摘要
针对传统粒子群算法(PSO)中影响粒子飞行状态倾向性的选择系数c1、c2固定不变的问题,考虑到微粒子在相应的自身认知能力与全局社会能力影响下能发挥出不同寻优速度与效率这一现象,对影响粒子飞行过程中追寻个体历史最佳与全局位置最佳二者选择倾向性的相关系数进行自适应动态调整。提出一种自适应性粒子群算法(adaptive particle swarm optimization,APSO)。仿真实验表明APSO具有良好的适应性,能够有效提高微粒子寻优的精度与效率,减少算法迭代优化所需时间,提高布控区域的覆盖率。
Considering the selection coefficient cI ,c2 affecting particle flight tendencies are fixed intraditional particle swarm algorithm （PSO） , and that particles exhibit different optimization speedand efficiency influenced by corresponding cognitive ability and global ability, it makes adaptive dy-namic adjustment on correlation coefficient which influences the choice preferrence between the indi-vidual history best and the best overall position during flight trace. It proposes a self-adaptive parti-cle swarm optimization （Adaptive Particle Swarm Optimization, APSO）. Simulation results showthat APSO has better adaptability, and it can effectively improve the accuracy and efficiency of parti-cle optimization, reduce the time required in algorithm iterative optimization, and improve the cover-age of surveillance area.
出处
《信息工程大学学报》
2015年第5期557-561,共5页
Journal of Information Engineering University
基金
国家自然科学基金资助项目(61201380)
国家科技重大专项资助项目(2014ZX03006003)
关键词
选择系数
个体与全局
自适应性
覆盖率
selection coefficient
individual and global
adaptability
coverage
