• 전기학회논문지
• /
• 제60권3호
• /
• pp.639-647
• /
• 2011

In this paper, we introduce a design methodology of data-centroid Radial Basis Function neural networks with extended polynomial function. The two underlying design mechanisms of such networks involve K-means clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on K-means clustering method for efficient processing of data and the optimization of model was carried out using PSO. In this paper, as the connection weight of RBF neural networks, we are able to use four types of polynomials such as simplified, linear, quadratic, and modified quadratic. Using K-means clustering, the center values of Gaussian function as activation function are selected. And the PSO-based RBF neural networks results in a structurally optimized structure and comes with a higher level of flexibility than the one encountered in the conventional RBF neural networks. The PSO-based design procedure being applied at each node of RBF neural networks leads to the selection of preferred parameters with specific local characteristics (such as the number of input variables, a specific set of input variables, and the distribution constant value in activation function) available within the RBF neural networks. To evaluate the performance of the proposed data-centroid RBF neural network with extended polynomial function, the model is experimented with using the nonlinear process data(2-Dimensional synthetic data and Mackey-Glass time series process data) and the Machine Learning dataset(NOx emission process data in gas turbine plant, Automobile Miles per Gallon(MPG) data, and Boston housing data). For the characteristic analysis of the given entire dataset with non-linearity as well as the efficient construction and evaluation of the dynamic network model, the partition of the given entire dataset distinguishes between two cases of Division I(training dataset and testing dataset) and Division II(training dataset, validation dataset, and testing dataset). A comparative analysis shows that the proposed RBF neural networks produces model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• 디지털융복합연구
• /
• 제11권12호
• /
• pp.41-47
• /
• 2013

국내 미디어기업에 대한 경영전략 사례 연구가 극히 드문 상황에서, 본 연구는 대표적인 뉴미디어 콘텐츠 기업인 CJ E&M에 대한 사례분석을 실시했다. 분석틀로는 CJ E&M의 출범과 다각화 과정에 대해서는 진화론적 관점을 적용하고, 출범 이후는 외부환경분석과 자원준거관점에 기반한 내부역량분석을 실시했다. CJ E&M은 그동안 수평 수직 다각화를 통해 규모의 시너지를 높이고, 타 미디어기업과는 달리 콘텐츠 강화 전략을 꾸준하게 전개한 점이 돋보인다. 또 엔터테인먼트 미디어 산업에 대한 열정과 의지의 리더십, 전문성 높은 인적 구성, 유연하고 창의적인 기업 문화, 효과적인 콘텐츠 포트폴리오 구성 등이 CJ E&M을 국내 미디어 선도기업으로 이끈 원동력이라고 평가할 수 있다. 그러나 경쟁과 성과를 지나치게 강조하는 조직 분위기와 오너의 의지에 따라 주요 사업이 결정되는 기업구조는 오히려 조직문화를 경색시키고 글로벌 경쟁력을 약화시킬 수 있다는 점을 주지할 필요가 있다.

• 한국대기환경학회지
• /
• 제17권1호
• /
• pp.85-96
• /
• 2001

The objective of this study is to develop the water spraying which can effectively by applied to the control or suppression of the fugitive dust generated from the scrap metal handling area at the Port of Inchon. As a first step toward this goal, we carried out some preliminary analyses on the chemical composition, physical shape, and particle size distribution of the sample dust. Next, to quantitatively investigate the effect of adding surfactants to the spraying water on the wettability of the sample dust, the Standard Sink Test was carried out for four different surfactants and at six different concentrations using the surfactants considered in this study. Results of from the preliminary analysis indicated that the main chemical component consisting of the sample dust is Goethite(FeO(OH)) and that the particles smaller than 10 ${\mu}{\textrm}{m}$ in geometric diameter occupy about 36% of the sample dust in mass. This result implies that the fugitive dust generated from the scrap metal handling area at the Port of Inchon should affect the environment nearby more than we have expected. This is because of relatively large mass percentage of the small metal particles less than 10${\mu}{\textrm}{m}$ in geometric diameter, what we may call respirable particles. As for the results of the Standard Sink Test, higher surfactant concentration tends to result in the higher wettability of the sample dust for the surfactants considered in this study, which in turn ensures the high particle collection efficiency of the droplets generated from the water spraying system. Based upon this preliminary results, studies to develop more sophisticated scaled model for dynamic test in underway and the effort to find the best surfactants as well as the optimum operating conditions are being made at the same time.

