• 한국지능시스템학회논문지
• /
• 제21권1호
• /
• pp.12-18
• /
• 2011

본 논문에서는 RBF 뉴럴 네트워크에서 은닉층 활성함수에 Interval type-2 퍼지개념을 적용한 새로운 RBF 뉴럴 네트워크를 설계하였다. 퍼지 시스템 분야에서 불확실한 정보에 대한 Type-1 퍼지집합의 성능을 보안하고자 Type-2 퍼지집합이 제안되었으며, 멤버쉽함수 안에 다시 멤버쉽함수를 생성함으로써 불확실한 정보를 좀 더 효과적으로 다루고자 하였다. 따라서 본 논문에서는 RBF 뉴럴 네트워크의 은닉층 활성함수에 type-2 퍼지집합의 개념을 적용하여 불확실한 정보에 대한 모델 성능을 개선하고자 하였다. 나아가 연결가중치를 상수항이 아닌 1차식으로 구성된 다항식을 사용하여 최종출력을 입력-출력의 관계식으로 표현하였다. 연결가중치는 기존의 경사하강법(Gradient Descent Method; GDM) 대신 conjugate gradient method(CGM)을 사용하여 파라미터를 동조하고, 은닉층의 활성함수는 공간탐색 진화 알고리즘(Space Search Evolutionary Algorithm; SSEA)을 이용하여 가우시안 함수의 중심점 및 분포상수를 동조하여 모델의 성능을 개선시킨다. 제안된 모델의 성능을 평가하기 위해 가스로 시계열 데이터를 사용하였으며, 결과를 기존 모델과 비교하였다.

• Nuclear Engineering and Technology
• /
• 제25권2호
• /
• pp.193-203
• /
• 1993

재장전노심의 핵연료 장전모형 설계를 위한 기존의 알고리즘 탐색방법의 단점을 보완하기 위한 새로운 경험적 탐색방법을 개발하였다. 노심의 핵연료 장전모형으로 고려될 수 있는 수없이 많은 경우의 수를 줄이기 위하여 일반적 핵연료 배치규칙, 영역별 배치방법 그리고 장전모형의 집단화 방법을 이용하였다. 비슷한 장전모형을 모아서 집단화시키는 기준으로 엔트로피 이론을 이용하였다. 또한 PROLOG언어를 이용하여 주어진 배치규칙에 따라 장전모형을 탐색하는 프로그램을 만들었다. 장전모형들의 노심내 출력분포 해석에는 2군 nodal코드인 MEDIUM-2D를 사용하였다. 이와같은 방법을 사용한 결과 수백개 정도의 장전모형 집단을 찾아낼 수 있었고, 여기에 가연성 독봉 배치규칙에 따라 가연성 독봉을 배치한 결과 장전모형 집단의 수를 수십개까지로 감소시킬 수 있었다. 이러한 장전모형 집단들로부터 실제로 이용 가능한 장전모형을 찾아내기 위하여, 주기길이 최대화방법과 첨두 출력 최소화방법을 사용하였다. 그 결과 고리 3호기 제10주기의 예상 재장전모형보다 주기길이는 길고 첨두출력은 낮은 장전모형을 찾아낼 수 있었다.

• 한국실내디자인학회논문집
• /
• 제17권5호
• /
• pp.157-165
• /
• 2008

This study confirm the value and the importance of the human sensibility study to add up the new meaning, and to suggest a new value of Korean sensibility study through the understanding of current statusand trend of the sensibility study in the space. The method of the study was to collect date related to the sensibility study and to analyze it focusing on its details. The date was collected from researches published on the website since the establishment of Korean Institute of Interior Design and Architectural Institute of Korea and selected at the keyword search comer. The data was extracted under keywords of research object, research purpose, research method, and analysis method. And then it was quantified with HAYASH lll program and used for analyses according to its pattern and feature. The study shows that nowadays categories representing the current status and trend of the sensibility studies in space consist of the environment, the human, and the space. The contemporary study for sensibility puts the importance on a object and a subject of the study like the environment harmonized with human and space, the humans the subject that essentially uses the spate, and the space for the architecture and the interior that puts human in. Accordingly, the study for human sensibility should develop into the study for the design focused on the intangible relationship such as 'information', 'elements for space design', 'sensibility' beyond the existing tangible categories of environment, human, and space. In addition, in the method ways of study and analysis, those studies for the sensible relationship are required to develop into new types of study applying research methods of various studies beyond the traditional border between human studies, social science, and natural science.

• Computers and Concrete
• /
• 제13권4호
• /
• pp.457-482
• /
• 2014

This paper summarizes available literature on the optimization of reinforced concrete (RC) beams. The objective of optimization (e.g. minimum cost or weight), the design variables and the constraints considered by different studies vary widely and therefore, different optimization methods have been employed to provide the optimal design of RC beams, whether as isolated structural components or as part of a structural frame. The review of literature suggests that nonlinear deterministic approaches can be efficiently employed to provide optimal design of RC beams, given the small number of variables. This paper also presents spreadsheet implementation of cost optimization of RC beams in the familiar MS Excel environment to illustrate the efficiency of the exhaustive enumeration method for such small discrete search spaces and to promote its use by engineers and researchers. Furthermore, a sensitivity analysis is performed on the contribution of various design parameters to the variability of the overall cost of RC beams.

