[1]李奕轩,田云娜,王凯欣.多策略改进的鱼鹰优化算法及其应用[J].计算机技术与发展,2025,(01):132-139.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0261]
　LI Yi-xuan,TIAN Yun-na,WANG Kai-xin.Improved Osprey Optimization Algorithm Based on Multiple Strategies and Its Application[J].,2025,(01):132-139.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0261]
点击复制

多策略改进的鱼鹰优化算法及其应用()

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

卷:
期数:
2025年01期
页码:
132-139
栏目:
人工智能
出版日期:
2025-01-10

文章信息/Info

Title:
Improved Osprey Optimization Algorithm Based on Multiple Strategies and Its Application
文章编号:
1673-629X(2025)01-0132-08
作者:
李奕轩; 田云娜; 王凯欣
延安大学 数学与计算机科学学院,陕西 延安 716000
Author(s):
LI Yi-xuan; TIAN Yun-na; WANG Kai-xin
School of Mathematics and Computer Science,Yan’an University,Yan’an 716000,China
关键词:
鱼鹰优化算法; 螺旋振荡; 莱维飞行; 差分变异; 柯西突变
Keywords:
osprey optimization algorithm; screwing oscillator; Levy flight; differential variation; Cauchy mutation
分类号:
TP301.6
DOI:
10.20165/j.cnki.ISSN1673-629X.2024.0261
摘要:
针对鱼鹰优化算法存在全局搜索能力不足、探索与开发不平衡以及易陷入局部最优的问题,提出一种多策略融合改进的鱼鹰优化算法。 首先,在探索阶段引入分种群螺旋振荡策略,提高算法的全局搜索能力和寻优精度;其次,在开发阶段引入基于动态因子调整的莱维飞行策略,提高搜索个体跳出局部最优的能力,平衡算法的探索与开发;最后,利用多差分柯西突变策略提高种群的多样性,强化算法跳出局部最优的能力,同时提升算法的收敛速度。 将该算法与 7 种优化算法在 30 个 CEC2017 测试函数、12 个 CEC2022 测试函数上进行实验对比,并使用 Wilcoxon 秩和检验与 Friedman 检验进行统计测试。 实验结果表明,该算法在 CEC2017 测试函数集、CEC2022 测试函数集上 Friedman 平均排名均为第一。 与原算法和其他 6 个对比算法相比,该算法在寻优精度、收敛速度以及稳定性方面表现优异。 通过求解 3 个实例约束优化问题,验证了该算法能有效解决实际优化问题。
Abstract:
Aiming at the problems of insufficient global search ability,unbalanced exploration and development,and easy to fall into local optimum in the osprey optimization algorithm,an improved osprey optimization algorithm based on multi-strategy fusion is proposed.Firstly,the sub - population spiral oscillation strategy is introduced in the exploration stage to improve the global search ability and optimization accuracy of the algorithm. Secondly,in the development stage,the Levy flight strategy based on dynamic factor adjustment is introduced to improve the ability of searching individuals to jump out of local optimum and balance the exploration and development of the algorithm. Finally,the multi-difference Cauchy mutation strategy is used to improve the diversity of the population,strengthen the ability of the algorithm to jump out of the local optimum,and improve the convergence speed of the algorithm. The proposed algorithm is compared with seven optimization algorithms on 30 CEC2017 test functions and 12 CEC2022 test functions,and Wilcoxon rank sum test and Friedman test are used for statistical testing. The experimental results show that the proposed algorithm ranks first in Friedman average on CEC2017 test function set and CEC2022 test function set. Compared with the original algorithm and other six comparison al-gorithms,the proposed algorithm performs well in terms of optimization accuracy, convergence speed and stability. By solving three examples of constrained optimization problems,it is verified that the proposed algorithm can effectively solve practical optimization prob-lems.
更新日期/Last Update: 2025-01-10