Abstract
Particle swarm optimization (abbr. PSO) is one of the most effective optimization algorithms. The PSO contains many control parameters. These causes, the performance of the serching ability of the PSO is significantly alternated. In order to analyze the dynamics of such PSO system, some papers pay attention to deterministc PSO systems which does not contain any stochastic factors. Based on the deterministic PSO, we proposed a canonical deterministic PSO. In this article, the canonical deterministic PSO is used. The state of the transmission of the position information of the global best can be regarded as a network structure. The network structure influences the performance of seraching ability. In this article, we analyze the performance focused on the degree and the average distance used in graph theory.
| Translated title of the contribution | Study of network structure on the searching performance of D-PSO |
|---|---|
| Original language | Japanese |
| Pages (from-to) | 27 - 30 |
| Journal | IEICE technical report. Circuits and systems |
| Volume | 110 |
| Issue number | 165 |
| State | Published - 26 Jul 2010 |