localization of micro-leaks. These methods offer novel solutions for addressing pipeline leakage challenges.
• A PC based FE model as an innovative learning tool in structural mechanicsMiyamori, Yasunori; Komuro, Kakeru; Suto, Soushi; Kadota, Takanori; Saito, Takehiko; 10.3217/978-3-99161-057-1-077Recent developments in digital technology make it possible to create interactive 3D models for structural analysis in a user-friendly software environment. In addition, the technology for creating 3D models of real structures using point cloud data for various users is also developing rapidly. Such rapid change may raise concerns about a developing gap between traditional structural mechanics education based on beam theory and the modern structural analysis scheme. However, if it becomes possible to efficiently perform structural analysis by recreating structures in cyberspace without special skills, it will be beneficial to cultivate a sense of structural mechanics for beginners. In this study, we investigated the feasibility and simplicity of constructing a point cloud-based structural analysis model to capture the general trend of deformation and stress in a structure. Specifically, we constructed a point cloud model from photographs of an approximately 2m long steel plate girder bridge using the SfM technique, meshing the model, and assembled a solid FE model using only general functions of packaged software without any special operations for model modification. We compared the strain in the constructed model under static loading with the actual measured values for the bridge. The model properly calculated the deflection shape, and it will help learners understand structural mechanics.
• RTK-Enabled UAV for Structural Health Monitoring Without GCPsAnn, Hojune; Yu, Yong-Rae; Kang, Gi-Sang; Oh, Juheum; Lee, Jong-Jae; Koo, Ki-Young; 10.3217/978-3-99161-057-1-078This study presents a structural shape monitoring system combining Real-Time Kinematic (RTK) technology with an Unmanned Aerial Vehicle (UAV) for georeferencing without Ground Control Points (GCPs). Traditional GCP-based methods, though accurate, require substantial field efforts, limiting efficiency. This approach employs RTK-enabled UAVs for direct georeferencing, achieving sub-2 cm positioning accuracy. Experiments involved detecting shape changes using a target structure with attached brick shapes of varying depths (8 mm, 20 mm, 44 mm, and 84 mm). Successful detection was achieved for depths of 20 mm or greater, with limitations for smaller depths due to sensor and resolution constraints. Depth and volume estimation errors, initially 11% and 3%, were reduced to 6% and 1% through point cloud registration, improving alignment and geometric accuracy. The study also identified challenges like occlusions and patternless surfaces, which impacted reconstruction quality. These findings highlight the system’s potential to enhance structural health monitoring, offering an efficient, scalable solution for infrastructure inspections, with applications in civil engineering and beyond.
• Advanced and Efficient Monorail Facility Inspections Using Optical Measurement Technologies, Including Laser and ImagingNakamura, Motoki; Kurashige, Hirotoshi; Yamazaki, Hiroshi; Hayashi, Ryohei; Inoue, Kousuke; 10.3217/978-3-99161-057-1-079To support safe operation of monorail systems, periodic track inspections for line maintenance and structural inspections of the concrete girders and steel beamway that form the track are necessary, and efficient inspections and highly accurate inspection results are required. Since conventional inspections centered on manual measurement and visual inspection, there were issues in terms of the safety risk associated with work in high places and work efficiency. Furthermore, in visual inspections, it was not possible to gain an adequate understanding of deterioration over time due to the difficulty of quantitative evaluation. This paper presents an example of a non-contact inspection technique in which industrial area cameras, laser displacement sensor and a high-speed 2D laser profiler were installed on the inspection vehicle, and track displacement, visible deterioration trolley wire wear are measured while the vehicle is traveling. Although inspection work at the work site had required half a year with the conventional technique, the introduction of this technique shortened the inspection time to only 4 days, realizing improved efficiency and reducing the man-power required in site work. Future goals include sustainable infrastructure maintenance and improvement of track safety through efforts in trend analysis of the progress of deterioration based on an expanded range of inspection items and analysis of various types of accumulated data.
• Application Method of SfM/MVS Technique Combined with Point Cloud Data for Inspection of Steel BridgesYamashita, Ko; Kato, Jun; 10.3217/978-3-99161-057-1-080The Structure from Motion/Multi-View Stereo (SfM/MVS) technique, which enables reconstruction of the three-dimensional shape of an object based on multi-view images, is a useful technique for inspection of infrastructure such as bridges in order to gain a comprehensive understanding of the state of damage. However, in SfM/MVS, the camera position and three-dimensional coordinates of the object are estimated based on feature points in images, making it difficult to apply this technique to steel structures, which have fewer feature points on surface textures due to coating. This paper presents a case in which SfM/MVS was successfully applied to a steel structure, specifically a steel bridge, by substituting the incomplete polygon model constructed in the conventional SfM/MVS process with a current-state CIM model created from existing drawings and point cloud data acquired with a terrestrial 3D laser scanner.
• Re-meshing Method for Finite Element Model Updating based on Extracting Structural Anomalous Information from Point Cloud DataWang, Jiexiu; Nishio, Mayuko; 10.3217/978-3-99161-057-1-081Finite element analysis (FEA) is widely used to evaluate civil structures’ performance. To consider detected structural anomalies due to damage in FEA, it is required to represent the anomalous areas in the original finite element (FE) model and update the mesh configuration. This study proposes an approach for updating the shell-element FE models of thin-walled structures with anomalous areas by the point cloud data (PCD)-based CV method, focusing on surface planar anomalies. In this approach, the Iterative Closest Point (ICP) algorithm was used for the alignment of the point cloud with the FE model. The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) on HSV values of PCD was used to extract points of anomalies. The anomalous region is denoted as its boundary points detected by the Alpha-shape algorithm. Constrained Delaunay Triangulation (CDT) generates the new meshes over a constrained area based on points related to anomalies. An experimental study was conducted for validation using a steel plate structure with various stickers attached to simulate the anomalies. The proposed CV-based FE model updating method was validated by comparing the PCD-based updated model and the manually updated model in terms of geometric and analytical accuracy. Most of the corresponding anomalous regions in the two models show a high degree of consistency, except for some areas affected by the low quality of the PCD, which, however, do not have a significant impact on the FEA results. With the same thickness reduction of anomalies, the analysis results indicate that there is only a minimal error between the two models. The proposed method is feasible as a substitute for manual rebuilding, facilitating the automation of the FE model updating with anomalies.
• Synthetic environment for close-range photogrammetry-based surface friction assessment of road infrastructuresPeng, Cheng; Jiang, Yi; Li, Shuo; 10.3217/978-3-99161-057-1-082Quantification of friction demands is a major task in management of road infrastructures. The use of pavement texture measurement in friction assessment offers the potential for describing road frictional characteristic in a non-contact manner. However, surface macrotexture profiles tested by stationary measurements and high-speed laser systems provide limited range of texture information at high frequency scales. To achieve adequate outdoor road surface reconstruction at ultra-high resolution and low cost, this research develops a synthetic environment for ground truth reference and efficient generation of data and experiment. To illustrate the approach, a photo-realistic computer graphics model of asphalt pavement surface is produced and virtually scanned using candidate image acquisition plans. Then, in-depth quality assessment of the corresponding 3D point cloud reconstruction models is performed. In this way, suggest use of a close-range photogrammetric pavement surface scan method using Structure-from-Motion (SfM) technology and its requirements for friction-oriented texture quantification in terms of spatial resolution, camera movement, and illumination configuration is put forth. The effectiveness of the synthetic environment and the optimized experiment setup is demonstrated through a field survey on three roads. Finally, the obtained point cloud datasets are used in texture feature characterization and friction number prediction modeling processes.
• Developing a Deep Learning-Based Method to Segment Bridge Members by using 2D Cross Sectional Point CloudsHidaka, Nao; Hashimoto, Naofumi; Watanabe, Ei; Uchiyama, Daisuke; 10.3217/978-3-99161-057-1-083For efficient maintenance and repair, 3D modeling of bridges for converting analytical models and visualizing deformation locations is being advanced. Instead of manually creating models from drawings and ledgers, automating model generation from point cloud data, capable of capturing as-is geometry quickly and widely, can improve efficiency. Generating 3D modeling of bridges from point cloud data requires segmenting each member, but dynamically setting thresholds for shape features and positional relationships is challenging due to point density and missing points. While deep learning can dynamically set thresholds, in the case of point cloud data, it is impractical to prepare sufficient training data, and the number of inputting points is inadequate for setting appropriate thresholds. Therefore, this research focuses on two aspects: most bridges consist of members with swept cross sections along longitudinal direction, and deep learning classification methods for 2D images are highly developed. The aim is to segment members based on deep learning on 2D cross-sectional point cloud data obtained by slicing along longitudinal direction. This reduces the number of inputting points and increases training data. Additionally, fine cross sections enable segmentation close to 3D. The multiple patterns of learning methods, training data processing, and procedures of segmentations are compared to identifying highly accurate segmentation methods.
• A 3D Virtual Assembly Method for Cable-Stayed Bridge Closure Using Laser ScanningXu, Yan; Cao, Zhenzhen; 10.3217/978-3-99161-057-1-084The mid-span closure segment assembly of a cable-stayed bridge after cantilever construction traditionally relies on using a total station to measure the longitudinal distances of key control points to determine the closure trimming length. This approach neglects the rotation of the cantilever section, reduces the inherently three-dimensional (3D) assembly problem to a one-dimensional (1D) longitudinal assembly. To overcome this limitation, this paper presents a 3D virtual pre-assembly method for the closure segment based on point clouds captured from a single station. An efficient and accurate two-stage registration method based on 2D image matching and 3D visibility simulation was developed to align incomplete measured point clouds with the design model. The estimated poses of the two cantilever ends were used for the virtual assembly of the closure segment. An optimization model for 5-DOF geometric information of the closure segment was established and calibrated using a particle swarm optimization. The proposed method was validated during the construction monitoring of a large-span railway cable-stayed bridge, demonstrating its reliability and practical effectiveness.
• Short- and long-term monitoring of bridges using terrestrial laser scanning dataMoser, Thomas; Lienhart, Werner; 10.3217/978-3-99161-057-1-085Terrestrial laser scanning (TLS) is commonly used to capture 3D point cloud data of the environment. In this article we demonstrate that TLS data can also be used to measure long-term and daily deformations of bridges. On one hand full dome laser scanning is used to determine the deformations of entire bridge pillars whereas scanning total stations are well suited to capture segments and profiles of bridge pillars. We highlight that an accurate point cloud registration and appropriate processing algorithms are crucial to reliably determine deformations in the millimeter range. The capabilities of our approach are demonstrated on two large highway bridges where the bending of bridge beams due to temperature changes and one side sun illumination are investigated.
• External Magnetization based Elasto-Magnetic Sensing Technique for Tension Monitoring of Aged PSC StructuresKim, Junkyeong; 10.3217/978-3-99161-057-1-086This study proposes a novel non-contact electromagnetic (EM) sensing system tailored to detect magnetic flux density changes associated with prestressed tendons embedded in concrete structures. The research emphasizes the optimization of sensor head geometry and coil configurations, including single, multi-solenoid, and Halbach array arrangements, to enhance external magnetic field detection at distances representative of real structural applications. Analytical formulations based on closed-form magnetic field equations were validated through finite element analysis (FEA) using ANSYS Maxwell. Results confirm that concentric Halbach-arrayed multi-solenoids outperform conventional configurations in delivering high-density magnetic fields beyond structural surfaces, particularly at target distances up to 30 cm. The verified modeling framework supports further development toward practical integration into structural health monitoring (SHM) systems.
• Development of SFCW Radar System for Concrete Structure InspectionLee, Sangho; Cho, Keunhee; Choi, Ji-Young; Lee, Joo-Hyung; Kwahk, Imjong; Joh, Changbin; 10.3217/978-3-99161-057-1-087Ground Penetrating Radar (GPR) is commonly used for internal inspection of concrete structures. However, the fixed design parameters of commercial GPR limit adaptability for specific inspection conditions and integration with emerging technologies. This study presents the development of a Stepped-Frequency Continuous Wave (SFCW) radar for concrete structure assessment, based on numerical simulations and experimental validation. Parametric analysis was performed to evaluate the influence of frequency bandwidth, antenna spacing, synthetic aperture length, and beamwidth on imaging performance, particularly in detecting embedded reinforcing bars. B-scan data were generated and processed using Delay and Sum Algorithm (DSA) for an image focusing, and image resolution was evaluated in both azimuth and range directions. Based on the simulation results, an SFCW radar prototype was built, and its performance was assessed through tests on reinforced concrete specimens. The experimental results confirmed the system’s capability to detect internal targets. The findings suggest that the proposed radar system offers improved flexibility and adaptability for concrete inspection compared to conventional commercial GPR.
• A Novel System Identification-Based Method for Rebar Radius Estimation in Radar SAR-Based Non-Destructive TestingPark, Kwang-Yeun; Lee, Joo-Hyung; Joh, Changbin; 10.3217/978-3-99161-057-1-088Non-destructive testing of reinforced concrete commonly utilizes electromagnetic waves, such as radar, to obtain internal structural information. When probing around rebars using electromagnetic equipment, hyperbolic-shaped images are often generated. Typically, image focusing techniques, including Hyperbolic Summation, Kirchoff Migration, Phase-shift Migration, Omega-k Migration, and Back-projection-based Focusing, which are based on Synthetic Aperture Radar (SAR) algorithms, are applied to analyze these hyperbolic images. However, these conventional methods cannot accurately determine the size of rebars and face limitations when inspecting doubly reinforced concrete due to shadow regions created by surface-layer rebars, which obscure the internal rebars. To address these challenges, this study proposes a novel approach that analyzes hyperbolic images based not on the image itself, but on the information related to wave propagation distances. In this method, the rebar cross-section is assumed to be a circle with an arbitrary radius, and a hyperbolic equation is established accordingly. The radius is determined by solving the equation using a system identification (SI)-based approach that minimizes the error between the measured hyperbola and the theoretical one. As with many conventional SI techniques, this problem is highly ill-posed, requiring the introduction of regularization methods to stabilize the solution.
• Non-contact non-destructive monitoring of concrete structures using pulsed Laser and microphonesMajhi, Subhra; Mukherjee, Abhijit; Sane, Nihar; Sharma, Siddhant; 10.3217/978-3-99161-057-1-089Concrete structures are at several stages of deterioration across the world. The presence of chloride ions salts as in the case of marine infrastructure or due to the application of de-icing salts, can further aggravate the rate of deterioration. Inspections of large concrete structures are predominantly undertaken through visual inspections. Detailed inspections are undertaken using piezo-generated ultrasonics. These inspections can be time and resource-intensive as the piezo devices need to contact the structure during measurements and their energy outputs are limited. Rapid inspections of large civil engineering structures would require a non-contact, high-energy source means of measurement. In our approach, we used a high-energy pulsed laser for excitation and an acoustic microphone for reception towards monitoring concrete structures. Defects in concrete like debonding and honeycombing, were simulated in the concrete specimens. The pulsed laser was used to excite the concrete specimen and the resultant waves generated due to this excitation were measured using focused cardioid microphones. The characteristic features in a typical waveform were first identified in pristine specimen. Subsequently, features corresponding to defects are extracted from the acquired signals using the signals from the pristine signal as a reference. The variations in these features were localised and their veracity was associated with the embedded defects in the specimen. As a result, the location and the nature of the defect were inferred. Thus, through this work, a framework for using pulsed lasers and microphones for non-contact non-destructive detection of defects in concrete is demonstrated.
• AI-Powered vehicle classification for scalable infrastructure monitoringIacussi, Leonardo; Giulietti, Nicola; Lucci, Alessandro; Lucenti, Giuseppe; Zappa, Eamanuele; Chiariotti, Paolo; Cigada, Alfredo; 10.3217/978-3-99161-057-1-090Real-time monitoring of road infrastructure is crucial in addressing the challenges posed by the increasing volume of vehicles and the need for timely maintenance to manage structural aging. Traditional Weight-in-Motion (WIM) systems provide accurate measurements of vehicle load, axle configurations, and speed but are costly to install and require road closures, hindering widespread deployment. This study introduces an innovative method for estimating traffic load by repurposing acceleration-based Structural Health Monitoring (SHM) systems integrated with an AI powered vision system which enables to classify vehicles, estimate their weight, speed and finally assess traffic load over time with a scalable and cheaper solution. Vehicles have been classified into three macro classes: cars, lightweight trucks and heavy trucks. A comparative analysis has been performed between load estimation using only the AI-powered vision system, based on YOLO object detection, and an enhanced approach that integrates acceleration data. The combined method demonstrated significantly improved accuracy in weight estimation. The methodology was tested on an highway viaduct and the results validated by using a reference WIM system. The findings underscore the potential of this integrated approach to provide cost-effective and scalable solutions for traffic load estimation and structural health assessment.
• Structural condition monitoring through information transferring with dimensional expansionLee, Jaebeom; Lee, Seungjun; Yoon, Dong-Jin; 10.3217/978-3-99161-057-1-091With the rapid development of sensor technologies and computational methodologies, real-time structural health monitoring (SHM) has gained significant attention in the field of civil engineering. Infrastructures, such as long-span bridges and dams, are often equipped with diverse sensor arrays to enable continuous monitoring of their structural conditions. However, conventional SHM typically require extended data collection periods post-sensor installation, which can delay their practical applications. To address this challenge, this study introduces a novel methodology termed information transferring with dimensional expansion, which leverages transfer learning principles to enhance anomaly detection capabilities in newly instrumented structures. By referencing datasets from similar existing infrastructures, this approach mitigates the dependency on extensive initial data while ensuring reliable anomaly detection. Validation through a case study on a long-span bridge in Republic of Kore demonstrates the method’s efficiency and accuracy, highlighting its potential to revolutionize SHM practices by enabling immediate operationalization upon sensor deployment. This research contributes to advancing SHM systems, emphasizing scalability and adaptability for diverse structural applications.
• Unsupervised Anomaly Detection for Structural Health Monitoring: A Vibration-Based Approach Using Isolation ForestSoltani, Emad; Gueniat, Florimond; Salami, Mohamed Reza; 10.3217/978-3-99161-057-1-092Structural health monitoring (SHM) plays a crucial role in ensuring the safety and longevity of critical infrastructure, such as bridges. SHM refers to continuous, sensor-based, and automated monitoring that complements traditional inspection methods by providing real-time data on structural performance. This paper proposes an unsupervised machine learning approach to SHM using vibration data, aiming to address the challenges of data scarcity and the difficulty of collecting labelled damage examples. The methodology combines statistical and spectral feature extraction with an Isolation Forest anomaly detection model, trained solely on healthy data to identify potential damage. The feature extraction process includes key metrics such as root mean square, entropy, and spectral centroid, which capture both time-domain and frequency-domain characteristics of the vibration signals. The Isolation Forest model is trained on these features to distinguish between normal and anomalous patterns, making it well-suited for applications where labelled damage data is unavailable. Results from FE simulation show high accuracy (95.5%), precision (91.75%), and recall (100%), demonstrating the effectiveness of the method in distinguishing damage from healthy states. The proposed approach provides a scalable and data-efficient solution for real-time damage detection in civil infrastructure, with significant potential for deployment in large-scale monitoring systems. Future work will focus on experimental validation and improving the model’s robustness in real-world conditions.
