[1]李汶娟,李 广,聂志刚.多项式变异和自适应权重优化的阿奎拉鹰算法[J].计算机技术与发展,2024,34(02):163-170.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 024]
　LI Wen-juan,LI Guang,NIE Zhi-gang.Polynomial Variance and Adaptive Weight Optimization for Aquila Algorithm[J].,2024,34(02):163-170.[doi:10. 3969 / j. issn. 1673-629X. 2024. 02. 024]

点击复制

多项式变异和自适应权重优化的阿奎拉鹰算法()

分享到：

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

卷:

34

期数:

2024年02期

页码:

163-170

栏目:

人工智能

出版日期:

2024-02-10

文章信息/Info

Title:

Polynomial Variance and Adaptive Weight Optimization for Aquila Algorithm

文章编号:

1673-629X(2024)02-0163-08

作者:

李汶娟; 李 广; 聂志刚

甘肃农业大学 信息科学技术学院,甘肃 兰州 730070

Author(s):

LI Wen-juan; LI Guang; NIE Zhi-gang

School of Information Science and Technology,Gansu Agricultural University,Lanzhou 730070,China

关键词:

Tent 混沌映射; 动态转换概率策略; 多项式变异扰动策略; 自适应权重; 阿奎拉鹰算法

Keywords:

Tent chaotic mapping; dynamic conversion probability strategy; polynomial variance perturbation strategy; adaptive weight; Aquila optimizer

分类号:

TP301. 6

DOI:

10. 3969 / j. issn. 1673-629X. 2024. 02. 024

摘要:

针对基本阿奎拉鹰算法存在收敛精度低、易陷入局部最优的问题,通过在全局搜索阶段引入多项式变异扰动策略,在局部开发阶段引入自适应权重优化策略,改进了阿奎拉鹰算法的局部探索能力,并且引入了 Tent 混沌映射初始化种群,增加种群多样性,引入动态转换概率策略来平衡全局探索和局部开发的比重,故提出多项式变异和自适应权重优化的阿奎拉鹰算法。 采用基本阿奎拉鹰算法、哈里斯鹰算法、灰狼算法、鲸鱼算法、海鸥算法做对比,9 个基准测试函数和 2 个工程优化问题对改进后的算法进行寻优性能验证,结果表明:改进后的算法在多数测试函数上取得较好的寻优效果,在工程优化问题中,效果优于多数对比算法。 证明了改进后的算法具有更快的收敛速度和精度,并在工程应用中取得较好效果。

Abstract:

To address the problem that the basic Aquila algorithm has low convergence accuracy and is prone to fall into local optimum,by introducing a polynomial variance perturbation strategy?

in the global search phase and an adaptive weight optimization strategy in thelocal exploitation phase, the local exploration ability of Aquila is improved. A Tent chaos mapping is introduced to initialize thepopulation and increase the population diversity,and a dynamic transformation probability strategy is introduced to balance the weight ofglobal exploration?

and local exploitation,so the Aquila algorithm with polynomial variance and adaptive weight optimization is proposed.The basic Aquila algorithm, Harris Hawks algorithm, Gray Wolf algorithm, Whale algorithm, and Seagull algorithm are used forcomparison,and 9 benchmark test functions and 2 engineering optimization problems are used to verify the improved algorithm ’ soptimization-seeking performance. The results show that the improved algorithm achieves better optimization-seeking results on most ofthe test functions and outperforms most of the comparison algorithms in engineering optimization problems. It is proved that the improvedalgorithm has faster convergence speed and accuracy,and achieves good results in engineering applications.

常用功能