首页|多策略混合的改进沙猫群优化算法及其应用

多策略混合的改进沙猫群优化算法及其应用

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
为了增加基本沙猫群算法的搜索效率、收敛效率，增加种群的多样性，增强全局搜索能力，提出一种多策略混合的改进沙猫群算法(improved sand cat swarm optimization algorithm，IMSCSO)。利用Sine映射进行初始化以得到分布更加均匀的种群;在攻击行为中根据个体适应度的大小对攻击区间进行区域划分，减小攻击范围，增加搜索效率;将向量的线性变换引入到搜寻行为中，通过对系数矩阵的设计增加收敛效率;利用聚集度圆增加算法跳出局部最优的能力;通过引入生存机制的黄金正弦策略增强其局部开发的能力。使用12个基本测试函数对改进后的算法进行测试，经过Wilcoxon秩和检测、时间复杂度分析、李亚普诺夫稳定性分析验证改进算法的优越性。最后使用改进沙猫群算法对SVM参数进行优化，并应用于轴承故障检测，证明算法在实际应用中的有效性。
Improved sand cat swarm optimization algorithm based on multi-strategy mixing and its application
In order to improve the search efficiency and convergence efficiency of the basic sand cat swarm algorithm,increase the diversity of the population,and enhance global search capabilities,an improved sand cat swarm optimization algorithm(IMSCSO)with multi-strategy mixing is proposed.Sine mapping is used for initialization to obtain a more evenly distributed population.In the attack behavior,the attack interval is divided according to the size of the individual adaptability to reduce the attack range and increase the search efficiency.The linear transformation of vectors is introduced into the search behavior,and the convergence efficiency is increased by the design of the coefficient matrix.The aggregation circle is used to increase the ability of the algorithm to jump out of the local optimum.Its ability of developing locally is enhanced with golden sinusoidal strategies which cite survival strategies.The improved algorithm is tested by 12 basic test functions,and the superiority of the improved strategy is verified by Wilcoxon rank sum detection,time complexity analysis,and Lyapunov stability analysis.Finally,the improved sand cat swarm algorithm is used to optimize the SVM parameters and applied to bearing fault detection,which proves the effectiveness of the algorithm in practical applications.

sand cat swarm algorithmSine mappingregional divisionlinear transformationaggregation circlegolden sinusoidal strategysurvival mechanisms

回立川、于千皓

展开 >

辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛 125105

沙猫群算法 Sine映射 区域划分 线性变换 聚集度圆 黄金正弦策略 生存机制

辽宁省教育厅辽宁省高等学校基本科研项目辽宁省教育厅辽宁省高等学校基本科研项目

LJ2017QL009LJKMZ20220675

2024

10.13195/j.kzyjc.2023.1095

控制与决策
东北大学

控制与决策

CSTPCD北大核心
影响因子：1.227
ISSN：1001-0920
年,卷(期)：2024.39(10)