--> ABSTRACT: Fractured Basement Reservoir Characterization for Fracture Distribution, Porosity and Permeability Prediction, by Lefranc, Marie; Farag, Sherif; Dubois, Agnes; Souche, Laurent; #90155 (2012)

Datapages, Inc.Print this page

Fractured Basement Reservoir Characterization for Fracture Distribution, Porosity and Permeability Prediction

Lefranc, Marie; Farag, Sherif; Dubois, Agnes; Souche, Laurent
Schlumberger, Ho Chi Minh City, Viet Nam.

Fractured basement reservoirs are challenging to model due to the complexity of the fracture network. This paper demonstrates an integrated approach to model these unconventional reservoirs through application of Continuous Fracture Modeling (CFM) and Discrete Fracture Network modeling (DFN). The main objectives of this workflow are to identify the potential flow contributing fractures and to reduce economic risks by optimizing the identification of new well targets.

An innovative workflow has first been developed to minimize the uncertainties on the potential flow contributing fracture sets identification, which are estimated from: 1) the analyses of borehole images, sonic measurements, conventional log data, production and mud log data, 2) the laterolog resistivity and 3) the semi-automated fracture trace extraction analysis generating P33 log (volume of fractures per volume of rock). Once the correlation between these three fracture indicators has been validated, the fracture intensity can be populated in the inter-well space.

The next step is the selection of 3D properties that can be used to propagate the fracture flow indicator away from the wells. Fracture drivers used here are optimized post-stack seismic attributes (including Ant-Tracking), seismic inversion data and optimized 3D forward geomechanical properties. 3D fracture intensity models (CFM) are then constructed using robust method based on Artificial Neural Network methods (ANN).

Finally, fracture properties are computed through DFN modeling. An advanced workflow has been created to estimate accurately: fracture distribution, geometry and orientation with well data calibration. The DFN scale-up is then performed through a novel combination of Oda and Flow based methods and allows obtaining 3D porosity and permeability distribution. It is ensured that the scaled-up DFN model can reproduce dynamic reservoir history such as that from well tests and production logs. The final model can be used to predict the flow rates and recovery of prospective wells.

 

AAPG Search and Discovery Article #90155©2012 AAPG International Conference & Exhibition, Singapore, 16-19 September 2012