PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A geologically-sound representation for evolutionary multi-objective subsurface identification
Vijay Pratap Singh, Marc Schoenauer and Michel Léger
In: IEEE Congress on Evolutionary Computation - CEC'05, 02-sept 2005, Edinburgh - UK.

Abstract

A new representation for evolutionary subsurface identification from surface or well geological data is proposed. It does not require any guess of the underlying fault structure by the expert, but generate geologically balanced layouts. The idea is to represent the subsurface as the combination of, first, a geologically initial set of horizontal layers and, second, fault parameters like shape and displacements. Based on volume and bed-length preservation, a morphogenesis process then gives the structure at present time. A first implementation of this representation is tested on an artificial geological inverse problem in foothills region: the fault locations and dips are considered as two different objectives, and the Epsilon-MOEA multi-objective evolutionary algorithm is applied. The first results show the efficiency of the chosen representation.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
ID Code:1725
Deposited By:Marc Schoenauer
Deposited On:28 November 2005