Summary
2011 International Symposium on Nonlinear Theory and Its Applications
2011
Session Number:B4L-D
Session:
Number:B4L-D1
Analysis of Dynamical Characteristic of Particle Swarm Optimization
Takuya Shindo, Kenya Jin’no,
pp.556-559
Publication Date:2011/9/4
Online ISSN:2188-5079
Summary:
A particle swarm optimization (PSO) is one of the powerful systems for solving global optimization problems. The searching ability of such PSO depends on the inertia weight coefficient, and the acceleration coefficients. Since the acceleration coefficients are multiplied by a random vector, the system can be regarded as a stochastic system. In order to analyze the dynamics rigorously, we pay attention to deterministic PSO which does not contain any stochastic factors. Especially, we propose a canonical deterministic PSO (CD-PSO). First, we compare with the searching ability between the standard PSO and the CDPSO. The deterministic PSO system must converge to the fix point without divergence. On the other hand, the standard PSO may diverge depending on the random parameter. Due to this divergence property, the standard PSO has high performance compared to the deterministic PSO. To overcome this weakness of the deterministic PSO, we proposed the reacceleration procedure for the deterministic PSO. In this article, the reacceleration operation is applied to the CD-PSO. Also, we confirm the performance of the CD-PSO with the reacceleration operation.