Industrial use of multivariate statistical analysis for process monitoring and control

被引:3
作者
Champagne, M [1 ]
Dudzic, M [1 ]
机构
[1] Tembec Inc, Temiscaming, PQ J0Z 3R0, Canada
来源
PROCEEDINGS OF THE 2002 AMERICAN CONTROL CONFERENCE, VOLS 1-6 | 2002年 / 1-6卷
关键词
D O I
10.1109/ACC.2002.1024870
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Multivariate statistical analysis has come of age over the last decade. Dofasco, a leading North American provider of Solutions in Steel((TM)) and Tembec, a multinational, integrated forest products company have applied this technology in many industrial applications. The intent of this paper is to give an introduction to the application of on-line multivariate statistical analysis in industrial settings, and its impact on daily operating costs, product quality, throughput, equipment reliability and environmental impact. This paper will introduce the reader to the basic concepts of multivariate statistical analysis for both continuous and batch processes. As well, the paper examines the types of problems to which multivariate statistical analysis can be applied in an industrial process, and our experiences in deploying robust, on-line applications.
引用
收藏
页码:594 / 599
页数:6
相关论文
共 16 条
[1]  
[Anonymous], ASQC STAT DIV NEWSLE
[2]  
Bendwell N., 2000, Control Systems 2000. Quantifying the Benefits of Process Control. Preprint, P185
[3]  
CHAMPAGNE M, 2000, TAPPI PULP C BOST MA
[4]  
DUDZIC M, 2000, IFAC WORKSH AUG 22 2, P438
[5]  
DUDZIC M, 1999, IEEE ADV PROC CONTR
[6]  
Eriksson L., 1999, INTRO MULT MEGAVARIA
[7]   PARTIAL LEAST-SQUARES REGRESSION - A TUTORIAL [J].
GELADI, P ;
KOWALSKI, BR .
ANALYTICA CHIMICA ACTA, 1986, 185 :1-17
[8]  
Ivanov I., 2000, Control Systems 2000. Quantifying the Benefits of Process Control. Preprint, P291
[9]  
IVANOV I, 2002, CONTROL SYSTEMS 2002
[10]   PROCESS ANALYSIS, MONITORING AND DIAGNOSIS, USING MULTIVARIATE PROJECTION METHODS [J].
KOURTI, T ;
MACGREGOR, JF .
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 1995, 28 (01) :3-21