关键词: 微粒群算法, 参数选择, 收敛性, 粒子轨迹, 惯性权重

Abstract: Particle Swarm Optimization（PSO） is a novel optimization technology and has been applied successfully to various optimization problems.But the convergence of the algorithm has been studied insufficiently.The author studies the particle’s trajectory and convergence of the existing PSO algorithm，on the basis of their characteristics，proposes a new PSO algorithm with a new inertia weight along broken line.This algorithm preserves the advantages of the incremental and reduced inertia weight，and overcomes their shortcomings.Simulated experiments achieve good results.

Key words: Particle Swarm Optimization（PSO）, parameter selection, convergence, particle’s trajectory, inertia weight