Journal of Bridge Engineering

In recent years, the π-shape structure, that is, two pile shafts supporting a reinforced concrete beam and its superstructure, is increasingly used for bridge foundations. Considering the construction quality and the requirement of settlement tolerance, the pile lengths of the π-shape structure are sometimes different when passive or active. However, the effect of the unequal pile lengths on the dynamic characteristics of the π-shape structure has not been investigated. Therefore, this paper analyzes the dynamic response of the π-shape structure with unequal pile lengths. A beam–unequal length piles–soil coupled vibrating system is established in the frequency domain considering both pile–beam–pile propagation and pile–soil–pile interaction. In order to account for the pile–soil–pile interaction, a fictitious soil pile model is employed to compensate for the length difference between the piles. The developed model and the analytical results are verified through comparison with the results of rigorous finite-element analysis and a model test. The verified model and its corresponding solutions are then utilized to conduct a comprehensive parametric study. Some conclusions derived from this study can extend the application of the π-shape structure in engineering practice.

In recent years, the Lode angle was found to have a significant effect on plasticity and fracture strain of ductile metal under shear dominant loading, and the classical J₂ plasticity theory generally overestimates the cases under shear dominant loading. In practice, shear dominant loading is common, that is, fillet-welded joints, partial-penetration welded joints, and metallic shear dampers. Therefore, it is important to study plasticity of structural steel under shear dominant loading and develop corresponding plasticity models for accurate evaluation of the plasticity and fracture of metallic structures. A plasticity model considering the effect of the Lode angle was simplified based on previous studies. Eight types of specimens with different configurations were designed to generate different initial stress states. Three different material tests were conducted to figure out little anisotropy in different locations of steel plate. The integration process for the simplified plasticity model were studied, and a user-defined subroutine in ABAQUS was developed for the model. A postnecking modification method for the true stress–true strain was recommended to evaluate the postnecking behaviors of structural steel through comparison of experimental and numerical simulation results. The simplified single parameter plasticity model proved to have high accuracy for specimens under various stress states, and a detailed calibration process for the material parameter was also introduced.

There are many problems in the commonly used anticorrosion technologies of the main cable, which has been difficult to meet the anticorrosion requirements of the main cable. In order to solve the problems, a new dehumidification system of conveying dry air from the inside of the main cable is put forward. The dehumidification system delivers dry air through the composite air supply conduit embedded in the main cable, and the dry air takes away the moisture in the main cable during the flow process, so as to achieve the purpose of anticorrosion and dehumidification. In this study, the influence of the temperature and humidity environment on the corrosion rate of the high strength steel wire is investigated. Then a dehumidification test of the main cable is carried out. The temperature and humidity variation, dry air pressure distribution, and dehumidification time of the main cable are tested. The anticorrosion effect and mechanism of the new dehumidification system are studied. The results show that the corrosion of the high strength steel wire is very slow in a humidity environment which is below 60%, and the corrosion rate increases with the increase of the temperature and humidity. The dry air of the new dehumidification system can achieve the full section coverage of the main cable, and the relative humidity at each position of the main cable can be reduced to less than the critical humidity of the high strength steel wire corrosion (RH = 60%). So the entire main cable can be well protected against corrosion. The dehumidification time of the main cable calculated by theory is basically consistent with that of the dehumidification test. With the increase of the dry air delivery velocity, the dehumidification time of the main cable is shorter. Therefore, the dehumidification efficiency of the main cable can be improved by properly increasing the delivery velocity of the dry air. The new dehumidification system can provide a more effective anticorrosion method for the main cable of the suspension bridge.

The impact effect of vehicle loads is an important and challenging problem in bridge engineering. In the last few decades, experimental and numerical researches on this problem have been extensively carried out, but a satisfactory formula that can take various factors into account has not been obtained. The vehicle–bridge interaction (VBI) under the excitation of road roughness is essentially a nonstationary random vibration problem. In this paper, the random load of a bridge is expanded into an expression in both the time and frequency domains using Fourier series; the dynamic response of the bridge under the moving random load is solved by the modal superposition method; an analytical solution of the second-order statistical characteristics of the vibration response of the bridge in the form of series is obtained; and on these bases, a theoretical formula of the impact factor (IM) spectrum is obtained. This approach is more scientific and reasonable in theory than the traditional deterministic approach because the theoretical IM spectrum is based on the second-order statistical characteristics of bridge dynamic responses. Moreover, the theoretical IM spectrum can more comprehensively reflect the various influencing factors of vehicle impact effects on the bridge.

