Comparative analysis of Granger causality and transfer entropy to present a decision flow for the application of oscillation diagnosis

Show simple item record

dc.contributor.author Lindner, Brian
dc.contributor.author Auret, Lidia
dc.contributor.author Bauer, Margret
dc.contributor.author Groenewald, Jeanne W.D.
dc.date.accessioned 2019-06-18T13:28:53Z
dc.date.issued 2019-07
dc.description.abstract Causality analysis techniques can be used for fault diagnosis in industrial processes. Multiple causality analysis techniques have been shown to be effective for fault diagnosis. Comparisons of some of the strengths and weaknesses of these techniques have been presented in literature. However, there are no clear guidelines on which technique to select for a specific application. These comparative studies have not thoroughly addressed all the factors affecting the selection of techniques. In this paper, these two techniques are compared based on their accuracy, precision, automatability, interpretability, computational complexity, and applicability for different process characteristics. Transfer entropy and Granger causality are popular causality analysis techniques, and therefore these two are selected for this study. The two techniques were tested on an industrial case study of a plant wide oscillation and their features were compared. To investigate the accuracy and precision of Granger causality and transfer entropy, their ability to find true connections in a simulated process was also tested. Transfer entropy was found to be more precise and the causality maps derived from it were more visually interpretable. However, Granger causality was found to be easier to automate, much less computationally expensive, and easier to interpret the meaning of the values obtained. A decision flow was developed from these comparisons to aid users in deciding when to use Granger causality or transfer entropy, as well as to aid in the interpretation of the causality maps obtained from these techniques. en_ZA
dc.description.department Electrical, Electronic and Computer Engineering en_ZA
dc.description.embargo 2020-07-01
dc.description.librarian hj2019 en_ZA
dc.description.uri http://www.elsevier.com/locate/jprocont en_ZA
dc.identifier.citation Lindner, B., Auret, L., Bauer, M. et al. 2019, 'Comparative analysis of Granger causality and transfer entropy to present a decision flow for the application of oscillation diagnosis', Journal of Process Control, vol. 79, pp. 72-84. en_ZA
dc.identifier.issn 0959-1524 (print)
dc.identifier.issn 1873-2771 (online)
dc.identifier.other 10.1016/j.jprocont.2019.04.005
dc.identifier.uri http://hdl.handle.net/2263/70231
dc.language.iso en en_ZA
dc.publisher Elsevier en_ZA
dc.rights © 2019 Elsevier Ltd. All rights reserved. Notice : this is the author’s version of a work that was accepted for publication in Journal of Process Control. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. A definitive version was subsequently published in Journal of Process Control, vol. 79, pp. 72-84, 2019. doi : 10.1016/j.jprocont.2019.04.005 . en_ZA
dc.subject Transfer entropy en_ZA
dc.subject Granger causality en_ZA
dc.subject Causality analysis en_ZA
dc.subject Fault diagnosis en_ZA
dc.subject Plant wide oscillation en_ZA
dc.subject Root cause diagnosis en_ZA
dc.title Comparative analysis of Granger causality and transfer entropy to present a decision flow for the application of oscillation diagnosis en_ZA
dc.type Postprint Article en_ZA


Files in this item

This item appears in the following Collection(s)

Show simple item record