기계적 분산을 이용한Epoxy Nanocomposites의 전기적 특성연구

장추문 중부대학교 대학원 2008 국내석사

These days the miniaturized, highly-efficient, and reliable electrical equipment has been being required due to the extra high voltage according to the gradually increasing power demand. Therefore, it is focused on the development of the nanometer-sized composite which has the better mechanical solidity and electrical property than the composite mixed with the micrometer-sized filler. In case of the nanometer-sized composite, it shows us powerful combination with epoxy thanks to the increase of surface area, and it brings about the remarkable improvement of mechanical solidity. Moreover it is reported that it increases dielectric strength and improves every electrical property like the uprising of destructive voltage. This thesis shows that the electrical property occurs in applying the nanometer-sized inorganic materials like SiO2(10nm), Al2O3(15nm), and TiO2(5nm) as fillers to epoxy resins. It is important to disperse the nanometer-sized inorganic fillers evenly inside the sample so as to make such nanometer composite. And therefore we applied defoaming and agitating method introducing the principle of rotation and revolution of a planet to the making of epoxy resin nanometer composite. As a result, TiO2 revealed the lowest current value, and Al2O3 showed the highest in the current-voltage characteristics(TiO2<SiO2<Al2O3). Also, in case of SiO2, the more the content of filler increased, the higher current value it showed. And when we changed the temperature under the same electric field, the current value radically increased at around 130℃. This means that it corresponds to the glass transition temperature(Tg) of epoxy. At this point, Al2O3 revealed its own property according to the more radical temperature change, and that's why Al2O3 has its own heat property. Al2O3 showed the most destructive strength in the dielectric destruction property, and SiO2 showed the lowest value in the micron-sized sample. But considering the fact of the current-voltage property, we can understand that the value of nanometer sized samples results in the similar one each other and that they improve to some extent more than unfilled sample or micron-sized sample. In case of dielectric property, the dielectric constant in epoxy nanometer sized composite was almost similar to unfilled sample at the normal temperature, and at 150℃, the smaller filler became, the larger the dielectric constant. This indicates that if the temperature of the sample is more highly increased than Tg value, dielectric constant ascends by means of the phase-change inside the sample, and that it is supposed to be influenced by the size of filler. We can also realize that dielectric loss moves to the higher frequency band according to the reduction of the sample and the uprising of the temperature. Besides, at the frequency band less than 100Hz, electric conductivity of the sample increased in accordance with the reduction of filler size. Likewise impedance decreased in accordance with the reduction of filler size, too, and it decreased in accordance with the uprising of temperature and frequency. We can also grasp that capacitance is maintained regularly, but impedance increases according to the uprising of temperature(about 130℃) and the reduction of size(10nm).

