• Proceedings of the KIPE Conference
• /
• 2019.07a
• /
• pp.11-13
• /
• 2019

This paper proposes a novel high step-up non-isolated DC-DC converter, suitable for regulating dc bus in various inherent low voltage micro sources especially for photovoltaic (PV) and fuel cell sources. This novel high voltage Non-isolated Boost DC-DC converter topology is best replacement, where high voltage conversion ratio is required without the transformer and also need continuous input current. Since the proposed topology utilizes the stack-based structure, the voltage gain, and the efficiency are higher than other conventional non-isolated converters. Switches in this topology is easier to control since its control signal is grounding reference. Also, there is no need of extra gate driver and extra power supply for driver circuit, which reduces the cost and size of system. In order to show the feasibility and practicality of the proposed topology principle operation, steady state analysis and simulation result is presented and analyzed in detail. To verify the performance of proposed converter and theoretical analysis 360W laboratory prototype is implemented.

• Composites Research
• /
• v.34 no.3
• /
• pp.180-185
• /
• 2021

Due to the formation of random three dimensional network structure, which cause a lot of scattering of phonons, the thermal conductivity is low when the liquid crystalline epoxy is cured with amine-based curing agent. This problem is solved by using a cationic initiator that can make mesogen groups to be stacked structure. In this experiment, the thermal stability is compared by investigating the activation energy of isothermal decomposition through TGA of an epoxy using an amine-based curing agent and a cationic initiator. As a result, the energy of the activation of the epoxy using a cationic initiator is high. Compared with the previous experiments, the thermal stability is similar to the thermal conductivity.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.977-985
• /
• 2020

Aiming at the problem of complex feature extraction and low accuracy in human action recognition, this paper proposed a network structure combining batch normalization algorithm with GoogLeNet network model. Applying Batch Normalization idea in the field of image classification to action recognition field, it improved the algorithm by normalizing the network input training sample by mini-batch. For convolutional network, RGB image was the spatial input, and stacked optical flows was the temporal input. Then, it fused the spatio-temporal networks to get the final action recognition result. It trained and evaluated the architecture on the standard video actions benchmarks of UCF101 and HMDB51, which achieved the accuracy of 93.42% and 67.82%. The results show that the improved convolutional neural network has a significant improvement in improving the recognition rate and has obvious advantages in action recognition.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2002.10a
• /
• pp.170-174
• /
• 2002

In this study, the material characterization and the dynamic behavior of 3D orthogonal woven composite materials has been studied under transverse central low-velocity impact condition by means of the micromechanical model using finite elements. To build up the micromechanical model considering tow spacing and waviness, an accurate unit structure is stacked in x-y-z direction repeatedly. First, the mechanical properties of 3D orthogonal woven composites are obtained by means of virtual experiment using full scale Finite Element Analysis based on the DNS concepts, and the computed elastic properties are validated by comparison to available experimental results[9]. Second, using the implementation of this validated micromechanical model, 3D transient finite-element analysis is performed considering contact and impact, and the impact behavior of 3D orthogonal woven composite is investigated. A comparison study will be carried out in terms of energy absorption capabilities.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2005.11a
• /
• pp.53-56
• /
• 2005

In this study, the material characterization and the dynamic behavior of 3D orthogonal woven composite materials has been studied under transverse central low-velocity impact condition by means of the micromechanical model using finite elements. To build up the micromechanical model considering tow spacing and waviness, an accurate unit structure is stacked in x-y-z direction repeatedly. First, the mechanical properties of 3D orthogonal woven composites arc obtained by means of virtual experiment using full scale Finite Element Analysis based on the DNS concepts, and the computed elastic properties arc validated by comparison to available experimental results. Second, using the implementation of this validated micromechanical model, 3D transient finite-clement analysis is performed considering contact and impact, and the impact behavior of 3D orthogonal woven composite is investigated. A comparison study with the homogenized model will be carried out in terms of global and local behaviors.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2009.10a
• /
• pp.489-492
• /
• 2009

We report white OLED with high color temperature based on simple stacked structure for medical display applications. White OLED was fabricated with two emitting materials of fluorescent blue dopant and phosphorescent yellow dopant. We achieved luminance efficiency of 16.2cd/A and CIE color coordinates of (0.305, 0.317) at 10mA/$cm^2$. In particular, the correlated color temperature was higher than 7,000K, enough for display applications.

