http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
하승재,고승현,조민성,전영표 한국공업화학회 2019 한국공업화학회 연구논문 초록집 Vol.2019 No.1
An impregnation process is important to enhance the properties of synthetic graphite. Because the graphite block or electrode typically have channels as well as pores due to the evaporation of low molecular weight components during the heat treatment. These channels and pores degrade the properties of the synthetic graphite such as physical strength, electrical property, and so on, and this is why the impregnation is necessary for high quality synthetic graphite. Process conditions of temperature, pressure and time for the impregnation process are investigated. The temperature and pressure ranges were 220 - 260 °C at 20 °C increments and 1 - 40 bar at 20 bar increments, respectively. The time was experimented with 1, 2 and 4 hours respectively. The experimental evidence show that the properties of synthetic graphite can significantly enhanced with the help of impregnation process based on the petroleum impregnation pitches we synthesized.
DEA를 이용한 육용종계 농가의 효율성 영향 요인에 관한 연구
하승재,한용희 한국자료분석학회 2019 Journal of the Korean Data Analysis Society Vol.21 No.6
In this paper, productivity influencing factors have been examined for the broiler breeder farms and the efficient productivity direction was suggested by data envelopment analysis (DEA). Breeding score was based on 1 year records (July 2017 to July 2018) of 43 broiler breeder farms. Efficiency analysis input factors are area (pyeong), score of facility, and the number of hens. Output factor are hatching eggs per 65 weeks, hatchability. Output based CCR model, BCC model and super BCC model were adopted in order to figure out the efficient productivity direction. The results of this study were as following. The number of efficiency DMU (decision making unit) from CCR model is 10 and 13 DMU from BCC model. Based on BCC model, pure technology efficiency and scale efficiency comparison shows that technology and scale of farms are the major factors of inefficiency. The technology caused the 30 inefficient DMU and scale of farm caused 33. It means that inefficiency of productivity did not differ much between technology and the scale of farm. If the major inefficient factor was scale, it is important to expand or minimize the size of farm appropriately in order to improve the efficiency. If the major inefficient factor was technology, operation system and maintenance system should be improved. 본 연구는 국내 A사와 거래 중인 43개 위탁 육용종계 농가의 2017년 7월에서 2018년 7월까지의 사육성적 자료를 DEA(data envelopment analysis) 기법을 이용하여 분석하여 육용 종계농가의 생산 효율성을 분석하고 생산성 개선 방향을 제시하였다. 육용종계 농가의 생산성을 높이는 것을 목적으로 하기 때문에 산출기준 CCR모형 및 BCC모형을 적용하였으며 투입 변수로는 면적(평), 시설점수, 편입수를, 산출 변수로는 보정 종란수와 발생율을 사용한 결과, CCR 모형에서는 10개, BCC 모형에서는 13개 농가가 효율적인 DMU(decision making unit)로 나타났다. 또한 BCC 모형을 이용하여 순수기술효율성(PTE)과 규모효율성(SE)을 비교하여 비효율의 원인을 기술 또는 규모의 요인으로 구분한 결과, 규모가 비효율의 원인인 농가는 33개인 반면 기술적 요인이 비효율의 원인인 농가는 30개로 나타났다. 마지막으로 비효율적인 농가들의 효율성 개선 방안으로 비효율의 주 요인이 기술적인 요인인 경우 운영체계 및 관리기술의 향상을 통해, 비효율의 주 요인이 규모인 경우 규모의 적절한 변경을 통해 효율적인 농가로 개선할 수 있음을 확인하였다.
반도체 장비 센서데이터와 딥러닝을 활용한 불량예측모델에 관한 연구
하승재,김도연,구교연,신용태 한국IT정책경영학회 2021 한국IT정책경영학회 논문지 Vol.13 No.2
This study proposes a model that predicts product quality based on AI by using deep learning from sensor data generated in facilities during the semiconductor manufacturing process. In semiconductor factories, there is a system for predicting defects called FDC (Fault Detection and Classification), but there is a limitation in setting and managing sensor standards for process managers due to the large number of alarm frequencies and high complexity automation systems. It is proposed to automatically provide sensor reference information by learning sensor data using deep learning of sensor data. The judgment result using the facility sensor data for one year for the same recipe of the same facility in one process is Accuracy, Precision, Recall, and F1-score were provided, and it was confirmed that the performance is superior to the existing FDC system.
