Journal of Bridge Engineering

The deterioration or damage of stay cables can threaten the safety of bridges. It is difficult to judge whether a stay cable is deteriorated only by the variation in cable force because cable force also varies with temperature. Therefore, an early warning method for abnormal cable forces of cable-stayed bridges considering structural temperature changes is proposed in this paper. In the proposed method, a baseline model is established to characterize the normal relationship between the healthy cable force and structural temperatures, so the abnormal cable force can be detected by inspecting the residual between the identified cable force to be detected and the cable force predicted by the baseline model. First, the frequency–temperature relationship (FTR) model obeyed by a healthy cable is established by analyzing the mechanism of cable force change caused by structural temperature changes in bridge components from the perspective of deformation coordination. Then, the prediction error is inspected by constructing a mean value control chart of the modeling error generated from the baseline FTR model. An early warning will be triggered when the prediction error exceeds the control limits. Finally, the proposed method is applied to an actual cable-stayed bridge to verify its effectiveness. The results show that the modeling and prediction capabilities of the established FTR model are satisfactory and that the abnormal changes in cable forces can be detected by the proposed early warning method with satisfactory accuracy.

Abstract

Emerging technologies are a strategic ally in the efficient management and preservation of pavements and bridges. Toward this end, different methodologies, such as weigh-in-motion (WIM), have been developed for the detection of overweight vehicles and traffic monitoring. Bridge weigh-in-motion (BWIM) systems can be installed and serviced without interrupting traffic flow. This paper presents a new approach for processing traditional strain response data. This proposal extends the BWIM concept of influence area from response time-histories. The dimensionality of the strain response waves is reduced by the calculation of the centroid and the area under the curve. Subsequently, these two dimensions are used as inputs to an ordinary Kriging (OK) metamodel to predict the gross vehicle weight (GVW) of the passing traffic. The OK methodology allows for the strategic selection of vehicles to train the BWIM metamodel. An example of the application of Kriging metamodeling (KM) to BWIM through the instrumentation of an in-service highway bridge located in Costa Rica is presented. Experimental horizontal strain data along with corresponding weight measurements from a static permanent weigh station were available for 90 trucks to validate the proposed enhanced BWIM methodology.

Practical Applications

This research addresses the current concerns regarding United States infrastructure. The ASCE’s 2021 Report Card for America’s Infrastructure ( ASCE 2021) scores a general C- for all 18 categories reported (aviation, D+; bridges, C; ports, B-; rail, B; roads, D; transit, D-; among others). In the bridge category, the use of “living bridges” is mentioned, referring to sensors installed on the structure, as an innovative solution to provide continuous feedback on structural conditions. The proposed methodology in this paper can provide information on the proper instrumented bridge’s structural condition, and the type of traffic traveling over the route the bridge services. This information can be used to feed the databases that support Transportation Asset Management Plans, which are pointed out in the ASCE’s Report Card as federally required tasks to maintain and preserve properly the transportation infrastructure through sceneries and projections.

The accuracy of frequency prediction is markedly affected by the nonlinearity and model degradation caused by multiple environmental conditions, which may hinder the prediction accuracy and detectability of damage. Therefore, this paper proposes a local correlation model (LCM) between multiorder bridge frequencies and multiple environmental factors for early warning of abnormal frequencies. First, partial least-squares analysis was conducted to extract several environmental principal components sensitive to modal frequencies. The most relevant local data set for each online environmental sample was selected according to similarity measurements based on Euclidean distance metrics. On this basis, more accurate environment–frequency relation models were formed using relatively simple local linear regression models. To filter out the residual environmental variability not suppressed by the LCM and to enlarge slightly abnormal frequency variation, a warning index (i.e., the weighted Mahalanobis distance) was defined using the residual subspatial reconstruction of principal component analysis. Finally, the validity of the proposed method was verified on a cable-stayed bridge. The results show that in contrast to conventional methods, the proposed LCM can accurately describe complicated frequency variations under changing environmental conditions by considering both the nonlinearity of environmental conditions and the time-varying properties of relation models. The detectability of frequency anomalies induced by sudden events can be effectively improved.

Continuous rigid-frame bridge (CRFB), a type of bridge adopting the unique form of pier-girder consolidation, combines a T-shaped rigid frame with a continuous girder. CRFBs located in western China are likely to suffer destructive earthquakes, which may result in serious damage to bridges. To study the longitudinal seismic responses and damage control of different structure types of the CRFB under strong ground motions, this work develops three nonlinear dynamic analysis models considering the initial internal force state and longitudinal girder-pier pounding effect based on the OpenSEES (version 3.3.0) platform. Moreover, the present study comparatively analyzes the displacement responses, internal force of piers, pounding force and times, and damage condition of the three models and investigates the effectiveness of tuned inerter-based dampers (TIBDs) in controlling seismic responses of the CRFB. The numerical results show that the longitudinal seismic responses of the main piers in the three models are obviously different, especially when subjected to near-fault ground motions. The peak pounding force at the abutments is much larger than at the transition piers, while the pounding times are the opposite. Under near-fault ground motions with a peak ground acceleration (PGA) of 0.4g, the main piers may be damaged moderately or even more seriously, and the bearings are completely destructive. The TIBDs can effectively control the maximum seismic responses and damage degree of the CRFB under pulse-like near-fault ground motions. Among them, both the tuned viscous mass damper and tuned inerter damper are ideal and useful devices. This study can provide useful references for the seismic design and performance analysis of CRFBs.

