• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.413-435
• /
• 2018

Although the accuracy of handwritten character recognition based on deep networks has been shown to be superior to that of the traditional method, the use of an overly deep network significantly increases time consumption during parameter training. For this reason, this paper took the training time and recognition accuracy into consideration and proposed a novel handwritten character recognition algorithm with newly designed network structure, which is based on an extended nonlinear kernel residual network. This network is a non-extremely deep network, and its main design is as follows:(1) Design of an unsupervised apriori algorithm for intra-class clustering, making the subsequent network training more pertinent; (2) presentation of an intermediate convolution model with a pre-processed width level of 2;(3) presentation of a composite residual structure that designs a multi-level quick link; and (4) addition of a Dropout layer after the parameter optimization. The algorithm shows superior results on MNIST and SVHN dataset, which are two character benchmark recognition datasets, and achieves better recognition accuracy and higher recognition efficiency than other deep structures with the same number of layers.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.4
• /
• pp.285-293
• /
• 2021

An optimal operation method based on weighted efficiency for a two-stage boost converter is proposed in this study. Detailed loss analysis of the converter is performed to derive the optimal operation method according to the load and input voltage fluctuations, and the optimal DC-bus voltage is derived by applying the weighted efficiency method. The proposed method can satisfy optimal efficiency in the main operation region without a complicated control method. Using 1kW typical two-stage boost converter and is verified three types of weighted efficiency formulas and loss analysis are utilized to derive high-efficiency optimal DC-bus voltage from each load range.

• Journal of Power Electronics
• /
• v.15 no.3
• /
• pp.634-643
• /
• 2015

This paper presents a high frequency design approach for improving efficiency over a wide load range in the self-driven phase-shifted full-bridge converters for server power systems. In the proposed approach, a detailed ZVS analysis of the lagging leg switches in both the continuous conduction mode (CCM) and the discontinuous conduction mode (DCM) is presented. The optimum dead time and the determination of the appropriate operation mode are given for high efficiency according to the load conditions. Finally, the optimum operation conditions are defined to achieve a high-efficiency. A laboratory prototype operating at 80 kHz, rated 1 kW (12 V-83.3 A), is built to verify proposed theoretical analysis and evaluations. The experimental results show that the maximum efficiency is achieved as 95% and 83.5% at full load and 5% load conditions, respectively.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.27 no.3
• /
• pp.228-236
• /
• 2022

The charging methods of electric vehicles are divided into wired charging and wireless charging. Restrictions on the use of charging infrastructure for wireless charging vehicles currently exist because most charging infrastructure uses the wired charging method. Thus, wired and wireless integrated charging system has been studied. In this system, a wireless charging system especially requires a control method for high-efficiency operation in consideration of a change in a coupling coefficient. Therefore, this paper introduces two control methods for the high-efficiency operation of wireless charging that can be applied to wired and wireless integrated charging systems. In addition, loss analysis is performed through PSIM simulation to select a more advantageous method for high-efficiency operation among the two control methods. To verify the simulation-based loss analysis result, the two control methods are applied to the actual wireless charging system, and the efficiency is compared through the experiments Based on the experimental results, a control method suitable for high-efficiency operation of the wireless charging method is selected.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2000.01a
• /
• pp.23-24
• /
• 2000

A new PDP cell structure called CSP (Charge Storage Pad) improves the luminous efficiency by 1.6 times and prevents cross talk between adjacent cells. The CSP, which is a conducting material, is inserted between the dielectric layer and the MgO film in the front plate. This CSP produces a longer time-averaged discharge path to get a high luminous efficiency and confines the discharge to prevent cross talk.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.24 no.4
• /
• pp.259-267
• /
• 2019

An active clamp flyback (ACF) converter applies a clamp circuit and circulates the energy of leakage inductance to the input side, thereby achieving a zero-voltage switching (ZVS) operation and greatly reducing switching losses. The switching losses are further reduced by applying a gallium nitride field effect transistor (GaN-FET) with excellent switching characteristics, and ZVS operation can be accomplished under light load with boundary conduction mode (BCM) operation. Optimal design is performed on the basis of loss analysis by selecting magnetization inductance based on BCM operation and a clamp capacitor for loss reduction. Therefore, the size of the reactive element can be reduced through high-frequency operation, and a high-efficiency and high-power-density converter can be achieved. This study proposes an optimal design for a high-efficiency and high-power-density BCM ACF converter based on GaN-FETs and verifies it through experimental results of a 65 W-rated prototype.

