• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.101.4-101
• /
• 2001

Evolutionary algorithm has been receiving a remarkable attention due to the model-free and population-based parallel search attributes and much successful results are coming out. However, there are some problems in most of the evolutionary algorithms. The critical one is that it takes much time or large generations to search the global optimum in case of the objective function with multimodality. Another problem is that it usually cannot search all the local optima because it pays great attention to the search of the global optimum. In addition, if the objective function has several global optima, it may be very difficult to search all the global optima due to the global characteristics of the selection methods. To cope with these problems, at first we propose a preprocessing process, grid-filtering algorithm(GFA), and propose a new distributed evolutionary ...

• Journal of Korean Society of Transportation
• /
• v.16 no.1
• /
• pp.151-164
• /
• 1998

본 논문은 통행배정과 교통신호제어기의 결합문제를 풀기 위한 새로운 해법의 제시를 목적으로 한다. 통행배정과 신호제어 결합모형은 네트웍 디자인 문제(Network Design Problem)로 비선형 비분리 목적함수(Nonlinear and Nonseparable Objective Function)와 비선형제약 및 비컴백스 집합(Nonlinear and Non-Convex Set)형태로 인해 다수의 국지해(Multiple Local Optima)를 갖는 특징이 있다. 따라서 이렇게 복잡하고 난해한 문제를 푸는 해법은 많은 국지해중에 가장 최소한 값(Global Optima)을 찾을수 있는 방법을 제공하여야한다. 전체최적해(Global Optima)를 찾 을 수 있는 기존의 방법들은 확률적최적화방법(Stochastic Optimization Methods)에 속한다. 본연구에서는 이러한 방법중 금속공학에서 발 견된 모의담금빌법(Simulated Annealing Method)에 근거한 해법을 제시한다. 이방법이 통행배정과 신호제어 결합문제에 적용되는지 검토하기 위해 이해법의 수렴성(Convergence)을 증명했으며 또한 실제 프로그램된 모형을 작은 고안된 네트워크에 적 용했다. 마지막으로는 개발된 해법의 실용성을 실험하기 위해 두 가지의 보다 큰 도로망에 적용 및 분석을 했다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.5
• /
• pp.52-65
• /
• 1997

For the degradation of severe noise and ill-conditioned blur the optimization function has the solution spaces which have many local optima around global solution. General restoration methods such as inverse filtering or gradient methods are mainly dependent on the properties of degradation model and tend to be isolated into a local optima because their convergences are determined in the convex space. Hence we introduce genetic algorithm as a searching method which will search solutions beyond the convex spaces including local solutins. In this paper we introudce improved evaluation square error) and fitness value for gray scaled images. Finally we also proposed the local fine tunign of window size and visit number for delicate searching mechanism in the vicinity of th global solution. Through the experiental results we verified the effectiveness of the proposed genetic operators and evaluation function on noise reduction over the conventional ones, as well as the improved performance of local fine tuning.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.12
• /
• pp.1210-1218
• /
• 2011

GA (Genetic Algorithms) are efficient for searching for global optima but may have some problems such as premature convergence, convergence to local extremum and divergence. These phenomena are related to the evolutionary operators. As population diversity converges to low value, the search ability of a GA decreases and premature convergence or converging to local extremum may occur but population diversity converges to high value, then genetic algorithm may diverge. To guarantee that genetic algorithms converge to the global optima, the genetic operators should be chosen properly. In this paper, we analyze the effects of the selection operator, crossover operator, and mutation operator on convergence properties, and propose the sweeping method of mutation probability and elitist propagation rate to maintain the diversity of the GA's population for getting out of the premature convergence. Results of simulation studies verify the feasibility of using these sweeping operators to avoid premature convergence and convergence to local extrema.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.05a
• /
• pp.134-137
• /
• 2000

Genetic algorithms(GAs) have been widely used as a method to solve optimization problems. This is because GAs have simple and elegant tools with reproduction, crossover, and mutation to rapidly discover good solutions for difficult high-dimensional problems. They, however, do not guarantee the convergence of global optima in GA-hard problems such as deceptive problems. Therefore we proposed a Schema Co-Evolutionary Algorithm(SCEA) and derived extended schema 76988theorem from it. Using co-evolution between the first population made up of the candidates of solution and the second population consisting of a set of schemata, the SCEA works better and converges on global optima more rapidly than GAs. In this paper, we show advantages and efficiency of the SCEA by applying it to some problems.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.9-11
• /
• 1999

