Multi-objective Process Optimizations of Injection Molding for Storage Box Frames Based on Grey Correlation Analysis and Entropy Weight Method
Taking the outer frame of the storage box as an example,the Moldflow mold flow analysis software was used to simulate the injection processes of two casting system solutions named cross type and H type.By comparing and analyzing the filling time,flow front temperature,volumetric shrinkage and overall warpage deformation,it is finally determined that the H-type casting system scheme is more reasonable.For the defects of volumetric shrinkage and overall warpage deformation in the initial analysis,the mold temperature,melt temperature,holding time and filling pressure were chosen as the factors to design the orthogonal tests and simulate the injection molding process,finally obtaining the analysis results.The volumetric shrinkage and overall warpage deformation results were processed by signal-to-noise ratio,and the weights of the two were calculated by the entropy weighting method,resulting to obtain the gray correlation degree.Finally,the influence degrees of each process parameter on the gray correlation degree and the optimal combination of process parameters were obtained by extreme difference analysis.The optimal combination of process parameters is shown as follows:a mold temperature of 50 ℃,a melt temperature of 250 ℃,a holding time of 8 s,a filling pressure of 90％.Compared with the initial process,the volume shrinkage under the optimal process parameter combination is decreased by 0.888 percent points,and the overall warpage deformation is reduced by 1.366 mm.The optimal process combination was verified in the actual trial molding.The results show that the surface of the trial molded products has no obvious appearance defects,which meets the dimensional and assembly requirements.Through the combination of entropy weight method and grey correlation analysis,the multi-objective optimization in the optimization processes of injection molding process parameters is successfully realized,proving the feasibility of this method.
万方数据
维普
