Journal of Bridge Engineering

Welded residual stress and stochastic traffic at the deck-to-rib welded joints in the orthotropic steel deck (OSD) of bridges highly affects the fatigue life of bridges. This study presents a strain energy-based method to evaluate the service life of welded joints in the OSD considering the combined effects of traffic load and welded residual stress. Thermomechanical analysis is performed for the welding process of the welded joints using a finite-element model to analyze the distribution of welded residual stress at critical locations of the welded joints. An analysis model is developed to simplify the nonlinear superposition of vehicle-induced stress and welded residual stress. A modified strain energy-based fatigue evaluation approach is presented based on the superposition of stresses. Based on the presented model, a new W_a–N_f curve is proposed to take the effects of welded residual stress into consideration. The presented method is validated against experimental results, and then it is applied to investigate the fatigue life of welded joints in an in-service cable-stayed bridge. The theoretical and inspection results are compared and analyzed. The results show that there is a better coincidence between the predicted fatigue life using the presented approach and the inspected fatigue life in contrast with the conventional approach of neglecting the effects of welded residual stress.

The stress state around a rib-to-deck (RD) joint in an orthotropic steel bridge deck is extremely sensitive to wheel-load location in the transverse direction. Fatigue cracks of various types can occur around the RD joint. In this study, a bending fatigue-test system was used to simulate an RD joint under the critical-load case. The fatigue tests for two groups of RD joint specimens with weld-penetration ratios of 15 and 75% were conducted to assess the influence of the penetration ratio on the fatigue-crack type and its fatigue resistance. Fatigue-test results suggest that the fatigue resistances of the RD joint specimens with 75 and 15% penetration ratios were approximately Class B and between Class C and Class D, respectively, per Japanese standards. Furthermore, results show that increasing the weld-penetration ratio can prevent root-to-throat fatigue cracks and enhance the fatigue resistance RD joints.

The effectiveness of an innovative retrofitting method using ultrahigh-performance fiber- concrete (UHPFRC) was investigated experimentally on RC bridge piers with deficient lap splice details. The objective is to study the performance of the proposed retrofitting technique for eliminating the bond failure mode in lap splice regions of rectangular columns with a cross-sectional aspect ratio exceeding 2. The strengthening technique consists of substituting the normal concrete around lapped bars in the splice region by UHPFRC. The test program includes unidirectional reverse cyclic tests conducted on five rehabilitated and one control rectangular large-scale specimens. The longitudinal reinforcement ratio ranged from 1.30 to 1.67% with bar diameters (d_b) equal to 25, 30, 35, and 45 mm. A splice length at the column base of 24d_b and stirrup spacing of 300 mm were used for all specimens to replicate column design practice before the introduction of modern seismic design specifications. The selected self-compacting UHPFRC mix contained 234 kg/m³ (3% by volume) of 10 × 0.20-mm straight fibers. The failure of all retrofitted specimens was progressive and ductile because they all failed due to the successive rupture of the dowel bars in the footing at a very high displacement ductility ratio. The lapped splice splitting failure mode was eliminated, whereas no longitudinal bar buckled. The UHPFRC cover integrity contributed to eliminate all failure modes associated with concrete damage (crushing, spalling). Recommendations for applying the proposed technique to deficient column rehabilitation are suggested.

