A Differential Evolution algorithm based on Ordering of individuals

Yang LOU*, Junli LI

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

9 Citations (Scopus)

Abstract

Differential Evolution (DE) algorithm has been shown to be powerful for many real optimization problems. To improve the robustness and convergence speed, we propose a novel method, which is ordering of individuals in the population. This has changed the structure of population in traditional DE algorithm, which is always randomly generated and randomly evolved. Ordering of individuals improved the stability of the evolved solution immensely. Differential Evolution algorithm based on Ordering of individuals (ODE) is the basic pattern of the applications of ordering and combined with other means, ordering would get a better performance to strengthen the robustness of DE algorithm. By the experimental testing of benchmark functions, the results show ODE algorithm has a better performance than DE algorithm especially in robustness.

Original languageEnglish
Title of host publication2010 The 2nd International Conference on Industrial Mechatronics and Automation
PublisherThe Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Pages105-108
Number of pages4
Volume2
ISBN (Electronic)9781424476565
ISBN (Print)9781424476541
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event2010 2nd International Conference on Industrial Mechatronics and Automation, ICIMA 2010 - Wuhan, China
Duration: 30 May 201031 May 2010

Conference

Conference2010 2nd International Conference on Industrial Mechatronics and Automation, ICIMA 2010
Country/TerritoryChina
CityWuhan
Period30/05/1031/05/10

Keywords

  • differential evolution algorithm
  • Odering

Fingerprint

Dive into the research topics of 'A Differential Evolution algorithm based on Ordering of individuals'. Together they form a unique fingerprint.

Cite this