Journal of Bridge Engineering

Scour is the major cause for many bridge failures and damage to piers and abutments. Scour is not easily discernible because it is hidden under the channel flow. Over the years, a number of sensors have been developed for detection of scour depth. Development, testing, and field implementation of a new and simple type of scour sensor is described in this paper. The scour depth detection concept is based on measuring the fundamental frequency of vibration of a rod embedded in the riverbed. The sensor uses a single fiber-optic Bragg grating (FBG) sensor for transduction of the vibration frequency. The inverse relationship between the fundamental frequency and the length of the sensor rod is used for detection of the scour depth. A computational approach is developed based on the Winkler spring reaction soil model for automated calibration of the scour sensor during installation in the riverbed. The scope of the research included development of the theoretical basis for the sensor, establishment of the computational methodology for detection of the riverbed foundation properties, proof-of-concept laboratory tests, small-scale field verification tests, and installation and remote monitoring of scour in a multispan scour critical bridge in Illinois. The results include laboratory test data from the measurements in soil, simulated scour tests in a hydraulic flume, and real-time data from remote monitoring of scour at the bridge site.

A cable-stayed bridge recently constructed by the Ohio Department of Transportation incorporates measures put forth by a Federal Highway Administration study to mitigate stay motion. In following recent trends, the stays at this bridge are built without the use of grout for the purposes of inspection and, if necessary, replacement. Several experiments were performed to determine the viability of using traditional vibration techniques, which assume an integral sheath, to estimate cable tension with this new configuration.

Different from postfailure maintenance in many perspectives, preventive maintenance of civil engineering structures is another highly crucial measure, not only for achieving efficient distribution of limited budget over existent aging infrastructures but also for maximizing their service lifespans. In the context of road bridges, their functional failure often causes serious impact on safety and logistics, which could turn out to be detrimental to the social economy. To appropriately maintain a huge number of aging infrastructures, a strategic maintenance program, facilitating both the global- and local-diagnosis approaches, that is effective in assessing early damage is of high demand. In this study, fatigue damage of concrete bridge decks, which is a common form of deterioration among bridges, was examined by sensitive nondestructive testing methods utilizing propagation of stress waves. Specifically, the fatigue damage process of concrete decks due to repeated traffic loads is visualized by means of active and passive elastic wave techniques, namely the elastic wave tomography and acoustic emission techniques. In the experiment, a full-scale concrete deck was subjected to repeatedly moving wheel load, to induce fatigue damage to the concrete. At three stages of after initial loading, after 10,000 passage and after 20,000 passages of 150-kN wheel loading, the fatigue test was suspended temporarily, and elastic waves were transmitted into the concrete to inspect interior structure with elastic waves' velocity. Applying static load with gradual increment in magnitude, acoustic emission testing was then conducted to extract characteristic acoustic emission (AE) parameters with regard to the intrinsic damage. Promising elastic wave parameters for quantifying the damage, which have been identified through experimental studies, were later verified using in situ deck specimens hewed out from an actual bridge. These experiments showed that by using sparsely arrayed AE sensors for measurement, followed by extracting AE frequency features, global investigation of the integrity of bridge decks could be carried out. Once the area of interest was identified through analysis of AE data, detailed information such as cross-sectional damage could be visualized by employing ultrasonic testing and tomographic reconstruction procedure.

The measurement system configuration is an important task in structural health monitoring in that decisions influence the performance of monitoring systems. This task is generally performed using only engineering judgment and experience. Such an approach may result in either a large amount of redundant data and high data-interpretation costs, or insufficient data leading to ambiguous interpretations. This paper presents a systematic approach to configure measurement systems where static measurement data are interpreted for damage detection using model-free (non–physics-based) methods. The proposed approach provides decision support for the following two tasks: (1) determining the appropriate number of sensors to be employed and (2) placing the sensors at the most informative locations. The first task involves evaluating the performance of the measurement systems in terms of the number of sensors. Using a given number of sensors, the second task involves configuring a measurement system by identifying the most informative sensor locations. The locations are identified based on three criteria; i.e., the number of nondetectable damage scenarios, the average time to detection, and the damage detectability. A multiobjective optimization is thus carried out leading to a set of nondominated solutions. To select the best compromise solution in this set, two multicriteria decision-making methods, Pareto-Edgeworth-Grierson multicriteria decision making and the preference ranking organization method for enrichment evaluation, are employed. A railway truss bridge in Zangenberg, Germany, is used as a case study to illustrate the applicability of the proposed approach. The measurement systems are configured for situations where measurement data are interpreted using two model-free methods; i.e., moving principal component analysis and robust regression analysis. The results demonstrate that the proposed approach is able to provide engineers with decision support for configuring measurement systems based on the data-interpretation methods used for damage detection. The approach is also able to accommodate the simultaneous use of several model-free data-interpretation methods. It is also concluded that the number of nondetectable scenarios, the average time to detection, and the damage detectability are useful metrics for evaluating the performance of measurement systems when data are interpreted using model-free methods.

