리튬-이온 배터리팩의 전압안정화회로 설계

황호석,남종하,최진홍,장대경,박민기 전력전자학회 2004 전력전자학회 논문지 Vol.9 No.2

PDA, 스마트폰, UPS 및 전기자동차와 같은 전기적 장치의 전력공급을 위하여 배터리가 직렬로 연결된 다중 셀을 만들어 일반적으로 사용한다. 이 경우 개별 셀 전압의 편차가 발생되면 배터리의 수명과 용량은 낮아지게 된다. 셀에서 전압의 안정화상태를 유지하기 위한 셀 전압을 안정화시키는 효율적인 방법은 없어서는 안 될 중요한 사항이다. 본 논문에서는 휴대형 가전기기에 적용하기 위해 마이크로컨트롤러를 사용한 다중 셀용 밸런싱 회로의 설계에 대하여 제안한다. 밸런싱 시스템은 충전되는 주기 동안 밸런싱 동작을 이행하며 마이크로컨트롤러로서 제어된다. 제안된 방법은 충전기와 레코드를 사용하여 실험을 통해 증명하였다. 실험결과에서 개별 배터리의 용량, 수명, 성능이 향상됨을 보여준다. For a power source of usual electronic devices such as PDA, smart phone, UPS and electric vehicle, the battery made of serially connected multiple cells is generally used. In this case, if there are some unbalanced among cell voltages, the total lifetime and the total capacity of the battery are limited to a lower value. To maintain a balanced condition in cells, an effective method of regulating the cell voltage in indispensable. In this paper, we propose the design of a balancing circuit for electronic appliances. The balancing system was controlled by a micro-controller which enables to implement the balancing action during charging period. Proposed method has been verified by the experiment using the charger and recorder. The experimental results show that the individual battery equalization can improve battery capacity and battery lifetime and performance through an extended operational time.

Delphi-AHP 방법을 이용한 저탄소수소 생산 기술 가치평가 모델 개발

황호석,김의식,장영신,김정환,김광준 한국수소및신에너지학회 2023 한국수소 및 신에너지학회논문집 Vol.34 No.2

Recently, low hydrogen carbon production technology is drawing interest due to lower production costs. Although the pace of research in this field has been accelerating, there is no well-established criteria for evaluation. The most of current evaluation methods needs information related to technology. However the technology is not enough to provide effective evaluation criteria because the technology is not fully developed. In this study, we propose an integrated Delphi-analytic hierarchy process (AHP) method and low carbon hydrogen production technology evaluation model. Experts opinion is used to provide evaluation criteria for the technology. In this study, integrated Delphi-AHP method are utilized for determining factors and calculating their numerical importance based on experts opinion. Then, sensitivity analysis is performed to verify the robustness of the analysis and scenarios of potential changes. As many as 11 factors are identified by Delphi method. Then, numerical importance of the factors are calculated by AHP. Sensitivity analysis is performed. It shows that intellectual property right (IPR) is always more important than other factors. This study proposes the numerical standard for the low carbon hydrogen production technology evaluation. The proposed model can be used for technology evaluation or commercialization.

엘라스틱 넷 벌점항을 이용한 비선형 지지벡터기계의 변수선택

황호석,최호식,박창이 한국데이터정보과학회 2023 한국데이터정보과학회지 Vol.34 No.1

Support vector machines (SVMs) are learning methods seeking the optimal hyperplane with the maximal margin based on training data. SVMs are widely used in classification problems because they can be applied in various problems and deal with high dimensional data effectively. However, in the presense of many noise variables, the prediction accuracy of a SVM can degrade and the interpretation of the final model is difficult. To remedy these disadvantages, a variable selection method based on the LASSO (least absolute shrinkage and selection operator) penalty can be considered. The elastic net penalty has been proposed in regression to overcome the shortcoming of the LASSO that the LASSO tends to select only one variable among the correlated variables when varaibles are strongly correlated. In this article, we consider a variable selection method in nonlinear classifications adopting the COSSO (component selection and smoothing operator), called the COSSO-SVM. The EL-COSSO-SVM adopting the elastic net instead of the LASSO is introduced. Our method is illustrated to improve the performances both in variable selection and prediction of the COSSO-SVM through simulation studies and real data analyses. 지지벡터기계 (support vector machines)는 주어진 훈련 데이터를 이용하여 마진 (margin)을 최대화 하는 최적 초평면 (hyperplane)을 찾는 학습방법이다. 지지벡터기계는 다양한 문제에 적용이 가능하며 고차원데이터를 효과적으로 다룰수 있기 때문에 분류문제에서 폭넓게 사용된다. 그러나 지지벡터기계는 잡음변수 (noise variable)들이 많은 경우 예측력이 저하될 수 있고 선택된 최종 모형의 해석이 어렵다. 이러한 단점을 보완하고자 LASSO (least absolute shrinkage and selection operator) 벌점에 기반한 변수선택법을 고려할 수 있다. 회귀분석에서 변수들간에 강한 상관관계가 있는 경우에 그 변수들중에서 하나만 선택하는 LASSO 벌점의 문제점을 보완하기 위해 엘라스틱 넷 (elastic net, EL) 벌점이 제안되었다. 본 논문에서는 비선형 분류문제의 변수선택법인 COSSO (component selection and smoothing operator)를 적용한 COSSO-SVM을 고려한다. 비선형 커널의 성분선택을 위해 LASSO 대신 엘라스틱 넷 벌점항을 적용한 EL-COSSO-SVM을 소개한다. 모의실험과 실제데이터에 대한 분석을 통하여 제안한 방법이 COSSO-SVM보다 변수선택과 예측력 측면에서 성능이 향상됨을 보인다.

