Journal of Bridge Engineering

The current AASHTO specification requirements for the web bend-buckling strength-reduction factor (R_b) do not consider the influence of longitudinal stiffeners on the flexural resistance of steel I-girders subsequent to the theoretical bend-buckling of a slender web. This is a deficiency that can have significant impact, especially in regions of continuous-span girders subjected to negative moment. This paper evaluates the contribution of single-web longitudinal stiffeners to the yield limit state flexural resistance of slender-web steel I-girders, often referred to as the plateau resistance for the lateral torsional and flange local buckling limit states. Based on test simulation studies, the authors recommend a cross-section model that captures the response of the postbuckled web for both homogeneous and hybrid girders subjected to uniform moment. The proposed model offers up to 60% improvement relative to the current AASHTO specifications.

The current AASHTO specifications impose a penalty on the strengths corresponding to the compression flange in slender-web cross sections due to bend-buckling of the web in flexural compression. This web bend-buckling factor (R_b) does not account for any potential benefits from the influence of web longitudinal stiffening. In a companion paper, the authors developed a cross-section model that can be used to estimate the flexural capacity of I-girders at the yield limit state. In this paper, an improved handling of combined web bend-buckling and lateral torsional buckling (LTB) of longitudinally stiffened I-girders is proposed based on finite-element test simulations. Results from the proposed prediction model are compared to available experimental tests. In addition, the R_b calculated from the proposed model, used in conjunction with the current specification flange local buckling (FLB) equations, is shown to provide an improved characterization of the FLB capacity of these types of members.

The composite inverted T-slab bridge system provides an accelerated bridge construction alternative for short- to medium-span bridges with spans ranging from 6.1 to 20 m. The system consists of adjacent precast inverted T-slabs with a cast-in-place (CIP) concrete topping. Such a composite bridge system offers a shallow superstructure depth ideal for sites with stringent vertical clearance requirements. When concentrated loads are applied to a bridge of this type, the bridge deforms as a two-way flat plate. This paper presents an analytical and experimental investigation to study the relationship between transverse bending and reflective cracking. Transverse bending moment demands were quantified using a finite-element model and compared with tested transverse bending moment capacities provided by several subassemblage specimens, which feature two precast cross-sectional shapes and three transverse connections. It was concluded that all tested specimens performed well at service load levels. The detail that features a precast inverted T-slab with tapered webs and no mechanical connection between the adjacent inverted T-slabs and CIP topping is the simplest and most economical.

In the design and performance evaluation of RC structures, different quantities, such as ultimate strength, serviceability, and ductility, need to be considered. Because these quantities deal with both strength and deflection, the nonlinear load-deflection prediction of RC bridges becomes essential. In this paper, an analytical procedure is proposed for predicting the flexural behavior of intact and corroded RC beams with or without lap splices considering the bond stress-slip behavior at the steel–concrete interface. The accuracy of the proposed procedure is verified through several experimental and numerical case studies. Last, the proposed procedure is applied to predict the flexural behavior of intact and corroded T-beams of a RC bridge, and the results are verified through the finite-element analyses.

For multispan suspension bridges, frictional resistance between the main cables and the saddles is essential for counterpoising unbalanced cable tension between different spans. The use of friction plates in saddles is one of the most common methods for increasing frictional resistance, and the corresponding evaluation method is vital for the design of suspension bridges. In this study, an analytical model for the calculation of frictional resistance between cables and saddles is presented. Large-scale model tests were performed to simulate the realistic friction conditions of a typical multispan suspension bridge in China, namely the Yingwuzhou Yangtze River Bridge, to validate the model. The main components of frictional resistance were analyzed, and the effects of vertical friction plates in enhancing frictional resistance were investigated by model tests and theoretical analyses. The results indicate that the presented analytical model provides an effective and feasible tool for evaluating the total frictional resistance and the contribution of friction plates in practice. The main components of total frictional resistance, including the frictions at the base and side contact interfaces between the cable and the saddle trough, are quantified, proving the former to be dominant. The parameter analyses reveal that the frictional resistance can be increased by 57% by incorporating six vertical friction plates for the actual saddle.

