摘要
针对电动车头罩出现安装困难的问题,本实验以减小产品翘曲量和体积收缩率为优化目标。首先在正交试验的基础上,运用信噪比和灰色关联度分析得出初步最佳工艺参数;然后以初步最佳工艺参数为基础,影响产品质量最大的四个因素为调整手段,建立正交试验,导入已经训练好的BP神经网络中进行预测,得出最佳工艺参数;最后用CAE进行模拟验证,最佳工艺参数下翘曲量为1.540 mm,体积收缩率为6.709%,符合生产要求。提出的优化方法能够有效提高制品质量,缩短产品生产周期,为多级注塑成型工艺参数的优化提供了一种可靠的解决方案。
Aiming at the problem of difficulty in installing the head hood of electric vehicle, this experiment aims to reduce the volume of warpage and the volume shrinkage of the product as the optimization target. First of all, on the basis of the orthogonal experiment, the initial optimum process parameters are obtained by using the signal to noise ratio and the grey correlation analysis. Then, based on the initial optimal technological parameters, the four factors that affect the quality of products are the orthogonal test, and the BP neural network is trained to predict the best process parameters. Finally, the CAE is used to simulate the results. The warpage is 1.540 mm under the optimum process parameters and the volume shrinkage is 6.709%, which is in line with the production requirements. The proposed optimization method can effectively improve the quality of products, shorten the production cycle, and provide a reliable solution for the optimization of multistage injection molding process parameters.
出处
《塑料科技》
CAS
北大核心
2018年第3期91-96,共6页
Plastics Science and Technology
关键词
电动车头罩
信噪比
灰色关联度
BP神经网络
多级注塑成型
Electric hood cover
Signal-to-noise ratio
Gray relational grade
BP neural network
Multi-stage injection molding
