폐 산세 용액으로부터 공침 반응에 의한 지르코늄 회수 시 BaF₂ 입도 영향 및 Ba₂ZrF₈의 진공증류 특성

최정훈,하이크 네르시시안,한슬기,김영민,박철호,강종원,나기현,김정훈,이종현,Choi, Jeong Hun,Nersisyan, Hayk,Han, Seul Ki,Kim, Young Min,Park, Cheol-Ho,Kahng, Jong Won,Na, Ki Hyun,Kim, Jeong hun,Lee, Jong Hyeon 한국자원리싸이클링학회 2017 資源 리싸이클링 Vol.26 No.6

핵연료 피복관은 필거링 및 열처리 공정에 의해 제조되며, 핵연료 피복관 표면의 불순물 및 산화층을 제거하기 위해 산세 공정이 진행된다. 산세 공정 중 산세 용액으로 지르코늄이 용해되어 포화상태가 될 시 폐 산세 용액은 전량 폐기 되며 용액 내 지르코늄 또한 폐기된다. 그러므로 지르코늄을 재활용하기 위해 $BaF_2$를 폐 산세 용액에 첨가하여 공침 반응에 의해 $Ba_2ZrF_8$을 형성시켰다. 한편, 전해제련을 통해 Zr을 회수하기 위해서는 $Ba_2ZrF_8$을 전해질로 사용해야하지만, 용융점이 $1053^{\circ}C$로 높다. 따라서 $ZrF_4$를 첨가하여 용융점을 낮추어야한다. 본 연구에서는 $Ba_2ZrF_8$을 진공 증류를 통해 $BaF_2$ 및 $ZrF_4$로 분리하는 연구를 진행하였다. 먼저, $BaF_2$ 입자크기($1{\mu}m$, $35{\mu}m$, $110{\mu}m$)에 따른 침전 특성을 실험하였다. 그리고 진공 증류를 통해 수득된 $BaF_2$를 볼밀링을 통해 분쇄하였으며, 침전 공정을 거치지 않은 $BaF_2$와 침전 효율을 비교하였다. Nuclear fuel cladding tube is fabricated by pilgering and annealing process. In order to remove impurity and oxygen layer on the surface, pickling process is carried out. When Zirconium(Zr) is dissolved and saturated in acid solution during the pickling process, all the waste acid including Zr is disposed. Therefore, $BaF_2$ is added into the waste acid to extract Zr and $Ba_2ZrF_8$ is subsequently formed. To recycle Zr by electrowinning process, $Ba_2ZrF_8$ is used as electrolyte, but it has high melting point ($1053^{\circ}C$). $ZrF_4$ should be added into $Ba_2ZrF_8$ to decrease the melting point. In this paper, it was investigated that $Ba_2ZrF_8$ was separated to $BaF_2$ and $ZrF_4$ by vacuum distillation. Firstly, $BaF_2$ with different particle size ($1{\mu}m$, $35{\mu}m$, $110{\mu}m$) was added into the waste acid and the respective precipitation property was estimated. $BaF_2$ obtained by vacuum distillation was shattered by ball-milling with different time. The precipitation efficiency was compared with $1{\mu}m$ of ${BaF_2}^{\prime}s$ one, which was not used as precipitation agent.

흡수율과 재생율을 동시 고려한 천연가스복합발전 공정 연계 이산화탄소 포집 공정의 재생 조건 최적화

최정훈,주영환,Jeong Hun Choi,Young-Hwan Chu 한국화학공학회 2023 Korean Chemical Engineering Research(HWAHAK KONGHA Vol.61 No.3

Natural Gas Combined Cycle(NGCC) recently receives lots of attention as an attractive form of power plants by virtue of its low carbon emission compared with coal-fired power plant. Nevertheless, it also needs carbon capture process since it is difficult to completely suppress carbon emission even for the NGCC. A simulation study has been performed to optimize operating condition of a carbon capture process using MEA considering low partial pressure of carbon dioxide in NGCC emission gas. For accurate optimization, overall process model including both NGCC and the carbon capture process has been built with a simulation software. Then, optimization in which various performance indices such as carbon dioxide absorption rate, solvent regeneration rate and power loss in the NGCC are simultaneously reflected has been done. Especially, it is noticeable that this study focuses on not only the amount of energy consumption but also the absorption and regeneration performance of carbon capture process. The best result considering all the performance indices has been achieved when the reboiler temperature is 120 ℃ and the reason has been analyzed.

Cone Beam Computed Tomography를 이용한 상악 임플란트 식립 전후의 골밀도 변화에 관한 연구

최정훈,이주민,김용덕,신상훈,정인교,Choi, Jeong-Hun,Lee, Ju-Min,Kim, Yong-Deok,Shin, Sang-Hun,Chung, In-Kyo 대한악안면성형재건외과학회 2010 Maxillofacial Plastic Reconstructive Surgery Vol.32 No.3

Purpose: This study examined the significance of increased bone density according to time after implantation on maxilla using demographic data with CBCT and compared the bone density between before vs. after implantation using the Hounsfield index. Materials and Methods: Twenty-five implant site on maxilla were selected. Cone-beam computerized tomography (CBCT) scans were used for the analysis. The implant sites were evaluated digitally using the Hounsfield scale with EzImplant TM and the results were compared over time. Statistical data over time was carried out to determine the correlation between the recorded Hounsfield unit (HU) over time and gender difference using repeated ANOVA. Results: The bone density of implantation site over time showed an increase in the HU mean values. Immediately after implantation, bone density was significantly increased than bone density before implantation. Until 6 month follow-up, bone density showed stable increasement. There is no significant difference on gender. Conclusions: Using CBCT, bone density increased over time after implantation on maxilla. Bone density measurements using CBCT might provide an objective assessment of the bone quality as well as the correlation between bone density and stability of implant.

국방 연구개발사업의 "운용자-개발자"간 연계성 향상을 위한 아키텍처 개발 방법에 대한 연구

최정훈,강석중,Choi, Jeong-Hun,Kang, Seok-Joong 한국군사과학기술학회 2010 한국군사과학기술학회지 Vol.13 No.6

In this paper, we have proposed Architecture Development methodology which can connect both operational view and system view. The Functional Architecture can connect both user and developer, and it is located between Activity analysis and System analysis. We suggest the new architecture methodology using the Functional Architecture and it provides effect to analyze the connection between user(military) and developer(enterprise) in Defence R&D and the new Architecture with the feedback analyze activity on a point of system view and the new architecture make the functional architecture.

지능망 서비스 시스팀을 위한 그래픽 사용자 인터페이스 개발 환경

최정훈(Jeong Hun Choi),김성규(Sung Kyu Kim),홍진표(Jin Pyo Hong) 한국정보과학회 1992 한국정보과학회 학술발표논문집 Vol.19 No.1

윈도우 시스팀을 기반으로한 효과적이고 사용하기 쉬운 사용자 인터페이스를 제공하는 응용 소프트웨어가 확산됨에 따라 통신 시스팀 분야에서도 그래픽 사용자 인터페이스를 이용한 시스팀의 연구 및 개발이 활발히 이루어지고 있다. 본 그래픽 사용자 인터페이스 개발 환경은 소프트웨어 재사용의 개념에 입각하여 지능망 서비스 시스팀의 그래픽 사용자 인터페이스 개발시 필요한 주요 기능들을 객체 지향적 라이브러리로 표준화한 그래픽 지원 환경을 제공한다. 이 새로운 개발 환경은 특히 지능망을 비롯한 통신 시스팀의 운영관리 및 감시기능을 그래픽 화면으로 구성하려는 개발자에게 일관성있고 생산성을 향상시킬 수 있는 그래픽 환경을 제공해주며, X 윈도우 시스팀을 이용하는 일반 응용 프로그래머에게 유용한 도구를 이용될 수 있다.

전기 유압식 브레이크 HILS를 통한 회생제동 시스템 특성 해석

최정훈(Jeong Hun. Choi),조배균(Bae-kyoon Cho),박진현(Jinhyun Park),황성호(Sung-Ho Hwang) 유공압건설기계학회 2011 유공압건설기계학회 학술대회논문집 Vol.2011 No.6

Nowadays, various researches about eco-friendly vehicles such as hybrid electric vehicle, fuel cell vehicle and electric have been actively carried out. Since most of these green cars have electric motors, the regenerative energy technology can be used to improve the fuel economy and the energy of vehicles. The regenerative brake is an energy recovery mechanism which slow a vehicle by converting its kinetic energy into electric energy, which can be either used immediately or stored until needed. This technology plays a significant role in achieving the high energy usage. However, there are some technical problems in driving safety and ride comfort of the vehicles. In this paper, the Hardware-in-the-loop Simulation(HILS) of Electro-hydraulic Brake(EHB) is performed to analyse the characteristics of the regenerative braking system for fuel cell electric vehicle. The performance simulator of the fuel cell vehicle is developed and transient response characteristics of the regenerative braking system are analyzed in the various driving situations.

