A nature inspired adaptive inertia weight in particle swarm optimisation
by Madhuri Arya; Kusum Deep; Jagdish Chand Bansal
International Journal of Artificial Intelligence and Soft Computing (IJAISC), Vol. 4, No. 2/3, 2014

Abstract: The selection of an appropriate strategy for adjusting inertia weight w is one of the most effective ways of enhancing the performance of particle swarm optimisation (PSO). Recently, a new idea, inspired from social behaviour of humans, for adaptation of inertia weight in PSO, has been proposed, according to which w adapts itself as the improvement in best fitness at each iteration. The same idea has been implemented in two different ways giving rise to two inertia weight variants of PSO namely globally adaptive inertia weight (GAIW) PSO, and locally adaptive inertia weight (LAIW) PSO. In this paper, the performance of these two variants has been compared with three other inertia weight variants of PSO employing an extensive test suite of 15 benchmark global optimisation problems. The experimental results establish the supremacy of the proposed variants over the existing ones in terms of convergence speed, and computational effort. Also, LAIW PSO comes out to be the best performer out of all the algorithms considered in this study.

Online publication date: Sat, 28-Jun-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 Artificial Intelligence and Soft Computing (IJAISC):
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