摘要
在系统试验的基础上建立反向传播(BP)神经网络模型,即BP网络,预测了给定工艺条件下(螺杆转速、注射行程、背压)预塑时的熔体温度分布,并分析不同工艺条件对轴向温差的影响。通过MATLAB建立BP模型,得到加工参数与熔体温度之间的非线性映射关系。试验证明所建网络具有较好的预测效果。发现,轴向温差随螺杆转速和注射行程的增大而增大,随背压的增大而降低。
A BP neural network is established on system experiments. Melt temperature distribution in the preplastic is predicted at prediction processing conditions (screw speed, injection stroke and back pressure). The effect of processing conditions on the axial temperature difference was analysed. Nonlinear mapping relationship between process- ing parameters and melt temperature is obtained through BP model established by MATLAB. The Experiments show that the neural network has a good prediction effect. The results show that axial temperature difference increases with increaseof screw rotation and injection stroke,and decreases with increase of back pressure.
出处
《现代塑料加工应用》
CAS
北大核心
2014年第4期55-57,共3页
Modern Plastics Processing and Applications
关键词
注射成型
反向传递网络
预塑阶段轴向温差
均匀性
injection molding
back propagation neural network
plasticizing phase
axial temperature difference
homogeneity