Impact of Particle Swarm Optimization Parameters on its Convergence
Narender Kumar Jain, Uma Nangia, Jyoti Jain · 2018
Particle swarm optimization algorithm is an intelligent optimization technique applied to solve optimization problem in the field of Engineering, optimization design of electrical networks, aircraft, control systems, optimum design of chemical processing equipment, design of civil engineering structure, economic load dispatch, load flow, management system, medical etc. Various parameters of PSO are: population size(P), inertia weight, acceleration constants random number, Maximum number of iteration. In this paper an attempt is made to understand the PSO algorithm by solving Rosenbrock benchmark function manually. The impact of variations of parameters on PSO convergence has been studied. Each parameters is varied systematically keeping all other parameters fixed to some value. Rosenbrock benchmark function has been used to study the impact of parameters variations. It was found that PSO is sensitive to the values of these parameters, combination of various parameters for best accuracy and faster convergence has been determined.