Journal of Bridge Engineering

The current design guidelines for concrete barriers have not been updated much since the 1990s despite (1) several experimental, analytical, and computational studies that addressed both the demand and capacity sides of the design problem and (2) substantial increases in the velocity and weight of certain types of trucks specified in the crash testing guidelines that led to significantly higher impact demands than in the previous times. This state-of-the-art paper critically reviews the findings of previous studies and critiques their proposed design models. It highlights the lack of consensus regarding the failure mechanisms that should be considered in the design of concrete barriers and contrasts the results of a variety of proposed models with current design guidelines. Based on the review, recommended models for barrier design are proposed to aid in future revisions of the design specifications. Future research directions are also identified and discussed.

In this study, three reinforced concrete (RC) columns with and without corrosion and different reinforcement details are tested under lateral cyclic loading. One of these columns is well-confined to represent modern RC bridge piers that are designed according to the current seismic design codes, and the second column has the same detail with corrosion damage. The third column is a lightly confined corroded column to represent aging RC bridge piers that are not designed to the current seismic design codes. The experimental results showed that corrosion significantly impacts the ductility loss more than the strength loss of the tested corroded columns. In addition, although the uncorroded column was designed according to the current seismic design code, severe inelastic buckling in the vertical bars was observed during the cyclic tests.

Many of the current fatigue design specifications for cross frames, especially those in composite steel bridge systems, have primarily been based on computational studies and component-level laboratory experiments. To fully understand the behavior of cross frames when subjected to in-service truck traffic and to critically evaluate legacy design provisions, in-service field monitoring of cross frames in various composite systems is important. Therefore, select cross frames in three bridges in the greater Houston area were instrumented with strain gage sensors—one straight bridge with normal supports, one straight bridge with skewed supports, and one horizontally curved bridge with radial supports. For each bridge, rainflow counting techniques and other postprocessing procedures were implemented for a monitoring period of one month to compare the measured cross-frame response. Among other key observations, the measured data demonstrated that load-induced force effects in the cross frames of the skewed bridge system exceeded those in the bridges with normal or radial supports and that cross-frame response was highly sensitive to the longitudinal and transverse truck positions. In all cases, though, the measured damage accumulated on all instrumented cross-frame members during their respective monitoring periods, if extrapolated to the entire service life, would likely not cause significant load-induced cracking in the critical welded gusset-to-member connections.

The fatigue provisions for bridge design in North America and Europe are highly simplified for the sake of practical implementation, in comparison with the state of the art in understanding of the true fatigue behavior. Historically, one significant simplification in the calibration of damage equivalence factors has been to assume that trucks cross the bridge one at a time, when in reality trucks are known to platoon, and multiple truck presence is not uncommon on longer span structures with higher traffic volumes. While code writers have taken steps more recently to consider multiple presence effects, these efforts have been limited, and a clear understanding of the influence of key design parameters on multiple presence effects has not yet been fully established. Against this background, the present work considers real-time multilane weigh-in-motion (WIM) data and compares fatigue damage from recorded loading events over years to the damage resulting from the same vehicles crossing the bridge individually, to better understand the significance of multiple vehicle presence effects on fatigue damage. For both finite and infinite life fatigue design, the results show that multiple vehicle presence effects can be significant, depending on traffic volume, road configuration, and influence line characteristics. Multiple presence factors are therefore defined, allowing practitioners to consider multiple presence effects appropriately when designing bridges where they may be relevant.

The ability of structural health monitoring (SHM) systems to detect defects and damage in existing structures is currently a key issue in the life-cycle management of civil infrastructure. In this respect, the use of dynamic identification to detect the loss of prestress load, grouting defects, cracking, and corrosion in post-tensioned concrete (PC) bridges is receiving special interest as an attractive tool that may drive predictive maintenance for disaster risk mitigation, motivating widespread implementation of SHM systems in engineering practice. In this study, an experimental program on reduced-scale PC beams representative of typical girder bridge decks, at the Department of Structures for Engineering and Architecture, University of Naples Federico II, is presented. The specimens were characterized by different prestressing load levels, grout injection conditions of post-tensioning ducts, and strand rupture during testing. Operational modal analysis (OMA) was carried out at multiple steps of each test, which were associated with different damage states of the specimens. OMA results show that both natural frequencies and mode shapes of PC girders are not significantly influenced by slight cracking corresponding to service load levels, while indicating some frequency variations in the case of local strand failure.

