Journal of Bridge Engineering

Accurately determining the current state of the main cables of a suspension bridge is critical to assessing the safety of the bridge. These cables are composed of thousands of individual bridge wires, each of which deteriorates over time at a different rate. Current inspection methods provide an incomplete picture of the condition across the cable section, so cable strength estimation methods that rely on this inspection data involve considerable uncertainty. Furthermore, there is no method for estimating the continuing decline in cable strength following an inspection due to ongoing corrosion. This paper lays the groundwork for a time-dependent corrosion-rate model for bridge wires by using monitored environmental parameters from the cable interior. To establish this model, a methodology to estimate the annual corrosion rate as a function of environmental variables was proposed. First, experimental data on the corrosion rate of carbon steel from previous studies were analyzed using machine learning methods. Temperature, relative humidity, pH, and Cl⁻ concentration were determined to be the most relevant variables for predicting the corrosion rate. Next, cyclic corrosion tests were performed by subjecting bridge wires to various levels of these environmental variables, and the resulting data were used to augment the experimental data set from previous studies. Finally, a corrosion-rate model that predicts the annual corrosion rate of bridge wires was developed using the augmented data set.

Reliably assessing the condition and safety of suspension bridges poses the problem of determining the remaining strength of bridge cables in their present state. Corrosion of the steel wires, which together comprise the main cable, is universally recognized as the main cause of the reduction in strength. Hidden by the protective cable wrapping, this corrosion often goes undetected and unmonitored for years, discovered only during costly and intrusive inspections. This paper presents a method for estimating the remaining capacity of cables that is based on monitored environmental condition data. The long-term corrosion rate of bridge wires, $C\left(t\right)$, can be described by the exponential expression $C\left(t\right)=A{t}^n$ where $A$ is the annual corrosion rate for a metal free of corrosion products, $t$ is the time in years, and $n$ is an exponent dependent on the type of metal as well as the prevalent environmental conditions. In this study, a novel approach that relies on tensile test data gathered during a bridge inspection, coupled with environmental condition data typical to the locality of the bridge was used to quantify $n$. Temperature and relative humidity distributions across the cable section required to estimate the corrosion rate were correlated to externally monitored inputs using data from a full-scale mock-up cable subject to cyclic temperature and humidity conditions. Using the developed method, the evolution of cable strength over time under typical environmental conditions was simulated for a 100-year-old cable (Williamsburg Bridge in New York City) as well as a new, hypothetical bridge cable composed of galvanized wires. In the numerical simulation of the Williamsburg Bridge cable, the reduction in cable strength between the years 1988 and 2100 was estimated between 3.2 and 7.0%. Extending this concept, the methodology presented in this paper for estimating the remaining strength of suspension bridge cables may be readily adapted to other bridges and can be used to complement the current best practices for bridge inspection.

In this study, an automated acoustic scanning system was developed for rapid delamination detection on concrete bridge decks. A new ball-chain impact source was designed by combining the advantages of chain-drag (rapid testing speed) and impact-echo tests (analysis in the frequency domain). Conventional steel link chains and the new ball chains were investigated as acoustic excitation sources. Acoustic signals were processed by short-time Fourier transform (STFT) in the frequency range 0.5–5 kHz. Compared with the conventional chain-drag test, the ball-chain results show better signal-to-noise ratio (S/N), higher sensitivity, and repeatability to delamination identification. The automated scanning system includes excitation sources (ball chains), acoustic sensors (microphones), a geographic positioning system (GPS), data acquisition (oscilloscope), and a signal-processing algorithm. Acoustic scanning results are integrated with positioning data to generate an image of the scanned area in a map view. The system was validated in the field on a concrete bridge deck and was found to show satisfactory accuracy, efficiency, and repeatability.

Ultrahigh performance concrete (UHPC) is gaining attention from the bridge community as a compelling option for closure pours that complete prefabricated bridge element (PBE) connections; these connections typically use field-cast UHPC cast over noncontact lap-spliced reinforcing bars to achieve structural continuity between elements. The goal of this study was to investigate the behavior of different commercially available UHPC-class materials for prefabricated bridge deck connections using noncontact lap splices. Two types of experiments were conducted to assess performance: a direct tension pullout test and a component-level prefabricated-bridge-deck connection test. The objective of the pullout testing was to investigate the effect of the UHPC fiber-volume fraction on UHPC-to-rebar bond strength and compare the performances of different UHPC specimens with existing design guidance. Component-level testing was used to investigate the bond between the UHPC and precast concrete, cracking behavior under cyclic loading, behavior during long-term cyclic loading, and monotonic ultimate-loading behavior. Findings indicated that available design guidance held for different UHPC-class materials. Furthermore, test results were used to propose additional guidance related to the UHPC fiber-volume fraction and cover requirements.

