多峰搜索的动态微粒群算法

Particle swarm optimization with dynamic-population for multi-peak searching

对多峰搜索问题提出了一类动态微粒群算法。该算法通过变换函数将多峰问题中的所有峰变为等高峰,从而保证每个峰都有同等机会被找到;在搜索过程中采用群体规模动态可调的进化方式,使得初始群体可以任意指定,从而克服了标准微粒群算法由于无法事先知道多峰函数峰值点个数而很难确定合适群体大小的困难。实验表明了该算法可以尽可能多地找到峰值点。

A class of dynamic-population particle swarm optimization for searching peaks of some multi-peak functions was proposed. This algorithm transformed all peaks of multi-peak problems into those peaks equally high by functional transformation, in order to find all peaks with the same probability. During the searching the size of particle swarm could be tuned to get any initial size of swarm. So it could solve the problem of determining swarm size because the number of peaks of the given multi-peak function could not be obtained in standard particle swarm optimization. The experiments manifest that the algorithm can search peaks of function as much as possible.