Research On Process Optimization Of Thin-wall Injection Molded Product Integrating Virtual And Real

With the application of non-metallic materials in industries such as aerospace,automobile manufacturing,home plumbing,etc.,thin-walled injection molded products have developed rapidly.Therefore,it is particularly important to ensure high-quality thin-wall injection molded products.The main quality indicators of thin-walled injection molded products are weight and warpage.Moreover,the quality index is related to the mold structure,material selection,process parameters and machine performance.Under the premise of the mold structure,material selection and machine performance determination,when disturbance occurs during mold debugging or machine operation,it is quickly recommended by the optimization system It is particularly important to combine a set of process parameters and input them to the control panel of the machine.This paper mainly studies the optimization of process parameters in the molding process of thin-walled injection products.Based on numerical simulation theory,the mold flow analysis software Moldflow and MATLAB analysis software with fitting and optimization functions are used to simulate and analyze the molding process and process parameters.Optimize and realize the integration of simulation and machine experiment by adjusting the machine parameters of the simulation system.The main work of the paper is as follows:1.Using SPSS software to design the Taguchi orthogonal experiment table and using Moldflow to carry out the mold flow simulation experiment to optimize the mold structure.By means of mean analysis and analysis of variance,the influence trend and order of influence of process parameters on quality indicators were obtained,and the weight and warpage deformation were integrated into one quality indicator using comprehensive coefficient analysis method and analyzed.It can be seen from the degree of influence that the degree of influence on the weight from strong to weak is:holding pressure time,melt temperature,holding pressure and injection time;the degree of influence on the amount of warpage from strong to weak is:holding pressure Time,melt temperature,holding pressure and injection time;the degree of influence on the comprehensive index from strong to weak in turn is:holding pressure time,holding pressure,injection time and melt temperature.2.Use two sets of fitting-optimization algorithms(ie neural network-dynamic control system and response surface-genetic algorithm)to establish the mapping function of process parameters and quality indicators and optimize the function globally.The results show that the order of weight optimization of various strategies from good to poor is as follows:NN-DCS algorithm,RSM-GA algorithm,orthogonal experiment weight combination,orthogonal experiment synthesis coefficient combination,orthogonal experiment warpage Excellent combination,initial recommended combination of mold flow analysis.The order of optimization of various strategies for the amount of warpage deformation from good to poor is as follows:NN-DCS algorithm,orthogonal experiment comprehensive coefficient combination,RSM-GA algorithm,orthogonal experiment warpage optimal combination,orthogonal experiment weight Excellent combination,initial recommended combination of mold flow analysis.Moreover,the optimization performance of the neural network-dynamic control system(NN-DCS)for weight and warpage is the best,but it also has the characteristics of slow operation and complex algorithms.3.Due to differences in material properties and machine parameters,there are differences in the quality results of simulation and machine experiments.Under the premise of determining the material,through analyzing the influence of parameters such as screw diameter,hydraulic response time,hydraulic enhancement ratio and other parameters on the quality index.The results show that the hydraulic response time of the machine parameters has an impact on the quality indicators.Therefore,adjusting the hydraulic response time of the machine parameters of the simulation system provides a way to realize the virtual and real integration of simulation and machine experiments.