Existing long-span truss bridges typically undergo significant retrofits to improve their resistance against extreme load events, such as the seismic loads, for which the bridge may not have been designed or considered originally. However, the conventional seismic retrofit measures recommended in most existing codes and specifications may not be effective in providing sufficient alternative load path (ALP) redundancy to a bridge that is vulnerable to sudden member-loss scenarios because of the significantly different magnitude of demands. This paper proposed a performance-based retrofit (PBR) approach for long-span truss bridges that are vulnerable to sudden member-loss scenarios. The performance of the bridge has been evaluated by using the demand-to-capacity ratio (DCR) or strain ratio (SR) on the member/component level or the displacement factor on the bridge system level as indicators. Based on the analysis results, the argument is made that the proposed PBR approach allows bridge designers to meet the predetermined performance objectives for sudden member-loss loading conditions. Increase in weight of steel because of such the ALP retrofits is less than 10%. The retrofitted bridge would meet the desired performance levels in the event of sudden loss of any members on the primary trusses of the bridge.

To increase the structural stiffness of multitower cable-stayed bridges, an analysis theory of crossed cable is developed, and a new-layout scheme of crossed cable for the multitower cable-stayed bridges is presented. The area ratio of the crossed cables at the same anchorage position was derived by examining their restraint stiffness to the bridge tower. Accordingly, the anchorage position of the crossed cable on the main span with a maximum restraint stiffness was examined, and an asymmetrical arrangement of the crossed cables on the main span was proposed. The finite-element models of three- and four-tower cable-stayed bridges were established. The influences of the asymmetrical arrangement of crossed cables on the deformation of the tower and girder, the stress of the bridge tower, and the dynamic characteristics of the bridge were analyzed. The results from the analysis demonstrate that the area ratio of crossed cables is equal to the ratio of the square of the cable length when their restraint stiffness is maximum. When the height–span ratio of the multitower cable-stayed bridge was between 0.2 and 0.3 and the anchorage position of the crossed cable was 0.69–0.73 times the span length from the middle tower, the restraint stiffness of the crossed cables was maximum. Under a uniform live load, the mechanical properties of asymmetrical arrangements were better than those of the symmetrical arrangement of crossed cables. When the crossed cable as adjusted to an asymmetrical arrangement for three- and four-tower cable-stayed bridges, the tower top displacement and tower bending moment were considerably reduced, and the vertical frequency of the bridge was increased. Note that the asymmetrical arrangement could considerably improve the restraint of crossed cables to the middle tower, increase the structural stiffness of the bridge, and improve the stress of the middle tower.

The cyclic performance of substandard reinforced concrete (RC) bridge substructures seismically retrofitted using titanium alloy bars (TiABs) is assessed using full-scale experiments and nonlinear numerical models. The retrofit technique uses TiAB ligaments to simply and effectively strengthen columns with inadequate flexural lap splices and provides confinement through hoop action when combined with continuous TiAB spirals. The failure characteristics, flexural ductility, and hysteretic behavior of unretrofitted substandard (i.e., vintage) and TiAB retrofitted columns were evaluated using quasi-static, reversed cyclic tests of full-scale bridge column–footing specimens modeled after the characteristics of highway bridges in Oregon constructed prior to the introduction of seismic design provisions. Foundation rocking was examined and quantified through implementation of a soil subgrade simulant. Experimental results demonstrate improved flexural response and ductility with reduced concrete damage and residual drifts of RC columns when retrofitted with TiAB. The measured responses of the test specimens were used to calibrate three-dimensional (3D), nonlinear, finite-difference models used to investigate the effect of soil–structure interaction (SSI) on the rocking moment capacity for as-built and retrofitted column–footing specimens. The model was then used to perform a parametric study to evaluate retrofit details (flexural strength and stiffness) and soil characteristics (strength and stiffness) and their relationships to engineering demand parameters such as displacement and strength, which indicated that incorporation of SSI will permit the design of more efficient and economical seismic retrofits.

