基于混沌迁移的社会学习天牛群算法
|
摘要
针对天牛须算法搜索精度较低和易陷入局部最优的缺陷,提出了社会学习天牛群算法(SLBSA)。首先,采用拉丁超立方抽样(LHS)产生多样性较好的天牛群;其次,将迭代过程类比生物觅食的过程,采用基于相似度的混沌迁移策略,避免天牛群陷入局部最优;最后,在天牛群中引入社会学习策略保证天牛群在更新时能进行充分高效的信息交换。为了验证SLBSA的有效性,将其与天牛群算法(BSO)及粒子群算法(PSO)通过标准测试函数进行测试和对比,验证了SLBSA具有更快的收敛速度和更强的全局搜索能力。
The beetle antennae search(BAS)has the shortcomings of low search accuracy and falling into local optimum easily.To solve these problems,a social learning beetle swarm algorithm(SLBSA)is proposed in this paper.Firstly,the Latin hypercube sampling(LHS)is used to generate the beetle swarm with better diversity.Secondly,the iterative process is compared with the biological foraging process,and the chaotic migration strategy based on similarity is adopted to avoid the beetle swarm falling into the local optimum.Finally,the social learning strategy is introduced to ensure the full and efficient information exchange during the updating process.In order to verify the effectiveness,the proposed SLBSA is tested in comparison with the beetle swarm optimization(BSO)and the particle swarm optimization(PSO)on the benchmark functions.Experimental results show that the SLBSA has the faster convergence speed and stronger global search capability.
The beetle antennae search(BAS)has the shortcomings of low search accuracy and falling into local optimum easily.To solve these problems,a social learning beetle swarm algorithm(SLBSA)is proposed in this paper.Firstly,the Latin hypercube sampling(LHS)is used to generate the beetle swarm with better diversity.Secondly,the iterative process is compared with the biological foraging process,and the chaotic migration strategy based on similarity is adopted to avoid the beetle swarm falling into the local optimum.Finally,the social learning strategy is introduced to ensure the full and efficient information exchange during the updating process.In order to verify the effectiveness,the proposed SLBSA is tested in comparison with the beetle swarm optimization(BSO)and the particle swarm optimization(PSO)on the benchmark functions.Experimental results show that the SLBSA has the faster convergence speed and stronger global search capability.
引文
[1]JIANG X Y,LI S.BAS:Beetle antennae search algorithm for optimization problems[J].Interna-tional Journal of Robotics and Control,2018,1(1):1-4.
[2]JIANG X Y,LI S.Beetle antennae search without parameter tuning(BAS-WPT)for multi-objective optimization[J].Eprint Ar Xiv,2017,1(1):1-4.
[3]邵良杉,韩瑞达.基于天牛须搜索的花朵授粉算法[J].计算机工程与应用,2018,54(18):188-194.
[4]陈婷婷,殷贺,江红莉,等.基于天牛须搜索的粒子群优化算法求解投资组合问题[J].计算机系统应用,2019,28(2):171-176.
[5]赵玉强,钱谦.一类带学习与竞技策略的混沌天牛群搜索算法[J].通信技术,2018,51(11):60-66.
[6]LEIGH R.Genetic algorithms in engineering systems[J].Computing and Control Engineering,1997,9(2):80.
[7]韦鹏飞,徐永海,王金浩,等.基于拉丁超立方采样的节点敏感设备暂降免疫水平评估[J].电工技术学报,2018,33(15):3415-3425.
[8]张晶,熊晓雨,鲍益波.基于KCF相似度的TLD目标跟踪算法[J].计算机工程与科学,2019,41(2):293-301.
[9]马理胜,张均东,任光,等.基于混沌迁移及无参数变异差分进化算法的舰船电力系统网络重构[J].上海海事大学学报,2015,36(3):76-81.
[10]赵耿,李琬璐,马英杰,等.基于混沌序列降低FBMC系统峰均比方法的研究[J].计算机应用与软件,2018,35(9):195-198,204.
[11]CHENG R,JIN Y C.A social learning particle swarm optimization algorithm for scalable optimization[J].Information Sciences,2015,291:43-60.
[12]张伟,宋学官,石茂林,等.基于代理模型的机械式挖掘机动臂轻量化设计[J].机械设计与制造,2019(4):1-4.
[13]田杰,谭瑛,孙超利,等.代理模型辅助进化算法在高维优化问题中的应用[J].机械设计与制造,2018,334(12):277-280.
[14]董健,钦文雯,李莹娟,等.基于改进反向传播神经网络代理模型的快速多目标天线设计[J].电子与信息学报,2018,40(11):2712-2719.
[15]WANG H D,JIN Y X.A random forest-assisted evolutionary algorithm for data-driven constrained multiobjective combinatorial optimization of trauma systems[J].IEEE Transactions on Cybernetics,2018:1-14.
[2]JIANG X Y,LI S.Beetle antennae search without parameter tuning(BAS-WPT)for multi-objective optimization[J].Eprint Ar Xiv,2017,1(1):1-4.
[3]邵良杉,韩瑞达.基于天牛须搜索的花朵授粉算法[J].计算机工程与应用,2018,54(18):188-194.
[4]陈婷婷,殷贺,江红莉,等.基于天牛须搜索的粒子群优化算法求解投资组合问题[J].计算机系统应用,2019,28(2):171-176.
[5]赵玉强,钱谦.一类带学习与竞技策略的混沌天牛群搜索算法[J].通信技术,2018,51(11):60-66.
[6]LEIGH R.Genetic algorithms in engineering systems[J].Computing and Control Engineering,1997,9(2):80.
[7]韦鹏飞,徐永海,王金浩,等.基于拉丁超立方采样的节点敏感设备暂降免疫水平评估[J].电工技术学报,2018,33(15):3415-3425.
[8]张晶,熊晓雨,鲍益波.基于KCF相似度的TLD目标跟踪算法[J].计算机工程与科学,2019,41(2):293-301.
[9]马理胜,张均东,任光,等.基于混沌迁移及无参数变异差分进化算法的舰船电力系统网络重构[J].上海海事大学学报,2015,36(3):76-81.
[10]赵耿,李琬璐,马英杰,等.基于混沌序列降低FBMC系统峰均比方法的研究[J].计算机应用与软件,2018,35(9):195-198,204.
[11]CHENG R,JIN Y C.A social learning particle swarm optimization algorithm for scalable optimization[J].Information Sciences,2015,291:43-60.
[12]张伟,宋学官,石茂林,等.基于代理模型的机械式挖掘机动臂轻量化设计[J].机械设计与制造,2019(4):1-4.
[13]田杰,谭瑛,孙超利,等.代理模型辅助进化算法在高维优化问题中的应用[J].机械设计与制造,2018,334(12):277-280.
[14]董健,钦文雯,李莹娟,等.基于改进反向传播神经网络代理模型的快速多目标天线设计[J].电子与信息学报,2018,40(11):2712-2719.
[15]WANG H D,JIN Y X.A random forest-assisted evolutionary algorithm for data-driven constrained multiobjective combinatorial optimization of trauma systems[J].IEEE Transactions on Cybernetics,2018:1-14.