• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.176-176
• /
• 2021

한반도의 강수를 예측하는 것은 수자원 관리 측면에서 매우 중요하다. 한반도의 강수는 연별 변동 뿐 아니라 계절별로 변동을 갖는다. 우리는 이 중 건기(Dry period)의 가뭄과 그 이후 농번기(3월, 4월)에 영향을 미치는 초겨울(11월, 12월)의 강수를 예측은 수자원 관리에 있어서 중요한 의미를 갖는다. 본 연구에서는 Regularized regression 모형인 Elastic net model을 이용하여 중장기 (7개월 이상)기반으로 초겨울의 강우 예측의 가능성에 대해서 논하고자 한다. 특히, 본 연구에서는 우리나라의 초겨울 강우의 변동이 대서양의 대규모의 대기 순환과 밀접한 관계를 보이는 것을 확인하였으며 이를 논하기 위해서 Sea Surface Temperature (SST) 등의 자료를 사용하여 분석하였다. 이 시간적 지체 효과를 갖고 있는 대기 순환은 Eurasia 지역을 기반으로 횡적인 순환과 관련이 깊은 것으로 파악되었다. 본 연구의 결과는 앞으로 우리나라의 가뭄관리에 유용하게 활용될 것으로 기대된다.

• Coupled systems mechanics
• /
• v.12 no.5
• /
• pp.461-479
• /
• 2023

The paper deals with the finite element modelling of the free vibration and structural behavior of a particular four-floor reinforced concrete structure subjected to static equivalent seismic loads and supported by a shallow foundation system called SNSF (Spider Net System Footing). The two FE models are a simple 2D Matlab model and a detailed 3D model based on solid elastic elements using Altairworks (Hypermesh and Optistruct). Both models can simulate the soil structure interaction. We concentrate on the behavior of a representative cell involving two columns on five levels. The influence of the boundary conditions on the external vertical planes of the domain are duly studied. The Matlab model appears relevant for a primary estimation of frequencies and stiffness of the whole structure under vertical and lateral loads.

• Computers and Concrete
• /
• v.33 no.6
• /
• pp.739-754
• /
• 2024

This study presents an innovative AI-driven approach to assess the ultimate axial load in Double-Skinned Profiled Steel sheet Composite Walls (DPSCWs). Utilizing a dataset of 80 entries, seven input parameters were employed, and various AI techniques, including Linear Regression, Polynomial Regression, Support Vector Regression, Decision Tree Regression, Decision Tree with AdaBoost Regression, Random Forest Regression, Gradient Boost Regression Tree, Elastic Net Regression, Ridge Regression, and LASSO Regression, were evaluated. Decision Tree Regression and Random Forest Regression emerged as the most accurate models. The top three performing models were integrated into a hybrid approach, excelling in accurately estimating DPSCWs' ultimate axial load. This adaptable hybrid model outperforms traditional methods, reducing errors in complex scenarios. The validated Artificial Neural Network (ANN) model showcases less than 1% error, enhancing reliability. Correlation analysis highlights robust predictions, emphasizing the importance of steel sheet thickness. The study contributes insights for predicting DPSCW strength in civil engineering, suggesting optimization and database expansion. The research advances precise load capacity estimation, empowering engineers to enhance construction safety and explore further machine learning applications in structural engineering.

• Nuclear Engineering and Technology
• /
• v.56 no.3
• /
• pp.824-831
• /
• 2024

In order to investigate the occurring processes during long-term dry storage of spent fuel assemblies, a joined project called SPIZWURZ, between the Karlsruhe Institute of Technology and the Gesellschaft für Anlagen-und Reaktorsicherheit (GRS), was started. Aim of the SPIZWURZ project is the determination and quantification of the influence of texture and elastic strain on diffusion and solubility of hydrogen in three different zirconium alloys used in western Europe during a long-term cooling transient (1 K/d) starting at 400 ℃. The strain in the cladding of an irradiated spent fuel rod shall be measured. Models predicting the formation of radial oriented hydrides will be validated, improved, and implemented in the GRS fuel rod performance code TESPA-ROD. This paper describes the SPIZWURZ project and already obtained first results.

• The Korean Journal of Applied Statistics
• /
• v.29 no.6
• /
• pp.1117-1128
• /
• 2016

In genetic association studies with high-dimensional genomic data, regularization procedures based on penalized likelihood are often applied to identify genes or genetic regions associated with diseases or traits. A network-based regularization procedure can utilize biological network information (such as genetic pathways and signaling pathways in genetic association studies) with an outstanding selection performance over other regularization procedures such as lasso and elastic-net. However, network-based regularization has a limitation because cannot be applied to high-dimension genomic data with a group structure. In this article, we propose to combine data dimension reduction techniques such as principal component analysis and a partial least square into network-based regularization for the analysis of high-dimensional genomic data with a group structure. The selection performance of the proposed method was evaluated by extensive simulation studies. The proposed method was also applied to real DNA methylation data generated from Illumina Innium HumanMethylation27K BeadChip, where methylation beta values of around 20,000 CpG sites over 12,770 genes were compared between 123 ovarian cancer patients and 152 healthy controls. This analysis was also able to indicate a few cancer-related genes.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.16 no.1
• /
• pp.17-22
• /
• 1980

