融合均值榜样的反向互学习水母搜索算法
- 分享到:
摘要：
为解决水母搜索算法(jellyfish search algorithm ,JS)的洋流运动缺乏多样性、群内运动缺乏引导性、种群间信息无交流 ,造成搜索速度慢 、稳定性差及易早熟的问题 ,构建了一种融合均值榜样的反向互学习水母搜索算法(oppositional-mutuallearningjellyfish search algorithm based on mean-valueexample,OMLJS) .首先在水母跟随洋流运动(全局搜索)部分 ,利用前两代水母的平均位置代替只考虑上一代水母的平均位置来引导水母个体的位置更新,提高算法的全局搜索能力;其次在水母的群内主动运动(局部搜索)部分,利用最优个体代替随机个体来引导水母进行更有效的搜索 ,加快算法的收敛速度 ;然后在水母进入下一次迭代前增加对水母种群进行动态反向互学习步骤 ,增加种群多样性及增强种群间的信息交流 ,达到互补另外两个策略 ,提高算法的整体优化性能。选用12个经典的基准测试优化函数 ,将 OMLJS与5个对比算法从解的平均值 、最优值及方差进行对比分析,并用于求解最小生成树问题 ,OMLJS能够更快地找到最小生成树.实验结果表明 ,OMLJS的收敛速度、求解精度明显提高 .
In order to solve the problems thatthe lack ofdivercity ofthejellyfish search algorithm(JS) in ocean current motion, thelack ofguidancein intra-swarm motion, and no exchangeofinformation between populations, which resultin slow search speed, poor stability, and susceptibility to precocious maturation, an oppositional-mutuallearningjellyfish search algo- rithm based on mean-valueexample(OMLJS) isconstructed. Firstly, in thepartofthejellyfish following ocean currents(global search) , the average position ofthe previoustwo generationsofjellyfish isused instead ofonly considering theaverageposition ofthe previous generation to guidethe position updateofindividualjellyfish, which improvestheglobalsearch abilityoftheal- gorithm ; secondly, in the partofjellyfish's intra-swarm active movement(localsearch) , the optimal individual is used instead ofrandom individualto guide thejellyfish to perform more effective search, which accelerates the convergence speed ofthe al- gorithm ; Then, before thejellyfish enters the nextiteration, a dynamic reverse mutual learning step is added to the jellyfish population to increase the diversity ofthe population and enhance the information exchange between the populations, so as to complement the other two strategies and improve the overall optimization performance of the algorithm. Twelve classical benchmark testoptimization functionsareselected to compareand analyzeOMLJS withfivecomparativealgorithmsin termsof the mean value, optimalvalueand varianceofthe solutions, and are used to solve the minimum spanning tree problem , where OMLJS is able to find the minimum spanning tree faster. The experimentalresults show thatthe convergence speed and solu- tion accuracy of OMLJS are significantly improved.
作者：
段艳明 , 肖辉辉 ,谭黔林
DuanYanming, Xiao Huihui , Tan Qianlin
机构地区：
河池学院大数据与计算机学院；广西蚕桑生态学与智能化技术应用重点实验室
引用本文：
段艳明,肖辉辉,谭黔林.融合均值榜样的反向互学习水母搜索算法[J].betway官方app 学报(自然科学版),2024,52(4):111-119.
Duan Yanming, Xiao Huihui, Tan Qianlin. Oppositional-mutual learning jellyfish search algorithm based on mean-value example[J] . Journal of Henan Normal University(Natural Science Edition) , 2024,52(4) :111-119. DOI:10. 16366/j. cnki.1000-2367. 2023. 04. 25. 0001.
基金：
国家自然科学基金(61972184;61562032)；河池学院高层次人才科研启动项目 (2019GCC012)
关键词：
水母搜索算法 ;均值榜样学习 ;反向互学习 ; 时间控制机制 ;最小生成树问题
jellyfish search algorithm ; mean-value example learning; oppositional-mutuallearning; time controlmecha- nism; minimum spanning tree problems
分类号：
TP18
