http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
In this work we have investigated the effects of deposition, hydrogenation and argonation time on the physical properties in microcrystalline silicon (μc-Si) films prepared by plasma-enhanced chemical vapor deposition (PECVD) from highly argon-diluted silane gases. The physical properties were studied using photoluminescence, dark conductivity, Raman spectroscopy, and transmission electron microscopy (TEM) measurements. It was observed that physical characteristics caused by crystalline phase dominated with increasing deposition time. Cross-sectional TEM micrographs show that crystalline phase with inverse cone shape are formed, resulting in an increase in the volume fraction of crystalline phase with deposition time. Post-growth hydrogenation and argontion was performed by exposing μc-Si:H films of about 190 nm thick to hydrogen and argon plasma in situ after deposition. It was observed that physical characteristics caused by crystalline phase increases with hydrogenation and argonation. Cross-sectional transmission electron microscopy image of μc-Si:H films after hydrogenation and argoantion shows the crystalline grains with approximately inverse cone shape. The μc-Si:H using argon dilution suggests that hydrogen itself may not be the unique factor in the formation of μc-Si:H films.
Neural network potential : from code development to application
The molecular dynamics (MD) simulation is a favored method in materials science for understanding and predicting material properties from atomistic motions. In classical MD simulations, the interaction between atoms is described by an empirical interatomic potential, so the reliability of the simulation hinges on the accuracy of the underlying potential. Recently, machine learning (ML) based interatomic potentials are gaining attention as they can reproduce potential energy surfaces (PES) of ab initio calculations, with a much lower computational cost. Therefore, an efficient code for training ML potentials and inferencing PES in new configurations would widen the application range of MD simulations. In this dissertation, we announce an open-source package, SNU Interatomic Machine-learning PotentiaL packagE - version Neural Network (SIMPLE-NN) that generates and utilizes the ML potential based on the artificial neural network with the Behler-Parrinello type symmetry function as descriptors for the chemical environments. SIMPLE-NN uses the Atomic Simulation Environment (ASE) package and Google Tensorflow for high expandability and efficient training, and also supports the in-house code for quasi-Newton method. Notably, the package features a weighting scheme based on the Gaussian density function (GDF), which significantly improves accuracy and reliability of ML potentials by resolving sampling bias that exists in typical training sets. For MD simulations, SIMPLE-NN interfaces with the LAMMPS package. We demonstrate the performance and usage of SIMPLE-NN with examples of SiO2. In addition, we also generate the neural network potential for Si oxidation process at the Si/SiO2 interface. Since the process consists of various reactions accompanied by dynamic bond breaking/forming and requires long simulation time, the simulation of the process is challenging with classical MD or AIMD. To describe the accurate modeling for the oxidation process, the training set is carefully selected to include overall reaction occur in the process. The optimized potential is validated with DFT calculation data and we confirm that our potential shows consistent results with DFT results. Using the potential, we also perform the large-scale simulation with >10,000 atoms during >10 ns. As a result, we propose the package for utilizing the neural network potential and validate the usefulness of the NNP with complex system. 