• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.1117-1120
• /
• 2013

사람의 질병은 여러 요인의 복합적인 작용으로 발생하는데 이 중 유전적인 요인에는 유전자 간의 상호작용을 들 수 있다. 마이크로어레이(Microarray) 데이터로부터 유전자의 활성화 및 억제 관계를 밝히려는 다양한 시도는 계속되어왔다. 그러나 마이크로어레이 자체가 갖는 불안정성과 실험조건 수의 제약이 커다란 장애가 되어 왔다. 이에 생물학적 사전 지식을 포함하는 방법들이 제안되었다. 본 논문에서는 질병과 관련된 유전자 간의 상호작용의 집합을 질병 모듈이라 정의하고 이를 유전자 알고리즘으로 학습한 베이지안 네트워크(Bayesian network)로 추론하는 방법을 제안한다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.111-117
• /
• 2000

경제적 로트 일정계획 문제(Economic Lot Scheduling Problem : ELSP)는 지난 수십 여 년간 많은 연구가 이루어진 생산일정계획 문제 중의 한 분야이다. 이 문제는 NP-hard 문제이기 때문에 수많은 발견적 기법이 제안되고, 사용되어져 왔다. 그 중에서도 Dobson［1］의 발견적 기법이 그 수행도의 우수성으로 보아 최고의 기법으로 여겨지고 있는데, 본 연구에서는 Dobson［1］의 시변 로트 크기(time varying lot size) 접근방법에 유전자 알고리즘을 이용한 새로운 발견적 기법을 제안하고, 수치실험을 통해서 새로운 기법이 기존의 Dobson［1］의 기법보다 더 우수하다는 것을 보이고자 한다.

• Journal of the Korean Geotechnical Society
• /
• v.18 no.6
• /
• pp.117-127
• /
• 2002

A deterministic approach of slope stability, which is generally corresponding to the model of a simple non-linear function for slopes, is problematic in that it does not account the versatile characteristics of ground layers in an effective way. To resolve this problem, this study proposes a new way of analyzing slope stability, so-called “genetic algorithm method, ” so as to reflect some particular conditions pertaining to the grounds under concern. Similarities and differences in slope stability that may exist between homogeneous and multiple ground layers are examined in a competitive manner, Overall, though similarities deemed a little bit salient, the algorithm method turned out to be very applicable to estimating the validity of slope stability. Furthermore, an additional effort to consider long-standing sequential and dynamic changes in both the amount of rainfall and the underground water level is made in order to improve the results.

• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.669-680
• /
• 2001

Genetic Algorithms are optimization algorithm that mimics biological evolution to solve optimization problems. Genetic algorithms provide an alternative to traditional optimization techniques by using directed random searches to locate optimal solutions in complex fitness landscapes. Hybrid genetic algorithm that is combined with local search called learning can sustain the balance between exploration and exploitation. The genetic traits that each individual in the population learns through evolution are transferred back to the next generation, and when this learning is combined with genetic algorithm we can expect the improvement of the search speed. This paper proposes a genetic algorithm based Cellular Learning with accelerated learning capability for function optimization. Proposed Cellular Learning strategy is based on periodic and convergent behaviors in cellular automata, and on the theory of transmitting to offspring the knowledge and experience that organisms acquire in their lifetime. We compared the search efficiency of Cellular Learning strategy with those of Lamarckian and Baldwin Effect in hybrid genetic algorithm. We showed that the local improvement by cellular learning could enhance the global performance higher by evaluating their performance through the experiment of various test bed functions and also showed that proposed learning strategy could find out the better global optima than conventional method.

