Digital Pore Alteration for Shale Rock Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for analyzing rock properties in shale samples are hindered by the indistinguishability of dead oil and organic matter solids in SEM images, leading to inaccurate porosity and permeability estimates due to the low permeability and hazardous solvent extraction processes.
Innovation Solution
The implementation of digital pore growing methods and systems that utilize SEM images to differentiate between dead oil and organic matter solids by converting matrix voxels to pore voxels based on dead oil estimates, allowing for more accurate rock property analysis without the need for solvent extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional solvent extraction is used to remove dead oil, then porosity and permeability measurements can be obtained, but the process is time-consuming (days to months) and uses hazardous solvents
Solution Approach 1:
The patent replaces the mechanical/chemical solvent extraction system with a digital image processing system. Instead of physically removing dead oil using solvents over days or months, the system uses SEM imaging combined with machine learning algorithms to digitally identify and differentiate dead oil from rock matrix and pore spaces, achieving rapid measurement without actual solvent treatment
Solution Approach 2:
The patent creates a digital copy or virtual model of the rock pore structure through SEM imaging and computational processing. Rather than physically treating the actual core sample, the system analyzes digital representations (images and 3D reconstructions) to estimate porosity and permeability, eliminating the need for time-consuming physical solvent extraction
2Loss of time
If SEM imaging is used for DRP analysis of shale samples, then the core cleaning process is avoided, but dead oil becomes indistinguishable from organic matter solids leading to inaccurate porosity estimates
Solution Approach 1:
The patent applies image processing techniques that enhance contrast and differentiate materials based on their visual characteristics in SEM images. The machine learning model is trained to recognize subtle differences in grayscale intensity, texture, and spatial distribution patterns between dead oil and organic matter solids, effectively creating visual differentiation where none naturally exists in standard SEM imaging
Solution Approach 2:
The patent transforms the analysis from relying on single imaging parameters to using multiple parameters simultaneously. The machine learning model analyzes combinations of grayscale intensity, local texture features, spatial context, and morphological characteristics to distinguish dead oil from organic matter, converting an unsolvable single-parameter problem into a solvable multi-parameter classification problem
3Measurement precision
If dead oil is counted as pore space in DRP analysis, then porosity is overestimated, but if dead oil is counted as organic matter, then porosity accuracy decreases due to indistinguishability in SEM images
Solution Approach 1:
The patent introduces machine learning algorithms as an intermediary layer between the raw SEM images and the porosity calculation. This intermediary automatically performs the complex differentiation task of distinguishing dead oil from organic matter and pore spaces, transforming an extremely complex manual analysis problem into a standardized computational process that can be applied consistently across all samples
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables precise estimation of rock properties such as porosity, permeability, and fluid flow characteristics by accounting for dead oil in digital models, improving the accuracy of rock property analysis and reducing the reliance on hazardous solvents.
Implementation Method 1
a focused ion beam scanning electron microscope (FIB-SEM) to remove material from a rock sample
Implementation Method 2
a focused ion beam scanning electron microscope (FIB-SEM) to remove material from a rock sample and generate images or videos of the pore structure
Data Source
AI summary
A method includes receiving images of a rock sample. The method also includes modifying a set of voxels related to one or more of the received images by applying a digital pore growing operation that changes non-pore voxels surrounding a pore space to pore voxels, wherein the digital pore growing operation is based at least in part on a predetermined dead oil estimate. The method also includes estimating a property of the rock sample based at least in part on the modified set of voxels.