基于综合关联度的注塑成型工艺参数多目标稳健设计  

The Multi-Objective Robust Design on Craft Parameters of Plastic Injection Based on Synthesis Relation Degree

作　　者：黄风立[1,2] 林建平[1] 钟美鹏[2] 许锦泓[2] 

机构地区：[1]同济大学机械工程学院,上海200092 [2]嘉兴学院机电工程学院,浙江嘉兴314001

出　　处：《高分子材料科学与工程》2011年第3期162-166,共5页Polymer Materials Science & Engineering

基　　金：浙江省科技计划支撑和引导计划面上项目(2007C21041);浙江省教育科研计划项目(Y200909445;Y200907535)

摘　　要：针对注塑成型过程中的多个质量指标,提出了基于综合关联度、Kriging模型及混合交叉变异蚁群算法的注塑成型工艺参数稳健优化方法。首先给出了综合关联度的定义及计算方法,再给出了Kriging模型的拟合方法和混合交叉变异的蚁群算法;在多目标稳健设计建模中,各分目标采用质量工程中的6σ原则;在混合交叉变异的蚁群算法中,提出了基于实验设计方法的算法参数确定方法;结合实例验证了提出的方法具有较好的工程实际意义。Aiming at the multi quality factors in the process of plastic injection,the robust optimum method of craft parameters in the process of plastic injection based on synthesis relation degree,Kriging model and ant colonies algorithm with the operation of crossover and mutation were put forward.Firstly,the definition and calculational method of synthesis relation degree were given,and the approximate model based on Kriging model and ant colonies algorithm with the operation of crossover and mutation were given also.In the robust design modeling of multi objections,the principle of 6 in quality control was applied to each objection.In the ant colonies algorithm with the operation of crossover and mutation,the selective method of algorithm parameters based on the method of experimental design was put forward.The given method is testified by an instance,and it has a good application in engineering.

关 键 词：注塑成型 综合关联度 KRIGING模型 蚁群算法 

分 类 号：TQ320.662[化学工程—合成树脂塑料工业]