A Machine Learning-Based Method for Assessing CO2-Induced Permeability Changes Due to Water-Stone Reaction

Abstract

The application relates to the technical field of carbon sequestration geology engineering, and particularly discloses a CO2 water-rock reaction induced permeability change evaluation method based on machine learning, which comprises the following steps: firstly, obtaining multi-focal surface image data of a sample after reaction on a scanning electron microscope platform; automatically identifying a reaction area by using a trained U-Net instance model to obtain porosity and specific surface area information; then automatically selecting corresponding calculation coefficients based on area-depth information of dissolution pores to perform normalization calculation and directly output permeability and spatial distribution thereof. The application does not need destructive detection, realizes quantitative characterization of physical properties of rock permeability, significantly improves analysis accuracy and efficiency, supports batch processing statistics, and provides reliable data support for carbon sequestration safety research and sequestration simulation.

Description

Technical Field

[0001] This application relates to the field of carbon sequestration geological engineering technology, and in particular to a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction. Background Technology

[0002] Carbon dioxide geological sequestration involves injecting carbon dioxide into deep reservoirs, where it reacts with the surrounding rocks under specific temperature and pressure conditions to achieve long-term stable storage. During this process, reactions such as carbonation dissolution, secondary precipitation, and ion exchange alter the microscopic pore structure and mineral interfaces of the rocks, leading to complex spatiotemporal evolution of porosity and pore throat topology. These changes directly affect the mineralization rate and final sequestration capacity, and also alter the formation's seepage patterns and pressure propagation mechanisms, thus impacting the safety and operational window of the sequestration process. In laboratory and pilot project assessments, conducting relevant reaction experiments using rock samples can characterize subtle differences in pore morphology and geometry before and after the reaction. This is crucial for understanding the reaction process, establishing the relationship between petrology and permeability, and providing important constraints for numerical simulations. Therefore, developing an objective and traceable quantitative assessment method for porosity changes is of significant practical and engineering value for sequestration stability assessment, parameter inversion, and optimization of field monitoring schemes.

[0003] Existing technologies typically use core displacement experiments to characterize microscopic evolution: after performing water-rock reaction on core samples taken in the field using water-rock reaction equipment, displacement pressure and flow rate are obtained through displacement experiments, and the dissolution profile is obtained and the area distribution is statistically analyzed using Darcy's formula. This type of method is relatively mature, but it mainly stays at the one-dimensional displacement level and is difficult to give the volumetric permeability changes and distribution. At the same time, due to the high equipment / time costs and strict requirements for core sample preparation and platform, it is difficult to meet the needs of routine high-throughput testing in laboratories, resulting in unstable test results. Summary of the Invention

[0004] To address the limitations of traditional two-dimensional displacement detection methods, such as limited accuracy, high cost, and low throughput, this application provides a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction.

[0005] This application provides a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction, employing the following technical solution: A machine learning-based method for assessing CO2-induced permeability changes caused by water-rock reaction includes the following steps: Sample preparation and image acquisition: The rock slice samples were placed in a water-CO2 environment for reaction; multi-focal images of the rock slice samples before and after the reaction were acquired on the scanning electron microscope platform, and the image sequence was obtained and the imaging parameters were recorded. Image preprocessing and front-to-back registration: The image sequence is subjected to noise suppression, illumination equalization and distortion correction, and geometric registration of the front and back fields of view is completed based on image markers or features; Dissolution region identification: The contours of dissolution regions are extracted using a trained instance segmentation model, and the area and number of each dissolution region are output. Dissolution depth reconstruction and calibration: The image sequence is reconstructed in three dimensions using a focal length evaluation algorithm, and the focal length change is converted into actual height through a calibration method to obtain the effective dissolution depth of each dissolution area; Permeability calculation: Based on the area of ​​each dissolution region and the corresponding effective dissolution depth, the porosity, specific surface area and permeability of the dissolution region in the field of view are calculated using the field-of-view volume normalization algorithm; Results Output and Quality Control: Output the penetration rate and regional statistical reports, and perform repeatability and consistency verification.

[0006] Furthermore, in the dissolution region identification step, the instance segmentation model is a deep learning model based on the U-Net architecture, which is trained using a dataset of dissolution region images containing manual annotations.

