Accurate differentiation between damage-related anomalies and data errors is a critical challenge in bridge monitoring. This paper presents a data-driven framework for anomaly detection and classification, addressing the question: How can anomalies be classified in multi-sensor bridge monitoring to distinguish structural changes from noise? The framework combines an adapted anomaly taxonomy with a deep neural network trained on synthetic data. It is validated using long-term monitoring data from a railway bridge, incorporating strain gauges, accelerometers, and an inclinometer. In offline training, the model achieves high precision, recall, and F1-scores, effectively detecting anomaly classes across sensor types. For online prediction, it provides anomaly type percentages and visualizations over daily, weekly, and annual timeframes, distinguishing frequent noise-related anomalies from rare anomalies signaling structural changes. Requiring one month of training data, the framework delivers a scalable solution for bridge monitoring and lays the groundwork for future self-learning anomaly detection in infrastructure management.

The bridge network is progressively aging, with an alarming proportion of bridges over 100 years. This situation engenders substantial risks to the overall reliability of transportation networks, requiring innovative methods for efficient management. Monitoring can provide a direct source of information about structural behavior generating alerts when changes occur. Real-time alerts enable effective infrastructure management and decision-making during damage or anomalous situations. However, monitoring can result in a large amount of data that is often difficult to convert into valuable information in real time. This paper presents an approach for real-time detection of abrupt damage occurrence in bridges using unsupervised anomaly detection algorithms and strain/acceleration measurements. The approach incorporates the separation of measurements into events having the same loading nature and the construction of three feature matrices based on statistical features, time-frequency features, and wavelet spectrum features. It includes the evaluation of five anomaly detection algorithms including Isolation Forest, One-Class Support Vector Machine, Robust Random Cut Forest, Local Outlier Factor, and Mahalanobis Distance. The approach is illustrated with a case study of a steel-bascule-railway bridge, that has experienced a brittle cracking event during monitoring. Results highlight the robustness of One-Class Support Vector Machine, Isolation Forest, and Local Outlier Factor algorithms in promptly detecting abrupt changes across different features. The separation of strain and acceleration data into loading-based events, coupled with the comparison of previous and new event features, provides robust feature matrices for effective damage detection. Enhanced detection and higher scores are particularly attributed to time-frequency domain features during damage occurrence. The presented approach can be used as a base on how to perform real-time anomaly detection within the context of bridge monitoring.

The trend in recent years to build longer and slender bridges has created new challenges in terms of safety and structural integrity. This study focuses on the effects of windinduced vibrations on the hangers of the Bomarsund Bridge in Åland, Finland. Those close to the mid-span have exhibited significant vibrations, requiring a thorough investigation to understand the response of the bridge to wind action. Computational Fluid Dynamics (CFD) simulations were performed using ANSYS Fluent to calculate the aerodynamic coefficients characterizing the given section (i.e. Strouhal number, drag, shedding frequencies, etc.) and to investigate the vortex-induced vibration (VIV) phenomena at different wind speeds. The results showed inconsistent drag coefficients at varying wind speeds and lower drag for geometries with rounded edges. The study highlighted the significant dependence of the Strouhal number on wind speed, challenging traditional geometry-based estimations. These findings can be used to implement effective measures to mitigate wind-induced vibrations.

Infrastructure owners face challenges in effective maintenance decision-making due to the process's multidisciplinary nature, spanning mathematics to cognitive science. This study delves into enhancing maintenance for complex infrastructure systems, specifically in scenarios where a single primary owner must consider the preferences and requisites of multiple stakeholders. In such unstructured problems characterised by diverse perspectives and potentially conflicting interests in uncertain environments, systematic decision analysis is paramount. In response, this paper proposes a generic conceptual model to outline the primary critical consideration of decision-making through a holistic scheme of the decision problem; it facilitates discovering shared views when defining the decision problem, eliciting objectives early in the decision-making process, and ensuring that selected agendas align with desired outcomes and core values. It provides valuable guidance for decision-modellers in infrastructure rehabilitation. These claims are reinforced with the application of the proposed model to case studies, illustrating its practicality in real-world settings.