This paper primarily introduces a new lateral isolation system proposed for the seismic control of a long-span cable-stayed bridge, namely, the Yongning Yellow River Bridge, with high seismic activity and a heavyweight concrete girder. Elastic cables (used in pairs to provide a uniaxial tension/compression constraint) in association with a fluid viscous damper (FVD) for supplemental energy dissipation were employed to form a new lateral isolation system at the girder–tower connections. The cable pairs provided the essential lateral stiffness in service conditions, the desired flexibility and sufficient deformation capacity for seismic isolation, and the ability to recenter the girder following an earthquake. At the pier locations, buckling restrained braces (BRBs) were used for lateral isolation of the piers. A three-dimensional nonlinear finite-element model was developed, and three lateral earthquake-resisting systems (i.e., the traditional fixed system, an isolation system using steel dampers, and the presented system) were analyzed through nonlinear time-history analysis. Sensitivity analyses were conducted to determine the design parameters of the new system, and the serviceability limit states were also checked. A further comparison of the seismic behaviors and cost-effectiveness between a typical isolation system with steel dampers and the proposed system was undertaken. Results show that the new lateral isolation system can provide a capable, cost-effective, durable, and resilient solution for lateral seismic control in long-span bridges in critical seismic conditions.

Welded joints in steel bridge decks are vulnerable to the fatigue damage caused by heavy-loaded trucks. A realistic probabilistic model of truck loads provides a basis for simulating the fatigue stress spectrum of these welded joints, where the fatigue reliability assessment can subsequently be carried out. In this paper, a stochastic fatigue truck load model was developed for probabilistic modeling of fatigue stress ranges to investigate the fatigue reliability of welded steel girder bridges. To deal with the uncertainty-induced computational complexity, a framework including deterministic finite-element-based hot-spot analysis and probabilistic modeling approaches is presented. In addition, a learning machine integrating uniform design and support vector regression is used to substitute the time-consuming finite-element model. The development of both the framework and the learning machine provides a reasonable, efficient, and accurate probabilistic fatigue damage model. Finally, a limit-state function of fatigue damage is established with the consideration of traffic parameters, including the growth factors of traffic volume and the vehicle weight. A prototype steel box-girder bridge is presented as a demonstration to illustrate the feasibility of the proposed stochastic fatigue truck load model and the corresponding framework. Parametric studies indicate the impact of traffic parameters on fatigue reliability indices of the welded joint in the lifecycle. The stochastic fatigue truck load model provides a new approach for probabilistic modeling of fatigue damage and reliability assessment of welded steel bridges.

Steel girders require stability controls for different construction stages before the lateral-torsional support from the concrete deck occurs. The load-carrying capacity of bridge girders and bracing forces generated in their bracings are very sensitive to the girders’ initial imperfections in terms of both the magnitude and distribution along the span. Relatively little knowledge is available in this matter; however, decisions on the “worst” shape of imperfections that gives a conservative resistance and/or maximizes the bracings forces is often not an easy task in practice. The present paper reports the test results (in terms of the load-carrying capacities and bracing forces) of a large-scale bridge of twin I-girder type in which the location of the intermediate cross beam was varied across the depth of the main girders. Moreover, extensive numerical investigations were performed to study the effects of some relevant shape of imperfections on both the load-carrying capacities of the studied bridges and the magnitude of bracing forces generated in the cross beams that were involved. The test results showed that the load-carrying capacity of steel girders can exceed their lowest theoretical eigenvalue because of the presence of initial imperfections. In the finite-element analyses of the studied bridge cases, in some cases, the girders followed their shape of geometric imperfections and reached a load value that was greater than the lowest eigenvalue of the systems.

In box girder bridges, the quantity and distribution of reinforcement to be put in concrete elements of sections can be evaluated only by considering the deformation of the cross section in addition to the longitudinal analysis of the static scheme, establishing the entire state of stress of box sections. This leads to a need to evaluate the interaction between internal forces obtained by the global analysis and the ones obtained by the local analysis of the cross sections. The frame effect implies the elastic deformation of slabs and webs, whereas eccentrically applied loads lead to cross-section distortion with the loss of the box shape. Hence, the reinforcement is strongly influenced by a significant interaction between longitudinal shear and transverse bending moments, because the same reinforcement bars, although acting on different planes, face these two stress resultants. Moreover, the longitudinal reinforcement in the webs plays an important role in the failure model. In this paper, an analytical model of interaction between longitudinal shear and transverse bending is presented, based on the stress field theory and on the application of the static theorem of plasticity. The proposed model allows engineers to establish the effective quantity of transverse and longitudinal reinforcements of box girder elements, both for ordinary and prestressed concrete girders. In the model, the variation of stress field inclination between the limit state of first cracking (serviceability) and the ultimate limit state is considered, and the related consequences are discussed. Interaction domains of longitudinal shear and transverse bending are sketched and supplied through dimensionless graphs to be used by designers for actual cases of engineering practice. A case study of a prestressed girder bridge is presented to show the reliability of the proposed approach. The analytical model is validated through comparisons with experimental data; finally, the influences of prestressing, warping phenomena, and cross-section distortion are discussed.

