• 대한의용생체공학회:의공학회지
• /
• 제36권5호
• /
• pp.211-220
• /
• 2015

Recently, model-based iterative reconstruction (MBIR) has played an important role in transmission tomography by significantly improving the quality of reconstructed images for low-dose scans. MBIR is based on the penalized-likelihood (PL) approach, where the penalty term (also known as the regularizer) stabilizes the unstable likelihood term, thereby suppressing the noise. In this work we further improve MBIR by using a more expressive regularizer which can restore the underlying image more accurately. Here we used a spline regularizer derived from a linear combination of the two-dimensional splines with first- and second-order spatial derivatives and applied it to a non-quadratic convex penalty function. To derive a PL algorithm with the spline regularizer, we used a separable paraboloidal surrogates algorithm for convex optimization. The experimental results demonstrate that our regularization method improves reconstruction accuracy in terms of both regional percentage error and contrast recovery coefficient by restoring smooth edges as well as sharp edges more accurately.

• 한국강구조학회 논문집
• /
• 제16권2호통권69호
• /
• pp.201-213
• /
• 2004

본 논문에서는 유전자 알고리즘과 보-기둥 접합부, 부재, 그리고 구조물 전체의 재료 및 기하학적 비선형 거동을 고려할 수 있는 개선소성힌지해석 방법을 접목시킨 평면 반강접 강골조 구조물의 최적설계법을 제안하였다. 개선소성힌지해석에서는 강골조 구조물의 기하학적 비선형성을 고려하기 위해 보-기둥 요소의 안정함수를 사용하였으며, 재료적 비선형성을 고려하기 위해 잔류응력, 소성힌지, 반강접 접합부 그리고 기하학적 불완전성 등에 의한 점진적인 강성감소모델을 사용하였다. 최적설계시 마이크로 유전자 알고리즘과 재생산을 위한 개체 선택 도구로 토너먼트 선택방법을 사용하였으며, 적합도 함수는 목적함수 및 벌칙함수로 나타낸 무제약 함수값의 조합으로 구성하였다. 목적함수로는 구조물의 중량을, 제약조건으로는 하중-저항능력, 사용성, 연성도, 그리고 시공성에 관한 기준을 사용하였다. 강접 및 반강접 접합부를 갖는 강골조 구조물의 최적설계결과의 비교를 통해 본 연구에서 제시한 방법의 적합성을 검증하였다.

• 산업경영시스템학회지
• /
• 제20권44호
• /
• pp.93-101
• /
• 1997

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 application of penalty function that can be adapted to transform a constrained optimization problem into an unconstrained one and premature convergence of solution. Thus, we developed an improved GAs to solve this problems, and two examples are given to demonstrate the effectiveness of the methodology developed in this paper.

• 전기학회논문지
• /
• 제57권9호
• /
• pp.1624-1627
• /
• 2008

This paper deals with a linear matrix inequality (LMI) approach to the design of a PID controller for an attraction type magnetic levitation system. First, we convert the $H_ {\infty}$ PID controller problem into a static output feedback problem. We then solve the static output problem by using the recently developed penalty function method. Numerical experiments show the effectiveness of the proposed algorithm.

• 대한전기학회논문지:전력기술부문A
• /
• 제49권7호
• /
• pp.339-344
• /
• 2000

This paper presents an improved optimal power flow algorithm, which solves an optimization problem with equality constraints with converted inequality constraints. The standard OPF and the penalty function method should do reconstructing active constraints among the inequality constraints so that the activation of the inequality constraints has been imposing an additional burden to solve OPF problem efficiently. However the proposed algorithm converts active inequality constraints into the equality constraints in order to preclude us from reconstructing the procedures. The effectiveness of the new OPF algorithm is validated by applying the IEEE 14 bus system.

• Journal of Information Processing Systems
• /
• 제15권6호
• /
• pp.1503-1516
• /
• 2019

In this paper, to improve the accuracy of long sentence similarity calculation, we proposed a sentence similarity calculation method based on a system similarity function. The algorithm uses word2vector as the system elements to calculate the sentence similarity. The higher accuracy of our algorithm is derived from two characteristics: one is the negative effect of penalty item, and the other is that sentence similar function (SSF) based on word2vector similar elements doesn't satisfy the exchange rule. In later studies, we found the time complexity of our algorithm depends on the process of calculating similar elements, so we build an index of potentially similar elements when training the word vector process. Finally, the experimental results show that our algorithm has higher accuracy than the word mover's distance (WMD), and has the least query time of three calculation methods of SSF.