재료과학 분야에서 분자동역학 시뮬레이션은 원자들의 움직임을 바탕으로 재료의 성질을 예측하고 이해하는데 널리 이용되는 방법론이다. 고전적인 분자동역학 시뮬레이션에서는 원자 간 포텐셜을 이용하여 원자들 간의 상호작용을 모사하므로 시뮬레이션의 신뢰도는 원자 간 포텐셜에 큰 영향을 받지만, 이를 제작하기 위해서는 많은 노력이 필요하다. 반면, 최근 연구가 진행되고 있는 기계 학습을 이용한 포텐셜의 경우 제일원리 계산을 바탕으로 하여 이를 모사할 수 있는 원자 간 포텐셜을 훈련하는 방식을 사용하기 때문에 다양한 시스템에 쉽게 적용이 가능하며 포텐셜의 정확도 역시 우수하다. 따라서, 기계 학습을 이용한 포텐셜을 빠르게 제작하고 이를 활용할 수 있는 코드가 있다면 이를 통해 분자동역학으로 고려할 수 있는 범위를 더욱 넓힐 수 있을 것이라 기대된다. 본 논문에서는 인공신경망 포텐셜을 제작하고 활용하기 위한 오픈소스 코드 SNU Interatomic Machine-learning PotentiaL packagE - version Neural Network (SIM\\PLE-NN)에 대해 소개한다. SIMPLE-NN은 고차원 인공신경망 및 Behler-Parrinello 가 제안한 symmetry function을 활용하여 인공신경망 포텐셜을 제작하는 코드로, 확장성 및 훈련 과정의 효율성을 위해 Atomic Simulation Environment (ASE) 및 Google Tensorflow 를 사용하였다. 추가로, 훈련 데이터의 비균일성을 개선하기 위해 Gaussian density function을 이용한 가중치를 적용하여 포텐셜의 정확성과 신뢰도를 향상시켰다. 본 코드의 성능은 SiO2를 활용하여 검증하였다. 또한, 본 연구에서는 Si/SiO2 계면에서 일어나는 Si의 산화과정을 모사할 수 있는 인공신경망 포텐셜을 제작하였다. 해당 과정은 원자 간 결합이 끊어지고 새로 생성되는 과정이 포함된 다양한 반응으로 구성되어 있고 이를 확인하기 위해서는 장기간의 시뮬레이션이 필요하기 때문에 기존의 고전적인 분자동역학이나 제일원리 계산을 이용한 분자동역학으로는 확인하기 어렵다. 본 논문에서는 Si의 산화과정을 정확하게 모사할 수 있는 인공신경망 포텐셜을 만들기 위하여 산화 과정에서 일어날 수 있는 전체 반응을 포함할 수 있도록 훈련 데이터를 정하였다. 완성된 포텐셜을 검증하기 위하여 이를 이용한 시뮬레이션과 DFT 계산 결과를 비교하였고 그 결과 본 포텐셜으로 얻은 결과가 DFT를 통해 얻은 결과와 일치함을 확인하였다. 또한, 해당 포텐셜을 이용하여 10,000개 이상의 원자로 구성된 구조를 10 ns 이상 시뮬레이션하는 것이 가능했다. 결론적으로, 본 논문에서는 인공신경망 포텐셜을 사용하기 위한 코드를 개발하였으며 인공신경망 포텐셜을 이용해 복잡한 현상을 시뮬레이션하여 인공신경망 포텐셜의 유용성을 증명하였다.
조직내 커뮤니케이션이 직무만족에 미치는 영향 : 경기도 북부지역 초등교사를 대상으로
이규현 수원대학교 경영대학원 2003 국내석사
The objective of this research is to contribute to the development of elementary school system by analysing the effect of communication on the job satisfaction in the elementary school teachers. It can be accomplished through looking over how communication is made in elementary school system and how communication is related to the job satisfaction. For carrying through this study, the following three questions were presented. First, is there a significant difference of communication in elementary school system by the characteristic of population? Second, is there a significant difference of the job satisfaction of elementary school teacher by the characteristic of population? Third, how does communication in elementary school system effect the job satisfaction? The major findings of this study are as follows First, in the formal communication up-ward-communication showed the highest satisfaction and down- ward-communication and horizontal-communication were followed. Informal communication showed the lowest level in satisfaction. Second, the job satisfaction was high, generally. The job satisfaction of itself was the highest however it is showed that satisfaction of the system and the satisfaction of reward was low. Third, the effect of communication in the system were successful if up-ward-communication and the horizontal-communication were active. On the other hand, the informal communication as well as communication with people of lower status does not seem to influence any significant changes in their satisfaction levels. From these results, in order to raise satisfaction levels in elementary teachers, suggest that communication with people with high status and the horizontal-communication need to be expanded and active constructively. The horizontal-communication such as hoppy club, the staff meeting, the social meeting, has to be institutionalized. On the other hand, the administration needs the communication skills needed and further more, use informal communication positively. The teacher group has many common factors such as education. But teacher are so individual that they are reluctant to open their mind. We need that informal organization does not role as a social group but also attribute to make an open atmosphere between teachers.
