• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.207-220
• /
• 2022

Structural health monitoring (SHM) plays a vital role in the maintenance and operation of constructions. In recent years, autonomous inspection has received considerable attention because conventional monitoring methods are inefficient and expensive to some extent. To develop autonomous inspection, a potential approach of crack identification is needed to locate defects. Therefore, this study exploits two deep learning-based segmentation models, DeepLabv3+ and Mask R-CNN, for crack segmentation because these two segmentation models can outperform other similar models on public datasets. Additionally, impacts of label quality on model performance are explored to obtain an empirical guideline on the preparation of image datasets. The influence of image cropping and label refining are also investigated, and different strategies are applied to the dataset, resulting in six alternated datasets. By conducting experiments with these datasets, the highest mean Intersection-over-Union (mIoU), 75%, is achieved by Mask R-CNN. The rise in the percentage of annotations by image cropping improves model performance while the label refining has opposite effects on the two models. As the label refining results in fewer error annotations of cracks, this modification enhances the performance of DeepLabv3+. Instead, the performance of Mask R-CNN decreases because fragmented annotations may mistake an instance as multiple instances. To sum up, both DeepLabv3+ and Mask R-CNN are capable of crack identification, and an empirical guideline on the data preparation is presented to strengthen identification successfulness via image cropping and label refining.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.6 no.1
• /
• pp.27-33
• /
• 1997

A neural network approach has been developed to determine the depth of a surface breaking crack in a steel plate from ultrasonic backscattering data. The network is trained by the use of feedforward three-layered network together with a back-scattering algorithm for error correction. The signal used for crack insonification is a mode converted 70$^{\circ}$transverse wave. A numerical analysis of back scattered field is carried out based on elastic wave theory, by the use of the boundary element method. The numerical data are calibrated by comparison with experimental data. The numerical analysis provides synthetic data for the training of the network. The training data have been calculated for cracks with specified increments of the crack depth. The performance of the network has been tested on other synthetic data and experimental data which are different from the training data.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.7 no.6
• /
• pp.258-269
• /
• 1999

It is important to evaluate fatigue damage of in-service material in respect to assure safety and remaining fatigue life in structure and mechanical components under cyclic load . Fatigue damage is represented by mathematical modelling with crack growth rate da/dN and cycle ration N/Nf and is detected by X-ray diffraction and ultrasonic wave method etc. But this is estimated generally by single parameter but influenced by many test conditions The characteristics of it indicates fatigue damage has complex fracture mechanism. Therefore, in this study we propose that back-propagation neural networks on the basis of ration of X-ray half-value breath B/Bo, fractal dimension Df and fracture mechanical parameters can construct artificial intelligent networks estimating crack growth rate da/dN and cycle ratio N/Nf without regard to stress amplitude Δ $\sigma$.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.3
• /
• pp.247-253
• /
• 2007

Determination of crack depth in filed using the self-calibrating surface wane transmission measurement and the cutting frequency in the transmission function (TRF) is very difficult due to variations of the measurement conditions. In this study, it is proposed to use the measured full TRF as a feature for crack depth assessment. A Principal component analysis (PCA) is employed to generate a basis of the measured TRFs for various crack cases. The measured TRFs are represented by their projections onto the most significant principal components. Then artificial neural networks (NNs) using the PCA-compressed TRFs is applied to assess the crack in concrete. Experimental study is carried out for five different crack cases to investigate the effectiveness of the proposed method. Results reveal that the proposed method can be effectively used for the crack depth assessment of concrete structures.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.9
• /
• pp.2374-2385
• /
• 1994

SiCw/Al composite material is especially attractive because of their superior specific strength, specific stiffness, corrosion fatigue resistance, creep resistance, and wear resistance compared with the corresponding wrought Al alloy. In this study, Fatigue crack growth behavior and fatigue crack path morphology(FCPM) of SiC whisker reinforced Al 6061 alloy with 25% SiC volume fraction and Al 6061 allay were performed. Result of the fatigue crack growth test sgiwed that fatigue crack growth rate of SiCw/Al 6061 composite was slower than that of Al 6061 matrix therefore it was confirmed that Sic whisker have a excellent fatigue resistance. And Al 6061 matrix had only FCPM perpendicular to loading direction. On the other hand SiCw/Al 6061 composite had three types in fatigue crack path morphology. First type is that both sides FCPM of artificial notch are perpendicular to loading direction. Second type is that a FCPM in artifical notch has slant angle to loading direction and the other side FCPM is perpendicular to loading direction. Third type is that both sides FCPM of notch have slant angle to loading direction. It was considered that this kinds of phenomena were due to non-uniform distribution of SiC whisker and confirmed by SEM observation for fracture mechanism study.

