基于灰色关联和Kriging模型的汽车饰件注塑工艺优化  

Process Parameter Optimization of Injection Molding for Automotive Interior Parts Based on Grey Relational and Kriging Model

作　　者：张庆[1] 葛东东[1,2] 何也能 ZHANG Qing;GE Dongdong;HE Yeneng(Zhejiang Industry Polytechnic College,Shaoxing,Zhejiang 312000,China;University of Shanghai for Science and Technology,School of Mechanical and Electrical Engineering,Shanghai 200093,China)

机构地区：[1]浙江工业职业技术学院,绍兴312000 [2]上海理工大学机械工程学院,上海200093

出　　处：《塑料》2023年第1期180-186,共7页Plastics

基　　金：浙江省教育厅一般科研项目(Y202043858)。

摘　　要：为减小某汽车内饰件注塑成型质量缺陷,提出了基于灰色关联分析和Kriging代理模型的注塑成型质量多目标优化方法。以塑件的Z方向翘曲变形值、体积收缩率、缩痕指数为优化目标,影响了塑件成型质量的熔体温度、模具温度、保压压力、保压时间等注塑工艺参数,应用最优拉丁超立方试验设计方法结合模流分析进行数据采样,基于灰色关联分析(GRA),将多目标优化问题转化为单目标优化问题,利用层次分析法(AHP)确定各优化目标对灰色关联度的影响权重,建立灰色关联度与各工艺参数之间的Kriging代理模型,采用自适应粒子群算法(APSO)在代理模型内进行全局寻优,得到最佳工艺参数组合。结果表明,Kriging代理模型的预测值与实际结果基本吻合,优化后,Z方向翘曲变形值、体积收缩率、缩痕指数分别为0.2686 mm、2.235%、0.3971%,与优化前相比,分别降低了1.21%、30.11%、68.53%,因此,提出的方法能实现注塑成型质量多目标优化。In order to reduce the quality defects of injection molding for a certain automobile interior parts,a multi-objective optimization method of injection molding quality based on gray relational analysis and Kriging agent model was proposed.Taking the Z-direction warpage deformation value,volume shrinkage rate and sink mark index as the optimization targets,the injection molding process parameters,such as,melt temperature,mold temperature,holding pressure,holding time,which could affect the molding quality of plastic parts,were used as test influencing factors.The optimal Latin hypercube test design method combined with mold flow analysis was used for data sampling.Based on grey relational analysis(GRA),the multi-objective optimization problem was transformed into a single-objective optimization problem,and the influence weights of various optimization targets on the gray relational grade were determined by analytic hierarchy process(AHP),then a Kriging agent model between gray relational grade and various process parameters was established,finally adaptive particle swarm optimization(APSO)was used for global optimization in the agent model to obtain the best combination of process parameters.The results showed that the predicted value of the Kriging agent model was basically consistent with the actual result,and after optimization,the Z-direction warpage deformation,the volume shrinkage rate and the sink mark index were 0.2686 mm,2.235%,0.3971%,respectively.Compared with before optimization,the reductions were 1.21%,30.11%,and 68.53%,respectively.The above indicated that the proposed method could effectively achieve multi-objective optimization of injection molding quality.

关 键 词：Kriging代理模型 灰色关联分析 汽车内饰件 注塑成型 多目标优化 

分 类 号：TQ320.66[化学工程—合成树脂塑料工业]