A Study on Weightless Particle Swarm Optimization with a Globally Best Particle

Authors

  • Idris Abd Latiff

Abstract

Particle Swarm Optimization (PSO) has been established as an efficient computational intelligencetool since its introduction. Much of the improvement made on the particle swarm algorithm centered on the effect of a parameter called the inertia weight. In this study, the effect of the absence of inertia weight on the performance of PSO algorithm has been analyzed. A new term called the global particle has been introduced in the velocity update equation. This term has been able to compensate the absence of the inertia weightin order to maintain the convergence ability of the algorithm. Test results with standard objective functions demonstrate the necessity of having the inertia weight and justify the effort and research spent on developing many variants based on this important parameter. The results also show where the inertia weight term can be omitted to save computational cost.

Downloads

Published

2020-05-10

Issue

Section

Articles