• Korean Journal of Metals and Materials
• /
• v.49 no.8
• /
• pp.628-634
• /
• 2011

SA508 Gr.4N Ni-Cr-Mo low alloy steel, in which Ni and Cr contents are higher than in commercial SA508 Gr.3 Mn-Mo-Ni low alloy steels, may be a candidate reactor pressure vessel (RPV) material with higher strength and toughness from its tempered martensitic microstructure. The inner surface of the RPV is weld-cladded with stainless steels to prevent corrosion. The goal of this study is to evaluate the microstructural properties of the clad interface between Ni-Cr-Mo low alloy steel and stainless weldment, and the effects of post weld heat treatment (PWHT) on the properties. The properties of the clad interface were compared with those of commercial Mn-Mo-Ni low alloy steel. Multi-layer welding of model alloys with ER308L and ER309L stainless steel by the SAW method was performed, and then PWHT was conducted at $610^{\circ}C$ for 30 h. The microstructural changes of the clad interface were analyzed using OM, SEM and TEM, and micro-Vickers hardness tests were performed. Before PWHT, the heat affected zone (HAZ) showed higher hardness than base and weld metals due to formation of martensite after welding in both steels. In addition, the hardness of the HAZ in Ni-Cr-Mo low alloy steel was higher than that in Mn-Mo-Ni low alloy steel due to a comparatively high martensite fraction. The hardness of the HAZ decreased after PWHT in both steels, but the dark region was formed near the fusion line in which the hardness was locally high. In the case of Mn-Mo-Ni low alloy steel, formation of fine Cr-carbides in the weld region near the fusion line by diffusion of C from the base metal resulted in locally high hardness in the dark region. However, the precipitates of the region in the Ni-Cr-Mo low alloy steel were similar to that in the base metal, and the hardness in the region was not greatly different from that in the base metal.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.6
• /
• pp.555-561
• /
• 2020

In this study, the applicability of Hunt's analytical solution for a two-layered leaky aquifer system, which was developed to estimate stream depletion due to the groundwater pumping of the upper shallow aquifer, was evaluated. The 5-year averaged stream depletions were estimated using Hunt's analytical solution for various combinations of hydraulic characteristic values such as transmissivity, storage coefficient of the two aquifers, interlayer leakage coefficient, stream-well distance, hydraulic conductivity of the streambed, and stream width. Through comparison with the numerical solution accurately simulated with a MODFLOW groundwater flow model, the analytical solution derived by regarding the stream width as a point was evaluated. It was found that the error in the stream depletion calculated by the analytical solution can be reduced to less than 0.05 when the stream-well distance is greater than the stream width or when the stream depletion factor (SDF) is more than about 3,000 days. In addition, when the streambed hydraulic conductivity is less than 1 m/d, the hydraulic diffusion coefficient of the lower aquifer layer is less than 100 ㎡/d, the hydraulic diffusion coefficient ratio of the upper and lower aquifer layers is 5 or more, and the leakage coefficient between the layers is less than 0.0004 m/d, the overall analytical solutions were overestimated compared with the numerical solutions.

• Clean Technology
• /
• v.27 no.1
• /
• pp.61-68
• /
• 2021

Selective catalytic reduction (SCR) is widely used as a method of removing nitrogen oxide in large-capacity thermal power generation systems. Uniform mixing of the injected ammonia and the inlet flue gas is very important to the performance of the denitrification reduction process in the catalyst bed. In the present study, a computational analysis technique was applied to the ammonia injection system design process of a denitrification facility. The applied model is the denitrification facility of an 800 MW class coal-fired power plant currently in operation. The flow field to be solved ranges from the inlet of the ammonia injection system to the end of the catalyst bed. The flow was analyzed in the two-dimensional domain assuming incompressible. The steady-state turbulent flow was solved with the commercial software named ANSYS-Fluent. The nozzle arrangement gap and injection flow rate in the ammonia injection system were chosen as the design parameters. A total of four (4) cases were simulated and compared. The root mean square of the NH3/NO molar ratio at the inlet of the catalyst layer was chosen as the optimization parameter and the design of the experiment was used as the base of the optimization algorithm. The case where the nozzle pitch and flow rate were adjusted at the same time was the best in terms of flow uniformity.

• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.616-625
• /
• 2021

Deep neural networks are an approximation method that approximates an arbitrary function to a linear model and then repeats additional approximation using a nonlinear active function. In this process, the method of evaluating the performance of approximation uses the loss function. Existing in-depth learning methods implement approximation that takes into account loss functions in the linear approximation process, but non-linear approximation phases that use active functions use non-linear transformation that is not related to reduction of loss functions of loss. This study proposes parametric activation functions that introduce scale parameters that can change the scale of activation functions and location parameters that can change the location of activation functions. By introducing parametric activation functions based on scale and location parameters, the performance of nonlinear approximation using activation functions can be improved. The scale and location parameters in each hidden layer can improve the performance of the deep neural network by determining parameters that minimize the loss function value through the learning process using the primary differential coefficient of the loss function for the parameters in the backpropagation. Through MNIST classification problems and XOR problems, parametric activation functions have been found to have superior performance over existing activation functions.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.6
• /
• pp.363-369
• /
• 2021

High-definition road maps are used as the basic infrastructure for autonomous vehicles, so the latest road information must be quickly reflected. However, the current drawing and structural editing process of high-definition road maps are manually performed. In addition, it takes the longest time to generate road lanes, which are the main construction targets. In this study, the point cloud of the road lane markings, in which color types(white, blue, and yellow) were predicted through the PointNet model pre-trained in previous studies, were used as input data. Based on the point cloud, this study proposed a methodology for automatically drawing and structural editing of the layer of road lane markings. To verify the usability of the 3D vector data constructed through the proposed methodology, the accuracy was analyzed according to the quality inspection criteria of high-definition road maps. In the positional accuracy test of the vector data, the RMSE (Root Mean Square Error) for horizontal and vertical errors were within 0.1m to verify suitability. In the structural editing accuracy test of the vector data, the structural editing accuracy of the road lane markings type and kind were 88.235%, respectively, and the usability was verified. Therefore, it was found that the methodology proposed in this study can efficiently construct vector data of road lanes for high-definition road maps.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.373-380
• /
• 2022

This paper presents a deep learning based group synchronization that supports networked immersive interactions between remote users. The goal of group synchronization is to enable all participants to synchronously interact with others for increasing user presence Most previous methods focus on NTP-based clock synchronization to enhance time accuracy. Moving average filters are used to control media playout time on the synchronization server. As an example, the exponentially weighted moving average(EWMA) would be able to track and estimate accurate playout time if the changes in input data are not significant. However it needs more time to be stable for any given change over time due to codec and system loads or fluctuations in network status. To tackle this problem, this work proposes the Deep Group Synchronization(DeepGroupSync), a group synchronization based on deep learning that models important features from the data. This model consists of two Gated Recurrent Unit(GRU) layers and one fully-connected layer, which predicts an optimal playout time by utilizing the sequential playout delays. The experiments are conducted with an existing method that uses the EWMA and the proposed method that uses the DeepGroupSync. The results show that the proposed method are more robust against unpredictable or rapid network condition changes than the existing method.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.429-440
• /
• 2021

With the recent development of deep convolutional neural network learning, deep learning techniques applied to single image super-resolution are showing good results. One of the existing deep learning-based super-resolution techniques is RDN(Residual Dense Network), in which the initial feature information is transmitted to the last layer using residual dense blocks, and subsequent layers are restored using input information of previous layers. However, if all hierarchical features are connected and learned and a large number of residual dense blocks are stacked, despite good performance, a large number of parameters and huge computational load are needed, so it takes a lot of time to learn a network and a slow processing speed, and it is not applicable to a mobile system. In this paper, we use the residual dense structure, which is a continuous memory structure that reuses previous information, and the residual dense channel attention block using the channel attention method that determines the importance according to the feature map of the image. We propose a method that can increase the depth to obtain a large receptive field and maintain a concise model at the same time. As a result of the experiment, the proposed network obtained PSNR as low as 0.205dB on average at 4× magnification compared to RDN, but about 1.8 times faster processing speed, about 10 times less number of parameters and about 1.74 times less computation.

