• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.325-334
• /
• 2023

Computer vision-based damage detection enables non-contact, efficient and low-cost bridge health monitoring, which reduces the need for labor-intensive manual inspection or that for a large number of on-site sensing instruments. By leveraging recent semantic segmentation approaches, we can detect regions of critical structural components and identify damages at pixel level on images. However, existing methods perform poorly when detecting small and thin damages (e.g., cracks); the problem is exacerbated by imbalanced samples. To this end, we incorporate domain knowledge to introduce a hierarchical semantic segmentation framework that imposes a hierarchical semantic relationship between component categories and damage types. For instance, certain types of concrete cracks are only present on bridge columns, and therefore the noncolumn region may be masked out when detecting such damages. In this way, the damage detection model focuses on extracting features from relevant structural components and avoid those from irrelevant regions. We also utilize multi-scale augmentation to preserve contextual information of each image, without losing the ability to handle small and/or thin damages. In addition, our framework employs an importance sampling, where images with rare components are sampled more often, to address sample imbalance. We evaluated our framework on a public synthetic dataset that consists of 2,000 railway bridges. Our framework achieves a 0.836 mean intersection over union (IoU) for structural component segmentation and a 0.483 mean IoU for damage segmentation. Our results have in total 5% and 18% improvements for the structural component segmentation and damage segmentation tasks, respectively, compared to the best-performing baseline model.

• Journal of Korean Society of Steel Construction
• /
• v.17 no.5 s.78
• /
• pp.549-559
• /
• 2005

A railway bridge with a double composite section is proposed to enhance the structural performance of existing two-girder bridges because the governing design parameter of railway bridges is the flexural stiffness. The concrete deck in negative moment regions is neglected in the design of continuous composite bridges assuming the concrete slab has no resistance to tension. Therefore, the flexural stiffness of the composite section in the negative moment region is reduced resulting in the increase of the depth of the steel section. In order to resolve this disadvantage, several methods are suggested and the double composite section is one of the excellent solutions for extending the span length and increasing the flexural stiffness. In this study, push-out tests on lying studs and mixed stud shear connection with lying and vertical studs were performed to investigate the behavior of the shear connection in the double composite section. Static strength of the shear connection was evaluated through the test results and numerical analyses.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.633-640
• /
• 2023

In this study, numerical analysis was performed based on field investigation to derive an appropriate reinforcement method by analyzing the displacement behavior characteristics of concrete blocks generated in the direct fixation track on the bridges of the serviced light rail transit. The track of this study was a direct fixation track on a sharp curved track, and the problem of movement of the concrete blocks installed on the bridge deck in the longitudinal and lateral directions occurred. In this study, based on the finite element model using 3D solid elements, the behavior of the direct fixation track that could be occurred under operating load conditions was analyzed. In addition, the reinforcement effect of various reinforcement methods was analyzed. As a result of analyzing the lateral displacement before and after reinforcement, it was analyzed that the maximum lateral displacement after reinforcement under the extreme lateral wheel loads significantly decreased to about 3% (about 0.1mm) compared to before reinforcement. In addition, as a result of examining the generated stress of the filling mortar, bridge decks, and reinforcing bar, it was analyzed that all of them secured a sufficient safety factor of 2.6 or higher, and the optimal conditions for the reinforcement method were derived. Therefore, it is judged that the number of anchoring reinforcements and symmetrical anchor placement reviewed in this study will be effective in controlling the occurrence of lateral displacement of concrete blocks and securing the structural integrity of bridges and concrete blocks.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.973-978
• /
• 2011

In this study, effects due to differences of creep and shrinkage which is caused by material differences such as concrete characteristic and age in an asymmetric PSC box girder were evaluated. For this purpose the prestress distribution, creep and shrinkage were analyzed with the FE analysis program, LUSAS 14.3. As a result of the prestress, the stress distribution was stable. In case of the analysis result which was conducted with 1,000 days response time, the shear stress between PC section and CIP section is satisfied with design shear strength.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.1056-1061
• /
• 2007

Railroad roadbed was consisted into structure types that earthwork, tunnel, bridge and joint sections. Joint section was affected a large factor confidence and safety of the train running by stiffness transition zone that track substructure stiffness change section as like between tunnel and earthwork from ballast track to concrete track. These problems are the results of increased dynamic wheel loads, which also lead to wear and tear on vehicle components and contribute to poor ride quality. The study presented in this paper was conducted on model test by using Wheel Moving Loading System.

