• Smart Structures and Systems
• /
• 제15권3호
• /
• pp.735-749
• /
• 2015

The major difficulty of using Bayesian probabilistic inference for system identification is to obtain the posterior probability density of parameters conditioned by the measured response. The posterior density of structural parameters indicates how plausible each model is when considering the uncertainty of prediction errors. The Markov chain Monte Carlo (MCMC) method is a widespread medium for posterior inference but its convergence is often slow. The differential evolution adaptive Metropolis-Hasting (DREAM) algorithm boasts a population-based mechanism, which nms multiple different Markov chains simultaneously, and a global optimum exploration ability. This paper proposes an improved differential evolution adaptive Metropolis-Hasting algorithm (IDREAM) strategy to estimate the posterior density of structural parameters. The main benefit of IDREAM is its efficient MCMC simulation through its use of the adaptive Metropolis (AM) method with a mutation strategy for ensuring quick convergence and robust solutions. Its effectiveness was demonstrated in simulations on identifying the structural parameters with limited output data and noise polluted measurements.

• 한국컴퓨터정보학회논문지
• /
• 제17권4호
• /
• pp.129-137
• /
• 2012

여왕벌의 생식방식을 모방하여 만든 여왕벌 유전자알고리즘은 유전자알고리즘의 성능을 대폭 향상시켰다. 그러나 여왕벌 유전자알고리즘에서는 여왕벌을 하나만사용하여 진화를 수행함으로서 개체들이 지나치게 해당 여왕벌이 있는 쪽으로 몰리는 문제를 발생하였으며 이는 결국 유전자 알고리즘의 성능저하를 가져왔다. 본 논문에서는 이러한 문제를 해결하고자 각 세대에서 가장 적합도가 좋은 여왕벌과 더불어 개체의 적합도가 부모 개체에 비하여 가장 크게 증가한 두 번째 여왕벌을 도입한 다중 여왕벌 진화 알고리즘을 제안한다. 다중 여왕벌을 도입함으로서 개체가 지역 최적해에 빠질 가능성이 줄어들고 지역 최적해에 빠진 경우에도 보다 쉽게 지역 최적해를 빠져나올 수 있게 되어 성능향상이 가능하였다. 4개의 함수최적화 문제에 적용시켜본 결과 본 논문에서 제안한 방법이 기존의 방법보다 대부분의 경우에서 성능이 향상됨을 볼 수 있었다.

• 한국지능시스템학회논문지
• /
• 제15권3호
• /
• pp.375-381
• /
• 2005

일반적으로 유전자 알고리즘은 최적 시스템을 디자인하는데 주로 이용된다. 하지만 알고리즘의 성능은 적합도 함수나 시스템 환경에 의해 결정된다. 두 개의 개체군이 꾸준히 상호작용하고 공진화 하는 공진화 알고리즘은 이러한 문제를 극복할 수 있을 것으로 기대된다. 본 논문에서는 GA가 풀기 어려운 GA-hard problem을 풀기 위하여 저자가 제안한 3가지 공진화 모델을 설명한다. 첫 번째 모델은 찾고자하는 해와 환경을 각각 경쟁하는 개체군으로 구성해 진화하는 방법으로 사용자의 환경설정에 의해 지역적 해를 찾는 것을 방지하는 경쟁적 공진화 알고리즘이다. 두 번째 모델은 호스트 개체군과 기생(스키마) 개체군으로 구성된 스키마 공진화 알고리즘이다. 이 알고리즘에서 스키마 개체군은 호스트 개체군에 좋은 스키마를 공급한다. 세 번째 알고리즘은 두 개체군이 서로 게임을 통해 진화하도록 하는 게임이론에 기반한 공진화 알고리즘이다. 각 알고리즘은 비주얼 서보잉, 로봇 주행, 다목적 최적화 문제에 적용하여 그 유효성을 입증한다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.111.3-111
• /
• 2002

$\textbullet$introduction $\textbullet$CO-evolution Algorithm $\textbullet$Subsumption Architecture $\textbullet$Neural Network $\textbullet$Khepera Robot

• 한국공간구조학회논문집
• /
• 제20권1호
• /
• pp.95-102
• /
• 2020

The structural design of arch roofs or bridges requires the analysis of their unstable behaviors depending on certain parameters defined in the arch shape. Their maintenance should estimate the parameters from observed data. However, since the critical parameters exist in the equilibrium paths of the arch, and a small change in such the parameters causes a significant change in their behaviors. Thus, estimation to find the critical ones should be carried out using a global search algorithm. In this paper we study the parameter estimation for a shallow arch by a quantum-inspired evolution algorithm. A cost functional to estimate the system parameters included in the arch consists of the difference between the observed signal and the estimated signal of the arch system. The design variables are shape, external load and damping constant in the arch system. We provide theoretical and numerical examples for estimation of the parameters from both contaminated data and pure data.

