Show simple item record

dc.contributor.authorAskarian, Mahdieh
dc.contributor.authorEscudero Bakx, Gerard
dc.contributor.authorGraells Sobré, Moisès
dc.contributor.authorZarghami, Reza
dc.contributor.authorJalali Farahani, Farhang
dc.contributor.authorMostoufi, Navid
dc.contributor.otherUniversitat Politècnica de Catalunya. Departament d'Enginyeria Química
dc.contributor.otherUniversitat Politècnica de Catalunya. Departament de Ciències de la Computació
dc.date.accessioned2016-01-26T11:08:16Z
dc.date.available2017-01-03T01:30:36Z
dc.date.issued2016-01-04
dc.identifier.citationAskarian, M., Escudero, G., Graells, M., Zarghami, R., Jalali Farahani, F., Mostoufi, N. Fault diagnosis of chemical processes with incomplete observations: A comparative study. "Computers & chemical engineering", 04 Gener 2016, vol. 84, p. 104-116.
dc.identifier.issn0098-1354
dc.identifier.urihttp://hdl.handle.net/2117/82031
dc.description.abstractAn important problem to be addressed by diagnostic systems in industrial applications is the estimation of faults with incomplete observations. This work discusses different approaches for handling missing data, and performance of data-driven fault diagnosis schemes. An exploiting classifier and combined methods were assessed in Tennessee-Eastman process, for which diverse incomplete observations were produced. The use of several indicators revealed the trade-off between performances of the different schemes. Support vector machines (SVM) and C4.5, combined with k-nearest neighbourhood (kNN), produce the highest robustness and accuracy, respectively. Bayesian networks (BN) and centroid appear as inappropriate options in terms of accuracy, while Gaussian naive Bayes (GNB) is sensitive to imputation values. In addition, feature selection was explored for further performance enhancement, and the proposed contribution index showed promising results. Finally, an industrial case was studied to assess informative level of incomplete data in terms of the redundancy ratio and generalize the discussion. (C) 2015 Elsevier Ltd. All rights reserved.
dc.format.extent13 p.
dc.language.isoeng
dc.publisherPergamon Press
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/es/
dc.subjectÀrees temàtiques de la UPC::Enginyeria química
dc.subject.lcshChemical processes
dc.subject.otherFault diagnosis
dc.subject.otherMissing data
dc.subject.otherIncomplete observations
dc.subject.otherClassification
dc.subject.otherImputation
dc.subject.otherMachine learning
dc.subject.otherissing data
dc.subject.othertolerant control
dc.subject.othersoft sensors
dc.subject.otherclassification
dc.subject.otherinference
dc.subject.otherinformation
dc.subject.otheralgorithms
dc.subject.othervalues
dc.subject.otherpls
dc.subject.otherpca
dc.titleFault diagnosis of chemical processes with incomplete observations: A comparative study
dc.typeArticle
dc.subject.lemacProcessos químics -- Gestió
dc.contributor.groupUniversitat Politècnica de Catalunya. GPLN - Grup de Processament del Llenguatge Natural
dc.contributor.groupUniversitat Politècnica de Catalunya. CEPIMA - Center for Process and Environment Engineering
dc.identifier.doi10.1016/j.compchemeng.2015.08.018
dc.description.peerreviewedPeer Reviewed
dc.relation.publisherversionhttp://www.sciencedirect.com/science/article/pii/S0098135415002793
dc.rights.accessOpen Access
drac.iddocument16977971
dc.description.versionPostprint (author's final draft)
dc.relation.projectidinfo:eu-repo/grantAgreement/MICINN/6PN/DPI2009-09386
dc.relation.projectidinfo:eu-repo/grantAgreement/MINECO/1PE/DPI2012-37154-C02-01
dc.relation.projectidinfo:eu-repo/grantAgreement/AGAUR/2014SGR1092
upcommons.citation.authorAskarian, M., Escudero, G., Graells, M., Zarghami, R., Jalali Farahani, F., Mostoufi, N.
upcommons.citation.publishedtrue
upcommons.citation.publicationNameComputers & chemical engineering
upcommons.citation.volume84
upcommons.citation.startingPage104
upcommons.citation.endingPage116


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

Except where otherwise noted, content on this work is licensed under a Creative Commons license: Attribution-NonCommercial-NoDerivs 3.0 Spain