基于数据分析的化工过程监测  

Process monitoring based on data analysis for chemical engineering

作　　者：李秀喜[1] 钱宇[1] 江燕斌[1] 

机构地区：[1]华南理工大学化工学院,广东广州510640

出　　处：《计算机与应用化学》2004年第4期538-542,共5页Computers and Applied Chemistry

基　　金：国家自然科学基金(20376025);国家杰出青年科学基金(202225620);广东省"千百十"人才基金

摘　　要：为了改善主元分析对带噪声过程的监测性能,本文结合小波包分析消噪性能与主元分析提取变量间相关性能的特点,提出了一种小波包主元分析方法。给出了基于小波包主元分析的过程监测的算法实现。并在此基础上,对 TE 过程进行了监测性能仿真。结果表明小波包主元分析方法有较好的监测性能。To improve monitoring performance for chemical process with noise,in the paper,a wavelet packet principal component a-nalysis(WPPCA)is proposed.It integrates ability of wavelet packet in de-noise and ability of PCA to de-correlate the variables by ex-tracting a linear relationship.An application structure of process monitoring based on the proposed method is given.Finally,the pro-posed approach is successfully applied to Tennessee Eastman process for dynamic monitoring.The simulation result shows that thewavelet packet PCA has better monitoring performance.

关 键 词：过程监测 数据分析 小波包分析 主元分析 

分 类 号：TQ086.3[化学工程]