改进增广最小二乘的轧机垂振系统参数辨识

Parameter Identification of Rolling Mill Vertical Vibration System Based on Improved Augmented Least Squares

数字孪生是物理实体的数字化表现,它通过数学模型实时监测和控制物理实体,因此对数学模型的精度要求非常高。针对轧机垂振系统模型精度低的问题,提出了改进增广最小二乘法对轧机垂振系统进行辨识。建立了垂振系统数学模型,并通过系统离散化推导出动力学方程,确定了辨识模型的结构。以现场采集的数据为基础,采用改进最小二乘法对模型参数进行辨识。实验结果表明,实测振幅与机理模型输出振幅的均方根误差为0.82,而与辨识模型输出振幅的均方根误差只有0.62,有效提高了模型的精度。

The digital twin is the digital representation of the physical entity.It monitors and controls the physical entity in real time through the mathematical model,so the accuracy of the mathematical model is very high.Aiming at the problem of low accu⁃racy of the rolling mill vertical vibration system model,an improved augmented least square method is proposed to identify the rolling mill vertical vibration system.The mathematical model of the vertical vibration system is established,and the dynamic equation is derived through the discretization of the system,and the structure of the identification model is determined.Based on the data collected on site,the improved least square method is used to identify the model parameters.The experimental results show that the root mean square error of the measured amplitude and the output amplitude of the mechanism model is 0.82,while the root mean square error of the output amplitude of the identification model is only 0.62,which effectively improves the accu⁃racy of the model.