뇌수막종의 종양주위부종에 대한 임상적 고찰

최정훈 ( Jeong Hoon Choi ),김창현 ( Chang Hyun Kim ),문재곤 ( Jae Gon Moon ),이호국 ( Ho Kook Lee ),이민진 ( Min Jin Lee ),최종훈 ( Jong Hun Choi ) 대한뇌종양학회 2007 대한뇌종양학회지 Vol.6 No.1

Objective:The extent of peritumoral brain edema(PTBE) in meningioma is very variable and PTBE can adversely affect the clinical course. Many influenceable factors have been investigated, but the mechanism of that is not fully understood. Recent reports have focused on the arterial blood supply pattern and immunohistopathological aspect. The authors investigated several clinical and immunohistopathological factors that may influence the development of PTBE and postoperative edema patterns in meningioma. Methods:We studied 30 cases of intracranial meningioma which were histologically proven. We focused on identifying the interaction of the location, the tumor and edema volume, the histological subtype, the preoperative edema index(EI), the biological activity indicated by the VEGF and the MIB-1 labeling index(LI), the postoperative EI. Results:There was no statistically significant correlation between the location and the PTBE. There was statistically significant correlation between the histology and the PTBE. But there was no significant correlation between the histology and the decrease in postoperative remaining PTBE. The MIB-1 LI was positively related to the PTBE formation in meningiomas. Also MIB-1 LI was related to the decrease in postoperative remaining PTBE. The VEGF was positively correlated with the PTBE formation in meningiomas but there was no statistically correlation with the decrease in postoperative remaining PTBE. Conclusion:The greater MIB-1 LI results in greater PTBE formation and more decreased in postoperative PTBE. The greater VEGF expression results in greater PTBE formation.

코로나19 유행에 따른 야간 영업 제한 조치가 서울 미세먼지 농도에 미치는 영향

최정훈(Jeong-Hun Choi),곽경환(Kyung-Hwan Kwak) 강원대학교 환경연구소 2021 Journal of the Environment Vol.14 No.1

국방 아키텍처프레임워크(MND-AF)의 활용성에 관한 연구

최정훈(Jeong-Hun Choi),강석중(Seok-Joong Kang) 한국통신학회 2010 韓國通信學會論文誌 Vol.35 No.10

신경망 모델을 이용한 한글 필기체 온라인 인식

최정훈(Jeong Hun Choi),권희용(Hee Yong Kwon),김춘석(Choon Seok Kim),황희융(Hee Yeung Hwang) 한국정보과학회 1990 정보과학회논문지 Vol.17 No.5

본 논문에서는 한글 자소의 특징을 학습한 신경 회로망을 이용해 위치와 크기에 무관한 필기체 한글 온라인 인식 시스템을 구현하였다. 마우스 입력을 획의 모양과 이전 획과의 상대적 위치로 표현한 특징 벡터로 변환해 신경망 입력으로 사용했다. 초성, 중성, 종성을 각 획수별로 분류해 17개 신경망에 분산 학습함으로써 조합 가능한 11, 172자를 문자 단위로 학습할 때보다 학습 패턴 개수가 최소화한다. 또한 필순, 획수, 모양, 시작 위치 등의 다양한 변형을 수용해 필기체 한글 인식의 많은 문제점을 해결했다. 획 단위로 표현된 가변 길이 입력에서 각 자소를 구분할 수 있는 적합한 인식 알고리즘을 제시한다. 신경 회로망은 오류역전파 학습규칙을 갖는 Multilayer Perceptron Network을 이용하였다. KSC 5601-1987 표준안 한글 2350자 중 앞의 564자를 실험한 결과 95% 이상의 인식률을 보였다. In this paper, an on-line recognition system for hand-written Korean characters using Neural Network model was implemented. A new representation scheme was proposed to provide "size and position indenpendency". It converts the mouse input into characteristic vector composed of "stroke type" and "relative start position" Then the neural network uses the characteristic vector as it's input. Because the possible number of Korean characters is too many to learn on character by character basis, it is necessary to learn Korean alphabets separately. Each Korean character is composed of three part, "first consonant", "middle vowel", and "last consonant(optional)", Moreover, input is done stroke by stroke. So, learning is done separately on 17 groups based on the number of strokes in each alphabet. Therefore, variations in "stroke order", "number of strokes", "stroke type", and "start position" are allowed, and appropriate algorithm that recognize each alphabet from variable-length characteristic vector was devised. We used a Multilayer Perceptron Network with the Error Back Propagation learning rule. The recognition rate is about 95% in the experiment using KS C 5601 1987.