A new initializing mechanism in Particle Swarm Optimization
Jiyong Du, Fengming Zhang, Guorong Huang, Yang Ji · 2011
Particle Swarm Optimization (PSO) is known to suffer from premature convergence prior to discovering the true global minimizer. In this paper, a novel initializing mechanism is proposed, which aims to liberate particles from the state of premature convergence. This is done by automatically initializing the swarm once particles have converged to local minima, which is detected by the proposed criterion. An inertia weight function is also designed to balance the global and local search ability. The adaptive weight PSO with initializing mechanism (IAWPSO) provides an efficient mechanism by making good use of the state of the swarm at premature convergence. Results suggest that IAWPSO is less problem-dependent and consequently provides more consistent performance than the comparison algorithms across the benchmark suite used for testing.