Two plied twine, of which yarn is cored by abacca fibers and wrapped by polyethylen fibers, has been used for the material to web the wing net in the anchovy boat seine. This conventional twine revealed the disadvantage of which tensile strength goes from bad to worse. To improve the disadvantage, authors developed the new twine which is plied with three yarns, each yarn is cored by polyvinyliden chloride fibers and wrapped by polyethylen fibers. To compare the physical properties of the two kinds of twine, some factors are valued and the following results are found. 1. The conventional twine shrinks 4 to 7 percent when it was soaked in fresh water for 24 hours, whereas the improved one shrinks little. 2. The specific gravity of the conventional twine is valued 1. 04 to 1. 025 after soaking the twine in fresh water for 24 hours, in spite of including abacca to make the gravity heavier than that of sea water, whereas the improved ones are valued 1. 06 to 1. 09 without soaking. 3. The stiffness of the two kinds of twine differs little. 4. The breaking strength and the elastic recovery of the improved twine are greater than the conventional one. 5. Mesh size of the actual fishing net, webbed by the improved twine and employed in commercial fishing for 6 months, varied little.

• Nuclear Engineering and Technology
• /
• v.49 no.8
• /
• pp.1733-1739
• /
• 2017

Molecular dynamics simulations were performed to investigate the uniaxial tensile properties of nanocrystalline $U_{0.5}Th_{0.5}O_2$ solid solution with the Born-Mayer-Huggins potential. The results indicated that the elastic modulus increased linearly with the density relative to a single crystal, but decreased with increasing temperature. The simulated nanocrystalline $U_{0.5}Th_{0.5}O_2$ exhibited a breakdown in the Halle-Petch relation with mean grain size varying from 3.0 nm to 18.0 nm. Moreover, the elastic modulus of $U_{1-y}Th_yO_2$ solid solutions with different content of thorium at 300 K was also studied and the results accorded well with the experimental data available in the literature. In addition, the fracture mode of nanocrystalline $U_{0.5}Th_{0.5}O_2$ was inclined to be ductile because the fracture behavior was preceded by some moderate amount of plastic deformation, which is different from what has been seen earlier in simulations of pure $UO_2$.

• Nuclear Engineering and Technology
• /
• v.42 no.4
• /
• pp.450-459
• /
• 2010

The present paper investigates the collapse pressure of cylinders with intermediate thickness subjected to external pressure based on detailed elastic-plastic finite element (FE) analyses. The effect of the initial ovality of the tube on the collapse pressure was explicitly considered in the FE analyses. Based on the present FE results, the analytical yield locus, considering the interaction between the plastic collapse and local instability due to initial ovality, was also proposed. The collapse pressure values based on the proposed yield locus agree well with the present FE results; thus, the validity of the proposed yield locus for the thickness range of interest was verified. Moreover, the partial safety factor concept based on the structural reliability theory was also applied to the proposed collapse pressure estimation model, and, thus, the priority of importance of respective parameter constituting for the collapse of cylinders under external pressure was estimated in this study. From the application of the partial safety factor concept, the yield strength was concluded to be the most sensitive, and the initial ovality of tube was not so effective in the proposed collapse pressure estimation model. The present deterministic and probabilistic results are expected to be utilized in the design and maintenance of cylinders subjected to external pressure with initial ovality, such as the once-through type steam generator.

• Nuclear Engineering and Technology
• /
• v.52 no.12
• /
• pp.2949-2957
• /
• 2020

Fatigue crack growth model has been developed for dissimilar metal weld joints of a piping component under cyclic loading, where in the crack is located at the center of the weld in the circumferential direction. The fracture parameter, Stress Intensity Factor (SIF) has been computed by using principle of superposition as K_H + K_M. K_H is evaluated by assuming that, the complete specimen is made of the material containing the notch location. In second stage, the stress field ahead of the crack tip, accounting for the strength mismatch, the applied load and geometry has been characterized to evaluate SIF (K_M). For each incremental crack depth, stress field ahead of the crack tip has been quantified by using J-integral (elastic), mismatch ratio, plastic interaction factor and stress parallel to the crack surface. The associated constants for evaluation of K_M have been computed by using the quantified stress field with respect to the distance from the crack tip. Net SIF (K_H + K_M) computed, has been used for the crack growth analysis and remaining life prediction by Paris crack growth model. To validate the model, SIF and remaining life has been predicted for a pipe made up of (i) SA312 Type 304LN austenitic stainless steel and SA508 Gr. 3 Cl. 1. Low alloy carbon steel (ii) welded SA312 Type 304LN austenitic stainless-steel pipe. From the studies, it is observed that the model could predict the remaining life of DMWJ piping components with a maximum difference of 15% compared to experimental observations.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.65-73
• /
• 2023

In this paper, we propose a data augmentation method based on CNN(Convolutional Neural Network) learning for efficiently obtaining concrete crack image datasets. Real concrete crack images are not only difficult to obtain due to their unstructured shape and complex patterns, but also may be exposed to dangerous situations when acquiring data. In this paper, we solve the problem of collecting datasets exposed to such situations efficiently in terms of cost and time by using vector and thickness-based data augmentation techniques. To demonstrate the effectiveness of the proposed method, experiments were conducted in various scenes using U-Net-based crack detection, and the performance was improved in all scenes when measured by IoU accuracy. When the concrete crack data was not augmented, the percentage of incorrect predictions was about 25%, but when the data was augmented by our method, the percentage of incorrect predictions was reduced to 3%.