반도체 후공정 MVP 공정설비 물류자동화

이상진 공주대학교 대학원 2015 국내석사

국 문 초 록 반도체 후공정 MVP 공정설비 물류자동화 이 상 진 공주대학교 대학원 기전공학과 본 연구를 통하여 반도체 후공정의 현실을 직시하여 특히 외관 검사 공정인 MVP(Marking Visual Inspection Packing) 공정에서 개선 주제를 설정하여 심층 분석을 통한 개선 방향을 파악 하고자한다. 더불어 반도체 후공정 외관 검사 공정의 출하 방법이 고객요구 및 공정변화에 따라 능동적 대처 가능한 방법을 도출 실시를 통하여 인력을 효율적으로 운영하여 결과적으론 생산성 및 품질 향상에 기여를 통하여 제조 경쟁력을 확보하고자 한다. 반도체 후공정 외관검사 공정은 검사가 필요한 제품을 Customer Tray에 옮겨 담아 설비에 투입한 후 검사를 통하여 양품과 불량을 구분 양품만을 Customer Tray에 담아 적층 후 상부에 Cover Tray를 담아 진공 포장한다. 최근 Customer Tray에 담던 양품의 제품을 Tape & Reel 형태로 제품을 담아 Tape & Reel 형태로 진공 포장하여 출하 하고 있다. 이 과정에 다수의 인력이 참여하게 됨에 따라 참여 인력의 투입 빈도를 파악하여 가장 빈도가 높은 작업인 Reel 교환을 자동화 Item 및 두 번째로 투입 빈도가 높은 Customer Tray를 투입에 필요한 작업을 자동화 Item으로 선정하였다. 상기의 개선 Item을 물류 자동화를 적용 다의 작업을 무인화 함으로 1인당 1대의 Tape & Reel 설비운영에서 5대에서 10대 까지 운영이 가능하게 되었다. 더불어 광학계를 활용 육안으로 검사하는 공정을 자동화에 적용 Human Error를 제거함으로 제조 비용감소 및 생산성 및 품질 향상에 기여하였다. Abstract Logistics Automation of MVP Processing Facilities for Semiconductor Back-End Processing Lee Sang-jin Kongju National University Graduate School (Mechatronics) The purpose of this study is to accurately understand the reality of semiconductor back-end processing and to identify, through an in-depth analysis, possible improvements of the process of marking, visual inspection, and packing (MVP) in particular. In addition, this study aims to discover methods that will enable the shipping schemes of visual inspection of semiconductor back-end processing to actively respond to customer demand and changes in process. This will ensure competitiveness in manufacturing by way of more efficient human resources management and the resulting improvement in productivity and quality. In visual inspection of back-end processing, the semiconductor products (devices) to be inspected are placed in a customer tray and then put into the inspection system. Good devices are separated from defective ones, piled up in customer trays, and then, vacuum-packed with the cover tray on top. Recently, the final good devices are vacuum-packed in T&R (Tape and Reel) instead of in customer tray before shipping. As a number of workers participate in this process, two most labor-intensive stages are selected as automation items; reel-exchanging stage and customer-tray-placing stage. This automation makes it possible for one operator to handle up to 10 T&R systems at the same time instead of one. Further application of this type of automation in visual inspection leads to the elimination of human errors, and therefore, results in improvements in cost, productivity and quality in manufacturing.

Technical Developments for Hyperpolarized 13C Magnetic Resonance Selective Metabolic Imaging : 초분극화 13C 자기공명 물질대사 영상을 위한 선택적 영상 획득 기술 개발

