http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Wavelet-like convolutional neural network structure for time-series data classification
Seungtae Park,Haedong Jeong,Hyungcheol Min,Hojin Lee,Seungchul Lee 국제구조공학회 2018 Smart Structures and Systems, An International Jou Vol.22 No.2
Time-series data often contain one of the most valuable pieces of information in many fields including manufacturing. Because time-series data are relatively cheap to acquire, they (e.g., vibration signals) have become a crucial part of big data even in manufacturing shop floors. Recently, deep-learning models have shown state-of-art performance for analyzing big data because of their sophisticated structures and considerable computational power. Traditional models for a machinery-monitoring system have highly relied on features selected by human experts. In addition, the representational power of such models fails as the data distribution becomes complicated. On the other hand, deep-learning models automatically select highly abstracted features during the optimization process, and their representational power is better than that of traditional neural network models. However, the applicability of deep-learning models to the field of prognostics and health management (PHM) has not been well investigated yet. This study integrates the “residual fitting” mechanism inherently embedded in the wavelet transform into the convolutional neural network deep-learning structure. As a result, the architecture combines a signal smoother and classification procedures into a single model. Validation results from rotor vibration data demonstrate that our model outperforms all other off-the-shelf feature-based models.
Lee, Seungtae,Kim, Seona,Choi, Sihyuk,Shin, Jeeyoung,Kim, Guntae The Electrochemical Society 2019 Journal of the Electrochemical Society Vol.166 No.12
<P>Mixed ionic and electronic conductors (MIECs) show greatly enhanced electrochemical properties as cathode materials for intermediate-temperature solid oxide fuel cells (IT-SOFCs). Among promising MIEC candidates, La<SUB>0.5</SUB>Ba<SUB>0.25</SUB>Sr<SUB>0.25</SUB>Co<SUB>0.8</SUB>Fe<SUB>0.2</SUB>O<SUB>3-</SUB><I><SUB>δ</SUB></I> (LBSCF) is chosen as a cathode material in this study due to its high oxygen ion diffusivity and oxygen reduction reaction (ORR) activity. It is well known that the composite configuration with superior oxygen ion conductor, e.g., La<SUB>0.9</SUB>Sr<SUB>0.1</SUB>Ga<SUB>0.8</SUB>Mg<SUB>0.2</SUB>O<SUB>3-</SUB><I><SUB>δ</SUB></I> (LSGM), results in the improvement of cell performance. Here, we successfully fabricate the LBSCF-LSGM composite cathode by infiltration method at a low sintering temperature (< 800°C), and thus nanoparticles are uniformly distributed on the LSGM porous scaffold. A half-cell exhibits an extremely low area-specific resistances (ASRs) of 0.010, 0.020, and 0.049 Ω cm<SUP>2</SUP> at 700, 650, and 600°C, respectively. Further, the peak power density about 0.9 W cm<SUP>−2</SUP> is attained at 700°C, suggesting that infiltrated LBSCF-LSGM composite demonstrates sufficient ORR activities at an intermediate temperature and thus to be a highly promising cathode material for IT-SOFCs.</P>
Biodiversity of Arthropods between GM and non-GM Rice Field
SeungTae KIM,SueYeon LEE,HunSung KIM,ChangGyu PARK,SeJin KIM,MyungPyo JUNG,JoonHo-LEE 한국응용곤충학회 2008 한국응용곤충학회 학술대회논문집 Vol.2008 No.05
Biodiversity of arthropods in Cnaphalocrocis medinalis (Lepidoptera: Crambidae) resistant GM rice (CryIAc1) and non-GM rice fields. Sampling was conducted 15 times using sweeping net and electric aspirator. Biodiversity was analyzed with species richness and Shannon diversity index (H’). Total 28,275 arthropod individuals (12,413 in GM rice and 15,862 individuals in non-GM rice) were collected and there were 22 families, 34 genera and 36 species belonging to 8 orders. There were 19 families, 29 genera, 31 species in GM rice and 20 families 32 genera 34 species in non-GM rice fields. There was no significant difference in species richness and species diversity (H’) between GM and non-GM rice. Species diversity was significantly higher in non GM rice in the insect pest group only in middle and late of August .