This paper analyzes the shear behavior of prismatic and tapered continuous beams with corrugated steel webs (CSWs) under different external prestressing tendon layouts. For girder bridges with CSWs, the existing design specifications consider only the shear force caused by the external prestressing force when calculating the shear stress of CSWs under the external prestressing tendons. However, in a tapered continuous beam with CSWs, in addition to the shear force caused by the prestressing force, the prestressing bending moment and the horizontal component of prestressing can also induce additional shear stress, which results in a redistribution of the shear stress on the section. This finding has not been reported previously. Then, the reasons for the shear behavior differences of tapered and prismatic continuous beams with CSWs under external prestressing tendons are explored. The study reveals the factors affecting the distribution of shear stress on indeterminate static beams with CSWs under external prestressing and proves that the traditional method used for predicting the shear stress on CSWs is not applicable to tapered cases. In addition, it is found that the direction of the internal forces and moments also affects the shear-bearing ratios of CSWs and concrete slabs. Consequently, a unified analytical formula is proposed for calculating the shear stresses of the CSWs induced by external prestressing, and it can be applied to both tapered and prismatic continuous beams with CSWs under external prestressing tendons.

Abstract

A bridge influence line (BIL) and a bridge influence surface (BIS) reflect the relationship between the bridge responses and the loads on the bridge and have been commonly used in techniques such as bridge damage detection, bridge safety evaluation, bridge model correction, bridge weigh-in-motion, and so on. Conventionally, a BIL can be extracted from the bridge response under a moving vehicle with known axle loads, while a BIS can then be obtained by lateral interpolation from BILs obtained for key transverse positions. However, in the traditional BIL-extracting methods, the transverse distance between the coaxial wheels was not considered; that is, the coaxial left and right wheel loads are simply treated as one concentrated load (usually referred to as the axle load). Therefore, the obtained BIL is the bridge response under two loads rather than one. Hence, errors may be introduced to the obtained BILs and propagated to the interpolated BIS. Moreover, the potentially significant effect of the unbalance of coaxial wheel loads on BIL identification is ignored. In this research, a new method for determining a genuine BIL and BIS is proposed. The wheel load rather than the axle load was taken into calculation in this method, where the effect of the transverse distance between the coaxial wheels and the unbalance of the coaxial wheel loads was naturally taken into consideration. Laboratory experiments and numerical simulations were performed to verify the effectiveness of the proposed method. Furthermore, a comprehensive parametric analysis was conducted to investigate the effects of some important parameters such as vehicle velocity, lateral deviation of the center of gravity of vehicles, axle count, and road surface condition on the identification performance. The results show that the BIL and BIS can be identified with satisfactory accuracy.

Practical Applications

Both the bridge influence line (BIL) and the bridge influence surface (BIS) have been commonly used in techniques such as bridge damage detection, bridge safety evaluation, bridge model correction, and bridge weigh-in-motion because they reflect the relationship between the bridge responses and the loads on the bridge. However, in the traditional vehicle-based BIL/BIS calibration methods, the track width of the calibration vehicle was ignored, leading to the accuracies of the calibrated influence lines/surfaces being affected by the track width and the transverse unbalance of the calibration vehicle. To address this problem, a new method, which can eliminate the negative effects of the track width and transverse unbalance, is proposed in this study. It provides a more accurate solution for bridge influence lines/surfaces than the traditional methods. The obtained bridge influence line/surface offers a more accurate input for related technologies and applications such as the bridge weigh-in-motion techniques and therefore improves the accuracy of these related technologies.

A numerical model is presented for a rocking bridge bent specimen consisting of two precast concrete columns with posttensioning (PT) bars, two footings, and a cap beam, with or without a buckling restrained brace (BRB). First, a numerical model of the posttensioned bridge bent specimen without a BRB is developed; emphasis is placed on modeling the rocking mechanism at the joint between precast concrete members joined only through PT bars. The numerical model is extended to a hybrid specimen consisting of a two-column posttensioned bridge bent with a BRB as an external energy dissipation device. Experimental results obtained from quasi-static cyclic tests of the posttensioned and hybrid specimens are compared with the numerical models. Satisfactory agreement of the numerical models with the experimental results is observed in terms of structural response, hysteretic energy, and PT-bar forces. The numerical models are subsequently extended to a previously tested three-column bridge bent for two configurations: (1) a posttensioned bridge bent with PT bars, and (2) a hybrid bridge bent with PT bars and two BRBs. A fiber-based bridge bent model is built, which includes soil–structure interaction (SSI) using simplified springs. A comparison of the posttensioned and hybrid bridge bents using far-field and near-field ground motions shows that the hybrid bridge bent displays superior performance compared with the posttensioned-only bridge bent.