Yang, Seungwook 연세대학교 일반대학원 2016 국내박사

자기공명영상은 비침습적으로 다양한 대조도를 가지는 인체구조와 질병을 영상화할 수 있는 기법이다. 보다 정량적인 적용분야에서는 핵자기공명분광법을 이용해 질병생리학적 접근을 통한 진단 및 치료계획이 시행되고 있다. 하지만 13C와 같은 동위원소들은 임상적으로 물질대사와 같은 여러 생리학적 기전에 대한 중요한 정보를 제공할 수 있는 가능성이 있음에도 불구하고 전통적으로 그 신호민감도가 매우 낮다는 한계를 가지고 있다. 최근 동적 핵 분극화 기술 개발로 13C 기질의 생체 내 신호대잡음비를 10,000배 이상 증폭시킬 수 있게 되어, 이를 이용한 생체 내 물질대사를 영상화 할 수 있게 되었다. 하지만 이러한‘초분극화’신호는 열평형상태 회복에 의한 빠른 신호감쇄와 같은 새로운 문제들을 가지기 때문에, 이를 극복하기 위한 새로운 기술의 개발이 필요하다. 본 논문에서는 초분극화 13C를 이용한 자기공명 물질대사 영상을 위한 기술 개발과, 이에 필요한 배경지식들에 대해 다루고 있다. 첫 번째로는 초분극화 [1-13C] 피루브산의 보다 정확한 생체 외 물질대사 정량화를 돕기 위한 기술을 논하며, 이는 젖산탈수소효소 및 니코틴아마이드 아데닌 다이뉴클레오타이드 효소를 이용해 생체 물질대사 과정을 생체 외에서 재현되었다. 대사산물의 신호를 증폭시키기 위해 간단한 주파수-선택적 펄스열을 이용해 피루브산과 젖산의 신호를 번갈아 가며 얻는 자기공명분광 교차획득기법 및 이에 적합한 정량화 기법을 소개하였다. 두 번째로는 이전 기법을 보다 발전시켜 초 고자장에서 발생하는 공간적 화학이동현상을 이용해 대사물질들의 선택적인 영상을 획득하는 기법을 제시하였다. 초 고자장환경에서는 서로 다른 공명주파수를 가지는 화합물간의 주파수축에서의 분리가 상대적으로 멀어지게 되며, 이는 영상획득에서 공간적인 분리로 이어지게 되는데, 여기서 본 기법을 적용하게 되면 서로 다른 두 화합물로부터 발생되는 신호를 서로 다른 두 영상 평면에 위치시켜 동시에 영상을 획득하는 것이 가능해진다. 본 연구에서는 이 기법을 초분극화 [1-13C] 피루브산을 이용한 소동물 신장영상 획득에 적용하였고, 결과에 대한 분석을 제시하였다. Magnetic resonance imaging (MRI) allows non-invasive and flexible detection and imaging of anatomies and diseases. For more quantitative applications, nuclear magnetic resonance (NMR) spectroscopy has been used to probe disease physiology in vivo, with the goal of improved diagnosis and treatment planning. However, spectroscopic methods have been traditionally challenged by low sensitivity, especially for signal detection of 13C isotopes. 13C probes has great potential in providing information regarding tissue / disease metabolism, but their use has been hampered both by their low NMR sensitivity and low in vivo concentrations. Recent technological advances utilizing dynamic nuclear polarization (DNP) has enabled dramatic signal increase of 13C substrates with over 10,000-fold enhancement for liquid-state polarization, allowing real-time imaging of the injected 13C probe and their downstream metabolic products. The ‘hyperpolarized’ signal detection, however, is characterized by number of challenges including rapid transient-state signal decay, which necessitates development of advanced data acquisition techniques. In this dissertation, technical developments for metabolic imaging with hyperpolarized 13C substrate is presented, along with necessary background knowledge and instrumentations required to conduct such experiments. First, a technique for quantitative modeling of hyperpolarized [1-13C] pyruvate metabolism is presented, where biological enzymes including lactate dehydrogenase (LDH) and nicotinamide adenine dinucleotide (NADH) are used in vitro to mimic in vivo metabolism. Numerical simulations were performed to estimate signal levels arising from interleaved, spectrally-selective excitation scheme, then actual in vitro experiments were performed by designing tailored RF excitation pulse and incorporating it into the MR spectroscopic pulse sequence. The proposed method has shown that increasing the signal arising from newlycreated downstream metabolites can increase the accuracy of metabolism kinetics quantifications. In the second work, a technique which allows interleaved acquisition of metaboliteselective images in ultra-high environment is presented. With increased main magnetic field strength, the spectral dispersion (i.e. chemical shift) also increases, increasing spatial displacement between different resonances. With the proposed technique, this spatial displacement is exploited by low-bandwidth excitation RF pulse combined with slice-selection gradient reversal technique in resolving signals from two different metabolites at two different slice positions. The presented technique was applied for metabolite-selective in vivo [1-13C] pyruvate and lactate image acquisitions.

인버터 高周波 低減을 위한 새로운 空間벡터 變調技法에 관한 硏究

이상택 漢陽大學校 大學院 2000 국내석사

본 논문에서는 인버터의 出力단에 나타나는 電流의 高調波 성분을 低減하여 전동기의 振動, 消音 및 高調波에 의한 熱 損失을 低減시키기 위한 연구로 인버터 出力電流에 포함된 高調波 성분을 測定하여 檢出된 高調波 電流를 電流분석 한 후 制御된 高調波 補償電流의 주입으로 高調波 電流를 低減하는 방법을 제안하였다. 본 논문에서 제안된 알고리즘은 기존의 空間벡터 變調시의 基準電壓 벡터에 高調波 성분을 고려한 역 위상의 補償벡터를 합성하여 인버터에 基準電壓 벡터를 인가하므로 高調波 電流의 발생이 減少되며, 실시간으로 인버터의 出力電流가 軌還 되어지므로 전동기의 特性이 바뀌더라도 適用 가능하다. 본 논문에서는 이론적 해석과 컴퓨터 시뮬레이션을 수행하였고, 제안한 空間벡터 變調技法에 의하여 인버터의 出力電流의 高調波 電流분이 減少되고 THD가 개선됨을 확인하였다.

