Improving the search efficiency of differential evolution algorithm by population diversity analysis and adaptation of mutation step sizes
by Dhanya M. Dhanalakshmy; M.S. Akhila; C.R. Vidhya; G. Jeyakumar
International Journal of Advanced Intelligence Paradigms (IJAIP), Vol. 15, No. 2, 2020

Abstract: The aim of this research work is to improve the efficiency of differential evolution (DE) algorithm, at the cases of its unsuccessful searches. Initially, this work discusses and compares different methods to measure the population diversity of DE algorithm implemented for DE/rand/1/bin variant for a set of benchmarking functions. A method which well demonstrates difference in population diversity evolution at successful and unsuccessful cases of DE search is identified based on comparison. This work is then extended to detect unsuccessful searches in advance using the evolution of population diversity measured by the identified method. On detecting a search as unsuccessful, a parameter adaptation strategy to adapt the mutation step size (F) is added to DE algorithm to recover from it. The improved DE algorithm, which comprises of the logic of adapting F value based on the population diversity, is compared with its classical version and found outperforming. The comparison results are reported in this paper.

Online publication date: Fri, 14-Feb-2020

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
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 Advanced Intelligence Paradigms (IJAIP):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your 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