具有稚虫迁徙机制的S型自适应混沌蜉蝣算法  

S-type Adaptive Mayfly Algorithm Based on Larval Migration Mechanism

作　　者：张恒齐 钱谦[1,2] ZHANG Heng-qi;QIAN Qian(Faculty of Information Engineering and Automation,Kunming University of Science and Technology;Yunnan Key Laboratory of Computer Technology Applications,Kunming University of Science and Technology)

机构地区：[1]昆明理工大学信息工程与自动化学院 [2]昆明理工大学云南省计算机技术应用重点实验室

出　　处：《化工自动化及仪表》2024年第2期262-273,共12页Control and Instruments in Chemical Industry

基　　金：云南省科技厅基础研究专项(批准号:202101AT070082)资助的课题。

摘　　要：针对蜉蝣优化算法(MA)全局搜索能力较弱,对子代优秀个体有效信息利用不够充分的缺点,以及在计算中后期易陷入局部收敛,进而影响优化效果的问题,提出一种具有稚虫迁徙机制的S型自适应混沌蜉蝣优化算法(S-AMA)。S-AMA算法首先采用Logistic混沌映射产生蜉蝣种群,增加算法初期种群的多样性;随后,将蜉蝣生命周期进行数学建模,并引入S型生命系数替换原有的重力惯性系数,动态调整算法探索能力和开发能力间的平衡性;最后,根据蜉蝣在缺氧环境下的种群活动,引入稚虫迁徙机制强化子代优秀个体摆脱局部最优的能力,进而更加充分地搜索最优解附近的区域,以增强算法的收敛精度。实验部分将S-AMA应用于标准函数测试集,并分别进行优化对比实验、Wilcoxon秩和检验。结果表明:与对比算法相比,S-AMA算法具有更好的寻优能力、收敛速度及鲁棒性。Considering Mayfly algorithm's(MA) poor searching ability,shortcomings in the individual information utilization of excellent offspring and the troubles to fall into local convergence in the middle and late calculation which affects the optimization effect,S-type adaptive mayfly algorithm based on larval migration mechanism(S-AMA) was proposed to solve above problems.Firstly,the S-AMA has Logistic chaotic mapping used to initialize both male and female mayfly populations and increase the diversity of the populations.Secondly,it has the Mayfly life cycle modeled and the S-type life coefficient added to replace the original gravity inertia coefficient to dynamically adjust the balance between the exploration ability and development ability of the algorithm;finally,through considering the life cycle in oxygen-deficient environment of a mayfly,a larval migration mechanism was introduced to strengthen the ability of the superior offspring to get out of local optimum and improve the convergence accuracy of the algorithm so as to enhance the convergence accuracy of the algorithm and search the region near the optimal solution more fully.In the experimental part,S-AMA was applied to the standard function test set,and the optimization contrast experiment and Wilcoxon rank sum test were performed respectively.The results show that the S-AMA algorithm has better optimization ability,convergence speed and robustness than the comparison algorithms.In the experiment,making use of standard function test set to compare the searching ability of the S-AMA with some other algorithms and implement the Wilcoxon rank sum test shows that,as compared to the contrast algorithm,the S-AMA has good optimization ability,convergence speed and strong robustness.

关 键 词：S型自适应混沌蜉蝣优化算法(S-AMA) 稚虫迁徙机制 混沌映射 S型生命系数 莱维飞行 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]