The serviceability of pedestrian bridges may be affected by the discomfort due to vibrations felt by the users crossing the deck. Design guidelines recommend avoiding critical frequency ranges and provide acceptance criteria for maximum accelerations to assess user discomfort for serviceability conditions. The paper presents selected results from an experimental campaign on a two-hinged steel arch pedestrian bridge, analysing its dynamic response to both ambient and human-induced vibrations. Activities such as walking, running, and jumping are investigated. In this field, many research works are available in the scientific literature, but the vibration analysis of a steel arch footbridge with tapered truss cross-section is still missing. The first six vibration modes of the bridge are identified using the Frequency Domain Decomposition technique, while damping ratios are estimated through Enhanced Frequency Domain Decomposition. Walking and running tests reveal a small shift in the bridge's forced response frequencies compared to its free response. Jumping tests are analysed by isolating specific modal responses and estimating modal damping ratios based on free vibration data after the jump. The study also compares the maximum vertical accelerations recorded during these tests with the acceptance limits provided by technical standards. Overall, the paper provides insight into the vibration assessment of a steel arch footbridge with tapered truss cross-section and offers practical indications for field-data interpretation, as well as reference values of natural frequencies, damping ratios and accelerations for this common kind of infrastructures.

This study presents a machine learning-driven comparative analysis to predict frost resistance in rubberized concrete, focusing on the interpretability and performance of three models: Stepwise Linear Regression (SLR), Stepwise Polynomial Regression (SPR), and Classification and Regression Tree (CART). Leveraging historical experimental data, the methodology integrates rigorous preprocessing via Interquartile Range (IQR) outlier detection and tenfold cross-validation to ensure robustness. Key variables, including rubber content (0–20% fine aggregate replacement), water-cement ratio, and freeze–thaw cycles, were analyzed to balance durability and sustainability goals. The CART model offered interpretable decision rules (test R² = 97.01%) identifying critical thresholds like rubber content ≤ 15%, providing a clear guideline for maximizing durability. Comparatively, the study advances prior Artificial Neural Network (ANN)-based approaches by delivering mathematical transparency (SLR, SPR) and actionable insights (CART). These outputs directly guide mix design: the SPR equation enables precise prediction of frost resistance for specific combinations of mix parameters, while the CART rules establish safe application limits. This moves beyond prediction to offer practical tools for optimizing sustainable concrete mixes under frost exposure, prioritizing both accuracy and implementable insight.

The issue of exposed reinforcement, especially in bridges subjected to aggressive environmental conditions, truck impacts, and patch repairs, is an important concern of the civil engineering community due to the increasing number of ageing reinforced concrete infrastructures and the important economic resources directed to bridge repairs. This paper presents a comprehensive review of the current research focused on the structural performance of reinforced concrete beams with exposed or unbonded reinforcement. It evaluates experimental investigations, analytical modelling efforts, and finite element simulations to understand the effects of bond loss on flexural strength, stiffness, ductility, and failure mechanisms. Results from over 20 key studies indicate that flexural strength can be reduced by up to 55% when the length of exposed reinforcement exceeds 80–90% of the beam span particularly in simply supported beams tested under a central point load, and especially in beams with a high reinforcement ratio of approximately 1.5%. In contrast, lightly reinforced beams under similar exposure conditions have shown strength reductions ranging from 0% to 30%, highlighting the importance of reinforcement ratio and test setups. The ductility of beams with 50% exposed length has been reported to decrease by as much as 60%, and stiffness losses were found to be approximately proportional to the exposed length. A comprehensive review of the current literature reveals that in certain configurations where exposed bars remain adequately anchored, the beams maintained an appropriate proportion of their original flexural capacity, suggesting that conservative assumptions in design may lead to unnecessary and costly interventions. The findings emphasise the need for improved predictive models and design guidelines to assess performance and ensure safety during repair and continued service. This systematic review integrates more than twenty key studies, identifies the principal parameters influencing the flexural behaviour of beams with exposed reinforcement, and highlights the current lack of normative provisions, hence offering a consolidated foundation for future experimental research and code development.

