• Journal of Ocean Engineering and Technology
• /
• v.25 no.6
• /
• pp.66-71
• /
• 2011

Recently, the production of shipbuilding and offshore plant industries, with a trend toward large structures, has led to an increased use of high strength ultra-thick plates. The use of ultra-thick plates increases the welding tasks, and the welding process generates distortion and residual stress in the weldment because of the rapid heating and cooling. Welding distortion and residual stress in the welded structure resulte in many troubles such as deformation and life deterioration. In particular, the welding residual stress has an important effect on welding deformation, fatigue, buckling strength, brittleness, etc. The purpose of this study was to evaluate the residual stress at a multi-pass weldment using an experimental method for EH36 high-tension steel. In this experimental method, AIS3000 was used to measure the residual stress of a welded part, HAZ, and base metal; EPMA and XRD were used to study the material properties.

• Journal of the Korean Vacuum Society
• /
• v.13 no.4
• /
• pp.157-163
• /
• 2004

Dependence of residual compressive stress of diamond-like carbon (DLC) films on relative humidity was investigated. Polymeric, graphitic and diamond-like carbon films were prepared by r.f.-PACVD using methane or benzene with the negative self bias voltage of the substrate ranging from -100 to -800 V. In-situ measurements of the residual stress were carried out in an environment chamber where the relative humidity was varied from 10％ to 90％. In dense DLC film of high residual compressive stress and hardness, we could not observe any change in the residual compressive stress with relative humidity. However, in the cases of graphitic and polymeric DLC films, abrupt change in the residual stress occurred by changing the relative humidity. The quantity of the stress change was inversely proportional to the film thickness, which means that the stress change with humidity is not due to the penetration of the water molecule into the film structure, but due to surface interaction between water molecules and film surface.

• Journal of the Korean Society for Railway
• /
• v.12 no.2
• /
• pp.183-189
• /
• 2009

The track structure of Gyungbu High Speed Railway line from Daegu to Busan is concrete track. It has a very strict specification for residual settlement because of its rigid type structural characteristics. The residual settlement of it comes from the residual settlement of the subgrade and the ground. The residual settlement of railway subgrade composed of crushed rock and soil might be major parts of total residual settlement depending on the field compaction qualities. Therefore, it is a key to minimize the residual settlement of the subgrade for a successful concrete track construction. In this paper, total 31 large scale compaction tests were performed to understand the compaction behaviors of the crushed rock-soil mixture. The test specimens were constituted with soil, crushed shale and mudstone taken from two sites under construction. The compaction tests were performed with the variations of rock types, #4 sieve passing contents, maximum particle size, and moisture contents. The influence of those factors on maximum dry unit weights of crushed rock-soil mixture was evaluated.

• Journal of Digital Convergence
• /
• v.19 no.6
• /
• pp.251-258
• /
• 2021

In this paper, we proposed an encoder-decoder model utilizing residual learning to improve the accuracy of the U-Net-based semantic segmentation method. U-Net is a deep learning-based semantic segmentation method and is mainly used in applications such as autonomous vehicles and medical image analysis. The conventional U-Net occurs loss in feature compression process due to the shallow structure of the encoder. The loss of features causes a lack of context information necessary for classifying objects and has a problem of reducing segmentation accuracy. To improve this, The proposed method efficiently extracted context information through an encoder using residual learning, which is effective in preventing feature loss and gradient vanishing problems in the conventional U-Net. Furthermore, we reduced down-sampling operations in the encoder to reduce the loss of spatial information included in the feature maps. The proposed method showed an improved segmentation result of about 12% compared to the conventional U-Net in the Cityscapes dataset experiment.

