一种基于粒子群算法的生产线缓冲区容量分配技术  

PSO-based Production Line Buffer Capacity Allocation Technology

作　　者：刘军[1] 马超 LIU Jun;MA Chao(School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050)

机构地区：[1]兰州理工大学机电工程学院,兰州730050

出　　处：《计算机与数字工程》2018年第3期465-470,484,共7页Computer & Digital Engineering

摘　　要：针对受随机故障等随机事件影响的直线型生产线系统,提出一种多种群粒子群分析技术,解决生产线在缓冲区总量固定、生产率最大的目标条件下的缓冲区容量优化分配技术问题。该技术将一定规模的粒子群平分成多个种群,分别按照ω线性递减策略的粒子群算法的规则进化,并对各种群中粒子群算法附以不同的惯性权重,通过"移民算子"实现各种群的协同进化。既保持各子群进化的独立性,又保证子群间进化的合作性。实验证明,该技术用于小型、大型生产线系统较传统算法具有收敛精度高、鲁棒性好、局部搜索能力强等优点,可以在较小的迭代次数内搜索到全局最优解,对克服早熟收敛有显著效果。For serial line disturbed by random events,an analysis method of multi-population particle swarm optimization is proposed to deal with buffer capacity allocation problem with an objective of maximum productivity for a buffer capacity fixed.A certain number of particle swarms are equally divided into multiple populations which evolve in a rule with theωlinear decreasing strategy.Particle swam optimizers of populations are endowed with different inertia weight and the cooperative coevolution is achieved with immigration operator,which keeps evolution independence and cooperativeness of populations.Numerical results shows that the method is better than traditional algorithms in convergence precision,robustness,local search capability and can ob?tain global optimal solution in less iterations when it is used for small and large-scale production line system and it also shows remarkable result for overcoming premature convergence.

关 键 词：粒子群算法 缓冲区容量分配 直线型生产线 多种群 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]