• 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.

• 한국산학기술학회논문지
• /
• 제13권12호
• /
• pp.5709-5715
• /
• 2012

플라스틱 소재를 사용한 사출성형부품은 차량 경량화뿐만 아니라 충격흡수를 통한 운전자와 보행자 보호, 제품 디자인의 자율성등의 우수한 장점을 가지고 있어 그 사용량이 급속히 증가하고 있다. 유리섬유함유 플라스틱재료를 이용한 레귤레이터 레일의 사출성형시 불량률최소화를 위해 유한요소해석과 실험계획법을 사용하여 성형공정중의 휨변형량(warpage)과 수축변형를 최소화할 수 있는 최적의 성형조건을 제시하고자 한다. 또한 유리섬유의 유동성과 냉각효율의 확보를 위해 최적의 게이트 및 런너방식을 설정한다. 사출금형을 제작하여 제품을 성형제작한 후 휨변형등의 성능확인을 통해 해석방법의 유효성과 최적화의 실효성을 확인하고자 한다.

• International Journal of Control, Automation, and Systems
• /
• 제2권2호
• /
• pp.256-262
• /
• 2004

Conventionally, profile non-uniformity has been characterized by relying on approximated profile with angle or anisotropy. In this study, a new non-uniformity model for etch profile is presented by applying a discrete wavelet to the image obtained from a scanning electron microscopy (SEM). Prediction models for wavelet-transformed data are then constructed using a back-propagation neural network. The proposed method was applied to the data collected from the etching of tungsten material. Additionally, 7 experiments were conducted to obtain test data. Model performance was evaluated in terms of the average prediction accuracy (APA) and the best prediction accuracy (BPA). To take into account randomness in initial weights, two hundred models were generated for a given set of training factors. Behaviors of the APA and BPA were investigated as a function of training factors, including training tolerance, hidden neuron, initial weight distribution, and two slopes for bipolar sig-moid and linear function. For all variations in training factors, the APA was not consistent with the BPA. The prediction accuracy was optimized using three approaches, the best model based approach, the average model based approach and the combined model based approach. Despite the largest APA of the first approach, its BPA was smallest compared to the other two approaches.

• Korean Journal of Geomatics
• /
• 제1권1호
• /
• pp.95-101
• /
• 2001

In a route plan, the route selection is a complicated problem to consider the spatial distribution and influence through overall related data and objective analysis on the social, economic and technical condition. The developed system in this study was compared and estimated by deciding a practical section for its validity and efficiency. Using Geographic Information System (GIS), the various information required for route selections in database was constructed, the characteristics of subject area by executing three-dimensional terrain analysis was grasped effectively, and the control point through buffering, overlay and location operation was extracted. An optimum route was selected by calculating the sum of alternatives to the sub-criteria weight, and from this result, there is a difference between real route and proposed route according to the prioritization of decision criteria based on the importance. This research could be constructed and applied geospatial information to the reasonable route plan and an optimum route selection efficiently using GIS. Therefore, the applications are presented by applying Analytic Hierarchy Process (AHP) to the decision-making of information needed in route selection.

• Advances in nano research
• /
• 제16권4호
• /
• pp.353-363
• /
• 2024

The incorporation of nanoplatelets in composite and polymeric materials represents a recent and innovative approach, holding substantial promise for diverse property enhancements. This study focuses on the application of nanocomposites in the production of sports equipment, particularly soccer balls, aiming to bridge the gap between theoretical advancements and practical implications. Addressing the longstanding challenge of suboptimal interaction between carbon nanofillers and epoxy resin in epoxy composites, this research pioneers inventive solutions. Furthermore, the investigation extends into unexplored territory, examining the integration of glass fiber/epoxy composites with nanoparticles. The incorporation of nanomaterials, specifically expanded graphite and graphene, at a concentration of 25.0% by weight in both the epoxy structure and the composite with glass fibers demonstrates a marked increase in impact resistance compared to their nanomaterial-free counterparts. The research transcends laboratory experiments to explore the practical applications of nanocomposites in the design and production of sports equipment, with a particular emphasis on soccer balls. Analytical techniques such as infrared spectroscopy and scanning electron microscopy are employed to scrutinize the surface chemical structure and morphology of the epoxy nanocomposites. Additionally, an in-depth examination of the thermal, mechanical, viscoelastic, and conductive properties of these materials is conducted. Noteworthy findings include the efficacy of surface modification of carbon nanotubes in preventing accumulation and enhancing their distribution within the epoxy matrix. This optimization results in improved interfacial interactions, heightened thermal stability, superior mechanical properties, and enhanced electrical conductivity in the nanocomposite.

• 대한토목학회논문집
• /
• 제7권3호
• /
• pp.141-154
• /
• 1987

본(本) 연구(硏究)에서는 전최적화(全最適化) 과정(過程)을 two-Levels로 나누었다. Level-1에서는 작용하중(作用荷重) 및 설계응력(設計應力)을 정규분포(正規分布)로 하는 확률변수(確率變數)로 하여 각부재(各部材)가 허용파괴확률(許容破壞確率)을 초과(超過)하지 않도록 단면(斷面) 최적화(最適化)하고 Level-2에서는 트러스의 절점좌표(節點座標)를 변수(變數)로 하여 형상(形狀) 최적화(最適化)한 것이다. Level-1에서는 유도(誘導)된 비선형계획문제(非線型計劃問題)를 SUMT문제(問題)로 교환(交換)시켜 Modified Newton-Raphson Method에 의한 SUMT법(法)을 채택(採擇)하고 Level-2에서는 Powell Method의 일방향(一方向) 조사법(調査法)에 의해 목적함수(目的函數)만이 최소(最小)가 되도록 하는 기법(技法)을 도입(導入)하여 형상(形狀) 최적화(最適化)를 하였다.

