Parameter estimation of aquifer models using neural networks and comparison to spatial interpolation

Mantoglou, A

dc.contributor.author	Mantoglou, A	en
dc.date.accessioned	2014-03-01T01:51:55Z
dc.date.available	2014-03-01T01:51:55Z
dc.date.issued	2002	en
dc.identifier.issn	00017132	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/26510
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-0036973369&partnerID=40&md5=68b79cc0bb475ab0a52e3f601f660fd0	en
dc.subject.other	flow modeling	en
dc.subject.other	groundwater	en
dc.subject.other	transmissivity	en
dc.subject.other	aquifer characterization	en
dc.subject.other	conference proceeding	en
dc.subject.other	flow modeling	en
dc.subject.other	groundwater flow	en
dc.subject.other	parameterization	en
dc.subject.other	transmissivity	en
dc.title	Parameter estimation of aquifer models using neural networks and comparison to spatial interpolation	en
heal.type	journalArticle	en
heal.publicationDate	2002	en
heal.abstract	A new parameterization of transmissivity based on Neural Networks is developed and by defining an appropriate network topology it is possible to express mathematically the network response as a sum of one-dimensional functions similarly to the Turning Bands Method. By selecting appropriate neural activation functions it is possible to get flexible and concise parameterizations that can handle both gradual as well as abrupt large-scale changes of the real transmissivity map. A simple regularization is proposed that can dampen the erratic high frequency components in the estimated parameters. Various examples indicate that the proposed parameterization can handle well various types of transmissivity variations and is particularly suited when the true transmissivity map exhibits specific sorts of heterogeneity with large anisotropies or abrupt changes along lines.	en
heal.journalName	Acta Universitatis Carolinae, Geologica	en
dc.identifier.volume	46	en
dc.identifier.issue	2-3	en
dc.identifier.spage	190	en
dc.identifier.epage	194	en