• Molecular & Cellular Toxicology
• /
• 제1권1호
• /
• pp.17-25
• /
• 2005

There are some critical drawbacks in the use of biomarkers for a global assessment of the toxicological impacts many chemicals and environmental pollutants have, primarily due to an individual biomarker's specificity for an explicit chemical or toxicant. In other words, the biomarker-based assessment methodology used to analyze toxicological effects lacks a high-throughput capability. Therefore, eco-toxicogenomics, or the study of toxicogenomics with organisms present within a given environmental locale, has recently been introduced with the advent of the so-called "-omics" era, which began with the creation of microarray technologies. Fish are comparable with humans in their toxicological responses and thus data from toxicogenomic studies performed with fish could be applied, with appropriate tools and implementation protocols, to the evaluation of environments where human or animal health is of concern. At present, there have been very active research streams for developing expression sequence tag (EST) databases (DBs) for zebra fish and rainbow trout. Even though few reports involve toxicogenomic studies with fish, a few groups have successfully fabricated and used cDNA microarrays or oligo DNA chips when studying the toxicological impacts of hypoxia or some toxicants with fish. Furthermore, it is strongly believed that this technology can also be implemented with non-model fish. With the standardization of DNA microarray technologies and ample progress in bioinformatics and proteomic technologies, data obtained from DNA microarray technologies offer not only multiple biomarker assays or an analysis of gene expression profiles, but also a means of elucidating gene networking, gene-gene relations, chemical-gene interactions, and chemical-chemical relationships. Accordingly, the ultimate target of eco-toxicogenomics should be to predict and map the pathways of stress propagation within an organism and to analyze stress networking.

• 한국항공우주학회지
• /
• 제48권7호
• /
• pp.523-528
• /
• 2020

스프링-메스 시스템의 원리를 이용하여 수평 구동형 MEMS 관성 스위치를 설계하였다. 본 MEMS 스위치는 외부에서 발생하는 가속도를 감지하여 점화안전장치를 장전시키는 역할을 한다. 성능 모델링을 통하여 다양한 가속도 조건에서의 구동 양상을 분석하였다. 시뮬레이션 결과 가속도의 기울기가 10g/msec 이하인 경우에 MEMS 스위치는 10g에서 잘 작동하는 것으로 나타났다. 반면에, 설계 변수들의 공차를 10%로 고려한 시뮬레이션 결과 스프링 폭과 길이에 의해 임계 동작 가속도가 규격(10±2g)을 벗어났다. 제작 공정상 10% 이하의 공차 관리가 어려운 스프링 폭을 두 배로 늘렸을 때 규격을 만족하는 것을 확인하고 설계보완을 제안하였다.

• Architectural research
• /
• 제22권1호
• /
• pp.13-21
• /
• 2020

The purpose of this study was to evaluate the variation in building energy predictions caused by simulation settings related to building envelop thickness. The study assessed the ceiling depth impact on skylight energy performance through OpenStudio integrated Radiance and EnergyPlus simulation programs. A ceiling as deep as 1.5 to 3m was analyzed for skylight to roof ratios from 1% to 25%. The results indicated that the building ceiling depth negatively affected the capability of skylights to significantly reduce building energy consumption. Through a parametric analysis, the study concluded that 8%, 9%, 10% and 11% skylight to roof ratio were optimal in terms of total building energy consumption for a ceiling depth of 1.5m, 2m, 2.5m and 3m, respectively. In addition, the results showed that the usually recommended 5% skylight to roof ratio was only efficient when no ceiling depth was included in the simulation model. Furthermore, the study indicated that the building energy saved by the optimal skylight of each ceiling depth decreased as the ceiling depth deepened. The highest total building energy reduction was 9%, 7%, 5% and 3% for a ceiling depth of 1.5m, 2m, 2.5m and 3m, respectively. This study induced that the solar heat gains and daylight visible transmittance by ceiling depth were crucial in the predictions of skylight energy performance and should not be neglected through building simulation simplifications as it is commonly done in most simulation programs' settings.

• 한국지능시스템학회논문지
• /
• 제11권5호
• /
• pp.406-412
• /
• 2001

본 논문은 기존의 퍼지 단층 퍼셉트론 알고리즘의 학습 시간과 수렴성을 개선하기 위해 인간 신경계의 생리학적 뉴런 구조를 분석하며 퍼지 논리를 이용한 새로운 뉴런 구조를 제시하고, 이를 바탕으로 생리학적 퍼지 단층 퍼셉트론(P-FLSP: Physiological Fuzzy Single Layer Perceptron)에 대한 학습 모형과 학습 알고리즘을 제안한다. 제안된 학습 알고리즘의 성능을 평가하기 위해 Exclusive OR 문제, 3-bit parity 문제 그리고 차량 번호판 인식 문제 등에 적용하여 피곤의 피지 단층 퍼셉트론 알고리즘과 성능을 비교, 분석하였다. 실험 결과에서는 제안된 학습 알고리즘(P-FSLP)이 기존의 퍼지 단층 학습 알고리즘보다 지역 최소화에 빠질 가능성이 감소하였으며 학습 시간과 수렴성도 개선되었을 뿐만 아니라, 영상 인식등에 대한 응용 가능성도 제시되었다.