[0007] Furthermore, in the dissolution depth reconstruction and calibration step, the calibration method is the step standard method, which uses step standard samples with known height differences to perform multi-focal plane scanning and establish the conversion relationship between focal length change and actual height.

[0008] Furthermore, in the dissolution depth reconstruction and calibration step, the calibration method is the platform stepping method, which uses precise control of the platform to move at equal step lengths and perform multi-focal plane scanning, records the focal length changes and correlates them with the stepping distance to reconstruct the relative height map.

[0009] Furthermore, in the permeability calculation step, the formula for calculating the porosity φ is:

[0010] in, A i Let i be the area of ​​the i-th etched region. d i Let be the effective dissolution depth of the i-th dissolution region. d max The maximum dissolution depth, l i_max This represents the side length of the square corresponding to the dissolved region.

[0011] Furthermore, in the permeability calculation step, the formula for calculating the specific surface area S is:

[0012] in, Ai Let i be the area of ​​the i-th etched region. d i denoted as the effective dissolution depth of the i-th dissolution region.

[0013] Furthermore, in the permeability calculation step, the permeability... k The calculation formula is:

[0014] in, φ i Let be the porosity of the i-th dissolution region. S i Let be the specific surface area of ​​the i-th etched region.

[0015] Furthermore, in the results output and quality control steps, the repeatability and consistency verification includes repeatedly measuring the same field of view and calculating the relative standard deviation of the permeability, as well as batch processing and stitching of multiple adjacent fields of view to verify the stability and consistency of the results.

[0016] This application also provides a system for implementing a machine learning-based method for assessing CO2-induced permeability changes caused by water-rock reaction, comprising: The image acquisition and calibration unit is used to control the scanning electron microscope platform to complete multi-focal plane image acquisition and parameter recording. The image preprocessing and registration unit is used to preprocess the image and complete the registration of the image before and after the reaction. The erosion area identification unit is used to load and run the instance segmentation model to identify and extract the erosion area; The dissolution depth reconstruction and calibration unit is used to generate a relative height map and perform depth calibration. The permeability calculation and result output unit is used to calculate the permeability and generate reports. The quality control and batch processing unit is used to manage batch processing workflows and perform quality verification.

[0017] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction.

[0018] In summary, this application includes at least one of the following beneficial technical effects: This application achieves automatic identification and three-dimensional quantification of dissolution zones after CO2 water-rock reaction by combining multifocal imaging with instance segmentation modeling using scanning electron microscopy. This method can non-destructively acquire information on the area, depth, and spatial distribution of dissolution zones, and directly calculate permeability values ​​using a field-of-view volume normalization algorithm. This technical approach overcomes the limitation of traditional core displacement experiments, which can only provide one-dimensional permeability data, achieving accurate conversion from two-dimensional images to three-dimensional physical property parameters, significantly improving the accuracy and spatial characterization capability of permeability assessment.

[0019] This method establishes a standardized workflow from image acquisition and processing to parameter calculation, supporting high-throughput batch analysis, significantly improving detection efficiency and reducing human error. Repeatability verification and multi-view stitching consistency checks ensure the reliability and traceability of the results. This technical system provides crucial data support for the study of water-rock reaction mechanisms in carbon sequestration, reservoir parameter sensitivity analysis, and large-scale numerical simulations, and has significant engineering application value. Attached Figure Description

[0020] Figure 1 This is a flowchart of a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction, according to an embodiment of this application. Figure 2 This is a schematic diagram of the CO2 water-rock reactor in the embodiments of this application; Figure 3 This is an image of a rock slice sample after reaction, as described in the embodiments of this application. Figure 4 This is a schematic diagram of the identification of the dissolved area in the embodiments of this application; Figure 5 This is a zoning statistical report of the dissolution regions 1-10 in the embodiments of this application. Detailed Implementation

[0021] The following is in conjunction with the appendix Figure 1-5 This application will be described in further detail.

