Sciact
  • EN
  • RU

GPU-based algorithm for evaluating the electrical resistivity of digital rocks Full article

Journal Computers and Mathematics with Applications
ISSN: 0898-1221
Output data Year: 2021, Volume: 82, Pages: 200-211 Pages count : 12 DOI: 10.1016/j.camwa.2020.11.005
Tags Digital rock physics; Electrical resistivity; FFT-based preconditioning; Finite-differences; Iterative solver
Authors Khachkova T. 1 , Lisitsa V. 2 , Reshetova G. 2 , Tcheverda V. 1
Affiliations
1 Institute of Petroleum Geology and Geophysics SB RAS, 3 Koptug ave., Novosibirsk, 630090, Russian Federation
2 Sobolev Institute of Mathematics SB RAS, 4 Koptug ave., Novosibirsk, 630090, Russian Federation

Funding (1)

1 Russian Science Foundation 19-77-20004

Abstract: We present an algorithm for numerical simulation of electric current in rock samples and numerical upscaling of heterogeneous samples’ electrical resistivity (conductivity). The solver is oriented on solving strongly heterogeneous problems so that partially saturated multi-mineral core samples can be treated. The algorithm is based on the Krylov-type solver, where the preconditioner is the inverse Laplace operator for homogeneous media. To compute the preconditioner, we use the spectral method in the directions normal to the current and solving series of 1D problems by Thomas algorithm. The solver is implemented using Graphic Processor Units and allows handling samples of the size up to 4003 voxels by a single GPU. © 2020 Elsevier Ltd
Cite: Khachkova T. , Lisitsa V. , Reshetova G. , Tcheverda V.
GPU-based algorithm for evaluating the electrical resistivity of digital rocks
Computers and Mathematics with Applications. 2021. V.82. P.200-211. DOI: 10.1016/j.camwa.2020.11.005 WOS Scopus OpenAlex
Identifiers:
Web of science: WOS:000604261400012
Scopus: 2-s2.0-85097056985
OpenAlex: W3106992641
Citing:
DB Citing
Scopus 20
OpenAlex 22
Web of science 13
Altmetrics: