The particle swarm optimization (PSO) is a population-based optimization method inspired by flocking behavior \r\nof birds and human social interactions. So far, numerous modifications of PSO algorithm have been published, \r\nwhich make the PSO method more complex. Several improved PSO versions succeed in keeping the diversity of the \r\nparticles during the searching process, but at the expense of convergence speed. This paper is aimed at increasing \r\nthe rate of convergence and diversity of solutions in the population via two easy techniques: \r\n(a) Applying improved acceleration coefficients \r\n(b) Dividing search space into blocks. In particular, the second technique is efficient in the case of functions \r\nwith optimal design variables situated in the one block. Hence, instead of proposing more complex variant of PSO, \r\na simplified novel technique, called Partitioned Particle Swarm Optimizer (PPSO), has been proposed. In order to \r\nfind optimal coefficients of this method, an extensive set of experiments were conducted. Experimental results and \r\nanalysis demonstrate that PPSO outperforms nine well-known particle swarm optimization algorithms with regard \r\nto global search.
Loading....