Digital inspections methods have gained increasing interest in the civil engineering field for assessing the condition of various structures including buildings and bridges. These methods are used primarily to evaluate the external surface condition of the structure. However, in the case of bridges, their application when inspecting the interior of these structures remains uncommon. This paper presents a case study for inspecting the interior of a concrete box girder bridge, the Strängnäs bridge. The inspection involved gathering data using a commercial drone, creating a photogrammetrical model, detecting and quantifying damages by employing a Convolutional Neural Network (CNN). The analysis successfully detected 60 cm long concrete cracks, a total area of 3.5 m2 leakage and corrosion over 40 cm. The study addressed the difficulties such as insufficient lighting, lack of GPS signal and dust clouds reducing visibility. Despite these obstacles, the study demonstrated the effectiveness of indoor drone-based inspections.

Bridge monitoring is a useful tool for alerting to changes in behavior and supporting informed bridge management decisions. However, the triggering of alerts is often based on predefined thresholds, which results in incomplete anomaly detection, leading to excessive false positives or negatives. This highlights the need for an efficient anomaly detection approach adapted to bridges. This paper presents a supervised classification framework for detecting and classifying anomalies in bridge monitoring data. The framework involves labelling data from existing datasets and training a classification algorithm to detect similar anomalies in new measurements. The focus is on strain measurements labelled using a predefined time series taxonomy, as illustrated by a case study of a bascule railway bridge. The framework demonstrates high accuracy in detecting and classifying anomalies, making it easy to identify their causes. It triggers alerts only when necessary and provides a reliable method for detecting changes in behavior.

Structural health monitoring with accelerometers provides notable benefits over strain gauges, particularly in installation time and cost efficiency. However, effective local damage assessment necessitates access to local stress histories. This paper proposes a methodology that integrates two distinct approaches to identify and predict stress and strain across various bridge locations from sparse monitoring via acceleration data. The proposed model is validated using strain histories and accelerations collected from the composite railway Bryngeån Bridge in Sweden during its in-service conditions. Initially, a deep learning algorithm for sequence data is employed to forecast strain histories from acceleration data gathered across various bridge locations. Subsequently, the local response function method is implemented, utilizing experimental data collected from the bridge and employing localized models of its substructures, allowing predictions of the bridge's local strain. By integrating these methods, the approach enables accurate prediction of stress ranges and cycles for critical non-instrumented parts, minimizing the need for extensive direct instrumentation and providing a cost-effective, efficient solution for operational structural health monitoring.

This paper proposes an innovative approach for detecting and quantifying concrete cracks using an adaptive threshold method based on Median Absolute Deviation (MAD) in images. The technique applies limited pre-processing steps and then dynamically determines a threshold adapted for each sub-image depending on the greyscale distribution of the pixels, resulting in tailored crack segmentation. The edges of the crack are obtained using the Laplace edge detection method, and the width of the crack is obtained for each centreline point. The method’s performance is measured using the Probability of Detection (POD) curves as a function of the actual crack size, revealing remarkable capabilities. It was found that the proposed method could detect cracks as narrow as 0.1 mm, with a probability of 94% and 100% for cracks with larger widths. It was also found that the method had higher accuracy, precision, and F2 score values than the Otsu and Niblack methods.

When designing railway bridges, it is of utmost importance to consider the dynamic effects causedby moving train loads. In order to do so, finite element models are constructed and subjected tomoving load models presented in the Eurocodes. Oftentimes, the Eurocodes provide parameter valuesof adequate precision for practical purposes, however for research purposes, more accurate values aresought. In order to obtain these material parameter values; an error domain model falsification(EDMF) procedure is used to perform a model updating of an FE model of the single span steel-concrete composite railway bridge named the Bryngeå Bridge on Botniabanan in northern Sweden.The parameters with uncertain values to be updated were obtained using a sensitivity analysis anddetermined to be the following: modulus of elasticity of the concrete deck (𝐸c), density of the concrete(𝜌c), density of the ballast (𝜌b), longitudinal stiffness contribution of the continuity of the rails pastthe bridge span (𝑘rail), and support viscous damping from the post installed viscous dampers at themovable supports (𝑐support). The model updating using EDMF is applied to a simplified 2D model ofthe bridge, comparing the calculated results with measured data for accelerations, strains in the steelgirders, and beam end rotations at the movable support of the bridge. The results indicate that accurateparameter values are found for the Bryngeå Bridge, and that falsification is an efficient approach toperform model calibration. Resulting distributions imply that the parameter 𝑘rail can be discarded asa material parameter for the analysis of the Bryngeå Bridge.