제주도에서 채집된 게류 3종(Macromedaeus orientalis, Eriphia smithii, Ocypode cordimana)을 한국 미기록종으로 보고한다. M. orientalis는 스쿠버 다이빙을 통해 20m 깊이와 하조대에서 채집되었다. 반면, E. smithii와 O. cordimana는 각각 E. smithii는 조하대 암반지대의 틈에서 채집되었고 O. cordimana는 상조간대의 모래해변의 굴에서 채집되었다. 현재 E. smithii는 Eriphia속의 유일한 종으로 보고되는 종이다. 이번에 보고되는 3종을 포함해 한국산 게류는 총 203종이 된다. Three species of crabs, Macromedaeus orientalis, Eriphia smithi, and Ocypode cordimana, from Jejudo Island are recorded for the first time in Korea. M. orientalis was collected under the rocks by SCUBA diving in 20 m and at low tidal mark. However, E. smithi and O. cordimana were collected in crevices of the rock at low tidal mark and in a burrow of sand beach at supra tidal mark, respectively. At present, E. smithi is the only species of the genus represented in Korean waters. Including the new records in this study, the established brachyuran crab fauna in Korea now stands at 203 species.
이규현 韓國外國語大學校 大學院 2006 국내석사
뿌쉬낀이 생의 후반으로 갈수록 철학과 사상을 담은 작품을 쓰고 있기 때문이다. 뿌쉬낀은 자신의 철학과 사상을 자연이라는 매개체를 통해 형상화 하고 있다. 자연을 제2의 나로 간주하고 자신의 사상과 감정을 상징적으로 표현했던 것이다. 이러한 이유로 자연의 형상은 작가의 입장, 감정, 사상, 철학을 대변하며 작품을 분석하고 인물을 이해하는데 중요한 요소가 된다. 그렇기 때문에『예브게니 오네긴』에 나타난 자연의 모습은 작품을 이해하고 작가의 의도를 파악하는데 중요한 단서로 작용하게 될 것으로 사료된다.
온라인 리뷰 유용성 평가에 영향을 주는 결정요인 : FCB 그리드 모델을 중심으로
이규현 연세대학교 정보대학원 2017 국내석사
Due to the rapid development of IT, most product purchases are being done online. It has helped to reduce physical costs like price and time. Since, consumers make purchases without any experiences about the products, consumers need to acquire and select a variety of information in order to make smart purchasing decisions. To obtain information about the products that the consumers want to buy, the first information they receive is the purchase reviews of the products and they invest a lot of money and time in selecting a Huge number of reviews. Therefore, this study constructs a review’s helpfulness evaluation model by grasping the characteristics of helpfulness reviews that can provide useful information to consumers according to the characteristics of the products. Using data from ‘Amazon.com’, a global online market place, this study found that reviews with similar characteristics to products based on FCB grids model received review’s helpfulness from consumers. This study suggests the helpfulness reviews that can help the consumer to make a reasonable decision in the online market and help the product sellers to grasp the consumers’ thoughts about their products and supplement their products’ faults. 급격한 IT의 발달로 인하여, 대부분의 상품 거래는 온라인 상에서 이루어지고 있다. 온라인에서의 상품 거래는 가격과 시간 등. 물리적 비용을 줄 일 수 있게 해주었다. 하지만 상품을 직접 체험해 보지 않고 구매를 하기 때문에, 소비자들은 구매 의사결정에 있어 현명한 소비를 위해 여러 정보들을 취득하고 선별해야 하는 필요성이 있다. 소비자들이 구매하고자 하는 상품에 대한 정보를 취득하기 위해 가장 먼저 접하는 정보는 해당 상품의 구매 리뷰이며, 수 많은 리뷰를 선별함에 있어 많은 비용을 투자하게 된다. 따라서 본 연구는 상품의 성격에 따라 소비자에게 유용한 정보를 제공 가능한 유용한 리뷰의 성격을 파악하여 리뷰의 유용성 평가 모델을 구축하였다. 세계적인 온라인 마켓인 ‘Amazon.com’의 데이터를 활용하여 검증한 결과, FCB 그리드로 나눈 상품의 성격과 유사한 성격을 가진 리뷰일수록 소비자들에게 리뷰의 유용성 평가를 받음을 알 수 있었다. 이처럼 본 연구는 온라인 마켓 시장에서 소비자의 구매 의사결정에 도움을 주며, 판매자의 상품에 대한 소비자들의 생각을 파악하고 보완할 수 있는 유용성 있는 리뷰를 선별할 수 있는 연구라는 것에 대한 시사점이 있다.