• Proceedings of the KIPE Conference
• /
• 2006.06a
• /
• pp.131-133
• /
• 2006

A high efficiency half bridge converter for PDP in whole operation load range is proposed in this paper. The ZVS(Zero Voltage Switching) in whole load range and high efficiency in low load range is achieved because the proposed converter is switched by fixed frequency asymmetrical duty ratio and resonant operation.

• Proceedings of the KIPE Conference
• /
• 2017.07a
• /
• pp.38-39
• /
• 2017

This paper proposes a new control method for an AC-DC Buck converter which is utilized as a front-end converter of a 2-stage high power density adapter. In the conventional adapter applications, 2-stage configuration shows higher power transfer efficiency and higher power density than those of the single stage flyback converter. In the 2-stage AC-DC converter, the boost converter is widely used as a front-end converter. However, an efficiency variation between high AC line and low AC line is large. On the other hand, the proposed conduction band control method for a buck front-end converter has an advantage of small efficiency variation. In the proposed control method, switching operation is determined by a band control voltage which represents output load condition, and an AC line voltage. If the output load increasesin low AC line, the switching operation range is expanded in half of line cycle. On the contrary, in light load and high line condition, the switching operation is narrowed. Thus, the proposed control method reduces switching loss under high AC line and light load condition. A 60W prototype which is configured the buck and LLC converter with the proposed control method is experimented on to verify the validity of the proposed system. The prototype shows 92.16% of AC-DC overall efficiency and 20.19 W/in 3 of power density.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.9
• /
• pp.1239-1248
• /
• 2018

This paper proposes a fault tolerant and high efficiency operation algorithm for a 4 LEG DC/DC converter for a battery energy storage system(BESS) forming a main power source in a standalone DC micro grid. The BESS for the main power supply in the stand-alone DC micro-grid is required to operate at high speed according to fault tolerant control and load by operating at all times. Fault-tolerance control changes the short-circuit fault to an open-circuit fault by using a fuse in case of leg fault in 4 legs, and operates stably through phase shift control. In addition, considering the loss of the power semiconductor, the number of LEG operation is adjusted to operate at high efficiency in the full load region. In this paper, fault tolerant control and high efficiency operation algorithm of DC/DC converter for BESS in standalone DC micro grid is presented and it is proved through simulation and experiment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1761-1777
• /
• 2021

An image with infrared features and visible details is obtained by processing infrared and visible images. In this paper, a fusion method based on Laplacian pyramid and generative adversarial network is proposed to obtain high quality fusion images, termed as Laplacian-GAN. Firstly, the base and detail layers are obtained by decomposing the source images. Secondly, we utilize the Laplacian pyramid-based method to fuse these base layers to obtain more information of the base layer. Thirdly, the detail part is fused by a generative adversarial network. In addition, generative adversarial network avoids the manual design complicated fusion rules. Finally, the fused base layer and fused detail layer are reconstructed to obtain the fused image. Experimental results demonstrate that the proposed method can obtain state-of-the-art fusion performance in both visual quality and objective assessment. In terms of visual observation, the fusion image obtained by Laplacian-GAN algorithm in this paper is clearer in detail. At the same time, in the six metrics of MI, AG, EI, MS_SSIM, Q_abf and SCD, the algorithm presented in this paper has improved by 0.62%, 7.10%, 14.53%, 12.18%, 34.33% and 12.23%, respectively, compared with the best of the other three algorithms.