Since 2006, the number of M3.0 and larger earthquakes occurring in Oklahoma has increased dramatically. Four M5.0 and larger events have caused damage to residential structures, which raises a concern about the potential for damage to Oklahoma’s highway bridges and their components. This study evaluates the potential for damage by assessing the seismic response of the most common bridge class in Oklahoma. The Oklahoma DOT bridge inventory is used to determine the most typical bridge class, and a representative bridge is modeled using nonlinear finite elements. A series of transient analyses were conducted to assess its performance under a suite of recorded bidirectional ground motions (GMs) from the September 3, 2016, M5.8 Pawnee earthquake (the largest event to date). Transient time history analyses were performed and responses (bearing deformation and column curvature) were recorded and presented. Slightly nonlinear responses were observed for the measured GMs. An incremental dynamic analysis was performed to assess the response of the typical highway bridge under higher intensity shaking closer to the epicenter in which seismic stations were not present. The measured GMs from seismic station GS.OK005 were incrementally scaled to AASHTO design levels (S₁ = 0.10g) and to intensities derived by a USGS GM mapping product near the epicenter (S₁ = 0.20g). Bearing responses indicative of slight damage, such as failure of anchor bolts and sliding of bearings, were predicted, and maximum column curvatures reached 80% of their yield curvature. This study has shown that future earthquakes with comparable or higher levels of shaking may well damage bridges, especially close to the epicenter in which shaking intensities are higher.

This article reports on research to experimentally analyze the reliability of ground-coupled ground-penetrating radar (GPR) for the detection and imaging of subsurface delamination in concrete. Experimental studies were conducted that included a RC deck mock-up with synthetic delamination targets and two highway bridge decks. Both of the bridge decks had areas of delamination as determined through sounding, infrared thermography, and some limited physical sampling. Receiver operating characteristics (ROC) analysis was used to evaluate the reliability of GPR when implemented using the procedure described in the current ASTM standard that employs signal reflections from uppermost steel bars. The overall results indicate that GPR had limited ability to detect and image areas of delamination in concrete using the aforementioned ASTM procedure. After almost four decades of implementation of GPR for the condition assessment of concrete bridge decks, confusion remains regarding the ability of GPR to directly detect areas of delamination. The findings from this research should make an important contribution to clarifying this issue to improve decision making for maintenance and repair when considering GPR as a condition assessment tool.

The design of pedestrian bridges (PBs) is typically governed by the serviceability limit state under human-induced excitation. A comprehensive evaluation of the reliability level achieved in designing for this limit state has not yet been reported. This paper attempts to address this gap for metal structures through a comprehensive structural reliability-based evaluation of design guidelines currently used in North America and Europe. An advanced first-order second-moment (AFOSM) method is used to determine reliability levels under different loading scenarios considering uncertainties in the pedestrian-induced walking loads, structural properties, and comfort limits. The results show that the guidelines do not achieve sufficiency under the design traffic. Moreover, suburban or urban PBs with frequently occurring design traffic densities of 0.2–0.8 pedestrians per square meter achieve very low reliability levels under infrequent traffic densities. Significant disagreement in the reliability levels obtained by the different guidelines is observed. Based on this evaluation, it is proposed that current design provisions be calibrated to a higher reliability index under design crowd densities and that traffic-dependent comfort limits be adopted. The reliability level achieved by incorporating a model error term, previously proposed by the authors to better align model predictions with observations, is also evaluated. The key results from this evaluation show that the uncertainty in the model error term has a positive impact on the reliability estimates; thus, this term can be regarded as deterministic.

This study concerns the structural behavior of long-span cable-supported bridges after the sudden rupture of a cable. The load carried by the failed cable must be redistributed to the remaining structure. The cable adjacent to the failed cable receives most of the redistributed load and becomes the critical member. The aim of this study is to find the stress increase ratio of this member. To do this, an analytical approach based on differential equations of the system is used. Then, an approximation function for a simplified bridge model in a cable-loss scenario is derived. The least-squares method is also used to minimize the error of the approximation function. The proposed approximation function has been checked by numerical models, and its accuracy has been proven. The results show that due to increases in the ratio of the bending stiffness of the girder to the axial stiffness of the cables, the stress increase ratio of the critical cable decreases.

This paper introduces a finite-element (FE) investigation into the behavior of integral abutment bridges (IABs) under alternate cycles of expansion and contraction of the bridge due to seasonal temperature variations. As an example, a multiple-span, reinforced concrete, solid-slab bridge was proposed and analyzed using an elastoplastic two-dimensional FE model. The bridge abutment is supported on steel H-piles. The proposed FE model has the capability of simulating both the construction of the bridge and the backfilling process using a multistage numerical technique. The mobilized earth pressures and changes in these pressures due to thermal effects were predicted for different bridge lengths. The results of the analyses showed that the design earth pressures are notably affected by the bridge length, the number of temperature-increase cycles, and the stiffness of the backfill material, but to a lesser extent by the relative stiffness between the bridge deck and abutment. The pressure distributions predicted here may be useful when updating the current IAB design guidelines.

