• 한국CDE학회논문집
• /
• 제9권3호
• /
• pp.183-191
• /
• 2004

This paper deals with smooth curve fitting of data corrupt by noise. Most research efforts have been concentrated on employing the smoothness penalty function with the estimation of its optimal parameter in order to avoid the 'overfilling and underfitting' dilemma in noisy data fitting problems. Our approach, called DBSF(Differentiation-Based Smooth Fitting), is different from the above-mentioned method. The main idea is that optimal functions approximately estimating the derivative of noisy curve data are generated first using genetic programming, and then their integral values are evaluated and used to recover the original curve form. To show the effectiveness of this approach, DBSP is demonstrated by presenting two illustrative examples and the application of estimating the principal dimensions of bulk cargo ships in the conceptual design stage.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1996년도 가을 학술발표회 논문집
• /
• pp.168-175
• /
• 1996

This paper deals with the variable-node element for fluid flow and the adaptive h-version mesh refinement algorithm. The transient element has been formulated by the Galerkin approach in which the pressure term is replaced with the penalty function. The present element having variable mid-side node and is suitable for constructing a locally refined mesh avoiding the use of the highly distorted elements. A modified Gauss quadrature is needed to integrate the element matrices to solve the trouble associated with the discontinuity of derivatives of shape functions. Several numerical examples show that the proposed element can be effectively used in the h-version adapt ive mesh refinement

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2004년도 춘계학술대회 학술발표 논문집 제14권 제1호
• /
• pp.462-465
• /
• 2004

수치적인 데이터를 분류하기 위한 대표적인 방법은 퍼지 규칙을 사용하는 것이다. 하지만, 이러한 방법은 퍼지 소속 함수를 어떻게 정의하느냐에 따라 퍼지 분류의 성능이 크게 영향을 받는다는 문제점과 퍼지 규칙을 쉽게 이해하기 위해 가능한 퍼지 규칙의 수를 적게 유지해야한다는 문제점이 있다. 본 논문에서는 효과적이며 이해하기 쉬운 퍼지 규칙을 생성하기 위해 기울기 강하법을 기반으로 하는 소속 함수 학습 방법을 제안한다. 에러율을 감소하기 위해 Penalty 연산과 Reward 연산을 통해 소속 함수가 반복적으로 조절된다. 새로운 소속 함수는 Coverage 연산에 의해 생성된다. 또한 이해하기 쉬운 퍼지 규칙을 최적화하기 위해 학습된 소속 함수를 퍼지 결정 트리에 적용한다. 본 논문에서 제안한 알고리즘의 타당성을 확인하기 위해 벤치 마크 데이터인 Iris, Wisconsin Breast Cancer, Pima. Bupa 데이터를 이용하여 실험 결과를 보인다. 실험 결과를 통해 제안한 알고리즘이 기존의 C4.5와 FID 3.1 알고리즘보다 더 효과적이거나 비슷한 성능을 보임을 알 수 있다.

• 한국통신학회논문지
• /
• 제31권8A호
• /
• pp.735-744
• /
• 2006

In this paper, the optimum pump light powers of optical phase conjugator(OPC) are numerically investigated as a function of amplifier spacing in 1,200 km $8{\times}40$ Gbps WDM systems with 0.1, 0.4, 0.8, or 1.6 ps/nm/km dispersion coefficient. It is confirmed that the variation of optimal pump light power dependence on amplifier spacing for NRZ transmission system is smaller than that for RZ transmission system through the evaluations and analysis of eye opening penalty(EOP) characteristics. And, in both cases of NRZ and RZ transmission, the variation of optimal pump light power is more increased as amplifier spacing becomes longer. Additionally, it is confirmed that the best amplifier spacing in NRZ and RZ transmission system is 50 km.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.44.2-44
• /
• 2001

