基于改进的BP神经网络无缝钢管连轧轧制力的预测

Prediction on rolling force in continuous rolling of seamless steel pipe based on improved BP neural network

无缝钢管连轧过程具有多变量、强耦合、非线性等特点,传统的数学模型无法对一些参数进行精确地预测。为了提高连轧过程中轧制力预测的精度,采用改进初始值选取方法来优化BP神经网络,建立改进的BP神经网络的轧制力预测模型。首先,采集某钢厂历史生产数据,进行预处理,通过灰色关联度确定影响轧制力的主要因素;然后,对初始值进行设置,利用MATLAB编写仿真程序对连轧机组轧制力进行预测。结果表明:基于改进的BP神经网络的轧制力预测模型具有很强的学习能力和表达能力,轧制力预测精度得到了很大的提高,对实际的生产具有重要意义。

The seamless steel pipe rolling process has the characteristics of multivariable,strong coupling and nonlinear,and the traditional mathematical model can not predict some parameters accurately.Therefore,in order to improve the accuracy of rolling force prediction during rolling process,BP neural network was optimized by the improved initial value selection method,and the improved BP neural network rolling force prediction model was established.First,the historical production data of a steel mill were collected and preprocessed,and the main factors affecting rolling force were determined by grey correlation degree.Then,the initial value was set,and the simulation program was written by MATLAB to predict the rolling force of continuous rolling mill.The results show that the rolling force prediction model based on the improved BP neural network has strong learning ability and expression ability,and the rolling force prediction accuracy is greatly improved,which is of great significance to the actual production.