Eventos Anais de eventos
ENCIT 2022
19th Brazilian Congress of Thermal Sciences and Engineering
Study of the impact of microCT image resizing on numerically estimated porosity and permeability results
Submission Author:
Ingrid Carneiro , SC
Co-Authors:
Ingrid Carneiro, Celso Peres Fernandes, Fabio Bagni, Francisco Hilario Bezerra
Presenter: Ingrid Carneiro
doi://10.26678/ABCM.ENCIT2022.CIT22-0431
Abstract
Petrophysical data such as porosity and permeability are fundamental in a reservoir area because they're inputs for reservoir characterization. Currently, with the advances in the micro computed tomography (microCT) area, it was possible not only a greater visualization of the internal structure of the rock but also to numerically estimate rock properties. This combination of microtomographic imaging and simulation is known as Digital Rock Physics (DRP). DRP can be used to complement laboratory investigations, such as, for example, obtaining properties in damaged rocks that can no longer be used in experimental tests. Estimating petrophysical properties from high quality core-plug images can become computationally expensive, taking days. To address this problem, resample algorithms such as the Lanczos algorithm can be applied to the grayscale image or binary image, decreasing the quality and consequently reducing the computational cost of the simulations. In this work, we propose a study to assess the impact of resizing using the Lanczos algorithm, increasing from 1.1 to 2 times the image resolution size, on the estimated porosity and permeability values. Stokes equations are used to model the single-phase flow inside the porous space. These equations are discretized using the Finite Volume Method implemented in the OpenFOAM software. Simulation results show that the resample performed on the grayscale images showed porosity and permeability results closer to the original values than the results obtained after resampling in the binary image. Furthermore, after analyzing the results, we verify that the porosity changes less after resample than the permeability, since the porosity doesn’t depend on the connections of the pore space. To ensure the reliability of the results, they were compared with those obtained in the commercial software Pore. The porosity values obtained in the two softwares were similar, whereas the permeability values were a little discrepant, since the Pore software uses a different model for flow, called the pore network.
Keywords
3D image processing, pore-scale simulation, brazilian outcrop
