论文部分内容阅读
针对某汽车塑件注塑成型时成型末端翘曲量较大导致尺寸变差的问题,结合注塑成型CAE工艺分析后发现,引起产品充填末端翘曲变形大的主要原因为注塑后冷却收缩不均,针对此问题,将CAE仿真分析和RBF神经网络的预测分析相结合,对注塑工艺参数中的保压工艺和冷却工艺进行了优化设计,CAE分析方案采用(冷却+填充+保压+翘曲),RBF神经网络采用聚类法和梯度算法,应用改善翘曲的L_(27)(38)设计试验方案进行神经网络训练和检验,应用混合正交法(L_(36)(2*6 3*2))进行二次水平密化优选参数,通过优化,找到了改善翘曲的注塑工艺方案,优化的注塑工艺方案能较好的指导产品的批量生产,对其它同类注塑产品的生产有较好的实践参考意义。
Aiming at the problem of large size warping caused by large amount of warpage at the molding end of an automobile plastic injection molding, combining the CAE process analysis of injection molding, it is found that the main cause of large warping deformation of the product filling end is uneven cooling shrinkage after injection molding, To solve this problem, CAE simulation analysis and RBF neural network prediction analysis are combined to optimize the pressure-holding process and cooling process in the injection molding process parameters. The CAE analysis program adopts (cooling + filling + packing + warping) The RBF neural network is trained and tested by using the L_ (27) (38) experimental design with improved clustering and gradient algorithms. The mixed orthogonal method (L_ (36) (2 * 6 3 * 2)) for the second time the level of optimization of the parameters of the optimization, found to improve the warpage of the injection molding process, the injection molding process optimization program can better guide the product mass production of other similar injection products are better Practical reference significance.