Current specifications for concrete barrier design (AASHTO-LRFD Section 13) indicate that the yield line method (YLM) should be used to ensure that a barrier has the adequate flexural capacity to resist the demand imposed by the truck impact. However, ensuring appropriate shear behavior of the parapet is not clearly addressed. Two recent experimental studies suggested that punching shear, rather than flexural yielding, can be the dominant failure mode of concrete barriers subjected to lateral loading. These studies employed idealized static loading setups in the laboratory, which are not necessarily accurate representations of truck impact loading on a barrier. The objective of this study is to use computational simulation to systematically investigate the failure pattern of concrete barriers subjected to lateral truck impact and assess their behavior in both flexure and shear. A set of pushover tests on half-scale concrete walls are selected from the literature and the failure modes of the tested parapets are simulated and analyzed in detail. Static loading demands from the tests are compared with those from impact loading as computed from high fidelity truck impact simulations and the results used to assess the potential for punching shear failure. A punching shear model is proposed. It is shown that the punching shear capacity of the MASH concrete barrier considered, which has standard details, is substantially larger than its flexural capacity. Therefore, it is conservative to continue to design barriers for flexural capacity as is the current practice.

Scouring of pile group foundations is a common phenomenon for cross-river bridges and can produce significant damage in earthquake-prone regions. This study experimentally investigated the seismic failure mechanism and postearthquake vertical load-carrying capacity of scoured pile group foundations. Three identical 2 × 3 reinforced concrete (RC) pile group specimens were embedded in homogeneous medium density sand with an overall scour depth equal to five times the diameter of a single pile, and then were subjected to lateral cyclic loads applied to the pile cap in order to produce a predetermined damage state in the piles. Pushover in the vertical–downward direction (pushdown) was finally applied on these damaged specimens exhibiting a permanent lateral displacement to evaluate their residual load-carrying capacities. Experimental results show that the leading pile was more prone to seismic damage, as both the first aboveground and first belowground plastic hinges originally occurred on it. The embedded depth of potential plastic hinges in leading, middle, and trailing piles gradually increased. In addition, the extension of pile damage had a significant influence on the residual vertical load-carrying capacity and the corresponding vertical failure mode of the pile group. Reductions of 10.4%, 47.5%, and 73.8% in the vertical load-carrying capacity of these scoured pile group specimens were recorded when they previously suffered a displacement ductility of 1.75, 3.5, and 5.0, respectively. Based on the experimental results, a linear degradation formula on the normalized postearthquake vertical load-carrying capacity of pile groups with respect to the displacement ductility was developed. The experimental results presented in this paper could be used to validate the ductility capacity and residual vertical load-carrying capacity of pile groups numerically evaluated by using three-dimensional nonlinear finite-element models. This research represents also a first step toward the development of a rapid postearthquake assessment approach for bridges with pile group foundations.

Highway bridge superstructures may be built using only precast concrete elements, such as deck bulb tees (DBTs). Formerly flange connections between those members have been made with welded steel inserts and grout keys, but these have not held up well under heavy truck traffic. Recently, agencies have started to investigate the use of alternative connections in which bars project laterally from the tee flanges and are connected by a narrow pour strip made from ultra-high-performance concrete (UHPC). They have shown better performance, but are typically made using proprietary UHPC, which is expensive. This paper reports tests and analyses of longitudinal joints between simulated DBTs connected using a nonproprietary UHPC. It includes an analysis of the deck to determine the force demands, material testing information including supplementary characteristics such as bond strength, and tests on full-scale panels simulating the deck portion of the tees. It was found that the strength demand is driven by a prescriptive requirement in the AASHTO LRFD specifications rather than by a specific force demand and that the nonproprietary UHPC was able to provide the required strength using a joint no more than 250-mm wide.

The addition of railings inevitably changes the aerodynamic shape of a main girder, which in turn has a great effect on the vortex-induced vibration (VIV) of a long-span bridge. To ensure the railing has a positive effect, this study presents an inclined railing with improved columns arranged with a spanwise distance of 1–5H (H is the height of the bridge). The design of the improved column is inspired by a passive vortex generator (PVG). The streamwise vortices developed downstream of PVG columns can trigger the three-dimensional instability of a wake, making the shear layer less susceptible to rolling up into mature vortex streets, which are the source of VIV. An experimental study is conducted to compare the influence of traditional and inclined railings on the VIV performance of a closed-box section bridge. It is shown that the addition of a traditional railing deteriorates the VIV performance of the bare girder. However, both vertical and torsional VIVs can be completely suppressed by installing an inclined railing. The results of wake analysis indicate that an inclined railing can greatly decrease the streamwise fluctuation velocity and reduce the spanwise correlation of the streamwise velocity in the wake.

The impact force time history obtained from a crash test or high-fidelity simulation is usually contaminated by numerical noise. The noise signal usually appears as a series of short duration, high amplitude spikes in the time domain that complicate the interpretation of the test or simulation results. The moving average method has been widely used to remove noise from the original data by smoothing the force time history over a moving time window of fixed size, typically 25 or 50 ms. Many researchers use the peak impact force from the smoothed signal as the equivalent static force for structural design purposes. However, there is no theoretical basis for this practice. High fidelity finite-element simulations are used to investigate the adequacy of this empirical assumption by comparing the dynamic response of typical bridge piers and barriers to their responses under the equivalent static force obtained by the moving window method. It is shown that this practice is unconservative for the cases considered. The empirical mode decomposition (EMD) method is used to decompose the original impact force data into noisy modes and useful dynamic signals. A typical impact force signal consists of high-frequency noise, pulselike signals, and monotonic loading components, the last two of which should be considered in the impact design of bridge structures.