【作者】 黄宴委； 彭铁根； 吴惕华；

【Author】 Huang Yanwei, Peng Tiegen, Wu Tihua (Department of Automation, Shanghai Jiaotong University, Shanghai, 200030 China)

【机构】 上海交通大学自动化系；

【摘要】 为了获得轧制过程的仿真模型,本文利用支持向量机建立轧制力模型,并将所建立的轧制力模型分别对各输入变量进行偏微分,从而进行轧制过程出口厚度灵敏度系数计算。使轧制模型简化,可以解决解析方法难以求解的轧制过程“代数环”问题,有利于在线监控。模型仿真表明基于灵敏度的轧制厚度模型与解析模型对比,具有很高的精度,可以有效、准确地分析在干扰因素作用下各变量的变化情况。更多还原

【Abstract】 In order to obtain a model for simulation of the rolling process, this paper presents that the rolling force is modeled based on support vector machine (SVM), and a method for the calculations of the sensitivity for the rolling process can be obtained by differentiating SVM, which can eliminate the "algebraic loop" in the analytical model for rolling process and can be fit for online supervision. Moreover, the simulation results of the rolling process model by the sensitivity based on SVM compared with those of the analytic model indicate that modeling by sensitivity based on SVM is valid and it can calculate the variations of the variables with the perturbations with high precision.更多还原