基于数据驱动的热连轧厚度建模与控制方法

针对轧机厚度机理模型逐渐不满足现有的控制精度要求的现象, 提出了一种基于数据驱动的热轧带钢厚度预测与控制方法.该方法通过对输入空间数据进行在线聚类划分, 在各子空间使用最小二乘支持向量机 (least square support vector machine, LS-SVM) 在线算法建立非线性模型, 并预测系统的输出值, 利用预测控制方法求得控制量, 根据控制器加权策略得到全局控制量.仿真结果证明了该方法的有效性.

Hot Rolling Thickness Modeling and Control Method Based on Data Driven

The thickness of the rolling mill mechanism model does not gradually satisfy the current control accuracy requirements,therefore,a modeling and control method based on the data driven of strip thickness is presented.In this method,the subtractive clustering is adopted to divide the input space into several clusters.Least square support vector machine(LS-SVM) is utilized to estimate the model of the nonlinear system and forecast the output value in each cluster subset.The linear predictive control algorithm is used to implement the predictive control.The global control values are obtained by weighted strategy of the controllers.Numerical simulations show the effectiveness of the proposed method.