Skip to Main content Skip to Navigation
Theses

Contrainte des modèles génétiques de réservoirs par une approche de reconnaissance statistique de forme

Abstract : The object of this thesis is modeling subsurface heterogeneity. We adapted the multiple-point (MP) simulation approach that reproduces and conditions complex geometrical patterns provided by unconditional genetic models. Initially, the MP approach was applicable only under the assumption of a certain spatial stationarity of the heterogeneity. To extend this approach to the non-stationary case, two algorithms appeared in the literature: the Tau model and the classification method. Both reveal the geometrical artifacts without necessarily honoring non-stationary constraints. In this work, we proposed a new algorithm of non-stationary MP simulation. It avoids the inconveniences of the existing algorithms and integrates continuous spatial constraint. The experimental results also show that our algorithm has a wider range of applicability than the existing ones.
Document type :
Theses
Complete list of metadatas

Cited literature [78 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00004452
Contributor : Ecole Mines Paristech <>
Submitted on : Friday, December 12, 2008 - 8:00:00 AM
Last modification on : Thursday, September 24, 2020 - 4:34:05 PM
Long-term archiving on: : Friday, September 10, 2010 - 1:02:46 PM

Identifiers

  • HAL Id : pastel-00004452, version 1

Citation

Tatiana Chugunova. Contrainte des modèles génétiques de réservoirs par une approche de reconnaissance statistique de forme. Mathématiques [math]. École Nationale Supérieure des Mines de Paris, 2008. Français. ⟨NNT : 2008ENMP1554⟩. ⟨pastel-00004452⟩

Share

Metrics

Record views

508

Files downloads

2061