• Computers and Concrete
• /
• 제4권2호
• /
• pp.83-100
• /
• 2007

Applying a direct formulation for the enrichment of the displacement field an extended finite element (XFEM) scheme for modeling of cohesive crack growth is developed. Only elements cut by the crack is enriched and the scheme fits within the framework of standard FEM code. The scheme is implemented for the 3-node constant strain triangle (CST) and the 6-node linear strain triangle (LST). Modeling of standard concrete test cases such as fracture in the notched three point beam bending test (TPBT) and in the four point shear beam test (FPSB) illustrates the performance. The XFEM results show good agreement with results obtained by applying standard interface elements in FEM and with experimental results. In conjunction with criteria for crack growth local versus nonlocal computation of the crack growth direction is discussed.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2007년도 하계종합학술대회 논문집
• /
• pp.381-382
• /
• 2007

In this paper, crack detection technique of concrete bridge is proposed robust against shadow and noise. Our technique consists of two steps. In the first step, crack candidate region is detected by preprocessing. Preprocessing techniques such as median filter, isolated point filter and morphological techniques, use utilized for detection of crack candidate regions. In the final step, crack is detected from crack candidate region by considering any connectivity between cracks. By experimental results, performance is improved 6.8% over the existing method.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2022년도 가을 학술논문 발표대회
• /
• pp.245-246
• /
• 2022

Typically, the methods of crack detection on concrete structures include some problems, such as a low accuracy and expensive. To solve these problems, we proposed a neural network-based crack search method. The proposed algorithm goes through three convolutions and is classified into crack and non-crack through the softmax layer. As a result of the performance evaluation, cracks can be detected with an accuracy of 99.4 and 99.34 % at the training model and the validation model, respectively.

• 한국구조물진단유지관리공학회 논문집
• /
• 제26권4호
• /
• pp.40-47
• /
• 2022

이 연구의 목적은 철근콘크리트 벽체의 내력, 변형 및 균열진전에 대한 개구부 모서리의 사인장 균열 제어를 위한 다양한 보강방법들이 효율성을 평가하는 것이다. 개구부 모서리의 사인장 균열제어 보강방법으로서 이 연구에서 제안한 균열제어 띠, 기존의 경사보강근 배근방법, 그리고 응력분산곡면판을 선택하였다. 균열제어 띠에서는 균열치유 기능성이 부여된 것도 함께 고려하였다. 개구부 주위의 체적변화 및 균열 폭 진전을 평가하기 위하여 개구부 모서리에서 사인장 균열 유도를 위한 축소 벽체 실험을 수행하였다. 실험결과 균열제어 띠는 경사보강근 및 응력분산곡면판에 비해 개구부 주위의 체적변화 감소 및 개구부 모서리 균열 폭 제어에 효율적이었다. 균열 치유 기능이 포함된 균열제어 띠는 사전 실험에서 발생한 균열들을 치유하면서 축소 벽체의 내력 향상 및 균열 폭 감소에 가장 좋은 성능을 보였다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.195-206
• /
• 2022

Crack detection is essential for inspection of existing structures and crack segmentation based on deep learning is a significant solution. However, datasets are usually one of the key issues. When building a new dataset for deep learning, laborious and time-consuming annotation of a large number of crack images is an obstacle. The aim of this study is to develop an approach that can automatically select a small portion of the most informative crack images from a large pool in order to annotate them, not to label all crack images. An active learning method with difficulty learning mechanism for crack segmentation tasks is proposed. Experiments are carried out on a crack image dataset of a steel box girder, which contains 500 images of 320×320 size for training, 100 for validation, and 190 for testing. In active learning experiments, the 500 images for training are acted as unlabeled image. The acquisition function in our method is compared with traditional acquisition functions, i.e., Query-By-Committee (QBC), Entropy, and Core-set. Further, comparisons are made on four common segmentation networks: U-Net, DeepLabV3, Feature Pyramid Network (FPN), and PSPNet. The results show that when training occurs with 200 (40%) of the most informative crack images that are selected by our method, the four segmentation networks can achieve 92%-95% of the obtained performance when training takes place with 500 (100%) crack images. The acquisition function in our method shows more accurate measurements of informativeness for unlabeled crack images compared to the four traditional acquisition functions at most active learning stages. Our method can select the most informative images for annotation from many unlabeled crack images automatically and accurately. Additionally, the dataset built after selecting 40% of all crack images can support crack segmentation networks that perform more than 92% when all the images are used.

