Variable selection and data pre-processing in NN modelling of complex chemical processes

Papadokonstantakis, S; MacHefer, S; Schnitzlein, K; Lygeros, AI

dc.contributor.author	Papadokonstantakis, S	en
dc.contributor.author	MacHefer, S	en
dc.contributor.author	Schnitzlein, K	en
dc.contributor.author	Lygeros, AI	en
dc.date.accessioned	2014-03-01T01:23:19Z
dc.date.available	2014-03-01T01:23:19Z
dc.date.issued	2005	en
dc.identifier.issn	0098-1354	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/16906
dc.subject	Data preprocessing	en
dc.subject	Neural networks	en
dc.subject	Process modelling	en
dc.subject	Variable entropy	en
dc.subject	Variable information	en
dc.subject	Variable selection	en
dc.subject.classification	Computer Science, Interdisciplinary Applications	en
dc.subject.classification	Engineering, Chemical	en
dc.subject.other	Condensation reactions	en
dc.subject.other	Database systems	en
dc.subject.other	Mathematical models	en
dc.subject.other	Neural networks	en
dc.subject.other	Generalization accuracy	en
dc.subject.other	Neural network models	en
dc.subject.other	Process yield	en
dc.subject.other	Data processing	en
dc.subject.other	chemical processing	en
dc.subject.other	modeling	en
dc.subject.other	neural network	en
dc.title	Variable selection and data pre-processing in NN modelling of complex chemical processes	en
heal.type	journalArticle	en
heal.identifier.primary	10.1016/j.compchemeng.2005.01.004	en
heal.identifier.secondary	http://dx.doi.org/10.1016/j.compchemeng.2005.01.004	en
heal.language	English	en
heal.publicationDate	2005	en
heal.abstract	The neural network models represent nowadays a powerful tool for complicated process identification. However, because of the fact that they belong to the category of data-driven ""black box"" models, they cannot avoid the consequences of the ""garbage in-garbage out"" rule. This work proposes a simultaneous data balancing-variable selection procedure, which is based on traditional statistical techniques and modern information theoretic approaches. It is implemented on a complicated dataset of restricted quality, which refers to a commercial aldol condensation unit (BASF). Based on the pre-processed database a neural model for the prediction of the process yield has been developed. The results verify the importance of the pre-processing stage in terms of generalization accuracy as well as of simpler network structure due to the data-variable selection procedure. Finally, an analysis of the model trends has been implemented to assess qualitative characteristics of the model, which was then used in industrial test runs and resulted in an improvement of the process operation. © 2005 Elsevier Ltd. All rights reserved.	en
heal.publisher	PERGAMON-ELSEVIER SCIENCE LTD	en
heal.journalName	Computers and Chemical Engineering	en
dc.identifier.doi	10.1016/j.compchemeng.2005.01.004	en
dc.identifier.isi	ISI:000229955500012	en
dc.identifier.volume	29	en
dc.identifier.issue	7	en
dc.identifier.spage	1647	en
dc.identifier.epage	1659	en