Flooding occurrences have become increasingly severe, posing a serious danger to end-user safety and bridge resilience. As flood fragility assessment is a valuable tool for promoting the resilience of bridges to climate change, it is of great importance to push the development of such methods. However, flood fragility has not received as much attention as seismic fragility despite the significant amount of damage and costs resulting from flood hazards. There has been little effort to estimate the flood fragility of bridges considering various flood-related factors and the corresponding failure modes. To this end, a fragility-based approach that can explicitly address the scour-hole geometry and flood-induced lateral load is presented. First, a three-dimensional finite-element model with pile foundations and surrounding soil was established to estimate the failure mode under various flood scenarios. The loadings on pile foundations were characterized by vertical loading from the superstructure, horizontal loading from the flood-induced lateral load, and the scour effect simulated through a time-history analysis. Then, all potential failure modes of bridge pile foundations in various flood scenarios were summarized. Based on extensive parameter investigations using the deterministic method, the dominant failure mode of penetration failure was determined, and a failure envelope was fitted to guide the design of the pile foundation. Upon establishing the failure mode, a probabilistic fragility analysis considering uncertainties in hydraulic, structural, and geological parameters was finally conducted using the Latin hypercube sampling (LHS) method. The results showed the effects of variation on the fragility of the pile foundation, highlighting that the deterministic analysis without considering the uncertainties in model parameters leads to underestimating the risk due to the penetration failure and the significant influence region.

This paper presents experimental testing evaluating the ability of a bridge deck deicing system to mitigate concrete deterioration from thermal stresses, frost action, and early-age cracking. Two experimental bridge deck models were constructed with embedded heat exchanger tubing and instrumented with thermocouples and strain gauges. Model 1 evaluated the efficiency of a deicing system in deicing and mitigating concrete deterioration from thermal stresses and frost action in concrete bridge decks, while Model 2 tested the effect of a deicing system on early-age cracking in bridge decks. The models were tested in a cold chamber laboratory under conditions representative of Montana's winter weather, with the system circulating warm fluid through the decks. Results showed the system succeeded in increasing concrete temperatures at all depths. While the system did not always raise temperatures above freezing, the consistent increases suggested that the system could enable deicing and mitigate frost action given certain conditions and higher inlet fluid temperatures. The system also successfully decreased thermal movement strain by up to 40% and reduced thermal shrinkage by decreasing the difference between peak cured and stabilized temperatures. Although the system did not eliminate thermal gradients, the maximum gradient induced was insignificant. Further testing is needed on strength impacts. Overall, the deicing system showed promise to reduce thermally induced deterioration in concrete bridge decks by regulating temperatures.

To predict the future condition of a bridge, statistical models for the time-in-condition rating (TICR) can estimate the time that a bridge stays in a given condition and then predict the future condition of the bridge. However, existing research typically uses the probability density functions as the likelihood function when the TICR is estimated by Bayesian updating, in which the change in the condition rating (CR) between two consecutive inspections is assumed to occur at the later inspection, which ignores the uncertainty of the time of the condition change. This assumption will introduce an error, which is particularly significant when the two consecutive inspections are separated over a long time. In addition, a large amount of existing bridge inspection data in China has not been fully recorded; for instance, a lot of bridge inspection data only contains the CR of the bridge from the last inspection. Current research that is based on the TICR has difficulty using this incomplete data. To solve these difficulties, this paper proposes a probability-based likelihood function for the Bayesian updating of the TICR models, which could estimate the distribution of the TICR more accurately using fully recorded or single data. The accuracy of the proposed method is verified with numerical examples, and the results from different methods are discussed. Then, the effect of using complete and single data are examined. The proposed method is applied to the CR of the superstructures of reinforced concrete bridges in Beijing that uses the real inspection data, and the future deterioration risk is evaluated using the updated TICR models.

A novel concept of hybrid reinforcement in the form of glass fiber‒reinforced polymer (GFRP) longitudinal column bars combined with steel longitudinal column bars and high-strength post-tensioning all-threaded rods is investigated in this research to construct bridge columns for seismic regions. Two half-scale specimens were constructed using grouted ducts to connect the columns to the footings as an accelerated bridge construction method. The first column was reinforced with a combination of mild steel and GFRP longitudinal bars (hybrid column) and the second with all-mild steel longitudinal bars, and both columns were confined with a double layer of GFRP spirals; this arrangement improves column longevity since GFRP materials do not corrode. Both specimens utilized high-strength all-threaded rods for post-tensioning. Carbon fiber‒reinforced polymer jackets were externally wrapped at the column bottom to confine the concrete in the plastic hinge region. The seismic behavior of both precast concrete specimens was assessed through quasi-static cyclic tests. Both columns exhibited satisfactory strength and deformability. The combination of post-tensioning bars with GFRP longitudinal bars improved the self-centering capability. The residual displacement of the hybrid column was decreased by 46% compared with the all-steel-reinforced column at 6.0% drift ratio. The elastic nature of the GFRP longitudinal bars provided substantial self-centering by reducing the residual displacement of the hybrid specimen thereby enhancing seismic resilience.