The dampers of cable-stayed bridges play a crucial role in bridge seismic resistance. Traditional research requires a large number of trial calculations of damper parameters, and the application of machine learning methods to optimize the seismic performance of dampers in cable-stayed bridges has a great significance. This article is based on the parameter analysis data of dampers for a single tower cable-stayed bridge. Firstly, the advantages and disadvantages of central composite design and comprehensive experimental method were compared and analyzed. Then, the response surface fitting method was optimized using support vector egression. Finally, the optimal damper parameters were studied using particle swarm optimization algorithm. Analysis shows that there is significant nonlinearity in the structural response under earthquake action. The use of support vector machines and particle swarm optimization algorithms can accurately and efficiently fit and optimize damper parameters. From this, it can be concluded that the machine learning method combining support vector machine and particle swarm optimization has good accuracy and applicability in optimizing the seismic performance of cable-stayed bridge dampers, and can be further extended to other research fields.

In bridge health monitoring systems, there exists the dynamic nonlinear dependency of failure among the multiple monitoring points of bridge components. To investigate the impacts of this dynamic nonlinear dependency on the time-varying reliability indices of bridges, this study focuses on relevant research.It proposes a time-varying reliability prediction method for bridge components based on the monitoring data, a Bayesian optimization long short-term memory network (BO-LSTM) and Gaussian-Copula theory. The specific research contents are as follows: (1) Bridge monitoring data are first subjected to filtering, and based on the filtered monitoring signals, a BO-LSTM model is established to dynamically predict the extreme stresses of the existing bridge; (2) A trivariate Gaussian Copula model, which accounts for the dynamic nonlinear dependency of failure among three control monitoring points, is constructed to characterize the nonlinear dependencies among the monitoring data; (3) the validity of the proposed model and method is validated using monitoring data from the Fumin Bridge in Tianjin City of China. The above research results will provide the theoretical foundation and application for bridge reliability prediction and assessment.

Compared with the two-tower suspension bridge, the seismic response of the middle tower is crucial for the seismic design of triple-tower suspension bridges. A finite element model was developed for a suspension bridge. The seismic responses of a triple-tower suspension bridge with an elastic cable (EC) or a fluid viscous damper (FVD) installed between the tower and the deck were analyzed, and the application of an energy dissipation central buckle (EDCB) which was composed of multiple buckling-restrained braces (BRBs) installed at the mid-span in the triple-tower suspension bridge was discussed. Several novel seismic mitigation systems, such as the combination of EDCB and FVD, and the combination of EDCB and EC, were implemented to control the seismic responses of triple-tower suspension bridges. The effects of different seismic mitigation systems on the seismic responses of suspension bridges were studied. Additionally, the seismic performance of triple-tower suspension bridges was evaluated using the analytic hierarchy process (AHP) method, and the selection of parameters for seismic mitigation systems in suspension bridges was discussed. The results show that the longitudinal displacement (LD) of suspension bridges can be effectively controlled by using EC or FVD; however, the internal forces of the middle tower significantly increase due to the use of EC. Applying an energy dissipation central buckle (EDCB) is an effective measure to reduce the internal forces of the middle tower in triple-tower suspension bridges. The combination of EDCB and traditional seismic mitigation devices is highly beneficial for improving the seismic performance of triple-tower suspension bridges. The analytic hierarchy process (AHP) method can be used to select the optimal parameters for seismic mitigation systems. The results indicate that Case 23 is the optimal solution for controlling the seismic responses of the triple-tower suspension bridge.

