Deep causal mining for plant-wide oscillations with multilevel Granger causality analysis

Tao YUAN, Gang LI, Zhaohui ZHANG, S. Joe QIN

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Referred Conference Paperpeer-review

11 Citations (Scopus)

Abstract

Plant-wide disturbance such as oscillations are common in large-scale complex controlled processes whose effects propagate to many units and may deteriorate overall control performance. It is important to capture the major causal relationship within the plant and diagnose the root cause along with complete propagation paths. This paper presents a novel multilevel Granger causality framework for root cause diagnosis. The high level is regarded as a group-wise analysis, which is clustered by dynamic time warping-based K-means method and investigated using group Granger causality. The low level is individual causal reasoning within each group where a partial least squares modified Granger causal test is developed to overcome multicollinearity issue. The proposed causality analysis framework is validated through a benchmark industrial case study to show its effectiveness and superiority.
Original languageEnglish
Title of host publication2016 American Control Conference (ACC)
PublisherInstitute of Electrical and Electronics Engineers
Pages5056-5061
Number of pages6
ISBN (Electronic)9781467386821
ISBN (Print)9781467386838
DOIs
Publication statusPublished - Jul 2016
Externally publishedYes
Event2016 American Control Conference, ACC 2016 - Boston, United States
Duration: 6 Jul 20168 Jul 2016

Publication series

NameProceedings of the American Control Conference
PublisherInstitute of Electrical and Electronics Engineers
ISSN (Print)0743-1619
ISSN (Electronic)2378-5861

Conference

Conference2016 American Control Conference, ACC 2016
Country/TerritoryUnited States
CityBoston
Period6/07/168/07/16

Bibliographical note

The authors are grateful for the financial support from the China Scholarship Council and the Texas-Wisconsin-California Control Consortium, and for the industrial data provided by the Eastman Chemical Company

Fingerprint

Dive into the research topics of 'Deep causal mining for plant-wide oscillations with multilevel Granger causality analysis'. Together they form a unique fingerprint.

Cite this