摘要
用神经元网络的方法代替传统轧制力模型的计算。解决与精轧机组设定模型有关的轧制力预报问题。经实测数据，传统方法（设定模型）与神经元网络方法获得的数据统计比较表明，神经元网络方法所得预报值优于传统方法。
This paper discusses using the neural network to predict the rolling force which is the key-point of the set-up model for the finisher of hot strip mill. The statistic comparison of the measured rolling force of 1700 mm hot strip mill with the predicted value from the neural network and also with the predicted value from traditional models, shows that the neural network get better result than traditional models.
出处
《钢铁》
CAS
CSCD
北大核心
1996年第1期54-57,39,共5页
Iron and Steel
关键词
神经网络
BP网络算法
轧制力
预报
轧机
neural network, BP network calculus, prediction of rolling force
