• 한국전산구조공학회논문집
• /
• 제18권1호
• /
• pp.51-59
• /
• 2005

SLSV(single loop single vector)방법은 신뢰성기반 최적설계(reliability-based design optimization, RBDO)에서 중첩된 반복과정을 제거함으로써 최적설계의 과도한 계산비용 문제에 대한 해결책을 제시하고 있지만, 종종 수렴하지 못하거나 잘못된 해가 얻어지는 등의 불안정성, 부정확성 문제를 가지고 있어 그 활용이 제한적이다. 본 논문에서는 수정된 HMV(hybrid mean value)방법, Inactive Design, Active MPP(most probable point) Design의 적용을 통해 SLSV방법에 있어서 안정성과 효율성을 효과적으로 개선시킬 수 있는 수정된 SLSV방법을 제안하였고 또한 다양한 예제를 통해 수정된 SLSV방법의 유용성을 검증하였다.

• 대한기계학회논문집A
• /
• 제27권11호
• /
• pp.1809-1814
• /
• 2003

Differential evolution (DE) algorithm is presented and applied to global optimization in this research. DE suggested initially fur the solution to Chebychev polynomial fitting problem is similar to genetic algorithm(GA) including crossover, mutation and selection process. However, differential evolution algorithm is simpler than GA because it uses a vector concept in populating process. And DE turns out to be converged faster than CA, since it employs the difference information as pseudo-sensitivity In this paper, a trial vector and its control parameters of DE are examined and unconstrained optimization problems of highly nonlinear multimodal functions are demonstrated. To illustrate the efficiency of DE, convergence rates and robustness of global optimization algorithms are compared with those of simple GA.

• Journal of the Korean Data and Information Science Society
• /
• 제25권2호
• /
• pp.447-454
• /
• 2014

In many practical machine learning and data mining applications, unlabeled training examples are readily available but labeled ones are fairly expensive to obtain. Therefore semi-supervised learning algorithms have attracted much attentions. However, previous research mainly focuses on classication problems. In this paper, a semi-supervised regression method based on support vector regression (SVR) formulation that is proposed. The estimator is easily obtained via the dual formulation of the optimization problem. The experimental results with simulated and real data suggest superior performance of the our proposed method compared with standard SVR.

• 한국지능시스템학회논문지
• /
• 제20권3호
• /
• pp.375-379
• /
• 2010

Support vector machine (SVM)은 튼튼한 이론적 배경을 가지고 있고 구조적 위험을 성공적으로 최소화하기 때문에 추천가 시스템과 같은 다양한 패턴 인식 분야에서 사용되고 있다. 하지만 SVM이 초평면을 결정할 때 이상점들은 margin 손실들을 가지고 있기 때문에 이들은 초평면을 결정하는데 매우 중요한 역할을 하고 있다. 그 이유로 SVM은 이상점들에게 매우 민감한 문제점을 갖는다. 강인한 SVM을 위해 우리는 이상점들의 margin 손실의 최대치를 제한하지만 이것은 non-convex 최적화 문제를 포함한다. 따라서 본 논문에서는 non-convex 최적화 문제에 적합한 유전자 알고리즘을 이용하여 강인한 SVM을 설계하는 방법을 제안한다. 제안하는 알고리즘의 우수성을 보여주기 위하여 UCI repository에서 선택된 여러 데이터베이스들을 이용한 실험을 수행하였다.

• 대한수학회논문집
• /
• 제25권1호
• /
• pp.139-147
• /
• 2010

A generalized nondifferentiable fractional optimization problem (GFP), which consists of a maximum objective function defined by finite fractional functions with differentiable functions and support functions, and a constraint set defined by differentiable functions, is considered. Recently, Kim et al. [Journal of Optimization Theory and Applications 129 (2006), no. 1, 131-146] proved optimality theorems and duality theorems for a nondifferentiable multiobjective fractional programming problem (MFP), which consists of a vector-valued function whose components are fractional functions with differentiable functions and support functions, and a constraint set defined by differentiable functions. In fact if $\overline{x}$ is a solution of (GFP), then $\overline{x}$ is a weakly efficient solution of (MFP), but the converse may not be true. So, it seems to be not trivial that we apply the approach of Kim et al. to (GFP). However, modifying their approach, we obtain optimality conditions and duality results for (GFP).