• Deep generative models to mitigate data scarcity in bridge structural health monitoringFarhadi, Sasan; Corrado, Mauro; Acquesta Nunea, Danilo; Ventura, Giulio; 10.3217/978-3-99161-057-1-093Structural health monitoring is essential for ensuring the safety, reliability, and longevity of infrastructural assets. However, conventional monitoring measurements face significant challenges, such as being labor-intensive, costly, and time-consuming. In recent years, the rise of machine learning and deep learning has data analysis frameworks, offering a promising solution to these challenges. Despite this, developing reliable and robust approaches that generalize well to unseen scenarios often requires large amounts of training data. This presents a challenge, mainly with regulatory constraints and difficulties in collecting data, particularly for rare events. To address the issue of data scarcity, this study proposes a generative data augmentation approach using a Wasserstein Generative Adversarial Network (WGAN). This approach generates high-quality short-time Fourier transform (STFT) spectrograms, which are transformed into image-like data, from in-situ acceleration signals for model training. The collected signals, recorded from real-world bridges during various events such as hammering, drilling, environmental noise, and, most importantly, the rare event of wire breakage in prestressed concrete bridges, are processed and fed into the WGAN model to synthesize additional data. This improves the diversity and robustness of training datasets. Evaluation of the generated spectrograms using various performance metrics, such as Structural Similarity Index Measure, Peak Signal-to-Noise Ratio, and Fréchet Inception Distance, demonstrates that the proposed method offers a scalable and cost-effective solution for enhancing the training dataset, particularly in scenarios where event data is sparse, such as prestressing wire breakage.
• Smart adaptive triggering strategy for edge intelligence enabled energy-efficient sensingCui, Shuaiwen; Yu, Xiao; Fu, Yuguang; 10.3217/978-3-99161-057-1-094Achieving both energy efficiency and high triggering accuracy is a critical multi-objective optimization challenge in Structural Health Monitoring (SHM), particularly for power-constrained wireless edge devices deployed in dynamic environments. Traditional empirical and static-threshold-based methods often struggle to simultaneously have low miss trigger and false trigger rate and lack adaptability to varying environmental and operational conditions. This study proposes a multi-stage adaptive triggering strategy built upon a Feedback Control (FC) framework, driven by Bayesian Optimization (BO) as the optimization engine, and accelerated by Digital Twin (DT) for data augmentation and Neural Networks (NN) for real-time contextual understanding and robust inference. The strategy dynamically refines triggering thresholds based on simulated insights and partial observations, enabling rapid adaptation and generalization across scenarios. Experimental validation through simulations and onboard deployments demonstrates that the proposed method improves F-beta performance by over 30% compared to conventional empirical methods. This approach provides a promising pathway toward intelligent, energy-efficient, and sustainable SHM sensing through fast feedback, reduced deployment cost, and minimized missed critical events.
• Post-tensioned wire breaks detection method using distributed acoustic sensing in bridges & viaducts.Lakshmanan, Dinesh; Muñoz, Felipe; Urricelqui, Javier; Jimenez-Rodriguez, Marco; 10.3217/978-3-99161-057-1-095Monitoring the structural integrity of civil infrastructures, such as bridges and viaducts is crucial, as non-visible damage like post-tensioned wire breaks can lead to catastrophic failures, endangering public safety. In this study, we simulate post-tensioned wire breaks by generating controlled mechanical impacts using a sclerometer. These impacts are applied at varying distances from optical fibers cables attached to the tendons of two different structures. A novel detection framework is developed using distributed acoustic sensing (DAS) technology to identify post-tensioned wire breaks in a suspension and a concrete bridge while effectively distinguishing between vehicular noise, environmental noise, and actual wire break events. For suspension bridges, a spectrogram-based template matching approach is implemented, leveraging sub-band selection and image-based analysis to enhance sensitivity to break events while suppressing false positives from environmental noise. In concrete bridges, a deep learning-based convolutional neural network (CNN) model achieves 96% classification accuracy, outperforming traditional methods in detecting wire breaks with high precision. These approaches provide a real-time, reliable solution for structural health monitoring, offering significant advancements in distinguishing critical break events from background interference, improving bridge safety and maintenance strategies.
• Detection of steel fractures in existing prestressed bridges with DFOSBurger, Harald; Schramm, Nicholas; Fischer, Oliver; 10.3217/978-3-99161-057-1-096Prestressed concrete bridges are designed to limit cracking. Aging can lead to prestressing steel fractures with strain changes without visible cracks. To detect and locate fractures, long-term monitoring with distributed fiber optic sensors in combination with acoustic emission sensing is useful. This study focuses on fiber optic sensing based on Rayleigh backscattering. A long-term monitoring system was installed to ensure the service of two prestressed concrete bridges in Munich. Their prestressing steel is at risk of stress corrosion cracking. For reliable operation of the fiber optical measurements, tests were carried out both in the lab and in the field to investigate the measurement signal in conditions that are as isolated as possible. On the bridges data is collected at three-month intervals starting in 2021. Furthermore, strain changes caused by temperature fluctuations and traffic loads were captured. Moreover, valuable insights are being gained in a long-term operation. The investigations show that long-term fiber optical strain measurements are useful to assess the structural behaviour of existing prestressed bridges over several years and can ensure safety of structures.
• Distributed fiber optic sensing of bridges with stress corrosion crackingGoldyn, Michal; Herbers, Max; Richter, Bertram; Zdanowicz, Katarzyna; Marx, Steffen; 10.3217/978-3-99161-057-1-097Stress corrosion cracking (SCC) of prestressing steel represents a critical threat to the long-term safety and serviceability of aging bridge infrastructure. This phenomenon occurs within the cross-section and leads to the initiation and propagation of cracks, ultimately causing the rupture of the prestressing wires, which may ultimately result in sudden bridge failure. This underscores the need for reliable monitoring solutions. Traditional non-destructive testing techniques, while valuable, often lack high-resolution capabilities. In contrast, Distributed Fiber Optic Sensing (DFOS) has emerged as a transformative approach, offering high-resolution, continuous monitoring of strain distribution and crack development in concrete structures. This study demonstrates the practical application of DFOS technology for detecting and quantifying crack propagation in operational bridge structures affected by the risk of active SCC. By considering field investigations on four infrastructure projects the research evaluates DFOS performance for structures vulnerable to SCC. The paper demonstrates the technology’s capability to monitor crack dynamics under operational conditions as anomalies in the crack pattern may indicate early symptoms of structural damage caused by SCC. By bridging knowledge gaps in the application of DFOS for infrastructure safety, the study advances the role of fiber optic sensing in addressing SCC challenges, ultimately contributing to the development of more resilient and sustainable bridge monitoring systems.
• Lifetime elongation of existing prestressed bridges with a lack of structural integrity using DFOSNeumann, Sören; Burger, Harald; Lamatsch, Sebastian; Fischer, Oliver; 10.3217/978-3-99161-057-1-098Existing prestressed concrete bridges without minimum shear or flexural reinforcement are at increased risk of sudden failure, even if they appear undamaged. To prevent such a sudden failure surface mounted Distributed Fiber Optical Sensors (DFOS) can detect local strain changes prior to cracking and visible damage. Failure mechanism analysis, laboratory tests and numerical simulations are used to identify relevant strain indicators. These strain changes could be detected by DFOS even if the failure location is not known exactly in advance. From these results, limits for a universal monitoring concept could be derived considering the individual failure mechanisms and the limitations of the sensor system.
• Concrete signature in long-term Distributed Fiber Optic Strain Sensing: Challenges and opportunities for Structural Health MonitoringUlbrich, Lisa; Abbozzo, Alessia; Jesse, Frank; di Prisco, Marco; 10.3217/978-3-99161-057-1-099In recent years, Distributed Strain Sensing (DSS), utilizing optical frequency domain reflectometry of Rayleigh backscatter, has gained significant prominence in the realm of Bridge Structural Health Monitoring (BSHM). Its key advantage is its ability to provide continuous strain monitoring with high spatial resolution (0.65 mm) and an accuracy of up to 1 μm/m. This capability facilitates the monitoring of deformations and defects, as well as precise crack detection, the assessment of crack width and others. However, publications on DSS based on Rayleigh backscattering often report local effects that are not linked to disturbances caused by the measurement principle, such as noise or anomalies in strain readings. These effects can complicate the evaluation of DSS data, particularly concerning crack detection and width measurement, as well as tasks like detecting tension wire breakages. There have been theories, suggesting that these local effects may stem from micro-cracks or inhomogeneities within the concrete matrix, yet further investigations into this phenomenon are lacking. Within this paper the phenomenon of local effects, henceforth referred to as the concrete signature, has been investigated on multiple time scales during. The analysis utilizes data from the openLAB research bridge in Bautzen, Germany. Possible reasons for the development and behavior of concrete signature are discussed, along with the challenges and opportunities associated with addressing it.
• Integrated Sensor Technology for Basalt-Reinforced Segmental Lining ElementsEvangelatos, Alexandros; Heindler, Lukas; Galler, Robert; Thurner, Thomas; 10.3217/978-3-99161-057-1-100Current climate protection goals are driving research toward large-scale structural health monitoring solutions for critical, long-lifespan infrastructure, alongside the usage of low CO₂ emission materials. In this work, we present an innovative approach that combines both aspects: the use of continuous basalt fibers assembled into sustainable reinforcement structures, and low-cost measurement systems for integrated long-term condition monitoring, specifically applied to tunnel segment monitoring. As a proof-of-concept for the integration of sensor elements into basalt reinforced concrete structures, we utilized stranded steel wire to create custom strain gauge sensors for integrated strain measurements, paired with a custom-designed resistive bridge-based measurement system to evaluate the feasibility of enabling low-cost condition monitoring. Mechanical tests were conducted on basalt-reinforced concrete specimens under both tensile and compressive loading. The results indicate that the system can measure even low strain values for the sensor-enhanced structures down to approximately 5 μm/m and a standard deviation of 2.1 μm/m, achieving a sensing performance close to state-of-the-art measurement systems and externally applied standard strain gauges. The study demonstrates the potential for cost-effective condition monitoring of individual tunnel segments with sustainable basalt reinforcement. Further optimizations of the system are anticipated in future projects.
• Wind Input and Acceleration & Displacement Outputs Monitoring System for High-Guyed Masts in ROSEHIPS ProjectWang, Miaomin; Gould, Duncan; Stafford, Richard; Koo, Ki-Young; Brownjohn, James; 10.3217/978-3-99161-057-1-101This paper presents a battery-operated wireless long-term response measurement system for high-guyed masts in the UK, as a part of ROSEHIPS project. The monitoring system aims to capture wind speed and direction as input and 3D accelerations and 3D displacements as outputs for a target period of three months. GNSS (Global Navigation Satellite System) based time-synchronisation methods were used in all subsystems for accelerometer, anemometer, and Computer Vision based cameras. Epson E-M352 sensors were used to measure triaxial accelerations with extreme accuracy 0.2ug/ÖHz at multiple points along the height of the mast, together with an anemometer. Low-power consumption features of ESP32 microprocessor unit was utilized to achieve a longer battery life. To remotely monitor its 3D displacement, a wireless stereo vision system was developed using Raspberry Pi. The displacement is measured in the image plane of each camera, and the data is then uploaded to the cloud for 3D reconstruction. The measurement accuracy was validated through an outdoor test, where the two cameras were approximately 100 meters apart, and the target was located about 200 meters from both cameras. The results showed a measurement error of approximately 1 mm. The time synchronisation between the accelerometer and the stereo vision system was also evaluated. By using the system to track an accelerometer mounted on a cantilever, the time-sync error was found to be less than 1 ms.
• Wireless Multi Sensor Monitoring of Engineering StructuresRennen, Markus; 10.3217/978-3-99161-057-1-102Engineering structures are generally understood to be predominantly functional objects, such as bridges, tunnels, cranes, silos etc. Failure of these structures not only causes physical damage but can also lead to interruption of production, disturbance of infrastructure or traffic and thus disruption of operations with commercial impact for both the asset owner as well as concerned commuters or customers. Therefore, in-situ monitoring is of major importance. The challenge lies in the need to observe specific parameters in difficult to access locations, under demanding environments, or with high data rate requirements. These conditions often exceed the capabilities of geodetic observation techniques. Wireless Condition Monitoring (WCM) nowadays can breach the gap by implementing a variety of sensors and maintenance-free hardware without the requirements of line-of-sight or cables. Compact Nodes with internal and external sensors and low power consumption are versatile and provide long battery life. Remote access allows adjustment or temporary changes of configuration settings (e.g. recording intervals). Automated data transfer to cloud-based visualization platforms enables continuous data access with configurable alerts, allowing for proactive evaluation of structural health. The article presents a number of practical field examples that address the challenges mentioned above, while highlighting the specific requirements for interpreting the collected data – supported by examples of result validation using independent techniques.
• Practical approach to calibrating wireless sensors for use in structural health monitoring in an outdoor environmentPetschacher, Michael Markus; Krüger, Markus; 10.3217/978-3-99161-057-1-103When sensor systems are used on outdoor structures (bridges, tunnels, etc.), they are exposed to a wide range of environmental influences. In particular, temperature can significantly affect the quality and accuracy of measurements. While most commonly available sensors are calibrated at temperatures of around 20°C before use, but the influence of variable temperatures is rarely considered. Furthermore, the measuring systems used for these sensors, particularly wireless sensor systems, are often only calibrated for room temperature. For this reason, this paper presents calibration procedures for monitoring systems including the sensors used (here displacement sensors are used as an example). The aim is to provide a practical routine for structural monitoring applications. This involves simulating typical temperature changes in a climate chamber (-20°C to +50°C) while measuring the temperature-induced strain of steel, and analyzing the reproducibility and temperature response of the entire measurement system. Other external influences affecting measurement quality are also discussed, and these are considered when determining the overall measurement uncertainty. This helps to define the requirements and limitations of measurement systems for structural health monitoring, particularly for bridges. The resulting findings should support the standardization process for assessing the suitability of monitoring systems for future SHM applications.
• eNodes: GNSS Time-Synchronised Wireless Accelerometer Measurement Nodes capable of operating indoorsKoo, Ki Young; Wang, Miaomin; Zhu, Zuo; Brownjohn, James; 10.3217/978-3-99161-057-1-104This paper presents time-synchronized wireless acceleration measurement nodes, named eNodes, capable of operating indoors by preserving timing information with a temperature-controlled crystal oscillator (TCXO). While GNSS-based time synchronization is commonly effective for outdoor measurements with available GNSS signals, it does not work indoors, such as inside high-rise buildings or box-girder bridges. To extend GNSS-based time synchronization to indoor applications, timing information is acquired outdoors both before and after the indoor deployment. The TCXO maintains this timing information accurately ensuring a stable and accurate frequency. Each eNode is equipped with an Epson M352 MEMS accelerometer, which offers extremely low noise of 0.2 μg/√Hz, and an ESP32 microprocessor unit. Real-time data transmission is enabled by a Wi-Fi mesh network. A series of experiments were conducted to evaluate the time-synchronization accuracy of the eNodes.
• Study on the Propagation Law of Magnetic Induction Signals for Wireless Communication in Underground StructuresZhang, Yiyan; Zhang, Dongming; Shi, Jingkang; Chen, Mingtao; Liu, Erwu; 10.3217/978-3-99161-057-1-105Despite advancements in electromagnetic wave-based communication, challenges such as high attenuation, medium variability, and large antenna requirements persist. Magnetic induction (MI) communication has emerged as a promising alternative, offering stable transmission characteristics and reduced near-field path loss. While previous studies have explored MI waveguide models and relay coil applications, experimental validation of through-the-ground MI transmission, particularly in homogeneous media, remains limited.This study investigates magnetic signal propagation in uniform underground environments through numerical simulations and experimental validation. A finite element model was developed using COMSOL to simulate magnetic signal transmission, focusing on coil geometry and medium properties. Experimental validation was conducted using a custom-built outdoor platform, where mutual inductance coils were employed to measure signal transmission in both air and soil. Key parameters, including coil spacing and medium permeability, were analyzed to evaluate path loss.Results demonstrate excellent agreement between simulations and experiments, confirming that soil’s air-like permeability results in minimal path loss over short distances. The study highlights permeability as the dominant factor in signal attenuation, with soil moisture and composition showing negligible effects. These findings validate the theoretical framework for MI transmission in homogeneous media and provide practical insights for optimizing MI-based communication systems in applications such as agricultural monitoring and underground utility networks. Future work should focus on long-distance transmission and the impact of enhanced power levels to further refine system performance in real-world scenarios.
• Smart Pavement Subsurface Monitoring with Distributed Embedded Passive RF Sensor NetworkEng, Kent X.; Haas, Zygmunt J.; Das, Samir R.; Djurić, Petar; Stanaćević, Milutin; Glisic, Branko; 10.3217/978-3-99161-057-1-106Pavement subsurface deterioration can lead to catastrophic road failures, often caused by long-term settlements and moisture accumulation. These effects develop gradually and are difficult to identify through manual inspection. Although structural health monitoring (SHM) systems have been developed to address these challenges, most are insufficient to collect 3D spatial information due to their limitations. Current systems primarily focus on surface and base courses, neglecting subsurface courses which sustain loads and provide stability. Therefore, it is essential to develop a long-term and scalable monitoring system for subsurface courses. Radio frequency sensing system has great potential to fill the gap. This study aims to further quantify the uncertainty of using distributed embedded passive radio frequency (RF) sensors in pavement subsurface courses. Laboratory experiments were conducted to investigate the uncertainty sources of the relationships between channel information and structural changes. Key challenges include correlating collected data with subsurface changes and finding the sources of uncertainties. The results demonstrate the effect of system topology on the relationships between channel information and structural changes. These prove the system's applicability in subsurface spatial monitoring. By addressing implementation challenges and decoupling monitored parameters, the system could be further advanced for real-world deployment.
• A Wireless Passive RFID Patch Antenna Strain SensorWei, Chengkai; Shi, Jingkang; Guan, Zhenchan; 10.3217/978-3-99161-057-1-107With the aging of civil infrastructures, strain monitoring is essential for predictive maintenance. However, current sensing technologies mainly rely on active battery powered sensors, leading to substantial expenses and low placement granularity. This paper proposes a wireless, passive RFID patch antenna strain sensor, characterized by a favorable linear relationship between its resonant frequency and the applied strain. In this paper, simulation is carried out by using the COMSOL multi-physics coupling software. The solid mechanics field is coupled with the electromagnetic field, and the frequency-domain scanning is conducted after the model generates strain. The simulation results are compared with the experimental results in the literature to determine their correctness. According to the simulation results of the scattering parameter S11, a patch antenna sensor is designed and fabricated, and corresponding experiments are conducted to detect the variations of the spectral curve before and after the sensor is embedded in concrete, thereby verifying its validity. Finally, the sensor is optimized based on the experimental results.
• 25-year Field Monitoring of the Tsing Ma Suspension Bridge in Hong KongXia, Yong; Zhang, Lu; Lu, Tian; Wang, Xiaoyou; 10.3217/978-3-99161-057-1-108The Tsing Ma Suspension Bridge in Hong Kong has been the world’s first batch of bridges equipped with a longterm health monitoring system since 1997. For the first time, this study reports the first-hand field monitoring data of the bridge from 1997 to 2022. The 25-year data provide an invaluable and rare opportunity to examine the long-term characteristics of loads, bridge responses, and their relationships, thereby enabling the assessment of the bridge’s load evolution and structural condition over time. The current status and recent update of the health monitoring system are also reported. This study is the first to report the one-quarter-century status of a structural health monitoring system and the behavior of a long-span suspension bridge. This research provides a benchmark for many other bridge monitoring systems worldwide.
