[1]胡树斌,魏霖静.基于混合策略改进的麻雀搜索算法[J].计算机技术与发展,2023,33(04):146-153.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 022]
　HU Shu-bin,WEI Lin-jing.Improved Sparrow Search Algorithm Based on Hybrid Strategy[J].,2023,33(04):146-153.[doi:10. 3969 / j. issn. 1673-629X. 2023. 04. 022]

点击复制

基于混合策略改进的麻雀搜索算法()

分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年04期

页码:

146-153

栏目:

人工智能

出版日期:

2023-04-10

文章信息/Info

Title:

Improved Sparrow Search Algorithm Based on Hybrid Strategy

文章编号:

1673-629X(2023)04-0146-08

作者:

胡树斌¹ ; 魏霖静²

1. 甘肃农业大学 理学院,甘肃 兰州 730070;

2. 甘肃农业大学 信息科学技术学院,甘肃 兰州 730070

Author(s):

HU Shu-bin1 ; WEI Lin-jing2

1. School of Science,Gansu Agricultural University,Lanzhou 730070,China;

2. School of Information Science & Technology,Gansu Agricultural University,Lanzhou 730070,China

关键词:

麻雀搜索算法; 精英反向学习; 黄金正弦算法; 莱维飞行; 函数优化

Keywords:

sparrow search algorithm; elite opposition-based learning; golden-sine algorithm; Levy flight; function optimization

分类号:

TP18

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 04. 022

摘要:

为克服基本麻雀搜索算法( Sparrow Search Algorithm,SSA) 存在的迭代过程中种群多样性减少、易陷入局部最优停滞、早熟收敛等不足,提出一种基于混合策略改进的麻雀搜索优化算法( EGSSA) 。 首先,迭代时利用精英反向学习机制在个体求解过程中计算精英反向解,提升算法寻优过程中种群丰富性和种群质量,为提高算法全局寻优能力以及收敛精度奠定基础;其次,将黄金正弦策略用以改进发现者位置更新方式,协调算法局部开发和全局搜索能力,并且提高算法的收敛能力;最后,在加入者的更新中采用了莱维飞行随机步长,以使新加入者所搜索的范围更为全面,一定程度避免算法陷入“早熟收敛冶。 选取 12 个基准测试函数进行仿真实验,分别与 4 种元启发式算法以及 3 种改进麻雀搜索算法进行比较,实验结果以及 Wilcoxon 秩和检验均表明,EGSSA 算法具有更好的全局探索能力和稳定性以及较强跳出局部极值的能力。

Abstract:

In order to overcome the shortcomings of the basic sparrow search algorithm ( SSA) in the iterative process, such as populationdiversity reduction,local optimal stagnation and premature convergence,an improved sparrow search optimization algorithm ( EGSSA)based on hybrid strategy is proposed. Firstly,the elite opposition-based learning mechanism is used to calculate the elite reverse solutionin the individual solution process in the iteration,so as to improve the population richness and population quality in the optimizationprocess of the algorithm,and lay the foundation for improving the global optimization ability and convergence accuracy of the algorithm.Secondly,the golden sine strategy is used to improve the location update method of the discoverer, coordinate the local development andglobal search ability of the algorithm,and improve the convergence ability of the algorithm. Finally,the random step size of Levy flight isintroduced in the update of the joiner ’s position, which makes the search direction of the joiner more comprehensive, and avoids thealgorithm from falling into premature convergence. Twelve benchmark functions are selected for simulation experiments and compared

with four other meta-heuristic algorithms and three other improved sparrow search algorithms,respectively. The experimental results andthe Wilcoxon rank sum test both show that the EGSSA algorithm has better global exploration ability and stability,as well as a strongability to jump out of local extremum.

常用功能