• Proceedings of the KIPE Conference
• /
• 2015.11a
• /
• pp.221-222
• /
• 2015

ITER AC/DC Converter의 부하는 초전도 코일이며 이에 필요한 컨버터는 총 6종류(2상한:TF, 4상한:PF, CS, VS, CCU/L, CCS)가 있다. 이중 VS 컨버터(${\pm}1050V$, ${\pm}22.5kA$)는 6대가 직렬로 접속되어 운전되고 CS 컨버터(${\pm}1050V$, ${\pm}4.5kA$)는 4대가 직렬로 접속되어 운전한다. 이들 컨버터용 제어기의 개발 단계에서 실 부하상태를 준비하는 것은 어렵기 때문에 $RTDS^{TM}$ (Real Time Digital Simulator)를 이용하여 제어 대상인 High Power 부분과 초전도 코일의 동적 시스템 모델을 HILS(Hardware-in-the-Loop Simulation)로 구축하였다. 본 논문에서는 HILS 구축에 대한 상세한 내용과 이를 활용하여 Control 시스템을 검증한 결과를 서술하였다.

• Korean Chemical Engineering Research
• /
• v.56 no.6
• /
• pp.779-783
• /
• 2018

In this work, Mathematical modeling was carried-out to predict the charging/discharging characteristics of VRLA (Valve regulated lead acid) battery, which is mainly used as a 12 V lead acid battery for automobile. And It also carried-out how it's characteristics would be changed due to aging. A mathematical modeling technique, which has been mainly used to predict behavior of Lithium-ion batteries, is applied to commercial 70 Ah VRLA battery. The modeling result of Voltage was compared with result of constant current charge / discharge test. From this, it can be seen that the NTGK model can be applied to the lead acid battery with high accuracy. It was also found that the aging of lead-acid battery can be predicted by using it.

• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.280-283
• /
• 2013

최근 원자력 발전소, 의료 시스템, 항공 시스템 등과 같이 사람의 생명과 밀접하게 관련되어 있는 소프트웨어로 제어하는 시스템들이 점차 늘어나고 있다. 차량에서 또한 차량 제어 소프트웨어의 오작동으로 인한 잦은 사고로 인하여 운전자와 탑승자의 생명을 위협 받고 있다. 이러한 문제로 인하여 차량시스템 제어 소프트웨어도 안전성 확보를 위한 기술로 차량에 통신 기술을 접목한 차량 통신 기술에 대한 관심이 높아지고 있다. 차량 운전자 뿐 아니라 탑승자의 안전과 밀접하기 때문에 많은 연구가 진행되고 있다. 이러한 많은 연구 중 긴급메시지전송 시스템은 차량 간 통신(V2V)을 통한 운전자의 안전성 확보에 대한 연구다. 본 논문에서는 차량 긴급메시지 전송에 필요한 모듈을 구조적으로 나누고 이를 통하여 긴급메시지 전송시스템 구조의 안전성을 평가한다. 긴급메시지 전송시스템의 안정성을 검증하기 위하여 오토마타 모델링을 통한 시스템 구조를 설계하고 검증을 위해 CTL 논리식 정의, SMV(Symbolic Model Verifier)검증도구를 통한 시스템 안전성 모델 검사를 하였다.

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.80-90
• /
• 2021

Using optical music recognition technology based on deep learning, we propose to apply the results derived to VR games. To detect the music objects in the music sheet, the deep learning model used YOLO v5, and Hough transform was employed to detect undetected objects, modifying the size of the staff. It analyzes and uses BPM, maximum number of combos, and musical notes in VR games using output result files, and prevents the backlog of notes through Object Pooling technology for resource management. In this paper, VR games can be produced with music elements derived from optical music recognition technology to expand the utilization of optical music recognition along with providing VR contents.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.555-560
• /
• 2021

Recently, as people's interest in facial skin beauty has increased, research on skin disease recognition for facial skin beauty is being conducted by using deep learning. These studies recognized a variety of skin diseases, including acne. Existing studies can recognize only the single skin diseases, but skin diseases that occur on the face can enact in a more diverse and complex manner. Therefore, in this paper, complex skin diseases such as acne, blackheads, freckles, age spots, normal skin, and whiteheads are identified using the Inception-ResNet V2 deep learning mode with multi-label classification. The accuracy was 98.8%, hamming loss was 0.003, and precision, recall, F1-Score achieved 96.6% or more for each single class.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.115-124
• /
• 2022

In this paper, YOLO and EAST models are tested to analyze their performance in text area detecting for real-world and normal text images. The earl ier YOLO models which include YOLOv3 have been known to underperform in detecting text areas for given images, but the recently released YOLOv4 and YOLOv5 achieved promising performances to detect text area included in various images. Experimental results show that both of YOLO v4 and v5 models are expected to be widely used for text detection in the filed of scene text recognition in the future.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.538-540
• /
• 2021