• Towards Accurate Road Health Monitoring: A Damage Detection System Using FBG SensorsGolmohammadi Tavalaei, Seyed Ali; Hasheminejad, Navid; Ghaderiaram, Aliakbar; Van den bergh, Wim; Hernando, David; 10.3217/978-3-99161-057-1-109Advancements in road infrastructure health monitoring through sensor networks offer a transformative solution to the limitations of traditional inspection methods by enabling more accurate, real-time assessments of structural conditions. However, once appropriate sensors are selected and deployed, a key challenge remains: converting raw sensor data into meaningful health indicators (HIs) that effectively capture structural changes indicative of potential damage. A health indicator (HI) is a crucial metric derived from structural health monitoring (SHM) data, designed to reflect the current condition and damage state of a monitored structure. This study presents a machine learning-based approach leveraging principal component analysis (PCA) to develop a sensitive and damage-specific HI by extracting and ranking the most relevant current features. The proposed method is first validated through experimental fatigue testing using a four-point bending machine under random thermal conditions. To further evaluate its effectiveness and reliability in real-world applications, the approach is applied to field data collected from a network of fiber Bragg grating (FBG) sensors embedded in asphalt pavement. By analyzing strain measurements, the study demonstrates that the PCA-based HI successfully detects structural changes, providing a robust and data-driven solution for real-time infrastructure monitoring.
• Etched fiber Bragg grating sensor-based groundwater salinity monitoring for seawater intrusionJiang, Hongtao; Guo, Junyi; Shi, Bin; Sun, Mengya; Wei, Guangqing; 10.3217/978-3-99161-057-1-110As global climate change drives rising sea levels, coastal regions face growing threats from seawater intrusion. This process increases groundwater salinity, accelerating steel corrosion and compromising the structural integrity of concrete infrastructure. However, addressing these challenges is limited by existing salinity monitoring technologies, which often suffer from slow response times and low sensitivity in in-situ conditions. This study proposes a salinity sensor based on Fiber Bragg Grating (FBG), enhanced by chemically etching the fiber cladding to create an etched FBG (EFBG). This modification improves sensitivity to external refractive indices for accurate salinity detection. A three-layer waveguide dispersion model simulated wavelength shifts during sensor etching and testing using MATLAB, revealing how etching diameters impact sensitivity and confirming a linear relationship between wavelength and seawater salinity. To improve EFBG durability and reduce hydrofluoric acid damage, the etching process was divided into rapid, stable, and fine stages. Results demonstrated that higher etching levels increased sensitivity, achieving a sensitivity coefficient of up to 29.432 pm/% in specific conditions. The EFBG salinity sensor offers high sensitivity, fast response, compact size, corrosion resistance, and interference immunity, making it ideal for in-situ groundwater salinity monitoring in aquifers and mitigating risks to coastal structural health.
• Geo-hazard DFOS Monitoring and its ApplicationsBin, Shi; Zhu, Honghu; Zhang, Chengcheng; Sun, Mengya; Zhang, Wei; Zhang, Taiyin; Guo, Junyi; 10.3217/978-3-99161-057-1-111As the geological body on which humans depend, rock and soil are constantly moving under the action of natural forces and human activities. Their instability often triggers geo-hazards, posing severe threats to the environment, infrastructure safety, and sustainable development. High-quality data acquisition and effective monitoring are essential for geo-hazard prevention and mitigation. The stability of rock and soil is governed by mechanical discontinuous interfaces, which are classified into material, state, and movement interfaces. This paper focuses on distributed fiber-optic sensing (DFOS) technology as an advanced tool for geo-hazard monitoring and early warning. The paper summarizes the authors' achievements over the past two decades in DFOS-based geo-hazard monitoring theory, sensing techniques, and application systems. Key advancements include strain-sensing coupling theory, moisture and seepage monitoring methods, disaster identification and prediction models, and integrated fiber-optic sensing technology platforms. Three representative cases are presented, demonstrating the application of DFOS to monitor the material interface of stope overlying rock, the state interface of land subsidence, and the movement interface of a reservoir slope. Finally, future research directions for fiber-optic sensing in rock-and-soil disaster monitoring are outlined.
• Structural Behaviors of Prestressed Double-T Slab under Loadings with Seasonal EffectsLiang, Yitian; Glišić, Branko; 10.3217/978-3-99161-057-1-112Understanding the effects of temperature on structural behavior is critical in structural health monitoring (SHM), especially for prestressed concrete components with complex geometries due to their complicated internal strain distributions. Temperature-strain relationships in structural components can offer valuable insights into various structural properties, such as the coefficient of thermal expansion (CTE), and boundary and continuity conditions of structures. However, these relationships can be influenced by variability of ambient environmental conditions, especially ambient temperature variations, which can create thermal gradients and complicate the interpretation of relationships and identification of structural properties. This study presents a preliminary study on the structural behaviors of a prestressed slab with a double-T cross-section through a series of static and dynamic loading tests conducted across four seasons, with ambient temperatures ranging from 8°C to 24°C. The proposed approach utilizes long-gauge strain sensors embedded within the slab to continuously capture strain and temperature data. By analyzing the temperature-strain relationships derived from on-site loading test measurements, the study aims to evaluate how the structural behaviors of the prestressed double-T slab, which can reflect the structural properties, such as boundary and continuity conditions, change under different environmental temperatures. The preliminary results show clear variations in strain changes for the same loading condition under different temperatures. These variations suggest that environmental temperatures and thermal gradients could affect load response and boundary restraints. Furthermore, the findings demonstrate that the embedded long-gauge strain sensors effectively capture the temperature changes and strain distribution under the loadings, enabling the assessment of structural behaviors of the slab with seasonal effects. This research highlights the importance of accounting for environmental factors in structural health monitoring and provides new perspectives for understanding and predicting the behavior of structures with complex geometrical properties (e.g., double-T slab) under varying conditions.
• Vibration Analysis of Ship Hulls using Fiber Bragg GratingRoberts, Gethin Wyn; Àarberg, Irena; Lienhart, Werner; 10.3217/978-3-99161-057-1-113The paper outlines an ongoing research project, incorporating Fiber Bragg Grating (FBG) systems to measure and detect the vibration in a ship’s hull. The causes of such vibration are due to the various engines and motors on board, as well as the force induced by the movement of the sea, and the vibration induced into the hull by the propellor. Five ships in all have been monitored using the FBG system, using both 3 sensor rosettes, and chains of 10 sensors. All the sensors used were glued to the ships’ hulls and various sea trials carried out. The tests included gathering data with the engines switched off, the engines running at various speeds, both whilst stationary in the harbour as well as whilst sailing. Change detection is the main application of such monitoring, and such change is evident and detected due to a broken flange on one ship, and a new engine on one ship. Data were gathered before and after such effects were changed. The induced vibration due to the propellor is also evident.
• Structural monitoring of Zeeland Bridge - improved structural identification by combining a modular model updating framework with a mobile measurement setup during load testsBesseling, Floris; Kortendijk, Coen; De Bruijn, Janno; Lourens, Eliz-Mari; 10.3217/978-3-99161-057-1-114To reduce uncertainties associated with its structural re-assessment, the Zeeland Bridge in the Netherlands is currently the subject of a field lab, which will run for 2 years. In this contribution, the structural identification approach, the model updating concept and the first measurement campaign are presented, followed by some preliminary measurement results. The present stage focusses on load testing of the bridge to obtain insight into the possibly varying response in different spans of the bridge. Previously, parametric studies to expose input-output parameter dependencies were performed on a representative sub- system of the bridge, and the results are used to assist in the design of a measurement campaign and the development of a robust model updating strategy for the bridge. The results of the first measurements allow for evaluation of the actual performance of the bridge when subjected to heavy truck loads. This information will be used as a basis for further development of the updating approach.
• Scotiabank Saddledome Roof Structure Monitoring ProgramLimaye, Vidyadhar; Roy, Subharajit; LeBlanc, Mark; Whittaker, Elizabeth; Paranjape, Atul; Alleyne, Lindsay; 10.3217/978-3-99161-057-1-115Scotiabank Saddledome, an indoor arena in Calgary, Alberta, constructed in 1983 with a hyperbolic paraboloid (saddle shaped) roof, has concave cables running in the east west direction to support gravity loads and convex cables in the north-south direction to support lateral loads. The stranded cables, encased in concrete and anchored into a ring beam, are not visible to detect the signs of corrosion. Events held in the arena require suspension of entertainment loads from the roof structure which, when coupled with snow loads, pose a major safety concern. The installation of an acoustic monitoring system in 1999 to detect breaks in cable strands did not perform as intended. In 2014 the roof membrane was damaged by a significant hailstorm exposing concrete to moisture infiltration. In 2022, a Building Condition Assessment of the roof recommended further investigation of the bonded cable system. Learning from strand failures at the Arizona Veterans Memorial Coliseum resulting in costly remediation work, a Structural Health Monitoring (SHM) system with strategically mounted sensors and a laser-based deflection measuring device, was implemented in July 2023. The objective of this SHM program is to collect data on monitoring parameters for roof movements continually over a period of 3 years in order to identify trends and implement an active alarm system based on data collected in the first year. This paper presents the field application of SHM for risk management of a complex roof structure.
• Long-term monitoring and data processing of a continuous prestressed concrete bridgeSavard, Marc; Laflamme, Jean-François; 10.3217/978-3-99161-057-1-116The Grand-Mere Bridge in the province of Québec, Canada, built in 1977, is a cast-in-place, segmental box-girder bridge measuring 285 m (935 ft) in length. Several problems arose during the construction of this bridge and an increasing deflection combined with localized cracking were noted after only a few months of operation. These defects were mainly due to insufficient prestressing, causing high tensile stresses in the deck and possible corrosion of the prestressing steel. A few years after strengthening of the bridge in 1992, a long-term monitoring program was implemented, including vibrating wire sensors (strain and crack sensors), inclinometers and temperature sensors. So far, more than 20 years of data have been collected and processed, leading to the recommendation of the rehabilitation of the structure using stay cables to ensure that the structure performs well until its scheduled replacement. This paper presents the instrumentation strategies, the various trends observed in the data and the relevant interpretations derived from them. In the context of damage detection, finite-element models have been developed and calibrated on measurements. Data indicate that the addition of stay cables eliminated the progression of permanent deflection and provided the structural system with added strength and redundancy. Lessons learned from this investigation are presented, along with a discussion of the conditions required for successful electronic monitoring.
• Smart Structural Health Monitoring with Acoustic EmissionHäuserer, Michael; Trattnig, Horst; 10.3217/978-3-99161-057-1-117It is common sense that civil infrastructure like tunnels or bridges are coming in age and needs to be renewed in the next years. These issues are, amongst others, related to massive increased traffic load nowadays compared to the time of construction and material issues which turned out after many years of operation. In concrete structures with tendons, the stress corrosion cracking of the used steel is one of the main issues and leads to tendon failure with significant impact on the stability of the structure. SHM with Acoustic Emission is used successfully for many years to detect tendon failures. The products on the market have been adapted since many years to the needs of the customers and the full measurement chain from a self-checking smart sensor network, high performance and scalable data acquisition systems to automate data analysis, processing and alarming is available. A cloud-based dashboard rounds up the package and makes processed data available for customers. Since May 2024 a guideline from DGZfP “Richtlinie SE 05 Detektion von Spanndrahtbrüchen mit Schallemissionsanalyse” is available and give a general frame about the approach, definitions ad help to specify tenders in a correct way.
• The Collapse of the Carola Bridge – Forensic Engineering and Palliative MonitoringFiedler, Max; Schacht, Gregor; Ritter, Robert; Marx, Steffen; Scheerer, Silke; Clages, Luis; Ebell, Gino; Czeschka, David; Voigt, Chris; 10.3217/978-3-99161-057-1-118The events of September 11, 2024, will remain etched in the collective memory of Germany’s bridge engineering community. The sudden and unannounced partial collapse of a prestressed concrete bridge rightfully reverberated across society at large. The structure in question was the Carola Bridge in Dresden. This architecturally refined and exceptionally slender bridge is, with good reason, regarded by professionals as an icon of its time's structural engineering. Even by today’s standards, its design and construction would pose a considerable challenge. This paper presents the main findings from investigations undertaken to determine the cause of the collapse and attempts to reconstruct the failure process. Additionally, the acoustic monitoring system implemented to safeguard the remaining superstructures is also presented.
• Understanding the Dynamic Behavior of Large Sign Structures Under Wind LoadingLinderman, Lauren Elizabeth; Johnson, Nicole; Nguyen, Lam; Schillinger, Dominik; Guala, Michele; French, Catherine W.; 10.3217/978-3-99161-057-1-119Dynamic Messaging Signs (DMS) are much larger and heavier roadside signs than typically placed on their respective support systems. The excess weight and size of these signs, in conjunction with their breakaway support systems, introduces wind-induced vibration problems not seen in the past. The AASHTO LRFD Specification for Structural Supports for Highway Signs, Luminaires, and Traffic Signals (SLTS), including interim revisions through 2022, does not yet address vibration design for these nontraditional roadside signs. The DMS support system, specifically the friction fuse connection, is susceptible to the formation of stress concentrations and potential fatigue issues. A DMS was instrumented with strain gages, accelerometers, anemometers, and temperature sensors to characterize both the wind loading and response of the structure. A dynamic numerical model was validated with experimental field data and used to evaluate the fatigue life of the DMS instrumented in the field. The results of the dynamic analysis performed with the validated FEM model differed significantly from the analysis with the equivalent static pressure equation for natural wind gusts prescribed in the AASHTO Specification, which highlights the importance of considering the dynamic behavior of these heavier sign panels. Extension of the dynamic method to models of other large DMS in service showed a greater fatigue stress and corresponding shorter estimate of the fatigue life.
• 6-Component Operational Modal Analysis of wind turbines for damage detectionMüller, Laurin; Dhabu, Anjali; Bernauer, Felix; Donner, Stefanie; Bode, Kay; Hadziioannou, Céline; 10.3217/978-3-99161-057-1-120The rapid expansion of the wind energy sector has necessitated remote monitoring of wind turbines to ensure safe, reliable, and cost-effective operations. While traditional inspection methods remain in use, there is an increasing shift toward passively monitored, real-time solutions to detect and localize potential damage. The present study makes a novel attempt to explore the potential of 6-component seismic data for use in structural damage detection frameworks for wind turbine monitoring. Measuring both translational and rotational ground motions is a relatively recent advancement in Structural Health Monitoring, offering valuable insights into the dynamic behavior of towers. In the present study, two 6-component (6-C) seismometers were placed at the foundations of two different wind turbine types in the wind park of Kirchheilingen, Germany. The monitoring campaign lasted 7 weeks and focused on capturing vibrational data during operation. By analyzing these signals, in conjunction with Supervisory Control and Data Acquisition (SCADA) Data from the turbine operator, the research aims to identify patterns indicative of structural damage, such as changes in modal frequencies, damping ratios, or signal coherences. It will contribute to the development of scalable, cost-efficient SHM systems tailored for the wind energy industry. Furthermore, the insights gained could inform future design improvements and predictive maintenance strategies, ultimately supporting the sustainable growth of renewable energy infrastructure.
• Estimation of Wind Turbine Foundation Settlement and Error Modeling Using High-Resolution Dual-Orbit Satellite DataDallari, Veronica; Bassoli, Elisa; Grassi, Francesca; Mancini, Francesco; Vincenzi, Loris; 10.3217/978-3-99161-057-1-121The demand for renewable energy sources is increasing, making it essential to develop effective maintenance plans for existing infrastructure. This study represents the initial step in a process designed to estimate the settlement of onshore wind turbine foundations, as well as its associated uncertainties. The method relies on high-resolution dual-orbit satellite data, which help to reduce cost and time required for instrument installation and on-site inspections. The turbine is modeled as a 1D rigid body and is assumed to be firmly constrained to the foundation slab. The proposed formulations allow for the estimation of the turbine motion components – translations in the W-E and vertical directions and rotation along the S-N axis –, which can be generally linked to foundation settlement. The components are determined by solving a linear system which accounts for the mean annual velocities of the Permanent Scatterers on the wind turbine surface, turbine height and incidence angles of satellite orbits. At the present stage, analytical formulations for the a posteriori estimation of the motion component uncertainties are proposed, with a particular focus on the positioning error in elevation of Permanent Scatterers. To assess the accuracy of these expressions, Monte Carlo numerical simulations are conducted. The strong agreement between numerical and analytical results demonstrates that the turbine motion components can be estimated with high accuracy.
• Distributed fibre optic sensing of decommissioned wind turbine blades under bendingWang, Chao; Zhang, Shaoqiu; Ruane, Kieran; Jaksic, Vesna; Li, Zili; 10.3217/978-3-99161-057-1-122The decommissioning of wind turbine blades (WTBs) presents significant environmental challenges due to their non-biodegradable composition. To promote sustainable reuse and repurposing, it is essential to establish effective structural health monitoring (SHM) techniques that can accurately assess the residual performance of decommissioned WTBs. This study investigates the feasibility and applicability of distributed fibre optic sensing (DFOS) as an advanced monitoring tool for evaluating the structural integrity of decommissioned WTBs intended for reuse in civil engineering applications. A four-point bending test was conducted on a WTB segment, with DFOS deployed alongside other monitoring techniques, including strain gauges, and digital image correlation (DIC). The DFOS measurements demonstrated strong agreement with those obtained from strain gauges and DIC, with negligibly small variations in strain magnitude, validating its accuracy and reliability for continuous strain monitoring. The results further confirmed sufficient load-bearing capacity of the WTB segment, indicating its potential for second-life structural applications. This study highlights the capability of DFOS in providing high-resolution, distributed strain measurements, offering a promising approach for assessing the suitability of decommissioned WTBs for reuse. Future research aims to incorporate material characterisation studies and long-term monitoring to establish standardised frameworks for the sustainable repurposing of WTBs, contributing to a circular economy in the wind energy sector.
• Prediction of urban wind speed during tropical cyclones using a novel deep learning-based spatiotemporal modelZeng, Yuan-Jiang; Chen, Zheng-Wei; Ni, Yi-Qing; Chan, Pak-Wai; 10.3217/978-3-99161-057-1-123Tropical cyclones (TCs) stand as one of the most destructive extreme weather events, posing significant threats to human safety and urban infrastructure. One critical phenomenon associated with TCs is the occurrence of strong winds; thus, accurate prediction of urban wind speed during TCs can provide essential information for decision-making, which is vital for enhancing urban resilience. This study proposes a deep learning-based model that accounts for the spatial and temporal dependencies of wind speed data collected from sensors of meteorological stations while addressing the impacts of climate change. The model integrates temporal and spatial encodings with measured time series data, enabling the capturing of long-term temporal dependencies that reflect periodic weather patterns and climate change through the attention mechanism of a Transformer architecture. The outputs derived from this computation are further utilized to identify dynamic patterns of wind speed during TCs. Additionally, a graph neural network (GNN) is integrated to capture spatial dependencies, considering the non-Euclidean distribution of meteorological stations. To evaluate the performance of the proposed model, wind speed measurements from Hong Kong between 2000 and 2023 are used for training and testing. Comparative analyses with sequence-to-sequence models and GNN-recurrent neural network or GNN-Transformer hybrid models demonstrate that the proposed model enhances prediction performance.
