• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.120-122
• /
• 2006

The paper concerns the hybrid optimization of fuzzy inference systems that is based on Hierarchical Fair Competition-based Parallel Genetic Algorithms (HFCGA) and information data granulation. The granulation is realized with the aid of the Hard C-means clustering and HFCGA is a kind of multi-populations of Parallel Genetic Algorithms (PGA), and it is used for structure optimization and parameter identification of fuzzy model. It concerns the fuzzy model-related parameters such as the number of input variables to be used, a collection of specific subset of input variables, the number of membership functions, the order of polynomial, and the apexes of the membership function. In the hybrid optimization process, two general optimization mechanisms are explored. Thestructural optimization is realized via HFCGA and HCM method whereas in case of the parametric optimization we proceed with a standard least square method as well as HFCGA method as well. A comparative analysis demonstrates that the proposed algorithm is superior to the conventional methods.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.87-90
• /
• 2001

This paper presents a design of new type biped robot and dynamic walking simulation for this system. The robot is distinguished from other one by which has a parallel mechanism type trunk and lead-screw type actuators to drive the joints of the trunk, knee and ankle. The basic consideration on the design is that it is able to accommodate itself to human's daily environments without any other modification of around and also to operate its upper limbs more smoothly with a spine functional trunk. It is designed according to a human with about 130 cm height and about 30 kg weight. And it also is able to dynamically walk on an even ground. It has constructed with total 14 DOFs which have two legs, a hip, and a trunk. The joints of each leg and trunk are adopted with a parallel structure which has good kinematic characteristics and take light weight. To test of the capacity of joint actuators and to analysis of the dynamic properties of the biped robot, optimized trunk trajectory is determined by means of an approximated FFT method based on ZMP criteria, and dynamic simulation is performed using DADS with a 1.1 time/step velocity on the even ground during four steps.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.204-206
• /
• 2007

In this study, we introduce the optimization method of fuzzy inference systems that is based on Hierarchical Fair Competition-based Parallel Genetic Algorithms (HFCGA) and information data granulation, The granulation is realized with the aid of the Hard C-means clustering and HFCGA is a kind of multi-populations of Parallel Genetic Algorithms (PGA), and it is used for structure optimization and parameter identification of fuzzy model. It concerns the fuzzy model-related parameters such as the number of input variables to be used, a collection of specific subset of input variables, the number of membership functions, the order of polynomial, and the apexes of the membership function. In the optimization process, two general optimization mechanisms are explored. The structural optimization is realized via HFCGA and HCM method whereas in case of the parametric optimization we proceed with a standard least square method as well as HFCGA method as well. A comparative analysis demonstrates that the proposed algorithm is superior to the conventional methods. Particularly, in parameter identification, we use the UNDX operator which uses multiple parents and generate offsprings around the geographic center off mass of these parents.

• 정보관리연구
• /
• 제29권3호
• /
• pp.1-26
• /
• 1998

순차 Java 프로그램을 병렬 머쉰에서 실행할 경우 루프는 전체 수행 시간 중 많은 부분을 차지하므로 병렬성 검출의 기본이 된다. 본 논문은 기존에 작성된 단일 루프 구조를 갖는 Java 프로그래밍 언어에서 종속성 분석을 수행하여 묵시적 병렬성을 검출하는 방법을 제안한다. 또한 재구성 컴파일러에 의하여 병렬 코드를 생성하는 방법과 Java 원시 프로그램을 Java 프로그래밍 언어 자체에서 지원하는 다중스레드 기법으로 변환하는 방법을 제안한다. 스레드 문장으로 변환된 프로그램에 대해 루프의 반복계수와 쓰레드 수를 매개변수로 하여 성능 분석을 하였다. 재구성 컴파일러에 의한 장점은 사용자의 병렬성 검출에 대한 오버헤드를 줄이고, 순차 Java 프로그램에 대한 효과적인 병렬성 검출을 가능하게 한다.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2006년도 하계학술대회 논문집 Vol.7
• /
• pp.155-156
• /
• 2006

Single crystalline ZnO fims were successfully grown on r-plane sapphire substrate by plasma-assisted molecular beam epitaxy. Epitaxial relationship between the ZnO film and the-r-plane sapphire was determined to be [-1101]$Al_2O_3\;{\parallel}$ [0001]ZnO, [11-20]$Al_2O_3\;{\parallel}$ [-1100]ZnO based on the in-situ RHEED analysis and confirmed again by HRXRD measurements. Grown (11-20) ZnO films showed faceted structure along the <0001> direction and the RMS roughness was about 4 nm.

