• Journal of Digital Convergence
• /
• v.19 no.7
• /
• pp.271-282
• /
• 2021

This study was conducted to develop a model for predicting the length of stay for premature infants through machine learning. For the development of this model, 6,149 cases of premature infants discharged from the hospital from 2011 to 2016 of the discharge injury in-depth survey data collected by the Korea Centers for Disease Control and Prevention were used. The neural network model of the initial hospitalization was superior to other models with an explanatory power (R²) of 0.75. In the model added by converting the clinical diagnosis to CCS(Clinical class ification software), the explanatory power (R²) of the cubist model was 0.81, which was superior to the random forest, gradient boost, neural network, and penalty regression models. In this study, using national data, a model for predicting the length of stay for premature infants was presented through machine learning and its applicability was confirmed. However, due to the lack of clinical information and parental information, additional research is needed to improve future performance.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.6B
• /
• pp.571-578
• /
• 2010

Manning's roughness coefficients were reevaluated for the computation of river flow of the Han River, the Nakdong River and the Geum River. The roughness coefficients were estimated by two methods. One is based on the assumption that roughness is primarily a function of grain diameter and the other is based on the findings that roughness may vary significantly with the flow discharge. The roughness coefficients adopted in each river improvement master plan have been compared with those obtained using the FLDWAV in this study, and their applicabilities have been reviewed, using the FLDWAV and HEC-RAS models. The design flood water levels computed by the abovementioned models with the roughness coefficients proposed in this study have shown good agreement with the measurements of time variation. The roughness coefficients computed using the FLDWAV model showed nearly no close correlation with the various hydraulic characteristic factors, such as grain size and river depth, etc.. Finally the design flood water levels and levee safety about the downstream part from the Paldang Dam of the Han River has been reviewed using HEC-2 model with roughness coefficients of this study and the results indicated that some parts of the existing levees were short of safety.

• Journal of Astronomy and Space Sciences
• /
• v.27 no.3
• /
• pp.253-262
• /
• 2010

The electrical power system (EPS) of Korean satellites in low-earth-orbit is designed to achieve energy balance based on a one-orbit mission scenario. This means that the battery has to be fully charged at the end of a one-orbit mission. To provide the maximum solar array (SA) power generation, the peak power tracking (PPT) method has been developed for a spacecraft power system. The PPT is operated by a software algorithm, which tracks the peak power of the SA and ensures the battery is fully charged in one orbit. The EPS should be designed to avoid the stress of electronics in order to handle the main bus power from the SA power. This paper summarizes the results of energy balance to achieve optimal power sizing and the actual trend analysis of EPS performance in orbit. It describes the results of required power for the satellite operation in the worst power conditions at the end-of-life, the methods and input data used in the energy balance, and the case study of energy balance analyses for the normal operation in orbit. Both 10:35 AM and 10:50 AM crossing times are considered, so the power performance in each case is analyzed with the satellite roll maneuver according to the payload operation concept. In addition, the data transmission to the Korea Ground Station during eclipse is investigated at the local-time-ascending-node of 11:00 AM to assess the greatest battery depth-of-discharge in normal operation.

• Journal of Soil and Groundwater Environment
• /
• v.16 no.2
• /
• pp.61-76
• /
• 2011

This study was objected to provide suggestions for best management practices to restore the cultural and historical values of the wells in Palaces as well as their water qualities. Water resources in the five Palaces in Seoul Metropolitan, including Gyeongbokgung, Changdeokgung, Changgyeonggung, Jongmyo Shrine, and Deoksugung, were surveyed for their physical flows and chemical compositions from April to July in 2010. Ground waters in most wells were found at depths within 5 m from the ground surface, showing typical water-table aquifer systems. Hydraulic gradients indicate water resources in Gyeongbokgung, Changdeokgung, and Changgyeonggung flowing toward south, and toward east in Deoksugung area. Especially, water-level fluctuation data at S-10 in Deoksugung implied the influence of groundwater discharge facility. In Jongmyo Shrine, water was not detected in wells, indicating the water level was lower than the well depth. Based on the water chemistry and stable isotope analyses, water resources and their qualities appeared to be formed by the water-rock interaction along the groundwater paths. S-10 (Deoksugung) and S-14 (Changgyeonggung) samples were contaminated with nitrate ($NO_3$) in levels of higher than Korean drinking water standard, 10 mg/L as $NO_3$-N, but once in four sampling campaigns. In the situation that water resources in Palaces still maintain natural characteristics, the materials that will be used for the restoration and improvement of the Palace water supplies should be carefully selected not to disturb the natural integrity. In addition, because the wells are located in the center of metropolitan area, a systematic monitoring should be applied to detect and to manage the potential impacts of underground construction and various pollution sources.

