基于ICA的工业过程监控方法及其在TEP中的应用  

Industry Process Monitoring Based on ICA and Its Applications in TEP

作　　者：郭旭红[1] 朱忠奎[1] 刘刚[2] 张娟娟[1] 

机构地区：[1]苏州大学机电工程学院,苏州215021 [2]苏州热工研究院,苏州215004

出　　处：《系统仿真学报》2009年第8期2331-2334,共4页Journal of System Simulation

基　　金：江苏省高校自然科学研究计划(08KJB460005)

摘　　要：针对工业过程的监控,论文采用独立分量分析(ICA)方法提取过程的独立分量,并把正常工况下的独立分量构成特征子空间,将待监测的数据投影到该子空间,提出以投影系数的变化作为监控指标的监控方法。仿真应用表明独立分量分析能够提取有效的TEP状态特征,监控指标可以有效的对系统进行监控。For the performance monitoring of the industry process, the ICA was adopted to extract the independent components （ICs）, which represents the current operating condition. Furthermore, these ICs span the characteristic space of normal operating condition （NOC）, and by projection of process variables onto the space of NOC, the changes of coefficients matrix for monitoring were obtained. The presented method was utilized to monitor the Tennessee-Eastman Process （TEP） and the results show that the proposed method based on ICA is effective in showing the process condition.

关 键 词：独立分量分析 过程 监控 TEP 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]