摘要
粒子群算法(Particle Swarm Optimizer,PSO)在处理较大数据的寻优问题时存在运行时间过长,易陷入局部最优的问题。文章基于传统粒子群算法,利用基于因素空间的特征选取法,优化特征数据组合,减少算法运行时间,利用多种群粒子群优化模型思想和模糊数学思想相结合提出基于因素空间的模糊多种群粒子群算法,增强了算法全局寻优能力并加快收敛速度,实现了算法的总体最优,改善了传统粒子群优化算法容易收敛到局部最优的缺陷。研究结果表明,与传统粒子群算法、动态多种群粒子群算法、基于密度峰值的多种群粒子群算法相比,文章所提出的基于因素空间的模糊多种群粒子群算法具有更优的寻找全局最优解能力和收敛速度。
Particle Swarm Optimizer(PSO)has the problems of long running time and easy to fall in-to local optimality when dealing with the optimization of large data.In this paper,based on the tradi-tional particle swarm optimization algorithm,the feature selection method based on factor space is used to optimize the combination of feature data and reduce the running time of the algorithm.The fuzzy multi-population particle swarm optimization algorithm based on factor space is proposed by com-bining the idea of multi-population particle swarm optimization model with the idea of fuzzy mathemat-ics,which enhances the global optimization ability of the algorithm and speeds up the convergence rate,thus achieving the overall optimization of the algorithm.The traditional particle swarm optimiza-tion algorithm is easy to converge to the local optimal defect.The results show that compared with the traditional particle swarm optimization algorithm,dynamic multi-population particle swarm optimiza-tion algorithm and multi-population particle swarm optimization algorithm based on density peak,the fuzzy multi-population particle swarm optimization algorithm based on factor space proposed in this pa-per has a better ability to find the global optimal solution and convergence speed.
作者
钟育彬
范书衡
ZHONG Yu-bin;FAN Shu-heng(School of Mathematics and Information Science,Guangzhou University,Guangzhou 510006,China)
出处
《广州大学学报（自然科学版）》
CAS
2023年第6期77-81,共5页
Journal of Guangzhou University:Natural Science Edition
关键词
因素空间
多种群粒子群算法
优化问题
factor space
multi-population particle swarm optimization
problem of optimization
