Improved particle swarm optimization: Catching the big wave on the surf

Abstract

Particle swarm optimization (PSO) is relatively a new population-based intelligence algorithm and exhibits good performance on optimization. However, during the optimization process, the particles become more and more similar, and gather into the neighborhood of the best particle in the swarm, which makes the swarm premature convergence probably around the local solution. PSO technique can be augmented with an additional mutation operator that provides diversity and helps prevent premature convergence on local optima. In this paper, diversity concept is evaluated in commonly used PSO algorithms including constriction factor PSO, inertial weight PSO, Gaussian mutation PSO, and a new vibrational mutation PSO combining the idea of mutation strategy related to periodicity. New algorithm is tested and compared with selected PSO algorithms. The results give insight into how mutation operator affects the nature of the diversity and show that the addition of mutation operator with periodicity concept can significantly enhance optimization performance.