一种RBF神经网络在某冷连轧机组轧制力计算中的组合应用

Application of the combination of a sort of RBF neural network in roll force calculation of certain tandem cold mill

为提高冷连轧机组轧制力计算精度,采用RBF神经网络与数学模型计算轧制力,轧制力偏差在11%以内,远小于单纯使用数学模型;而后组合使用长期数据修正网络、短期数据修正网络、速度修正网络(三者都为RBF神经网络)进一步提高轧制力预报精度。上述方法已成功用于某2130mm冷连轧机组过程控制系统,误差在±5%以内。这充分证明,RBF网络可以成功用于轧制过程控制。

To improve precision of roll force calculation of TCM, first RBF neural network and mathematical model is used to calculate roll force, the error is within 11% and is less than using mathematical model only; and then a combination of long-time-data correcting network,short-time-data correcting network and speed correcting network（all of these three are RBF neural networks） is used to improve precision more. The method above has already been used in process control system of certain 2130turn TCM successfully and the error is within 50%. It proves well that RBF neural network can be used in rolling process control successfully.