• Steel and Composite Structures
• /
• v.10 no.4
• /
• pp.297-315
• /
• 2010

This paper is concerned with free vibration characteristics and natural frequency of horizontally curved composite plate girder bridges. Three-dimensional finite element models are developed for the girders using the software package LUSAS and analyses carried out on the models. The validity of the finite element models is first established through comparison with the corresponding results published by other researchers. Studies are then carried out to investigate the effects of total number of girders, number of cross-frames and curvature on the free vibration response of horizontally curved composite plate girder bridges. The results confirm the fact that bending modes are always coupled with torsional modes for horizontally curved bridge girder systems. The results show that the first bending mode is influenced by composite action between the concrete deck and steel beam at low subtended angle but, on the girders with larger subtended angle at the centre of curvature such influence is non-existence. The increase in the number of girders results in higher natural frequency but at a decreasing rate. The in-plane modes viz. longitudinal and arching modes are significantly influenced by composite action and number of girders. If no composite action is taken into account the number of girders has no significant effect for the in-plane modes.

• Steel and Composite Structures
• /
• v.11 no.1
• /
• pp.21-38
• /
• 2011

This study examines the influence of curved, steel, I-girder bridge configuration on girder end reactions and cross frame member forces during seismic events. Simply-supported bridge finite element models were created and examined under seismic events mimicking what could be experienced in AASHTO Seismic Zone 2. Bridges were analyzed using practical ranges of: radius of curvature; girder and cross frame spacings; and lateral bracing configuration. Results from the study indicated that: (1) radius of curvature had the greatest influence on seismic response; (2) interior (lowest radius) girder reactions were heavily influenced by parameter variations and, in certain instances, uplift at their bearings could be a concern; (3) vertical excitation more heavily influenced bearing and cross frame seismic response; and (4) lateral bracing helped reduce seismic effects but using bracing along the entire span did not provide additional benefit over placing bracing only in bays adjacent to the supports.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2010.05a
• /
• pp.79-80
• /
• 2010

Age-dependent laboratory tests for a curved composite box girder were carried out to investigate age-dependent effects of concrete on toraional behavior of a curved girder. Time-dependent incremental finite element method predicted the toraional behavior as well as flexural including axial behaviors of the test specimen.

• Steel and Composite Structures
• /
• v.37 no.2
• /
• pp.117-136
• /
• 2020

Curved steel-concrete composite box girder has been widely adopted in urban overpasses and ramp bridges. In order to investigate its mechanical behavior under complicated and combined bending, shear and torsion load, two large-curvature composite box girders with interior angles of 25° and 45° were tested under static hogging moment. Based on the strain and deflection measurement on critical cross-sections during the static loading test, the failure mode, cracking behavior, load-displacement relationship, and strain distribution in the steel plate and rebar were investigated in detail. The test result showed the large-curvature composite box girders exhibited notable shear lag in the concrete slab and steel girder. Also, the constraint torsion and distortion effect caused the stress measured at the inner side of the composite beam to be notably higher than that of the outer side. The strain distribution in the steel web was approximately linear; therefore, the assumption that the plane section remains plane was approximately validated based on strain measurement at steel web. Furthermore, the full-process non-linear elaborate finite element (FE) models of the two specimens were developed based on commercial FE software MSC.MARC. The modeling scheme and constitutive model were illustrated in detail. Based on the comparison between the FE model and test results, the FE model effectively simulated the failure mode, the load-displacement curve, and the strain development of longitudinal rebar and steel girder with sufficient accuracy. The comparison between the FE model and the test result validated the accuracy of the developed FE model.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.8 no.1
• /
• pp.9-14
• /
• 2008

In the case of horizontally curved bridges, the use of curved composite box girder bridges are increased due to its functionality and for aesthetical reason. As it compared with the open section, the steel box girder bridges have advantages to resistant of distortion and corrosion. In practice the grid analysis is conducted by utilizing only the cross beam. Since the stiffness of the concrete slab is not included in the grid analysis, the cross beam is induced the distribution of the live load. In this study the affects of the radius of curvature, the number of diaphragm and cross beam to the load distribution of the curved steel box girder bridge was investigated by applying the finite element method. The results indicate that the curvature of curved bridge had a large affect of the load distribution and as the curvature was increased the load distribution factor was increased. A single diaphragm at the center of girder is important role for the load distribution effects and structural stability, but additional diaphragm did not affect it as much. The affects of the cross beam to the load distribution were investigated and its influence was minor. It can be safely concluded that the addition of cross beam does not aid the purpose of the live load distribution. And the stiffness of concrete slab for the load distribution effects should be concerned in the design of curved steel box girder bridges.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.5A
• /
• pp.407-415
• /
• 2010