• 정보처리학회논문지D
• /
• 제10D권3호
• /
• pp.459-470
• /
• 2003

데이터웨어하우징의 애트리뷰트 벡터나 멀티미디어 데이터베이스의 특징 벡터는 모두 고차원 데이터를 이루고 있기 때문에, 이러한 고차원 데이터를 효율적으로 검색하기 위해서는 고차원 색인 기법이 요구된다. 이를 위하여 다수의 고차원 색인 기법들이 제안되었는데, 제안된 대부분의 색인 기법들이 차원의 수가 증가할수록 검색 성능이 급격히 저하되는 ‘차원 저주(dimensional curse)’ 문제를 지니고 있다. 셀-기반 필터링(Cell-Based Filtering : CBF) 기법은 이러한 차원 저주 문제를 해결하기 위해 제안되었다. 그러나 CBF 기법은 데이터의 양이 증가할수록 선형적으로 검색 성능이 감소하며, 이를 극복하기 위해 병렬 처리 기법을 사용하는 것이 필요하다. 본 논문에서는 데이터 디클러스터링(declustering) 방법으로 수평 분할 방식을 사용한 병렬 CBF 기법을 제안한다. 아울러 제안한 병렬 CBF 기법의 성능을 최대화하기 위하여, 병렬 CBF 기법을 다수의 서버로 구성된 Shared Nothing(SN) 구조의 클러스터 아키텍쳐 하에서 구축한다. 또한 SN 구조의 클러스터 아키텍쳐에 적합한 데이타 삽입 알고리즘, 범위질의 처리 알고리즘, k-최근접 질의 처리 알고리즘을 제시한다. 마지막으로 제안하는 병렬 CBF 기법이 기존 CBF 기법과 비교하여 서버 개수에 비례하여 우수한 검색 성능을 달성함을 보인다.

• 대한기계학회논문집B
• /
• 제40권9호
• /
• pp.597-604
• /
• 2016

본 연구에서는 GPU를 이용한 비압축성 유동장의 병렬연산을 위하여, P2P1 유한요소를 이용한 분리 알고리즘 내의 행렬 해법인 이중공액구배법(Bi-Conjugate Gradient)의 CUDA 기반 알고리즘을 개발하였다. 개발된 알고리즘을 이용해 비대칭 협착관 유동을 해석하고, 단일 CPU와의 계산시간을 비교하여 GPU 병렬 연산의 성능 향상을 측정하였다. 또한, 비대칭 협착관 유동 문제와 다른 행렬 패턴을 가지는 유체구조 상호작용 문제에 대하여 이중공액구배법 내의 희소 행렬과 벡터의 곱에 대한 GPU의 병렬성능을 확인하였다. 개발된 코드는 희소 행렬의 1개의 행과 벡터의 내적을 병렬 연산하는 커널(Kernel)로 구성되며, 최적화는 병렬 감소 연산(Parallel Reduction), 메모리 코얼레싱(Coalescing) 효과를 이용하여 구현하였다. 또한, 커널 생성 시 워프(Warp)의 크기에 따른 성능 차이를 확인하였다. 표준예제들에 대한 GPU 병렬연산속도는 CPU 대비 약 7배 이상 향상됨을 확인하였다.

• 한국세라믹학회지
• /
• 제49권3호
• /
• pp.205-215
• /
• 2012

It has been studied the crystal and block structures of the hexagonal ferrites with M, W, Y and Z types prepared by various coprecipitation-oxidation method. The structures have been refined with a Rietveld analysis of the powder X-ray diffraction pattern with high precision ($R_{WP}$ <0.09, $R_I$ <0.03). The density difference between the S-blocks was proportioned to the cobalt contents in hexagonal ferrites, but that between the R or T-blocks was relatively small. Compared with the blocks and cation-oxygen polyhedra in BaM ($BaFe_{12}O_{19}$), those were bulky to the normal direction for the c-axis in $Co_2W$ ($BaCo_2Fe_{16}O_{27}$) and to the parallel direction for the c-axis in $Co_2Y$ ($Ba_2Co_2Fe_{12}O_{22}$) and $Co_2Z$ ($Ba_3Co_2Fe_{24}O_{41}$). The S-blocks of $Co_2W$, $Co_2Y$, and $Co_2Z$ were unstable and distorted. Because the T-block of $Co_2Z$ was unstable, the T-block was decomposed into the Ba-rich phase and $Co_2W$ at high temperatures above $1200^{\circ}C$. A standard powder X-ray diffraction pattern for $Co_2Z$ was proposed as well.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2000-2003
• /
• 2003

In recent years, robots have been used widely in industrial field and have been expanded as a result of continuous research and development for high-speed and miniaturization. The goal of this paper is to design the serial manipulator through kinematic analysis and to control the position and orientation of end-effector with respect to time. In general, a structure of industrial robot consists of several links connected in series by various types of joints, typically revolute and prismatic joints. The movement of these joints is determined in inverse kinematic analysis. Compared to the complicated structure of parallel and hybrid robot, open loop system retains the characteristic that each link is independent and is controlled easily. AC servo motor is used to place the robot end-effector toward the accurate point with the desired speed and power while it is operated by position control algorithm. The robot end-effector should trace the given trajectory within the appropriate time. The trajectory of end-effector can be displayed on the monitor of general personal computer through Opengl program.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1999년도 가을 학술발표회 논문집
• /
• pp.387-394
• /
• 1999

Artificial Neural Network (ANN) is a computational model inspired by the structure and operations of the brain. It is massively parallel system consisting of a large number of highly interconnected and simple processing units. The purpose of this paper is to verify the applicability of ANN to predict experimental results through the use of measured experimental data. Although there have been accumulated data based on hysteretic characteristics of structural element with cyclic loading tests, it is difficult to directly apply them for the analysis of elastic and plastic response. Thus, simple models with mathematical formula such as Bi-Linear Model, Ramberg-Osgood Model, Degrading Tri Model, Takeda Model, Slip type Model, and etc, have been used. To verify the practicality and capability of this study, ANN is adapted to several models with mathematical formula using numerical data To show the efficiency of ANN in nonlinear analysis, it is important to determine the adequate input and output variables of hysteretic models and to minimize an error in ANN process. The application example is Beam-Column joint test using the ANN in modeling of the linear and nonlinear hysteretic behavior of structure.