Particle swarm optimisation with population size and acceleration coefficients adaptation using hidden Markov model state classification
by Oussama Aoun; Malek Sarhani; Abdellatif El Afia
International Journal of Metaheuristics (IJMHEUR), Vol. 7, No. 1, 2018

Abstract: Particle swarm optimisation (PSO) is a metaheuristic algorithm based on population, it succeeded in solving a large number of optimisation problems. Several adaptive PSO algorithms have been proposed to enhance the performance of the original one. In particular, parameter adaptation has become a promising issue of PSO. In this paper, we propose an adaptive control of two PSO parameters using hidden Markov model (HMM) classification to enhance PSO performance, called HMM adaptive control of PSO (HMM-ACPSO). That is, we integrate HMM to have a stochastic control of states at each iteration. Then, the classified state by HMM is used to adapt PSO with both acceleration parameters and population size. Furthermore, several strategies varying the swarm are adopted according to the classified state. We performed evaluations on several benchmark functions to test the HMM-ACPSO algorithm. Experimental results reveal that our suggested scheme gives competitive results comparing with PSO variants regarding both solution accuracy and convergence speed.

Online publication date: Sun, 20-May-2018

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 Metaheuristics (IJMHEUR):
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 [email protected]