The variations of harmony search and its current research trends

Xiaolei Wang*, Xiao Zhi Gao, Kai Zenger

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review

1 Citation (Scopus)

Abstract

Since the HS was firstly assessed and utilized in the optimal cost design of water distribution network in 2006, it has attracted growing research interest in handling a large variety of optimization problems. As aforementioned, the original HS has its own inherent drawbacks (slow convergence and outdated information). Hence, a lot of relevant literature proposes many variations of the HS, which can be classified into two major categories: modification based on the regular HS and hybridization with other NIC algorithms. The former targets the improvement of parameters or operators of the HS to enhance the quality of solutions, while the latter incorporates some optimization approaches to overcome its drawbacks and improve the performance. From the viewpoint of structure, the cooperation in the fusion strategies of the hybrid HS methods could be in the manner of either cooperator or embedded operators. This chapter first gives an overview of the variations of the HS and the current research trends, and a modified HS method for constrained optimization is next discussed in details.

Original languageEnglish
Title of host publicationSpringerBriefs in Applied Sciences and Technology
PublisherSpringer
Pages21-30
Number of pages10
Edition9783319083551
DOIs
Publication statusPublished - 2015
MoE publication typeA3 Book section, Chapters in research books

Publication series

NameSpringerBriefs in Applied Sciences and Technology
Number9783319083551
ISSN (Print)2191-530X
ISSN (Electronic)2191-5318

Keywords

  • Adaptive parameters
  • Constrained optimization
  • Harmony search method
  • Hybrid harmony search methods
  • Parameter tuning
  • Penalty function

Fingerprint

Dive into the research topics of 'The variations of harmony search and its current research trends'. Together they form a unique fingerprint.

Cite this