Reconstruction based fault prognosis for continuous processes

Gang LI, S. Joe QIN*, Yindong JI, Donghua ZHOU

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

120 Citations (Scopus)

Abstract

In this paper, a multivariate fault prognosis approach for continuous processes with hidden faults is proposed based on statistical process monitoring methods and multivariate time series prediction. It is assumed that the fault is a slowly time-varying autocorrelated process and can be completely reconstructed. Fault magnitude is estimated first via reconstruction, then predicted by a vector AR model with wavelet based denoising. Given the fault direction, a new index is proposed to detect the fault, which integrates fault detection and prognosis together. Case studies on a continuous stirred tank reactor and the Tennessee Eastman process demonstrate the effectiveness of the proposed approaches. © 2010 Elsevier Ltd.
Original languageEnglish
Pages (from-to)1211-1219
Number of pages9
JournalControl Engineering Practice
Volume18
Issue number10
Early online date20 Jun 2010
DOIs
Publication statusPublished - Oct 2010
Externally publishedYes

Funding

This work was supported by National 973 project under Grants 2010CB731800 and 2009CB320602 and the NSFC under Grants 60721003, 60736026 and 60931160440. S. Joe Qin acknowledges the financial support from the Changjiang Professorship by the Ministry of Education of PR China.

Keywords

  • Multivariate fault prognosis
  • Principal component analysis
  • Reconstruction based estimation
  • Vector AR model
  • Wavelet based denoising

Fingerprint

Dive into the research topics of 'Reconstruction based fault prognosis for continuous processes'. Together they form a unique fingerprint.

Cite this