Abstract: A multi-objective optimization based on suitable-distribution particle swarm (SDPS) is discussed. There are simple rules and quick convergence speed for SDPS. Also, it can ensure the final solutions more dispersive and symmetrical. A new method is proposed to update the external memory and fix the fit-sharing radius. In order to compare with other algorithms, the proposed algorithm is used to solve some well-known multi-objective functions. It is proved that the proposed algorithm is able to find the solutions which are much more exact and efficient convergence to the true Pareto front and much more well-proportioned distribution over the Pareto front.

Key words: suitable-distribution, particle swarm, multi-objective optimization, Pareto front

摘要： 提出了一种基于适配粒子群的多目标优化方法。该方法给出的适配粒子群算法规则简单、收敛速度快，得到的解集有较好的分散性和均匀性。将提出的外部记忆体更新和适配半径选择的方法应用于经典的多目标函数中。结果表明，该优化方法能够快速准确地收敛于Pareto解集，并且使其对应的目标域均匀分布于Pareto最优目标域。

关键词: 适配, 粒子群, 多目标优化, Pareto最优目标域

CLC Number: 

• TP301.6