The near-surface wind field during typhoon landfall is significantly influenced by terrain, particularly in coastal mountainous regions. Based on the WRF model, the near-surface wind field in coastal mountain terrain under typhoon influence is studied by using the simulation results of Typhoon Dujuan (1521). A sensitivity analysis was conducted to determine the suitable simulation period (60 h before landfall to 12 h after landfall) and the appropriate physical schemes (Medium-Range Forecast boundary layer scheme and Betts-Miller-Janjic cumulus convection scheme) for Typhoon Dujuan. The comparison between simulated and observed data revealed a high correlation, with a Pearson correlation coefficient of 0.91 during the wind speed mutation period. A comprehensive analysis was then performed, combining horizontal and vertical wind fields, wind speed, and wind direction, with respect to the terrain. This analysis explored phenomena such as wind speed mutations at the station, the stable position of maximum wind speed, and alternating high-low wind speeds in the horizontal wind field. Results indicate that the initial simulation time significantly impacts the typhoon path in complex geographic areas. Selecting an appropriate simulation period effectively reduces terrain-induced interference on the typhoon trajectory. Wind speed mutations during landfall are primarily driven by dynamic changes in the typhoon circulation, highlighting the close relationship between typhoon weakening and these mutations. Mountainous terrain notably alters the near-surface wind field, especially when the typhoon structure interacts with elevated terrain, causing pronounced alternating high-low wind speed patterns. Additionally, during landfall, the typhoon vertical structure becomes unbalanced, further contributing to its weakening. Through a thorough analysis of the near-surface wind field and mountainous terrain, a better understanding of wind speed variations during typhoon landfall is achieved. These findings provide a theoretical foundation for enhancing the accuracy of typhoon predictions in complex terrain conditions, thereby improving the reliability of typhoon path and intensity forecasts.

In the field of bridge health monitoring, the accurate identification of moving loads is critical for structural safety assessment. To address the prevalent challenges in the existing methods, including improving the modeling of temporal correlations in stochastic moving loads and the ill-posed nature of the inverse identification problem, this paper proposes a novel AR(1)-Tikhonov-SVD framework driven by multi-point deflection responses. Initially, temporally correlated stochastic loads factor sequences are generated via the first-order autoregressive model (AR(1)), which are then coupled with nominal load pattern to construct the stochastic moving load models; Subsequently, the ill-posed system is mitigated through integrating Tikhonov regularization technology and singular value decomposition (SVD) technology, and the generalized cross-validation (GCV) is adopted to adaptively optimize the regularization parameters; Finally, validation was performed using a simply supported beam finite element model (FEM) with multi-point deflection data: (1) Relative Percentage Error (RPE) measured 8.41% under noise-free conditions; (2) RPE increased to 15.10% under 2% Gaussian noise contamination; and (3) increasing the number of measurement points yielded marked improvements in stochastic loads reconstruction accuracy, demonstrating the framework's robustness. This methodology transcends conventional deterministic identification frameworks, establishes a probabilistic paradigm for bridge safety assessment. The three innovative aspects of this work are: (1) Probabilistic stochastic moving load modeling using AR(1)-generated temporally correlated sequences coupled with nominal load patterns; (2) Self-adaptive ill-posedness resolution via Tikhonov-SVD-GCV integration enabling noise-robust reconstruction; (3) Spatio-temporal anti-noise verification through multi-point deflection synergy (9 sections) quantifying sensor-density efficacy.

With the growing demand for long-span bridges in mountainous canyons, wind resistance has become a critical design consideration. Accurately characterizing the complex, terrain-specific wind characteristics—through probabilistic modeling—is essential. However, conventional parametric distribution models often fail to capture the full complexity of wind behavior in such environments. This study addresses this challenge by analyzing one year of high-resolution wind speed and direction measurements collected at a representative mountainous site. We employ an adaptive bandwidth-optimized Gaussian kernel density estimation (KDE) method to construct marginal distribution models for wind characteristics—bypassing restrictive parametric assumptions and effectively capturing multimodal and asymmetric features inherent in the observed data. Building upon these nonparametric marginals, we further apply Copula theory to model the dependence structure between wind speed and direction, enabling a flexible and decoupled representation of their joint probabilistic behavior. This approach accurately captures the nonlinear interdependence between the two variables, which is often overlooked in traditional bivariate analyses. The key findings are as follows: (1) Wind fields in complex mountainous terrain exhibit pronounced directional confinement and “wind-locking” phenomena, with prevailing wind directions strongly aligned with local topographic orientation (e.g., valley axis). (2) The Gaussian KDE method demonstrates superior capability in representing the multimodal and skewed nature of wind data in such environments, outperforming conventional parametric fits. (3) Copula-based modeling effectively characterizes the nonlinear dependence between wind speed and direction, offering a robust and versatile framework for joint distribution modeling in complex terrain.

