Visual method for process monitoring and its application to Tennessee Eastman challenge problem

Author

Gu, Yi-Ming ; Zhao, W-Hong ; Hui Wang

Author_Institution

Inst. of Syst. Eng., Zhejiang Univ., Hangzhou, China

Volume

6

fYear

2004

fDate

26-29 Aug. 2004

Firstpage

3423

Abstract

Online process monitoring is extremely important for the successful operation of any process. A visual data-based method suitable for online monitoring of complex systems is proposed. The self-organizing map is used to project a high-dimensional vector of process data onto a 2D visualization space in which different process conditions are represented by different regions. The process state can be indicated by the trajectory in visualization space. The effectiveness of the proposed method is illustrated by the application on the Tennessee Eastman process. Online monitoring and fault detection can be carried in a more intuitionistic and practical manner by using this method.

Keywords

chemical industry; control engineering computing; data visualisation; fault diagnosis; large-scale systems; process monitoring; production engineering computing; self-organising feature maps; 2D visualization space; Tennessee Eastman challenge problem; complex systems; fault detection; high-dimensional vector; online process monitoring; self-organizing map; visual data-based method; Application software; Computerized monitoring; Data visualization; Fault detection; Knowledge engineering; Large-scale systems; Modems; Neural networks; Space technology; Systems engineering and theory;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on

Print_ISBN

0-7803-8403-2

Type

conf

DOI

10.1109/ICMLC.2004.1380378

Filename

1380378