• 전기전자학회논문지
• /
• 제22권2호
• /
• pp.420-426
• /
• 2018

가까운 미래에, 전자의 ballastic 혹은 near-ballastic 이동이 가능한 CNT(Carbon NanoTube)를 활용한 CNTFET(Carbon NanoTube Field Effect Transistor)은 현재의 실리콘 기반 트랜지스터를 교체할 유력한 후보 중 하나로 고려되고 있다. 고성능의 CNTFET으로 대규모 집적회로를 구현하기 위해서는 semiconducting CNT가 CNTFET 안에 동일한 간격과 높은 밀도로 정렬되어 배치되어야 하지만, CNTFET 공정의 미성숙으로, CNTFET 안의 CNT는 불규칙하게 배치하게 되고, 현존하는 HSPICE 라이브러리 파일은 불규칙한 CNT 배치에 의한 성능의 변화를 회로 레벨에서 평가할 수 있는 기능을 지원하지 않는다. 이러한 성능의 변화를 평가하기 위해서 선형 프로그래밍을 활용한 방법이 과거에 제안되었으나, CNTFET의 전류와 게이트 커패시턴스를 계산하는 과정에서 오차가 발생할 수 있는 문제점이 있다. 본 논문에서는 언급한 오차가 발생되는 이유에 대해서 자세히 논하고, 이 오차를 줄일 수 있는 새로운 방법을 제시하고자 한다. 시뮬레이션 검토 결과, 새롭게 제시된 방법이 기존 방법의 오차, 7.096%를 3.15%까지 줄일 수 있음을 보이고 있다.

• 한국지반공학회지:지반
• /
• 제14권5호
• /
• pp.191-204
• /
• 1998

플라이 애쉬의 폐기물 매립장 대용 차수재로서의 적용성을 평가하기 위하여 실내 실험과 모델 연구를 실시하였다. 실제의 매립장 차수층이 불포화 상태일 수 있는 점을 고려하여, 대용 차수재와 기존의 점토 차수재에서의 포화 및 불포화 상태에서의 물 흐름 특성을 비교하였다. 차수층에서의 오염물 이동 특성을 연구하기 위하여 오염물을 이용한 회분실험을 실시하였으며, 이때 폐놀을 대표 오염물질로 사용하였다. 모델은 물질 수지에 기초하였으며, 유한 차분법과 predictor-corrector법을 이용한 수치 해법으로 풀었고, 연립 미분방정식을 다루는테는 sequential method를 이용했다. 연구 결과 대용 차수재는 기존 차수재보다 폐놀 차단 능력이 더 크고 또 불포화대에서의 차수능력도 더 크다는 것을 알 수 있었다. 또한, 본 연구에서 고려된 흐름조건에서는 차수층을 포화대로 해석할 경우 통과하는 물의 양이 약간 과소평가되고 폐놀 차단 능력은 매우 과소평가됨을 알 수 있었다.

• Steel and Composite Structures
• /
• 제8권1호
• /
• pp.1-18
• /
• 2008

Observations from experiments and real fire indicate that restrained steel beams have better fire-resistant capability than isolated beams. Due to the effects of restraints, a steel beam in fire condition can undergo very large deflections and the run away damage may be avoided. In addition, axial forces will be induced with temperature increasing and play an important role on the behaviour of the restrained beam. The factors influencing the behavior of a restrained beam subjected to fire include the stiffness of axial and rotational restraints, the load type on the beam and the distribution of temperature in the cross-section of the beam, etc. In this paper, a simplified model is proposed to analyze the performance of restrained steel beams in fire condition. Based on an assumption of the deflection curve of the beam, the axial force, together with the strain and stress distributions in the beam, can be determined. By integrating the stress, the combined moment and force in the cross-section of the beam can be obtained. Then, through substituting the moment and axial force into the equilibrium equation, the behavior of the restrained beam in fire condition can be worked out. Furthermore, for the safety evaluation and repair after a fire, the behaviour of restrained beams during cooling should be understood. For a restrained beam experiencing very high temperatures, the strength of the steel will recover when temperature decreases, but the contraction force, which is produced by thermal contraction, will aggravate the tensile stresses in the beam. In this paper, the behaviour of the restrained beam in cooling phase is analyzed, and the effect of the contraction force is discussed.