PLS回归软测量方法在催化重整稳定油组分估计中的应用  

Application of Soft Sensing Method Based on PLS Regression for Estimating the Stabilizing Oil Components of Catalytic Reform Reactor

作　　者：刘瑞兰[1] 苏宏业[1] 褚健[1] 

机构地区：[1]工业控制技术国家重点实验室,浙江大学先进控制研究所,浙江杭州310027

出　　处：《化工自动化及仪表》2002年第5期44-47,共4页Control and Instruments in Chemical Industry

基　　金：国家杰出青年科学基金 ( 6 0 0 2 530 8);浙江省自然科学重点基金 (ZD990 5)资助项目

摘　　要：提出扰动分类法和线性部分最小二乘 (PLS)回归相结合的建立软测量模型的方法 ,并将它用于催化重整稳定油组分的估计中。仿真结果表明扰动分类法和PLS回归相结合建立的软测量模型简单。In this paper,a combined approach of disturbance classification and linear partial least squares (PLS) regression for modeling soft sensor is presented.The strategies for applying it to modeling the complex non linear process are described in detail by an example in which the stabilizing oil components of catalytic reform reactor are estimated.The simulation results prove the method for modeling soft sensor to be simple and practical.

关 键 词：组分估计 软测量 催化重整 PLS 成品油 

分 类 号：TE624.42[石油与天然气工程—油气加工工程] TE626.25