DIGITAL СORE: TEMPERATURE FIELD INFLUENCE ON TWO-PHASE FILTRATION OF FLUIDS IN ROCKS

Abstract

Link for citation: Katanov Yu.E., Yagafarov A.K., Aristov A.I. Digital сore: temperature field influence on two-phase filtration of fluids in rocks. Bulletin of the Tomsk Polytechnic University. Geo Аssets Engineering, 2023, vol. 334, no. 10, рр. 108-118. In Rus. Relevance. Computational fluid dynamics is a powerful tool for studying geological processes. Fluid flow equations in porous media can be solved directly with the resulting three-dimensional X-ray and computer tomographic images. Numerous studies of the geological features of rocks depend on heat transfer in porous materials. These processes include heat conduction in the rock material, convection between matter and the surrounding gas, electrical conductivity, viscous dissipation of fluids, kinetics of chemical reactions, and interfacial heat transfer at the solid/liquid interface. While energy transfer in porous media under conditions of local thermal equilibrium is well studied on the scale of Darcy's law, it is less often considered in porous models. Heat transfer can be accounted for by creating a «heat network» based on a porous network model representing a set of fluid-conducting channels with an appropriate solution to the unified energy balance equation. It is also necessary to take into account convective transfer of thermal energy by the flow of liquid phases (oil, water), each of which is represented by a separate network. Knowing the distribution of the temperature field in the texture of each lithological type of rock, it is possible to determine the probabilistic displacement of the fluid displacement front in the corresponding fluid-conducting space of the rock. Objective: neural network modeling of the effect of the temperature field on the filtration characteristics of hydrocarbons in the fluid-conducting space of rocks. Objects: polymictic sandstones of the Tyumen Formation. Methods. Digital reconstruction of rock texture was performed using artificial intelligence methods and neural networks; neural network algorithms of probabilistic displacement front of two-phase flow (oil, water) were developed in Python programming language; methodical approach to the study of thermal field propagation in rock texture and its influence on two-phase fluid flow was developed using Fourier law, Navier–Stokes equations and similarity criteria of hydrocarbon systems. Results. We developed algorithms for neural network modeling of the temperature field in the fluid-conducting space of a rock (polymictic sandstone) as well as neural network algorithms to estimate the displacement of two-phase filtration front under the corresponding influence of the temperature distribution in the texture of a digital core. The basic mathematical models of these algorithms are outlined. The source code of the algorithms is written in the Python programming language with additional use of non-commercial libraries. The results of neural network modeling have a high degree of reliability, confirmed by experiments with the data obtained in the laboratory of core studies (University of Tyumen, Tyumen), the laboratory of the V.I. Shpilman Research and Analytical Centre for the Rational Use of the Subsoil (Khanty-Mansiysk), laboratory of digital research in oil and gas in the framework of the technological project «Digital core» (Industrial University of Tyumen, Tyumen).