Process Monitoring and Fault Detection Method Based On Independent Component Analysis

Author

Wu, Yinghua ; Yang, Yinghua ; Qin, Shukai ; Chen, Xiaobo

Author_Institution

Sch. of Inf. Sci. & Eng., Northeastern Univ., Shenyang

Volume

2

fYear

0

fDate

0-0 0

Firstpage

5586

Lastpage

5589

Abstract

Multivariate statistical process control (MSPC) are based upon the assumption that the observed data must be subject to normal probability distribution, which sometimes can not be satisfied in many industrial applications. Independent component analysis (ICA) is a recently developed method, which can overcome the need of the data distribution. In this paper, a new method is introduced for process monitoring and fault detection based on ICA method. Use ICA to extract the independent components, use I², Ie² and SPE charts for fault detection. At last, the simulation results of fractional distillation process monitoring reveal this method is very effective

Keywords

fault diagnosis; independent component analysis; multivariable control systems; normal distribution; process monitoring; statistical process control; data distribution; fault detection; fractional distillation; independent component analysis; multivariate statistical process control; normal probability distribution; process monitoring; Data engineering; Data mining; Fault detection; Independent component analysis; Industrial control; Information science; Monitoring; Principal component analysis; Probability distribution; Process control; fault detection; independent component analysis (ICA); process monitoring;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on

Conference_Location

Dalian

Print_ISBN

1-4244-0332-4

Type

conf

DOI

10.1109/WCICA.2006.1714143

Filename

1714143