A recently constructed and commissioned in-service light rail transit bridge in Denver, Colorado was examined analytically and via field testing to study its live load performance. The bridge is a four-span horizontally curved structure, composed of chorded prestressed concrete bulb tee girders. The bridge was designed following the American Railway Engineering and Maintenance-of-Way Association (AREMA) guidelines from the Manual of Railway Engineering, which do not explicitly address transit rail structures. In-situ load testing and calibrated finite element analyses were conducted to study the live load performance and determine the effects of speed and centrifugal forces on dynamic amplification and live load distribution. Field-testing and model results indicated variations in live load distribution and amplification as train speed increased. In addition, live load distribution and amplification values differed from results determined using equations from AREMA and the AASHTO guidelines, documents that are commonly used to assist analysis and design due to a lack of available provisions tailored to transit rail bridges.

This paper presents the safety assessment of prestressed concrete box-beam (PCB) bridges under customized transport vehicle (CTV) load. First, the recorded data of 81,616 CTVs were used in the investigation of load characteristics, in which five typical CTV types were extracted and the corresponding critical CTVs were determined. Second, the effects of PCB bridges under critical CTV loads were compared with the design load effect of AASHTO and the Ministry of Communications and Transportation (MOCAT). Third, the safety assessment of PCB bridges under critical CTV loads was accomplished with the presented reliability-based load rating method, in which the partial factor of CTV load effect is optimized by considering the benefit and failure cost upon the passing of CTVs. Adopting different partial factors for the CTV load effect according to the gross vehicle weight is proposed. The multilevel ratio of indirect cost to direct cost was adopted to investigate its influences on target reliability index, partial factor, and assessment results. The analysis results show that the ratio of indirect cost to direct cost has a greater influence on target reliability index rather than the partial factor of CTV load effect and the assessment results. For the PCB bridges under CTV load, the negative bending moment may control the structure safety instead of the positive bending moment in some load cases. The proposed analysis framework and the results can be used in the safety assessment of other bridge types under CTV load and traffic permitting.

A long-span bridge supports a large number of vehicles, which are usually present when an earthquake strikes the bridge site due to the low predictability of earthquake events. Existing seismic analyses of long-span bridges have failed to characterize the comprehensive coupling effects among the bridge structure, vehicles, and spatially varying seismic excitations. A new full-response prediction methodology for the coupled bridge–traffic interaction system under spatially varying earthquake excitations was developed by capturing the interaction effects not only between the bridge and moving vehicles, but also between the earthquake excitations and the coupled bridge–traffic system. Different from most existing bridge seismic analyses, in which only traditional earthquake loads in terms of inertial forces are applied on the bridge structure, the new formulation also incorporated the coupled earthquake forces on both the bridge and vehicles, which were expressed as functions of the bridge–traffic coupling matrices and earthquake displacement inputs. The proposed methodology was numerically demonstrated on a prototype long-span bridge and traffic system under spatially varying earthquake excitations. The full responses of the bridge and vehicles were predicted when the bridge–traffic system was subjected to earthquake excitations. Numerical results suggest that the coupled earthquake force as derived in the present study has notable influence on the dynamic performance of the bridge and vehicles under seismic excitations.

Restrain devices, being of low cost and easy to install, are often used to limit the relative displacements between the decks and columns in a simply supported bridge. Although the current guidelines have provided different restrainer design methods, they do not adequately consider the dynamic interactions between the decks and columns in isolated bridge systems considering the flexibility of isolation systems. Additionally, due to a fiber-reinforced polymer (FRP) with a high tensile strength and shape memory alloy (SMA) with energy dissipation capacity and self-centering property, they are good candidates to be used as seismic restrainers. However, currently there is no appropriate design method for such restrainers in isolated bridges. The objective of this study is to propose a restrainer design procedure for simply supported bridges equipped with elastomeric isolation systems. Three types of restrainers, i.e., steel, SMA, and carbon fiber-reinforced polymer (CFRP) cables, are considered. The design procedure for restrainers is developed based on a linearized 2-degree-of-freedom (DOF) analytical model. A three-span simply supported highway bridge located in Vancouver, British Columbia, Canada, is chosen as a case study. The effectiveness of the proposed method is evaluated. The results show that the restrainers could remain functional and limit the relative displacement within a design value during earthquakes. The effectiveness factor, defined as the ratio of the yielding displacement to the peak restrainer displacement, ranges from 2.0 to 6.4. The relative displacements of the bridge retrofitted with steel, CFRP, and SMA cables at midspan supports could be decreased by 41.6, 49.2, and 58.1%, respectively, compared with the allowable design displacement. The total length of SMA cable restrainers is the smallest, and they are more efficient in limiting the relative displacement compared with elastic restrain devices.

