三层铝板复合轧制包覆率的神经网络模型

An Artificial Network Mkdel for the Pacoing-rate of Three-Layer Aluminum sheet After Rolling

三层铝复合轧制的包覆率与各个工艺参数之间存在着非常复杂的非线性关系，难以用准确的解析式来表达。我们利用对非线性系统逼近能力很强的神经网络进行了建模，形成了相应的工艺参数知识库，能较为准确地对复合轧制后的包覆率进行预测。

The packing-rate of three-layer aluminum after rolling has very complex nonlinear relations with other processing parameters,which is difficult to be described with accurate analytic expressions. Therefore, we use artificial network model to simulate the relations. Which can acculately predict the pocking-rate after rolling.