眞武湯이 Gentamicin-유도 신독성에 미치는 영향
Aminoglycosides, including gentamicin, have been used as antibiotics for the various infections by gram-negative bacteria. However, there are some restrictions for using these drugs. Gentamicin, a typical aminoglycoside, has the side effect of nephrotoxicity, including polyuria, glycosuria, proteinuria, glomerulonephritis, and uremia. The aims of this study were to examine the prevention or reduction effects of Jinmootang on the gentamicin-induced nephrotoxicity and to investigate the possible mechanisms on the effect of Jinmootang. The subcutaneous injections of 60mg of gentamicin per kg of body weight to Sprague-Dawley rats for 8 days induced typical symptoms of nephrotoxicity by aminoglycosides. 0.6ml of water extract Jinmootang (100ml/chup) was orally treated in the experimental animal. 24-hour urine was collected with the metabolic cage and plasma was sampled from the abdominal aorta. The plasma concentration of sodium was significantly decreased by the treatment of gentamicin but it was not significantly changed by the treatment of Jinmootang to the animal. The concentration of potassium was greatly decreased in the gentamicin-treated animals. However, it was returned to the normal level in the Jinmootang-treated animals. The concentrations of creatinine and urea were increased by gentamicin treatment. But, Jinmootang reduced these concentrations. Nevertheless, the osmolalities of plasma in both group were not different from each other. Even though the plasma concentration of aldosterone was not significantly changed, the mean value was increased by the gentamicin intoxication. The concentration of aldosterone was decreased by the treatment of Jinmootang. The reduction of aldosterone level in plasma could be a factor to improve the hypokalemia. The fractional excretion of potassium was much higher than normal by the treatment of gentamicin and it was decreased by 50% in the Jinmootang-treated rats. Therefore, the reabsorption of potassium was significantly increased by the treatment of Jinmootang, even though the filtered load of potassium in the experimental group was much higher than control. Even though the concentration of plasma aldosterone was decreased by the treatment of Jinmootang, the fractional excretion of sodium was not increased, slightly lower. These data suggested that Na reabsorption was increased in the proximal tubule by Jinmootang. The filtered load of glucose in the Jinmootang-treated group was greater than in control. Nevertheless, the fractional excretion of glucose in the experimental group was not different from that in control. These results indicate that glucose reabsorption was increase in the proximal tubule by Jinmootang treatment. The results of this study suggest that Jinmootang could improve the some nephrotoxic symptoms induced by gentramicin treatment. Hypokalemia, the reduced glomerular filtration rate, and dysfunctions of renal proximal tubule and distal nephron were significantly recovered to normal level. The increase of glomerular filtration rate by Jinmootang might contribute to eliminate the waste product, including creatinine and urea, and/or gentamicin through the kidney.