[0022] This application discloses a machine learning-based method for assessing permeability changes caused by CO2-water-rock reaction. (Refer to...) Figure 1 The machine learning-based method for assessing CO2-induced permeability changes due to water-rock reaction includes the following steps: Step 1, Sample Preparation and Image Acquisition, specifically includes: Step 1.1, Sample Preparation: like Figure 2As shown, the CO2 water-rock reaction apparatus includes a reactor, a temperature-controlled heating module, and a CO2 injection system. The reactor is a high-pressure reactor. The temperature-controlled heating module includes a heater for heating the reactor. The CO2 injection system includes a carbon dioxide cylinder and a carbon dioxide pump connected to the reactor. A vacuum pump is also connected to the reactor for evacuation. Temperature and pressure sensors are installed inside the reactor to monitor the temperature and pressure, respectively. Rock samples react with CO2 and water within the reactor to simulate the water-rock reaction of CO2 with the surrounding rock under temperature and pressure conditions during CO2 geological sequestration.

[0023] The rock sample used in this embodiment was a basalt sheet with a diameter of 25 mm and a thickness of 5 mm, and the surface was polished. Deionized water was added to the reactor, and pure CO2 gas was injected to a preset pressure of 10 MPa. The reaction temperature was set at 60°C, and the reaction time was 72 hours. After the reaction was completed, the rock sheet was removed (e.g., Figure 3 As shown in the figure, it undergoes pretreatment such as drying and conductive coating, and is ready for SEM imaging and analysis.

[0024] Step 1.2, Image Acquisition: SEM imaging was performed in SE2 mode, with the electron beam acceleration voltage set to 3.0 kV and the working distance to 7.2 mm. Multi-focal plane image sequences were acquired for specific fields of view (physical size approximately 3810 μm × 2650 μm) of rock slice samples before and after the reaction at a magnification of 30x.

[0025] Record imaging parameters, including magnification, working distance, stage coordinates, and scale information, to ensure dimensional consistency in subsequent calculations.

[0026] Step 2: Image preprocessing and registration (before and after registration): Noise suppression, illumination equalization, and distortion correction are performed on the image sequence. The pre- and post-reaction fields of view are registered under a global affine-local non-rigid hybrid model to form a standardized dataset, providing input for subsequent analysis.

[0027] Step 3, Identification of the dissolution area, including: Step 3.1, Model Training: Using an image dataset of manually labeled dissolution areas containing samples of this type of rock fragments, the instance segmentation model based on the U-Net architecture is trained. During the training process, data augmentation strategies including random rotation, translation, scaling, and contrast adjustment are adopted to improve the generalization ability of the model.

[0028] Step 3.2, Dissolution Area Extraction: Input the registered post-reaction SEM image from Step 2 into the trained model to automatically identify and extract 42 dissolution areas (numbered 1-42; see the schematic diagram of their outlines and numbers). Figure 4Output the area of ​​each dissolved region. And its position in the field of view.

[0029] Step 4: Reconstruction and calibration of dissolution depth: Based on a multi-focal plane image sequence, a relative height map of the field of view is reconstructed using a focal length evaluation algorithm. A platform stepping method (50nm step size) is employed for dimensional calibration; that is, by precisely controlling the platform to move at equal step sizes and recording focal length changes, the relationship between focal length and true height is established, thereby obtaining the effective dissolution depth of each dissolution region. .

[0030] Step 5, Permeability Calculation: Based on the area and effective dissolution depth of each dissolution region, the porosity of the dissolution region in the field of view is calculated by normalization according to the field volume. Specific surface area S and permeability k:

[0031]

[0032]

[0033] in, A i Let i be the area of ​​the i-th etched region. d i Let be the effective dissolution depth of the i-th dissolution region. d max The maximum dissolution depth is determined by taking a standard petrographic thin section thickness of 30 μm. l i_max This represents the side length of the square corresponding to the dissolved region; In this embodiment, taking region 1 (i.e., i=1) as an example, the area of ​​the erosion region is... Effective dissolution depth The maximum side length is Maximum dissolution depth Substituting the data into the porosity estimation formula yields the following result:

[0034] That is, the average porosity of this field of view (dissolution region 1) is approximately 1.291%. Further, substituting the data into the specific surface area estimation formula, we obtain:

[0035] Furthermore, substituting the data into the penetration rate estimation formula yields the following result:

[0036] That is, the estimated permeability of this field of view (dissolution zone 1) is approximately 0.002608 mD.