The objective of the study was to propose a technique to reduce the limitations of modal identification in damage detection using reduced field data for nondestructive structural health monitoring of a cable-stayed concrete bridge. Simply supported bridge models were constructed with predetermined damage at the midspan of the bridge. The technique necessitated the performance of linear and eigen analyses on the control beam and nonlinear analysis on the bridge with damage. Residuals from regression of the mode shape using the Chebyshev rational series on the modal frequencies and transformation and application into the fourth-order centered finite-divided-difference formula were shown. The use of the regressed-mode shapes for the RC bridge model showed very large residuals around the areas of the damage. The results showed that the method was successful in assisting to reduce the limitations of modal identification in locating damage on a bridge model with limited field data and was comparable to other techniques proposed by other researchers in terms of its simplicity.

Continuous awareness of the evolution of the structural condition of bridge structures is of great value for bridge owners, as it allows them to make informed decisions regarding the maintenance and management of these public facilities. Structural condition assessment via monitoring has gained in popularity in recent years, because it can provide structural engineers with plentiful information on the structural condition through various sensors. A key issue for successful application of monitoring technologies for condition assessment is how to realize meaningful interpretations of monitoring data. In this study, an approach to structural condition assessment of in-service bridge deck making use of long-term monitoring data of strain response is proposed and applied to the instrumented Tsing Ma Bridge. The proposed method consists of structural assessment at two levels: (1) deck truss component level and (2) deck cross-section level. As long-term monitoring data of dynamic strain under the in-service condition physically result from a combination of live loads and environmental variations, a wavelet multiresolution decomposition method is first applied to extract live-load effects from the raw measurement data. Based on the extracted data, stress-response histories are derived, and the statistical characteristics (sample means and sample SDs) of peak stresses are explored for stress-level assessment of deck truss components. Dynamic internal forces (axial force, shear force, and moment) of the monitored deck sections are subsequently evaluated by synthesizing the internal forces of the corresponding truss members. Then, internal force–level assessment of the deck sections is carried out for various loading conditions such as monsoon, typhoon, and with and without railway traffic.

Improving transportation infrastructure inspection methods and the ability to assess conditions of bridges has become a priority in recent years as the transportation infrastructure continues to age. Current bridge inspection techniques consist largely of labor-intensive subjective measures for quantifying deterioration of various bridge elements. Some advanced nondestructive testing techniques, such as ground- penetrating radar, are being implemented; however, little attention has been given to remote sensing technologies. Remote sensing technologies can be used to assess and monitor the condition of bridge infrastructure and improve the efficiency of inspection, repair, and rehabilitation efforts. Most important, monitoring the condition of a bridge using remote sensors can eliminate the need for traffic disruption or total lane closure because remote sensors do not come in direct contact with the structure. The purpose of this paper is to evaluate 12 potential remote sensing technologies for assessing the bridge deck and superstructure condition. Each technology was rated for accuracy, commercial availability, cost of measurement, precollection preparation, complexity of analysis and interpretation, ease of data collection, stand-off distance, and traffic disruption. Results from this study demonstrate the capabilities of each technology and their ability to address bridge challenges.

Comprehensive structural health monitoring systems have been developed and installed in several long-span suspension bridges around the world, aiming to monitor structural health conditions of the bridges in real time. Nevertheless, many key issues remain unsolved, such as how to take full advantage of the health monitoring system for effective and reliable damage detection of these complex structures. An innovative testbed was therefore established in a laboratory to allow researchers to recreate rational damage scenarios, apply different sensors and sensing networks, and test various damage detection algorithms. The design principles of the laboratory-based testbed are introduced. The paper will then outline the design and setup of a physical model for a long-span suspension bridge, which will consider various damage scenarios. Geometric measurements and modal tests were subsequently carried out to identify its geometric configuration and dynamic characteristics, respectively. The finite-element modeling of the physical bridge model was finally established using a commercial software package, which was followed by a finite-element model updating the use of the measured modal properties. This testbed, comprising of the delicate physical model and the updated finite-element model of a long-span suspension bridge, could solve a benchmark problem for the structural health monitoring of long-span suspension bridges.