In this paper, we proposed a smart target detection system that detects and recognizes a designated target to provide relative motion information when performing a target detection mission of a drone. The proposed system focused on developing an algorithm that can secure adequate accuracy (i.e. mAP, IoU) and high real-time at the same time. The proposed system showed an accuracy of close to 1.0 after 100k learning of the Google Inception V2 deep learning model, and the inference speed was about 60-80[Hz] when using a high-performance laptop based on the real-time performance Nvidia GTX 2070 Max-Q. The proposed smart target detection system will be operated like a drone and will be helpful in successfully performing surveillance and reconnaissance missions by automatically recognizing the target using computer image processing and following the target.

• Journal of Korean Society of Environmental Engineers
• /
• v.31 no.9
• /
• pp.825-830
• /
• 2009

Pharmaceutical compounds enter the water environment through the diverse pathways. Because their concentration in the water environment was frequently detected in the level of ppt to ppb, the monitoring system should be optimized as much as possible for finding appropriate management policies and technical solutions. One Factor At a Time (OFAT) approach approximating the response with a single variable has been preferred for the optimization of LC-MS/MS operational conditions. However, it is common that variables in analytical instruments are interdependent. Therefore, the best condition could be found by using the statistical optimization method changing multiple variables at a time. In this research, response surface analysis (RSA) was applied to the LC-MS/MS analysis of emerging antibiotic compound, sulfamethoxazole, for the best sensitivity. In the screening test, fragmentation energy and collision voltage were selected as independent variables. They were changed simultaneously for the statistical optimization and a polynomial equation was fit to the data set. The correlation coefficient, $R^2$ valuerepresented 0.9947 and the error between the predicted and observed value showed only 3.41% at the random condition, fragmentation energy of 60 and collision voltage of 17 eV. Therefore, it was concluded that the model derived by RSA successfully predict the response. The optimal conditions identified by the model were fragmentation energy of 116.6 and collision voltage of 10.9 eV. This RSA can be extensively utilized for optimizing conditions of solid-phase extraction and liquid chromatography.

• Journal of the Korean Society of Radiology
• /
• v.16 no.5
• /
• pp.595-602
• /
• 2022

This study tried to suggest the most efficient learning rate for accurate and efficient automatic diagnosis of medical images for chest X-ray pneumonia images using deep learning. After setting the learning rates to 0.1, 0.01, 0.001, and 0.0001 in the Inception V3 deep learning model, respectively, deep learning modeling was performed three times. And the average accuracy and loss function value of verification modeling, and the metric of test modeling were set as performance evaluation indicators, and the performance was compared and evaluated with the average value of three times of the results obtained as a result of performing deep learning modeling. As a result of performance evaluation for deep learning verification modeling performance evaluation and test modeling metric, modeling with a learning rate of 0.001 showed the highest accuracy and excellent performance. For this reason, in this paper, it is recommended to apply a learning rate of 0.001 when classifying the presence or absence of pneumonia on chest X-ray images using a deep learning model. In addition, it was judged that when deep learning modeling through the application of the learning rate presented in this paper could play an auxiliary role in the classification of the presence or absence of pneumonia on chest X-ray images. In the future, if the study of classification for diagnosis and classification of pneumonia using deep learning continues, the contents of this thesis research can be used as basic data, and furthermore, it is expected that it will be helpful in selecting an efficient learning rate in classifying medical images using artificial intelligence.

• Clean Technology
• /
• v.28 no.3
• /
• pp.238-248
• /
• 2022

The world is striving to transition to a carbon-neutral society. It is expected that using hydrogen instead of hydrocarbon fuel will contribute to this carbon neutrality. However, there is a need for combustion technology that controls the increased NOx emissions caused by hydrogen co-firing. Flameless combustion is one of the alternative technologies that resolves this problem. In this study, a numerical analysis was performed using the 1D opposed-flow diffusion flame model of Chemkin to analyze the characteristics of flameless combustion and the chemical reaction of methane-hydrogen fuel according to its hydrogen content and flue gas recirculation rate. In methane combustion, as the recirculation rate (K_v) increased, the temperature and heat release rate decreased due to an increase in inert gases. Also, increasing K_v from 2 to 3 achieved flameless combustion in which there was no endothermic region of heat release and the region of maximum heat release rate merged into one. In H₂ 100% at K_v 3, flameless combustion was achieved in terms of heat release, but it was difficult to determine whether flameless combustion was achieved in terms of flame structure. However, since the NOx formation of hydrogen flameless combustion was predicted to be similar to that of methane flameless combustion, complex considerations of flame structure, heat release, and NOx formation are needed to define hydrogen flameless combustion.