Computational approaches to understand NMR and Fluid Dynamics data in Porous Media...
Computational Physics approach to understand NMR data...Fluid dynamics in porous media is an important issue in different areas, especially in oil exploration. Several NMR studies are being developed to elucidate fluid dynamics in porous media and estimate properties such as permeability, tortuosity and characteristic lengths. In order to better understand the correlation between NMR data and the fluid dynamics restricted in porous media, we are developing advanced computational algorithms. Particularly, the correlation between T2 distributions and microstructure and physical properties of the porous media is one of the most important issues in the Oil Industry. Using a random walk method in position and phase space, we developed a software to obtain the T2-distribution based in the intrinsic diffusion of the fluid molecules in digital porous media constructed from 3D microCT images. To better understand this petrophysical problem, we have established a fruitful partnership with collaborators from CENPES/Petrobras.
|
Porous Media - Pore Network construction...Network Extraction algorithms from X-ray microcomputed tomography have become a routine method to obtain pore connectivity and pore morphology information from porous media.
The main approaches for this extraction are either based on Max Ball Algorithm or Medial Axis Extraction. The first is a robust method to separate the pore space into pore and throats, which provides the pore and throat sizes distributions directly. The second simplifies the medium by thinning the volume into a one voxel wide centerline that preserves the volume's topological information. Since each method has drastically different implementations, it is not usual to use both to characterize the porous structure. To solve this problem, we developed a method to extract a simplified centerline of porous materials by adding few more steps to the well-established Max Ball algorithm. Results for sandstones and carbonate rock samples show that this Medial-Axis network can be used for single phase flow simulations, as well as preserves the mediums morphology. To consolidate the results obtained by the proposed method, we are developing collaborations with partners from CENPES/Petrobras and IBM. |