融合均匀变异与高斯变异的蝙蝠优化算法

发布时间：2018-03-04 02:22

本文选题：蝙蝠算法　切入点：均匀变异　出处：《控制与决策》2017年10期 　论文类型：期刊论文

【摘要】：为提高蝙蝠算法的寻优精度和收敛速度,提出一种融合均匀变异和高斯变异的蝙蝠优化算法.算法引入变异开关函数,该函数使所有蝙蝠个体在任何时期都有概率发生变异,使种群保持较高的多样性和活跃性.同时在算法整个寻优过程中融入均匀变异和高斯变异,两种变异机制共同协作使算法首先快速定位到全局最优解区域,随后完成局部精确搜索.仿真结果表明,改进后的算法寻优性能显著提高,具有较快的收敛速度和较高的收敛精度.
[Abstract]:In order to improve the optimization accuracy and convergence speed of bat algorithm, a bat optimization algorithm combining uniform mutation and Gao Si mutation is proposed. The algorithm introduces mutation switch function, which makes all bat individuals have probability mutation at any time. At the same time, the uniform mutation and Gao Si mutation are incorporated in the whole optimization process of the algorithm. The two mutation mechanisms work together to make the algorithm locate to the global optimal region quickly. The simulation results show that the improved algorithm has better performance, faster convergence speed and higher convergence accuracy.
【作者单位】：河南大学管理科学与工程研究所;河南大学计算机与信息工程学院;河南大学智能网络系统研究所;
【基金】：河南省科技攻关项目(162102110109) 河南省科技攻关重点项目(142102210036)
【分类号】：TP18

【相似文献】