한국해양대학교

Detailed Information

Metadata Downloads

실수코딩 유전알고리즘과 단체법을 결합을 하이브리드 탐색법

Title
실수코딩 유전알고리즘과 단체법을 결합을 하이브리드 탐색법
Alternative Title
A Hybrid Search Method Combining a Real-Coded Genetic Algorithm with a Simplex Search Method
Author(s)
강원룡
Publication Year
2002
Publisher
한국해양대학교 대학원
URI
http://kmou.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002175414
http://repository.kmou.ac.kr/handle/2014.oak/9625
Abstract
One of the main drawbacks in applying genetic algorithms (GAs) to complex problems has been the high computational cost due to their slow convergence rate. This fact gives a difficulty in an attempt to use GAs for function optimization problem. To alleviate this difficulty, an island-model based hybrid search method which combines a real-coded genetic algorithm(RCGA) with a simplex search method is presented. Our motivation for employing the RCGA is to introduce best exploration into the hybrid search method, and the simplex search method is to introduce cost-effective exploitation. In an attempt to make effective use of the exploitation operation of the simplex search method in the proposed search framework, we use a parallel architecture where two algorithms run during the isolation time and exchanges migrants. To demonstrate the superiority of the proposed algorithm, it is compared with an alternative optimization technique, RCGA proposed by Michalewicz in two optimization problems for modeling a system with time delay and tuning the parameters of a PID controller.
Appears in Collections:
제어계측공학과 > Thesis
Files in This Item:
000002175414.pdf Download

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse