基于神经网络的注塑成型工艺优化

介绍了一种基于CAE把Taguchi实验设计方法和神经网络结合使用的注塑成型工艺优化方法，并通过一个简单的实例对该方法的可行性进行了验证。结果表明：神经网络结合Taguchi实验设计方法的优化算法，可以对注射成型过程中的注射压力最大值进行优化和预测；在进行最少次实验的结果上给出最佳实验因素水平组合，确定出最佳实验条件，并将实验因素对实验目标的影响大小排序，由此获得较重要的实验因素，从而进行注塑成型工艺优化及控制。

Optimization of Injection Molding Processing Based on Artificial Neural Network

A new method, which is the combination of Taguchi DOE method and BP ANN, was introduced in this paper; it could optimize the variables of injection molding processing in an easy and hurry way. An orthodoxy matrix was used in Taguchi method to reduce the quantity of the experiments that had to be carried out. In the results we could find out the optimal level and the influence degree of each experiment variable. BP ANN was used for its powerful non-linear and time-varied abilities to predict the values of the process variables in injection molding procedure. The combination of these two methods could get the best of each other and make the optimization more efficiently. An example was showed to prove its ability.