Journal of Bridge Engineering

Piles transmit structural loads through skin friction, end-bearing, or both to deeper and stronger soil layers. Surcharge loads, site grading, or dewatering activities induce downward movement in the soil that is adjacent to piles installed in a compressible soil layer. This movement causes negative skin friction stresses that act downward at the pile–soil interface, which causes an additional force denoted the drag force (Q_n) that is applied to the pile, which results in a larger axial load in the pile shaft. The end-bearing force and positive skin friction that develops in the pile part within stable (incompressible) soil resist the applied loads [e.g., dead load (Q_d) and Q_n]. This paper aimed to evaluate Q_n that was mobilized on driven H-piles that were installed in soft clay with three-dimensional (3D) nonlinear finite-element analysis. Two numerical models were validated against the field data from two instrumented H-piles, which were part of a three-span bridge (E-21) on Highway 418, Ontario, Canada. The calculated settlements and Q_n agreed well with measured field data. The validated numerical models were employed to conduct a parametric analysis to investigate the location of the neutral plane (NP) at which the skin friction changed from negative to positive. In addition, the performance of the piles that were installed in small and large groups was investigated, which considered Q_n and group effects. It was found that the group effect was negligible for piles that were installed in one row but were significant for piles that were installed in large groups. Finally, a group factor was proposed to calculate the drag force for piles in a group (Q_{(n)pile in group}).

In cable-stayed bridges, a cross-tie solution is considered an effective measure to enhance the in-plane stiffness of main cables while the damping increment of main cables is achieved by installing dampers on the main cables. However, the performance of passive viscous dampers reduces because of constraints on the installation location and the influence of ambient temperature. Recently, the cumulative benefits of cross-ties and external dampers have been explored by several researchers. The current study is motivated to explore the combined effect of the viscous damper and the high-damping-rubber (HDR) damper in a cable network. An analytical approach is used to develop a mathematical model of a two-cable network by installing the viscous damper on Main Cable 1 and the HDR damper on Main Cable 2. The effect of key system parameters is explored to observe the performance of an HDR damper in a two-cable hybrid system. Two numerical examples are presented to discuss the damping of the first 10 modes in a two-cable network equipped with both a viscous damper and an HDR damper. It is observed that stiffer cross-ties are a better choice in achieving the higher damping ratio in the fundamental mode of a two-cable network. Similarly, the performance of the HDR damper can also be increased by installing it on the neighboring cable with a high-frequency ratio.

Rolling friction pairs in the tower saddle of suspension bridges could significantly reduce the horizontal load that is transferred to the tower. To investigate the dependence of the rolling friction coefficient (μ) of a rolling friction pair on the normal load (f_n) and the cylinder radius (R), an experiment was conducted where a sandwich-like mechanism measured μ. Based on the μ–f_n–R relationship, a numerical model, which used one-dimensional (1D) Gaussian random nonplanar surface representation, was proposed in this research to calculate the overall rolling friction coefficient (μ_T) of a multiroller plate system. The experimental results showed that the dependence of μ on f_n and R could be divided into three stages: (1) roughness; (2) elastic; and (3) inelastic. In addition, μ was proportional to $\sqrt{f_n/R}$ in the elastic stage. The proposed numerical model could accurately calculate μ_T for a multiroller plate system. The μ_T rose with increasing surface nonplanarity and slightly increased with the number of rollers (N). This clarified that the μ–f_n–R relationship and proposed numerical model could help to quickly determine the size of the rollers and the flatness of plates during design and reduce the scale of the experiments required.

Abstract

Current standard evaluation procedures for compression gusset plates, the two-fold Whitmore section (Whitmore) buckling and the partial plane shear yielding (PPSY), as specified in The Manual for Bridge Evaluation (MBE), are conservative and may result in unnecessary strengthening and retrofitting. To provide accurate and rational load rating factors for decision-making, the two-strut buckling model, which is based on the lower bound theorem of limit analysis, is developed to evaluate compression diagonal gusset plates in steel truss bridges. The proposed model includes two struts, horizontal and vertical strut plates that are subjected to moments, axial compression, and shear forces, and satisfies equilibrium and the M–P–V interaction equation, where the moment on the vertical or the horizontal strut sections is M, the axial force on the vertical or horizontal strut sections is P, and the shear force on the vertical or horizontal strut sections is V. The predicted results were compared with 116 large-scale experimental and finite-element (FE) analytical tests that were reported in the National Cooperative Highway Research Program (NCHRP) 12-84 project, the MBE refined methods including the basic corner check (BCC) and the truncated Whitmore section (TWS) buckling. Of the three methods, the proposed two-strut buckling (TSB) model had the best professional factor [PF (P_test/P_predication)] and the relatively low coefficient of variation (COV). The procedures and concepts in the proposed model combine established engineering principles with simplicity and good engineering intuition to achieve uniformity when evaluating compression diagonal gusset plates in steel truss bridges.

