一种引入反向搜索机制果蝇优化算法研究.PDFVIP

安徽农业大学学报, 2018, 45(5): 966-970 Journal of Anhui Agricultural University [DOI] 10.13610/ki.1672-352x001 网络出版时间：2018/9/19 9:08:58 [URL] /kcms/detail/34.1162.s0942.002.html 一种引入反向搜索机制的果蝇优化算法研究 1 2 1 3 王咏梅 ，刘 磊 ，施培蓓 ，张 健 (1. 合肥师范学院计算机学院，合肥 230031 ；2. 合肥市公安信息中心，合肥 230001 ；3. 安徽大学计算机学院，合肥 230601) 摘 要：针对基本果蝇优化算法求解复杂优化问题时全局搜索能力差，种群多样性偏低等问题，提出一种引入 反向搜索机制的果蝇优化算法（RFOA ）。该算法通过在搜索趋于停滞时计算果蝇个体和进化方向的夹角，挑选出 一批和进化方向相反的果蝇个体并利用这些个体去探索新的最优解，从而跳出局部最优。通过标准测试函数进行仿 真测试，实验结果证明，解决部分较为复杂的优化问题时，RFOA 相比其他改进算法可以更有效地避免早熟收敛， 加快收敛速度，提高收敛精度。 关键词：果蝇优化算法；函数优化；反向搜索 中图分类号：TP301 文献标识码：A 文章编号：1672352X (2018)05096605 Research on fruit fly optimization algorithm with reverse searching mechanism 1 2 1 3 WANG Yongmei , LIU Lei , SHI Peibei , ZHANG Jian (1. School of Computer Science and Technology, Hefei Normal University, Hefei 230031; 2. Hefei Public Security Information Center, Hefei 230001; 3. Computer College, Anhui University, Hefei 230601) Abstract: The fruit fly optimization algorithm (FOA) was widely used to solve complex optimization prob- lem. However, its limited search space and swarm diversity weak its global search ability. To tackle this limitation, we proposed a novel reverse searching fruit fly optimization algorithm (RFOA). When the search of the fruit fly optimization algorithm was stagnated in calculating the angle of the fruit fly individual and evolution direction, we picked out a group of fruit fly individuals with opposite evolutionary direction, and use these individuals to explore new optimal solutions，so as to jump out of the local optimal. Simulation experiments were carried out by standar