Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Journal of Zhejiang University(Engineering Science)》 2009-01
Add to Favorite Get Latest Update

Impact of control strategies on multivariate statistical monitoring performance

YE Lu-bin,LIU Yu-ming,LIANG Jun(State Key Laboratory of Industrial Control Technology,Zhejiang University,Hangzhou 310027,China)  
The Tennessee Eastman(TE) process simulating the features of actual complex systems was taken as testing object to analyze the impact of system's control strategies when applying multivariate statistical monitoring methods in complex industrial processes.The similarities and differences between control strategies and the ability of restraining faults were compared under three control strategies including base control,decentralized control and plant-wide control.Fault detection rate was adopted as performance index for monitoring methods.Then the impact of control strategies on monitoring performance was detailedly analyzed in two typical multivariate statistical monitoring methods respectively based on principal component analysis(PCA) and dissimilarity of data.The simulations of 20 faults of TE process indicate that control strategies have significant impact on the monitoring performance,and choosing appropriate modeling variables and model parameters according to the characteristic of control strategies can effectively improve fault detection rate.
【Fund】: 国家自然科学基金资助项目(60574047);; 国家教育部博士点专项基金资助项目(20050335018);; 国家“863”高技术研究发展计划资助项目(2007AA04Z168)
【CateGory Index】: TP277
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved