求解多目标优化问题的灰色粒子群算法  

Grey particle swarm algorithm for multi-objective optimization problems

作　　者：于繁华[1] 刘寒冰[2] 戴金波[1] 

机构地区：[1]长春师范学院信息技术学院,吉林长春130032 [2]吉林大学交通学院,吉林长春130025

出　　处：《计算机应用》2006年第12期2950-2952,共3页journal of Computer Applications

摘　　要：鉴于基本粒子群算法无法解决高维多目标优化问题,提出了一种适合求解高维多目标优化问题的灰色粒子群算法(GPSO),该算法根据灰色关联能够很好地分析目标矢量之间的接近程度,并能掌握解空间全貌的特点,利用灰色关联度的大小来选取粒子群算法中的全局极值和个体极值。实验结果证明,该算法可行而有效,同时也拓展了粒子群算法的应用领域。Since the basic particle swarm algorithm cannot solve the problem of high-dimension objective optimization, a Grey Particle Swarm Optimization （GPSO） algorithm for hlgh-dimension objective optimization was proposed. This method can analyze the degree of approach between objective vectors. And the panorama of the solution space can be controlled by this way. The global maximum and individual maximum in the particle swarm algorithm can be selected according to the degree of grey relevancy. Test results show that GPSO is feasible and effective, and it extends the application field of particle swarm algorithm.

关 键 词：灰色粒子群算法 灰色关联 多目标优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]