• Proceedings of the KWS Conference
• /
• 2006.05a
• /
• pp.100-102
• /
• 2006

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2003.10a
• /
• pp.93-96
• /
• 2003

Laser beam welding process is a relatively new process in comparison with arc welding process, but it is expected to apply widely because of the many advantages, and research and development of that process is being progressed actively for the practical use. the application of this welding process has been restricted due to the high initial investment and the need of precise processing against the material, but cost reduction and thick plate welding in high speed have become practial by recent technological development, and this welding process to not only small parts in automobile, machinery and physicochemical field, but also a large structure and pipe line are being applied. In order to utilize this welding process appropriately to a steel structure, the properties of welding residual stresses and fracture toughness in welded joints are to be investigated for relibilty. On this study, after performing the finite element analysis, thermal and residual stress properties have been examined to the general structural steel (HT50) by laser beam welding. Besides, the property of fracture toughness has been investigated by the Charpy impact test and 3-points bending CTOD test carried out in the range of temperature between $-60^{\circ}C$ and $20^{\circ}C$. From the research results it is revealed that the maximum residual stress appears in the center of plate thickness and the fracture toughness is influenced by strength mis-match.

• Journal of the Korean Society of Groundwater Environment
• /
• v.6 no.4
• /
• pp.213-217
• /
• 1999

The river bank is one of the most important structure of fluvial hydraulic structure. Because the breaking of river bank is the cause of calamity, the durability and stability of river bank an very important factors. The breaking of river bank is the cause of the overflow of flood and the leakage of river bank. In this study, we investigated the leakage of river bank using the resistivity probing and estimated the volume of leakage using the weighted residual method The study basin of this study is the upstream of Sumji river basin and the factor of river bank is length 300 m and berm 2.0 m and width 4.5 m and height 4 m. We evaluated the leakage of river basin using using the resistivity probing and estimated the leakage volume using the weighted residual method. The result of this study, the leakage of river bank generated at the point of 39~45 m 80~90 m. 218~222 m. 214~250 m and the type of leakage is the rectangle and the polygon. And the leakage volume of this points evaluated 2.7$\times$$10^{-3}$ $\textrm{m}^3$/sec.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.413-435
• /
• 2018

Although the accuracy of handwritten character recognition based on deep networks has been shown to be superior to that of the traditional method, the use of an overly deep network significantly increases time consumption during parameter training. For this reason, this paper took the training time and recognition accuracy into consideration and proposed a novel handwritten character recognition algorithm with newly designed network structure, which is based on an extended nonlinear kernel residual network. This network is a non-extremely deep network, and its main design is as follows:(1) Design of an unsupervised apriori algorithm for intra-class clustering, making the subsequent network training more pertinent; (2) presentation of an intermediate convolution model with a pre-processed width level of 2;(3) presentation of a composite residual structure that designs a multi-level quick link; and (4) addition of a Dropout layer after the parameter optimization. The algorithm shows superior results on MNIST and SVHN dataset, which are two character benchmark recognition datasets, and achieves better recognition accuracy and higher recognition efficiency than other deep structures with the same number of layers.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.5
• /
• pp.248-254
• /
• 2005

In this Paper, the structure of a split bandwidth wideband speech coder and its highband coder for tone qualify elevation are Proposed. The lowband and highband by the split bandwidth method are encoded independently applying the G.729E and MLT (Modulated Lapped Transform) residual model. In the highband structure which is encoded by low bit rate of 4kbps, the MLT residual signals are distinguished to voice and unvoice signal . The voice signals are applied to MLT peak picking method by lowband pitch period. Because transformed MLT residual signals are represented by periodic signal that have periodic peak. The unvoice signals are applied to MLT which linear prediction spectral response is added and do vector quantization. Performance for proposed 15.8kbps wideband speech coder was verified through subjective listening test.

• Composites Research
• /
• v.16 no.6
• /
• pp.56-61
• /
• 2003

In this Paper, a fiber Bragg grating(FBG) sensor system is described and FBGs are well-suited for long term and extremely severe experiments, where traditional strain gauges fail. In the system. a reflect wave-length measurement method which employs a tunable light source to find out the center wave-length of FBG sensor is used. We apply the FBG system to nuclear energy Power Plant for structural integrity test to measure the displacement of the structure under designed pressure and to check the elasticity of the structure by measuring the residual strain. The system works very well and it is expected that it can be used for a real-time strain, temperature and vibration detector of smart structure.