基于动态调整惯性权重下改进学习因子的粒子群算法 |
| |
作者姓名: | 徐生兵 |
| |
作者单位: | 东莞理工学院城市学院计算机与信息科学学院,广东东莞523000 |
| |
基金项目: | 东莞理工学院城市学院青年教师基金项目(ZR15). |
| |
摘 要: | 粒子群算法针对高维复杂函数常存在早熟收敛问题,本文提出一种在已有动态调整惯性权重的基础上对学习因子进行改进的粒子群算法,使学习因子随着搜索的不同阶段改变认知学习因子和社会学习因子。比较五个标准测试函数的实验结果,表明改进后的算法得到的结果更优。
|
关 键 词: | 粒子群算法 动态调整惯性权重 学习因子 全局搜索 |
A new Modified Acceleration Coefficient in PSO Base on Dynamic Adjustment of Inertia Weights |
| |
Authors: | Xu Sheng-bing |
| |
Affiliation: | Xu Sheng-bing (Department of Computation and Information Science of City College of Dong Guan University of Technology GuangdongDongguan 523000) |
| |
Abstract: | Particle swarm optimization (pso) algorithm for high-dimensional complex functions often lead to premature convergence problem,this paper proposes a new modified acceleration coefficient in PSO base on dynamic adjustment of inertia weights. Comparing experimental results of five standard test functions,the results show that the improved algorithm is better. |
| |
Keywords: | particle swarm optimization algorithm dynamic change of inertia weights acceleration ceefficient globle searching |
本文献已被 CNKI 维普 等数据库收录! |