Live load of moving vehicles has a very important effect on the fatigue life of suspension bridges, which causes not only vertical deformation but also longitudinal deformation. In this study, a general analytical formulation for analyzing the quasistatic longitudinal displacement of suspension bridges under vertical live loads is developed, and the underlying deformation mechanism is revealed. First, the analytical vertical and longitudinal deformation equations for the single main cable subjected to live loads are formulated considering the geometric nonlinearity. Then, the relation of longitudinal displacements between the stiffening girder and the main cable for a single-span suspension bridge is established through analyzing the geometric configuration of deformed deck-suspender segment and imposing the null longitudinal force condition. The relation is further modified to incorporate the effect of central buckles (CBs). Compared with the finite-element (FE) prediction, the proposed analytical solution is quite accurate for both concentrated and distributed loads. It is found that the coupling of vertical and longitudinal displacement of main cables and the longitudinal constraint between the cables and girder, are responsible for the longitudinal displacement of the girder. The effects of sag-to-span ratio, CB, and inclined suspenders are studied. The longitudinal displacement of the girder can be reduced by about 20% when the sag-to-span ratio is varied from 1/9 to 1/11, and the CB with proper stiffness is more effective in reducing the longitudinal displacements. The proposed formulation can be conveniently applied for parameter optimization in the preliminary design stage so as to avoid tedious repetitive FE analysis.

Most commercial software allows moving-load analysis to be performed on finite-element bridge models. However, the available software does not allow the incorporation of the dynamic load induced by vehicles due to vibration resulting from pavement roughness. The present study, aims to utilize an uncoupled iterative scheme to incorporate road roughness into the numerical model in order to overcome this problem. The vehicle-induced stress history obtained using the uncoupled iterative scheme is then used to determine the fatigue life of a plate girder bridge. A fatigue damage model is proposed, which incorporates the load sequence effects and the segregation of load cycles into crack-initiation and propagation stages. Investigation of various influencing parameters reveals the significance of vehicle speed, vehicle arrival time, pavement category, and the eccentricity of the load and girder design variables on the fatigue life of the bridge.

Segmental posttensioned precast concrete (SPPC) is a promising approach to accelerated bridge construction (ABC). However, an impediment to the performance-based design of SPPC piers lies in the challenge of defining appropriate damage states and corresponding engineering demand parameter (EDP) limits. To address this, in this study, genetic programming (GP), a form of artificial intelligence, is used to develop drift ratio–based EDP limit models in order to identify the onset of four damage levels that are particularly introduced herein for SPPC piers. In this respect, a finite-element model using ABAQUS is developed and experimentally validated to simulate the response of SPPC piers under seismic loading. The model is then utilized to generate a data set of 316 cases, all within the feasible design space of SPPC piers. This data set is subsequently used to develop GP-based predictive equations that map the relationship between different SPPC pier design parameters (e.g., posttensioning strand ratio, concrete compressive strength, and gravity load level) and EDP (i.e., drift ratio) limits. The performance of the developed equations is evaluated through different measures, including fitness function and performance index. Ultimately, parametric and sensitivity analyses are conducted to demonstrate that the GP-based equations are robust enough to characterize the four damage states for SPPC piers.

Socket connections are suggested to connect a precast column with adjacent members in accelerated bridge construction owing to their large construction tolerance. However, a typical socket-embedded connection always causes a heavy footing or pile cap due to the required embedment length, which is not preferable in the seismic design of a bridge system. In this paper, a new type of socket connection, namely, the partially embedded connection between a precast rectangular column and a stepped pile cap, is proposed to overcome this defect. Three half-scale specimens, that is, a cast-in-place specimen and two such designed socketed specimens with different step heights, were fabricated and tested to verify the new socket connection design concept. Experimental results showed that all specimens formed the approximate plastic hinges at the column bases and only minor damage occurred in the pile caps and the connecting regions; the hysteretic curves of the socketed specimens were approximately the same as that of the cast-in-place specimen, which proves that this new socket partially embedded connection is reliable to transfer the horizontal bending moment from columns to pile caps in a column–pile cap system and, meanwhile, can not only reduce the pile cap volume but also achieve the expected seismic performance.

The steel–concrete hybrid joint (SCHJ) between a concrete girder and a steel–concrete composite girder in a cable-stayed bridge has a high risk of cracking under a hogging moment. This study investigated the crack resistance of the SCHJ. Based on a specific bridge, a half-scale SCHJ specimen was designed and tested. A particular loading setup was developed to reproduce the actual loading conditions of the SCHJ. The crack developments and strain distributions of the SCHJ specimen were observed. The SCHJ with fully infilled concrete cracked earlier than the composite girder. A numerical model was developed using the ABAQUS software version 6.14, and the results obtained by this model were in good agreement with the experimental results. The crack development laws of the concrete deck at typical sections in the SCHJ and composite girder in this bridge were derived. An analytical model was proposed to describe crack development in the SCHJ and composite girder. An SCHJ with partially infilled concrete was proposed to improve the crack resistance of the SCHJ compared with the composite girder, and design recommendations were provided for the novel SCHJ with partially infilled concrete in this bridge.