In current bridge design specifications, no provisions exist for live-load distribution (LLD) in the components of skewed integral bridges (SIBs). Thus, in this study, skew correction factors (SCFs) were developed for the girders, abutments, and piles of SIBs to adjust the readily available LLD formulae for integral bridges (IBs). For this purpose, two-dimensional (2D) and three-dimensional (3D) structural models (SMs) of numerous SIBs were constructed and then analyzed to calculate the LLD factors (LLDFs) for SIB components as a function of the skew angle and various structural parameters. The ratio of the LLDFs for SIBs to those for the same bridges with no skew were then used to formulate SCFs. Live-load (LL) effects in SIB components were then simply obtained by multiplying the SCFs with the results obtained from the readily available LLD formulae in the literature. The comparison of the results with those from structural analyses revealed that using the developed SCFs together with readily available LLD formulae for regular IBs yields reasonably good estimates of LL moment and shear in SIB components.

Vibration measurements are an established method for determining the tension force of stay cables. Whereas stay cables are structural members exhibiting significant geometrical nonlinearities, their behavior can be linearized to approximate their vibration characteristics. Based on measured natural frequencies, the cable force can thus be identified. This paper presents methods to facilitate such force identification using highly mobile measurement equipment—namely, modern microelectromechanical systems (MEMS)–based acceleration sensors connected to battery-operated microcontrollers as well as those integrated in smartphones. The paper systematically investigates the accuracy of measurement data obtained from such systems and the effect on the tension force calculated. It is shown that sensor resolution and sampling rate directly affect the accuracy of the force measurement and are discriminating criteria between competing systems. Furthermore, a study of the amplitude of excitation arising from manual and ambient wind excitation shows that resolution limitations of the sensor may prohibit a reliable identification of natural frequencies. A novel smartphone-based software framework is presented, which integrates measurements from the different types of sensor systems. Furthermore, it allows use of nonlinear finite-element–based analyses for the cable-force identification. The proposed methodology and implementation were validated on cables of the Queensferry Crossing during construction, for which direct force measurements were available to compare with vibration-based results. The technology has shown potential to allow very simple and cost-effective yet accurate tension-force measurements.

Research concerning tsunami forces on bridges has focused primarily on determining total forces either experimentally or numerically, often with the goal of establishing total force demand equations. However, this approach does not provide an understanding of which bridge components contribute to the total force demands as well as the individual component demands. Using a two-dimensional (2D) computational fluid dynamics approach, the influence of individual bridge components on total bridge forces was assessed in detail to show how this approach can provide critical information for designers, and was applied for bridges with varying superelevation and girder spacing as a proof-of-concept example to illustrate the benefits of this approach in understanding tsunami-like wave impacts on structures with complex geometries. The analyses resulted in the following findings. First, the upstream face of the upstream girder along with the traffic barriers carried the majority of the total horizontal impact load, which remained approximately constant while both the superelevation angle and girder spacing varied. Second, the opposing girder faces between pairs of girders experienced horizontal impact loading due to the compressed fluids between them. These loads were comparable in magnitude to the maximum horizontal impact load; however, these opposing girder forces canceled each other out, with this effect diminishing for increasing girder spacing. Despite its lack of influence on the overall demand, such behavior dictates the local design requirements of bridge girders to resist bore-induced forces. Third, the top deck face as well as the upstream and middle bottom deck faces sustained the majority of the vertical impact loading. As the superelevation angle was increased, a decrease in the total vertical load was observed, whereas when the girder spacing was decreased, an increase in the total vertical load occurred. Fourth, the upstream face of the upstream girder, traffic barriers, and top deck face carried the bulk of the total horizontal postimpact load, which varied parabolically with superelevation independent of the girder spacing. Finally, the deck primarily resisted the vertical postimpact load, which decreased linearly with an increase in superelevation angle, decreasing in magnitude for increased girder spacing.

