Pattern matching in historical batch data using PCA

Author

Singhal, Ashish ; Seborg, Dale E.

Author_Institution

Dept. of Chem. Eng., California Univ., Santa Barbara, CA, USA

Volume

22

Issue

5

fYear

2002

fDate

10/1/2002 12:00:00 AM

Firstpage

53

Lastpage

63

Abstract

The article seeks to answer the question: how can relevant information be extracted from huge historical databases? A pattern-matching methodology has been evaluated in a case study for a batch fermentation process. The proposed approach is both data driven and unsupervised. The new approach relies on PCA and a new similarity factor based on distance between the two datasets. The computational requirements are modest, allowing large databases to be processed in a relatively small amount of time.

Keywords

batch processing (industrial); data mining; fermentation; pattern matching; principal component analysis; statistical process control; Tennessee Eastman Challenge Process; acetone-butanol; batch fermentation process; data driven approach; historical batch data; huge historical databases; pattern-matching methodology; simulated chemical reactor; unsupervised approach; Chemicals; Data engineering; Data mining; Databases; Industrial plants; Manufacturing; Pattern matching; Preventive maintenance; Principal component analysis; Production;

fLanguage

English

Journal_Title

Control Systems, IEEE

Publisher

ieee

ISSN

1066-033X

Type

jour

DOI

10.1109/MCS.2002.1035217

Filename

1035217