惯性权值和学习因子对标准PSO算法性能的影响

doi:10.13438/j.cnki.jdzk.2019.04.005

吉首大学学报（自然科学版）

惯性权值和学习因子对标准PSO算法性能的影响

宋梦培，莫礼平，周恺卿

（吉首大学信息科学与工程学院，湖南吉首 416000）

出版日期:2019-07-25 发布日期:2019-07-31
通讯作者: 莫礼平（1972—），女，湖南安化人，高级实验师，主要从事智能计算及其应用、中文信息处理研究.
基金资助:
国家自然科学基金资助项目（61462029）；湖南省自然科学基金面上项目（2019JJ40234）；吉首大学研究生科研项目（JDY1816）

Influence of Inertia Weight and Learning Factor on Performance of Standard PSO Algorithm

SONG Mongpei, MO Liping, ZHOU Kaiqing

（College of Information Science and Engineering, Jishou University, Jishou 416000, Hunan China)

Online:2019-07-25 Published:2019-07-31

摘要/Abstract

摘要：

基于标准PSO算法，通过分析惯性权值和学习因子2类参数不同的取值策略对常用测试函数优化结果的影响，来探究2类参数对算法性能的影响.实验结果表明，2类参数恰当的动态改变不仅能明显提高单峰函数的寻优精度和收敛速度，而且能提高双峰和多峰函数的寻优概率；惯性权值主要影响算法的收敛速度，随着惯性权值的递增，算法收敛速度逐渐加快；学习因子主要影响算法的寻优精度，当反映粒子的自我学习能力和向群体最优粒子学习的能力的学习因子同增同减变化时，寻优精度提高；惯性权值递增结合2种学习因子的同增同减变化，或惯性权值递减结合2种学习因子的一增一减变化，均可使标准PSO算法性能得到显著提高.

关键词： 标准PSO算法, 惯性权值, 学习因子, 收敛速度, 寻优精度

Abstract:

Two types of parameters, inertia weight and learning factor, are the main factors affecting the accuracy and convergence speed of the standard PSO algorithm. Based on the standard PSO algorithm, the influence of different value-taking strategies of two types of parameters on the optimization results of commonly used test functions is analyzed to explore its impact on the performance of the standard PSO algorithm. The experimental results show that the proper dynamic change of the two types of parameters can significantly improve the optimization accuracy and convergence speed of the unimodal function, and can improve the optimization probability of the bimodal and multimodal functions. The inertia weight mainly affects the convergence speed of the algorithm. As the inertia weight increases, the convergence speed of the algorithm increases. The learning factor mainly affects the optimization precision of the algorithm. When the two learning factors increase and decrease, the precision of the optimization increases. The inertia weight increases with the two learning factors. The increase and decrease of the change, or the decrease of the inertia weight combined with the increase and decrease of the two learning factors, can significantly improve the performance of the standard PSO algorithm.

Key words: standard particle swarm optimization, inertia weight, learning factors, convergence speed, optimization accuracy

宋梦培，莫礼平，周恺卿. 惯性权值和学习因子对标准PSO算法性能的影响[J]. 吉首大学学报（自然科学版）, DOI: 10.13438/j.cnki.jdzk.2019.04.005.

SONG Mongpei, MO Liping, ZHOU Kaiqing. Influence of Inertia Weight and Learning Factor on Performance of Standard PSO Algorithm[J]. Journal of Jishou University（Natural Sciences Edition), DOI: 10.13438/j.cnki.jdzk.2019.04.005.

参考文献

[1] KENNEDY J, EBERHART R C. Particle Swarm Optimization[C]. Proceedings of IEEE International Conference on Neural Networks. Perth, Australia:IEEE Service Center, 1995: 1 942-1 948.
[2] EBERHART R C, SHI Yuhui. Comparing Inertia Weights and Constriction Factors in Particle Swarm Optimization[C]. Proceedings of the Congress on Evolutionary Computation Nagoya, Japan: IEEE Press, 2000: 84-88.
[3] SHI Yuhui, EBERHART R A. Modified Particle Swarm Optimizer[C]. IEEE World Congress On Computational Intelligence. Washington DC: IEEE Press, 1998: 69-73.
[4] SUGANTHAN P N. Particle Swarm Optimizer with Neighborhood Operator[C]. Proceedings of the Congress on Evolutionary Computation. Washington DC, Piscataway, NJ: IEEE Service Center, 1999: 1 958-1 962.
[5] NATSUKI HIGASHI, HITOSHI IBA. Particle Swarm Optimization with Gaussian Mutation[C]. Proceedings of the Swarm Intelligence Symposium, Indianapolis: IEEE Service Center, 2003: 72-79.
[6] LI Changhe, YANG Shengxiang. An Adaptive Learning Particle Swarm Optimizer for Function Optimization[C]. IEEE Congress on Evolutionary Computation, Trondheim: IEEE Service Center, 2009: 381-388.
[7] 张寅,曹德欣.混沌动态种群数粒子群优化算法[J].计算机工程与应用,2011,47(35):38-40.
[8] XIA Xuewen, LIU Jingnan, HU Zhongbo. An Improved Particle Swarm Optimizer Based on Tabu Detecting and Local Learning Strategy in a Shrunk Search Space Applied[J]. Soft Computing, 2014, 23(1): 76-90.
[9] MAHDAVI SEDIGHEH, RAHNAMAYAN SHAHRYAR. Shiri Mohammad Ebrahim Multilevel Framework for Large-Scale Global Optimization[J]. Soft Computing, 2016, 21(14): 4 111-4 140.
[10] CAMPOS MAURO, KROHLING RENATO A. Hierarchical Bare Bones Particle Swarm for Solving Constrained Optimization Problems[C]. Proceedings of the IEEE Congress on Evolutionary Computation. Cancun, Mexico: IEEE Press, 2013: 805-812.
[11] LI Jie, ZHANG Junqi, JIANG Changjun, et al. Composite Particle Swarm Optimizer with Historical Memory for Function Optimization[J]. IEEE Transactions on Cybernetics, 2015, 45(10): 2 350-2 363.
[12] SHOU Yongyi, LI Ying, LAI Changtao. Hybrid Particle Swarm Optimization for Preemptive Resource-Constrained Project Scheduling[J]. Neurocomputing, 2015, 148: 122-128.
[13] 赵玉磊,郭宝龙,吴宪祥,等. 基于双粒子群协同优化的ECT图像重建算法[J]. 计算机研究与发展,2014,51(9): 2 094-2 100.
[14] ZHANG Yong, GONG Dunwei, HU Ying, et al. Feature Selection Algorithm Based on Bare Bones Particle Swarm Optimization[J]. Neurocomputing, 2015, 148: 150-157.
[15] KOSHITI A, ARYA L D, CHOUBE S C. Voltage Stability Constrained Distributed Generation Planning Using Modified Bare Bones Particle Swarm Optimization[J]. Journal of the Institution of Engineers(India): Series B, 2013, 94(2): 123-133.

惯性权值和学习因子对标准PSO算法性能的影响

Influence of Inertia Weight and Learning Factor on Performance of Standard PSO Algorithm

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	黄金超. Kumaraswamy分布族参数的经验Bayes检验收敛速度[J]. 吉首大学学报（自然科学版）, 2021, 42(4): 9-15.
[2]	夏颖, 许贵桥. 一种拟Grünwald插值算子在B_a,Φ空间中的收敛速度[J]. journal6, 2009, 30(4): 19-22.
[3]	项雪艳, 何倩, 杨文善. 一些函数基于Chebyshev多项式的收敛性[J]. journal6, 2007, 28(2): 25-27.