A global optimization method based on simulated annealing and evolutionary strategy

Dar Yun Chiang, Jausung Moh

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A global optimization method is proposed to improve the conventional method of simulated annealing. By introducing the probability distribution function for the objective function and the concept of stable energy for detecting thermal equilibrium during annealing, the selection of initial temperature and equilibrium criterion becomes easy and effective. Furthermore, the efficiency and robustness of the proposed method is retained by employing the technique of region reduction and an adaptive neighborhood structure. In the case where multiple (global) optima may exist, a technique based on the method of simulated evolution is developed to circumvent the difficulty of convergence of population. Numerical studies of some standard test functions and an optimum structural design problem show that the proposed method is effective in solving global optimization problems.

Original languageEnglish
Title of host publicationInternational Conference on Intelligent Computing, ICIC 2006, Proceedings
PublisherSpringer Verlag
Pages790-801
Number of pages12
ISBN (Print)3540372717, 9783540372714
DOIs
Publication statusPublished - 2006
EventInternational Conference on Intelligent Computing, ICIC 2006 - Kunming, China
Duration: 2006 Aug 162006 Aug 19

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4113 LNCS - I
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference on Intelligent Computing, ICIC 2006
Country/TerritoryChina
CityKunming
Period06-08-1606-08-19

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'A global optimization method based on simulated annealing and evolutionary strategy'. Together they form a unique fingerprint.

Cite this