• 한국강구조학회 논문집
• /
• 제12권5호통권48호
• /
• pp.515-525
• /
• 2000

본 연구의 목적은 평면 트러스의 단면치수 및 형 상 최적화 알고리즘을 개발 하는 것이다. 본 연구에 적용된 최적화 기법은 무제약 축차선형화기법(SUMT)의 extended penalty method와 다변수(多變數)를 가지는 직접탐사법인 Hooke & Jeeves method이다. 상기(上記) 언급된 두 방법은 단면최적화와 형상최적화 과정의 각 단계에서 반복적 다 최적설계 과정에서 최종의 최적값이 구해질 때까지 단면최적화와 형상최적화 과정이 반복적으로 상호작용하게 된다. 트러스의 형상최적화에 관한 기존의 연구문헌에서는 최적화의 설계변수로서 부재의 단면적과 절점좌표를 사용하였다. 이렇게 할 경우, 압축재의 좌굴특성을 제대로 반영하기 어려우므로 한정된 조건으로 좌굴특성을 부여하게 되어 보다 실제적인 최적설계에 많은 제약을 가져오게 되므로 본 연구에서는 부재의 실제치수와 절점좌표를 최적화의 설계 변수로 취하게 되므로 부재치수의 변화에 따른 좌굴특성의 변화를 최대한 반영할 수 있다.

• 제어로봇시스템학회논문지
• /
• 제4권1호
• /
• pp.68-75
• /
• 1998

The Intelligent Cruise Control system, ICC, is a driver assisting system for controlling relative speed and distance between two vehicles in the same lane. The ICC may be considered as an extension of a traditional cruise control, not only keeping a fixed speed of the vehicle, but correcting the speed also to that of a slower one ahead. This paper presents a real-time self-tuning fuzzy control algorithm to develop ICC. The self-tuning fuzzy control law is adopted to reduce the effects of nonlinearities of the vehicle and various road environments. In the self-tuning algorithm an interior penalty method is applied to preserve the inherent order of membership functions and is modified as an on-line algorithm for real time application. Via simulations, the performance of the suggested control algorithm is compared with a PID and a fuzzy control without self-tuning. The suggested control algorithm is implemented on PRV III and the results of the test driving on a local road are given.

• 한국공간구조학회논문집
• /
• 제4권4호
• /
• pp.91-98
• /
• 2004

유전자 알고리즘은 가장 훌륭한 이산최적화 기법 중 하나이다. 그러나, 유전자 알고리즘은 무제약 최적화 기법이기 때문에 제약조건은 간접적으로 표현된다. 가장 일반적인 방법은 벌칙함수를 사용하여 제약 문제를 무제약 문제로 변환하는 것이다. 본 연구에서는 적합도에 벌점함수를 적용하여 거부전략, 벌점전략, 복합전략 등에 따른 3가지 함수를 구성하였다. 그리고, 이 적합도 함수들을 사용한 설계프로그램을 구현하고, 산형골조와 2층 3경간 골조의 설계문제에 적용시켜 설계결과를 비교하였다. 이를 통하여 유전자 알고리즘을 이용한 유용한 골조 설계프로그램의 구현이 가능할 것으로 판단된다.

• Journal of Mechanical Science and Technology
• /
• 제14권3호
• /
• pp.306-313
• /
• 2000

The homogenization method and the density function method are common approaches to evaluate the equivalent material properties for design cells composed of matter and void. In this research, using a new topology optimization method based on the homogenized material with a penalty factor and the chessboard prevention strategy, we obtain the optimal layout of a structure for the natural frequency of a designated mode. The volume fraction of nodes of each finite element is chosen as the design variable and a total material usage constraint is imposed. In this paper, the subspace method is used to evaluate the eigenvalue and its corresponding eigenvector of the structure for the designated mode and the recursive quadratic programming algorithm, PLBA algorithm, is used to solve the topology optimization problem.