Utmattning av stålkonstruktioner är ett skadeförlopp som kan delas i sprickinitiering, stabil tillväxt och brott. För broar hanteras detta vanligtvis med så kallade  livslängdsmetoder som täcker alla tre faser. Ett typexempel är nyttjandet av  förbandsklasser som beskriver livslängden i antal cykler för givna spänningsvidder. En  sådan metod ger dock inget direkt samband mellan den fysiska skadans status  (sprickan) och livslängden.

I tillämpningar av metoder för skadetålighet eller sannolikhetsbaserad inspektionsplanering måste skadans (sprickans) storlek uppskattas och bedömmas. Den  klassiska teorin för spricktillväxt baseras på Paris lag, som beskriver ett linjärt samband  mot spänningsintensitetsvidden om båda egenskaperna är logaritmerade. Data för  beskrivning tillväxtfasen är dock begränsade för stål från verkliga broar. Den publikation  som används regelmässigt, den brittiska BS 7910, baseras på sammanfattade tester  utförda på stål från offshore-tillämpningar. Dessa visar på stor spridning och  tillämpbarheten på broar är osäker.

Vid rivningen av järnvägsbron över Söderström sommaren 2019 plockades 20 provkroppar ut för utmattningsprovning. Dessa resultat har redovisats i tidigare publicerade rapporter. Föreliggande rapport visar kompletterande provningar för att uppskatta spricktillväxtparametrarna i Paris lag och tröskelvärdet för spricktillväxt. Totalt 13 provningar för det förra och 8 för det senare har genomförts. Rapporten redovisar provningens genomförande och de statistiska utvärderingarna av resultaten.

För lutningen m = 3 erhölls ett karakteristiskt intercept C_k=4,2×10^-13 vilket kan jämföras med värdet i BS 7910 om 5,2×10^-13. Data från provstavarna visar således på en långsammare spricktillväxt.

Gällande tröskelvärdet för spricktillväxt erhölls en fördelning efter en Bayesiansk uppdatering till ΔK_th ~ LN(220; 93,2) N/mm^3/2. Det karakteristiska värdet beräknades till ΔK_th,k=104 N/mm^3/2. Det senare kan jämföras mot värdet i BS 7910 om 63 N/mm^3/2.

The fatigue of steel structures is a damaging process divided between crack initiation, propagation and fracture. This is typically considered for bridges with so-called total life  methods covering all three phases. An example is using detail categories describing the  service life in the number of cycles for given stress ranges. Such methods do not,  however, give any direct relation between the physical damage (the crack) and the  service length.

To implement methods for damage tolerance or probability-based inspection planning,  the cracks' size must be estimated and evaluated. The classical theory for crack  propagation is based on the Paris law, describing a linear relation to the stress intensity  factor range in a log-log scale. However, data for describing the propagation phase is  limited, considering typical bridge steels. The established publication flaw evaluation in  steel structures, BS 7910, is based on test data from offshore steels. These tests are  afflicted with significant scatter, and the applicability to bridge steels is uncertain.

At the dismantling of the Söderström railway bridge in Stockholm, 20 specimens were  cut out for fatigue testing. These results have been reported in previously published  reports. The current report presents additional testing to estimate propagation  parameters in the Paris law and the threshold value for crack growth. A total of 13 tests  were performed for the former and 8 for the latter. The report presents the execution of  the tests and the evaluation of the results.

A characteristic intercept of C_k=4,2×10^-13 was reached for the slope m=3, which can be compared to the value in BS 7910 of 5,2×10^-13. Hence, the data from the tests show a slower crack propagation.

Regarding the threshold value for crack growth, a distribution ΔK_th ~ LN(220; 93,2) N/mm^3/2 was reached after a Bayesian update considering data. The associated characteristic value was calculated to be ΔK_th,k=104 N/mm^3/2, which can be compared to the value in BS 7910 of 63 N/mm^3/2.