一种自适应改变惯性权重的粒子群算法

针对标准粒子群算法收敛性和收敛速度的问题，分析标准粒子群算法惯性参数对算法性能优化的影响，提出一种自适应改变惯性权重的粒子群算法（ACPSO）。通过对粒子速度和位置变化过程的分析，并结合早熟收敛程度和个体适应值自适应的调整惯性权重，使得算法能在全局收敛性和收敛速度之间找到良好的平衡关系，并且通过典型的函数测试，表明此方法有效的控制了粒子群的多样性，而且具有良好的收敛速度。

An adaptive changed inertia weight particle swarm algorithm

Aimed at convergence and convergence rate of the standard particle swarm algorithm, analyzing inertial parameters of the standard particle swarm algorithm affects the performance optimization, an adaptive change in inertia weight of particle swarm optimization algorithm (ACPSO) is proposed. By analyzing the particle velocity and the process of the changing position, combined with the degree of premature convergence and individual adaptive value adjust the inertia weight, the algorithm has a good balance between the global convergence and convergence rate. And a typical function test shows that this method is effective in controlling the particle swarm diversity, and it also has good convergence rate.