조합 최적화 문제인 Traveling Salesman problems(TSP)을 Genetic Algorithm(GA)과 Local Search Heuristic인 Lin-Kernighan(LK) Heuristic[2]을 이용하여 접근하는 것은 최적해를 구하기 위해 널리 알려진 방법이다. 이 논문에서는 LK를 이용하여 주어진 TSP 문제에서 Local Optima를 찾고, GA를 이용하여 Local Optimal를 바탕으로 Global Optima를 찾는데 이용하게 된다. 여기서 이런 GA와 LK를 이용하여 TSP 문제를 풀 경우 해가 점점 수렴해가면서 중복된 유전자가 많이 생성된다. 이런 중복된 유전자를 제거함으로써 탐색의 범위를 보다 넓고 다양하게 검색하고, 더욱 효율적으로 최적화를 찾아내는 방법에 대해서 논하겠다. 이런 방법을 이용하여 rat195, gil262, lin318의 TSP문제에서 효율적으로 수행된다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.3
• /
• pp.587-592
• /
• 2018

An adaptive beam forming system using a phased array antenna improves communication quality by beam forming adaptively to a communication environment having an interference signal. For adaptive beam forming, a good combination of the phases of the excited signals to each radiating element of the phased array antenna should be calculated. In this paper, improved particle swarm optimization algorithm that adds a re-spreading procedure according to particle density was proposed to increase the probability of good phase shift combination output.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.4
• /
• pp.281-285
• /
• 2005

Learning and evolving are two basics for data mining. As compared with classical learning theory based on objective function with minimizing training errors, the recently evolutionary computing has had an efficient approach for constructing optimal model without the minimizing training errors. The global search of evolutionary computing in solution space can settle the local optima problems of learning models. In this research, combining co-evolving algorithm into statistical learning theory, we propose an co-evolutionary computing for statistical learning theory for overcoming local optima problems of statistical learning theory. We apply proposed model to classification and prediction problems of the learning. In the experimental results, we verify the improved performance of our model using the data sets from UCI machine learning repository and KDD Cup 2000.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.321-324
• /
• 2000

조합 최적화 문제인 Traveling Salesman problems(TSP)을 Genetic Algorithm(GA)[3]과 Local Search Heuristic Algorithm[8]을 이용하여 접근하는 것은 최적해를 구하기 위해 널리 알려진 방법이다. 본 논문에서는 TSP문제를 해결하기 위한 또 다른 접근법으로, 다수의 Ant들이 Tour들을 찾는 ACS(Ant Colony System) Algorithms[4][6][7]을 소개하고, ACS에서 Global Optima를 찾는 과정에서, 이미 이루어져 있는 Ant들의 Tour결과들을 서로 비교한다. Global Updating Rule에 의해 Global Best Tour 에 속해 있는 각 Ant Tour의 edge들을 update하는 ACS Algorithm에, 각 루프마다 Ant Tour들을 우성과 열성 인자들로 구분하고, 각각의 우성과 열성 인자들에 대해서 Global Updating Rule에 기반한 가중치를 적용(Weight Updating Rule)하므로서 기존의 ACS Algorithm보다 효율적으로 최적 해를 찾아내는 방법에 대해서 논하고자 한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.169.4-169
• /
• 2001

Conventional methods to solve the nonlinear programming problem range from augmented Lagrangian methods to sequential quadratic programming (SQP) methods. NPSOL, which is a SQP code, has been widely used to solve various optimization problems but is still subject to many numerical problems such as convergence to local optima, difficulties in initialization and in handling non-smooth cost functions. Recently, many evolutionary methods have been developed for constrained optimization. Among them, CEALM (Co-Evolutionary Augmented Lagrangian Method) shows excellent performance in the following aspects: global optimization capability, low sensitivity to the initial parameter guessing, and excellent constraint handling capability due to the benefit of the augmented Lagrangian function. This algorithm is ...