• 정보처리학회논문지B
• /
• 제8B권5호
• /
• pp.477-484
• /
• 2001

본 논문에서는 회귀용 support vector machine의 성능 개선을 위한 모멘텀과 kernel-adatron 기법이 조합형 학습알고리즘을 제안하였다. 제안된 학습알고리즘은 supper vector machine의 학습기법인 기술기상승법에 발생하는 최적해로의 수렴에 따란 발진을 억제하여 그수렴속도를 좀 더 개선시키는 모멘텀의 장점과 비선형 특징공간에서의 동작과 구현의 용이성을 갖는 kernel-adatorn 알고리즘의 장점을 그대로 살린 것이다. 제안된 알고리즘의 support vector machine을 1차원과 2차원 비선형 함수 회귀에 적용하여 시뮬레이션한 결과, 학습속도에 있어서 2차 프로그래밍과 기존의 kernel-adaton 알고리즘보다 더 우수하고, 회귀성능면에서도 우수한 성능이 있음을 확인하였다.

• 한국음향학회지
• /
• 제20권3호
• /
• pp.8-14
• /
• 2001

본 논문에서는 부밴드 적응 필터 구조에서 음향 신호에 부가된 잡음을 제거하기 위한 새로운 비용함수와 블록 공액 경사 알고리듬을 제안하였다. 제안한 비용함수를 위하여 부밴드로 나뉘는 신호를 블록으로 구성한 후 각 대역에서의 신호를 하나의 블록으로 조합하였다. 이러한 과정을 통해 제시된 비용함수는 부밴드 적응 필터 구조에서 적응 필터에 대한 2차 형식을 가짐으로서 제안한 알고리듬의 수렴성이 보장되었다. 또한 제안한 비용함수를 최소화하는 알고리듬으로 사용한 부밴드 블록 공액 경사 알고리듬은 전대역 블록 공액 경사 알고리듬에 비해 잡음제거 성능이 뛰어난 것을 컴퓨터 모의 실험으로 확인하였다.

• 대한기계학회논문집A
• /
• 제20권12호
• /
• pp.3804-3821
• /
• 1996

The configuration optimization is a structural optimization method which includes the coordinates of a structure as well as the sectional properties in the design variable set. Effective reduction of the weight of discrete structures can be obrained by changing the geometry while satisfying stress, Ei;er bickling, displacement, and frequency constraints, etc. However, the nonlinearity due to the configuration variables may cause the difficulties of the convergence and expensive computational cost. An efficient approximation method for the configuration optimization has been developed to overcome the difficulties. The method approximates the constraint functions based onthe second-order Taylor series expansion with reciprocal design variables. The Hessian matrix is approzimated from the information on previous design points. The developed algotithms are coded and the examples are solved.

• 전산구조공학
• /
• 제1권1호
• /
• pp.87-94
• /
• 1988

반복 비선형 계획법의 하나인 선형화 기법을 절대수렴의 전제아래 합성 구조물의 최적 설계에 응용한다. 선형화 기법은 설계문제의 제약조건을 선형화된 등가 제약조건으로 변형시키며 active-set 정책을 구사한다. 결과, 매 설계단계에서 풀어야 할 상태 및 수반 방정식의 수를 줄임으로써 실질적인 계산의 절감을 기한다. 기둥으로 받쳐진 판-보 구조물은 최적화 기법의 능력을 시험키 위한 합성구조물의 좋은 예로서, 설계결과 선형화 기법은 만족할만한 수렴치로써 최적해를 산출함을 알 수 있고 나아가 이 방법은 모든 종류의 최적화 문제에 적용될 수 있을 것으로 보인다.

• 한국지능시스템학회논문지
• /
• 제18권2호
• /
• pp.236-242
• /
• 2008

본 논문은 회전형 역 진자 시스템(Rotary Inverted Pendulum System : RIPS)에 대한 계층적 공정 경쟁 기반 유전자 알고리즘(Hierarchical Fair Competition-based Genetic Algorithms : HFCGA) 기반 최적 퍼지 제어기 설계를 제안한다. 회전형 역 진자 시스템의 제어를 위해 퍼지제어기를 사용하였으며, 이때 퍼지제어기의 규칙은 LQR(Linear Quadratic Regulator) 제어기를 기반으로 하여 설계하였다. 유전자 알고리즘은 전역해를 구할 수 있는 장점이 있어 많은 분야에 성공적으로 적용되고 있지만 조기수렴 문제로 인하여 지역해에 빠질 수 있다. 이러한 문제를 해결하기 위하여 병렬유전자 알고리즘이 개발되었으며, HFCGA는 병렬유전자 알고리즘을 개선한 방법 중의 하나이다. 본 논문에서는 퍼지 제어기의 파라미터의 최적화를 위해 계층적 공정 경쟁 기반 유전자 알고리즘을 사용하였다. 시뮬레이션 및 실험을 통하여 LQR 제어기, 기존 단순유전자 알고리즘(SGA)을 이용한 퍼지제어기와 제안된 HFCGA 기반 퍼지제어기의 성능 비교를 통하여 제안된 방법의 우수성을 보인다.