An increasing number of online structural health monitoring systems have been implemented on bridges, which are often comprised of a variety of local sensing and nondestructive evaluation (NDE) devices. However, before any new type of sensor and NDE technology, such as ultrasonic surface wave sensors, can be deployed for field implementation, they first need to be validated and characterized under realistic loads and environmental conditions in a controlled laboratory setting. This paper demonstrates the use of a hybrid testing setup to create a characterization testbed for more realistic evaluation of NDE and sensor technologies. A prototype bridge structure based on the Yellow Mill Pond Bridge in Connecticut was selected to demonstrate the features and issues of a hybrid simulation testbed for sensor characterization, such as flexibility in traffic load representation, and large-scale specimens even with small testing equipment. Using this prototype bridge structure, a series of hybrid simulation tests using real variable-amplitude traffic loading were performed on a 0.6-scale steel flange plate specimen with a welded cover plate. Features of this hybrid simulation testbed are discussed, including the complexity of the virtual model (e.g., the dynamic mass effects and damping of the structure) and stability issues that can develop in the test. Finally, the ability of the hybrid testing platform to evaluate sensor performance under varying environmental conditions was demonstrated for elevated temperatures. Although a strain gauge was used to demonstrate the technology, the sensor testbed can be used for general characterization of other types of NDE devices and sensors.

Safety of suspension or stay cable bridges depends on the durability of their cables; stay cable bridges may present a healthy aspect on observable lengths but may be damaged in the nonvisible parts (anchorages). It is important to be able to detect as early as possible the defects affecting them. In this study, the acoustic emission (AE) technique is employed to detect the presence and location of broken wires in anchorage. Allowing a cable to vibrate can induce AE from interwire fretting. The study of the behavior of a broken wire during bending contributed to the understanding of the origin of the AE, that is, interwire fretting and the definition of the most suited bending conditions. For the identification of interwire friction, the main parameters of the acoustic signals are the number of events, their energy, and frequency distribution. These parameters are very dependent of roughness, lubrication, contact strength between wires, and recovery length. Several cables were studied with various surface conditions including corroded, dry, and oiled.

Two different nondestructive test (NDT) methods, air-coupled impact-echo (IE) and infrared (IR) thermography are evaluated on a full-scale simulated reinforced concrete bridge deck containing simulated delamination and cracking defects. The IE data are presented as two-dimensional frequency maps and spectral B-scan lines. The IR data are presented as temperature maps on the concrete surface. The lateral boundaries of the detected delaminations are also indicated in the images. The results obtained from each of the individual NDT methods show reasonably good agreement with most of the actual defects. The advantages and limitations of each method to characterize defects are discussed. The consistency and sensitivity of each method are also investigated. Finally, a simple data fusion technique is proposed to improve effectiveness of the individual test data. The findings from this study demonstrate that the combination of air-coupled IE and IR thermography tests is a practical option for consistent and rapid in situ evaluation of reinforced concrete bridge decks.

An ultrasonic P-wave reflection imaging probe is developed and utilized to noninvasively image the geometry of a deep foundation supporting a bridge pier. The source ultrasonic transducer emits compressive waves into saturated soil that subsequently transmits to and reflects back from an embedded object, and the receiver transducer measures the reflections that are used to construct an image. The components of the system, including the custom transducer probe and data acquisition hardware, were integrated with the nees@UCLA cone penetration testing truck. Some fundamentals of propagation of ultrasonic waves in soil are presented first, including transducer directivity, effects of unsaturation, the size of the internal scale of the soil, and transducer coupling. The system components are then presented, followed by field testing. The probe successfully imaged a pile foundation in very soft saturated clay but not in stiffer and/or unsaturated soils. The probe could be very useful for nondestructive quality assurance of structural elements constructed in situ in soft saturated soils.