• Proceedings of the KSR Conference
• /
• 2008.11b
• /
• pp.1012-1017
• /
• 2008

In this study, We examine closely the capacity spectrum method which a kind of displacement-based method evaluated by displacement of structure as an alternative to the load-based analysis method. The displacement-based method can easily review the strength of structure, seismic performance, ductility. Seismic performance by using capacity spectrum method is divided into design response spectrum and capacity spectrum. We can diagram design response spectrum by deciding the design seismic factor depending on performance target, site classification, seismic level, return period as UBC-97. Capacity spectrum is a load-displacement curve obtained by Push-over analysis considering the geometric parameter and the material parameter. We execute the seismic performance evaluation by using the capacity spectrum method to reinforced concrete pier which has been seismic design. As a result, We confirmed that there is a yield point and a ultimate point close by design response spectrum of UBC-97.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.965-972
• /
• 2011

Unlike metallic materials, the importance of fatigue performance of concrete has been ignored. However, it is reported that environmental effects, if it cause deterioration, may increase the risk of fatigue failure under repeated loadings. In case of railroad bridges, the risk may increase due to highly periodic, repetitive, heavier nature of train load, which runs through the fixed passage called the track. Especially, when new material or structure is implemented for a main bridge member, experimental validation should be performed to avoid damage or failure due to unexpected behavior. In this paper, the fatigue performance of an IT girder is examined via a repeated loading test. The IT girder is a new type of a prestressed concrete (PSC) girder with two prestressed H-beams in the top of the girder, which provide additional sectional capacity, and it can be applied to the span longer than 30m which is a typical limit for a usual PSC girder. To obtain the fatigue performance, a 10m IT girder specimen is designed, and a repeated load test is performed by applying the cyclic load two million times. The fatigue performance of the girder is examined according to the Japanese and the CEB-FIB design codes. The fatigue test result shows that the IT girder satisfies both design codes.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.532-542
• /
• 2018

Because maglev trains have a propulsion and absorption force without contact with the rails, they can drive safely at high-speed with little oscillation. Recently, test model of a maglev propulsion train was produced and operated, and has since been chosen as a national growth industry in South Korea; there have been many studies and considerable investment in these fields. This study examined the dynamic responses due to bridge-maglev train interaction and basic material to design bridges for maglev trains travelling at high-speed. Depending on the major factors affecting the dynamic effects, the scope of this study was restricted to the relationship between dynamic responses. A concrete box girder was chosen as a bridge model and injured train and rail types in domestic production were selected as the moving train load and guideway analysis model, respectively. From the analysis results, the natural frequency of a bridge for a maglev train, which has a deflection limit L/2000, was higher than those of bridges for general trains. The dynamic responses of the girder of the bridge for a maglev train showed a substantial increase in proportion to the velocities of the moving train like other general bridge cases. Maximum dynamic response of the girder is shown at a moving velocity of 240km/h and increased with increasing moving velocity of train. These results can be used to design a bridge for maglev propulsion trains and provide the basic data to confirm the validity and verification of the design code.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.15 no.2
• /
• pp.51-60
• /
• 2011

In this study, structural safety capacity analysis of the overpass railway bridge between Konkuk Univ. and Guui station railroad has been performed. The overpass is expected to have suffered durability reduction by deterioration. The weight reduction of the overpass has been implemented to prevent further durability reduction and to improve performance capacity. To reduce the weight, 3 procedures of replacing concrete soundproofing wall to light-weight soundproofing wall, replacing gravel ballast to concrete ballast, and reducing the weight of trough have been performed. The analysis of static/dynamic behaviors and improved capacity of the light-weighted overpass bridge has been performed. The structural safety verification of the improved structure has been implemented by using rating factors of load carrying capacity of PSC I girder. The results have shown that the deflection has been reduced by 2.6mm and tensile strength has been improved by 1.07MPa, which indicate that the structural capacity has effectively been improved. Also, the natural frequency has improved by approximately 30% where vibration reduction and dynamic behavior improvement have been achieved. Moreover, in the rating factor evaluation based on analysis and test results, an improvement from 1.82 to 1.93 has been observed. Therefore, weight reduction method for the overpass is effective considering overall results.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.27 no.4
• /
• pp.305-312
• /
• 2007

Among many types of light rail transits (LRT), the rubber-tired automated guide-way transit (AGT) is prevalent in many countries due to its advantages such as good acceleration/deceleration performance, high climb capacity, and reduction of noise and vibration. However, AGT is generally powered by high-voltage electric power feeding system and it may cause electromagnetic interference (EMI) to measurement sensors. The fiber optic sensor system is free from EMI and has been successfully applied in many applications of civil engineering. Especially, fiber Bragg grating (FBG) sensors are the most widely used because of their excellent multiplexing capabilities. This paper investigates a prestressed concrete girder bridge in the Korean AGT test track using FBG based sensors to monitor the dynamic response at various vehicle speeds. The serviceability requirements provided in the specification are also compared against the measured results. The results show that the measured data from FBG based sensors are free from EMI though electric sensors are not, especially in the case of electric strain gauge. It is expected that the FBG sensing system can be effectively applied to the LRT railway bridges that suffered from EMI.