Digital Rock Analysis for Reliable Multiphase Permeability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the representative elementary volume (REV) in digital rock analysis are subjective, prone to errors, overestimate search regions, and fail to account for sample heterogeneity, particularly in multiphase fluid flow simulations.
Innovation Solution
The method involves deriving pore structure parameters from a pore-matrix model, partitioning the model into phase-matrix models based on phase distribution, and analyzing the dependence of these parameters on saturation to determine the REV, using high-resolution imaging and computational models to characterize porosity and permeability across different length scales.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional REV determination methods are used, then the process is simpler, but the measurement precision and reliability are poor due to subjectivity and errors
Solution Approach 1:
The patent replaces subjective manual REV determination with an automated computational system that uses digital rock models and fluid flow simulations. The system automatically identifies the REV by analyzing when permeability measurements converge across different domain sizes, eliminating human subjectivity and error while providing objective, repeatable results through computational algorithms
Solution Approach 2:
The patent creates digital copies of rock samples through high-resolution imaging (FIB-SEM, X-ray CT) to generate three-dimensional pore structure models. These digital replicas allow for repeated virtual experiments and measurements without destroying the original sample, enabling precise REV determination through multiple simulations and statistical analyses
2Measurement precision
If larger search regions are used for REV determination, then the representativeness improves, but the computational cost and time increase
Solution Approach 1:
The patent employs a dynamic, multi-scale approach where the analysis domain size is systematically varied to identify the convergence point. Instead of using a fixed large domain, the method progressively increases domain size and monitors when permeability values stabilize, allowing the optimal REV size to be determined adaptively based on the specific sample characteristics rather than using unnecessarily large domains for all cases
Solution Approach 2:
The patent performs preliminary analysis using smaller domains to estimate the REV size before conducting full-scale simulations. By first identifying the approximate convergence scale through initial tests, the method avoids the computational expense of immediately running extensive simulations on large domains, thus reducing overall computational time while maintaining accuracy
3Reliability
If conventional methods assume homogeneous media, then the analysis is simpler, but the reliability fails for heterogeneous samples like carbonates
Solution Approach 1:
The patent applies local quality analysis by examining permeability convergence at different locations and scales within the sample. Rather than assuming uniform properties throughout, the method identifies regions with consistent permeability behavior and uses these local convergence characteristics to define the REV, making the approach suitable for heterogeneous materials like carbonates where properties vary spatially
Solution Approach 2:
The patent changes the analysis parameter from assuming homogeneous media to evaluating permeability as a function of domain size and position. By monitoring how permeability values change with increasing domain size and comparing results across multiple realizations, the method reliably identifies the REV even in heterogeneous samples without requiring simplified homogeneity assumptions
Data Source
Figure 1~2
Figure 3A~5B
Figure 6
AI summary
The pore structure of rocks and other materials can be determined through microscopy and subjected to digital simulation to determine the properties of multiphase fluid flows through the material. To ensure reliable results, the digital rock model is first analyzed via a series of operations that, in some embodiments, include: obtaining a three-dimensional pore/matrix model of a sample; determining a flow axis; verifying that the dimension of the model along the flow axis exceeds that of a representative elementary volume (REV); selecting a flow direction; extending model by mirroring if pore statistics at a given saturation are mismatched for different percolating phases; and increasing resolution if the smallest nonpercolating sphere dimension is below a predetermined threshold. This sequence of operations increases reliability of results when measuring relative permeability using the model and displaying relative permeability measurements to user.