Journal of Bridge Engineering

Structural analysis methods for steel I-girder bridges can be grouped in three categories: 1D line-girder, 2D-grid, and 3D finite-element analysis. This paper discusses the qualities and limitations of these techniques, and it focuses on improvements that may be implemented to 2D-grid methods to capture horizontal curvature and support skew effects effectively at little or no additional computational cost compared to traditional 2D-grid methods. The paper focuses predominantly on the problem of analyzing the noncomposite response of the structure during construction, which is of essential importance for setting girder cambers and controlling the constructed geometry. The proposed modeling techniques also can be applied to the analysis of partially composite structures during staged deck placement and to the analysis of a structure in final composite condition. Analysis results for several example bridges with challenging geometries are presented to demonstrate the application and show the impact of the proposed modeling improvements.

A dense deployment of sensors is often beneficial for obtaining a more complete understanding of the dynamic behavior of complex structures, such as long-span bridges. Wireless smart sensor networks are advantageous in this regard because they are less expensive and easier to deploy than traditional wired sensor networks. A dense array of wireless smart sensor networks with 113 smart sensors was installed on a cable-stayed bridge in South Korea to record ambient acceleration. The deployment features the world’s largest wireless smart sensor network for civil structural monitoring. In total, 10 GB of data were collected from the deck sensor nodes during a 1-year period of deployment. The natural excitation technique, in conjunction with the eigensystem realization algorithm, was used to perform system identification. A comprehensive statistical analysis of modal properties, such as natural frequencies, modal damping, and mode shapes of the bridge, was performed. The relationship between temperature and natural frequency was also examined. The statistical analysis of system identification results demonstrates that using data obtained by a wireless smart sensor network provides a high resolution and confidence in identified modal properties.

A novel demountable shear connector for precast steel-concrete composite bridges is presented. The connector uses high-strength steel bolts, which are fastened to the top flange of the steel beam with the aid of a special locking nut configuration that prevents bolts from slipping within their holes. Moreover, the connector promotes accelerated construction and overcomes the typical construction tolerance issues of precast structures. Most importantly, the connector allows bridge disassembly. Therefore, it can address different bridge deterioration scenarios with minimum disturbance to traffic flow including the following: (1) precast deck panels can be rapidly uplifted and replaced; (2) connectors can be rapidly removed and replaced; and (3) steel beams can be replaced, whereas precast decks and shear connectors can be reused. A series of push-out tests are conducted to assess the behavior of the connector and quantify the effect of important parameters. The experimental results show shear resistance, stiffness, and slip capacity significantly higher than those of welded shear studs along with superior stiffness and strength against slab uplift. Identical tests reveal negligible scatter in the shear load-slip displacement behavior. A design equation is proposed to predict the shear resistance with absolute error less than 8%.

This paper presents a general approach for evaluating wind load factors based on measured wind data for RC columns. Reliability analyses were carried out by an advanced first-order second-moment reliability method to investigate the wind load–governed limit state for the RC pylons of five cable-supported bridges in Korea. The axial force-bending moment (P-M) interaction diagrams of the pylons define the limit-state function. Load and strength parameters were considered random variables in the reliability analysis. The statistical parameters of wind load were determined by Monte Carlo simulation. Based on the results of the reliability analysis, dead load factors were set to the bias factors of dead load components, and a P–M interaction diagram drawn by the mean values of the strength parameters was used to define a design equation. The failure point of the wind load was obtained by equating the nonexceedance probability at the failure point of wind load to the probability of safety corresponding to a given reliability index. An analytical form of the wind load factor was derived in terms of the statistical parameters of wind load and the target reliability index. The wind load factor was adjusted to be used with the dead load factors and the resistance factor specified in a reliability-based design code. Validity of the proposed load factors was verified through a reliability assessment of the pylon sections of the five bridges. It is shown that the proposed load factors secure the target reliability levels within a 2% error.

The installation of dampers close to cable anchorages is a common approach for stay-cable vibration mitigation. Inspired by the force-deformation relationship produced by actively controlled dampers, passive negative-stiffness dampers (NSDs) were proposed for stay cables in the past to achieve excellent vibration-control performance. However, a systematic comparison between passive NSDs and active control in cable vibration mitigation has rarely been reported in literature. This paper systematically compares vibration mitigation performances of a passive NSD to the performance of two active control methods, linear quadratic regulator (LQR) and output feedback control. The comparison indicates that a passive NSD can offer a stay cable with a high damping level comparable to that of LQR control. However, passive NSD will also decrease the modal frequencies of a stay cable, whereas LQR will increase the frequencies slightly. The dynamic response results also indicate that the active LQR control offers slightly better control performance than the passive NSD in various loading cases. The superiority of the LQR control over the passive NSD can be explained through an output feedback control approach. It is noteworthy that the NSD is regarded as more practical and reliable because it offers protection by completely passive means without the need for the feedback and actuation that are required in active control.

