A novel electromagnetism-like mechanism algorithm with modified Solis and Wets local search for global optimisation
by Chunjiang Zhang; Liang Gao; Xinyu Li; Qing Wu
International Journal of Services Operations and Informatics (IJSOI), Vol. 7, No. 2/3, 2012

Abstract: Electromagnetism-like mechanism (EM) algorithm, a meta-heuristic algorithm for global optimisation, utilises an attraction-repulsion mechanism to move the sample points towards the optimality. The original EM has a strong ability for diversification. And a simple random line search algorithm for local search has been added to improve its intensification. However, it plays a quite limited role and the performance of the original EM is not satisfactory. Therefore, in this paper, a modified Solis and Wets local search is proposed to improve the performance of EM. Two self-adapt parameters are introduced into Solis and Wets local search. In addition, an accelerated force formula is adopted for high-dimensional function optimisation. Some benchmark test problems have been used to evaluate the proposed algorithm. Results obtained are compared with those from other algorithms including original EM algorithm and all kinds of particle swarm optimisation methods. The comparisons show that the novel EM has achieved significant improvement.

Online publication date: Sat, 27-Dec-2014

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 Services Operations and Informatics (IJSOI):
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