• Korean Journal of Materials Research
• /
• v.23 no.6
• /
• pp.334-338
• /
• 2013

In this study, $BaTiO_3$ thin films were grown by RF-magnetron sputtering, and the effects of the thin film thickness on the structural characteristics of $BaTiO_3$ thin films were systematically investigated. Instead of the oxide substrates generally used for the growth of $BaTiO_3$ thin films, p-Si substrates which are widely used in the current semiconductor processing, were used in this study in order to pursue high efficiency in device integration processing. For the crystallization of the grown thin films, annealing was carried out in air, and the annealing temperature was varied from $700^{\circ}C$. The changed thickness was within 200 nm~1200 nm. The XRD results showed that the best crystal quality was obtained for ample thicknesses 700 nm~1200 nm. The SEM analysis revealed that Si/$BaTiO_3$ are good quality interface characteristics within 300 nm when observed thickness. And surface roughness observed of $BaTiO_3$ thin films from AFM measurement are good quality surface characteristics within 300 nm. Depth-profiling analysis through GDS (glow discharge spectrometer) showed that the stoichiometric composition could be maintained. The results obtained in this study clearly revealed $BaTiO_3$ thin films grown on a p-Si substrate such as thin film thickness. The optimum thickness was 300 nm, the thin film was found to have the characteristics of thin film with good electrical properties.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.12 no.3
• /
• pp.115-125
• /
• 1992

The variation of flow pattern caused by the topographical change of Kwangyang bay, is analyzed using the numerical tidal model for the depth-integrated two- dimensional long wave equation. The results of study are as follows. Due to pier construction, the area of water surface is deceased and the water inflow into the Kwangyang bay is reduced. For this result, at the outer bay of Myo island, the tidal range is slightly increased. And at the inner bay, water level is dropped generally, and especially at the time of low water tide, the phenomena of water level drop obviously appears. According to the variation pattern, flow velocities is lower than those of non-construction condition over the Kwangyang bay. But at the channel(from Kwangyang east stream) flowing into the east Kwangyang bay, for the contraction of channel profile, flow velocity is increased. The study based on the 100 year frequency design flood discharge from Sueocheon(river) and Dongcheon(river) which are flowing into the bay and Seomjin River flowing along the boundary of the bay is also performed. During the spring tide condition, the results showed the rise of water level about 1.2 m at Seomjin River Estuary and 0.3 m at inner bay is occurred.

• The Journal of Engineering Geology
• /
• v.25 no.4
• /
• pp.449-458
• /
• 2015

The compilation of a flood hazard map is an efficient technique in managing areas at risk of flooding in the case of a dam-break. A scenario-based numerical modeling approach is commonly used to compile a flood hazard map related to dam-break and to determine the model parameters that capture peak discharge, including breach formation and progress, which are important in the modeling method. This approach might be considered less reliable if an existing model is used without local validation. In this study, a dam-break model is linked to a routing model to identify flood-risk areas in the case of failure of the Sandae Reservoir Gyeongju, Gyeongbuk. Model parameters are extracted from a DEM, and maps of land use and soil texture. The simulation results are compared with on-site investigations in terms of inundation and depth. The model reproduces the inundation zone with reasonable accuracy.

• Journal of Digital Convergence
• /
• v.21 no.2
• /
• pp.17-25
• /
• 2023

The purpose of this study was to analyze the characteristics of elderly motorcycle accidents according to data from elderly inpatients to prepare prevention measures for the elderly against injury in motorcycle accidents. Chi-squared test, independent sample t-test, and canonical correlation analysis were performed on the Korea Disease Control and Prevention Agency's National Hospital Discharge In-depth Injury Survey data from 2015 to 2019, from which the records of 1,384 elderly inpatients hospitalized because of motorcycle accidents were obtained. intracranial injury(S06) was the most common care and treatment characteristic for both age groups. The most frequent injury site was the head and neck, and the most frequent injury type was a fracture. The above findings show that prevention education and policy formulation at the national level are necessary to identify and manage the factors of elderly motorcycle accidents. This study provides basic data for developing measures and policies to prevent and reduce injuries, making it significant for public health causes.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1661-1665
• /
• 2006

In a natural river, cross sections of a channel vary according to inner or outer parts of meandering. Generally, depth of outer parts is deeper than that of inner parts. This kind of cross section change by meandering can be demonstrated by Beta distribution. The objects of this research is a 3D simulation of primary and secondary flow in the meandering natural channel. FLOW-3D program, a numerical model using CFD technique, and LES method was used for this research. 3D simulations were conducted in the channels having Beta distribution cross sections which have beds of mortar, gravel and vegetation. Two types of water stages and discharge were applied to each channel. In this research, primary flows are located in the outer parts of a top of bend and secondary flows rotate in the bottom on outer parts.

• Journal of the Korean Society for Railway
• /
• v.18 no.4
• /
• pp.325-334
• /
• 2015

This paper deals with a control method for a hybrid energy storage system (ESS) with a battery and super capacitor for grid-connected operation. To improve the grid regulation and to minimize the depth of discharge of the battery, control algorithms for the hybrid ESS are proposed. A p-q control with predictive current control is used for the inverters for the super-capacitor and the battery. To verify the effectiveness of the proposed control method, simulation studies using the PSIM and RTDS are carried out for the actual grid model.