模具技术Issue(6)：28-35,8.
基于CRITIC权重法和灰色关联分析的微芯片封装多目标参数优化
Multi-objective parameter optimization of microchip package based on CRITIC weight method and grey correlation analysis
摘要
Abstract
The multi-objective optimization method for injection molding process parameters was proposed taking chip packaging as the research object.In order to solve the defects such as volume shrinkage and warping deformation in the injection molding process of microchip packaging,a Taguchi orthogonal experimental design with 5 factors and 4 levels was established with volume shrinkage rate and total warping value as the optimization indexes.Through the simulation analysis of Moldflow,the test results are represented by the signal-to-noise ratio,the weights of two optimization indicators are calculated by the CRITIC weight method,the multi-objective optimization is transformed into a single objective optimization problem by using gray correlation analysis,and the influence degree of 5 process parameters is obtained by range analysis.Mold temperature＞melt temperature＞pressure holding time＞pressure holding pressure＞injection time,and the optimal combination of process parameters is obtained:mold temperature is 155 ℃,melt temperature is 60 ℃,pressure holding time is 5 s,pressure holding pressure is 55 MPa,injection time is 0.9 s.After analysis and verification,the optimized volume shrinkage rate is 2.670％,and the warpage value is 0.026 9 mm.Compared with the parameters recommended by the Moldflow system,the final simulation results are reduced by 8.47％and 21.57％respectively,which prove that this method can improve the quality of plastic parts and effectively reduce the defects in the injection molding process.关键词
Moldflow/微芯片封装/CRITIC权重法/灰色关联分析/正交试验Key words
moldflow/microchip packaging/CRITIC weight method/grey correlation analysis/orthogonal test分类
化学化工引用本文复制引用
赵玲玲,王权,范海霞,张群..基于CRITIC权重法和灰色关联分析的微芯片封装多目标参数优化[J].模具技术,2024,(6):28-35,8.基金项目
天津市科技计划项目(编号:24ZYJDSS00030,22YFYSHZ00190)资助项目,天津市教委科研计划项目(编号:2022ZD023)资助项目. （编号:24ZYJDSS00030,22YFYSHZ00190）
