论文部分内容阅读
本文论述钢锭轧前过程动态模型的开发。以具有自由边界和可变边界的不稳定导热偏微分方程和相应的边界条件为背景,通过时间及空间时间离散化的方法,将一个无限维的分布参数问题描述为有限维的离散状态空间模型。模型给出钢锭在轧前过程的热状态。经大型工业试验验证,证明摸型精度能满足工程要求,从而奠定了现代控制理论和系统理论应用的基础。以所开发的状态空间模型为基础,本文还研究了降价状态观测器及卡尔曼滤波器在实现均热炉内钢锭温度分布在线估计中的应用。最后,论述为实现均热炉最小能耗的高级计算机控制策略及最调度。所论述的研究方法同样可以应用于大多数分布参数过程,如连续加热炉、换热器,热处理及固定床反应器等。
This article discusses the development of the dynamic model before ingot rolling process. Based on the unstable thermal differential equations and the corresponding boundary conditions with free boundary and variable boundary, the problem of an infinite-dimensional distribution parameter is described as a finite-dimensional discrete state space model by time and space time discretization . The model gives the hot state of the ingot during the rolling process. The large-scale industrial test proved that touch type accuracy can meet the engineering requirements, thus laying the foundation for modern control theory and system theory application. Based on the developed state-space model, this paper also studies the application of the reduced-price state observer and Kalman filter in on-line estimation of ingot temperature distribution in a soaking furnace. Finally, it discusses the advanced computer control strategy and the most scheduling to achieve the minimum energy consumption of soaking furnace. The methodology discussed also applies to most distributed parameter processes such as continuous furnaces, heat exchangers, heat treatment and fixed bed reactors.