• Corrosion Science and Technology
• /
• v.13 no.3
• /
• pp.112-119
• /
• 2014

In this work, it is aimed to develop the fabrication method of axial stress corrosion cracking (SCC) defects having various sizes, on the outer diameter surface of the steam generator (SG) tubings. To control the length of the artificial SCC defect, the specific area of the SG tubing samples was exposed to an acidic solution after a sensitization heat treatment. During the exposure to an acidic solution, a direct current potential drop (DCPD) method was adopted to monitor the crack depth. The size of the SCC defect was first evaluated by an eddy current test (ECT), and then confirmed by a destructive examination. From the comparison, it was found that the actual crack length was well controlled to be similar to the length of the surface exposed to an acidic solution (5, 10, 20 or 30 mm in this work) with small standard deviation. From in-situ monitoring of the crack depth using the DCPD method, it was possible to distinguish a non-through wall crack from a through wall crack, even though the depth of the non-through wall crack was not able to be precisely controlled. The fabrication method established in this work was useful to simulate the SCC defect having similar size and ECT signals as compared to the field cracks in the SG tubings of the operating Korean PWRs.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.22 no.3
• /
• pp.69-74
• /
• 2018

Fiber-Reinforced Cementitious Composites (FRCCs) have electrical conductivity by inserting reinforced conductive fibers into a cementitious matrix. Such characteristic allows us to utilize FRCCs for crack monitoring of a structure by measuring electrical responses without sensor installation. However, the electrical responses are often sensitively altered by temperature variation as well as crack initiation. The temperature variation may disturb crack detection on the measured electrical responses. Moreover, as sensing probes for measuring electrical reponses increase, undesired contact noises are often augmented. In this paper, a self-sensing impedance circuit is specially designed for reducing the number of sensing probes. The crack initiation and temperature variation effects on the self-sensing impedance responses of FRCCs are experimentally investigated using the self-sensing impedance circuit. The experiment results reveal that the electrical impedance response are more sensitively changed due to temperature variation than crack initiation.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.20 no.1
• /
• pp.112-117
• /
• 2016

For artificial ground greening construction without root-proofing layer, this research reviewed the effect on watertightness of asphalt waterproofing layer by plant roots penetration based on crack width, and crack penetrated roots. Experiment on concrete crack width was performed with three conditions such as 0.15, 0.30, and 0.45 mm, and all three conditions confirmed that all plant roots penetrations were made through crack area in 12 ~ 18 months. In addition, according to evaluation of effects on waterproofing layer by crack penetrated plant roots and in condition of 0.45 mm crack width, it indicated that penetration is made on asphalt waterproofing layer in 12 months due to roots penetration.

• Bulletin of the Society of Naval Architects of Korea
• /
• v.21 no.1
• /
• pp.35-42
• /
• 1984

In the present study, rotating bending fatigue tests have been carried out in three kinds of carbon steel specimens; an annealed low carbon steel, an annealed high carbon steel and quenched-tempered high carbon steel; with a small artificial surface defect that might exist in real structures. Fatigue crack lengths have been observed by a method of replication in order to investigate the growth characteristic of fatigue crack in the viewpoints of strength of materials and fracture mechanics. The main results obtained are as follows: 1) The effect of a small surface defect upon the reduction of fatigue limit is considerably large, and the rate of fatigue limit reduction grows in the following order; annealed low carbon steel(mild steel), annealed high carbon steel, quenched-tempered high carbon steel. 2) When the growth rate of surface crack length(2a) was investigated in the viewpoints of fracture mechanics based upon $ ${\Delta}K_{\varepsilon}$, the dependence of stress level and of surface defect size disappear, and there exists a linear relationships between d(2a)/dN and ${\Delta}K_{{\varepsilon}t},\;\Delta_{{\varepsilon}t}\sqrt{{\pi}a}$, on log. plot, i.e, $d(2a)/dN={C{\cdot}{\Delta}K_{\varepsilon}}^3_t$, where ${\Delta}_{{\varepsilon}t}\sqrt{{\pi}a}$ a is the cyclic total strain intensity factor range.

• Journal of the Korean Ceramic Society
• /
• v.43 no.7 s.290
• /
• pp.439-444
• /
• 2006

Artificial graphite generally manufactured by carbonization sintering of shape-body of kneaded mixture using granular cokes as filler and pitch as binder, going through pitch impregnation process if necessary and finally applying graphitization heat treatment. Graphite materials are used for core internal structural components of the High-Temperature Gas-cooled Reactors (HTGR) because of their excellent heat resistibility and resistance of crack progress. The HTGR has a core consisting of an array of stacked graphite fuel blocks are machined from IG-110, a high-strength, fine-grained isotropic graphite. In this study, crack stabilization and micro-structures were measured by bend strength and fracture toughness of isotropic graphite grade IG-110. It is important to the reactor designer as they may govern the life of the graphite components and hence the life of the reactor. It was resulted crack propagation, bend strength, compressive strength and micro-structures of IG-110 graphite by scanning electron microscope and universal test machine.