This paper presents a new concept for controlling of under actuated robot manipulators with avoiding obstacles using switching computed torque method (SCTM). One fundamental approach of this algorithm is to use the partly stable controllers (PSCs) in order to fulfill the ultimate control objective. Here, we use genetic algorithms (GA)in order to employ the optimum control action for a given time frame with the available set of elemental controllers, depending on which links/variables are controlled, i.e. the selection of optimum switching sequence of the control actions. The proposed approach models links of the robot using evolving ellipses and then introduces a penalty scheme for the objective function of GA when it detects collisions. An under actuated robot manipulator, which has three detrees-of-freedom is taken into consideration so as to illustrate the design procedure. Simulation results show the e.ectiveness of the proposed method.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2003년도 춘계공동학술대회
• /
• pp.8-13
• /
• 2003

This paper considers a continuous-review two-echelon inventory control problem with one-to-one replenishment policy incorporated and with lost sales allowed where demand arrives In a stationary Poisson process The problem Is formulated using METRIC-approximation in a combined approach of pricing and (S-1.S) Inventory policy, for which an iterative solution algorithm is derived with respect to the corresponding one-warehouse multi-retailor supply chain. Specifically, decisions on retail pricing and warehouse inventory policies are made in integration to maximize total profit in the supply chain. The objective function of the model consists of sub-functions of revenue and cost (holding cost and penalty cost). To test the effectiveness and efficiency of the proposed algorithm, numerical experiments are performed The computational results show that the proposed algorithm is efficient and derives quite good decisions

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1996년도 춘계공동학술대회논문집; 공군사관학교, 청주; 26-27 Apr. 1996
• /
• pp.519-524
• /
• 1996

This paper describes the application of Genetic Algorithms(GAs) to nonlinear constrained mixed optimization problems. Genetic Algorithms are combinatorial in nature, and therefore are computationally suitable for treating discrete and integer design variables. But, several problems that conventional GAs are ill defined are applicaiton of penalty function that can be adapted to transform a constrained optimization problem into an unconstrained optimization problem into an unconstrained one and premature convergence of solution. Thus, we developed an modified GAs to solve this problems, and two examples are given to demonstrate the effectiveness of the methodology developed in this paper.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 춘계학술대회 논문집
• /
• pp.516-520
• /
• 1996

This paper focues on development of a timer selection algorithm for IEEE802.4 token bus networks to serve large-scale integrated manufacturing systems. Such factory automation networks have to satisfy delay constraints imposed on time-critical messages while maintaining as much network capacity as possible for non-time-critical messages. Therefore, the object of this paper is to develop timer selection algorithm to minimize a certain penalty function. This paper presents an algorithm based on Genetic Algorithm. The efficacy of the algorithm has been demonstrated by a series of simulation experiments.

• Asia pacific journal of information systems
• /
• 제16권4호
• /
• pp.1-25
• /
• 2006

This research approached the negotiation problem from the seller's perspective in the make-to-order manufacturing industry in order to solve the multi-attribute automated negotiation problem. To this end, more than 'two negotiation attributes are defined and a preferred value of each negotiation alternative is calculated based on the entropy value of each attribute. And then, the best option is selected according to the utility function of each negotiator, a penalty of negotiation delay, and conviction of possibility of negotiation success. The suggested automated negotiation methods in this study can be employed for the development of advanced automated negotiation systems.

• 대한기계학회논문집
• /
• 제19권3호
• /
• pp.713-723
• /
• 1995

In intelligent machine tools, a computer based control system, which can adapt the machining parameters in an optimal fashion based on sensor measurements of the machining process, should be incorporated. In this paper, the technology for adaptively optimizing the cutting conditions to maximize the material removal rate in face milling operations is proposed using the exterior penalty function method combined with multilayered neural networks. Two neural networks are introduced ; one for estimating tool were length, the other for mapping input and output relations from experimental data. Then, the optimization of cutting conditions is adaptively implemented using tool were information and predicted process output. The results are demonstrated with respect to each level of machining such as rough, fine and finish cutting.