摘要
本文将均匀搜索粒子群算法(Uniform search Particle Swarm Optimization,简称UPSO)的位置更新公式变换为一个差分方程,求解差分方程得到非递推的位置更新公式,推导解的收敛条件并求出了UPSO对学习系数c及惯性系数w的收敛区域,最后通过6个Benchmark函数仿真实验对收敛区域的正确性进行验证,实验结果表明学习系数和惯性系数在收敛区域内时的UPSO收敛,不在收敛区域外时UPSO发散.
The uniform search particle swarm optimization （UPSO） algorithm formula was tsansformed into a differential e quation. Solving the differential equation, we get a nonrecurrence location update formula, and the UPSO＇ s convergence region for learning coefficient c and inertia coefficient w were concluded by deducing the solution convergence conditions. Finally simulation experiments were provided on the selected location of the region of convergency by 6 Benchmark functions. Experimental results show that UPSO converges when the learning coefficient and inertia/coefficient are in the convergence region and diverge outside convergence region.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2012年第6期1115-1120,共6页
Acta Electronica Sinica
基金
国家自然科学基金(No.11172342)
教育部"新世纪优秀人才支持计划"资助项目(No.NCET-110674)
陕西省自然科学基金项目(No.2012JM8043)
关键词
粒子群算法
均匀搜索粒子群算法
particle swarm optimization
uniform search particle swarm optimization