In Korea, more than 90% of the total number of steel bridges built for 40~70 m span length is a steel box-girder bridge type. A steel box-girder bridge is suitable for long span or curved bridges with outstanding flexural and torsional rigidity as well as good constructability and safety. However, a steel box-girder bridge is uneconomical, requiring many secondary members and workmanship such as stiffeners and ribs requiring welding attachments to flanges or webs. Therefore, in US and Japan, a plate girder bridge, which is relatively cheap and easy to construct is generally used. One type of the plate girder bridge is the two- or three-main girder plate bridge, which is a composite plate girder bridge that minimizes the number of required main girders by increasing the distance between the adjacent girders. Also, for the simplification of girder section, the stiffener which requires attachment to the web is not required. The two-main steel girder plate bridge is a representative type of plate girder bridges, which is suitable for bridges with 10 m effective width and has been developed in the early 1960s in France. To ensure greater safety of two- or three-main girder plate bridges, a larger steel section is used in the bridge domestically than in Europe or Japan. Also, the total number of two- or three-main girder plate bridge constructed in Korea is significantly less than the steel box girder bridge due to a lack of designers' familiarity with more complex design detailing of the bridge compare to that of a steel box girder bridge design. In this study, a new construction method called Turn Over method is proposed to minimize the steel section size used in a two- or three-main girder plate bridge by applying prestressing force to the member using confining concrete section's weight to reduce construction cost. Also, a full scale 20 m Turn Over girder specimen and a Turn Over girder bridge specimen were tested to evaluate constructability and structural safety of the members constructed using Turn Over process.

• Steel and Composite Structures
• /
• v.29 no.2
• /
• pp.161-173
• /
• 2018

Point bending is often used for cambering and curving structural steel girders. An analytical solution, applicable in the elasto-plastic range only, that relates applied loads to the desired curve was recently developed for inducing horizontal curves using four-point bending. This solution does not account for initial residual stresses and geometric imperfections built-in hot-rolled sections. This paper presents results from a full-scale test on a hot-rolled steel section curved using four-point bending. In parallel, a numerical analysis, accounting for both initial geometric imperfections and initial residual stresses, was carried out. The models were validated against the experimental results and a good agreement for lateral offset and for strain in the elasto-plastic and post-plastic ranges was achieved. The results show that the effect of initial residual stresses on deformation and strain is minimal. Finally, residual stresses due to cold bending calculated from the numerical analysis were assessed and a revised stress value for the service load design of the curved girder is proposed.

• Journal of Korean Society of Steel Construction
• /
• v.28 no.5
• /
• pp.365-371
• /
• 2016

In the case of the superstructure which is consist of two I girders and slab, the section can behave as II section, so that the neutral axis with respect to out of plane direction flexure can be regarded as major axis. Therefore in-plane flexural mode might govern the free vibration mode. Meanwhile, horizontally curved girders always experience not only bending moments but also torsional moments although the primary load is usually supposed to be gravitational load. The interaction due to bending and torsional moments make the behavior complicated and torsional mode may govern the free vibration mode. In other words, structure can have different dynamic characteristic due to its initial curvature. In this research, using 3-dimensional sell elements, free-vibration analyses are carried out due to initial curvature. The analysis models are assumed to be composite and non-composite and finally natural frequency and eigen mode are discussed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.6 no.2
• /
• pp.17-22
• /
• 1986

The primary objective of this study is to compare the effects that are caused by shrinkage and creep of a concrete bridge deck during its construction. In this study four different bridge structures were compared. Two straight box girders and two curved box girders were compared for stress changes in positive moment region and negative moment region due to the effects of concrete. The effects on displacement behavior by the assumed section length by concrete placement were also studied. The analyses were performed by using Vlasov equation and finite difference numerical method to solve the governing differential equation.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.11
• /
• pp.520-526
• /
• 2019

I-type girders are widely applied as very economical sections in plate girder bridges. There has been research on developing composite laminated panels, curved plates reinforced with closed-end ribs, and new forms of ribs and compression flanges for steel box girders. However, there is a limitation in analyzing the exact cause of local buckling caused by an I-type girder's webs. Therefore, an I-type girder's web was modeled using the finite element analysis program LUSAS 17.0 before and after reinforcement. We checked for the minimum thickness criteria presented in the Korea highway bridge design code, and the cause of buckling after performing a linear elastic buckling analysis of dead and live loads was analyzed. Before reinforcement, an eigenvalue (λ₁) at the 1st mode was 0.7025, the critical buckling load was smaller than the applied load, and there is a buckling. After reinforcement, when applying vertical and horizontal stiffeners to the web part of the girder at support, a Nodal line was formed, the eigenvalue was 1.5272, and buckling stability was secured. To improve buckling trace of the girder at the support, an additional plate was applied to the web at the support to ensure visual and structural safety, but buckling occurs at center of web. The eigenvalue (λ₁) was 3.5299, and this method is efficient for reinforcing the web of the support.