• Graph network representation and intelligent evaluation for service performance of bridge clustersLi, Shunlong; Wang, Jie; 10.3217/978-3-99161-057-1-124As the most vulnerable part of the infrastructure transportation network, bridges will inevitably encounter problems such as aging and degradation throughout their entire service life [1]. The maintenance costs for all bridges within the region are increasing year by year [2]. When the financial conditions are insufficient to fully cover the costs, many domestic provinces and cities adopt the maintenance and repair plans based on single bridges relying on experience and the "fire-fighting" post-event repair mode [3]. There are few scientific management models that focus on the overall service performance of the regional bridge clusters. This leads to either excessive or insufficient detection and maintenance. Therefore, there is an urgent need for a systematic intelligent assessment framework for existing bridge clusters. However, current research on service performance evaluation and prediction for large-scale bridge networks suffers from multiple limitations, such as limited research objects, simplified modeling forms, difficulty in quantitative assessment, generalized prediction outcomes, and insufficient consideration of maintenance decision-makings [4]. Therefore, this study focuses on graph network representation and intelligent evaluation for service performance of bridge clusters. Firstly, a systematic comparative analysis of two distinct graph network representation methodologies (undirected and directed network) is conducted based on actual bridge cluster cases of different scales. Secondly, Second, tailored intelligent assessment frameworks of vulnerability are developed for each representation. Finally, benchmarking against evaluation outcomes reveals critical performance differentials across methodologies. This work thus establishes a theoretical foundation for intelligent operation and maintenance strategies in bridge network management.
• Lightweight vision fundamental model-based structural surface crack segmentation using model distillationGuo, Yapeng; Li, Shunlong; 10.3217/978-3-99161-057-1-125Vision fundamental models demonstrate considerable competitiveness in structural surface crack segmentation due to their strong generalization ability. Vision fundamental models improve the fitting capacity for various objects by increasing image encoder complexity. However, for crack segmentation, the excessive number of these parameters leads to slow running speeds and large space occupation. This paper presents a lightweight Segment Anything Model (SAM)-based crack segmentation method using model distillation technology, aiming for consistent crack image embedding. Firstly, end-to-end automatic crack segmentation is achieved by modifying the SAM model through the addition of a crack segmentation head. Secondly, model distillation is employed to transfer features from the heavy-parametric encoder in SAM with minimal loss. Comparative analysis of cutting-edge crack segmentation techniques across eight frequently utilized datasets demonstrates their effectiveness and precision. The findings reveal the potential of mobile deployment of civil structure damage identification based on vision fundamental models.
• Spatial-Temporal Graph Model for Environmental Temperature and Traffic Flow Prediction of City RegionsLin, Chenglong; Xu, Yang; 10.3217/978-3-99161-057-1-126Recent research for correlation prediction from spatial-temporal monitoring data of bridge groups has explored graph neural networks and state space models, offering new angles and advanced algorithms. However, current research still faces significant challenges: (1) constructing suitable graph structures to accurately reflect complex spatial-temporal correlations, (2) designing an effective spatial-temporal neural network to capture spatial-temporal dependencies during the service state evolution of bridge groups, and (3) fully making use of spatial-temporal monitoring data to boost prediction accuracy and efficiency. To tackle these challenges, this study introduces a graph selective state space model for spatial-temporal prediction of environmental temperature and traffic flow for bridge groups. Firstly, a spatial-temporal graph structure is set up to account for data characteristics in both spatial and temporal aspects and forecast the dynamic evolution of bridge group system. Then, a state space model is built to produce a structured state space sequence and introduce a selective mechanism to dynamically adjust model behaviors and optimize computational resources. Lastly, through decomposing and reintegrating spatial-temporal features of monitoring data for bridge groups under different complexities, validation experiments are performed to show the efficacy, universality, and efficiency using multi-type, multi-scale, and multi-granularity spatial-temporal monitoring data of environmental temperature and traffic flow.
• Ultimate flexural strength analysis of serving concrete main girders considering bridge deck pavementCui, Hongtao; Li, Zhonglong; Guo, Yapeng; Li, Shunlong; 10.3217/978-3-99161-057-1-127Concrete girders are a type of widely used structures in small and medium span bridges. The ultimate flexural strength of concrete main girders serves as a critical foundation for assessing the structural performance of small and medium span simply supported girder bridges, and is essential for ensuring their safe operation. After the installation of prefabricated concrete girders, cement and asphalt concrete are sequentially poured on the top surface of the girders as bridge decks. The bridge deck pavement and concrete girder jointly bear the overall external loads. However, during the design and operation stages, the bridge deck pavement is typically regarded as secondary dead load when estimating bearing capacity, without considering its inherent reinforcement effect on the main girder. Thus, understanding the damage mechanism and destructive behaviour of concrete girders whilst considering the deck pavement effect is crucial for bridge safety assessment. The currently prevalent laboratory-based research method using scaled models can effectively elucidate the failure mechanisms of concrete girder members by controlling the experimental environment, the findings cannot be directly extrapolated to evaluate the service performance of actual bridge structures.
• Data-Driven Monitoring Solutions for Concrete Structures: Long-Term Insights with CorroDec2G SensorsSteffes, Christian; 10.3217/978-3-99161-057-1-128Passive sensor technologies offer a robust solution for monitoring concrete structures. These technologies utilize advanced methods to measure critical parameters such as moisture, corrosion, and temperature within the concrete. By employing RFID technology, the sensors operate wirelessly and require no maintenance. With a lifespan exceeding 80 years, passive sensors enable comprehensive and efficient long-term structural monitoring. This article provides insights into the installation methods, the benefits of the cloud-based data platform, and how this technology contributes to improved safety, durability, and the preservation of infrastructure.
• Numerical dataset for benchmarking of drive-by bridge monitoring methodsCantero, Daniel; Sarwar, Zohaib; Malekjafarian, Abdollah; Corbally, Robert; Makki Alamdari, Mehrisadat; Cheema, Prasad; Aggarwal, Jatin; Noh, Hae Young; Liu, Jingxiao; 10.3217/978-3-99161-057-1-129Publicly accessible vehicle measurements for testing and validating drive-by bridge monitoring techniques are currently insufficient. Although relevant monitoring campaigns have been conducted, their results are limited and generally inaccessible to the research community. Consequently, this paper introduces a numerical dataset designed to advance drive-by monitoring methods. The dataset is freely available and can be downloaded from an online repository. It comprises numerically simulated vehicle responses generated using an open-source vehicle-bridge interaction model. The repository includes over half a million individual vehicle crossing events, covering various monitoring scenarios, bridge spans, damage locations, damage magnitudes, road profile conditions, and vehicle properties. This dataset is intended to serve as a reference solution and benchmark for future developments in drive-by bridge monitoring.
• Indirect footbridge damage classification using explainable deep learning: A field-testing studyLi, Zhenkun; Lan, Yifu; Lin, Weiwei; 10.3217/978-3-99161-057-1-130Structural health monitoring (SHM) has gained significant attention in recent decades due to several structural failures and the increasing maintenance demands from stakeholders. This urgency has been further amplified by the impact of predictive climate changes worldwide. Footbridges, as critical components of modern transportation systems, play a vital role in daily life and therefore require meticulous attention to their health conditions. Traditionally, monitoring footbridge conditions involves installing many sensors directly on the structure, which is often cost-prohibitive in engineering applications. Recent advancements have highlighted the indirect method of bridge health monitoring, where sensors are mounted on passing vehicles rather than the bridge itself. This approach is not only more economical but also easier to implement in practical engineering scenarios. This paper further extends the indirect monitoring method to classify footbridge damage using the responses of shared scooters. Advanced deep learning techniques are utilized to predict the severity of damage to the footbridge based on the vibrations recorded from shared vehicles. The proposed method was validated through field tests involving scooters and a footbridge. Furthermore, to interpret the outputs of the deep learning model, SHapley Additive exPlanations (SHAP) values were calculated, offering insights into the decision-making process of the model.
• Drive-by bridge modal identification under multi-source excitationsLi, Jiantao; Tan, Jie; 10.3217/978-3-99161-057-1-131The drive-by bridge modal identification (BMI) method, which employs a sensory system mounted on a moving vehicle, offers an efficient and cost-effective alternative for monitoring the health of bridge structures, particularly for short- to mid-span bridges. This technique allows for real-time, large-scale bridge assessments without the need for stationary sensors or traffic disruptions. However, extracting accurate modal parameters, such as frequencies and damping ratios, from vehicle responses is challenging due to the influence of multi-source excitations, including road surface roughness, random traffic loads, and dynamic vehicle-bridge interactions. These factors introduce noise and complexity that can compromise the reliability of the BMI method. To address these challenges, this study integrates an adaptive signal decomposition technique, Successive Variational Mode Decomposition (SVMD), with Operational Modal Analysis to accurately identify the modal frequencies and damping ratios from drive-by measurements. The impact of multi-source excitations on the vehicle-bridge interaction process is systematically investigated, and key factors affecting the accuracy and reliability of BMI under such conditions are analyzed. Based on these findings, recommendations are made to improve the robustness and precision of the drive-by BMI method. This work might contribute to advancing the practical implementation of BMI in real-world bridge health monitoring applications.
• Field test on tunnel indirect damage identification from moving train responseLi, Qi; Xie, Xiongyao; Zeng, Kun; 10.3217/978-3-99161-057-1-132To explore the potential application of the tunnel damage identification method based on train acceleration, a three-axis accelerometer was installed on a metro train carriage to collect acceleration signals. The original signals are segmented and aligned according to the stations, with data analyzed in terms of station sections. Next, the probability density distribution, fast Fourier transform spectrum, and one-third octave spectrum of the signal are calculated. A time-frequency domain fast analysis software for acceleration data is then developed. By comparing changes in time-frequency domain features, the anomalous section of the tunnel is identified. The results confirm that the tunnel damage identification method based on train acceleration is applicable for real-world metro tunnels.
• A methodology for data collection and aggregation in population-based structural health monitoring ecosystemsLim, David Ren-Huang; Ferguson, Alan J.; Brennan, Daniel S.; Hester, David; Woods, Roger; 10.3217/978-3-99161-057-1-133Population-based Structural Health Monitoring (PBSHM) provides insights based on data derived from comparing multiple structures' responses, providing a shift towards an integrated data domain. This presents significant challenges in data collection and integration of data across diverse structural populations, such as sensor systems, environmental data, and maintenance records and requires substantial engineering effort. This fragmentation of data across different formats and systems creates substantial engineering overhead when integrating new data sources, limiting the practical implementation of population-based approaches. This paper introduces a structured data flow architecture for systematic data collection and aggregation in PBSHM ecosystems by defining distinct functional components within the data collection process and enabling the structured integration of diverse data sources. The results establish a foundation for scalable PBSHM data collection, supporting the broader transition towards integrated structural health monitoring ecosystems.
• Towards a plug and play population-based structural health monitoring aggregation pipeline design for resource constrained systemsLim, David Ren-Huang; Ferguson, Alan J.; O’Higgins, Connor; Hester, David; Woods, Roger; 10.3217/978-3-99161-057-1-134The practical implementation of Population-based Structural Health Monitoring (PBSHM) often involves distributed which face challenges from limited compute resources, power budgets, and variable communication bandwidths, for example, IoT devices with battery-powered wireless sensor network gateways uploading data over metered connections. Building upon the data flow architecture presented in our first paper in this special session, this paper demonstrates how inherent state within data streams' can be leveraged for optimisation. This paper introduces a novel, plug-and-play aggregation pipeline specifically designed to address these limitations. We present an optimised data representation and transmission strategy that minimises computational and bandwidth requirements at the network edge. By leveraging efficient data serialisation techniques, our pipeline achieves a significant reduction in data payload size with negligible information loss, thereby enhancing the scalability and financial viability of PBSHM systems. This work validates an enabling technology for the real-world deployment of large-scale, low-power monitoring ecosystems. It does so by comparing two data formats, JavaScript Object Notation (JSON) and Concise Binary Object Representation (CBOR), using monitoring data from a long-term bridge campaign. The results show a reduction in the volume of transmitted data by up to 12.2 times.
• Advancing PEAR: Development of a Bridge Benchmark Datasets for PBSHM ResearchO’Higgins, Connor; Gowdridge, Tristan; Hester, David; Worden, Keith; Brennan, Daniel S.; 10.3217/978-3-99161-057-1-135Population-Based Structural Health Monitoring (PBSHM) is an emerging field in Structural Health Monitoring that leverages data from multiple structures to enhance the assessment of individual structures. Unlike traditional SHM, which generally relies on data from a single structure, PBSHM utilises collective knowledge from a population to facilitate increasing the knowledge on an individual structure. Transfer learning enables the inference from a source structure to a target structure within the population. One of the limitations of this method is that a lot of transfer-learning methods require data models that are trained using substantial amounts of high-quality data which can be difficult to obtain. To support PBSHM research, the concept of the Population-based SHM Engineered Asset Resource (PEAR) has been introduced. PEAR is conceptualised as a benchmark dataset containing semi-realistic structures and associated data intended to drive the development and validation of PBSHM methodologies. This work advances the PEAR prototype by developing complete populations for two types of bridges, along with their associated data. The pipelines for generating these populations are presented, detailing how they produce structural data and PBSHM-specific models. Additionally, a simple analysis of the generated populations is conducted, demonstrating their utility in PBSHM research and showcasing the potential of PEAR as a resource for current and future PBSHM research.
• A Transfer Learning approach for damage identification in operational viaductsMorleo, Eleonora; Limongelli, Maria Pina; Piscini, Andrea; Troielli, Edoardo; 10.3217/978-3-99161-057-1-136A major limitation in data-driven Structural Health Monitoring is the scarcity of labeled data for training machine learning models. Transfer Learning addresses this by enabling knowledge sharing across similar structures, reducing datasets distribution shift. This study proposes a novel Transfer Learning framework for damage identification in operational viaducts with similar spans, using modal frequencies as damage-sensitive features. Domain Adaptation is performed via Normal Condition Alignment, to map source and target features in a shared latent space. A baseline normal condition is established on source features through a linear regression model. Gaussian Mixture Models are trained on source residuals, and used to detect anomalies in the target domain, based on residual distributions. A real viaduct for which long-term monitoring data are available is used as a case study. The structure comprises two homogeneous datasets collected on the deck of similar spans. Source data pertain to a deck with extensive measurements, whereas target data refer to a second deck with a reduced dataset, due to sensor malfunctions. Damage is simulated in the target dataset by reducing the measured frequencies. Validation using data from real damaged scenarios will enable future scaling of the proposed framework to operational conditions, providing a practical tool for data-driven SHM of viaducts, enabling damage detection in under-instrumented areas by leveraging data from other spans.
• The future of conservation: Citizen Science models for the Photomonitoring of cultural heritageCosentino, Antonio; Clementi, Jessica; Molinari, Antonio; Sanvito, Veronica; Mazzanti, Paolo; 10.3217/978-3-99161-057-1-137This study investigates the effectiveness of photomonitoring as a remote sensing technique for cultural heritage conservation, focusing on the Aurelian Walls (Rome) and the Church of Santa Apollonia (Ferrara), Italy. Using mobile devices such as smartphones and tablets, structural changes—including brick detachment and vegetation growth—were detected through Structural Similarity Index (SSI) mapping. The results highlight both the advantages and limitations of mobile-based monitoring, emphasizing its flexibility and rapid deployment. Key challenges include variations in pixel size and lighting conditions, which influence data consistency. Despite these limitations, the study supports the potential of citizen science integration to enhance spatial and temporal data collection. By leveraging crowdsourced imagery, monitoring efforts can become more comprehensive and cost-effective. The findings align with broader citizen science initiatives, demonstrating how non-invasive, mobile-based techniques can contribute to sustainable heritage preservation. Future research should focus on optimizing data acquisition and processing methodologies to improve the robustness of this approach.
• Correlation of natural frequencies of bridges that are under similar environmental conditions.Swaile-Blemmings, Ellie; Hester, David; O’Higgins, Connor; Taylor, Su; Woods, Roger; Cross, Elisabeth J.; 10.3217/978-3-99161-057-1-138While Structural Health Monitoring (SHM) has potential to aid bridge managers, its adoption has been limited, with one of the challenges being determining a bridge’s condition without a historical reference point for that structure. Researchers have started investigating Population-Based Structural Health Monitoring (PBSHM) to tackle this, facilitating the sharing of data from comparable structures. The key advantage of PBSHM is that it potentially enables us to use data from one structure to make inferences about the health of another structure in the same population. Whilst, to date, populations that have been used for PBSHM have been defined using structural similarities alone, you might be missing out on information that could be useful for bridge managers, which raises the question: Could we define populations in a different way? This research investigates if it is potentially useful to define a population of bridges based on whether they experience the same environmental conditions. To answer this, long-term natural frequency data from two bridges close to each other are analysed to determine the level of correlation between them. This work shows that it may be potentially useful to define populations based on factors other than structural similarities, which allows greater opportunities for PBSHM.
• A Novel AI-Wavelet Based Framework for Benchmark Data Analysis in Structural Health MonitoringSilik, Ahmed; Noori, Mohammad; Farhan, Nabeel; Wang, Tianyu; Altabey, Wael A.; Wu, Zhishen; 10.3217/978-3-99161-057-1-139Structural Health Monitoring (SHM) plays a vital role in ensuring the safety, durability, and operational efficiency of critical infrastructure. Traditional SHM methods often fall short in detecting subtle damage patterns, particularly when faced with noisy signals, missing data, or the complex, time-varying behavior of real-world structures. To address these challenges, this study presents a hybrid framework that integrates Discrete Wavelet Transform (DWT) with a deep learning architecture combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. The proposed approach begins by segmenting long-duration acceleration signals into fixed-length windows and applying DWT to extract informative time–frequency features. CNN layers are then used to learn spatial representations from the transformed data, while LSTM layers capture temporal dependencies critical for detecting structural changes over time. The model is trained and evaluated using benchmark SHM datasets under both healthy and damaged states. Moreover, supervised learning is utilized for accurate damage severity classification, while unsupervised learning are used to facilitate anomalies detection without relying on labeled samples. Experimental results demonstrate improved performance in classifying damage conditions compared to conventional machine learning approaches. This framework offers a robust and scalable solution for data-driven SHM, supporting more accurate diagnostics and paving the way for predictive maintenance in complex monitoring environments.
• Preliminary results from a field application of dynamic monitoring on three spans of a railway bridgeMassarelli, Eleonora; Civera, Marco; Ventura, Giulio; Chiaia, Bernardino; 10.3217/978-3-99161-057-1-140Vibration-based Structural Health Monitoring is of the foremost importance for critical civil infrastructures, especially concerning the safety of the train transport network. In fact, even minor structural changes might cause derailment and potentially fatal accidents. This contribution reports some preliminary analyses carried out on 52 accelerometric recordings collected over two consecutive days from three spans of a railroad bridge. The acquisitions include several train passages and the quiet periods between them, when the structure was excited only by ambient vibrations (i.e. random microtremors), thus allowing Ambient Vibration Testing (AVT). Specifically, a newly developed Automated Operational Modal Analysis (AOMA) algorithm was applied. Its results are here compared to state-of-the-art commercial software (ARTeMIS). Some considerations regarding the effects of train passages are also briefly reported, as well as directions for current and future research work in this field.
