Particle swarm optimization in the presence of multiple global optima

Authors:
Sunny Choi;Blayne E. Mayfield
Affiliations:
Oklahoma State University, Stillwater, OK, USA;Oklahoma State University, Stillwater, OK, USA
Venue:
Proceedings of the 11th Annual conference on Genetic and evolutionary computation
Year:
2009

Citing 5
Cited 0

Neural network design

Neural network design
Swarm intelligence

Swarm intelligence
The particle swarm optimization algorithm: convergence analysis and parameter selection

Information Processing Letters
The particle swarm - explosion, stability, and convergence in amultidimensional complex space

IEEE Transactions on Evolutionary Computation
The fully informed particle swarm: simpler, maybe better

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Dynamic analyses of canonical particle swarm optimization (PSO) have indicated that parameter values of phi_max = 4.1 and constriction coefficient chi = 0.729 provide adequate exploration and prevent swarm explosion. This paper shows by example that these values do not prevent swarm explosion in some cases. In other examples it is shown that even when the swarm does not explode, the canonical PSO algorithm with these parameter values can still fail to converge indefinitely. A satisfactory analysis of PSO has yet to be made, and will require abandoning certain assumptions that oversimplify particle behavior.