• Composites Research
• /
• v.35 no.5
• /
• pp.340-346
• /
• 2022

In this study, an optimal structural design of composite helicopter blades is performed using the genetic algorithm-based optimizer PSGA (Particle Swarm assisted Genetic Algorithm). The blade sections consist of the skin, spar, form, and balancing weight. The sectional geometries are generated using the B-spline curves while an opensource code Gmsh is used to discretize each material domain which is then analyzed by a finite element sectional analysis program Ksec2d. The HART II blade formed based on either C- or D-spar configuration is exploited to verify the cross-sectional design framework. A numerical simulation shows that each spar model reduces the blade mass by 7.39% and 6.65%, respectively, as compared with the baseline HART II blade case, while the shear center locations being remain close (within 5% chord) to the quarter chord line for both cases. The effectiveness of the present optimal structural design framework is demonstrated, which can readily be applied for the structural design of composite helicopter blades.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2003.11a
• /
• pp.109-115
• /
• 2003

This study intends to provide an optimum design of flexible matrix composite driveshafts using a genetic algorithm. An objective function is defined as a combination of shaft flexibility, whirling stability and torsional buckling and the design variables are selected as ply angles and the shaft thickness. Results show that the genetic algorithm can successfully find an optimum solution at which the overall performance of the FMC shafts is significantly enhanced

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.696-702
• /
• 2006

Flutter characteristics of composite curved wing were investigated in this study. The efficient and robust system for the flutter optimization of general composite curved wing models has been developed using the coupled computational method based on both the standard genetic algorithm and the micro genetic algorithms. Micro genetic algorithm is used as an alternative method to overcome the relatively poor exploitation characteristics of the standard genetic algorithm. The present results show that the micro genetic algorithm is more efficient in order to find optimized lay-ups for a composite curved wing model. It is found that the flutter stability of curved wing model can be significantly increased using composite materials with proper optimum lamination design when compared to the case of isotropic wing model under the same weight condition.

• Journal of Digital Convergence
• /
• v.18 no.12
• /
• pp.481-488
• /
• 2020

Symbolic regression is an analysis method that directly generates a function that can explain the relationsip between dependent and independent variables for a given data in regression analysis. Genetic Programming is the leading technology of research in this field. It has the advantage of being able to directly derive a model that can be interpreted compared to other regression analysis algorithms that seek to optimize parameters from a fixed model. In this study, we propse a symbolic regression algorithm using parallel genetic programming based on a coarse grained parallel model, and apply the proposed algorithm to PMLB data to analyze the effectiveness of the algorithm.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.41 no.4
• /
• pp.275-283
• /
• 2013

In this paper, the structural optimum design method of composite rotor blade cross-section was investigated with the genetic algorithm. An auto-mesh generation program was developed for iterative calculations of optimum design, and stresses in the blade cross-section were analyzed by VABS (variational asymptotic beam sectional analysis) program. Minimum mass of rotor blade was defined as an object function, and stress failure index, center mass and blade minimum mass per unit length were chosen as constraints. Finally, design parameters such as the thickness and layup angles of a skin, and the thickness, position and width of a torsion box were determined through the structural optimum design method of composite rotor blade cross-section presented in this paper.

• Proceedings of the Korean Society Of Semiconductor Equipment Technology
• /
• 2005.05a
• /
• pp.151-156
• /
• 2005

본 논문에서는 비선형 역학 시스템에서 복합적 지능 알고리즘을 이용하여 시스템의 제어 성능을 개선시키는 방법에 대하여 논의하였다. 기존의 비선형 제어를 위한 뉴럴 네트워크 및 유전자 알고리즘은 학습이 종료된 후에 고정된 상태에서는 훌륭한 제어를 보여주지만, 환경 변화와 같은 변이 인자가 발생되면 제어가 제대로 되지 않으며 재학습을 해야만 한다. 이때 재학습에 드는 시간이 많이 걸리는 문제점이 있다. 제안하는 시스템에서는 변이 인자가 발생되었을 때의 상황을 항원으로 하는 면역 시스템을 기존 제어시스템에 추가하여 사용함으로써 보다 안정적이며 빠른 제어 수행이 가능함을 보일 것이다. 또한 기존에 하드웨어로 구성하기 어려웠던 유전 알고리즘을 하드웨어로 구성하기 쉽도록 유전 인자를 메모리 주소로 하는 알고리즘을 만들게 되었다.