Injection Molding Process Simulation And Structure Optimization Of Automotive Front-end Modules

With the continuous improvement of plastic products quality and performance and much more car enterprise use full plastic automotive front-end modules. It is very necessary to research on how to improve the quality of all plastic molded front-end modules through the optimization and improvement of CAE simulation technology for injection molding process all plastic front-end module.This paper use JAC B2 car front-end modules model from GENIUS as the research object. First, create a front-end module finite element analysis model to determine the establishment of its mesh, material selection and cooling systems, make a special effort to study on the effect of different gating system to filling time, flow front temperature, cavitation and weld marks of molding, ultimately determine a relatively optimum gating system. In order to study the impact of Injection time, injection pressure, mold temperature, melt temperature and dwell time on warping deformation and volume shrinkage, established the five factors and four levels orthogonal experiment, use Moldflow respectively simulate each orthogonal test and use the mean range and variance analysis results. Then select the optimum process parameters, and experimental verification. At the same times establish regression equation, the results of orthogonal test analysis and regression analysis are consistent.Subsequently use orthogonal test experimental data as samples to build a predictive model of BP neural network to predict nonlinear relationship between process parameters and laboratory parameters. At the same time use Moldflow to simulate different parameters, and compare predicted results and the simulation results gap. Then the 46 experiments of response surface methodology using BP neural network prediction, obtained forecast data as a result of the response surface method. Then obtain different significant factors by response surface analysis, analysis the combined effects of the significant factors in warping deformation amount and volume shrinkage and generate 3D images directly reflect this influence. Meanwhile obtain the optimal process parameters, and compare with the orthogonal experiment optimal process parameters after experimental verification. Finally, do gray relational grade calculation of response surface experimental results through the BP neural network prediction. Through gray correlation degree analysis of different process parameters for multi-objective optimization, in the same time meet the optimal parameters the minimum amount of warpage and volume shrinkage.