• 한국안전학회지
• /
• 제27권6호
• /
• pp.109-114
• /
• 2012

This study is the things to experiment and evaluate the performance of self-healing water-proofing on the concrete to be using the crystal growth composite waterproofing admixture. The cylinder to be making on the concrete by ${\varnothing}150{\times}300$ mm for evaluating the performance of self-healing water-proofing was aging 90 days and cut on a 50 mm. So, it prompted the crack and applied. After it measured the quantity of water to be flow the water throughout the crack part of the cylinder, it applied the basic formular of Darcy's law and calculated the coefficient of water permeability. So, it verified the performance of self-healing water-proofing on the basis of the changing shape of the water permeability. This experiment is the thing to be applied the general evaluation of the structure to demand the real watertightness on doing for the evaluating of performance of the quantity of water leak and self-healing water-proofing about the various penetration crack.

• 대한기계학회논문집A
• /
• 제26권3호
• /
• pp.528-536
• /
• 2002

As one step for the safety performance of LNG storage tank, the change in fracture toughness within the X-grooved weld heat-affected zone (HAZ) of newly developed 9% Ni steel, which was submerged arc (SA)-welded, was investigated. Both crack initiation fracture toughness and crack arrest fracture toughness were evaluated by the crack tip opening displacement (CTOD) tests and compact crack arrest (CCA) tests. As the evaluated region approached the fusion line, each test result shorted different tendency, that is, crack initiation toughness decreased while crack arrest toughness increased. The results were discussed through the observation of the microstructural change.

• Computers and Concrete
• /
• 제8권6호
• /
• pp.735-755
• /
• 2011

The capability of a multi-directional fixed smeared crack constitutive model to simulate the flexural/punching failure modes of fiber reinforced concrete (FRC) laminar structures is discussed. The constitutive model is implemented in a computer program based on the finite element method, where the FRC laminar structures were simulated according to the Reissner-Mindlin shell theory. The shell is discretized into layers for the simulation of the membrane, bending and out-of-plane shear nonlinear behavior. A stress-strain softening diagram is proposed to reproduce, after crack initiation, the evolution of the normal crack component. The in-plane shear crack component is obtained using the concept of shear retention factor, defined by a crack-strain dependent law. To capture the punching failure mode, a softening diagram is proposed to simulate the decrease of the out-of-plane shear stress components with the increase of the corresponding shear strain components, after crack initiation. With this relatively simple approach, accurate predictions of the behavior of FRC structures failing in bending and in shear can be obtained. To assess the predictive performance of the model, a punching experimental test of a module of a façade panel fabricated with steel fiber reinforced self-compacting concrete is numerically simulated. The influence of some parameters defining the softening diagrams is discussed.

• Smart Structures and Systems
• /
• 제15권6호
• /
• pp.1543-1567
• /
• 2015

In this study, a wireless crack sensor is developed for monitoring cracks propagating in two dimensions. This sensor is developed by incorporating a laser-based optical navigation sensor board (ADNS-9500) into a smart wireless platform (Imote2). To measure crack propagation, the Imote2 sends a signal to the ADNS-9500 to collect a sequence of images reflected from the concrete surface. These acquired images can be processed in the ADNS-9500 directly (the navigation mode) or sent to Imote2 for processing (the frame capture mode). The computed crack displacement can then be transmitted wirelessly to a base station. The design and the construction of this sensor are reported herein followed by some calibration tests on one prototype sensor. Test results show that the sensor can provide sub-millimeter accuracy under sinusoidal and step movement. Also, the two modes of operation offer complementary performance as the navigation mode is more accurate in tracking large amplitude and fast crack movement while the frame capture mode is more accurate for small and slow crack movement. These results illustrate the feasibility of developing such a crack sensor as well as point out directions of further research before its actual implementation.

• 한국기계가공학회지
• /
• 제3권4호
• /
• pp.88-94
• /
• 2004

High manganese steel was maintained stability of Non-Magnetics performance. Fatigue tests were carried out under constant stress amplitude, using a non-magnetic high manganese steel. The fatigue crack growth mechanism of the high manganese steel was clarified from results such as observation of crack growth path and fracture surface. The result of getting this study was shown as following: 1) Remarkably ${\Delta}Kth$ of the high manganese steel is big with about 3 times of the general steel product. 2) In the low ${\Delta}K$ value region, da/dN is dependent on Kmax, and in the high ${\Delta}K$ value region, it is dependent on ${\Delta}Keff$. The reason of this behavior is crack closure due to fracture surface roughness and fretting oxide. 3) It seems to ease the stress concentration of crack tip crack growth behavior in the ${\Delta}Kth$ vicinity by the generation of the secondary crack.