基于反向学习与混合位置中心的樽海鞘算法  

Salp Swarm Algorithm Based on Opposition-Based Learning and Hybrid Position Center

作　　者：蒋美琪 杨兴 罗聪敏 JIANG Meiqi;YANG Xing;LUO Congmin(College of Information Science and Technology,Chengdu University of Technology,Chengdu 610059 China)

机构地区：[1]成都理工大学信息科学与技术学院,四川成都610059

出　　处：《西华大学学报（自然科学版）》2021年第1期17-21,74,共6页Journal of Xihua University:Natural Science Edition

摘　　要：为解决樽海鞘算法收敛速度慢和易陷入局部最优解的问题,引入不同的优化策略,提出一种基于反向学习与混合位置中心相结合的樽海鞘算法。该算法在更新领导者位置时进行反向学习;在每次迭代更新中引入位置加权中心和位置均值中心,通过比较不同位置中心值的优劣程度得到更佳食物源位置。在8个不同类型的基准测试函数上的实验结果表明,改进的樽海鞘算法在求解精度、收敛速度上均有明显的提高,且具有更佳的鲁棒性。In order to improve Salp swarm algorithm convergence velocity and result precision,different optimization strategies are introduced,and an improved Salp swarm algorithm based on oppositionbased learning and hybrid position center is proposed in this paper.The improved Salp swarm algorithm selects leaders for opposition-based learning,introduces a position weighted center and a position mean center in each iteration update,and obtains a better food source position by comparing the advantages and disadvantages of different position center values.Through the experimental analysis of 8 different types of benchmark test functions,we found that the improved Salp swarm algorithm has not only a significant improvement in solution accuracy and convergence speed,but also has better robustness.

关 键 词：启发式算法 樽海鞘算法 反向学习 均值中心 位置加权中心 全局最优 寻优精度 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP301.6[自动化与计算机技术—控制科学与工程]