• Journal of Electrical Engineering and Technology
• /
• 제8권3호
• /
• pp.621-632
• /
• 2013

In this paper, we introduce advanced architectures of genetically-oriented Fuzzy Neural Networks (FNNs) based on fuzzy set and fuzzy relation and discuss a comprehensive design methodology. The proposed FNNs are based on 'if-then' rule-based networks with the extended structure of the premise and the consequence parts of the fuzzy rules. We consider two types of the FNNs topologies, called here FSNN and FRNN, depending upon the usage of inputs in the premise of fuzzy rules. Three different type of polynomials function (namely, constant, linear, and quadratic) are used to construct the consequence of the rules. In order to improve the accuracy of FNNs, the structure and the parameters are optimized by making use of genetic algorithms (GAs). We enhance the search capabilities of the GAs by introducing the dynamic variants of genetic optimization. It fully exploits the processing capabilities of the FNNs by supporting their structural and parametric optimization. To evaluate the performance of the proposed FNNs, we exploit a suite of several representative numerical examples and its experimental results are compared with those reported in the previous studies.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제21권3호
• /
• pp.167-180
• /
• 2017

According to the fact that the low convergence level of the complete elliptic integral of the first kind for the modulus which having values approach to one. In this paper we propose novelty of the complete elliptic integral which having new infinite series that consists of new modulus introduced as own modulus function. We apply scheme of iteration by substituting the common modulus with own modulus function into the new infinite series. We obtained so many new exact formulas of the complete elliptic integral derived from this method correspond to the number of iterations. On the other hand, it has been also obtained a lot of new transformation functions with the corresponding own modulus functions. The calculation results show that the enhancement of the number of significant figures of the new infinite series of the complete elliptic integral of the first kind corresponds to the level of quadratic convergence.

• Journal of the Optical Society of Korea
• /
• 제13권3호
• /
• pp.379-385
• /
• 2009

The vergence control of binocular stereoscopic camera is the most essential factor for acquiring high quality stereoscopic images. In this paper, we proposed a binocular stereoscopic camera vergence control method using disparity information by the simple image processing and estimate the quantity of vergence control using the Lagrange interpolation equation. The method of extracting disparity information through image processing is as follows: first the key-object in left & right images was extracted through labeling of the central area of the image, and then a simple method was used for calculating the disparity value of the same key-object in the labeled left and right images. The vergence control method uses disparity information and keeps the convergence distance of left & right cameras and the distance of the key-object the same. According to the proposed method, variance in the distance of the key-object and application of calculated disparity information of obtained left & right images to the quadratic Lagrange interpolation equation could estimate the quantity of vergence control, which confirmed that the method of stereoscopic camera vergence control can be simplified through experiments on various key-objects and other convergence distance.

• Structural Engineering and Mechanics
• /
• 제23권1호
• /
• pp.15-27
• /
• 2006

A locally varying temperature field or a mixture of two or more different materials can cause local variation of elasticity properties of a beam. In this paper, a new Euler-Bernoulli beam element with varying Young's modulus along its longitudinal axis is presented. The influence of axial forces according to the linearized 2nd order beam theory is considered, as well. The stiffness matrix of this element contains the transfer constants which depend on Young's modulus variation and on axial forces. Occurrence of the polynomial variation of Young's modulus has been assumed. Such approach can be also used for smooth local variation of Young's modulus. The critical loads of the straight slender columns were studied using the new beam element. The influence of position of the local Young's modulus variation and its type (such as linear, quadratic, etc.) on the critical load value and rate of convergence was investigated. The obtained results based on the new beam element were compared with ANSYS solutions, where the number of elements gradually increased. Our results show significant influence of the locally varying Young's modulus on the critical load value and the convergence rate.

• 한국산업융합학회 논문집
• /
• 제25권2_2호
• /
• pp.271-278
• /
• 2022

The solar energy has been widely used to reduce the fossil fuel and prevent the environmental pollution. The renewable energy including solar heat tends to spread due to carbon neutrality for main country of the world. Targets of solar collector are usually acquisitions of hot water or hot air. Especially, air-heating collector using solar heat is known as the technology for obtaining hot air. This study aims to investigate of characteristics of thermal flow when the hot air by air-heating collector using solar heat flows inside of indoor space. The thermal flow of heating indoor space was simulated using ANSYS-CFX program and thus the behaviors of hot air in indoor space were evaluated with standard k-𝜀 turbulence model. As the results, as the inlet velocity was increased, the behaviors of hot air became simple, and temperature range of 25~75℃ had almost no effect on behavior of flow. As the inlet temperature was increased, the temperature curve of indoor space from bottom to top was changed from linear to quadratic. Furthermore, it was confirmed that inlet velocity as well as inlet temperature also should be considered to heat indoor space equally by air-heating collector using solar heat.