4륜 조향 차량의 동역학적 모델링 및 제어에 관한 연구

김상현 漢陽大學校 大學院 1995 국내석사

두대 로봇 매니퓰레이터를 이용한 유연성 협조 제어에 관한 실험적 연구

이성준 漢陽大學校 大學院 1995 국내석사

SVD에 근거한 칼만필터를 이용한 시각추적 시스템의 구현에 관한 연구

김선중 漢陽大學校 大學院 1995 국내석사

유연힌지형 정밀 스테이지의 모델링 및 진동제어

김재익 금오공과대학교 대학원 2009 국내석사

This paper designs the precision positioning stage(PPS) composed of piezoelectric actuators and flexible hinges, and compares the finite elements method(FEM) model with the measured model. Measured model is described by the curve fitting scheme with the function of a random input voltage. The piezoelectric actuator shows nonlinear response characteristics including hysteresis due to the ferroelectric characteristics. This paper proposes input sharping and sliding mode control(SMC) algorithm. In input sharping control algorithm, we examine overshoot, residual vibration, effect of modeling error and hysteresis. This paper proposes the sliding mode control with an integrator to improve the ability of tracking response to a complex input signal. Finally, this paper considers coupling effects of a precision positioning stage with the sliding mode control technique.

입력성형기법을 이용한 유연 다관절 측정시스템의 잔류진동 제거에 관한 연구

박상원 금오공과대학교 대학원 2009 국내박사

Every positioning system, including high-speed and precise positioning stages and coordinate measuring machines (CMMs), undergoes residual vibration during the accelerating/decelerating period. Reducing such residual vibration is very crucial to enhance the speed and precision of positioning systems, which are often employed by manufacturing/measuring machines. Input shaping is known to be a very effective tool for suppressing such residual vibration without introducing any complicated sensors and feedback control. This thesis investigates the input shaping method for flexible multi-link measurement systems. To this end, this thesis starts with the dynamic analysis and input shaping method for a Cartesian coordinates based positioning system as a preliminary study on flexible multilink systems. Vibration characteristics of the positioning system are found to be affected not only by the structural dynamics but also by the servo actuators that consist of the mechanism, driving motor and controller. This thesis proposes an integrated dynamic model to accommodate both the structural dynamics and the servo actuators. Theoretical modal analysis with a commercial finite element code is carried out to investigate the dynamic characteristics of the experimental positioning stage. Experiments are performed to validate the theoretical modal analysis and estimate the equivalent stiffness due to the servo actuators. Input shaping is dealt with to suppress vibration of the positioning system. The effects of servo control gain are also investigated. The experiments show that input shaping effectively removes residual vibrations and then improves the performance of positioning systems. Flexible multi-link systems are often dominated by several vibration modes, which lead to a need for multi-mode input shapers. This thesis proposes a new formulation for multi-mode input shapers. The essence of the proposed method is to limit the number of impulses to n+1 for n-mode input shapers. The proposed design formula is very simple to implement. The proposed method is applied to two-mode, three-mode and four-mode input shapers. This thesis discusses characteristics of several input shapers designed under the same condition. Simulations and experiments show that the proposed formulation is very useful for designing multi-mode input shapers. The dynamic characteristics of flexible multi-link systems are more complicated than conventional system because of the existence of non-linear, time-varying properties. Residual vibration is obviously a critical issue in measurement systems because it affects measurement precision and speed. This thesis proposes a new multi-mode input shaping method in order to suppress the residual vibration in flexible multi-link measurement systems. The performance of the proposed input shaping method related to measurement accuracy and precision in a fabricated flexible two-link measurement system is also investigated. The experimental results prove that the proposed input shaping method is very effective to suppress residual vibration in flexible multi-link measurement systems.

CFD를 이용한 TURBO BLOWER 냉각 유로 최적화에 관한 연구

이현철 공주대학교 계약학과 대학원 2015 국내석사

In this paper, the cooling flow passage and the temperature distribution of the 200 HP turbo blowers were studied with computational fluid dynamics (CFD) to optimize the cooling capacity of the turbo blower. The flow was 3-dimensional, steady, compressible and turbulent. The analysis was performed by utilizing the heat generations of electrical components measured by the experiment. The optimal cooling passage was determined among several passages by the component temperature. The optimized model was successfully verified with the cooling experiment. Thus, a design process for the cooling passage and temperature management for a tubo blower were established through this study and can be utilized for the next model.