• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.7
• /
• pp.1259-1266
• /
• 2022

Recently, due to the specialization of structural design standards and evaluation methods, the classification rules are being integrated. A good example is the common international rules (CSR). However, detailed regulations are presented only for the cargo hold area where the longitudinal load is greatly applied, and no specific evaluation guidelines exist for the bow and stern structures. Structural design of the mentioned area is carried out depending on the design experience of the shipbuilder, and because no clear standard exists even in the classification, determining the root cause is difficult even if a structural damage problem occurs. In this study, an engineering-based solution was presented to identify the root cause of representative cases of buckling damage that occurs mainly in the stern. Buckling may occur at the panel wall owing to hull girder bending moment acting on the stern structure, and the plate thickness must be increased or vertical stiffeners must be added to increase the buckling rigidity. For structural strength verification based on finite element analysis modeling, reasonable solutions for load conditions, boundary conditions, modeling methods, and evaluation criteria were presented. This result is expected to be helpful in examining the structural strength of the stern part of similar carriers in the future.

• Journal of the Society of Disaster Information
• /
• v.18 no.4
• /
• pp.899-907
• /
• 2022

Purpose: Although many studies on seismic fragility analysis of general bridges have been conducted using machine learning methods, studies on curved bridge structures are insignificant. Therefore, the purpose of this study is to analyze the seismic fragility of bridges with I-shaped curved girders based on the machine learning method considering the material property and geometric uncertainties. Method: Material properties and pier height were considered as uncertainty parameters. Parameters were sampled using the Latin hypercube technique and time history analysis was performed considering the seismic uncertainty. Machine learning data was created by applying artificial neural network and response surface analysis method to the original data. Finally, earthquake fragility analysis was performed using original data and learning data. Result: Parameters were sampled using the Latin hypercube technique, and a total of 160 time history analyzes were performed considering the uncertainty of the earthquake. The analysis result and the predicted value obtained through machine learning were compared, and the coefficient of determination was compared to compare the similarity between the two values. The coefficient of determination of the response surface method was 0.737, which was relatively similar to the observed value. The seismic fragility curve also showed that the predicted value through the response surface method was similar to the observed value. Conclusion: In this study, when the observed value through the finite element analysis and the predicted value through the machine learning method were compared, it was found that the response surface method predicted a result similar to the observed value. However, both machine learning methods were found to underestimate the observed values.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.4
• /
• pp.45-53
• /
• 2023

In this study, to analyze the correlation of surface chloride and corrosion amount level according to the installation location of steel members exposed to the marine environment, the surface chloride and mean corrosion depth were evaluated by member units for box girder members of the offshore steel bridge and box specimens. The surface chloride was measured monthly using the Bresle method for one year. The corrosion amount was evaluated by converting the weight loss due to corrosion products generated in the monitoring steel plate into mean corrosion depth. As a measurement result of the surface chloride and corrosion amount, relative differences in surface chloride and mean corrosion depth were appeared depending on the shape or installation location of the steel members. Moreover, even if members of the same shape were installed in the same bridge, it was confirmed that the corrosion amount was increased locally and rapidly. The tendency of corrosion amount depending on the surface chloride was evaluated to analyze the correlation between surface chloride and corrosion amount, and the relation equations that can asseses the corrosion amount depending on the surface chloride were analyzed. From the results of the correlation between surface chloride and corrosion amount, it was found that the corrosion amount of the steel member affected by the surface chloride was varied up to about 1.15 times depending on the structural detail.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.3
• /
• pp.173-184
• /
• 2023

Presently, the general seismic fragility evaluation method for a bridge system composed of member elements with different nonlinear behaviors against strong earthquakes has been to evaluate at the element-level. This study aims to develop a system-level seismic fragility evaluation method that represents a structural system. Because the seismic behavior of bridges is generally divided into transverse and longitudinal directions, this study evaluated the system-level seismic fragility in both directions separately. The element-level seismic fragility evaluation in the longitudinal direction was performed for piers, bridge bearings, pounding, abutments, and unseating. Because pounding, abutment, and unseating do not affect the transverse directional damages, the element-level seismic fragility evaluation was limited to piers and bridge bearings. Seismic analysis using nonlinear models of various structural members was performed using the OpenSEES program. System-level seismic fragility was evaluated assuming that damage between element-levels was serially connected. Pier damage was identified to have a dominant effect on system-level seismic fragility than other element-level damages. In other words, the most vulnerable element-level seismic fragility has the most dominant effect on the system-level seismic fragility.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.1A
• /
• pp.1-13
• /
• 2010

This paper presents an investigation on the geometric nonlinear behavior of cable-stayed bridges using geometric nonlinear finite element analysis method. The girder and mast in cable-stayed bridges show the combined axial load and bending moment interaction due to horizontal and vertical forces of inclined cable. So these members are considered as beam-column member. In this study, the nonlinear finite element analysis method is used to resolve the geometric nonlinear behavior of cable-stayed bridges in consideration of beam-column effect, large displacement effect (known as P-${\delta}$ effect) and cable sag effect. To analyze a cable-stayed bridge model, nonlinear 6-degree of freedom frame element and nonlinear 3-degree of freedom equivalent truss element is used. To resolve the geometric nonlinear behavior for various live load cases, the initial shape analysis is performed for considering dead load before live load analysis. Then the geometric nonlinear analysis for each live load case is performed. The deformed shapes of each model, load-displacement curves of each point and load-tensile force curves for each cable are presented for quantitative study of geometric nonlinear behavior of cable-stayed bridges.