In this paper, a vibration-based approach for estimating the cable tension forces in cable-stayed bridges is detailed and applied to the ambient response of the Veterans’ Glass City Skyway Bridge. First, the frequency stabilization diagrams were derived using the covariance-driven stochastic subspace identification (SSI-COV) method. Second, natural frequencies of each stay cable were identified by applying the hierarchical clustering algorithm to the frequency stabilization diagrams. Finally, the identified natural frequencies along with measurable physical and geometrical characteristics of the stay cables were used as input to a cable model that includes geometric nonlinearity and flexural stiffness to obtain the stay-cable tensions. Following the proposed methodology, the combination of the SSI-COV method and hierarchical clustering algorithm can be used as a complementary tool alongside the Fourier transform-based analysis of the cable responses to make a distinction between physical and spurious modes of the stay cables.

An experimental program was performed to investigate the postfracture performance of composite twin I-girder bridges in the fatigue crack condition. An intact specimen and a damaged specimen were tested under one-point load to understand the effect of fracture on the bridge response and safety. In addition, nonlinear numerical analyses were performed to study the mechanical behavior of the models. The results showed that the fracture of the entire web and bottom flange at the midspan section can result in a considerable increase in the shear strain on the studs near the middle section. The fracture of the bridge in the midspan section governed by bending failure is found to be most critical in terms of the remaining load-carrying capacity, and the fracture near the support location governed by shear failure is found to be brittle, which can result in a sudden collapse of the bridge system. The effects of the fatigue crack on the safety of the composite twin I-girder bridge can be mitigated by using a continuous span instead of a simply supported span or by using a stronger and thicker concrete slab.

Built-up steel girders have been shown to be resistant to total member collapse in the event that a single component fails from brittle fracture or from fatigue. However, accurate methods to determine the remaining capacity of members with partially failed cross sections have not been addressed. Because the failure of a single component represents a localized failure, a simple flexural stress calculation of a member cross section, using only the remaining components, does not represent the actual member capacity. Additionally, a traditional linear stress distribution over the height of a flexural member is not accurate due to the load transfer, engagement of mechanical fasteners, and remaining portions of unfailed components. This paper will present the methods used in determining the localized stress distribution as well as how to estimate the appropriate levels of stress in partially failed built-up steel members. Results of experimental fatigue tests are presented along with an analytical parametric study investigating the localized stresses near a partially failed component. This information is critical in determining the strength capacity of members who have experienced a failure in one of the components, as well as determining the remaining fatigue life of built-up girders subjected to vehicular loads.

Conventional load-rating methods for bridge timber piles consider only concentric axial loads. However, in the case of abutment piles, earth pressure and surcharge loads exert a significant lateral force on the piles. There is need for a load-rating procedure that can account for combined axial loads and bending from lateral loads. This paper proposes a load-rating method specifically for deteriorated and fiber-reinforced polymer (FRP) retrofitted abutment timber piles. The new load-rating method accounts for various factors, including dead load, pile length, earth pressure, and timber pile deterioration. The proposed load-rating method is demonstrated on a prototype timber pile bridge model using finite-element methods. Results show that timber pile deterioration affecting as low as 10% of the pile cross section can significantly reduce the load-carrying capacity of abutment timber piles. In addition to the load-rating technique, this paper introduces a simplified elastic design method for retrofitting the tensile face of abutment timber piles using FRP composites. Both the load-rating and FRP design methods are presented as a series of charts that engineers can easily use to aid their decision-making process.

This study evaluated the fatigue performance of rib-to-deck welded joints in orthotropic steel decks (OSDs) with thickened edge U-ribs (TEUs). Fatigue performance of four rib-to-deck specimens with TEUs was investigated and compared with those of three rib-to-deck specimens with conventional U-ribs. Overall, the use of TEUs enhanced the fatigue strength of rib-to-deck joints. Finite-element modeling was conducted and validated by experimental data and sensitivity analyses. The validated finite-element model was then used to investigate the mechanism for the enhancement of fatigue strengths caused by use of TEUs and to evaluate the effects of weld penetration and open angle of rib-to-deck joints. Findings show that use of TEUs significantly improved the robustness of welded joints in terms of sensitivity to variations in the weld penetration. Appropriate increases in the open angle of rib-to-deck joints in OSDs with TEUs further improved fatigue performances. Finally, two full-scale orthotropic steel deck specimens were tested to further validate the effectiveness of TEUs.

This paper presents a five-step algorithm to determine the reasonable state of suspension bridges with spatial cables without theoretical derivation or original programming. The algorithm is extended from the common FEM analysis method with additional algebraic operation and flow control. The five-step algorithm is an efficient and convenient method for the general bridge designers in normal engineering practice and can be summarized as follows. In the first four steps, a main cable–only model is used considering the couple of the spatial main cable and inclined hangers. First, a set of ideal vertical components of the hanger forces is obtained by the rigid-supported continuous beam method. Second, the shape of the main cable in the midspan is calculated iteratively based on the designated sag. Third, the reasonable state of the main cable in the side span is determined by the force equilibrium of the IP point. Fourth, the anchor span is solved by the equilibrium of the splay saddle. Finally, a full-bridge model is activated and specially handled to complete the analysis. Next, a 1,588-m midspan suspension bridge with spatial cables is introduced as a case study. Several details and strategies are also presented. It is proved that the algorithm can avoid the theoretical derivation and simplify the programming on the premise of the necessary numerical precision.