사과 과수화상병 예찰을 위한 사과꽃 생물계절 단계 분류 연구

황호석(Hwang Ho-seok),김민준(Kim Min-jun),문애경(Aekyung Moon),이영학(Yeung-Hak Lee),정윤주(Yunju Jeong) 한국정보기술학회 2022 Proceedings of KIIT Conference Vol.2022 No.6

대용량 영상 데이터 학습을 위한 반자동 어노테이션 방법 연구

황호석(Hwang Ho Seok),문애경(Moon Ae Kyung) 한국통신학회 2023 한국통신학회 인공지능 학술대회 논문집 Vol.2023 No.9

다중셀에 적용된 전압안정화회로 설계

황호석(H.S.Hwang),남종하(J.H.Nam),장대경(D.K.Jang),최진홍(J.H.Choi),신익수(I.S.Sin) 전력전자학회 2003 전력전자학술대회 논문집 Vol.2003 No.11

For a power source of usual electronic devices such as PDA, smart phone, UPS and electric vehicle, the battery made of serially connected multiple cells is generally used. In this case, if there are some unbalanced among cell voltages, the total lifetime and the total capacity of the battery are limited to a lower value. To maintain a balanced condition in cells, an effective method of regulating the cell voltage in indispensable. In this paper introduced a method for battery balancing system using dissipation in the resistance.

객체 검출에서 클래스 압축 및 분할을 사용한 중세 한글 인식 인식률 향상

하영서,황호석,김민준,이창준,심재창 한국멀티미디어학회 2023 멀티미디어학회논문지 Vol.26 No.6

Artificial Intelligence, Deep Learning, Text Recognition, OCR Text recognition is an important method for converting documents into digital format by allowing the extraction of text from images. While optical character recognition (OCR) technology has been developed extensively over the years, it has lower performance when it comes to recognizing Hangeul (the Korean alphabet) due to its larger number of character types compared to other languages. In this study, we propose a technique for object detection that incorporates class compression and division to improve the accuracy of Hangeul text recognition. We conducted experiments comparing models with and without compression and division, as well as OCR, a well-established text recognition method. The results showed that the model utilizing compression and division achieved the highest accuracy, even with the smallest amount of data.

에지 디바이스를 활용한 실시간 말벌 모니터링 시스템의 설계 및 구현

정윤주,황호석,권영재,이철희 한국멀티미디어학회 2022 멀티미디어학회논문지 Vol.25 No.12

In this paper, an edge-enabled wasps monitoring system based on state of the art deep learning model is proposed to classify wasps species and transmit their information to clients for pest control and ecology research. We utilized and evaluated the lightweight models of YOLOv5, YOLOX, and YOLOv7 series to select the optimal object detection model for wasps. In order to improve the recognition performance of small objects such as wasps, a new tiling method is proposed which can reclassify objects located on boundary area between tiles. Our monitoring system is composed by local GPU-based edge computers, Server, and Web-based reporting system. In our experiment, the YOLOX- nano model showed the highest performance among YOLO-based models above in terms of recall, precision, Fi-score, and mAP@IOUs for our custom test data set. However, even the YOLOX-nano model also requires the tiling process for a better bounding box generation for small objects under the real environment. As a result, the proposed wasps monitoring system showed meaningful efficiency for applying in the field of pest control and ecology research.

에너지 IoT 생태계 조성을 위한 실감형 정보서비스 시스템 개발

안승갑,황호석,김영현,박명혜,An, Seung-Gap,Hwang, Ho-Seok,Kim, Yeong-Hyeon,Park, Myeong-Hye 한국정보처리학회 2018 정보처리학회지 Vol.25 No.2

산업용 이차전지의 열화진단에 관한 연구

장대경,황호석,박민기 서울産業大學校 2005 논문집 Vol.54 No.3

This paper proposes a new degradation diagnosis method of secondary battery in order to improve a battery economy. In the proposed method, a bad cell is detected by the measurement of DC internal resistance and the battery capacity is estimated by the electric discharge test in a short period of time, which can be used to determine the time to replace the battery. Through the computer simulation and actual experimentation, the proposed method's validity is confirmed by comparing and analyzing the performance. As a result, we show that it is possible to estimate the accurate capacity of secondary battery in a short space of time.