• Journal of Astronomy and Space Sciences
• /
• 제35권3호
• /
• pp.133-141
• /
• 2018

We report on our Galactic plane searches for magnetars in the archival Chandra X-ray Observatory (CXO) data. We summarize the properties of known magnetars and use them to establish a procedure for magnetar searches. The procedure includes four steps: source finding, spectral characterization, optical counterpart checks, and period searches. We searched 1,282 archival CXO observations, found 32,838 X-ray sources, and selected 25 intriguing candidates using the developed procedure. Although we do not firmly identify a magnetar among them, we significantly reduced the number of targets in future magnetar searches to be done with better X-ray telescopes.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 2008년도 연합학회학술대회
• /
• pp.211-224
• /
• 2008

As a lot of information and media are given to users in Internet space nowadays, users feel disoriented or "lost in space" intensively. So it is suggested that we have the system to reduce information overload and to propose effective and efficient information. In this study we present a visualizing technique which uses fisheye views on data adjacency to combine global context and local details for presentation of many results in limited space. Data Adjacency on graph theory is applied to set up degree of interest which is main focus in fisheye views. Graph theory is useful to solve the problem resulted from various combinational optimization, especially it has advantages to analyze issues in information space like Internet. To test the usability of the proposed visualization technique, we compared the effectiveness of different visualization techniques. Results show that our method is evaluated with respect to less time and high satisfaction for a task accomplishment.

• 산업경영시스템학회지
• /
• 제40권2호
• /
• pp.92-98
• /
• 2017

Data clustering determines a group of patterns using similarity measure in a dataset and is one of the most important and difficult technique in data mining. Clustering can be formally considered as a particular kind of NP-hard grouping problem. K-means algorithm which is popular and efficient, is sensitive for initialization and has the possibility to be stuck in local optimum because of hill climbing clustering method. This method is also not computationally feasible in practice, especially for large datasets and large number of clusters. Therefore, we need a robust and efficient clustering algorithm to find the global optimum (not local optimum) especially when much data is collected from many IoT (Internet of Things) devices in these days. The objective of this paper is to propose new Hybrid Simulated Annealing (HSA) which is combined simulated annealing with K-means for non-hierarchical clustering of big data. Simulated annealing (SA) is useful for diversified search in large search space and K-means is useful for converged search in predetermined search space. Our proposed method can balance the intensification and diversification to find the global optimal solution in big data clustering. The performance of HSA is validated using Iris, Wine, Glass, and Vowel UCI machine learning repository datasets comparing to previous studies by experiment and analysis. Our proposed KSAK (K-means+SA+K-means) and SAK (SA+K-means) are better than KSA(K-means+SA), SA, and K-means in our simulations. Our method has significantly improved accuracy and efficiency to find the global optimal data clustering solution for complex, real time, and costly data mining process.

• 농촌계획
• /
• 제18권3호
• /
• pp.111-122
• /
• 2012

This study is to evaluate the classification system of rural-villages remodeling components which is provided for improving the quality of life for rural community by improvement of settlement environment. To achieve this, rural-villages remodeling components are classified according to the spatial structure of rural area through analysis of literature, then we have examined the applicability through case studies after modification work which is based on experts's discussion and rearrangement by pilot investigation of researcher. In the classification system of rural-villages remodeling classified productivity area, residential area, community area in first group and this classification is divided into 6 space to production, 4 space to residence, 5 space to community in second group by literature search, pilot investigation of researcher and field survey. The subject elements surveyed a total of 123 through the literature search, additionally, 1 element at a space to production and space to community in field survey for types in zoning cases. As a result, categories and items are decided that it is included 125 target elements.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 춘계학술대회논문집
• /
• pp.117-119
• /
• 2005

For large structural eigen-analysis, the whole structure is divided into some partitioned structures and through synthesis of partitioned structural model the eigen-data of structure can be obtained. In that case, eigenvalue problem consists of semidefinite mass matrix form because of displacement constraint condition. In this work the eigenvalue problem is considered by means of several method, determinant search and null space reduction method.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제3권6호
• /
• pp.215-218
• /
• 2014

해슁(hashing)을 기반으로 한 근사 최근접 이웃 탐색(approximate nearest neighbors search, ANN search) 방법에서는 데이터 샘플들을 k-비트 이진 코드로 변환하는 해쉬 함수들을 이용함으로써 근접 이웃 탐색이 이진변환 공간에서 이루어지게 된다. 본 논문에서는 PCA 기반 군집화 방법인 Principal Direction Divisive Partitioning(PDDP)를 이용한 해슁 방법을 제안한다. PDDP는 가장 큰 분산을 가지는 클러스터를 선택하여 그 클러스터의 첫 번째 주성분 방향을 이용하여 두 개의 클러스터로 분할하는 과정을 반복적으로 시행하는 군집화 방법이다. 제안하는 해슁 방법에서는 PDDP에서 분할을 위해 사용하는 주성분방향을 바이너리 코딩을 위한 사영벡터로서 사용한다. 실험결과는 제안하는 방법이 다른 해슁 방법들과 비교하여 경쟁력 있는 방법임을 입증한다.