• 한국실내디자인학회논문집
• /
• 제15권1호
• /
• pp.30-38
• /
• 2006

The new aspect of contemporary architectural design is the computer simulation of morphogenesis and evolution of the organic body. Morphogenesis and evolution is the kind of emergence that is the process of complex pattern formation from simpler rules in complex system. The development comprises the sequence of pattern formation, differentiation, morphogenesis, growth. This study analyzes the application methodology of various biomorphism in contemporary architecture. The methods of generative application by computation in architecture are self-organization, differentiation, growth algorithm via MoSS. And the methods of evolution by computation are genetic algorithm, multi-parameter in environments, phylogenetic cross-over, competing as natural selection, mutation+external constraints, generative algorithm+genetic algorithm via Genr8.

• International Journal of Control, Automation, and Systems
• /
• 제6권2호
• /
• pp.282-287
• /
• 2008

In distributed mobile robot systems, autonomous robots accomplish complicated tasks through intelligent cooperation with each other. This paper presents behavior learning and online distributed evolution for cooperative behavior of a group of autonomous robots. Learning and evolution capabilities are essential for a group of autonomous robots to adapt to unstructured environments. Behavior learning finds an optimal state-action mapping of a robot for a given operating condition. In behavior learning, a Q-learning algorithm is modified to handle delayed rewards in the distributed robot systems. A group of robots implements cooperative behaviors through communication with other robots. Individual robots improve the state-action mapping through online evolution with the crossover operator based on the Q-values and their update frequencies. A cooperative material search problem demonstrated the effectiveness of the proposed behavior learning and online distributed evolution method for implementing cooperative behavior of a group of autonomous mobile robots.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.173-176
• /
• 2001

This paper proposes training algorithms of neuro-fuzzy systems. First, we introduce a structure training algorithm, which produces the necessary number of hidden nodes from training data. From this algorithm, initial fuzzy rules are also obtained. Second, the parameter training algorithm using evolution strategy is introduced. In order to show their usefulness, we apply our neuro-fuzzy system to a nonlinear system identification problem. It was found from experiments that proposed training algorithms works well.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권11호
• /
• pp.5574-5591
• /
• 2017

Inspired by the idea of Artificial Immune System, many researches of wireless sensor network (WSN) intrusion detection is based on the artificial intelligent system (AIS). However, a large number of generated detectors, black hole, overlap problem of NSA have impeded further used in WSN. In order to improve the anomaly detection performance for WSN, detector generation mechanism need to be improved. Therefore, in this paper, a Differential Evolution Constraint Multi-objective Optimization Problem based Negative Selection Algorithm (DE-CMOP based NSA) is proposed to optimize the distribution and effectiveness of the detector. By combining the constraint handling and multi-objective optimization technique, the algorithm is able to generate the detector set with maximized coverage of non-self space and minimized overlap among detectors. By employing differential evolution, the algorithm can reduce the black hole effectively. The experiment results show that our proposed scheme provides improved NSA algorithm in-terms, the detectors generated by the DE-CMOP based NSA more uniform with less overlap and minimum black hole, thus effectively improves the intrusion detection performance. At the same time, the new algorithm reduces the number of detectors which reduces the complexity of detection phase. Thus, this makes it suitable for intrusion detection in WSN.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.165.2-165
• /
• 2001

A number of previous researches in evolutionary algorithm are based on the study of facets we observe in natural evolution. The individuals of species in natural evolution occupy their own niche that is a subdivision of the habitat. This means that two species with the similar requirements cannot live together in the same niche. This is known as the competitive exclusion principle, i.e., complete competitors cannot coexist. In this paper, a new evolutionary programming algorithm adopting this concept is presented. Similarly in the case of natural evolution , the algorithm Includes the concept of niche obtained by partitioning a search space and the competitive exclusion principle performed by migrating individuals. Cell partition and individual migration strategies are used to preserve search diversity as well as to speed up convergence of an ...