• 한국지능시스템학회논문지
• /
• 제12권5호
• /
• pp.451-461
• /
• 2002

차분 진화는 다양한 형태의 목적함수를 최적화하는데 매우 효율적인 방법임이 입증되었다 차분 진화의 가장 큰 이점은 개념적 단순성과 사용의 용이성이다. 그러나 차분 진화의 수렴성이 제어 파라미터에 매우 민감한 단점이 있다. 본 논문은 새로운 교배용 벡터 생성법과 제어 파라미터의 적응 메커니즘을 결합한 적응성 있는 차분 진화를 제안한다. 이는 수렴성을 해치지 않으면서 차분 진화를 보다 강인하게 만들며 사용이 쉽도록 해준다. 12가지 최적화 문제에 대해 제안한 방법을 시험하였다. 적응성 있는 차분 진화의 응용 사례로써 이벤트 예측을 위한 교사 클러스터링 방법을 제안한다. 이 방법을 진화에 의한 이벤트 클러스터링이라 부르며 데이터 모델링 검증에 널리 사용되는 4 가지 사례에 대해 그 성능을 시험하였다.

• 한국경영과학회지
• /
• 제32권2호
• /
• pp.109-121
• /
• 2007

This paper focuses on algorithms based on the evolution, which is applied to various optimization problems. Especially, among these algorithms based on the evolution, we investigate the simple genetic algorithm based on Darwin's evolution, the Lamarckian algorithm based on Lamark's evolution and the Baldwin algorithm based on the Baldwin effect and also Investigate the difference among them in the biological and engineering aspects. Finally, through this comparison, we suggest a new algorithm to find more various solutions changing the genotype or phenotype search space and show the performance of the proposed method. Conclusively, the proposed method showed superior performance to the previous method which was applied to the constrained minimum spanning tree problem and known as the best algorithm.

• Structural Engineering and Mechanics
• /
• 제26권2호
• /
• pp.191-210
• /
• 2007

The three different approaches (reformulations) of evolution strategies (ESs) have been proposed in the literature as extensions of the technique for solving discrete problems. This study implements an extensive research on application, evaluation and comparison of them in discrete optimum design of steel frames. A unified formulation is first developed to explain these approaches, so that differences and similarities between their inherent search mechanisms can clearly be identified. Two examples from practical design of steel frames are studied next to measure their performances in locating the optimum. Extensive numerical experimentations are performed in both examples to facilitate a statistical analysis of their convergence characteristics. The results obtained are presented in the histograms demonstrating the distribution of the best designs located by each approach. In addition, an average improvement of the best design during the course of evolution is plotted in each case to compare their relative convergence rates.

• Industrial Engineering and Management Systems
• /
• 제10권3호
• /
• pp.203-208
• /
• 2011

Job shop scheduling is well-known as one of the hardest combinatorial optimization problems and has been demonstrated to be NP-hard problem. In the past decades, several researchers have devoted their effort to develop evolutionary algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) for job shop scheduling problem. Differential Evolution (DE) algorithm is a more recent evolutionary algorithm which has been widely applied and shown its strength in many application areas. However, the applications of DE on scheduling problems are still limited. This paper proposes a one-stage differential evolution algorithm (1ST-DE) for job shop scheduling problem. The proposed algorithm employs random key representation and permutation of m-job repetition to generate active schedules. The performance of proposed method is evaluated on a set of benchmark problems and compared with results from an existing PSO algorithm. The numerical results demonstrated that the proposed algorithm is able to provide good solutions especially for the large size problems with relatively fast computing time.

• Nuclear Engineering and Technology
• /
• 제54권10호
• /
• pp.3864-3877
• /
• 2022

We propose a new approach called PESA (Prioritized replay Evolutionary and Swarm Algorithms) combining prioritized replay of reinforcement learning with hybrid evolutionary algorithms. PESA hybridizes different evolutionary and swarm algorithms such as particle swarm optimization, evolution strategies, simulated annealing, and differential evolution, with a modular approach to account for other algorithms. PESA hybridizes three algorithms by storing their solutions in a shared replay memory, then applying prioritized replay to redistribute data between the integral algorithms in frequent form based on their fitness and priority values, which significantly enhances sample diversity and algorithm exploration. Additionally, greedy replay is used implicitly to improve PESA exploitation close to the end of evolution. PESA features in balancing exploration and exploitation during search and the parallel computing result in an agnostic excellent performance over a wide range of experiments and problems presented in this work. PESA also shows very good scalability with number of processors in solving an expensive problem of optimizing nuclear fuel in nuclear power plants. PESA's competitive performance and modularity over all experiments allow it to join the family of evolutionary algorithms as a new hybrid algorithm; unleashing the power of parallel computing for expensive optimization.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2005년도 창립60주년 기념 추계 학술대회
• /
• pp.214.1-214.1
• /
• 2005

• Advances in Computational Design
• /
• 제2권4호
• /
• pp.313-331
• /
• 2017

In this study, teaching-learning based optimization (TLBO) is improved by incorporating model of multiple teachers, adaptive teaching factor, self-motivated learning, and learning through tutorial. Modified TLBO (MTLBO) is applied for simultaneous topology, shape, and size optimization of space and planar trusses to study its effectiveness. All the benchmark problems are subjected to stress, displacement, and kinematic stability constraints while design variables are discrete and continuous. Analyses of unacceptable and singular topologies are prohibited by seeing element connectivity through Grubler's criterion and the positive definiteness. Performance of MTLBO is compared to TLBO and state-of-the-art algorithms available in literature, such as a genetic algorithm (GA), improved GA, force method and GA, ant colony optimization, adaptive multi-population differential evolution, a firefly algorithm, group search optimization (GSO), improved GSO, and intelligent garbage can decision-making model evolution algorithm. It is observed that MTLBO has performed better or found nearly the same optimum solutions.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 하계종합학술대회논문집
• /
• pp.893-896
• /
• 1998

A new evolutionary programming technique using multiple subpopulations with completely differnt evolution mechanisms is propsed to solve the optimization problems. Three subpopulations, each has different evolution charcteristics and uses different EP algorithms such as SAEP, AEP and FEP, are cooperating with synergy effect in which it increases the possibility to quickly find the global optimum of continuous optimization problems. Subpopulations evolve in differnt manner and the interaction among these leads to global minimum quickly.

• KIEE International Transaction on Electrical Machinery and Energy Conversion Systems
• /
• 제11B권3호
• /
• pp.81-85
• /
• 2001

Processes for optimizing the coil shape of deflection yoke are proposed A very accurate and practical winding modeler is developed and volume integral equation method (VIEM) is used for field calculation. Two steps of optimizations are done by using (1+1) evolution strategy. Those are dimensional optimization and pin-position optimization Various techniques are applied for reducing computational time for the optimization.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제12권1호
• /
• pp.29-35
• /
• 2012

This paper introduces an adaptive control method of strong mutation rate and probability for queen-bee genetic algorithms. Although the queen-bee genetic algorithms have shown good performances, it had a critical problem that the strong mutation rate and probability should be selected by a trial and error method empirically. In order to solve this problem, we employed the measure of convergence and used it as a control parameter of those. Experimental results with four function optimization problems showed that our method was similar to or sometimes superior to the best result of empirical selections. This indicates that our method is very useful to practical optimization problems because it does not need time consuming trials.