• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.35 no.4
• /
• pp.377-385
• /
• 2022

Thermal batteries are specialized as primary reserve batteries that operate when the internal heat source is ignited and the produced heat (450~550℃) melts the initially insulating salt into highly conductive eutectic electrolyte. The heat source is composed of Fe powder and KClO₄ with different mass ratios and is inserted in-between the cells (stacks) to allow homogeneous heat transfer and ensure complete melting of the electrolyte. An ideal heat source has following criteria to satisfy: sufficient mechanical durability for stacking, appropriate heat calories, ease of combustion by an igniter, stable combustion rate, and modest peak temperature. To satisfy the aforementioned requirements, Fe powder must have high surface area and porosity to increase the reaction rate. Herein, the hydrothermal and spray drying synthesis techniques for Fe powder samples are employed to investigate the physicochemical properties of Fe powder samples and their applicability as a heat source constituent. The direct comparison with the state-of-the-art Fe powder is made to confirm the validity of synthesized products. Finally, the actual batteries were made with the synthesized iron powder samples to examine their performances during the battery operation.

• Journal of Korean Society of Disaster and Security
• /
• v.16 no.4
• /
• pp.129-136
• /
• 2023

A special offshore bridge with a high pylon has special structural features.Special offshore bridges have inspection blind spots that are difficult to visually inspect. To solve this problem, safety inspection methods using drones are being studied. In this study, image data of the pylon of a special offshore bridge was acquired using a drone. In addition, an artificial intelligence algorithm was developed to detect damage to the pylon surface. The AI algorithm utilized a deep learning network with different structures. The algorithm applied the stacking ensemble learning method to build a model that formed the ensemble and collect the results.

• Geophysics and Geophysical Exploration
• /
• v.16 no.3
• /
• pp.171-179
• /
• 2013

An automated ultrasonic velocity measurement system adopting pulse-echo-overlap (PEO) method has been constructed, which is known to be a precise and versatile method. It has been applied to velocity measurements for 5 kinds of engineering plastic cores and compared to first arrival picking (FAP) method. Because it needs multiple reflected waves and waves travel at least 4 times longer than FAP, PEO has basic restriction on sample length measurable. Velocities measured by PEO showed slightly lower than that by FAP, which comes from damping and diffusive characteristics of the samples as the wave travels longer distance in PEO. PEO, however, can measure velocities automatically by cross-correlating the first echo to the second or third echo, so that it can exclude the operator-oriented errors. Once measurable, PEO shows essentially higher repeatability and reproducibility than FAP. PEO system can diminish random noises by stacking multiple measurements. If it changes the experimental conditions such as temperature, saturation and so forth, the automated PEO system in this study can be applied to monitoring the velocity changes with respect to the parameter changes.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.111-124
• /
• 2018

