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
Volume4113 LNCS - I
ISBN (Print)3540372717, 9783540372714
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
CountryChina
CityKunming
Period06-08-1606-08-19

Fingerprint

Evolutionary Strategy
Global optimization
Simulated annealing
Simulated Annealing
Global Optimization
Optimization Methods
Structural design
Probability distributions
Distribution functions
Annealing
Structural Design
Thermal Equilibrium
Probability Distribution Function
Global Optimum
Test function
Numerical Study
Objective function
Robustness
Optimization Problem
Temperature

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Chiang, D-Y., & Moh, J. (2006). A global optimization method based on simulated annealing and evolutionary strategy. In International Conference on Intelligent Computing, ICIC 2006, Proceedings (Vol. 4113 LNCS - I, pp. 790-801). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4113 LNCS - I). Springer Verlag.
Chiang, Dar-Yun ; Moh, Jausung. / A global optimization method based on simulated annealing and evolutionary strategy. International Conference on Intelligent Computing, ICIC 2006, Proceedings. Vol. 4113 LNCS - I Springer Verlag, 2006. pp. 790-801 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{b3f1b6237ae3461e8e81dcd4dd14fc6d,
title = "A global optimization method based on simulated annealing and evolutionary strategy",
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.",
author = "Dar-Yun Chiang and Jausung Moh",
year = "2006",
language = "English",
isbn = "3540372717",
volume = "4113 LNCS - I",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "790--801",
booktitle = "International Conference on Intelligent Computing, ICIC 2006, Proceedings",
address = "Germany",

}

Chiang, D-Y & Moh, J 2006, A global optimization method based on simulated annealing and evolutionary strategy. in International Conference on Intelligent Computing, ICIC 2006, Proceedings. vol. 4113 LNCS - I, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4113 LNCS - I, Springer Verlag, pp. 790-801, International Conference on Intelligent Computing, ICIC 2006, Kunming, China, 06-08-16.

A global optimization method based on simulated annealing and evolutionary strategy. / Chiang, Dar-Yun; Moh, Jausung.

International Conference on Intelligent Computing, ICIC 2006, Proceedings. Vol. 4113 LNCS - I Springer Verlag, 2006. p. 790-801 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4113 LNCS - I).

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

TY - GEN

T1 - A global optimization method based on simulated annealing and evolutionary strategy

AU - Chiang, Dar-Yun

AU - Moh, Jausung

PY - 2006

Y1 - 2006

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=33749574339&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749574339&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540372717

SN - 9783540372714

VL - 4113 LNCS - I

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 790

EP - 801

BT - International Conference on Intelligent Computing, ICIC 2006, Proceedings

PB - Springer Verlag

ER -

Chiang D-Y, Moh J. A global optimization method based on simulated annealing and evolutionary strategy. In International Conference on Intelligent Computing, ICIC 2006, Proceedings. Vol. 4113 LNCS - I. Springer Verlag. 2006. p. 790-801. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).