Enhancing the search ability of differential evolution through competent leader Online publication date: Sat, 12-Jul-2014
by Yiqiao Cai; Jixiang Du; Weibin Chen
International Journal of High Performance Systems Architecture (IJHPSA), Vol. 5, No. 1, 2014
Abstract: In the field of evolutionary algorithm (EA), differential evolution (DE) is successfully used in various scientific and engineering fields due to its strong global optimisation capability and simple implementation. However, in most of DE, the search is guided by the random or local optimal vectors. That is, DE does not effectively use the good information of population to guide the search. Therefore, to alleviate this drawback and enhance the search ability of DE, a competent leaders guiding strategy (cLGS) is proposed in this paper. The proposed cLGS is inspired by the natural phenomenon that good species usually contain useful information to guide the search of population. With the competent leaders, the good information of population can be utilised effectively during the evolutionary process. By incorporating cLGS into JADE which is a very competitive DE variant, the resulting algorithm, named JADE-cLGS, is proposed. In order to test the effectiveness of JADE-cLGS, a suite of benchmark functions is used. Experimental results demonstrate the high performance of JADE-cLGS by comparing with several DE variants.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of High Performance Systems Architecture (IJHPSA):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com