Bridge train derailments are among the most common and severe consequences of ship collisions with bridges. This study simulates the dynamic response of bridges and trains under such collisions using finite element analysis, integrating practical engineering considerations. The paper analyzes and calculates the dynamic response of both the train and the bridge structure under ship collision by establishing a ship-train-bridge coupled model. Building upon previous numerical simulations of ship-pier collisions, this paper closely focuses on the Chongqi Railway-Highway Bridge, constructing a detailed simulation model that closely mirrors the actual structure. Simulations and verifications of sea-river ship’s impacts were performed and a system model for ship-vehicle-bridge dynamic coupling was established. Through finite element analysis, an in-depth examination of this coupled system was conducted. The numerical results indicate that ship collision significantly intensifies the dynamic response of the train-bridge system: the risk of train derailment rises sharply, with transverse vibrations far exceeding those in a no-impact scenario, challenging system stability. At the same time, the transverse displacement and acceleration in the mid-span region of the bridge show a significant increase, which further highlights the profound impact of the impact on the transverse dynamic response of the bridge structure. Additionally, the train's operating speed and its specific position on the bridge significantly impact the safe train operation, while the train's location on the bridge directly affects the lateral dynamic response of the bridge structure.

Utilizing ultra-high performance concrete (UHPC) for strengthening reinforced concrete (RC) columns has emerged as a compelling approach, particularly for retrofitting. Unlike traditional normal strength concrete jacketing methods, UHPC offers a more efficient alternative. These methods often require substantial increases in column size to achieve desired enhancements in axial, shear, and flexural capacities. By quantitatively assessing the volume of material utilized against the resulting benefits in ultimate capacity and service life extension, it becomes evident that UHPC holds significant promise in retrofit and upgrading endeavors. In the present study, a total of four columns were experimentally tested, including one control column in need of retrofitting and three retrofitted columns, one with a fully reinforced normal strength concrete jacket and the other two with fully reinforced UHPC jackets. The experimental results reveal that the UHPC jacket increases the stiffness, strength, and energy absorption of the columns compared to the normal strength concrete jacket. This study evaluates the feasibility and performance of RC columns strengthened using UHPC. It examines hysteresis curves, failure modes, stiffness degradation, and sectional curvature to assess the efficacy of the strengthening and retrofitting methods.

This paper establishes a novel time-dependent structural reliability analysis framework for assessing corroded reinforced concrete (RC) piers, bridging the gap between probabilistic performance-based earthquake engineering and practical infrastructure management. The primary contribution is a holistic reliability model that uses the reliability index (

) to evaluate competing failure criteria of excessive maximum drift (seismic performance) and residual drift (post-earthquake functionality). Through Monte Carlo simulations integrated with a validated nonlinear finite element model, the study quantifies seismic performance evolution over a 75-year lifespan under two seismic hazard levels. Key findings reveal a critical transition where residual drift supplants maximum drift as the governing failure criterion as corrosion progresses. Crucially, this work introduces a new prognostic framework that defines lost service life (LSL) due to corrosion and recovered service life (RSL) after maintenance actions, linked by an effectiveness coefficient. This framework translates reliability metrics into actionable RSL charts that quantify the life-cycle extension benefits of corrective maintenance. The results provide engineers with a powerful tool to shift from time-based to condition-based management, enabling data-driven decisions on repair strategies and lifecycle planning.

Accurately predicting the ultimate shear capacity of perfobond rib (PBL) connectors is of great significance for the design of steel–concrete composite structures. This study predicted the ultimate shear capacity of PBL using machine learning (ML) methods. Initially, a dataset comprising 233 sets of PBL push-out test data was established. To enhance data quality, an Isolation Forest was used to identify and eliminate outliers from the dataset. Subsequently, four ML models—XGBoost, DT, RF, and ANN—were trained on this dataset to predict the ultimate shear capacity of PBL. By comparing and analyzing the prediction results, XGBoost demonstrated the best predictive performance with an R² value of 0.97, outperforming other models. Then, a visual analysis, including SHAP and PDP, was conducted on the XGBoost model, revealing the contribution levels of different features to the predicted values. The analysis found that the number of perforated holes (n) had the greatest impact. Moreover, based on the analysis of visualizations, recommended ranges of values for the input features are provided to maximize the ultimate shear capacity of the PBL connectors. In comparison with traditional formulas, the trained ML models exhibit superior accuracy. The MAE of XGBoost is approximately 10% of that of the traditional formulas, and its RMSE value is less than 20% of those.