Practical Applications

To provide accurate and rational load rating factors for decision-making for the bridge owners, the two-strut buckling model that is based on the lower bound theorem of limit analysis is developed to evaluate compression diagonal gusset plates in steel truss bridges. The predicted results were compared with 116 large-scale experimental and finite-element (FE) analytical tests that were reported in the National Cooperative Highway Research Program (NCHRP) 12-84 project, the Manual for Bridge Evaluation (MBE) refined methods, the basic corner check (BCC), and the truncated Whitmore section (TWS) buckling. Of the three methods, the proposed two-strut buckling model had the best professional factor [PF (P_test/P_predication)] and a low coefficient of variation (COV). The procedures and concepts in the proposed model combine established engineering principles with simplicity and good engineering intuition to achieve uniformity when evaluating compression diagonal gusset plates in steel truss bridges.

Scoured pile-group foundations (SPGFs) are susceptible to inelastic deformation during earthquakes. Hence, the ductile behavior of SPGFs has become a rising concern in academic and industrial communities. Traditionally, pushover analyses with single-point load patterns have been adopted to obtain the ductile characteristics of bridge pile-group foundations. However, this practice cannot properly account for the inertial effects from both the superstructure and the pile-cap. Accordingly, in this study, the well-known incremental dynamic analysis (IDA) was adopted to rigorously and comprehensively investigate the ductile behavior of SPGFs for the seismic design of bridges in flood-prone regions. To this end, a shake-table test on an SPGF-supported RC bridge bent model was firstly employed to validate the adopted numerical modeling technique. An in-depth parametric study was then conducted on the bridge bent with the damage potential of the SPGF embedded in cohesionless soil, considering various structural and geotechnical parameters. Based on the IDA results, the seismic failure process, structural limit states, and ductile performance indices of SPGFs were systematically investigated. It was found that the failure process of SPGFs obtained from IDA was consistent with that reported in a previous quasistatic test. Scoured pile-group foundations have a considerable displacement ductility capacity (quantified as 2.48 and 3.47, on average, for the easy-to-repair and ultimate states, respectively), as well as lateral strength enhancement capacity (i.e., the lateral strength values for the easy-to-repair and ultimate states were, on average, 1.40 and 1.46 times that of the first-yield state, respectively). The displacement ductility capacity of SPGFs can be improved by using more rows of piles, a larger pile diameter, a higher transverse confinement of the pile section, a lower longitudinal reinforcement ratio of the pile section, or a larger pile spacing. Those with a shorter aboveground pile height (scour depth), or supporting a shorter pier, exhibited a larger displacement ductility capacity.

Bridge pile capacity is a vital criterion used to assure the durability and stability of a bridge pile foundation. In fact, reliably predicting the pile capacity plays a significant role in supporting data-driven decisions for the design, construction, and quality assurance of bridge piles. While previous studies have examined black-box machine learning (ML) models for bridge pile capacity prediction, little-to-no studies were directed to integrating expert knowledge and large bridge pile data to develop an easy-to-interpret white-box ML model for estimating bridge pile capacity. Therefore, this study proposed a novel white-box expert knowledge–guided Bayesian belief network (BBN) to accurately estimate bridge pile capacity. The proposed BBN was developed based on (1) a comprehensive bridge pile data set of 2,735 data points collected from a large bridge project, (2) expert knowledge obtained from eight bridge and geotechnical experts using the systematic three-round Delphi method, (3) a variety of data preprocessing methods, and (4) parametric Bayesian learning applied to different graphical models. The performance of four different BBN models was assessed and compared based on an unseen testing set to evaluate the generalizability of the proposed BBN model. Model evaluation results indicated that the optimal BBN is a tree-augmented Bayesian network that can estimate the discretized capacity of bridge piles with an accuracy of 90.51%. The proposed BBN model was further validated by testing its generalizability performance on another data from a different location. This study contributed to the body of knowledge by providing a novel, intrinsically interpretable, and robust data-driven expert knowledge–guided model for accurately estimating the bearing capacity of bridge piles. Ultimately, this paper aims to attract more research and practical attention toward developing knowledge-based white-box models for advancing the predictive analytics of bridge pile-related data and decisions.

