Distance based multiple swarms formulation method in particle swarm optimization

Particle swarm optimization (PSO) is a solution to the functional optimization problem. PSO has been used in various fields because it has a simple algorithm and is characterized by being versatile with rapid convergence. However, because of its rapid convergence, it may fall into a local optimum especially when the problem has a large number of local solutions. In this paper, we propose a new type of PSO that forms swarms according to distances between particles, with the aim of addressing this local optima issue. Forming swarms in this way makes it possible for each particle to search the region of its own periphery in detail. The results of performance evaluation experiments using computer simulation verified that this method is less likely to get trapped in a local solution compared with conventional PSO.