摘要
由于传统的人工萤火虫算法(Glowworm Swarm Optimization,GSO)存在收敛速度慢和易陷入小区域搜索等缺点,提出了一种基于极值域萤光素值人工萤火虫算法(Extreme-Domain Glowworm Swarm Optimization,EDGSO)。在萤光素值更新过程中,算法限定了荧光素的变化范围,修改萤光素值极值域的上下限,以防止过早的陷入局部搜索。仿真实验的结果证明新的算法既能有效的防止过早陷入局部最优解的现象发生,同时又具备较强的全局搜索能力,进而使传统的人工萤火虫算法的性能得到较大的优化。
The traditional algorithm of glowworm swarm optimization has low convergence speed and iseasy to fall into the small area search, this paper proposes a Glowworm swarm optimization algorithm basedon extreme-domain luciferin. In the updating process of the luciferase value, the algorithm limits the scope ofthe fluorescein, modifies the range bound of the luciferin value, so as to avoid the algorithm falling into alocal optimum. Through the typical function test, experimental results show that the proposed algorithm hasstrong global search ability, and can avoid prematurity effectively, thereby improving the overall performanceof the glowworm algorithm.
出处
《控制工程》
CSCD
北大核心
2017年第1期148-152,共5页
Control Engineering of China
基金
湖北省教育厅科学技术研究计划指导性项目(B2014202)
关键词
极值域荧光素值
萤火虫算法
函数优化
Extreme-domain luciferin
glowworm swarm optimization
function optimization