http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
양성자 조사 기술을 이용한 NPT형 전력용 다이오드의 스위칭 속도 개선
김병길,배영호 위덕대학교 산업기술연구소 2005 산업기술연구소 논문집 Vol.9 No.1
양성자 조사 기술을 이용하여 전력용 반도체 소자의 스위칭 특성 개선 기술을 개발하였다. 양성자 주입 에너지를 변화시키며 시편을 제조하여 주입 조건에 따른 결함 형성 기구와 소자의 동작 특성을 분석하였다. NPT형 전력용 다이오드 소자에 주입 에너지 2.32 MeV, 2.55 MeV, 2.97 MeV의 조건으로 양성자를 주입시켜 결정 결함의 생성 위치를 변화시켰으며 각 조건으로 제조된 소자의 전류-전압 특성, 용량-전압 특성, 항복전압 특성, 역방향 회복시간을 측정하여 양성자를 주입하지 않은 소자와 비교 분석하였다. 주입에너지 2.97 MeV로 양성자가 주입된 소자의 경우, 순방향 전류 5A에서 전압강하는 최초 소자의 120 %인 1.1 V로 증가하였고, 역방향 전압 100 V에서 누설전류는 64 nA, 역방향 항복전압은 670 V로 최초 소자와 같은 값을 나타냈었다. 역방향 회복 시간은 양성자를 주입하지 않은 소자의 20 % 수준인 50 nA로 감소하였다. Proton irradiation technology has been used for improving the switching characteristics of a pn diode. The NPT type power diodes which were irradiated respectively with 2.32 MeV, 2.55 MeV, 2.97 MeV energy condition were analyzed by current-voltage, capacitance-voltage, and reverse recovery time measurements and compared with non irradiation device. In the case of 2.97 MeV irradiated device, forward voltage drop was increased to 1.1 V witch was 120% of its original device at forward current of 5 A, reverse leakage current was 64 nA at reverse voltage of 100 V, and reverse breakdown voltage was 670 V which was the same voltage as original device without irradiation. The reverse recovery time of device was 50 nA, reduced to about 20% compared to that of original device without irradiation.
안구적출에 따른 위둔덕의 칼슘결합단백질의 재구축 및 상호 연관성
안병수,고길석,안명수,김경주,권안성,정명섭,박춘매,조병옥,김진우,Samudra Acharya,Parmeshwar Narayan Amatya,장인엽 朝鮮大學校 附設 醫學硏究所 2007 The Medical Journal of Chosun University Vol.32 No.1
Background: Superior colliculus is a part of midbrain, and participates in the visual reflexes, It receives afferent fibers from optic nerve, visual cortex, and spinotectal tract. After optic deprivation, the microscopic structure of the superior colliculus changed. Calcium-binding proteins (CBPs) Play an important role in the neuronal protection, differentiation and reorganization of the central nervous system, Objectives and Methods: The effects of neonatal retinal deafferentation on a CBPs, calbindm D-28k (CB), Parvalbumin (PB) and calretimn (CR), and the existence of colocalization between the CBPs were examined immunohistochemically in the rat superior colliculus. Results: On the experimental (contralateral to enucleation) side of superior colliculus, the number of CB-immunoreactive (IR) cells was reduced (77.4% compared to control), but not fibers. The number of PB-IR neurons and fibers was also reduced on the experimental side (88.5% compared to control), In the other hand, the CR-IR cells were dramatically increased (642% compared to control), but CR-IR fibers were markedly decreased on the experimental side. The colocalization between CB-CR and PV-CR was rarely observed in the superior colliculus Conclusion: These results suggest that the changes of retinotectal projection may alter the expressional pattern of CBPs in different manners; relatively stable in CB- and PV-IR neurons and plastic in CR-IR neurons.
이경훈,문병석,남길정,강일환 대한상하수도학회 2002 상하수도학회지 Vol.16 No.2
To develop the forecasting model on hydrologic rainfall-runoff process, this fuzzy theory is applied in this study, that is the mathematical theory to process ambiguity and the neural network theory, that is applied human's empirical thinking to mathematical process. As the result of applying neural network theory, the model of linear-linear, linear-nonlinear in the transfer function and the model applying fuzzy theory showed the excellent expectation that a coefficient of determination is over 0.97, percent error of maximum flow is less 10%. This result showed that neural network and fuzzy theory have excellent applicability in the flood forecasting. And this study expect that neural network and fuzzy theory can be used as the approach to settle problems being occurred in the course of rainfall-runoff modeling.
이경훈,문병석,김강석,남길정 대한상하수도학회 2003 상하수도학회지 Vol.17 No.5
This study. which chose Sum-Jin river dam and Ju-Amdam of Sum-Jin river water system for objection region, forecasted inflow of dam to manage a multipurpose dam during flood period, using neural network model, executed HEC-5 to be inputted the foresaw and measured inflow data and investigated the result data. HEC-5 is dam imitative operating program. When no damaging channel capacity was 100% or 70%, this study investigated the flood control storage probability so that we understand concerning with no damaging channel capacity and the flood control storage capacity of reservoir. As a result that forecasted dam inflow through neural network model, we got excellent forecasting outcome that the coefficient of determination is over 0.9. and find to be able to apply HEC-5 because. as a result that £ was applied to HCE-5 forecasting outcome and surveyed inflow data. the forecasting outcome were more superior managing result than real outflow one at the phases of peak flow control rate, reservoir outflow rate, and reswyoir storage rate. finally this study investigated abut application of HEC5 model at the best suited dam management and regulation of no damaging channel capacity. as deciding the best suited outflow by adjusting no damaging channel capacity.
클러스터형 VOD서버에서 고가용성을 고려한 자체 복구 시스템
이좌형,서동만,방철석,김병길,정인범 강원대학교 정보통신연구소 2004 정보통신논문지 Vol.8 No.-
While clustered VOD (Video On Demand) servers that have been proposed for building large-scale VOD system recently, have the advantages of high availability and scalability, the system level reliability of the clustered VOD servers could be decreased by adding more servers to scale up the capacity of the server. To alleviate this reliability problem and support server-level fault tolerance, this paper proposes and analyzes self recovery system based on RAID-3 and 4 with pipelining concept. The proposed self recovery system could increase the efficiency of the disk by accessing the disk with large size block as in RAID-4. The network resource could be used efficiently and the workload of memory could be decreased by sending data in small size block as in RAID-3. By introducing pipeline concept, the recovery tasks are processed at several nodes in parallel so that the workload in CPU, network, and memory could be distributed to several nodes.