• Clean Technology
• /
• v.28 no.4
• /
• pp.316-322
• /
• 2022

The flow pattern at the inlet of the catalyst layer in a selective catalytic reduction (SCR) system is one of the key parameters influencing the performance of the denitrification process. In the curved diffusing parts between the ammonia injection grids and the catalyst layers, guide vanes are installed to improve flow uniformity. In the present study, a numerical simulation has been performed to investigate the effect of the geometrical configuration of the guide vanes on the aerodynamic characteristics of a denitrification facility. This application has been made to the existing SCR process in a large-scaled coal-fired power plant. The flow domain to be solved covers the whole region of the flow passages from the exit of the ammonia injection gun to the exit of the catalyst layers. ANSYS-Fluent was used to calculate the three-dimensional steady viscous flow fields with the proper turbulence model fitted to the flow characteristics. The root mean square of velocity and the pressure drop inside the flow passages were chosen as the key performance parameters. Four types of guides vanes were proposed to improve the flow quality compared to the current configuration. The numerical results showed that the type 4 configuration was the most effective at improving the aerodynamic performance in terms of flow uniformity and pressure loss.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.1
• /
• pp.51-58
• /
• 2023

In this paper, we propose an organ segmentation technique for the automatic extraction of medical diagnostic indicators from X-ray images. In order to calculate diagnostic indicators of heart disease and spinal disease such as VHS(vertebral heart scale) and Cobb angle, it is necessary to accurately segment the thoracic spine, carina, and heart in a chest X-ray image. A deep neural network model in which the high-resolution representation of the image for each layer and the structure converted into a low-resolution feature map are connected in parallel was adopted. This structure enables the relative position information in the image to be effectively reflected in the segmentation process. It is shown that learning performance can be improved by combining the OCR module, in which pixel information and object information are mutually interacted in a multi-step process, and the channel attention module, which allows each channel of the network to be reflected as different weight values. In addition, a method of augmenting learning data is presented in order to provide robust performance against changes in the position, shape, and size of the subject in the X-ray image. The effectiveness of the proposed theory was evaluated through an experiment using 145 human chest X-ray images and 118 animal X-ray images.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1246-1246
• /
• 2001

The quality of meat is highly variable in many properties. This variability originates from both animal production and meat processing. At the pre-slaughter stage, animal factors such as breed, sex, age contribute to this variability. Environmental factors include feeding, rearing, transport and conditions just before slaughter (Hildrum et al., 1995). Meat can be presented in a variety of forms, each offering different opportunities for adulteration and contamination. This has imposed great pressure on the food manufacturing industry to guarantee the safety of meat. Tissue and muscle speciation of flesh foods, as well as speciation of animal derived by-products fed to all classes of domestic animals, are now perhaps the most important uncertainty which the food industry must resolve to allay consumer concern. Recently, there is a demand for rapid and low cost methods of direct quality measurements in both food and food ingredients (including high performance liquid chromatography (HPLC), thin layer chromatography (TLC), enzymatic and inmunological tests (e.g. ELISA test) and physical tests) to establish their authenticity and hence guarantee the quality of products manufactured for consumers (Holland et al., 1998). The use of Near Infrared Reflectance Spectroscopy (NIRS) for the rapid, precise and non-destructive analysis of a wide range of organic materials has been comprehensively documented (Osborne et at., 1993). Most of the established methods have involved the development of NIRS calibrations for the quantitative prediction of composition in meat (Ben-Gera and Norris, 1968; Lanza, 1983; Clark and Short, 1994). This was a rational strategy to pursue during the initial stages of its application, given the type of equipment available, the state of development of the emerging discipline of chemometrics and the overwhelming commercial interest in solving such problems (Downey, 1994). One of the advantages of NIRS technology is not only to assess chemical structures through the analysis of the molecular bonds in the near infrared spectrum, but also to build an optical model characteristic of the sample which behaves like the “finger print” of the sample. This opens the possibility of using spectra to determine complex attributes of organic structures, which are related to molecular chromophores, organoleptic scores and sensory characteristics (Hildrum et al., 1994, 1995; Park et al., 1998). In addition, the application of statistical packages like principal component or discriminant analysis provides the possibility to understand the optical properties of the sample and make a classification without the chemical information. The objectives of this present work were: (1) to examine two methods of sample presentation to the instrument (intact and minced) and (2) to explore the use of principal component analysis (PCA) and Soft Independent Modelling of class Analogy (SIMCA) to classify muscles by quality attributes. Seventy-eight (n: 78) beef muscles (m. longissimus dorsi) from Hereford breed of cattle were used. The samples were scanned in a NIRS monochromator instrument (NIR Systems 6500, Silver Spring, MD, USA) in reflectance mode (log 1/R). Both intact and minced presentation to the instrument were explored. Qualitative analysis of optical information through PCA and SIMCA analysis showed differences in muscles resulting from two different feeding systems.