반도체 설비 센서 데이터를 활용한 딥러닝 기반의 불량예측 모델에 관한 연구
하승재 ( Seung-jae Ha ),이원석 ( Won-suk Lee ),구교연 ( Kyo-yeon Gu ),신용태 ( Yong-tae Shin ) 한국정보처리학회 2021 한국정보처리학회 학술대회논문집 Vol.28 No.1
본 연구는 반도체 제조 공정중 발생하는 센서 데이터를 활용하여 딥러닝기반으로 불량을 예측하는 모델을 제안한다. 반도체 공장에서는 FDC((Fault Detection and Classification)라는 불량을 예측하는 시스템이 있지만, 공정의 복잡도가 높고 센서의 종류가 많아 공정 관리자가 모든 센서의 기준을 설정 및 관리하는데 한계가 있다. 이를 해결하기 위해 공정 설비의 센서 데이터를 딥러닝을 활용하여 학습시켜 센서 기준정보로 임계치를 제공하고, 가공중 발생하는 센서 데이터가 입력되면 정상 여부를 판정하는 모델을 제안한다.
국제표준(ISO/ICE 18013-5) 기반의 모바일 운전면허증 연구
김의정,하승재,이재은,이남용 한국IT정책경영학회 2021 한국IT정책경영학회 논문지 Vol.13 No.2
In response to the development of mobile devices owned by individuals, ISO/IEC published a draft for the 'Mobile Driving License (mDL) application as of 2020. This draft describes the function of a mobile driver's license, but does not include the technology for specific implementation. The purpose of this paper is to study the system framework for converting the driver's license system operated by the National Police Agency into mDL based on international standards. As a result of research through the literature analysis method, the self-sovereign identity model is applied to the personal authentication model, which can be implemented with DID (Decentralized Identity) technology. The DID trust registry is more reliable than a centralized or distributed DB in a blockchain that can technically block data forgery and alteration. Due to the nature of operating in a closed network, the optimal method of configuring the blockchain by distributing write nodes to the issuing agency and configuring the read nodes to be used by other public institutions and private institutions is optimal.
Upcycling waste PET plastic to graphite
고승현,하승재,조민성,강다해,길현식,전영표 한국공업화학회 2019 한국공업화학회 연구논문 초록집 Vol.2019 No.1
Development of new technologies for converting waste plastics into value-added products is attracting widespread attention because of the global plastic waste crisis. Herein, we propose a route for converting waste plyethylene terephthalate (PET) to graphite that is one of the valuable carbon materials. PET was converted to graphite via a synthetic method of pyrolysis at 900 °C followed by catalytic graphitization at 2400 °C. This technique overcame the intrinsic non-graphitizable property of PET and yielded graphite showing high crystallinity with the maximum crystallite size of 20.9 nm in L<sub>c</sub> and the d<sub>(002)</sub> spacing of 3.373 Å.
김현경,하승재,이재은,한용희 한국IT정책경영학회 2021 한국IT정책경영학회 논문지 Vol.13 No.2
Influence Factors on Purchasing Intention of Influencer Shopping MallThe purpose of this study is to analyze the impact of the characteristics of influencer and its shopping malls on purchases and to identify the impact of influencer as a marketing tool. A survey of 262 copies of the questionnaire was used for analysis by conducting a Google survey of customers who purchased the influencer shopping mall. LISREL 8.7 and SPSS were used as statistical tools for hypothesis verification. As a result of empirical analysis for hypothesis verification, The fact that authenticity of the shopping mall's characteristics is a positive influence factor for both satisfaction and trust, that price and convenience of the shopping mall's characteristics are positive impacts on satisfaction, but that price and convenience are rejected in the relationship with trust, and that the characteristics of other influencer shopping malls, such as expertise, attractiveness, interactivity, and reliability, are all of the factors affecting satisfaction and trust, can be seen as a good indication of the customer's perspective on the influencer shopping mall.