|
Injection molded plastic products molding way , suitable for mass production of complex shape , the exact size of the plastic products . The better molding process conditions with the continuous improvement of the product quality requirements , how to take advantage of the injection molding process optimization technology is becoming more and more urgent. CAE technology can help engineers optimize the molding process , to reduce repeated tryout repair mode , but this can only be a reasonable process and can not get the optimum process . Therefore, study how to extract from the CAE analysis results effectively to assist in the process optimization of great significance to improve the quality of the products . This article first cover of the printer , for example , the discussion in the case of plastic raw materials , injection molding machine , mold structure determined , the process parameters on the quality of molded articles . Injection molding quality volume shrinkage , warpage and sink marks of index optimization goals packing pressure , mold temperature , melt temperature , injection time , dwell time , as the optimization variables , and practical experience in the range of parameter level. Taguchi method analysis of five molding process parameters on the quality of products , optimize the molding parameters , optimization of parameter combinations , and experimental verification . Statistical regression analysis to analyze the impact of the molding process parameters on the quality of products in the case of the single objective , and compared with the results of the analysis of the Taguchi method , regression analysis, network mapping model between the quality of the products and the molding process parameters , predict the optimal combination of parameters , in the case of multi-objective and experimental verification . Finally, select the elevator pedal protection trim , for example , CAE simulation data for the sample data , through the design of the network structure and learning algorithm used to establish BP artificial neural network - based contraction - warping - settlement plaque index prediction model . Network prediction and experimental results are compared , indicating that the trained neural network model can correctly predict the quality of products , greatly reducing the number of numerical simulation to optimize the molding process , shorten production time and improve part quality purposes .
|