Fractional order dynamics in a particle swarm optimization algorithm

This article reports the study of fractional dynamics during the evolution of Particle Swarm Optimization (PSO) algorithm. Some initial swarm particles are randomly changed, for simulating the system response, and its effect is compared with a non-perturbed reference. The perturbation effect in the...

ver descrição completa

Detalhes bibliográficos
Autor principal: Pires, E. J. Solteiro (author)
Outros Autores: Oliveira, P. B. Moura (author), Machado, J. A. Tenreiro (author), Jesus, Isabel S. (author)
Formato: conferenceObject
Idioma:eng
Publicado em: 2007
Assuntos:
Particle swarm optimization
Fractional calculus
Intelligent systems
Testing
Analytical models
Employee welfare
Texto completo:http://hdl.handle.net/10400.22/10089
País:Portugal
Oai:oai:recipp.ipp.pt:10400.22/10089
Descrição
Resumo:This article reports the study of fractional dynamics during the evolution of Particle Swarm Optimization (PSO) algorithm. Some initial swarm particles are randomly changed, for simulating the system response, and its effect is compared with a non-perturbed reference. The perturbation effect in the PSO evolution is observed in the perspective of the fitness time behavior of the best particle. the dynamics is represented throught the median of a sample of experiments, while adopting the Fourier analysis for describing the phenomena. The influence of the PSO parameters influence upon the global dynamics is also analyzed.

Registos relacionados