• A damage screening method of the concrete slab focusing on correlation of mode shapesKadota, Takanori; Daigo, Takuya; Tomioka, Akihiro; Miyamori, Yasunori; Oshima, Toshiyuki; 10.3217/978-3-99161-057-1-141In Japan, visual inspections have been conducted as every five years duty since 2014. To improve the quality of inspections, it is necessary to record the evaluation of the structure's condition and its performance until the next inspection. However, methods and technical level depend on the judgement of the road administrator. A city of Kitami, Hokkaido, Japan where our university is located, was merged with one city and three towns in 2006, and has the largest area ranking in Hokkaido area and 4th in Japan. The total number of bridges, viaducts and functional culverts over 2.0 m length became more than doubled to 524 bridges compared to before the merger. Therefore, it is necessary to establish a labor saving and cost-effective method to assess the performance of bridge structures. In recent years, it has become possible to easily measure structural responses by the improvement of sensor performance, and more research has been conducted on maintenance management methods. Among those, vibration characteristics have a significant effect on the stiffness and mass of a bridge member and will be the index that grasps easily various damage effects. In this study, a damage location screening method was investigated for concrete slab bridges where segregation was suspected, based on correlation between mode shapes by using the COMAC. As a result, it is shown that these techniques will identify the damaged location and be used as an effective method to screen damaged locations.
• Feasibility of micro-motion from SAR imagery for vibration-based SHMLotti, Alessandro; Vattulainen, Aleksanteri B.; Suppi, Chiara; Diaz Riofrio, Sebastian; Milillo, Pietro; Tubaldi, Enrico; Zonta, Daniele; Clemente, Carmine; 10.3217/978-3-99161-057-1-142This paper explores the use of single-pass Synthetic Aperture Radar (SAR) images to remotely measure the vibrational response of reflective ground targets, aiming to support vibration-based Structural Health Monitoring (SHM) of civil infrastructure. A Sub-Pixel Offset Tracking (SPOT) technique is applied to SAR imagery to reconstruct the time history of radial velocity, using the principle of micro-motion (m-m) effects induced by vibrating targets. Validation is by comparing SAR-extracted velocity profiles with synchronous ground measurements of a vibrating Corner Reflector (CR). Three representative test cases— including single-frequency, and amplitude-modulated vibrations—with maximum displacement values of 24.1 mm, 1.1 mm and 0.5 mm are analyzed using SAR images collected by the Umbra X-band SAR constellation. The extracted vibrational features are evaluated through time-domain correlation, spectral accuracy, and error metrics. Results confirm that SPOT can reconstruct velocity time histories and frequency content reliably for medium-to-high velocity scenarios (> 10 mm/s) and single-frequency signals. Even under low-velocity, complex signal conditions, the main frequency peaks are detectable, with negligible frequency errors and correlations of 0.61 (time) and 0.82 (frequency). This study demonstrates the potential of SAR m-m processing for fully remote vibration-based SHM, offering a scalable, installation-free alternative for assessing structural dynamics.
• Setting an optimal threshold for novelty detection in data-driven Structural Health MonitoringDe Corso, Alessio; Rainieri, Carlo; 10.3217/978-3-99161-057-1-143The data-driven approach to vibration-based Structural Health Monitoring aims to detect anomalies in the monitored modal properties. A key step in this framework is compensating for the normal variability in the data, which is due to the strong influence of environmental and operational variables on the structure’s dynamic behavior. The decision-making process is then formulated as a binary classification problem, supported by an appropriate alarm threshold to distinguish between normal and anomalous structural conditions. The threshold is typically set on the statistical distribution of the novelty index computed during the training phase, often assuming a Gaussian distribution of the data. However, anomaly detection requires a more refined modeling of the distribution tails. The present paper investigates the use of Extreme Value Theory for threshold setting, focusing on the Block Maxima sampling technique and the Generalized Extreme Value distribution. A comparison with conventional approaches demonstrates the significant accuracy achievable through the extreme value theory. The natural frequency time histories of the KW51 bridge are used as benchmark data to highlight the method’s effectiveness in improving the reliability of early damage detection.
• A Comprehensive Approach for Vision-Based Dynamic Monitoring of Structures and InfrastructurePonsi, Federico; Eslami Varzaneh, Ghita; Ghirelli, Giorgia; Bassoli, Elisa; Vincenzi, Loris; 10.3217/978-3-99161-057-1-144Structural monitoring is crucial for extending the service life of civil structures. Vibration-based monitoring is widely employed across various applications, leveraging both traditional and innovative sensing technologies. Among these, video-based methods have emerged as a promising and cost-effective approach for evaluating structural displacements at critical points. This paper presents a novel vision-based procedure enabling accurate three-dimensional structural displacement measurement using only a single camera. The method applies to assessing dynamic effects on bridges subjected to dynamic loads. The algorithm extracts displacements by tracking predefined targets over time. Special attention is given to reconstructing small 3D displacements from videos that inherently capture two-dimensional projections of the scene. The procedure is validated through experiments on a steel frame in a controlled environment, comparing displacement time histories with imposed vibrations from a shaking table. The originality of this work lies in achieving accurate 3D measurements with minimal equipment, offering a practical and innovative solution for structural health monitoring.
• Model Updating and Damage Detection for Bridge Integrity ManagementTemur, Eray; Limongelli, Maria Pina; Piscini, Andrea; Troielli, Edoardo; 10.3217/978-3-99161-057-1-145The integrity management of bridges is crucial for ensuring public safety and economic stability. In practice, Structural Health Monitoring data recorded during bridge operation is increasingly used to guide maintenance decisions. However, incorporating structural damage information more effectively can lead to optimal strategies for integrity management. In this study, we employ Bayesian Model Updating to develop a more reliable structural model. The updated finite element model is then used to train a variational autoencoder-based surrogate model for damage detection, localization, and severity estimation. The variational autoencoder model establishes a link between damage-related features and the modal properties derived from SHM data. Damage information supports maintenance decision-making through a predefined decision rule.
• On a data compression technique for acceleration signals from a railway bridgeYadav, Pranav; Gupta, Vaibhav; Saravanan, U.; 10.3217/978-3-99161-057-1-146Structural health monitoring (SHM) is essential for ensuring bridge safety and longevity. Under dynamic loads, such as train traffic, acceleration data from sensors offers valuable insights into the condition of the structure. Vehicle bridge interaction models required to predict the acceleration time histories involve numerous parameters for rail traffic. Also, model-based methods have a trade-off between high-fidelity, computationally intensive, and less accurate models. To overcome these limitations, this study introduces a deep learning (DL) algorithm to identify changes in the bridge. However, large datasets resulting from high-frequency sampling and long observation periods pose computational challenges as the train passes over the bridge. To address this, down sampling is employed, reducing data complexity while preserving essential features of the signal. The approach is demonstrated using acceleration data recorded at a node point of a railway truss bridge during train passage. An Autoencoder is employed, compressing high-dimensional data into a low-dimensional latent space, and a deep neural network (DNN) is applied to the latent space, incorporating a measurement loss function to estimate the system parameters. This framework ensures computational efficiency and data integrity, enabling precise system parameter estimation and showcasing its effectiveness in real-life bridge SHM.
• Intelligent Imaging: Transforming Concrete Assessment Methods with AIAhmad, Afaq; Plevris, Vagelis; Ullah, M.; Mir, Junaid; 10.3217/978-3-99161-057-1-147This study uses novel image processing techniques to explore the effects of Cement Replacement Materials (CRM) like silica fume and fly ash on concrete’s microstructure and durability. Cylindrical concrete specimens were prepared with mixed ratios of 1:2:4 and 1:3:6, incorporating water-cement (W/C) ratios of 0.4, 0.5, and 0.6 and CRM levels of 0%, 15%, and 25%. Images captured at various cylinder heights were analyzed using rectangle and nearest neighboring methods to quantify aggregate distribution and air void characteristics, including area, size, and spacing. Validation against manual measurements showed an error rate of less than 1.9%, underscoring the accuracy of these techniques. Results indicated that increasing CRM content reduced air void proportion and size, indicating improved durability. Additionally, CRM increased concrete homogeneity, with 25% of CRM samples exhibiting the lowest coefficient of variation (Cv) values (0.29–0.37), compared to higher Cv values (0.41–0.57) in non-CRM mixes. These findings highlight CRM’s potential to enhance concrete mix design for better structural performance and sustainability in construction applications.
• Monitoring of Non-Linearities in Fatigue Degradation of Metallic Materials Using Techniques beyond Stress and StrainBoller, Christian; Starke, Peter; 10.3217/978-3-99161-057-1-148Traditional fatigue assessment in metals is based on load sequences either measured or assumed, S-N (Wöhler) curves and the application of linearized damage accumulation rules. This requires a large amount of experimental effort to obtain materials data to be used for prognostics of which the result is often unsatisfactory. Furthermore, such assessment is mainly based on stress and strain as the loading parameters applied. However, materials fatigue degradation is a more complex process, far from being linear and not limited to stress and strain only. Material’s degradation is an issue. Without knowing a material’s prior loading history its degree of degradation can neither be assessed nor monitored on this basis. However, monitoring a material’s degree of degradation is a prerequisite to preserve a structure’s health over its Residual Useful Life (RUL). Available Non-Destructive Testing (NDT) techniques can be of a significant help. This paper shows how a metallic material’s non-linear fatigue behaviour can be visualized in a 3D plot characterizing the loading applied as an input parameter, the NDT parameter recorded as a material response and the relative fatigue life, hence the degree of degradation, as a resulting parameter respectively. It is shown how this resulting 3D viewing plane can be used to determine a material’s degree of fatigue degradation at virtually any stage of its operational life and it is demonstrated how this information can be used for a monitoring system in the sense of Structural Health Monitoring (SHM) to further track a structure’s RUL in a much more precise way than traditionally done so far.
• Towards structural health monitoring of clay-printed structuresVollmert, Jasper; Peralta, Patricia; Alatassi, Adel; Chmelnizkij, Alexander; Smarsly, Kay; 10.3217/978-3-99161-057-1-149Structural health monitoring (SHM) is a well-established practice to ensure safety and reliability of civil structures. With the increasing demand for environmentally responsible construction practices and the need to reduce the carbon footprint of construction projects, sustainable materials, such as clay, are gaining attention. Clay-printed structures introduce a novel domain to SHM that requires adaptations of established SHM strategies. Research on SHM strategies devised for clay-printed structures remains scarce, leaving a critical gap in understanding the long-term performance of clay-printed structures. Serving as a foundation for developing SHM strategies for clay-printed structures, this paper proposes a methodology to experimentally determine the structural behavior of clay-printed structures, including buckling, shrinkage, and load-bearing capacity, while identifying key factors critical for developing SHM strategies. The methodology proposed in this study incorporates condition assessment, constraint definition, design optimization, prototyping, and SHM strategy definition. The methodology is implemented for a wall component to experimentally determine shrinkage. Based on the structural behavior of the wall component, an SHM strategy is proposed that essentially consists of selecting appropriate SHM techniques, defining sensor placement, and establishing decision-making criteria. The results demonstrate the feasibility of constructing structurally stable clay-printed structures and provide key insights into SHM strategies for clay-printed structures, advancing sustainable construction practices.
• Icelandic turf houses: A one-year monitoring overviewTeeter, Kathryn Ann; Marteinsson, Björn; Kristófersdóttir, Ágústa; Sigurðardóttir, Alma; Sigurðardóttir, Dórótea H.; 10.3217/978-3-99161-057-1-150Structural Health Monitoring (SHM) is employed to provide insights and conservation guidelines for Icelandic turf houses through periodic and continuous monitoring of hygrothermal and geometrical parameters. The turf houses are unique vernacular structures that were the primary dwelling form in Iceland from settlement (~ 874 A.D.) through the 20th century. These structures, now maintained predominantly by the National Museum of Iceland, represent a cultural legacy but suffer from limited research on their materials, structural behavior, and long-term maintenance requirements. Key conservation challenges include water leakage, differential settlement, and geometrical distortion, highlighting the need to understand the hygrothermal performance of the turf roofs and the structural behavior of the timber frames carrying the roofs. To address these issues, a remote and unobtrusive SHM system was designed, tailored to the constraints of heritage buildings. This paper provides a one-year overview of monitoring at two sites: Keldur farm in southern Iceland and Laufás in the north. The study details the monitoring strategy, system design, installation, and operation, presenting results such as 3D point scans, hygrothermal data, and comparative analyses between the two sites. Findings from the study contribute to understanding the behavior of Icelandic turf houses, offering insights for their long-term conservation and ongoing management.
• Gas permeability under varying laboratory conditionsGrba, Damian; Zimmermann, Thomas; Strauss, Alfred; 10.3217/978-3-99161-057-1-151Durability is a critical factor in the long-term performance of concrete structures. Ensuring adequate surface quality – and thus extending the service life of concrete elements – relies significantly on proper curing during the hydration phase. To assess and monitor the effectiveness of the curing process, suitable testing methods are essential. Among these, gas permeability testing provides valuable insights into the porosity of the concrete, which directly affects its durability. This study investigates the influence of various curing methods on gas permeability using both laboratory and field testing. The results demonstrate that insufficient curing leads to higher gas permeability in the near-surface zone of concrete. Concrete specimens and structural components made from identical mix designs but subjected to different curing conditions are analyzed and compared. The findings also take into account the influence of concrete composition and environmental conditions on surface quality. Based on these insights, the study offers recommendations for a reliable evaluation of curing effectiveness through gas permeability as an indicator of surface integrity.
• Detailed material testing of adobe structures to complete a comprehensive SHM approach that includes laser scanning and ambient vibration studiesTakhirov, Shakhzod; Ergashev, Zukhritdin; Rakhmonov, Bakhodir; Gilani, Amir; Akhmedov, Mirzokhid; Shamansurov, Ravshan; 10.3217/978-3-99161-057-1-152The research team, comprising international and local experts, has been studying heritage adobe buildings in Uzbekistan for several years. A few field studies were conducted at the heritage sites of Uzbekistan and Karakalpakstan: Toprak-Kala, Chadra Hauli, Ulli Hovli, and many others. As the first step of the comprehensive research strategy, laser scanning was used to generate accurate digital twins of the heritage structures. A non-destructive ambient vibration study of selected structures was conducted in the second step. This in-situ testing measures resonant frequencies and mechanical properties at very small excitations, which is insufficient for accurate numerical modeling. This study was conducted to address this shortcoming. A few adobe structures constructed of pakhsa were selected. Since the pakhsa is made of clay, which is available in the vicinity of the construction site, the exact georeferencing of each structure was considered. A few cylindrical samples were bored out from the walls of the structures and tested at the University of California, Berkeley. The samples were instrumented with strain gages, and they were investigated in compression and split tests. As a result of this study, Young’s modulus, Poisson’s ratio, and the strength of each test specimen were measured. These parameters will be used to generate more accurate numerical models of the structures and assess the advantages of reinforcement strategies for heritage structures.
• Redundant Monitoring Strategies for Structural and Geohazard Assessment Using Wireless Tiltmeters and LiDAR on Linked Highway Bridges in ColombiaRestrepo, Víctor; Salazar, Héctor; Piedrahita, Jean; 10.3217/978-3-99161-057-1-153The bridge system composed of Los Grillos, Puente Nuevo, and Chorro Blanco, located along the roadway connecting Sogamoso and Aguazul, near the municipality of Pajarito (Colombia), is founded on an active large-scale landslide in shale bedrock. This mass movement exhibits variable displacement rates depending on rainfall frequency over time. On August 20, 2023, the Los Grillos Bridge collapsed as a result of cumulative ground displacements that compromised its foundations and piers. Between July and December 2024, an integrated monitoring system was implemented, combining observations from an Automated Total Station (ATS), distance measurements using LiDAR, and tilt data obtained from inclinometers. The primary objective of this system is to establish correlations and track both ground and structural displacements, thereby supporting local stakeholders and decision-makers in the operational management of the remaining bridges still in service for civilian traffic. This paper presents the principal findings and illustrates how the integration of data from multiple sensor technologies enhanced the understanding of differential behavior between the ground and the structures. The analysis includes the collapsed bridge as well as the two remaining bridges in the affected area, providing timely and valuable information to support safe roadway operations.
• Optimizing Bridge Recalculation: Uncertainty in SHM-Based Recalculation of Prestressed Concrete BridgesWalker, Maria; Eisermann, Cedric; Bartels, Jan-Hauke; Marx, Steffen; 10.3217/978-3-99161-057-1-154There are several ways to incorporate SHM data into the structural assessment of existing bridges. Beyond conventional model calibration, SHM can improve environmental effects and load estimates, thereby reducing the model uncertainty. However, the measurement data itself is also affected by epistemic uncertainty. This paper investigates the influence of selected data quality characteristics on the recalculation of prestressed concrete bridges, focusing on the example of coupling joints. A research bridge serves as a case study, equipped with temperature sensors recording data since February 2024 until today. A numerical FE model of the bridge provides a solid basis for simulations. A sensitivity analysis was carried out to identify the key parameters influencing the results. This includes the effect of the temperature gradient on the fatigue stress of the coupling joint. The study demonstrates the impact of representativeness and coverage of measurements in a spatial and temporal context on the estimated remaining service life of the structure. It highlights the importance of the correct selection of the sensor number and placement, and of the data collection period length. The results confirm the suitability of the proposed methodology for the systematic evaluation of monitoring concepts. However, further research is needed to derive specific recommendations for the design of monitoring systems for coupling joints. This work contributes to optimized SHM-based bridge recalculation by providing a basis for assessing the quality of monitoring concepts and its influence on structural analysis.
• Hangar Stressing on the 6th Street Viaduct Replacement, Los Angeles, CAThurlow, Paul; Estrada, Sergio; 10.3217/978-3-99161-057-1-155Spanning across Freeway 101, several rail roads and the LA river, the 6th Street Viaduct replacement project is one of the largest bridge projects in the City of Los Angeles. The original bridge was built in 1932 and became a backdrop to the film industry. The iconic bridge was demolished following the decision to replace it with the existing 6th Street Viaduct due to the structure becoming seismically vulnerable. At 3,060-ft-long and 100-ft-wide, the redevelopment of the new bridge – designed by Michael Maltzan, – includes 10 network arch spans, with a total of 388 hangers supporting the bridge deck. The bridge spans 101 Highway, the Los Angeles River. The hanger installation and stressing for the bridge was a complex procedure that would need careful attention to detail for loading the hangars before removal of formwork and for fine tuning the final load criteria. There were 18 load sequences per arch. The instrumentation and monitoring of the hangars while loading required a novel approach that started two years in advance of the works with development of a bespoke system, calibration and acceptance. During installation and works many lessons were learnt by all parties involved. The close working relationship with a desire to succeed between the site team and designers was as fascinating as the technical brilliance applied by all to deliver this section of the project in a safe manner, within the schedule and to the specification.
