Optimization of FHB5.10 Electrical Appliance Shell Injection Molding Process Based on Moldflow and BP Neural Network

The purpose of this article is to reduce the warpage deformation of thin-walled injection molded products and improve the prediction ability of their warpage deformation. Aiming at the thin-walled shell of FHB5.10 electrical apparatus with complex internal structure and many forming features such as porous, groove and scoliosis, the four-sided mosaic core-pulling mechanism is used to complete the forming of its side features in the die design. The key point is that the finite element simulation of each scheme in the orthogonal test is carried out by using Moldflow software, and the influence degree of each variable on the warping deformation of plastic parts is in the following order: holding time E > melt temperature A > mold temperature B > injection time C > cooling time F > holding pressure D. Finally, each process scheme of the orthogonal test is used as sample data to train the BP neural network. The results show that using BP neural network to predict the warping deformation of plastic parts has high prediction accuracy, which provides a reference for the production of injection molds.