기업집단 형태별 의결권 및 배당권의 괴리가 기업성과에 미치는 영향
이승태(Seungtae Lee) 한국국제회계학회 2010 국제회계연구 Vol.0 No.32
대규모기업집단은 사업내용을 지배하는 주체에 따라 그룹총수 중심의 기업집단과 주력회사 중심의 기업집단으로 구분된다. 기업집단의 형태에 따라 의결권과 배당권의 괴리현상이 기업성과에 미치는 영향은 상이할 것으로 예상된다. 그룹총수 중심의 기업 집단은 주력회사 중심의 기업집단에 비해 의결권과 배당권간의 괴리가 커 지배주주가 소수주주의 권익을 침해하여 기업성과는 하락할 것으로 예상되며 주력회사 중심의 기업집단은 의결권과 배당권간의 괴리가 상대적으로 작아 경영자는 독립적인 지위에서 기업을 경영함으로써 기업성과는 증가할 것으로 예상된다. 기업집단 형태에 따라 의결권과 배당권의 괴리가 기업성과에 미치는 영향이 상이한지를 검증하기 위해 2001년부터 2008년까지 대규모기업집단으로 분류된 기업을 그룹총수 중심의 기업집단과 주력 회사 중심의 기업집단으로 구분하여 각 집단별 기업성과에 대한 차별성을 검증한다. 아울러 2008년 금융위기가 기업성과에 차별적 영향을 미치는 지를 추가로 분석한다. 분석결과는 다음과 같다. 그룹총수 중심의 기업집단과 주력회사 중심의 기업집단 간 변수들의 평균의 차이가 있는가를 검증한 결과 기업집단 형태별 의결권과 배당권의 괴리(WEDGE)는 유의적으로 평균의 차이를 보였다. 이를 바탕으로 기업집단 형태별 기업성과를 회귀분석한 결과 그룹총수 중심의 기업집단은 의결권과 배당권의 괴리가 증가할수록 시가 대비 장부가치 비율로 측정한 기업성과와 유의적인 음(-)의 관계를 보였다. The large business group according to the subject which governs a business group is divided in the business group based heads of conglomerates(HC) and business group based flagship company(FC). Business performance of business group differ from wedge. The purpose of this paper is to study the distinctive properties of performance that can be changed by the corporate governance base head of group. Using the ownership structure of large business groups, I investigate the effect of a disparity between dividend right and voting right of large business group member firms of their performance. For this study, one research hypotheses is formulated based upon the prior studies; Larger wedge has smaller effect on performance in HC than FC. The primary findings are summarized in the following way. Using a sample of Large Business Group listed on the Korea Stock Exchange over the 2001-2008 period, the test results indicates that the business group HC exists a significant negative relation between the wedge and performance.
Wavelet-like convolutional neural network structure for time-series data classification
Park, Seungtae,Jeong, Haedong,Min, Hyungcheol,Lee, Hojin,Lee, Seungchul Techno-Press 2018 Smart Structures and Systems, An International Jou Vol.22 No.2
Time-series data often contain one of the most valuable pieces of information in many fields including manufacturing. Because time-series data are relatively cheap to acquire, they (e.g., vibration signals) have become a crucial part of big data even in manufacturing shop floors. Recently, deep-learning models have shown state-of-art performance for analyzing big data because of their sophisticated structures and considerable computational power. Traditional models for a machinery-monitoring system have highly relied on features selected by human experts. In addition, the representational power of such models fails as the data distribution becomes complicated. On the other hand, deep-learning models automatically select highly abstracted features during the optimization process, and their representational power is better than that of traditional neural network models. However, the applicability of deep-learning models to the field of prognostics and health management (PHM) has not been well investigated yet. This study integrates the "residual fitting" mechanism inherently embedded in the wavelet transform into the convolutional neural network deep-learning structure. As a result, the architecture combines a signal smoother and classification procedures into a single model. Validation results from rotor vibration data demonstrate that our model outperforms all other off-the-shelf feature-based models.
해수침식을 받은 모르타르의 저항성에 대한 결합재의 영향
이승태 ( Lee Seungtae ),김성수 ( Kim Seongsoo ),정호섭 ( Jung Hoseop ),김종필 ( Kim Jongpil ),박광필 ( Park Kwangpil ) 한국구조물진단유지관리공학회 2009 한국구조물진단유지관리공학회 학술발표대회 논문집 Vol.13 No.2
This paper reports an experimental finding related to seawater attack resistance of mortars made with 3 types of Portland cements. In ordert o evaluate the resistance, compressive strength and expansion of mortars exposed to seawater were regularly monitored. As a results, it was found that the resistance to seawater attack of mortars were greatly dependent on the cement components.