External dampers are commonly used on site to control excessive cable vibrations on cable-stayed bridges. The optimum damper size and the corresponding maximum attainable damping ratio of a damped cable can be predicted using existing design tools, such as the universal damping estimation curve. However, because an external damper is typically installed very close to the cable anchorage on the bridge deck, its effectiveness in suppressing vibrations of relatively long cables would be limited. The hybrid system solution can overcome this limitation by interconnecting a damped cable with its neighboring ones with transverse crossties to enhance its in-plane stiffness. However, when a single damped cable becomes a part of a hybrid system, its bending stiffness and sag would be altered, thus having an impact on the optimum damper size and the maximum achievable damping ratio. The current study evaluated how the configuration change of a damped cable from an isolated one to be a part of a hybrid system would affect its damping property. After a brief review of an analytical model of two-cable hybrid systems, a parametric study to investigate how variation in the crosstie stiffness and the spacing between the crosstie and the damper would influence the damping property of a hybrid system is presented. Damping estimation equations for predicting the optimum damper size and the corresponding maximum attainable damping ratio of two-cable hybrid systems are presented. The validity and applicability of the proposed damping estimation equations were examined using 12 symmetric and 12 asymmetric two-cable hybrid systems, the system parameters of which were taken as the same as those on real cable-stayed bridges. Results show that the proposed damping estimation equations can provide satisfactory prediction of the optimum damper size and the corresponding maximum achievable damping ratio for two-cable hybrid systems having different system properties and layouts. Numerical examples are presented to illustrate the application of the proposed damping estimation equations in the practical design of hybrid systems.

Deterioration of concrete bridge barriers as a result of corrosion of internal steel reinforcement in severe environmental conditions is a major problem. Glass-fiber-reinforced polymer (GFRP) bars are currently used as an alternative to the conventional steel reinforcement because of their corrosion resistance, long-term durability properties, and exceptionally high tensile strength. A recent design project conducted at Ryerson University on a TL-5 bridge barrier proposed the use of 15 M and 13 M GFRP bars as vertical reinforcement in the barrier front and back faces at 300 mm spacing, respectively; 15 M GFRP bars were used as horizontal reinforcement in the barrier wall. The connection between the deck slab and the barrier wall utilized GFRP bars with a 180° hook for proper anchorage. To qualify the developed GFRP-reinforced barrier for use in Canada, a vehicle crash test was performed according to the safety-performance evaluation guidelines of the 2009 AASHTO Manual for assessing safety hardware (MASH) for Test Level 5 (TL-5). The crash test involved a 36,000-V vehicle impacting the barrier at a target impact speed and impact angle of 80 km/h and 15°, respectively. This article summarizes the procedure and the results of the vehicle crash test conducted on the developed GFRP-reinforced barrier. Criteria to evaluate crash-test results showed that (1) the barrier controlled and redirected the vehicle to the lane; the vehicle did not penetrate, underride, or override the barrier; (2) no concrete detached elements, fragments, or other debris from the barrier penetrated the occupant compartment or presented undue hazard to others in the area; (3) the occupant compartment remained undeformed; and (4) the truck remained upright during and after the collision. As such, the developed barrier performed acceptably according to MASH TL-5. The calculated equivalent impact force, acceleration, deflection, and recorded strains caused by vehicle impact are presented.