Dynamic characteristics of cable-stayed bridges are widely accepted as valuable indicators to determine their performance in structural health monitoring (SHM). Although research has been extensively conducted in this area, such vibration-based physics methods still face great challenges in improving the effectiveness of damage identification from complex large-scale systems, particularly when other factors, including operational and environmental conditions, may cause high interference to the vibration response. Data-intensive machine learning techniques have been gaining attention due to their robustness for data classification. In this study, a framework was developed for data-driven structural diagnosis and damage detection using a support vector machine (SVM) integrated with enhanced feature extraction techniques for rapid condition assessment for large-scale cable-stayed bridges. The wavelet transform, Hilbert-Huang transform (HHT), and Teager-Huang transform (THT) were selected as three representative feature extraction methods. A kernel function-based SVM was used to facilitate the identification of damaged and undamaged cases. Numerical simulation was conducted to verify the effectiveness and accuracy of the proposed methods applied to a cable-stayed bridge. Results showed that the wavelet time-frequency analysis is more robust to noise than the HHT and THT, whereas the latter two transforms are more sensitive to capture damage/defects. Moreover, for regular signal data, the THT, due to the high time resolution, had the highest concentration and thus is the most sensitive compared with the other two methods. Parameters of interest, including impacts of damage level, damage location, sensor locations, and moving vehicle loading, are extensively discussed. All cases reveal that data-driven approaches could effectively map damage features over and under undamaged cases, dramatically enhancing the effectiveness and accuracy of data classification, which will greatly benefit in situ cable-stayed bridge assessment and management.

An innovative normal concrete (NC) with partial use of ultrahigh-performance concrete (UHPC) composite structure is proposed in this paper to reduce the long-term creep-induced deflection in continuous box-girder bridges. The creep properties of NC-UHPC composite columns were experimentally investigated, and test results showed that the creep coefficients in NC-UHPC composite columns decreased considerably compared with NC columns. A creep computational model of the NC-UHPC composite was further developed to describe long-term creep behaviors of the composite columns in good agreement with the test results. Based on the proposed creep model, the long-term behaviors of a long-span box-girder bridge, made of the NC-UHPC composite, were evaluated by finite-element analysis (FEA), and the influences of the UHPC area fraction and the length of embedded precast UHPC columns were also discussed. Compared with the traditional prestressed concrete (PC) box girder, the long-term creep-induced deflection at the midspan of the NC-UHPC composite bridge reduced by 54.5%. The NC-UHPC composite is, therefore, conceived to be one of effective solutions for long-term creep-induced deflection control of long-span continuous PC box-girder bridges.

This article presents the results of an experimental investigation to evaluate the effectiveness of the use of hybrid fiber-reinforced concrete (HyFRC) along with some additional detailing features at the pier–foundation interface for improved seismic performance. Two different fibers, namely, steel and polypropylene, of appropriate volume fraction were mixed with concrete to obtain HyFRC. Three different detailing strategies were considered at the region of the pier–foundation interface. Whereas normal reinforcement details were followed for Specimen 1, additional reinforcing bars (dowels) were used in the interface region for Specimen 2, and corrugated sheet duct was used below the interface region in Specimen 3. In this study, detailed strain measurements at predefined locations were correlated with the enhancement of displacement ductility and energy dissipation in different test specimens. Hybrid simulation was used with input excitations of different intensity levels to evaluate the seismic performance of HyFRC bridge piers. The simulation was conducted in a manner such that the specimens experienced different levels of damage. Further, after the completion of the hybrid test, cyclic tests were conducted to evaluate the ultimate capacities of these specimens. The piers with dowel reinforcement at the interface region were found to be better than the other two types for the considered seismic performance parameters, such as delayed growth of strain, improved damage tolerance, enhanced energy-dissipation capacity, higher stiffness at any drift ratio, and improved load-carrying capacity. Further, specimens with HyFRC outperformed all the corresponding specimens made of conventional concrete in terms of all of the aforementioned parameters characterizing enhanced seismic performance.

The vibration behavior of a stay cable subjected to abrupt rupture due to lateral force was studied using model experiments. Based on the experimental results, the performance of an existing cable-stayed bridge was studied using numerical analysis in cases where stay cables are abruptly ruptured by lateral force. The experimental results show that the impact force currently recommended is insufficient. The impact force is more affected by the tensile stress of the cable during rupture than by the initial stress. The results of numerical analysis show that the cable-stayed bridge analyzed in this study lacks serviceability performance against losing multiple cables in a collision incident.