Abstract

The progressive increase in loads and frequency of trains traversing over aging railway infrastructure can lead to large lateral forces and displacements that could potentially cause derailment and problems related to safety. Reliance only on design thresholds for safety assessment of existing bridges can lead to conservative evaluations, unnecessary replacements, and increased maintenance costs. However, there is a lack of data on lateral behavior of existing railroad infrastructure. This article examines the recommended lateral load and displacement specified in codes. It specifically analyzes the responses of railroad bridges under revenue service traffic, using field-collected data. The goal is to establish a contextual understanding of how modeling, field data, and the code relate to each other. First, this paper presents an investigation of measured lateral forces of existing railroad bridges using different instrumentation techniques such as truck performance detector, track loading vehicle, and linear variable differential transducers (LVDTs). Second, a series of lateral displacement field monitoring data of three railroad steel bridges using the LVDT and the time history analysis of two railroad timber bridges under train load events with different speed and axle load using laser are presented. Time history responses of railroad bridges under train load provide an objective qualification to evaluate their current condition. Finally, the data help to identify maximum lateral displacement, train speed, symmetry of bridge response associated with the construction, the number of cars, and freight versus passenger different responses. These measurements are recommended to be used for field evaluations of bridge lateral responses under traffic.

Practical Applications

This study focuses on lateral load and displacement in railroad bridges. The authors suggest using field monitoring data to ensure the safe operation of existing railroad bridges, because the design code’s conservative values may not be sufficient. To assess lateral load, the authors recommend using a pair of forces rather than a single concentrated lateral force. For lateral displacement assessment, the authors suggest the use of a field monitoring technique with LVDT or sensors. This approach enables the creation of a displacement time history that includes train speed and symmetrical movement data. The practical application of this method may be limited by budget, bridge access under revenue service traffic, and safety restrictions for sensor location. Additionally, the amount of data that can be collected depends on the number and location of sensors. Sensor placement must be guided by prior engineering knowledge of the bridge. Finally, data collection may be time-consuming, which may be cost-ineffective and render the whole exercise difficult. In summary, this study proposes a field monitoring approach for lateral load and displacement assessment that requires a careful consideration of practical limitations.

The corrosion of structural steel has a significant impact on the durability of plate girders servicing in marine environments. A high-performance configuration for a steel plate girder with a titanium–clad bimetallic steel (TCBS) web was suggested to eliminate the adverse effects of corrosion on the durability of bridges and industrial structures serving in corrosive environments. The patch-loading resistant experiment on the hybrid TCBS plate girder was innovatively conducted to clarify the feasibility of using TCBS in components with large deformation. Moreover, a numerical modeling method was suggested to research the patch-loading resistance response of hybrid TCBS plate girders. Based on the numerical analysis, a parametric study was performed to clarify the nonlinear bearing response of the hybrid TCBS plate girders subjected to patch loading, which included the effects of the stress–strain property of the structural steel, loading length, aspect ratio, and height-to-thickness ratio. Numerical results showed that current standards underestimated the ultimate bearing capacity of hybrid TCBS plate girders. Therefore, a suggestion for further improving the design methods of TCBS plate girders was recommended. Additionally, the research results of this paper yielded experimental evidence for the engineering application of bimetallic materials.

Fatigue vehicle load is essential for fatigue analysis, whose accuracy is directly associated with fatigue design. With the development of society, traditional fatigue vehicle load may fail to reflect modern traffic conditions, especially for transportation hubs in developed areas. Based on traffic flow data of a long-span steel box girder bridge, statistical analysis is performed and the two types of standard vehicle load are equivalent. Further, fatigue characteristics and fatigue life of two types of welded connection are compared based on the equivalent structural stress method. As a result, 2-axle vehicle takes the largest proportion, and the overload ratio of a 6-axle vehicle is the highest. Compared with fatigue vehicle load III, there should be a correction coefficient 1.89 and 2.30 for equivalent vehicle load based on average traffic flow and heavy lane traffic flow, respectively. For a single-side welded connection and a double-side welded connection, the equivalent structural stress will increase by 81.0%–131.2% for positions of concern when considering heavy lane traffic flow, which is about 21% higher than results considering average traffic flow. In addition, the fatigue life of these concerned positions will encounter a dramatic decrease. Fatigue life of the most unfavorable position is 492.4 years and cracks will occur in 90.0 years based on fatigue vehicle load III. However, it will decrease to 42.3 years and 7.7 years, respectively, when considering the traffic flow of the heavy lane.

