基于小世界邻域结构的微粒群算法研究

分析了邻域结构对微粒群算法的影响，针对收敛速度慢，早熟收敛等缺点，结合小世界网络的基本特性，提出了一种基于小世界邻域结构的微粒群算法。在该模型中，邻域内部的高聚集性有利于微粒的细致搜索，而邻域间少量的长程连接又能保证微粒在进化过程中更加全面、快捷地实现信息的有效共享，从而在提高收敛速度的同时防止陷入局部最优。将本模型与Gbest模型及环形结构进行比较，发现该算法不仅具有更快的收敛速度，而且能够获得更好的收敛效果。

Particle Swarm Optimization Based on Small-world Neighbourhood Structure

The impact of the neighbourhood structure on particle swarm optimization is analyzed in this paper. In view of the problem of slow search speed and premature convergence, the particle swarm optimization is combined with the basic characteristics of small-world model, and a novel particle swarm optimization based on small-world neighbourhood structure is proposed. In this model, highly clustered neighbourhood is beneficial to the subtle search, and a few long-range connections guarantee the particles to share information more comprehensively, more quickly and more effectively in the evolution process. Thus the search speed is improved and prevented from trapping into local optima at the same time. The comparison between this model with other models, such as Gbest model and ring model, is carried out. The experimental results show that the proposed method can not only significantly speed up the convergence, but also gain better convergence effect.