A Two-Stage Method for Identification of Dual-Rate Systems with Fast Input and Very Slow Output
A Two-Stage Method for Identification of Dual-Rate Systems with Fast Input and Very Slow Output
复制标题
快速输入和极慢输出双速率系统的两阶段识别方法
DOI：
10.1021/ie800948g
复制
发表时间：
2009-01
影响因子：
4.2
通讯作者：
Zhao, Jun
中科院分区：
文献类型：
--
作者：
Shao, Zhijiang;Chen, Xi;Qian, Jixin;Mo, Shengyong;Zhao, Jun
具有快速输入和非常缓慢输出的双速率系统在化学过程工业中非常常见。针对基于输出误差法的快速单速率模型辨识问题，提出了一种结合初始化和局部极小值处理技术的两阶段辨识方法。在其第一阶段，有限脉冲响应（FIR）模型直接从原始的双速率采样数据使用组合最小二乘准则和牛顿-拉夫森（N-R）方法。在第二阶段中，一个单一的和快速的速率输出误差模型的基础上开发的原始快速输入信号和推导出的慢速率模型的输出。除仿真实例外，该方法还应用于开发工业精馏塔的组成观测器。
A dual-rate system with fast input and very slow output is very common in the chemical process industry. To identify its original fast single-rate model based on the output error method, a two-stage method combining techniques for dealing with initialization and local minimum problems is proposed in this paper. In its first stage, a finite impulse response (FIR) model is developed directly from the original dual-rate sampled data using the combined least-squares criterion and Newton−Raphson (N−R) method. In the second stage, a single-and-fast-rate output error model is developed on the basis of the original fast input signal and the derived slow rate model output. Besides the simulation examples, the method is also applied to develop a composition observer for an industrial distillation column.