Understanding the importance of process alarms based on the analysis of deep recurrent neural networks trained for fault isolation

Gyula Dorgo, Peter Pigler, Janos Abonyi

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The identification of process faults is a complex and challenging task due to the high amount of alarms and warnings of control systems. To extract information about the relationships between these discrete events, we utilise multitemporal sequences of alarm and warning signals as inputs of a recurrent neural network–based classifier and visualise the network by principal component analysis. The similarity of the events and their applicability in fault isolation can be evaluated based on the linear embedding layer of the network, which maps the input signals into a continuous-valued vector space. The method is demonstrated in a simulated vinyl acetate production technology. The results illustrate that with the application of recurrent neural network–based sequence learning not only accurate fault classification solutions can be developed, but the visualisation of the model can give useful hints for hazard analysis.

Original languageEnglish
Article numbere3006
JournalJournal of Chemometrics
Volume32
Issue number4
DOIs
Publication statusPublished - Apr 2018

    Fingerprint

Keywords

  • deep learning
  • fault classification
  • visualisation of discrete events

ASJC Scopus subject areas

  • Analytical Chemistry
  • Applied Mathematics

Cite this