摘要
在现有自适应粒子群优化算法的研究基础上本文引入1种反弹机制(Rebound Mechanism),提出了1种改进的粒子群算法——反弹自适应粒子群优化算法。RAPSO能在搜索过程中充分利用粒子的飞行速度和方向等信息(下文称为动量信息),维持粒子的多样性以提升算法的搜索性能。通过比较,本文提出的RAPSO在一定程度上改进了现有的自适应粒子群算法的优化性能。运用RAPSO对催化裂化装置进行优化试验,其结果表明无论在单变量优化还是在多变量优化中,该装置的转化率都得到了一定程度的提高。
This paper presents a new method named Rebound Mechanism for improving traditional Particle Swarm Optimization(PSO)algorithm. Based on this method and a kind of Adaptive PSO algorithm this paper proposed a modified PSO algorithm named Rebound Adaptive Particle Swarm Optimization(RAPSO)algorithm.The proposed approach can maintain the population diversity and improve the PSO algorithm by taking full advantage of the momentum information,such as flight speed and orientation of particles.According to comparison,the RAPSO improved existing PSO algorithm in some degree.The results of a Fluid Catalytic Cracking(FCC)unit optimization experiment applied RAPSO indicated that the conversion rate increased a lot regardless of applying single-objective optimization or multiple-objective optimization.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2011年第6期745-748,共4页
Computers and Applied Chemistry
基金
中央高校基本科研业务费专项资金(2010121047)