Precast concrete segmental box girder bridges (PCSBs) are becoming increasingly popular in modern bridge construction. The joints in PCSBs are of critical importance, which largely affects the overall structural behavior of PCSBs. The current practice is to use unreinforced small epoxied keys distributed across the flanges and webs of a box girder cross section forming a joint. In this paper, finite-element analysis was conducted to simulate the shear behavior of unreinforced epoxied joints, which are single keyed and three keyed to represent multikeyed epoxied joints. The concrete damaged plasticity model along with the pseudodamping scheme was incorporated to analyze the key assembly for microcracks in the concrete material and to stabilize the solution, respectively. In numerical analyses, two values of concrete tensile strength were adapted: one using a Eurocode formula and one using the general assumption of tensile strength of concrete as 10%f_cm. The epoxy was modeled as linear elastic material because the tensile and shear strength of the epoxy were much higher than those of the concrete. The numerical model was calibrated by full-scale experimental results from literature. Moreover, it was found that the numerical results of the joints, such as ultimate shear load and crack initiation and propagation, agreed well with experimental results. Therefore, the numerical model associated with relevant parameters developed in this study was validated. The numerical model was then used for a parametric study on factors affecting shear behavior of keyed epoxied joints, which are concrete tensile strength, elastic modulus of epoxy, and confining pressure. It has been found that the tensile strength of concrete has a significant effect on the shear capacity of the joint and the displacement at the ultimate load. A linear relationship between the confining pressure and the shear strength of single-keyed epoxied joints was observed. Moreover, the variation in the elastic modulus of epoxy does not affect the ultimate shear strength of the epoxied joints when it is greater than 25% of the elastic modulus of concrete. Finally, an empirical formula published elsewhere for assessing the shear strength of single-keyed epoxied joints was modified, based on the findings of this research, to be an explicit function of the tensile strength of concrete.

Serviceability issues related to lightweight pedestrian bridges have attracted a great deal of attention in the literature since the London Millennium Bridge incident, which involved large amplitude motions during its inauguration. Various design recommendations have since been proposed to quantify the steady-state resonant response and to mitigate excessive levels of vibration in slender pedestrian bridges. In the current study, design provisions currently used in North America and Europe are evaluated. Two full-scale aluminum pedestrian bridges were tested in the laboratory, and their performance in terms of meeting serviceability limits under pedestrian walking loads was assessed in accordance with a number of guidelines and standards. The experiments involved walking tests under different pedestrian densities. The results show significant differences in the predicted responses, which are mainly attributable to differences in the DLFs adopted in the guidelines, lack of guidance on the appropriate walking speeds in crowd loading conditions, and the factor scaling the individual pedestrian load to multiple pedestrians. The results after applying the guidelines were found unconservative in the lateral direction. The simulated results showed sensitivity to changes in the natural frequency due to the added mass of pedestrians, which becomes especially important in lightweight bridges. Recommendations to address these issues and to reconcile the differences between the various guidelines and standards and with measurements, based on results already available in the literature, are proposed.

Lateral wind loads often control the stability of bridge girders during construction stages in which the deck is not yet structurally effective. In this study, the key objectives were to experimentally quantify aerodynamic wind-load coefficients (drag, lift, and torque) for common bridge girder shapes and to quantify shielding effects arising from aerodynamic interference between adjacent girders. Wind tunnel tests were performed on reduced-scale models of built-up steel plate girders and prestressed concrete Florida I-beams (FIBs). Aerodynamic properties were measured for individual girder cross-sectional shapes and for systems of multiple adjacent girders (tested both with and without bridge deck formwork in place). Results from the wind tunnel tests, which revealed negative drag coefficients at selected downstream girder positions, were synthesized into simplified wind-loading cases suitable for use in bridge design. Wind-load cases were separately developed for assessing overall multigirder system stability and for determining the required strength of girder bracing between individual girders.