Although numerous studies have been conducted on prestressed duct friction loss test methods, only a few have considered the error sensitivities of the test methods used. For this paper, to conduct a prestressed duct friction loss test, the test error sensitivity coefficients of strain distribution and tension force ratio methods were derived. The effects of prestressed tendon parameters on the error sensitivity coefficient were studied and validated through a statistical analysis involving large amounts of in situ test data. The results indicated that the tension force ratio method was sensitive to test error, especially when the difference in tendon angle was less than 3° and the length was less than 15 m. The test error sensitivity coefficient of k was greater than that of μ under the same conditions. To compensate for the shortcomings of the strain distribution and tension force ratio methods, the P-Δ method was proposed and a solution for the P-Δ method was derived. A comparative analysis indicated that the error sensitivity of the P-Δ method used to calculate the curvature friction coefficient μ was significantly less than that of the tension force ratio method, while test error amplification also occurred when the P-Δ method was used to calculate the wobble friction coefficient k.

Drilling crack-arrest holes is a simple technique, commonly used by bridge owners to control and/or repair fatigue cracks in steel bridges. Although well-established relationships exist for sizing the diameter of crack-arrest holes for in-plane fatigue loading, the effectiveness of such holes in preventing reinitiation of out-of-plane (distortion-induced) fatigue cracks is not well understood and has received little attention in the literature. This paper presents the findings of experimental and numerical studies on the effectiveness of crack-arrest holes for repairing cracks in bridge girders damaged by distortion-induced fatigue. The experimental study was performed on segments of plate girder loaded out-of-plane with cyclic forces applied through a crossframe. Crack-arrest holes of various diameters were drilled at the tips of fatigue cracks having different lengths, and their effectiveness in preventing crack reinitiation was evaluated. The experimental results were complemented with a suite of three-dimensional, solid-element, finite-element analyses. Parametric study was used to investigate the effect of crack-arrest hole diameter, placement, and crack length on stress demands in the web gap region of a girder similar to a physical model tested in the laboratory. The experimental and numerical findings of the study were evaluated in the context of common industry practices. The results showed that in bridge girders susceptible to distortion-induced fatigue damage, crack-arrest hole placement (inherently tied to cracking location and geometry) was the most important parameter and had much greater influence on the effectiveness of crack-arrest holes than hole diameter.

Seismic vulnerability is a critical issue in Italy, because the bulk of existing structures were designed before enforcement of the new national building code in 2008 and the ensuing seismic risk map, which assigns a nonnegligible peak ground acceleration (PGA) to previously nonseismic areas. Thus, seismic retrofitting becomes of primary importance even though often challenging, because remarkable limitations mostly due to structure geometry, site features, and economical issues should be tackled. All of the aforementioned aspects were a primary concern for the Tanaro and Bormida Bridges (Alessandria, Italy), discussed herein, of the A21 Highway. The main criticalities came from (1) the presence of three different superstructure deck types (simply supported RC and prestressed concrete beam decks, and a continuous prestressed concrete box girder deck), (2) geometrical and feasibility restraints (limited deck-to-deck gaps and bearing heights), and (3) instream foundations. Whereas vulnerability assessment showed the need for retrofitting, seismic isolation of the bridge with friction pendulum bearings (FPBs) was studied. The preliminary design was performed through the simplified linear single-degree-of-freedom (SDOF) model, and the final solution was checked by means of linear modal analyses via finite-element simulations. The use of a SDOF model proved to be very useful in designing the isolation system thanks to its flexibility and possibility of easily managing the main parameters. The final designed isolation system was shown to be very effective owing to the significant reduction (even higher than 80% in the present work) of the base shear force and the limitation of the seismic displacements allowed by high energy dissipation capability.

Bayesian dynamic linear models (BDLMs) are traditionally used in the fields of applied statistics and machine learning. This paper performs an empirical validation of BDLMs in the context of structural health monitoring (SHM) for separating the observed response of a structure into subcomponents. These subcomponents describe the baseline response of the structure, the effect of traffic, and the effect of temperature. This utilization of BDLMs for SHM is validated with data recorded on the Tamar Bridge (United Kingdom). This study is performed in the context of large-scale civil structures in which missing data, outliers, and nonuniform time steps are present. The study shows that the BDLM is able to separate observations into generic subcomponents to isolate the baseline behavior of the structure.