• 대한기계학회논문집A
• /
• 제34권9호
• /
• pp.1305-1310
• /
• 2010

복잡한 시스템 계의 설계정보를 효과적으로 추출하기 위해서 형상최적설계가 수행되는 경우, 일반적으로 유한요소해석 기법과 D-최적배열을 이용한 실험계획법이 연동된 반응표면모델을 구성하고 여기에 최적설계기법이 적용된다. 그러나, 설계변수에 형상공차와 같은 변동성이 존재하면 최적해의 강건성 확보를 위하여 설계변수의 형상공차를 확률론적인 변동성으로 고려한 추가적인 강건설계가 필요하다. 본 연구에서는 계산시간이 많이 소요되는 유한요소해석에 의한 강건설계문제에 설계변수의 표준정규분포를 고려한 반응표면모델을 구축하여 최적설계를 수행하므로서 손쉽게 강건최적값을 구하는 방법을 제안하였다. 승용차용 브레이크 디스크에 제안된 방법을 적용하여 열변형과 중량을 최소화하는 설계변수의 강건최적해를 구하고, 몬테카를로 시뮬레이션 추정결과와 비교하여 이의 적합성을 검증하였다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계 학술대회논문집(수송기계편)
• /
• pp.134-139
• /
• 2005

This paper presents a robust design of vehicle interior noise using Taguchi method and a substructure synthesis method with a hybrid model. Firstly, the proposed method identifies the critical process of the concerned interior noise through a TPA (Transfer Path Analysis). Secondly, a strategy for a robust design is discussed, in which the major noise factor among uncertainties in the process is quality distribution of rubber bushes connecting a cradle and a trimmed body. Thirdly, a virtual test model fer the process is developed by applying a substructure synthesis method with a hybrid modeling approach. Fourthly, virtual tests are carried out according to the predefined tables of orthogonal array in Taguchi robust design process. The process was performed under 2 sub-steps. The first step is sensitivity analysis of 31 panels, and the other step is weight optimization of mass dampers on sensitive panels. Finally, two vehicles with the proposed countermeasures were validated. The proposed method reduces 87.5% of trials of measurements due to the orthogonal arrays and increases robustness by 8.6dB of S/N ratio and decreases $5\;dB(A){\sim}10\;dB(A)$ of interior noise in the concerned range of RPM.

• 한국식품영양과학회지
• /
• 제42권11호
• /
• pp.1759-1766
• /
• 2013

본 연구는 오징어 간에 분포하는 효소를 활용한 은연어 육가수분해물 제조의 최적 조건과 생성된 가수분해물의 식품 및 생물 활성 특성을 조사하였다. 가열 처리한 은연어 육가수분해물 제조의 최적 조건은 가수분해 온도 $55^{\circ}C$, pH 5.5, 오징어 간의 첨가 비율 0.67%이었다. 총 구성 아미노산은 함량은 가열 및 산 처리 가수분해물에서 유의적인 차이를 보이지 않았으나, 총 유리 아미노산 함량은 조성은 산 처리가수분해물이 높았다. 한편 무기질 함량은 가열 처리 가수분해물이 높았다. Cd의 허용 기준을 고려할 때 가수분해물 제조 시 전처리 공정은 가열이 적합한 것으로 나타났다. 가열 처리한 가수분해물의 분자량은 1,000~9,500 Da이 40.0%, 500 Da이 6.7%, 200~250 Da은 12.6%, 그 이외 분자량 물질이 34.8%인 반면, 산 처리 가수분해물은 450~5,600 Da이 40.9%, 200~300 Da이 16.8%, 그 이외의 분자량이 42.3%에 해당하였다. 가수분해물은 $200{\mu}g/mL$의 농도에서 HepG2 세포에 대한 독성은 관측되지 않았다. 가수분해물의 잠재적인 적용 분야를 확인하기 위해 식품 및 건강 기능성에 대한 연구를 진행하고 있는 중이다.

• 한국지반신소재학회논문집
• /
• 제10권4호
• /
• pp.123-130
• /
• 2011

본연구의 선행연구에서는 주로 강모래를 포함한 각종 되메움재 후보군에 대하여 함수비, 건조단위중량, 입도분포 등의 차이에 따른 열저항 특성을 조사, 비교 분석하였다. 이 연구에서는 선행연구 결과를 토대로 하여 지중송전관로의 송전용량의 증대 및 열폭주(thermal runaway) 현상을 방지할 수 있는 새로운 되메움재의 개발에 주안점을 두었다. 연구를 위해 강모래, 재생모래, 쇄석, 석분과 같은 원재료에 플라이 애쉬(fly-ash), 슬래그(slag), 플록(floc)과 같은 입자가 작고 보습효과가 있는 재료를 혼합하여 혼합비와 함수상태에 따룬 열저항률의 변화를 측정, 분석하였다. 연구결과 단일재료만으로는 되메움재로의 사용이 어려울 것으로 판단되나, 쇄석의 입도분포 개선을 위해 평균입경이 작은 재생모래와 석분, 술래그 및 플록을 혼합한 결과, 최적함수비에서 $50^{\circ}C$-cm/Watt이하의 열저항률을 얻을 수 있고, 최적함수비 상태 후의 건조시에도 열저항률의 증가가 일어나지 않고 있어, 어느 정도 되메움재로서의 최적화에 근접한 것으로 판단된다.