Soft Computing

, Volume 21, Issue 8, pp 2129–2137

Global harmony search with generalized opposition-based learning

Methodologies and Application

DOI: 10.1007/s00500-015-1912-1

Cite this article as:
Guo, Z., Wang, S., Yue, X. et al. Soft Comput (2017) 21: 2129. doi:10.1007/s00500-015-1912-1
  • 163 Downloads

Abstract

Harmony search (HS) has shown promising performance in a wide range of real-world applications. However, in many cases, the basic HS exhibits strong exploration ability but weak exploitation capability. In order to enhance the exploitation capability of the basic HS, this paper presents an improved global harmony search with generalized opposition-based learning strategy (GOGHS). In GOGHS, the valuable information from the best harmony is utilized to enhance the exploitation capability. Moreover, the generalized opposition-based learning (GOBL) strategy is incorporated to increase the probability of finding the global optimum. The performance of GOGHS is evaluated on a set of benchmark test functions and is compared with several HS variants. The experimental results show that GOGHS can obtain competitive results.

Keywords

Evolutionary algorithm Harmony search Exploitation Opposition-based learning 

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Zhaolu Guo
    • 1
  • Shenwen Wang
    • 2
  • Xuezhi Yue
    • 1
  • Huogen Yang
    • 1
  1. 1.Institute of Medical Informatics and Engineering, School of ScienceJiangXi University of Science and TechnologyGanzhouChina
  2. 2.School of Information EngineeringShijiazhuang University of EconomicsShijiazhuangChina

Personalised recommendations