多策略改进的麻雀搜索算法  

Multi-strategy improved sparrow search algorithm

作　　者：回立川[1] 李瑶 李欢欢 于淼 王久阳 HUI Lichuan;LI Yao;LI Huanhuan;YU Miao;WANG Jiuyang(Faculty of Electrical and Control Engineering,Liaoning Technical University,Huludao 125105,China;Huludao Electric Power Supply Company,State Grid Liaoning Electric Power Company Limited,Huludao 125000,China)

机构地区：[1]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105 [2]国网辽宁省电力有限公司葫芦岛供电公司,辽宁葫芦岛125000

出　　处：《辽宁工程技术大学学报（自然科学版）》2023年第6期722-732,共11页Journal of Liaoning Technical University (Natural Science)

基　　金：辽宁省高等学校基本科研项目(LJ2017QL009);辽宁工程技术大学学科创新团队资助项目(LNTU20TD-32)。

摘　　要：针对麻雀搜索算法在迭代收敛时易陷入局部最优的问题,提出多策略改进的麻雀搜索算法(NLSSA)。利用邻域重心反向学习策略优化麻雀算法的初始种群,提高初始个体质量。通过Levy飞行策略的长短距离跳跃更新麻雀生产者位置,从而提升麻雀算法的局部极值逃逸能力。在跟随者位置更新机制中引入自适应权重,从而平衡麻雀算法的局部挖掘和全局寻优能力。为了验证所提NLSSA算法的性能,利用8个基准测试函数进行测验,测试结果与Wilcoxon符号秩检验结果表明,与麻雀搜索算法、粒子群优化算法、灰狼优化算法和其他改进的麻雀搜索算法相比,NLSSA算法在寻优精度、稳定性能和收敛速度方面的效果更佳。Aiming at the problems that the sparrow search algorithm tends to fall into local optimization during iterative convergence,a multi-strategy improved sparrow search algorithm(NLSSA)is proposed.Firstly,the neighborhood center of gravity reverse learning strategy is used to optimize the initial population of the sparrow algorithm and improve the initial individual quality,and then,the long and short distance jump of the Levy flight strategy is used to update the sparrow producer position,thereby improving the local extremum escape ability of the sparrow algorithm,and finally,the adaptive weight is introduced in the follower position update mechanism,so as to balance the local mining and global optimization ability of the sparrow algorithm.In order to verify the performance of the proposed NLSSA algorithm,this paper uses eight benchmark functions to test,and the test results and the Wilcoxon symbolic rank test show that the NLSSA algorithm has higher search accuracy,stability performance and convergence speed than sparrow search algorithm,particle swarm algorithm,grey wolf optimizer and other improved sparrow search algorithms.

关 键 词：麻雀搜索算法 邻域重心反向学习 Levy飞行策略 自适应权重 基准函数 

分 类 号：O348.8[理学—固体力学]