• SHM Application in Development of New Live Load Distribution factors for Timber BridgesLimaye, Vidyadhar; Vickers, Philip; Hoseinpour, Cyrus H.; Memon, Amjad; Clarke, Justin; 10.3217/978-3-99161-057-1-156The Nova Scotia Department of Public Works (NSDPW), based on their observations and research, believe that the simplified method of analysis of timber bridges included in the Canadian Highway Bridge Design Code (CHBDC) yields excessively conservative load ratings for typical timber girder bridges in the province. With over 2000 such bridges in their inventory, this study looks to improve conservative load ratings by developing more realistic live load distribution factors for timber bridges based on load testing and analytical work. NSDPW engaged SHM Canada to conduct load testing of selected timber bridges and to develop a new regime of realistic live load distribution factors derived from the test data. Six timber bridges of various characteristics were selected for this study to cover as wide a range of bridge configurations as possible in the provincial inventory. The collected displacement and strain measurement data were analyzed and used to calibrate a large number of analytical models and followed by statistical and mathematical formulation of the proposed simplified method. The new method incorporates distribution factors specifically developed for timber bridges, by taking into consideration various parameters such as span length, girder spacing, and mechanical properties of the girders, to offer a fast, reliable, and cost-effective approach for evaluation and management of the province’s timber bridge inventory.
• Full-Scale Bridge Testing: Lessons from the Demolition of the Steinavötn BridgeSigurðardóttir, Dórótea H.; Tsai, Ching Yi; Guðmundsson, Guðmundur Valur; 10.3217/978-3-99161-057-1-157Structural engineers, particularly bridge engineers, rarely have the opportunity to test full-scale bridges to failure due to the high costs involved. Such experiments are typically conducted on scaled-down specimens in controlled settings, which can introduce challenges in accurately correlating results to real-world full-scale behavior. In 2019, a unique opportunity arose when the Steinavötn bridge in southern Iceland was irreparably damaged by flooding. Built in 1964, the 102-meter-long reinforced concrete continuous beam bridge had two abutments and five piers and was part of Iceland’s national road system, connecting the ring road around the island. One of its piers suffered scour damage beyond repair, leading to the decision to demolish the structure. Before its demolition, a measurement campaign was conducted using long-gauge fiber-optic sensors to capture the bridge’s response to ultimate loading. This study presents the findings from the measurement campaign and bridge modelling, providing valuable insights into the ultimate load behavior of a full-scale bridge and advancing the understanding of structural performance under extreme conditions.
• Advancing high-fidelity Digital Twin Technology for Structural Health MonitoringAntonau, Ihar; Warnakulasuriya, Suneth; Ansari, Talhah; Wüchner, Roland; Löhner, Rainald; Nicolas Airaudo, Facundo; Antil, Habir; 10.3217/978-3-99161-057-1-158This study addresses the need for high-fidelity system identification in Digital Twin (DT) applications for Structural Health Monitoring (SHM). As infrastructure ages, its material properties may degrade due to various factors, including damage, corrosion, and fatigue. Accurate assessment of material properties is critical for ensuring safety and reliability. High-fidelity identification enables the detection of localized damages that traditional methods may not detect, directly impacting maintenance strategies and public safety. In this work, we present a formulation of the optimization problem that minimizes errors between observed and simulated displacements by varying material properties. Additionally, we utilize adjoint-based sensitivity analysis, combined with regularization techniques such as Vertex Morphing, to enhance the efficiency and robustness of the optimization process. Our case studies, which include detailed analyses of 2D and 3D structures using real-world data, demonstrate the effectiveness of our methods in accurately inferring material properties and revealing structural integrity. By implementing this advanced methodology, practitioners can achieve timely and accurate assessments of structural integrity, leading to better-informed decision-making regarding maintenance and safety protocols. This research contributes to the ongoing advancement of Digital Twin technology, promoting safer and more efficient infrastructure management.
• Advancing High Fidelity Finite Element Model Updating Using Cooperative Game Theory: A Novel Framework for Structural OptimizationEreiz, Suzana; Duvnjak, Ivan; Jiménez-Alonso, Javier Fernando; Bartolac, Marko; Košćak, Janko; Damjanović, Domagoj; 10.3217/978-3-99161-057-1-159High fidelity finite element model updating plays a critical role in ensuring the accuracy and reliability of structural models for complex infrastructure systems. This study focuses on the application of a cooperative game theory model to update high fidelity finite element model of a pedestrian suspension bridge. By treating the model updating process as a cooperative game model, game theory provides a novel framework for distributing and balancing multiple objectives inherent in this process. The proposed approach is compared with conventional finite element model updating methods to assess its efficiency, accuracy, and robustness. Key performance indicators, such as the reduction in discrepancy between experimental and numerical modal parameters and computational efficiency are evaluated. The cooperative game theory framework is shown to enable an optimized and balanced resolution of conflicting requirements in high fidelity model updating, resulting in improved alignment with observed structural behavior. The primary objective of this research is to demonstrate the potential of game theory as an innovative and effective tool for solving optimization problems in high fidelity FE model updating. The findings are expected to contribute to advancements in structural health monitoring by providing a robust methodology for enhancing the reliability of numerical models.
• Efficacy of decoupling techniques to extract the static strain response from the dynamic response of a bridge under a moving vehicle using a low pass filterSarath, R.; Saravanan, U.; 10.3217/978-3-99161-057-1-160Structural Health Monitoring of bridges is being used increasingly to ensure safe operation of bridges. Non-iterative and mechanics-based algorithms that were developed in the past to find the material property of a bridge or the live load moving over the bridge use the static strain response of the bridge. However, the field strain measurement of these response quantities has both static and dynamic components. To apply these non-iterative methods for live load or material property estimation, it is important to decouple the static components of the strain from its dynamic components. Hence, in the current study, the dynamic components of the bridge strain response are filtered to extract the static components using a low-pass filter. The adequacy of filtering is then measured based on the probability of the static maximum axial strain and average shear strain contained in the probabilistically determined dynamic response corresponding to different road roughness. The idea of relating the cutoff frequency to the bridge natural frequency is investigated. It is concluded that using a cutoff frequency of half the bridge natural frequency, one can sufficiently filter out the dynamic components under any case of vehicle speed, road roughness, and bridge natural frequency.
• Laboratory Testing of Old Bridge Girders: Preliminary ResultsKreslin, Maja; Kosič, Mirko; Šajna, Aljoša; Anžlin, Andrej; Hekič, Doron; Požonec, Vladimir; Triller, Petra; 10.3217/978-3-99161-057-1-161This paper presents key results on the laboratory testing of old girders removed from a flood-damaged bridge located near Ljubljana, Slovenia. The structure was widened in 1989 to accommodate pedestrians and cyclists by integrating prefabricated prestressed reinforced concrete T-girders. To assess the structural behaviour of the bridge, six girders were subjected to a rigorous testing program involving bending and shear tests in a laboratory setting. The tests were performed on girders with static lengths of 12.20 meters and 9.90 meters. The program aimed to evaluate the structural performance of the girders. Preliminary results indicate satisfactory structural behaviour of the prestressed T-girders under the applied loads, with insights into their performance under both bending and shear stresses. This study contributes valuable data for assessing the long-term behaviour of bridges. The outcomes are particularly relevant for optimising resource allocation in bridge rehabilitation projects and ensuring safety and functionality in transportation networks.
• Investigation of the causes of the unusual gap between the bridge and abutment using long-term monitoringIwabuchi, Sunao; Takehara, Tomohisa; Miyamori, Yasunori; Kadota, Takanori; Hinata, Yoichi; Oshima, Toshiyuki; 10.3217/978-3-99161-057-1-162Unusual gaps at the ends of bridge girders can occur due to the displacement of abutments or insufficient initial gaps during construction. However, when such irregularities arise, it is difficult to estimate the cause through visual inspection alone. As analyses on the causes of these gaps and their expansion are needed, investigations into appropriate remedial measures have become important. For these analyses, we developed a device called the Expansion-Gap Measurement System (EGMS), which enables the causes of unusual gaps to be estimated based on continuous monitoring for temperature-related changes in gap distance. By capturing seasonal variations in gap distance, the system allows us to estimate the progression in abutment movement and the causes of damage. This paper presents the case study of a simple bridge made with H-shaped steel girders with unusual gaps where the EGMS was installed for approximately three years to investigate the causes of a gap. It was found that during the high-temperature summer months, the gap closed completely, whereas in winter, the gap remained. Furthermore, it was confirmed that the abutment saw no lateral movement during the winter, indicating the minimal progression of displacement. Based on these findings, the cause of the unusual gap was attributed to either an error in the abutment’s initial placement during construction or to previous lateral displacement of the abutment.
• Monitoring the dynamic sensitivity of the Solkan footbridge to user-induced excitationKosič, Mirko; Hekič, Doron; Drygala, Izabela Joanna; 10.3217/978-3-99161-057-1-163This paper presents a comprehensive monitoring study of the dynamic sensitivity of the Solkan footbridge in Slovenia, with a focus on its response to user-induced actions. An extensive ambient vibration measurement campaign was carried out, during which 26 triaxial accelerometers were strategically deployed along the bridge to capture its modal characteristics in all three spatial directions. The structure’s dynamic response was monitored under both regular pedestrian traffic and elevated loading conditions during a local marathon event, allowing for the assessment of its behaviour across a broad spectrum of real-world scenarios. Preliminary measurements revealed reduced pedestrian comfort, primarily due to resonance effects resulting from bridge–user interaction. The study highlights the importance of field-based dynamic assessments in diagnosing performance issues and informing mitigation strategies. The findings contribute to the advancement of resilient and dynamically efficient design and maintenance practices for pedestrian bridges.
• InSAR as a Component of Geotechnical Monitoring During Subway Construction in PragueHlaváčová, Ivana; Kolomazník, Jan; Struhár, Juraj; Orlitová, Erika; 10.3217/978-3-99161-057-1-164A new subway line is under construction in Prague, Czechia. The drilling and construction potentially affect the structural health of buildings and infrastructure in the zone of influence. As part of geotechnical monitoring for the Metro Line D construction, subsidence is monitored through satellite interferometry (InSAR). Measurements from TerraSAR-X/PAZ are analysed using a customised PS-InSAR algorithm to capture evolving deformation patterns during construction. In addition, several add-on methods have been developed to provide targeted trend analysis and geotechnically relevant metrics. The retrospective “passportization stage” has been followed by standard monitoring stages with adaptive stage duration and frequency of satellite acquisitions. The nonlinear characteristics of displacement trends present challenges for InSAR. Particularly in X-band data, phase unwrapping errors compromise spatial interpretation, elevate noise levels, and diminish the reliability of results. Issues are addressed by tailored enhancements to the InSAR methodology, including advanced time series segmentation considering statistically significant differences in displacement velocities or noise levels. Validation confirms strong agreement between displacement trends measured through InSAR and conventional geotechnical methods.
• Utilizing PSDefoPAT® to analyze surface deformation of embankment damsEvers, Madeline; Thiele, Antje; 10.3217/978-3-99161-057-1-165Dams are used worldwide to, e.g., manage flooding, generate energy, or secure the freshwater supply. They play an essential role in the local economy. However, an operational or structural failure also poses a significant threat to the environment and the local economy. Therefore, it is vital to ensure their structural health and functionality. In this study, we present the ground surface deformation of the Parapeiros-Peiros Dam during the later stages of its construction, first filling, and shortly afterwards. Since data services, such as the European Ground Motion Service, surfaced and provide freely available ground motion datasets, one might think that in-house processing of SAR data for surface deformation monitoring of critical infrastructure is obsolete. In order to explore the advantages of in-house processing, we compare ground motion datasets generated by Fraunhofer IOSB and the European Ground Motion Service based on advanced differential synthetic aperture radar interferometry techniques. The dataset consists of sets of measuring points, their mean deformation velocity, and the associated displacement time series. Based on the mean velocity maps, we present a spatial analysis of the observed deformation patterns. In addition, we analyzed the temporal deformation pattern of individual measuring points by employing the Persistent Scatterer Deformation Pattern Analysis Tool (PSDefoPAT®). This tool can be used to fully automatically identify the statistically best fitting model to describe the temporal deformation pattern of a persistent or distributed scatterer (i.e., linear, quadratic, piecewise linear, or periodic) and provides insight into the dynamics of the surface deformation. It can aid with the analysis of changes in the structural health of the dam.
• Exploring InSAR Capabilities for Bridge structural health monitoring using TerraSAR-X and Sentinel-1 DataKwapisz, Maciej; Boros, Vazul; Gutjahr, Karl Heinz; Hlvacova, Ivana; Martinez, Cesar; Schlaffer, Stefan; Struhar, Juraj; Vorwagner, Alois; 10.3217/978-3-99161-057-1-166Remote sensing, in particular multi-temporal Synthetic Aperture Radar interferometry (MT-INSAR), is becoming an operational technique for landslide and subsidence monitoring, and it shows significant potential as an effective tool for bridge monitoring as well. In this case study, the possibilities of MT-INSAR-based structural health monitoring were demonstrated on a motorway bridge in Austria. The bridge is a perfect test object to compare the achieved accuracy due to the availability and good coverage of TerraSAR-X and Sentinel-1 data in combination with an in-situ deformation monitoring system. Due to the overlapping period of one year, a statistical evaluation of the obtained deformations along the bridge could be made. Another topic addressed in this contribution is the modelling of typical bridge deformation patterns, which are primarily caused by thermal expansion of the bridge. To detect critical displacement patterns, it is therefore necessary to separate the thermal component from the critical one. After completing this step, we applied and evaluated newly developed algorithms that detect changes in bridge deformation patterns and raise alarms when necessary. Furthermore, an interesting comparison was made between processed Sentinel-1 time series as provided by the Copernicus Land Monitoring Service via the European Ground Motion Service (EGMS) and the custom processing of the area of interest exclusively, utilizing site-specific temperature data.
• Satellite-based InSAR for monitoring and safeguarding high-voltage power pylons amid the energy transitionDörfler, Markus; Keuschnig, Markus; Hartmeyer, Ingo; 10.3217/978-3-99161-057-1-167Resilient high-voltage grids are essential for ensuring energy supply and preventing. However, climate change, the energy transition and the required expansion of electricity grids pose growing challenges for infrastructure operators in the energy sector. In Austria, landslides in alpine regions and decreasing groundwater levels in flat areas represent significant potential risks to power pylons. The Austrian Power Grid AG (APG), which operates the Austrian transmission grid, faces growing demands to detect damage at an early stage and to guarantee grid security amid changing climatic conditions. Satellite-based Interferometric Synthetic Aperture Radar (InSAR) potentially provides millimeter-precise, area-wide monitoring of ground motion and structural deformations. Periodic InSAR data updates (e.g. semi-annually) enable a complete and continuous analysis of all single structures such as power pylons and thus facilitate an assessment of the structural integrity of the entire grid. This allows the early identification of risks such as landslides or structural changes and the implementation of predictive maintenance. This paper highlights previous experiences and future potentials of integrating the InSAR technology into APG's workflows and risk management, which contributes to sustainable planning and increased grid stability in an increasingly complex system.
• Remote Sensing for Stability Assessment of River Bridges: Case Study of the Red River Bridge in Winnipeg, CanadaCusson, Daniel; Stewart, Helen; Regoui, Chaouki; Zhao, Junhui; Helmi, Karim; Telehanic, Boris; Murison, Evangeline; Thomson, Doug; Mufti, Aftab; Clark, Shawn; 10.3217/978-3-99161-057-1-168With Canada's transportation infrastructure aging, compounded by the effects of climate change, the need to enhance condition assessment through structural health monitoring is increasingly critical to ensure integrity, performance, public safety, and cost-effectiveness. Bridge pier scouring, caused by high river flow and turbulence that erode the surrounding bed material, poses a significant threat to bridge stability and can potentially lead to failure. Conventional scouring inspections are often time-consuming and costly. This paper presents a case study on the Red River Bridge in Winnipeg, Canada, where an innovative, multidisciplinary assessment of bridge stability was performed, including both environmental and structural investigations. The environmental investigation utilized multispectral satellite imagery and optical-band unmanned aerial vehicle (UAV) imagery, combined with large-scale particle image velocimetry (LS-PIV), to assess river flow and turbulence. An anomalous condition near a bridge pier, detected in multispectral satellite imagery, was confirmed by UAV photogrammetry and LS-PIV river current patterns. The structural investigation, detailed in this paper, incorporated Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) deformation measurements from satellite imagery, in-situ measurements on the bridge deck, and numerical bridge model predictions. This provided an assessment of the bridge's structural behavior and its potential connection to the condition observed near one of the bridge piers.
• Potential of InSAR for Structural Health Monitoring of Flood Protection SystemsDohnalík, Petr; Boros, Vazul; Kwapisz, Maciej; Leopold, Philip; Vorwagner, Alois; Thiele, Antje; Evers, Madeline; 10.3217/978-3-99161-057-1-169Extreme weather events that cause flooding endanger people, economic goods, and the environment. Flood protection systems, such as dams, dikes, and levees, defend these valuable assets and therefore their structural health should be monitored. The goal of this project is to investigate the potential of Interferometric Synthetic Aperture Radar (InSAR) satellites for the monitoring of dams and dikes. The ability to monitor flood protection systems depends on the availability of a sufficient number of measurement points. This is influenced by several factors, such as the type of the scanned surface area (e.g. vegetation cover, concrete), the orientation of the dike with respect to the satellite’s orbit, the temporal and spatial resolution of the SAR sensor, and the period for which the European Ground Motion Service (EGMS) provides the data. It is also of interest to study the differences between persistent scatterers and distributed scatterers. Furthermore, the correlation between surface subsidence detected by InSAR and the changes in water level, which pinpoint flood events, is also investigated. The use of Corner Reflectors or satellites with higher spatial resolution, are also some of the aspects to be explored in the next steps to investigate how to maximize the potential of InSAR satellites for the monitoring of flood protection systems.
• ISABHEL (Integrated SAtellite and ground-based monitoring for Bridge HEalth Lifetime assessment)Costantini, Vera; Chiaia, Bernardino; Civera, Marco; Ciavattone, Alberto; Ambrosio, Davide; Ranalletta, Carlo; Del Monte, Emanuele; Marini, Roberta; Mazzanti, Paolo; 10.3217/978-3-99161-057-1-170This paper presents ISABHEL (Integrated SAtellite and ground-based monitoring for Bridge HEalth Lifetime assessment) project, which demonstrates an integrated approach to Structural Health Monitoring (SHM) by combining satellite InSAR data, contact sensors, Photomonitoring™, and Finite Element Modeling (FEM). The system is implemented on two bridges over the Po River in Turin, Italy: the Amedeo VIII and the Regina Margherita bridges. Each technology complements the others, providing a comprehensive understanding of bridge behavior. The InSAR analysis using high-resolution COSMO-SkyMed data revealed slight asymmetric deformation in the Regina Margherita Bridge, with the western lane exhibiting higher deformation rates. Contact sensors were strategically designed to be positioned based on each bridge's specific vulnerabilities, with the Amedeo VIII bridge focused on static monitoring and the Regina Margherita bridge on dynamic monitoring. The calibrated FEM models will enable prediction of structural behavior and establish critical thresholds. A web platform integrating all data sources will provide real-time visualization and alerts. This paper presents the initial results of this ongoing project funded by ESA, which will be completed in the next months.