With rapid economic development in recent decades, long-span bridges have increasingly attracted attention worldwide. The invention of UHPC (ultrahigh performance concrete) materials has brought more alternatives for long-span bridge deck systems. Compared with the traditional orthotropic steel deck, the UHPC-based deck system greatly reduces the risk of deck fatigue cracking. Recently, a steel–UHPC composite bridge deck (SUCD) system using perfobond strip (PBL) shear connectors is proposed, which shows advantages in terms of self-weight and mechanical properties. However, research focusing on this new steel–UHPC composite bridge deck is limited. In this study, based on a field bridge (under construction), eight full-scale slabs for the SUCD are fabricated and tested to investigate the bending behavior of the structure under sagging and hogging moments. The effects of hole spacing in PBL and reinforcement ratio on the flexural behavior of the SUCD are also studied. In addition, the transverse bending behavior of the SUCD is investigated. To accurately predict the cracking moment and bending capacity of the SUCD, a related calculation method considering the hole effect of PBL is developed and validated by the experiment. Compared with other methods, the prediction demonstrates more accurate results. With the proposed prediction method, the effect of a single parameter on the bending performance is analyzed, and the main factors affecting the mechanical behavior of the SUCD are presented. The findings can provide a reference for the design and further study of the SUCD with PBL shear connectors.

Bridges play a significant role in urban areas, and their performance and safety are highly related to the carbon emissions of infrastructure systems. Previous studies have mainly offered maintenance policies that balance structural safety with overall costs. Considering the goal of achieving near-zero global carbon emissions by 2050, this study proposes a policymaking agent based on a convolutional autoencoder–structured deep-Q network (ConvAE-DQN) for managing deteriorating bridges at the network level while considering sustainability performance. This agent considers environmental, economic, and safety metrics, including spatially correlated structural failure probability, traffic volume, bridge size, and others, which are transformed into a multiattribute utility model to form the reward function. Reinforcement learning is employed to optimize the life-cycle maintenance planning to minimize the total carbon emissions and economic costs while maximizing regional safety performance. The proposed method is substantiated by developing sustainable life-cycle maintenance policies for an existing bridge network in Northern China. It is found that the proposed ConvAE-DQN policymaking agent could output efficient and sustainable life-cycle maintenance policies, which are annually stable and easy to schedule. The utility-based reward function enhances the stability and convergence efficiency of the algorithm. This study also assesses the impact of budget levels on network-level bridge safety and carbon footprint.

Ultrahigh-performance concrete (UHPC) strengthening is an efficient technique to improve the capacity of shear-deficient members. However, the applicability of UHPC strengthening on a T-beam has been scarcely investigated, particularly with regard to repair configurations not reaching beam supports involving a higher delamination potential. In this study, the shear behavior of concrete T-beams with cast-in-place UHPC strengthening is investigated with 10 concrete T-beams, including different strengthening configurations, layer thicknesses, and anchors at the repair interface. Beams with a UHPC bottom layer ending before the support are investigated for the first time. Load–deflection, lateral, and cross-sectional cracking patterns in each beam are analyzed. Besides, strain distributions of the beams are recorded and analyzed through a digital image correlation system to distinguish different failure modes. The efficiency of UHPC strengthening for improving the shear behavior of concrete beams is clarified, and recommendations are provided for a UHPC-strengthened beam design to avoid delamination. The testing results show that UHPC strengthening using lateral layers is the most efficient configuration for improving shear capacity and does not increase the sectional height, while a U-shaped jacket configuration is recommended when a substantial increase of beam stiffness is required. A combination of shear and separation failure is found in beams with a UHPC bottom layer ending before the support, and it is, therefore, suggested not to use a UHPC bottom layer alone due to the development of separation cracks. More ductile failure modes obtained with 50-mm lateral layers and a 25-mm U-jacket are suggested. Installation of anchors at the repair interface is recommended to achieve more ductile failure modes.

