Particle Swarm Optimization With an Aging Leader and Challengers

In nature, almost every organism ages and has \na limited lifespan. Aging has been explored by biologists to \nbe an important mechanism for maintaining diversity. In a \nsocial animal colony, aging makes the old leader of the colony \nbecome weak, providing opportunities for the other individuals \nto challenge the leadership position. Inspired by this natural \nphenomenon, this paper transplants the aging mechanism to \nparticle swarm optimization (PSO) and proposes a PSO with an \naging leader and challengers (ALC-PSO). ALC-PSO is designed \nto overcome the problem of premature convergence without \nsignificantly impairing the fast-converging feature of PSO. It \nis characterized by assigning the leader of the swarm with a \ngrowing age and a lifespan, and allowing the other individuals \nto challenge the leadership when the leader becomes aged. The \nlifespan of the leader is adaptively tuned according to the \nleader’s leading power. If a leader shows strong leading power, \nit lives longer to attract the swarm toward better positions. \nOtherwise, if a leader fails to improve the swarm and gets old, \nnew particles emerge to challenge and claim the leadership, \nwhich brings in diversity. In this way, the concept “aging” \nin ALC-PSO actually serves as a challenging mechanism for \npromoting a suitable leader to lead the swarm. The algorithm \nis experimentally validated on 17 benchmark functions. Its high \nperformance is confirmed by comparing with eight popular PSO \nvariants.