• SGAM – Smart Geotechnical Asset Management: Enhancing predictive maintenance with data-driven insights and Earth Observation technologiesBrunetti, Alessandro; Di Renzo, Maria Elena; Gaeta, Michele; Mazzanti, Paolo; Mastrantoni, Giandomenico; Valerio, Emanuela; 10.3217/978-3-99161-057-1-171Natural hazards such as landslides, subsidence, and liquefaction represent growing threats to critical infrastructure. Building upon the methodological foundation presented in the SGAM project, this work introduces enhancements to the Smart Geotechnical Asset Management (SGAM) framework, with particular emphasis on its hazard assessment component. The SGAM system integrates geotechnical monitoring, Earth Observation (EO) data, and machine learning techniques to support predictive maintenance of linear infrastructure. In this paper, we present methodological refinements, expanded geohazard integration, and new insights from recent applications. A synthesis geospatial layer supports proactive risk mitigation by highlighting high-priority intervention zones. These developments aim to improve data-driven infrastructure management.
• An EMI-based approach for Structural Health Monitoring of a Space Reinforced Concrete Frame StructureSapidis, George M.; Naoum, Maria C.; Papadopoulos, Nikos A.; Voutetaki, Maristella E.; Chalioris, Constantin E.; Rousakis, Theodoros C.; 10.3217/978-3-99161-057-1-172The electro-mechanical impedance (EMI) method represents a promising approach to structural health monitoring (SHM), attributable to its ability to simultaneously employ piezoelectric transducers for both actuation and sensing purposes. As a result, an extensive volume of literature has surfaced recently, analyzing the efficacy of the EMI method in reinforced concrete (RC) structural elements subjected to quasi-static loading sequences. Nevertheless, the investigation into applying the EMI method in dynamic loading environments must be more robust within the current body of research. This study evaluates the effectiveness of the EMI method for SHM of a one-bay, one-story RC space frame structure under the influence of earthquake excitations. Therefore, a shaking table was used to excite the RC frame with progressively increased ground excitation, wherein piezoelectric patches are strategically embedded in meticulously chosen locations. The embedded PZT sensors facilitate promptly diagnosing earthquake-induced damage to the RC frame. The experimental outcomes reveal that the EMI method effectively and expeditiously identified damage formation within the RC frame.
• Tactile Pressure Sensors to analyse Anchor Wall Behaviour in mid-scale ExperimentsSchleicher, Julian; Rebhan, Matthias J.; Daxer, Hans-Peter; Pamminger, Vera; Arnold, André; Tschuchnigg, Franz; 10.3217/978-3-99161-057-1-173Material ageing, such as corrosion of reinforcements or damage to geotechnical tension elements, can lead, among other effects, to load redistribution mechanisms at the interface of structure and the adjacent soil. These mechanisms are difficult to capture, as analytical calculations are not practical and three-dimensional numerical simulations, utilizing advanced constitutive laws require sufficient monitoring data to validate the calculations. Consequently, it is difficult to determine the reliability of damaged structures. To improve existing knowledge in this field, a test setup simulating an anchor wall was developed, to provide calibration data for numerical studies and to evaluate load redistribution mechanisms at the soil-structure interface in general. In this context, newly developed Tactile Pressure Sensors (TPS) were tested for their suitability as an additional monitoring tool, to record changes in compressive stress at the interface and consequently, to complement conventional monitoring devices used for deformation and force measurements. Such TPS can enable the visualization of an approximate compressive stress distribution across the entire interface, utilizing 576 individual sensing elements over an area of 1.50 m² to monitor changes in stress. Thus, additional data is provided to improve the evaluation of load redistribution mechanisms and to be used as validation data for numerical models.
• Long term monitoring of the Balladelaan bridge using smart AggregatesKortendijk, Coen N.; Hoekstra, Anke; Besseling, F.; Yang, Y.; Cheng, H.; 10.3217/978-3-99161-057-1-174The Balladelaan Bridge is a key bridge in the city of Amersfoort. The city has asked us to investigate the construction quality and remaining lifespan of this 5 span statically indeterminate cast-in-place concrete plate bridge, which was built in 1946. For this a pioneering long-term Structural Health Monitoring (SHM) initiative has been implemented in the form of a Field lab employing Smart Aggregates (SA’s) using Coda-Wave Interferometry (CWI), augmented with extensive temperature measurements, and a short-term measurement using Acoustic Emission (AE) sensors. This study represents an extensive application of Smart Aggregates for continuous monitoring of a bridge's structural health over an extended period. The study aims to address the suitability of SA’s for long term monitoring and accounting for environmental influences such as temperature and humidity on the wave speed in the concrete. Preliminary findings demonstrate a significant influence of temperature on wave speed readings, underscoring the necessity of temperature compensation in SHM analysis. After accounting for these environmental influences, the study generates critical insights into the bridge's integrity and performance. The outcomes of this research will not only enhance the understanding of the Balladelaan Bridge's condition but also establish a benchmark for future SHM projects utilizing Smart Aggregates and CWI technology.
• Structural Health Monitoring of Composite Plate using Piezoelectric TransducerHuynh, Thanh-Canh; Koh, Ghan Ghee; 10.3217/978-3-99161-057-1-175This research proposes an approach for detecting damage in composite structures using the cost-effective electromechanical (EM) impedance method. By employing a piezoelectric transducer driven at high frequencies, the technique facilitates localized damage identification by tracking changes in EM impedance signatures that correspond to alterations in the structural properties caused by damage. A numerical model of a composite plate with a surface-mounted piezoelectric patch is constructed to simulate EM impedance responses under different damage conditions within the composite layers. Damage is represented by localized reductions in stiffness, simulating common failure modes such as delamination or matrix cracking. The simulation results indicate that the EM IMPEDANCE response is highly sensitive to damage within the composite plate across a wide frequency range.
• Study on Methods for Identifying and Evaluating Damage to Cross-Sea Bridges Subjected to Ship CollisionsGuo, Jian; Cui, Yuhao; Wang, Zheng; Wu, Jiahui; 10.3217/978-3-99161-057-1-176Cross-sea bridges are crucial transportation links, ensuring smooth maritime traffic and protecting public safety and property. However, ship collisions pose a serious threat, potentially causing extensive damage to bridges, disrupting traffic, polluting the environment, and leading to casualties. Therefore, it is extremely important to develop methods for identifying and evaluating damage to cross-sea bridges caused by ship collisions. A study has been conducted utilizing a combined approach of numerical simulation and experimental validation to analyze the structural dynamic responses of bridges subjected to ship collisions. Based on the insights gained, a structural damage assessment method combining response surface method and Monte Carlo simulation has been introduced. This method takes into account factors such as the impact height and kinetic energy during ship-bridge collisions, establishing a comprehensive evaluation index system. This approach offers a holistic view of the damage state of bridges subjected to ship collisions, providing a scientific foundation for subsequent emergency response and repair strategies. Ultimately, the research aims to mitigate the adverse effects of ship collisions on the structural integrity of cross-sea bridges.
• An automatic system for the rapid post-earthquake safety assessment of bridgesAguero, Marlon; Skolnik, Derek; Ciudad-Real, Mauricio; 10.3217/978-3-99161-057-1-177This paper presents the development of an advanced system for the rapid post-earthquake safety assessment of bridges using advanced sensor technology. Upon completing the assessment, the system generates an automated report on bridge condition. To evaluate the serviceability of bridges, engineers frequently employ sensors, such as accelerometers, strain gauges, and tiltmeters to measure accelerations, displacements, strain, tilt, and deflections. The obtained data are essential for theoretical and practical reasons. Engineers analyze the data to gain insights into real-world bridge dynamics and to develop and validate models that inform design codes. On the practical level, the data are used in structural health monitoring (SHM) systems to enhance public safety by providing reliable data about bridge conditions, both long-term and after unexpected events, such as disasters and earthquakes. Traditionally, bridge inspections are conducted every two years to detect potential deterioration. However, these inspections are expensive and may need to be done more frequently following extreme events like earthquakes, fires, or bridge strikes. To reduce costs, we propose an innovative automatic rapid assessment system that uses measured bridge response data to initiate and minimize re-evaluation efforts. The system works by converting the acceleration data to displacements; subsequently, a threshold that defines the serviceability of the bridge is established. When one of the thresholds is exceeded, a report on bridge condition is automatically generated. This system is particularly useful in post-earthquake events and after other emergencies. In such situations, fast and reliable decision-making is a strong necessity, but also a serious challenge due to common human conditions, such as panic and fear. Rapid, automated generation of reports ensures accurate assessments of damage, which are crucial for the reduction of serious economic losses and the maintenance of reliable infrastructure access. In the paper we will discuss two case studies which illustrate the deployment of automatic, real-time assessment systems. As will be shown, these systems enhance the preparedness for disaster scenario and considerably improve bridge safety.
• Bridge superstructure vibrational analysis as means to detect scour in a medium span bridgeZhao, Junhui; Helmy, Kareem; Telehanic, Boris; Murison, Evangeline; Mufti, Aftab; Thomson, Douglas; 10.3217/978-3-99161-057-1-178Scour around bridge piers is a leading cause of failure but detecting it reliably and cost-effectively remains a challenge. Vibration analysis offers a potential solution by monitoring changes in vibrational mode frequencies, as scour reduces a bridge’s natural frequencies. This study measures traffic-induced vibrations on both bridge piers and decks, enabling continuous monitoring without disrupting traffic. An important consideration in using vehicle-induced vibrations is that each vehicle tends to preferentially excite certain vibrational modes, influenced by factors such as vehicle speed and span length. Therefore, obtaining a representative vibrational spectrum requires the passage of many vehicles. In this work, we found that the passage of 10 to 50 vehicles is necessary to reduce errors to the level required for robust vibrational analysis. Continuous monitoring of vibrational frequencies also enables compensation for seasonal variations in environmental factors such as temperature. In this study, a medium-span bridge was monitored over approximately 10 months. By correlating frequency data with temperature, we can account for environmental influences. After compensating for temperature, frequency changes exceeding ±1.7% can be detected and may indicate the presence of scour. The measured frequencies closely matched predictions from finite element model calculations.
• 2D sonar techniques for monitoring the canal bed morphology of entrances to navigation locksBastani, Mohsen; Kromanis, Rolands; 10.3217/978-3-99161-057-1-179Navigation locks are essential components of inland waterways. They enabling vessels to traverse sections with differing water levels. These structures are increasingly vulnerable to damages caused by (1) scour i.e., erosion of sediment due to natural water flow and ship-induced currents, and (2) sediment or debris obstructing lock gates. Scour-induced damage threatens the structural integrity of locks, leading to costly maintenance, prolonged closures, and economic and environmental consequences. Traditional monitoring methods, including visual inspections and fixed instrumentation, are often hampered by water turbidity, high costs, and susceptibility to debris damage. Sonar technologies provide a non-invasive, cost-effective alternative for detailed underwater imaging, even in challenging environments. This paper explores the application of 2D sonar imaging for monitoring navigation lock approaches, with a focus on bed morphology and scour progression. Using the Prinses Beatrix Lock in the Netherlands as a case study, the paper demonstrates the effectiveness of 2D sonar systems in (i) detecting morphological changes such as scour and sediment transport from bed protection layers and (ii) estimating ship drafts. The findings underscore the importance of integrating sonar-based structural health monitoring systems to extend the lifespan of navigation locks, enhance safety, and optimize maintenance strategies for aging waterway infrastructure.
• On-Line Health Monitoring of Underground Pipelines by Source Localization of Leak DamagesYoon, Dong-Jin; Lee, Sun-Ho; Park, Choon-Su; 10.3217/978-3-99161-057-1-180This study aims to perform leak detection on an old oxygen underground pipeline in operation. The target pipe is a 220-meter section of nominal diameter 80A steel pipe that supplies oxygen for welding in heavy industrial facilities. In order to observe the response characteristics of the coupled vibration propagating through the pipe, an impact experiment was conducted to experimentally derive the frequency band and propagation velocity of the coupled vibration in the tested compressible fluid transportation pipe. Conventional leak detection methods mainly depend on frequency-domain filtering because it is difficult to improve the signal-to-noise ratio through averaging in the time domain due to the random nature of the leak signal. In this study, we propose a leak detection algorithm with improved detection performance by utilizing an ensemble cross-correlation function that applies averaging in the τ-domain based on the deterministic arrival time difference characteristics of the leak signal. In addition, leak detection using two sensors is likely to misjudge the leak source near the sensor due to noise propagating outside the detection range, and a single damage positioning result is insufficient to determine the leak damage. Therefore, this study proposes a distributed measurement-based leak detection technique and a decision map based on multiple damage localization results. The experimental results confirmed that damage localization using coupled vibration is possible in compressible gas transport pipelines, and experimentally verified that leakage can be effectively detected with a location error rate of approximately 1.37% in industrial sites that are always in operation.
• Effects of grout-strand interface modelling on the degradation of external grouted post-tensioning tendonsVecino, Belén; Renedo, Carlos M.C.; Chillitupa-Palomino, Luis; Díaz, Iván M.; 10.3217/978-3-99161-057-1-181In recent years, relevant brittle fractures of external grouted post-tensioning tendons in bridges have been reported due to corrosion damage, compromising the structural safety and stability of the bridges. A previous detailed finite element (FE) modelling approach for grouted tendons has been developed by the authors and compared with experimental results presented in the literature, ensuring an accurate reproduction of experimental results by accounting for steel plasticity and large deformations. This modelling approach considered a bonded contact to model the strand-grout interface. That is, an immediate re-anchoring of the strands in the grout is assumed in case of failure and, consequently the influence of the bond stress-slip behaviour is not considered in the modelling approach. This paper presents an alternative FE modelling strategy where the differences with the previous one are: i) the modelling of the strands as beams instead of solid bodies, ii) the presence of the sheathing duct, and iii) a non-linear model to reproduce the strand-grout bond stress-slip behaviour. The objective is to investigate the influence of different models to define the grout-strand interface: a bonded model or a bond stress-slip model, while validating the author’s previous FE approach. These models are also compared with the experimental results from the literature. Normal stresses along the strands and in the grout are studied, and degradation curves are derived, that is, the effective tensile force and natural frequencies versus damage (defined as the percentage of broken strands). These degradation curves serve as a key performance indicator of the structural performance of the tendon for structural health monitoring systems, anticipating to severe damage and potentially dangerous scenarios.
• Autonomous peak-picking procedure for tension force estimation in cables and external post-tensioning tendonsChillitupa-Palomino, Luis; Renedo, Carlos M.C.; Garcia-Palacios, Jaime H.; Díaz, Iván M.; 10.3217/978-3-99161-057-1-182Cables and external post-tensioning tendons are key elements of modern strategic bridges; however, their structural integrity has been questioned due to some structural collapses registered during the last decades. In this context, cost-effective vibration-based monitoring strategies can be implemented to improve their maintenance by continuously estimating their tension force (a key damage indicator) from their natural frequencies. These frequencies may vary in time due to environmental changes, modification of the service loads, and tensioning processes produced during cable substitution manoeuvres. In monitoring systems that only use one accelerometer per cable (the most common situation in practice), a robust and accurate peak-picking method is required to adequately identify which are their actual almost harmonic natural frequencies and to which modal order they correspond to. Ideally, this method should be automated (to run continuously), autonomous (with as few hyperparameters as possible) and self-regulating (to discard poor quality spectra and outliers). Additionally, the method must be able to cope with two well-known phenomena experienced in practice that dirt cable spectra: i) the double peak effect, and ii) the presence of bracing belts between cables. Thus, this paper works on developing an autonomous peak-picking procedure to cope with the aforementioned phenomena for enabling a reliable tension force estimation method in cable structures. This methodology has been applied to a one-week monitoring data set of measurements of real external post-tensioning tendons of a road bridge in Spain.
• Field monitoring and mitigation for high-mode vortex-induced vibrations of cables in cable-stayed bridgesYan, Yuxuan; Wang, Xin; Hu, Hao; Guan, Hua; Wang, Kai; 10.3217/978-3-99161-057-1-183With the increase in the span of cable-stayed bridges, stay cables tend to become more flexible and longer, which leads to lower damping and structural natural frequency. Therefore, it is more vulnerable to wind load effects. This study aims to investigate the characteristics and mechanisms of vortex-induced vibrations (VIVs) of stay cables based on long-term field monitoring for a sea-crossing cable-stayed bridge. First, a vibration monitoring system with a high sampling frequency for cables was established. On this basis, the time history data of vibration acceleration for cables with different lengths were collected to study the vibration characteristics, such as vibration amplitudes and frequency distributions of cables. For the longer stay cables, the cable vibrations with over-limit acceleration amplitudes were observed in a wind velocity region of 6~9m/s with very high vibration frequencies, and the wind directions that caused the vibrations were almost vertical to the stay cables. The mechanism of cable vibrations was discussed and investigated based on the relationship between wind characteristics parameters and the vibration response of cables. VIV occurred because the control frequencies of the cable vibrations coincided with the high-mode vortex shedding frequencies of the cables. Finally, the effects of different types of dampers on suppressing VIV were compared and investigated.
• Field Application of TFC-based Electromagnetic Sensors for Monitoring Cross-sectional Loss in Tendons of BridgesLee, Joo-Hyung; Park, Kwang-Yeun; Choi, Ji-Young; Lee, Sanho; Joh, Changbin; 10.3217/978-3-99161-057-1-184External tendons in prestressed concrete (PSC) bridges are essential for structural performance and durability. However, internal damage to strands within grouted ducts is difficult to detect through visual inspection. To address this issue, the Korea Institute of Civil Engineering and Building Technology (KICT) developed a non-destructive testing (NDT) sensor based on the Total Flux Change (TFC) principle, derived from Faraday’s law of electromagnetic induction. The TFC-based electromagnetic (EM) sensor was designed for field use, featuring a lightweight, detachable frame and wireless data acquisition system. The sensor generates an alternating magnetic field via a primary coil and measures the induced voltage from a secondary coil. A reduction in strand cross-sectional area alters the magnetic flux, resulting in a measurable variation in the induced voltage. Field tests were conducted on the Jeongneungcheon Overpass in Seoul, targeting external tendons between piers. The sensor successfully measured voltage variations along the tendon length. In tendons with suspected damage, the voltage amplitude showed distinguishable reductions, unlike in undamaged control cases. Results were consistent with previous laboratory experiments on specimens with artificially induced strand loss. The findings confirm that the developed EM sensor enables efficient and intuitive detection of internal tendon damage with practical field applicability, even in actual bridge environments. This study demonstrates the feasibility of the TFC-based EM sensor for practical field applications and highlights its potential for integration into long-term bridge monitoring systems.
