分层模糊神经网络的轧钢产品质量建模

Quality Modeling of Steel Rolling Products Based on the Hierarchical Fuzzy Neural Network

轧钢生产过程是极其复杂的大系统,模糊神经网络进行质量建模将不可避免地陷入维数灾。分层模糊神经网络是解决维数灾的有效手段。文中采用分层模糊神经网络对轧钢产品质量进行建模,并用预处理后的数据进行训练和校验。仿真结果表明所建质量模型能够较好地拟合建模数据,表明了该方法的有效性。

The industrial process of steel rolling is extremely complex large scale system.The method of using fuzzy neural network to establish the quality model will inevitablely fall into the curse of dimensionality.The hierarchical fuzzy neural network is an effective mean to solve the curse of dimensionality.In this paper,hierarchical fuzzy neural network is used to establish the quality model of steel rolling process,then the pretreated data is used to training and validation data.The simulation result shows that the model constructed could approximate the modeling data well,and shows that the effectiveness of this method.