摘要
结合人工神经网络所表现出来的良好特性,利用正交试验获得的数据作为神经网络的训练样本,建立输入为工艺参数、输出为翘曲变形量的神经网络模型,并通过样本检验了ANN模型的准确性,从而缩短设定工艺参数的时间,在工艺参数取值范围内,采用ANN模型代替CAE软件模拟试验,结合正交试验法,对工艺参数进一步优化。结果表明:将神经网络与正交试验、数值模拟三者结合用于注射过程参数优化可以缩短优化工艺参数的时间,提高工艺设计效率,并能获得比单纯使用正交试验和数值模拟方法更为优化的结果。
The data got from an orthogonal experiment was used as the training sample to establish a neural network model in which the input is the technological parameters and the output is the warpage amount. The accuracy of the ANN model was proved by the sample. In this way, the time to set the technological parameters was shortened. The ANN model was used to substitute CAE numerical simulation test, and combined with orthogonal experiment method, the technological parameters was further optimized. The result showed that the integration of ANN, orthogonal experiment and numerical simulation method has greatly increased the technological design efficiency.
出处
《模具工业》
北大核心
2007年第7期1-5,共5页
Die & Mould Industry
基金
浙江省教育厅科研项目(20060357)
关键词
注射成型
人工神经网络
工艺参数
优化
翘曲量
injection molding, artificial neural network （ANN）
technological parameter
optimization
warpage amount