Abstract

In bridges, monitoring data usually correspond to normal operational and environmental conditions, resulting in a lack of damage-related data. For this reason, machine learning algorithms for damage detection are typically unsupervised. Conversely, numerical models are employable as surrogates for extreme events rarely encountered during the existence of a bridge and for common damage scenarios, enabling the training of supervised machine learning algorithms. In this paper, hybrid data obtained by integrating monitoring and numerical observations from the Z-24 Bridge benchmark are used for training supervised machine learning algorithms to classify damage. Special attention is dedicated to the numerical model that, while being simple enough to be used on thousands of runs, must capture the complex nonlinear behavior that typifies damaged conditions. A model updating technique is used for the preliminary calibration of the finite-element model. Numerical data are generated in a probabilistic manner starting from the initial finite-element model, assuming the Gaussian distribution of the uncertain parameters. Three undamaged scenarios and three damaged scenarios are modeled. Subsequently, several supervised learning classifiers are trained with the same hybrid database and a comparative summary of their effectiveness is presented. The paper introduces a novel soft classification method that accounts for the overlapping of observations belonging to different structural conditions in the feature space. This study proves that data generated from probabilistic-based finite-element models can be used for structural health monitoring and damage identification in the context of bridges, therefore providing a hybrid supervised approach that can be easily applied in practice.

Practical Applications

Because bridges are one-of-a-kind, expensive structures, considerable efforts are made to ensure they remain in service for as long as possible. Consequently, monitoring systems are being installed on more and more bridges. However, monitoring data are usually related to healthy states of the structure rather than damage occurrences. Therefore, digital replicas can be employed to simulate damage scenarios that may occur in the lifespan of a bridge. Hybrid data can be obtained by blending monitoring data with simulated data from the same bridge. This paper illustrates the construction of such a hybrid database, which covers a wide range of possible scenarios for a well-known bridge benchmark. This hybrid database is then used to train machine learning algorithms to distinguish between various changes in the dynamic behavior of the bridge as a way to evaluate the condition of the structure and identify damage. In practice, the hybrid approach can be implemented on monitored bridges to raise flags when flaws are detected, to localize and evaluate the severity of damage, and even to make a prognosis on the remaining life of the bridge.

Before the anticipated Nankai Trough earthquake, the effects of multiple cascading hazards on bridge performance and the associated functionalities of road networks must be assessed to develop recovery strategies. This paper presents a framework for assessing the probabilistic connectivity of road networks exposed to both ground motion and tsunami, considering the spatial correlations among hazard intensities. To evaluate the joint probabilities of bridge states (i.e., passable and impassable states) and the probabilistic connectivity of road networks, the total probability theorem was used to integrate spatially correlated seismic and tsunami hazard assessments into the fragility estimates that consider the cascading effects of ground motion and tsunami-induced damages to bridges. For illustrative purposes, the probabilistic connectivity of hypothetical road networks in Japan that could be affected by both ground motion and tsunami during the anticipated Nankai Trough earthquake was assessed using the proposed method. The results showed that seismic retrofitting of bridges located along the route where the probability of impassable states due to tsunami is very low can enhance the probabilistic connectivity of entire road networks.

Concrete beams (girders) prestressed with steel cables are widely used in highway bridges. The prestressed steel cables are prone to area loss due to corrosion. The high stress level and deteriorated bond due to corrosion further risks the structural integrity of these beams. Therefore, corrosion-resistant fiber-reinforced polymer (FRP) reinforcement has been investigated as a substitute for prestressed steel in pretensioned concrete beams. While generating data on such members, especially at a large scale, is a slow and costly affair, validated finite-element models (FEMs) can be leveraged to accelerate the process. This paper presents an FEM strategy for carbon fiber–reinforced polymer (CFRP) prestressed beams that is convenient yet sufficiently accurate. A series of AASHTO Type-I beams with a composite deck were previously tested by the authors and the data from those tests were used to validate the developed FEM approach. Thereafter, a parametric study was conducted using the developed FEM, and the results were used to investigate the behavior of such pretensioned concrete beams beyond the scope of the experimental work. The investigated parameters included prestress ratio, modulus, and reinforcement ratio of the prestressing CFRP reinforcement, and the properties of the concrete and the composite deck. The influence of these parameters on the cracking load, precracking and postcracking stiffnesses, flexural capacity, and maximum deflection of the beam was examined. The results indicate that the prestress ratio influences the cracking load and deflection in CFRP-prestressed beams, with FRP properties affecting postcracking behavior and failure modes. The tensile strength and rupture strain of FRP, respectively, affect the bending capacity and deflection, while the concrete properties have a minor impact on the beam behavior.