Delamination of the concrete cover above the upper reinforcing bars is a common problem in concrete bridge decks. Acoustic nondestructive evaluation is widely used to detect such delamination because of its low cost, fast speed, and ease of implementation. The accuracy of traditional acoustic approaches is dependent on the level of ambient noise, and the detection process is highly subjective. An automatic impact-based delamination detection (AIDD) system is described in this paper. In this system, the traffic noise is eliminated by a modified version of independent component analysis. Mel-frequency cepstral coefficients are then used as features for detection to eliminate subjectivity. The delamination detection is performed by a radial basis function neural network. The AIDD system was developed using mixed-language programming in MATLAB, LabVIEW, and C++. The performance of the system was evaluated using data from two bridges, and the results were satisfactory.

Through modern nondestructive inspection (NDI) techniques, fatigue cracks can be detected with varying accuracies in steel bridge structures, and consequently assessment and updating of fatigue reliability can be performed. A probabilistic fracture mechanics method based on the Bayes theorem using information from NDI has been demonstrated to update the relevant parameters of crack growth models effectively. Using the probabilistic characteristics of the NDI techniques, the probabilistic fatigue failure method, and the Bayes theorem, a procedure for assessing and updating the fatigue reliability of existing steel bridge components is presented. Data collected from two steel bridges, the Zhejiang Street Bridge and Ganjiang River Bridge, are used to illustrate the fatigue reliability assessment and updating method. In the case study, the suspender $\beta ‐T$ curves of the Zhejiang Street Bridge and Ganjiang River Bridge are determined after inspection and updating. It is found in this research that the fatigue reliability of the selected members in the existing steel bridges increases with no cracks detected, decreases with no size cracks detected, and changes remarkably with cracks detected.

This paper discusses monitoring of bonded posttensioned (PT) concrete elements using the acoustic emission technique. In particular, a statistical pattern recognition technique based on a multivariate outlier analysis is presented to identify initial yielding and the onset of failure. Experimental tests on large-scale single-tendon bonded PT concrete beams, subjected to multiple load cycles, will be presented to validate the proposed monitoring system.

Under certain weather conditions, the Veterans’ Glass City Skyway (VGCS) Bridge in Toledo, Ohio, accumulates ice on the stay sheaths in such a way that has led to large fragments falling into traffic lanes. To aid the Ohio Department of Transportation (ODOT) in their response to this, an automatic inference and monitoring system was implemented. This system utilizes existing weather station measurement data to infer possible events, alerting officials and providing tools to further assess the situation. The development of the decision, the algorithm, and implementation process are presented. The system was implemented in January 2011, and it warned of and monitored an icing event that occurred February 20–24, 2011. Ice fell shortly after the system indicated conditions were met for icefall. ODOT personnel used the ice inference system to assist them in managing the response to the icing event.

Vibration-based damage detection techniques typically use changes of modal characteristics of bridges as a possible indication of damage. However, structural damages are not the only cause of these changes. This study investigates the effect of temperature variations on modal characteristics of a two span steel-concrete composite bridge in North Carolina, and addresses the extent and reason of the daily changes observed in its dynamic properties. The field testing included measuring vibration responses, deflections, and temperatures of the bridge throughout a summer day. The dynamic characteristics of the bridge, derived from measurements at different times of the day, were compared with each other. Recorded temperatures and deflections of the bridge deck were used to address the observed changes in the natural frequencies of the bridge. Results of the field testing showed that temperature variations can induce modal variability on a daily cycle. A primary reason for this observation can be attributed to the temperature gradients measured on the bridge deck from night to noon.

Magnetic methods are progressing in the detection of corrosion in prestressing strands in adjacent precast, prestressed concrete box-beam bridges. This study is the first field trial of magnetic strand defect detection systems on an adjacent box-beam bridge. A bridge in Fayette County, Ohio, that was scheduled for demolition was inspected. The prestressed box beams had significant strand corrosion. The corroded strands showed discontinuities and a reduced cross-sectional area. These changes are reflected in the magnetic signatures of the prestressing steel. Corrosion in the prestressing steel was detected using two magnetic methods; namely, the magnetic flux leakage method and the induced magnetic field method. The purpose of these tests was to demonstrate the ability of the magnetic methods to detect hidden corrosion in box beams in the field and tackle the logistic problem of inspecting box beams from the bottom. The inspections were validated by dissecting the bottom of the box beams after the inspections. The results showed that the magnetic flux leakage method detects hidden corrosion and strand breaks with a sufficient amount of accuracy. Both magnetic field methods were also able to estimate corrosion by detecting the effective cross-sectional area of the strand. Thus, it was shown that the magnetic methods can be used with sufficient reliability to predict hidden corrosion in prestressing strands of box beams. The recommended actions to make magnetic inspection practical are discussed.