基于精英反向学习和对数螺旋的HHO算法  

Harris Hawks Optimization Algorithm Based on Elite Opposition Learning and Logarithmic Spiral

作　　者：唐剑兰 蔡茂国[1] 徐翔 TANG Jian-lan;CAI Mao-guo;XU Xiang(College of Electronics and Information Engineering,ShenzhenUniversity,Shenzhen Guangdong 518000,China)

机构地区：[1]深圳大学电子与信息工程学院,广东深圳518000

出　　处：《计算机仿真》2023年第9期364-370,410,共8页Computer Simulation

摘　　要：针对传统哈里斯鹰优化算法(Harris Hawks Optimization, HHO)在处理庞杂问题易出现局部最优、收敛速度慢、寻优精度低的缺点,提出一种ELSHHO算法来对其进行改进。首先引入精英反向学习策略来对种群进行初始化,可以有效增强初始种群的多样性;其次在种群位置更新时加入精英反向学习策略可以提高算法探索解空间的能力和解的质量从而降低寻优难度加快收敛速度;最后,通过引入对数螺旋因子来增强算法的局部搜索性能,提高寻优精度。使用具有单峰和多峰特征的10个测试函数来对改进的算法进行验证,通过实验得出,ELSHHO算法可以有效提高收敛速度和寻优精度。An ELSHHO algorithm is presented to address the drawbacks of classic Harris hawks optimization(HHO)in dealing with complicated problems,such as local optimization,sluggish convergence,and low optimization accuracy.Firstly,elite opposition-based learning approach was used to initialize the population,which can successfully increase the beginning population's variety.Secondly,adding elite opposition-based learning strategy in population location updates improved the ability of the algorithm to explore the solution spaceandthe quality of population,reduce the difficulty of optimization and speed up convergence;Finally,the logarithmic spiral factor was incorporated to enhance the algorithm's local search performance and optimization accuracy.The improved algorithm was verified by using 10 test functions with single peak and multi peak characteristics.The experimental results show that ELSHHO algorithm can effectively improve the convergence speed and optimization accuracy.

关 键 词：元启发式 哈里斯鹰优化 对数螺旋 精英反向学习 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]