The inverted pendulum model is currently a widely accepted pedestrian lateral force model in the study of pedestrian-induced lateral vibration. The inverted pendulum model proposes that pedestrians walking on a laterally vibrating structure maintain their balance and comfort by adjusting their lateral foot position rather than their stride frequency. In other words, the stride frequency of a pedestrian is unaffected by the lateral vibration of the structure. This viewpoint directly contradicts the synchronization theory, which suggests that pedestrians adjust their stride frequency to adapt to the lateral vibration of the structure. The observed phenomenon of phase drift in Bocian's experiments provided validation for the effectiveness of the inverted pendulum model. However, the simultaneous observation of phase pulling indirectly refuted the underlying assumption of the inverted pendulum model: that pedestrians' stride frequency remains unaffected by the lateral vibration of the structure. Bocian's experimental findings shattered the artificial boundary between the inverted pendulum model and the synchronization theory, indicating the existence of a hybrid mechanism in the pedestrian lateral force model. Regrettably, to date, no specific model for this hybrid mechanism has been developed. This study employs the Kuramoto model to describe the phase evolution of a pedestrian under the influence of structural lateral vibration. Based on the traditional inverted pendulum model (IPM), the phase evolution inverted pendulum model (PE-IPM), which takes into account the adjustment of pedestrian lateral stride frequency (or stride time), is developed to explain the potential hybrid mechanism in the pedestrian lateral force model. To validate the effectiveness of PE-IPM, detailed comparisons are made between its numerical simulation results of pedestrian lateral gait and Bocian's experimental findings, the results of which are in good agreement. The validation demonstrates that PE-IPM can capture the detailed characteristics of pedestrian lateral gait well and effectively address the issue of an underestimated equivalent damping coefficient in IPM, confirming that PE-IPM can be used as an improved pedestrian lateral force model for further investigations into pedestrian-induced lateral vibration.

When the high-density polyethylene sheath of bridge cables is not cracked, the galvanized high-strength steel wires are not corroded and only subjected to the fatigue loads of vehicles (called prefatigue). Some studies have confirmed that the corrosion rate and fatigue life of certain alloy materials can be affected by the prefatigue, which is not clear for the corroded steel wires in bridge cables. In this paper, the mechanical properties of high-strength steel wires considering the prefatigue effect under fatigue-corrosion alternating loading were studied, and the residual fatigue life was predicted. First, the corrosion characteristics of high-strength steel wires with different degrees of prefatigue and corrosion were explored through an optical microscope and a surface roughness measuring instrument. Then, the static mechanical properties and fatigue life of corroded steel wires were obtained. The results show that when the prefatigue effect is greater than the corrosion effect, the surface roughness of the corroded steel wires decreases significantly, leading to a smoother surface and a weaker stress concentration. Compared with the results in the “corrosion + fatigue” case, the fatigue life of the steel wires in the “fatigue (prefatigue) + corrosion + fatigue” case may increase. The prefatigue effect has no significant effect on the static performance of the corroded steel wires. Finally, a fatigue life prediction model of corroded steel wires considering the effect of prefatigue was established based on the aforementioned experimental data.

This study addresses the absence of a provincial scale database on chloride exposure for bridges, which specifically focuses on Ontario, Canada. Considering the region-based climate and traffic characteristics, a comprehensive database was developed based on chloride exposure prediction models. This database documents chloride exposure information under different Representative Concentration Pathways (RCPs) of greenhouse gas (GHG) emissions for bridge piers and decks across regions and times. This prediction model was validated against experimentally derived chloride profiles from concrete core samples. The database allows for the assessment of spatial and temporal patterns in chloride exposure, which considers the projected climate change conditions. To facilitate data interpretation, this study employs heat maps to transform the extensive provincial database into a visually accessible format. Bridge piers in Southern Ontario have higher surface chloride concentrations than those in Northern Ontario; however, bridge decks in Northern Ontario tend to exhibit higher surface chloride concentrations compared with those in Southern Ontario. Future projections suggest an overall increase in the surface chloride concentrations on bridge piers across most of Ontario under the RCP4.5 and RCP8.5 scenarios, which is largely attributed to increased traffic volume. However, certain southern regions might see a decrease under RCP8.5, which is a direct consequence of climate change. In contrast, surface chloride concentrations on bridge decks are expected to decline, a trend that is driven by diminishing snowfall and increasing traffic volume. In addition, these findings suggest that relying on RCP8.5 could lead to underestimating the climate change impacts compared with RCP4.5. This study highlighted the spatiotemporal patterns in highway bridge chloride exposure in the context of climate change.