InterPore2022

Relative permeability computations using large Digital Rock Physics simulations

2 Jun 2022, 13:45

15m

Oral Presentation (MS09) Pore-scale modelling MS09

Speaker

Mohamed Regaieg

Description

Mohamed Regaieg1*, Igor Bondino1, Clément Varloteaux2, Titly Farhana Faisal3, and Richard Rivenq1

1 TotalEnergies SE
2 Computational Hydrocarbon Laboratory for Optimized Energy Efficiency
3 Inria

Abstract

Direct Numerical Simulation (DNS) on systems larger than few millimeters is too computationally demanding. Pore Network Modeling (PNM) is a practical way to study the flow at pore scale on larger volumes while keeping reasonable running times. Recent numerical developments in digital rock physics, known as Generalized Network Modelling (GNM) (Raeini et al 2017), allow to reconstruct an upscaled version of the 3D segmented image of a rock in the form of a network of pore elements where the single-phase flow conductances in each pore are derived by solving the Stokes equation in the original geometry. In engineering terms, this hybrid solution allows to capture relevant flow information from the original Micro-CT image whilst keeping the overall cost of multi-phase computation manageable. In this work, OpenFOAM is called for the Stokes flow solution inside a pore network extraction platform called Gnextract (Raeini et al 2017), and TotalEnergies’s pore-scale network simulator DynaPNM (Regaieg et al 2017) is used in quasi-static mode after having been made fully parallel. All codes are run on TotalEnergies’s supercomputer PANGEA.

In the first part, we give an overview of TotalEnergies’s two-phase flow simulation workflow. Then, we describe how simulation runs in parallel mode allow to perform large uncertainty studies (thousands simulations / day) on images as large as (8480x8480x10000) voxels representing a rock volume of 46 cm3. We document the very large statistical dispersion in relative permeability results (due to the microscopic arrangement of oil-water contact angles) that is normally achieved when images as small as 1200x1200x1200 are used in simulation. We show how this finite size effect can be drastically reduced by simulating much larger and representative images, greatly improving the precision of the numerical result. Finally, relative permeability curves are computed compared to validation experiments.

Keywords: PNM, plug scale, quasi-static, speed-up

Presentation materials

Interpore2022 (1).pptx