Abstract

Wind barrier is an effective countermeasure to protect vehicles from the lateral wind on sea-crossing bridges. Previous studies focused on the sheltering efficiency of wind barriers and their effects on the aerodynamic/aeroelastic characteristics but ignored the analysis of intrinsic mechanism for dynamic stability, especially for flutter instability, which results only in a vague understanding of the role of wind barriers in flutter. Hence, this study paid attention to the effect mechanism of wind barriers on the flutter characteristics to generate a comprehensive understanding of the coupled relationship between wind barriers and the main structure. Wind tunnel experiments of closed box girders with different aspect ratios were first conducted to unearth the sensitivity of wind-barrier location, wind-barrier shape, wind-barrier porosity, and aspect ratios of main girders to flutter performance. The positive function of wind barriers was discovered, which was contrary to the common view that wind barriers had an adverse effect on streamlined structures. Then, the overall evolution of flutter process affected by wind barriers was investigated in detail from the flutter derivative contribution and modal participation. Simultaneously, local-region responses on the bridge surface to the installation of wind barriers were also studied by a proposed simplified local effect model based on surface fluctuating pressure. Also, the reason for the different influences of wind barriers was revealed according to the analysis from both overall and local levels.

Practical Applications

Wind barrier is an effective countermeasure to protect vehicles from the lateral wind on sea-crossing bridges. However, a modification of structural aerodynamic shape affected by wind barriers may have an adverse effect on aerodynamic stability, especially flutter instability. This study focuses on the wind-barrier effects on the flutter characteristics of closed box girders with different aspect ratios. The model test shows that there is no obvious positive or negative correspondence between wind barriers and structural flutter performance. If we can make full use of wind barriers, the flutter critical wind velocity can be improved by even 40%. Then, the flutter mechanism of wind barriers is investigated in detail based on the analysis of global derivatives, local derivatives, and unsteady pressure. Finally, the key flutter derivatives and the key function region of the deck surface are obtained, which may provide some references for the application of closed box girders in super long-span bridges.

Bridges and pavements are two major infrastructure components of a transportation network providing mobility of freight and commodities for economic vitality and access to a range of users as social benefits. However, the lack of a comprehensive infrastructure management system incorporating bridges and pavements inhibits accurate performance prediction, optimal maintenance actions, and the associated use of shrinking budgets. This paper presents an integrated probabilistic life-cycle connectivity framework for the performance analysis of transportation networks containing bridges and asphalt pavements by considering flexural and shear failure modes for prestressed concrete and steel bridges and four failure modes, including international roughness index, rut depth, alligator cracking, and transverse cracking, for asphalt pavements. In this framework, neural network–based deep learning models are used to assess the probabilistic performance of transportation networks and to provide guidance for the associated maintenance strategies. An existing transportation network consisting of bridges and asphalt pavement segments is selected to investigate its life-cycle connectivity reliability and component importance using the matrix-based system reliability method. Results show that the consideration of asphalt pavement failure probability has a significant effect on the probability of transportation network connectivity.

Previous research shows that an accurate simulation of vehicle–bridge systems is not feasible using beam models because they cannot adequately represent the torsional and transverse behavior of a bridge. Three-dimensional dynamic vehicle–bridge analytical solutions and FE models can also be cumbersome to develop and are prone to errors from idealization and modeling assumptions. To address these limitations, this paper presents a novel analytical vehicle–bridge simulation method that utilizes the experimentally estimated modal parameters of a bridge structure. The estimated modes from ambient vibration tests inherently enable the simulation to be valid for any generalized structural system and boundary condition because they truly represent the actual structure. In this paper, the mathematical derivation of the analytical model using plate vibration is presented in detail, the framework for the application of the model is outlined, and the proposed model is validated using a full-scale case study arterial highway bridge in the Canadian Province of New Brunswick. The proposed model offers a valuable solution applicable to real-time structural health monitoring and diagnostics, bridge weight in motion, and drive-by vehicle monitoring fields.

Driven by high-profile events, much attention has been paid to the progressive collapse behavior of building structures. Comparably less research has addressed bridges, especially long-span cable-stayed bridges. In this paper, the progressive collapse behavior of a prototype long-span cable-stayed bridge is computationally investigated through stay-cable removal analyses. The simulations are carried out by direct sudden removal of a single stay cable at different locations and sequential removal of multiple cables. Dynamic effects as well as material and geometric nonlinearities are accounted for in the simulations, which are used to investigate critical cable loss locations and system response to the various cable loss scenarios. The simulation results show that the prototype bridge exhibits good robustness against failure of a single stay cable. When multiple cables are removed, the bridge is less tolerant to cable losses near its supporting pylons than its midspan. For the former, it exhibits an instability-type partial collapse as opposed to an unzipping type of progressive collapse followed by pylon failure for the latter.