This study aims to develop a classification model for predicting the occurrence of hyperlipidemia, one of the chronic diseases. Prior studies applying data mining techniques for predicting disease can be classified into a model design study for predicting cardiovascular disease and a study comparing disease prediction research results. In the case of foreign literatures, studies predicting cardiovascular disease were predominant in predicting disease using data mining techniques. Although domestic studies were not much different from those of foreign countries, studies focusing on hypertension and diabetes were mainly conducted. Since hypertension and diabetes as well as chronic diseases, hyperlipidemia, are also of high importance, this study selected hyperlipidemia as the disease to be analyzed. We also developed a model for predicting hyperlipidemia using SVM and meta learning algorithms, which are already known to have excellent predictive power. In order to achieve the purpose of this study, we used data set from Korea Health Panel 2012. The Korean Health Panel produces basic data on the level of health expenditure, health level and health behavior, and has conducted an annual survey since 2008. In this study, 1,088 patients with hyperlipidemia were randomly selected from the hospitalized, outpatient, emergency, and chronic disease data of the Korean Health Panel in 2012, and 1,088 nonpatients were also randomly extracted. A total of 2,176 people were selected for the study. Three methods were used to select input variables for predicting hyperlipidemia. First, stepwise method was performed using logistic regression. Among the 17 variables, the categorical variables(except for length of smoking) are expressed as dummy variables, which are assumed to be separate variables on the basis of the reference group, and these variables were analyzed. Six variables (age, BMI, education level, marital status, smoking status, gender) excluding income level and smoking period were selected based on significance level 0.1. Second, C4.5 as a decision tree algorithm is used. The significant input variables were age, smoking status, and education level. Finally, C4.5 as a decision tree algorithm is used. In SVM, the input variables selected by genetic algorithms consisted of 6 variables such as age, marital status, education level, economic activity, smoking period, and physical activity status, and the input variables selected by genetic algorithms in artificial neural network consist of 3 variables such as age, marital status, and education level. Based on the selected parameters, we compared SVM, meta learning algorithm and other prediction models for hyperlipidemia patients, and compared the classification performances using TP rate and precision. The main results of the analysis are as follows. First, the accuracy of the SVM was 88.4% and the accuracy of the artificial neural network was 86.7%. Second, the accuracy of classification models using the selected input variables through stepwise method was slightly higher than that of classification models using the whole variables. Third, the precision of artificial neural network was higher than that of SVM when only three variables as input variables were selected by decision trees. As a result of classification models based on the input variables selected through the genetic algorithm, classification accuracy of SVM was 88.5% and that of artificial neural network was 87.9%. Finally, this study indicated that stacking as the meta learning algorithm proposed in this study, has the best performance when it uses the predicted outputs of SVM and MLP as input variables of SVM, which is a meta classifier. The purpose of this study was to predict hyperlipidemia, one of the representative chronic diseases. To do this, we used SVM and meta-learning algorithms, which is known to have high accuracy. As a result, the accuracy of classification of hyperlipidemia in the stacking as a meta learner was higher than other meta-learning algorithms. However, the predictive performance of the meta-learning algorithm proposed in this study is the same as that of SVM with the best performance (88.6%) among the single models. The limitations of this study are as follows. First, various variable selection methods were tried, but most variables used in the study were categorical dummy variables. In the case with a large number of categorical variables, the results may be different if continuous variables are used because the model can be better suited to categorical variables such as decision trees than general models such as neural networks. Despite these limitations, this study has significance in predicting hyperlipidemia with hybrid models such as met learning algorithms which have not been studied previously. It can be said that the result of improving the model accuracy by applying various variable selection techniques is meaningful. In addition, it is expected that our proposed model will be effective for the prevention and management of hyperlipidemia.

• Journal of the Korean Chemical Society
• /
• v.51 no.2
• /
• pp.178-185
• /
• 2007

Intermediate pentasil borosilicate zeolite-like materials have been crystallized by a novel method named steam-assisted conversion, which involves vapor-phase transport of water. Indeed, amorphous powders obtained by drying Na2O.SiO2.B2O3.TBA2O gels of various compositions using different boron sources are transformed into crystalline borosilicate zeolite belonging to pentasil family structure by contact with vapors of water under hydrothermal conditions. Using a variant of this method, a new material which has an intermediate structure of MFI/MEL in the ratio 90:10 was crystallized. The results show that steam and sufficiently high pH in the reacting hydrous solid are necessary for the crystallization to proceed. Characterization of the products shows some specific structural aspects which may have its unique catalytic properties. X-ray diffraction patterns of these microporous crystalline borosilicates are subjected to investigation, then, it is shown that the product structure has good crystallinity and is interpreted in terms of regular stacking of pentasil layers correlated by inversion centers (MFI structure) but interrupted by faults consisting of mirror-related layers (MEL structure). The products are also characterized by nitrogen adsorption at 77 K that shows higher microporous volume (0.160 cc/g) than that of pure MFI phase (0.119 cc/g). The obtained materials revealed high surface area (~600 m2/g). The infrared spectrum reveals the presence of an absorption band at 900.75 cm-1 indicating the incorporation of boron in tetrahedral sites in the silicate matrix of the crystalline phase.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.22 no.3
• /
• pp.275-283
• /
• 2002

To detect the position and depth of buried underground utilities, method of Ground Penetrating Radar(GPR) survey is the most commonly used. However, the skin-depth of GPR is very shallow, and in the places where subsurface materials are not homogeneous and are compose of clays and/or salts and gravels, GPR method has limitations in application and interpretation. The aim of this study is to overcome these limitations of GPR survey. For this purpose the site where the GPR survey is unsuccessful to detect the underground big pipes is selected, and soil tests were conducted to confirm the reason why GPR method was not applicable. Non-destructive high-frequency electromagnetic (HFEM) survey was newly developed and was applied in the study area to prove the effectiveness of this new technique. The frequency ranges $2kHz{\sim}4MHz$ and the skin depth is about 30m. The HFEM measures the electric field and magnetic field perpendicular to each other to get the impedance from which vertical electric resistivity distribution at the measured point can be deduced. By adopting the capacitive coupled electrodes, it can make the measuring time shorter, and can be applied to the places covered by asphalt an and/or concrete. In addition to the above mentioned advantages, noise due to high-voltage power line is much reduced by stacking the signals. As a result, the HFEM was successful in detecting the buried underground objects. Therefore this method is a promising new technique that can be applied in the lots of fields, such as geotechnical and archaeological surveys.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.25 no.4
• /
• pp.127-134
• /
• 2015

Aluminum nitride (AlN) single crystals have attracted much attention for a next-generation semiconductor application because of wide bandgap (6.2 eV), high thermal conductivity ($285W/m{\cdot}K$), high electrical resistivity (${\geq}10^{14}{\Omega}{\cdot}cm$), and high mechanical strength. The bulk AlN single crystals or thin film templates have been mainly grown by PVT (sublimation) method, flux method, solution growth method, and hydride vapor phase epitaxy (HVPE) method. Since AlN suffers difficulty in commercialization due to the defects that occur during single crystal growth, crystalline quality improvement via defects analyses is necessary. Etch pit density (EPD) analysis showed that the growth misorientations and the defects in the AlN surface exist. Transmission electron microscopy (TEM) and electron back-scattered diffraction (EBSD) analyses were employed to investigate the overall crystalline quality and various kinds of defects. TEM studies show that the morphology of the AlN is clearly influenced by stacking fault, dislocation, second phase, etc. In addition EBSD analysis also showed that the zinc blende polymorph of AlN exists as a growth defects resulting in dislocation initiator.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.2
• /
• pp.276-284
• /
• 2020

This paper proposes a new convolutional neural network (CNN) based spectrum sensing technique for cognitive radio communications. The proposed technique determines the existence of the primary user (PU) by using energy detection without any prior knowledge of the PU's signal. In the proposed method, the received signal is high-rate sampled to sense the entire spectrum bands of interest. After that, fast Fourier transform (FFT) of the signal converts the time domain signal to frequency domain spectrum and by stacking those consecutive spectrums, a 2 dimensional signal is made. The 2 dimensional signal is cut by the sensing channel bandwidth and inputted to the CNN. The CNN determines the existence of the primary user. Since there are only two states (existence or non-existence), binary classification CNN is used. The performance of the proposed method is examined through computer simulation and indoor experiment. According to the results, the proposed method outperforms the conventional threshold-based method by over 2 dB.

• Journal of the Korean Magnetics Society
• /
• v.7 no.2
• /
• pp.82-89
• /
• 1997

Dependence of magnetoresistance on the thickness of Cu, type and thickness of buffer layer, and the stacking number of multilayer in the form buffer /$[Co(17{\AA}/Cu(t{\AA})]_{20}$ were investigated. To evaluate effect of annealing on this samples, X-ray diffraction analysis, vibrating sample magnetometer analysis, and magnetoresistance measurement (4-probe method) were performed. The magnetoresistance ratio exhibits a maximum of 21% for the multilayer with Cu thickness of 24$\AA$ and Fe buffer layer thickness of 50$\AA$. Deposition of film under low base pressure induces in increase magnetoresistance ratio by preventing oxidation. The multilayer annealed below 30$0^{\circ}C$ temperature allowed larger textured grain without loss in the periodicity. Magnetoresistance ratios of the multilayer with Cu thickness of 24$\AA$ and 36$\AA$ were increased due to the increase in the antiferromagnetically coupled fraction after annealing.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.1A
• /
• pp.31-37
• /
• 2012

The phase distribution in concrete materials strongly affects its material properties. It is important to identify the spatial distribution of void in concrete because the void in concrete materials affects mechanical behavior and permeability significantly. Therefore, a proper method to describe the void distribution of a material is needed. In this research, CT(computed tomography) is used to examine and to quantify the void distribution of porous concrete specimens. 3D concrete digital specimens are created by subsequent stacking of 2D cross-sectional images from CT. Then, probability distribution functions such as two-point correlation, lineal-path and two-point cluster functions are used for void distribution characterization. It is confirmed that probability distribution functions obtained from CT images are effective in characterizing void distributions including the anisotropy and percolation.