• Exploration of edge computing for monitoring a four-story building frame modelTang, Qing-Chen; Kool, Emil; Colmenares, Daniel; Bayane, Imane; Karoumi, Raid; 10.3217/978-3-99161-057-1-185With the rapid development of sensing technologies in vibration-based monitoring systems, various kinds of devices are connected to exchange data with each other in virtue of cloud computing. However, challenges arise when transmitting and processing large volumes of data, particularly due to latency and bandwidth limitations. To address these issues, edge computing has emerged as a promising solution, enabling local data processing to reduce transmission delays and minimize data redundancy. In this paper, the possibility of edge computing on lightweight edge devices is explored including the KRYPTON® CPU data logger and the ESP32-S3 microcontroller. These two monitoring systems, one with accelerometers and the other with strain gauges, are deployed on a four-story building frame model under varying structural mass and damping conditions that affect dynamic properties. Each system autonomously collects and caches data (accelerations and strains) locally using embedded code, enabling reliable, low-latency edge processing. Experimental results demonstrate the systems' ability to detect changes in dynamic behavior, supporting applications in fatigue assessment and damage detection. The proposed approach is scalable to dense sensor networks for large-scale structural health monitoring, where edge computing significantly reduces reliance on cloud infrastructure.
• On the use of 6C seismic station for bending-to-shear and torsional building response assessment.Rossi, Yara Lorena; Guéguen, Philippe; Bernauer, Felix; Chen, Kate Huihsuan; Lin, Chin-Jen; Ku, Chin-Sang; Chen, Yaochieh; 10.3217/978-3-99161-057-1-186To understand the behavior of healthy structures and the changes to their condition over time, it is imperative to perform experimental analysis of existing civil engineering structures. Development of novel sensors measuring rotations enable to directly measure important components of the system parameters, as they provide additional information on the structural response. Long-term monitoring of existing structures using 6C sensors – 3C translation and 3C rotation, enables a more in-depth analysis of the system parameters; frequency, modeshape and damping. Specifically, the translation-based characterization of the torsional mode can be enhanced through direct measurement of the torsional angle, center of torsion and more precise frequency extraction. Furthermore, the distinction between shear and bending can be analysed through the ratio of angle and deflection, instead of using proxies like frequency ratios or shear wave velocity. In this study, we analyse how using 6C datasets for structural characterization of high-rise buildings provide further information to understanding the system parameters and their variations over time. We find that the rotational components significantly contribute to the understanding of the vibration behavior and thereby propose to include 6C sensors to enhance the characterization of structures.
• Integration of Seismic Interferometry and System Identification Techniques for Real-Time Structural Health Monitoring: Automated Detection of Shear-Wave Velocity Changes Using Skyscraper Data for ValidationKalkan, Erol; AlHamaydeh, Mohammad; Wen, Weiping; 10.3217/978-3-99161-057-1-187This study presents an integrated approach that combines seismic interferometry and system identification techniques for real-time Structural Health Monitoring (SHM), enabling the automated detection of changes in shear-wave velocity profiles for damage assessment. The methodology is validated using data from a 62-story residential skyscraper in San Francisco, one of the tallest buildings in the Western U.S., equipped with 72 uniaxial accelerometers across 26 floors. The building incorporates advanced structural components, including buckling-restrained braces, outrigger columns, and a tuned liquid damper to mitigate seismic and wind-induced responses. Data from the 2014 M6.0 South Napa and 2018 M4.4 Berkeley earthquakes, as well as ambient vibration recordings, are analyzed to establish baseline dynamic properties, including modal parameters, shear-wave profiles, and wave attenuation. We monitor wave propagation velocities, normal mode frequencies, and intrinsic damping through deconvolution interferometry, enabling real-time identification of structural stiffness changes. Shear-wave travel-time curves from deconvolution show reduced velocities below the 28th floor, coinciding with buckling-restrained braces, while higher velocities are observed above. This integrated methodology offers a robust framework for automated damage detection and real-time structural health assessment, demonstrating the potential to enhance the resilience and safety of high-rise structures during seismic events.
• Experimental assessment of GNSS-smartphone performance in monitoring dynamic motionPsimoulis, Panagiotis; Xue, Chenyu; Jayamanne, Mudalige Oshadee Jayamanne; Li, Guangcai; Geng, Jianghui; 10.3217/978-3-99161-057-1-188The recent advancements of GNSS technology have enabled multi-frequency and multi-GNSS observations even at high-rate measurements (up to 100Hz) with a few-mm to cm-level accuracy, broadening the potentials of GNSS application in monitoring dynamic motion of structures. Furthermore, recent studies have revealed the potential of low-cost consumer-grade GNSS receivers in deformation monitoring of civil engineering structures of even cm-level and indicated that the type of GNSS antenna is the main parameter affecting the quality of the GNSS data. In this study, we investigate the potential of dual-frequency smartphone-based GNSS measurements in monitoring dynamic motion of structures. The study is based on controlled experiments of static, slow and dynamic motion of various amplitude and motion frequency, where 1-Hz dual frequency GNSS smartphone measurements are assessed against more accurate geodetic measurements (GNSS and/or Robotic Total Station). The preliminary results show that the GNSS smartphone measurements may suffer from several cycle slips and strong multipath effects, due to the linear polarized GNSS antenna of the smartphone, but in several cases the GNSS smartphone measurements were able to express the dynamic motion. Also, in this study we examine the performance of the GNSS smartphone measurements in monitoring the dynamic response of Wilford Suspension bridge, under various patterns of dynamic loading.
• Identification of Structural Dynamic Loads- From Physical Methods to Physics Informed Deep Learning ParadigmLei, Ying; Liu, Lijun; Zhang, Fubo; 10.3217/978-3-99161-057-1-189In this study, some progresses on the identification of structural dynamic loads are reported. First, a series of improved Kalman filter with unknown inputs developed by the authors for the identification of joint structural dynamic systems and dynamic loads are briefly reviewed. Then, some identification of structural dynamic loads using the physical guided deep learning paradigm are presented, including the identification of multi dynamic load positions and time histories using physics informed and enhanced Generative adversarial neural network (GAN) and Convolutional Long Short-Term Memory (ConvLSTM), respectively, the identification of full-field wind loads on buildings using physical informed recursive convolutional neural network (CNN), and the identification of stochastic fluctuating wind power spectrum on high- rise buildings using physical guided CNN with partial structural responses. The load type of the network during testing can be different from that during training. Through numerical simulation, it is proved that the proposed methods can learn the nonlinear mapping relationship between the structural responses and the external dynamic loads, and can reconstruct the load time histories well. The proposed methods are verified by numerical simulation and the results show that the deep learning methods can identify the unknown multi dynamic load positions and time histories, full-field wind loads on buildings and the stochastic fluctuating wind power spectrum on high-rise buildings.
• Multi-purpose bridge strain data fusion for BWIM and structural monitoringPanigati, Tommaso; Limongelli, Maria Pina; 10.3217/978-3-99161-057-1-190
• Vehicle speed estimation using convoluted reciprocity for bridge structural monitoringCantero, Daniel; 10.3217/978-3-99161-057-1-191The estimation of vehicle speed is a critical first step in deriving vehicle weight from bridge responses. Various strategies have been developed to extract the speed of passing vehicles, primarily relying on sensors that capture signals with features related to the vehicle's axles. These signals are processed through diverse methods; however, existing strategies often fail to perform optimally across different structural configurations. To address these challenges, the convoluted reciprocity (CR) relationship was recently proposed, which was verified numerically and validated experimentally in a laboratory setting. In this document, the novel speed estimation strategy based on CR is applied to an operational bridge using signals from the SCSHM benchmark. The results confirm that CR provides a robust speed estimation method for cases when the signals lack individual axle features.
• Physics-Informed Surrogate Modeling of the SCSHM BenchmarkTemur, Eray; Limongelli, Maria Pina; 10.3217/978-3-99161-057-1-192This study presents a physics-informed surrogate modeling approach for the SCSHM Benchmark bridge using a dual-path LSTM Autoencoder architecture. By combining synthetic data from a finite element model and real strain measurements, the model effectively reconstructs structural responses under moving truck loads. Results show good agreement between predicted and measured strains. Limitations such as the absence of vehicle–structure interaction effects are discussed, with directions for future improvements.
• Vibrational analysis of the benchmark data setThomson, Douglas; 10.3217/978-3-99161-057-1-193Traffic induced vibration is a promising means of continuously monitoring structural behavior. The benchmark data set measures strain at points that will be subject to traffic induced vibration. However, magnitude and frequency spectrum of the induced vibration from an individual vehicle depends on many factors including the vehicle speed and axle weight distribution. Therefore, to obtain a spectrum that is representative the average vehicle induced vibration the vibration from many vehicles must be examined. In this work several methods for the analysis of the strain versus time data to extract traffic induced vibrational spectrums will be compared. Also, the number of vehicles that need to be analyzed to extract a repeatable vibrational spectrum will be examined. Typically, ~40-50 vehicles are needed to obtain a repeatable vibrational spectrum suitable to extract frequency peaks. This approach is used on the benchmark data set and changes in the vibration frequencies due to temperature induced structural changes can easily be observed. The temperature induced structural changes might be the basis training and testing data sets that could be used to evaluate the effectiveness of some damage detection algorithms.
• Study of Semi-Rigid Joints Effect on Global Stiffness of Space Steel Structure Based on Monitoring DataYuan, Cheng; Lu, Wei; Teng, Jun; Hu, Weihua; 10.3217/978-3-99161-057-1-194In conventional structural design and analysis of space steel structures, joints are typically idealized as either perfectly hinged or fully rigid connections. However, actual joint behavior deviates significantly from these idealized assumptions, with joint stiffness exhibiting semi-rigid characteristics that critically affect global structural performance. This discrepancy between simplified joint models and real-world conditions leads to substantial errors in predicting structural stiffness through numerical simulations. This paper presents a novel methodology integrating structure health monitoring with refined finite element (FE) modeling to quantify the semi-rigid joints effect on global stiffness space steel structure. The joint stiffness parameters are inversely identified through stress and deformation monitoring data using Bayesian inference techniques; A multi-scale FE model incorporating semi-rigid joint behavior is developed through component-level validation; The stiffness evolution mechanism is rigorously validated against full-scale monitoring data from the Shenzhen Nanshan Science-Technology Innovation Center's space frame during its service period. Key findings demonstrate that joint flexibility reduces global stiffness by 18-22% compared to rigid-joint assumptions, with stiffness degradation rates showing strong correlation to stress redistribution patterns. The proposed joint-characterization framework provides a physics-based approach for tracking long-term stiffness evolution in space steel structures, offering significant improvements over conventional design methods in both accuracy and predictive capability.
• Sustaining vertical giants: Autonomous monitoring solutions for the construction and lifecycle of tall buildingsŠpiranec, Lidija; 10.3217/978-3-99161-057-1-195As urban environments increasingly shift towards vertical development, the challenges associated with constructing and maintaining tall buildings have intensified. Real-time and dynamic monitoring systems play a vital role in addressing these issues by providing accurate positioning and deformation data. The integration of diverse monitoring technologies during and after the construction of high-rise buildings is crucial for ensuring structural integrity, safety, and efficiency. By combining geodetic and geotechnical monitoring techniques, these systems offer comprehensive insights into building behaviour. The fusion of technologies like GNSS, IoT sensors, and remote sensing, alongside traditional survey methods, ensures precise data acquisition and analysis. This hybrid approach is essential for optimising construction and maintenance processes, reducing costs, and enhancing safety. Furthermore, the ability to process and analyse large volumes of monitoring data efficiently is critical for transforming raw data into actionable insights, aiding decision-makers in understanding the magnitude and direction of structural movements. The successful implementation of these monitoring techniques on iconic high-rise buildings, such as the Burj Khalifa and One World Trade Center, highlights their importance in modern construction and post-construction maintenance. Ultimately, the intelligent use of integrated monitoring technologies supports sustainable and resilient urban development.
• Multi-scale digital twin for a high-rise structure combining ANN and monitoring dataShao, Hetian; Lu, Wei; Zheng, Wenchang; Hu, Weihua; Teng, Jun; Lui, Eric M.; 10.3217/978-3-99161-057-1-196The application of digital twin technology in high-rise buildings provides a comprehensive approach to maintaining construction safety, tracking project advancement, and evaluating service conditions. This paper proposes a novel multi-scale digital twin framework for high-rise structures. The macro-scale model is constructed using spring elements, taking into account the dynamic behavior of flexure-shear coupling in high-rise structures. The macro-scale digital twinning is achieved by updating the macro-scale model through the integration of modal monitoring data with Artificial Neural Networks (ANN). A multi-scale analysis method from the structural macro-level to components of the substructure is developed through information transfer at boundary nodes, achieving a balance between computational efficiency and the demand for accuracy of the local components. Integrated with multiple monitoring data sources, the proposed framework provides a technical pathway for multi-scale model updating, real-time response acquisition, and disaster risk assessment of high-rise structures.
• Assessment method for torsional performance of high-rise buildings based on period ratioZheng, Jiayi; Lu, Wei; Hu, Weihua; Teng, Jun; 10.3217/978-3-99161-057-1-197The torsional performance of a structure significantly impacts the safety and service life of high-rise buildings. Due to deviations between real structures and design models, it is essential to evaluate the torsional performance of in-service high-rise buildings. This paper proposes a method for evaluating the torsional performance of in-service high-rise buildings based on the measured period ratio. By abstracting the high-rise building as an equivalent cantilever beam model with unidirectional eccentricity, the free vibration equation of the structure is derived, and the relationship between the period ratio, stiffness ratio, eccentricity, and radius of gyration is analyzed. The results indicate that changes in the period ratio can reflect the torsional performance of the structure. Based on the Latin Hypercube Sampling (LHS) and Kernel Density Estimation (KDE) methods, the probability density function and cumulative distribution function of the period ratio are established, and a four-level classification method for torsional performance is proposed. Application to a 40-story high-rise building validates the method. The research results provide a new theoretical basis and practical guidance for evaluating the torsional performance of in-service high-rise buildings.
• Digitalization of existing measurement equipment as a valid basis for monitoring and structural behaviorBurtscher, Stefan; Huber, Peter; Tutschku, Morris; Schuch, Markus A.; Scharinger, Florian; Rebhan, Matthias J.; 10.3217/978-3-99161-057-1-198To determine damage and its effects, dense time series data is required, along with information from other sources like temperature changes that influence the main damage parameter and the structure itself. This allows the assessment of structural behavior, separating periodic and temperature-related effects from damage and ageing-related changes in load-bearing capacity. However, existing monitoring systems often lack proper documentation on measured values and their limits. Analog systems, suitable for early service life monitoring, provide readings at long intervals (years). Poor accessibility to remote measuring points further limits comprehensive time series data, including temperature correlations and other environmental correlations. This article presents an approach that can be used to digitize different types of sensors and measuring devices in order to enable the autonomous and continuous generation of measurement data. The examples range from displacement transducers to force measuring devices, which were already installed in analogue form on existing civil engineering structures. The aim is to use digitalization to demonstrate a simple and cost-effective approach on using existing measurement technology as an initial basis for giving a statement about the behavior of the structure, its state of preservation and thus, in addition to supplementing the inspection process, also serve as a starting point for further monitoring.
• Increasing the value of bridge SHM data by leveraging network-level open dataFidler, Paul R.A.; Cocking, Sam; Huseynov, Farhad; Bravo Haro, Miguel; Ubeda Luengo, Pedro; Middleton, Campbell R.; Schooling, Jennifer M.; 10.3217/978-3-99161-057-1-199Installing and maintaining structural health monitoring (SHM) systems on infrastructure assets can be expensive. These systems may produce large volumes of data that require processing and interpretation before the behaviour of the asset can be understood and assessed. However, in-depth understanding typically also requires knowledge of asset construction details and loading patterns. These data may be produced and stored using disparate systems, databases, and file types, creating additional challenges for data fusion and interoperability. Additionally, there has been an increasing trend towards public bodies providing access to their data either reactively because of freedom of information requests, or proactively to encourage use by researchers or to allow others to provide innovative products or services using the data in ways not anticipated by those generating and providing them. This paper presents potential strategies to leverage publicly available data from sources such as Network Rail Open Data Feeds, Rail Data Marketplace, OpenRail Data, OpenStreetMap and others, to contextualise and increase the value of SHM data. Data are considered from four instrumented railway bridges in the U.K., each of varying steel, concrete, and masonry construction. This paper presents scenarios by which these data might be used to gain network-level insights into other structures on the network and discusses the current difficulties in achieving this in practice.
• A comprehensive workflow for digitizing and determining condition indicators for bridge and building constructionOlipitz, Michael; Jung, Roland; 10.3217/978-3-99161-057-1-200This article describes a comprehensive workflow in several phases that integrates the latest information and communication technologies and enables significant improvements in the assessment of both our infrastructure structures, especially our bridge structures, and our building structures. For building construction, the workflow includes not only the PERIOD mode (= PM) for the normal situation, but also a so-called RESCUE mode (= RM), which provides significant support for emergency services in the event of natural disasters such as earthquakes. In all applications in infrastructure and building construction, a digital twin is created along the structural axis in Phase I and automatically converted into a BIM model (= BIMUAV). The resulting BIMUAV model forms the basis for documenting the general condition of the construction in Phase II, which is referred to as the level of maintenance (= LOM) and documented using component-specific damage catalogs. In both Phases I and II, autonomous multi-agent condition estimation for UAVs and innovative sensor technology (Lidar, GPS, UWB etc.) will be used, the application of which will be demonstrated on a specific bridge project. The anomalies represent performance indicators of the components or structure and are categorized according to component-specific damage catalogs, which also determine the respective degree of damage. The classification of anomalies into damage classes is automated using neural networks or AI. In the infrastructure sector, the algorithm in Phase III enables the asset management of bridge maintainers to conduct real-time condition analyses, service life predictions and estimates of the scope of upcoming refurbishment work using real monitoring data. In building construction, the archiving the LOM in the BIM model carried out in the PM in Phase II represents immense added value for the real estate. The rescue mode (RM) is specifically designed for emergency services and, based on simplified dynamic models in Phase III, enables rapid decision-making support for emergency services on site.
• Principles and Case Study of IMSGeo: Automatic Displacement Monitoring System for Construction SitesZaczek-Peplinska, Janina; Kowalska, Elżbieta Maria; Saloni, Lech; 10.3217/978-3-99161-057-1-201Displacement monitoring is a crucial aspect of the construction process, spanning all its stages. Surveying the changes occurring in the structure and its surroundings according to a suitable schedule is fundamental to ensuring work safety and mitigating investment risks. In this article, we highlight the distinctive features of the IMSGeo system, developed jointly by GEOalpin Ltd. and the Warsaw University of Technology (Department of Engineering Geodesy and Measuring Systems). The innovative solutions proposed in the system are characterized by the following integrated features: utilization of advanced adjustment algorithms within a cohesive system, adjustment of a multi-station network, analysis of reference system stability as an integral component of each measurement epoch, reflectless measurement of surfaces and structural elements of objects, presented as a unified 2D or 3D entity, capability to position measuring instruments (motorized/robotic total station) without the need for additional monitoring devices to ensure station stability, implementation of a fully mobile WEB platform for the presentation, interpretation, comprehensive analysis, and archiving of geodata, use of Internet cloud computing for data collection, analysis, presentation, and distribution of monitoring results, ensuring independence from local server infrastructure, user platform functionality designed based on survey research conducted among investors, contractors, inspectors, building supervision representatives, and property managers. The IMSGeo system does not require additional capital investments in infrastructure from investors or contractors and is highly available and scalable. The practical section of the article introduces the IMSGeo system's WEB platform and its implementation on a selected site in Warsaw, Poland.