• ETRI Journal
• /
• v.31 no.2
• /
• pp.225-227
• /
• 2009

A new planar metamaterial (MTM) with simultaneous negative values of permittivity (${\varepsilon}$) and permeability (${\mu}$) is proposed. Our MTM is composed of two identical copper patterns etched on both sides of dielectric laminate, which is very thin and easy to fabricate. Unlike conventional MTMs, the proposed structure shows a negative refractive index (NRI) behavior with respect to a normally incident wave. To explain the underlying principle of the NRI characteristics, an equivalent resonant circuit model based on surface current density distribution is investigated. An eigenmode analysis and a three-dimensional wave simulation for the stacked MTM prism are also performed to verify the existence of negative refraction. The experimental results from the transmission and reflection measurement ensure the validity of our design approach and show good agreement with the theoretically predicted effective medium parameters.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.289-297
• /
• 2019

In this paper, a wideband, circularly polarized patch antenna is proposed that leverages the unidirectional resonant modes of a circular patch mounted on top of a grounded dielectric-ferrite substrate. The proposed antenna is fed via the proximity coupling method and several parasitically coupled patches are placed on a dielectric superstrate to enhance the impedance bandwidth of the antenna. The resonant modes of the structure rotate only in the clockwise or counter clockwise directions. In the frequency range where the effective permeability of the ferrite layer is negative, the resonance frequencies of these modes differ significantly, which produces a large axial ratio (AR) bandwidth. For the proposed antenna, the numerical results show the 10 dB impedance bandwidth to be around 44% and the 3 dB axial ratio bandwidth to be higher than 64%.

• Advances in Computational Design
• /
• v.8 no.1
• /
• pp.37-59
• /
• 2023

Squat reinforced concrete (SRC) shear walls are a critical part of the structure for both office/residential buildings and nuclear structures due to their significant role in withstanding seismic loads. Despite this, empirical formulae in current design standards and published studies demonstrate a considerable disparity in predicting SRC wall shear strength. The goal of this research is to develop and evaluate hybrid and ensemble artificial neural network (ANN) models. State-of-the-art population-based algorithms are used in this research for hybrid intelligence algorithms. Six models are developed, including Honey Badger Algorithm (HBA) with ANN (HBA-ANN), Hunger Games Search with ANN (HGS-ANN), fitness-distance balance coyote optimization algorithm (FDB-COA) with ANN (FDB-COA-ANN), Averaging Ensemble (AE) neural network, Snapshot Ensemble (SE) neural network, and Stacked Generalization (SG) ensemble neural network. A total of 434 test results of SRC walls is utilized to train and assess the models. The results reveal that the SG model not only minimizes prediction variance but also produces predictions (with R²= 0.99) that are superior to other models.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2014.02a
• /
• pp.185.2-185.2
• /
• 2014

We have studied the atomic and electronic structure of graphene nanoribbons (GNRs) on a hexagonal boron nitride (h-BN) sheet with intercalated atoms using first-principles calculations. The h-BN sheet is an insulator with the band gap about 6 eV and then it may a good candidate as a supporting dielectric substrate for graphene-based nanodevices. Especially, the h-BN sheet has the similar bond structure as graphene with a slightly longer lattice constant. For the computation, we use the Vienna ab initio simulation package (VASP). The generalized gradient approximation (GGA) in the form of the PBE-type parameterization is employed. The ions are described via the projector augmented wave potentials, and the cutoff energy for the plane-wave basis is set to 400 eV. To include weak van der Waals (vdW) interactions, we adopt the Grimme's DFT-D2 vdW correction based on a semi-empirical GGA-type theory. Our calculations reveal that the localized states appear at the zigzag edge of the GNR on the h-BN sheet due to the flat band of the zigzag edge at the Fermi level and the localized states rapidly decay into the bulk. The open-edged graphene with a large corrugation allows some space between graphene and h-BN sheet. Therefore, atoms or molecules can be intercalated between them. We have considered various types of atoms for intercalation. The atoms are initially placed at the edge of the GNR or inserted in between GNR and h-BN sheet to find the effect of intercalated atoms on the atomic and electronic structure of graphene. We find that the impurity atoms at the edge of GNR are more stable than in between GNR and h-BN sheet for all cases considered. The nickel atom has the lowest energy difference of ~0.2 eV, which means that it is relatively easy to intercalate the Ni atom in this structure. Finally, the magnetic properties of intercalated atoms between GNR and h-BN sheet are investigated.