Carbon dioxide oil displacement analogue simulation method based on digital core model

By constructing a digital core model and combining it with machine learning and artificial intelligence algorithms, the problems of long simulation cycles and parameter deviations in real core models have been solved, achieving efficient and accurate simulation of the carbon dioxide flooding process, which is applicable to the evaluation of flooding efficiency for various reservoir types.

CN122452284APending Publication Date: 2026-07-24CNPC GREATWALL DRILLING COMPANY +1
0 Cites 0 Cited by

Patent Information

Application Number
CN202510119280.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-01-24
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing technologies using real core models have long experimental cycles and deviations in observation and measurement parameters, making it difficult to efficiently simulate carbon dioxide flooding processes.

Method used

A digital core model is constructed based on the pore structure of real core samples. By combining machine deep learning and artificial intelligence algorithms, a multi-parameter single-variable simulation is performed to simulate the carbon dioxide displacement process in underground oil reservoirs and the movement of fluids in the pore space.

Benefits of technology

It improves simulation efficiency, enables rapid and repeatable evaluation of oil displacement effects, and is applicable to low-permeability and ultra-low-permeability reservoirs, medium-to-high permeability sandstone heavy oil reservoirs and fractured oil and gas reservoirs, providing a more accurate evaluation of oil displacement efficiency.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a carbon dioxide oil displacement analogue simulation method based on a digital core model, relates to the field of oil-gas exploration and development, and solves the problems of long test period, insufficient deviation of observation measurement parameters and the like due to adoption of a real core model in the prior art. Real core pore structure characteristics are utilized, a digital core model is created, machine learning and an artificial intelligence algorithm are combined, different physical processes are comprehensively coupled, and according to parameters such as different crude oil properties (density and dynamic viscosity) and different CO2 injection speeds, the residual oil dynamic distribution characteristics in the CO2 displacement process in an underground oil reservoir are simulated with a single variable. The method is suitable for low-permeability and ultra-low-permeability oil reservoirs displaced by CO2, medium-high-permeability sandstone heavy oil reservoirs (CO2 assisted huff and puff) and fractured oil and gas reservoirs, and has important significance for evaluating the CO2 injection development effect of an oil field. According to the method, tedious processes such as repeated oil washing of a real rock core are avoided, and the working efficiency is improved.
Need to check novelty before this filing date? Find Prior Art