• Journal of Power Electronics
• /
• 제16권3호
• /
• pp.1097-1109
• /
• 2016

Fault detection and isolation are related to system monitoring, identifying when a fault has occurred, and determining the type of fault and its location. Fault detection is utilized to determine whether a problem has occurred within a certain channel or area of operation. Fault detection and diagnosis have become increasingly important for many technical processes in the development of safe and efficient advanced systems for supervision. This paper presents an integrated technique for fault diagnosis and classification for open- and short-circuit faults in three-phase inverter circuits. Discrete wavelet transform and principal component analysis are utilized to detect the discontinuity in currents caused by a fault. The features of fault diagnosis are then extracted. A fault dictionary is used to acquire details about transistor faults and the corresponding fault identification. Fault classification is performed with a fuzzy logic system and relevance vector machine (RVM). The proposed model is incorporated with a set of optimization techniques, namely, evolutionary particle swarm optimization (EPSO) and cuckoo search optimization (CSO), to improve fault detection. The combination of optimization techniques with classification techniques is analyzed. Experimental results confirm that the combination of CSO with RVM yields better results than the combinations of CSO with fuzzy logic system, EPSO with RVM, and EPSO with fuzzy logic system.

• 한국해양공학회지
• /
• 제14권2호
• /
• pp.121-127
• /
• 2000

In this paper a new technique of inverse fairing problem for ship hull is proposed. Recently Lu solved the inverse fairing problem for automobile's body that was made by one surface element. In this system however hull surface is constructed by Gregory's composite surface interpolation method. So reflection line at boundary position is used as a tool of solving inverse problem in surface fairing. But the results are not good. The new concepts of Normal vector line and Constrained reflection line are introduced as an alternative tool. Energy minimization method for Normal Vector Line curve net and the inverse method for Constrained Reflection Line by using optimization technique are examined And the final lines from this proposed surface fairing method shows good fairness.

• Journal of the Korean Statistical Society
• /
• 제33권1호
• /
• pp.25-34
• /
• 2004

In this paper we propose a prediction method on the regression model with randomly censored observations of the training data set. The least squares support vector machine regression is applied for the regression function prediction by incorporating the weights assessed upon each observation in the optimization problem. Numerical examples are given to show the performance of the proposed prediction method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제3권1호
• /
• pp.100-104
• /
• 2003

The SVDD(support vector data description) is one of the most well-known one-class support vector learning methods, in which one tries the strategy of utilizing balls defined on the kernel feature space in order to distinguish a set of normal data from all other possible abnormal objects. The major concern of this paper is to consider the problem of modifying the SVDD into the direction of utilizing ellipsoids instead of balls in order to enable better classification performance. After a brief review about the original SVDD method, this paper establishes a new method utilizing ellipsoids in feature space, and presents a solution in the form of SDP(semi-definite programming) which is an optimization problem based on linear matrix inequalities.

• Journal of Electrical Engineering and Technology
• /
• 제8권5호
• /
• pp.991-1001
• /
• 2013

Instead of building new substations or transmission lines, proper installation of flexible AC transmission systems (FACTS) devices can make the transmission networks accommodate more power transfers with less expansion cost. In this paper, the problem to maximize power system loadability by optimally installing two types of FACTS devices, namely static var compensator (SVC) and thyristor controlled series compensator (TCSC), is formulated as a mixed discrete-continuous nonlinear optimization problem (MDCP). To reduce the complexity of the problem, the locations suitable for SVC and TCSC installations are first investigated with tangent vector technique and real power flow performance index (PI) sensitivity factor and, with the specified locations for SVC and TCSC installations, a set of schemes is formed. For each scheme with the specific locations for SVC and TCSC installations, the MDCP is reduced to a continuous nonlinear optimization problem and the computing efficiency can be largely improved. Finally, to cope with the technical and economic concerns simultaneously, the scheme with the biggest utilization index value is recommended. The IEEE-14 bus system and a practical power system are used to validate the proposed method.