[0037] Step 6, Results Output and Quality Control: Output porosity increment and zonal statistical reports. The zonal statistical reports include the location, area, depth, and permeability k of each solution zone (e.g., ...). Figure 5 (As shown).

[0038] To verify the reliability of the results, three repeated measurements were performed on this field of view, and the relative standard deviation of the porosity increment was less than 10%. At the same time, four adjacent fields of view were batch-processed and stitched together to verify the spatial consistency of the results.

[0039] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A machine learning-based method for assessing CO2-induced permeability changes caused by water-rock reaction, characterized in that: Includes the following steps: Sample preparation and image acquisition: The rock slice samples were placed in a water-CO2 environment for reaction; multi-focal images of the rock slice samples before and after the reaction were acquired on the scanning electron microscope platform, and the image sequence was obtained and the imaging parameters were recorded. Image preprocessing and front-to-back registration: The image sequence is subjected to noise suppression, illumination equalization and distortion correction, and geometric registration of the front and back fields of view is completed based on image markers or features; Dissolution region identification: The contours of dissolution regions are extracted using a trained instance segmentation model, and the area and number of each dissolution region are output. Dissolution depth reconstruction and calibration: The image sequence is reconstructed in three dimensions using a focal length evaluation algorithm, and the focal length change is converted into actual height through a calibration method to obtain the effective dissolution depth of each dissolution area; Permeability calculation: Based on the area of ​​each dissolution region and the corresponding effective dissolution depth, the porosity, specific surface area and permeability of the dissolution region in the field of view are calculated using the field-of-view volume normalization algorithm; The formula for calculating porosity φ is: The formula for calculating specific surface area S is: Penetration k The calculation formula is: in, A i Let i be the area of ​​the i-th dissolution region. d i Let be the effective dissolution depth of the i-th dissolution region. d max The maximum dissolution depth, l i_max This represents the side length of the square corresponding to the dissolved region. φ i Let be the porosity of the i-th dissolution region. S i Let be the specific surface area of ​​the i-th dissolution region; Results Output and Quality Control: Output the penetration rate and regional statistical reports, and perform repeatability and consistency verification.

2. The method for assessing CO2-induced permeability changes based on machine learning according to claim 1, characterized in that: In the process of identifying the eroded region, the instance segmentation model is a deep learning model based on the U-Net architecture, which is trained using a dataset of eroded region images containing manual annotations.

3. The method for assessing CO2-induced permeability changes based on machine learning according to claim 1, characterized in that: In the dissolution depth reconstruction and calibration steps, the calibration method is the step standard method, which uses step standard samples with known height differences to perform multi-focal plane scanning and establish the conversion relationship between focal length changes and actual height.

4. The method for assessing CO2-induced permeability changes based on machine learning according to claim 1, characterized in that: In the dissolution depth reconstruction and calibration steps, the calibration method is the platform stepping method. The platform is precisely controlled to move at equal step lengths and perform multi-focal plane scanning. The changes in focal length are recorded and correlated with the stepping distance to reconstruct the relative height map.

5. The method for assessing CO2-induced permeability changes based on machine learning according to claim 1, characterized in that: In the results output and quality control steps, the repeatability and consistency verification includes repeatedly measuring the same field of view and calculating the relative standard deviation of the permeability, as well as batch processing and stitching of multiple adjacent fields of view to verify the stability and consistency of the results.

6. A system for implementing the machine learning-based method for assessing CO2-water-rock reaction-induced permeability changes as described in any one of claims 1-5, characterized in that: include: The image acquisition and calibration unit is used to control the scanning electron microscope platform to complete multi-focal plane image acquisition and parameter recording. The image preprocessing and registration unit is used to preprocess the image and complete the registration of the image before and after the reaction. The erosion area identification unit is used to load and run the instance segmentation model to identify and extract the erosion area; The dissolution depth reconstruction and calibration unit is used to generate a relative height map and perform depth calibration. The permeability calculation and result output unit is used to calculate the permeability and generate reports. The quality control and batch processing unit is used to manage batch processing workflows and perform quality verification.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements a machine learning-based method for assessing CO2-induced permeability changes caused by water-rock reaction, as described in any one of claims 1-5.

Images