• Korean Journal of Construction Engineering and Management
• /
• v.24 no.1
• /
• pp.31-39
• /
• 2023

To reduce facility management costs and safety concerns due to aging of facilities, it is important to estimate the future facilities' condition based on facility management data and utilize predictive information for management decision making. To this end, this study proposed a methodology to estimate facility elements' condition using XGBoost. To validate the proposed methodology, this study constructed sample data for road bridges and developed a model to estimate condition grades of major elements expected in the next inspection. As a result, the developed model showed satisfactory performance in estimating the condition grades of deck, girder, and abutment/pier (average F1 score 0.869). In addition, a testbed was established that provides data management function and element condition estimation function to demonstrate the practical applicability of the proposed methodology. It was confirmed that the facility management data and predictive information in this study could help managers in making facility management decisions.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.5A
• /
• pp.699-707
• /
• 2008

For most structural evaluation of bridge integrity, it is very important to measure the geometric profile, which is a major factor representing the global behavior of civil structures, especially bridges. In the past, because of the lack of appropriate methods to measure the deflection profile of bridges on site, the measurement of deflection has been restricted to just a few discrete points along the bridge, and the measuring points have been limited to the locations installed with displacement transducers. Thus, some methods for predicting the static deflection by using fiber optic strain sensors has been applied to simply supported bridges. In this study, a method of estimating the static deflection profile by using strains measured from suspension bridges was proposed. Based on the classical deflection theory of suspension bridges, an equation of deflection profile was derived and applied to obtain the actual deflection profile on Namhae suspension bridge. Field load tests were carried out to measure strains from FBG strain sensors attached inside the stiffening girder of the bridge. The predicted deflection profiles were compared with both precise surveying data and numerical analysis results. Thus, it is found that the equation of predicting the deflection profiles proposed in this study could be applicable to suspension bridges and the FBG strain sensors could be reliable on acquiring the strain data from bridges on site.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1A
• /
• pp.75-89
• /
• 2006

This paper presents a practical and realistic Life-Cycle Cost (LCC) optimum design methodology of steel bridges considering time effect of bridge reliability under environmental stressors such as corrosion and heavy truck traffics. The LCC functions considered in the LCC optimization consist of initial cost, expected life-cycle maintenance cost and expected life-cycle rehabilitation costs including repair/replacement costs, loss of contents or fatality and injury losses, road user costs, and indirect socio-economic losses. For the assessment of the life-cycle rehabilitation costs, the annual probability of failure which depends upon the prior and updated load and resistance histories should be accounted for. For the purpose, Nowak live load model and a modified corrosion propagation model considering corrosion initiation, corrosion rate, and repainting effect are adopted in this study. The proposed methodology is applied to the LCC optimum design problem of an actual steel box girder bridge with 3 continuous spans (40 m+50 m+40 m=130 m), and various sensitivity analyses of types of steel, local corrosion environments, average daily traffic volume, and discount rates are performed to investigate the effects of various design parameters and conditions on the LCC-effectiveness. From the numerical investigation, it has been observed that local corrosion environments and the number of truck traffics significantly influence the LCC-effective optimum design of steel bridges, and thus realized that these conditions should be considered as crucial parameters for the optimum LCC-effective design.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4A
• /
• pp.647-657
• /
• 2006

Probabilistic Risk Assessment considering statistically random variables is performed for the preliminary design of a Cable Stayed Bridge, which is Prestressed Concrete Bridge consisted of cable and plate girders, based on the method of Working Stress Design and Strength Design. Component reliabilities of cables and girders have been evaluated using the response surface of the design variables at the selected critical sections based on the maximum shear, positive and negative moment locations. Response Surface Method (RSM) is successfully applied for reliability analyses for this relatively small probability of failure of the complex structure, which is hard to obtain through Monte-Carlo Simulations. or through First Order Second Moment Method that can not easily calculate the derivative terms of implicit limit state functions. For the analysis of system reliability, parallel resistance system consisting of cables and plate girder is changed into series connection system and the result of system reliability of total structure is presented. As a system reliability, the upper and lower probabilities of failure for the structural system have been evaluated and compared with the suggested prediction method for the combination of failure modes. The suggested prediction method for the combination of failure modes reveals the unexpected combinations of element failures in significantly reduced time and efforts compared with the previous permutation method or system reliability analysis method, which calculates upper and lower bound failure probabilities.