A method for upgrading the pore structure of low-permeability, dense sandstone

Abstract

This invention provides a method for scaling up the pore structure of low-permeability to dense sandstone. A small plunger micron-CT scan is performed, and a field emission scanning electron microscope (FE-SEM) sample is prepared by cutting sections from the small plunger. The micron-CT slices are registered and matched with the corresponding micron-CT slices in the FE-SEM image, and the pores in the SEM image and the micron-CT slices are extracted. The pore distribution maps of the FE-SEM image and the corresponding micron-CT slices are compared to determine the location and type of unidentified pores in the micron-CT, and the parameter characteristics of the unidentified pores are statistically analyzed. Under the constraints of pore location, type, pore size, and other parameters, a random growth method is used to supplement the missing small pores into the digital core constructed by the CT method. The pore model before scaling up and the newly generated pores after scaling up are presented in three-dimensional image form and merged to obtain the upgraded pore model. This invention overcomes the problems of inconsistency between the upgraded results and the actual core and the difficulty in establishing three-dimensional pore structure models at different scales.

Description

Technical Field

[0001] This invention belongs to the field of geological reservoir technology, specifically relating to a method for upgrading the pore structure scale of low-permeability-tight sandstone. Technical Background

[0002] CT (Computed Tomography) is an important tool for studying reservoir pore structure, primarily characterizing the three-dimensional structure of samples. However, in image-based pore structure characterization, magnification and sample size are contradictory. For three-dimensional images, the smaller the sample being scanned for CT, the higher the resolution, allowing for the resolution of smaller pores; conversely, the larger the sample, the lower the resolution. Reservoir pore structures span a wide scale, making it difficult to obtain data volumes capable of finely resolving micron- and nanometer-level pores on a macroscopic scale directly using experimental techniques. Therefore, scale-up methods are necessary. Previous studies have primarily employed homogenization theory and porosity scale-up methods for scale-up.

[0003] The first method is mainly based on statistical data such as mineral composition and porosity, and numerical simulation based on homogenization theory to carry out scale upgrades. Although a three-dimensional data volume is established, the parameters such as the pore throat configuration relationship cannot represent the actual core situation. The second method is based on three-dimensional images of the core and focuses on the characterization of pore structure at different scales. However, this method only estimates porosity and does not establish a three-dimensional data volume. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for upgrading the pore structure scale of low-permeability, dense sandstone, comprising the following steps:

[0005] Step 1: Drill a small plunger with a diameter of 2 mm and perform micron-CT scanning on it. After scanning, cut a piece from one end of the sample to make a field emission scanning electron microscope sample.

[0006] Step 2: Register and match the micron-sized CT slices in the field emission scanning electron microscope (FE-SEM) image, and use automatic pore extraction technology to extract the pores in the FE-SEM image and the micron-sized CT slices;

[0007] Step 3: Compare the pore distribution map of the scanning electron microscope image and its corresponding micron CT slice to determine the location and type of unidentified pores in the micron CT, and statistically analyze the parameter characteristics of the unidentified pores to provide parameters for the scale upgrade in Step 4. In this invention, the radius of the unidentified pores is less than 0.5 μm, and the pores are mainly intergranular pores, mainly developed between quartz and feldspar grains.

[0008] Step 4: The pores that were not identified by micron CT in Step 3 are the pores that need to be upgraded. Using their location, type and parameters as boundary conditions, the random growth method is used to supplement the missing small pores into the digital core constructed by the CT method.

[0009] Establishing a three-dimensional pore model requires upgrading pores and throats smaller than 0.5 μm. The specific steps are as follows: Extract a 250×250×250 data volume from the CT data, convert the 16-bit CT grayscale image to an 8-bit grayscale image, first filter the data volume, and set different thresholds corresponding to the EDS energy spectrum to scale up the pores between quartz and feldspar grains; perform an expansion operation on the quartz grain voxels, i.e., expand them outward by one voxel, and assign the expanded quartz voxel a value of 1; perform an expansion operation on the feldspar voxels, and assign the expanded feldspar voxels a value of 1; add the two together, and the voxels with a statistical value of 2 represent the edges of the quartz and feldspar grains, thus expanding the 250×250×250 data volume to a 1250×1250×1250 data volume.

[0010] Step 5: Present the pore model before scale upgrade and the newly generated pores after scale upgrade in the form of three-dimensional images, and merge them to obtain the upgraded pore model.

[0011] Steps 1-3 of this invention preprocess CT data volumes and field emission scanning electron microscope (FET) images to clarify the differences between FET images and CT slices when characterizing the porosity features of the same region, thereby determining the location and boundary parameters for scale upgrading. This overcomes the problem in homogenization theory-based upgrading methods where the upgraded results do not match the actual samples (as it generates pores at random locations, resulting in some differences from real core samples); it also overcomes the problem in porosity upgrading methods where a three-dimensional pore structure model cannot be established (it mainly focuses on the changes in porosity or other parameters at different scales, but it is difficult to determine three-dimensional models at different scales).

[0012] Step 4 defines the pore generation process, which generates pores at specific locations to achieve the goal of dimensional scaling.

[0013] Steps 5-6 show a comparison before and after the scale upgrade. The scale upgrade greatly improves porosity and connectivity, thus improving the accuracy of porosity identification. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of step 1 in the embodiment;

[0015] Figure 2a FE-SEM image of an embodiment

[0016] Figure 2b Micron-sized CT slices for examples

[0017] Figure 2c Pore ​​maps based on FE-SEM for an example

[0018] Figure 2d Pore ​​map based on micron-sized CT slices for an example

[0019] Figure 3 Comparison of pore distribution between emission scanning electron microscope and CT slices in an embodiment.

[0020] Figure 4a This is one of the CT scale upgrade flowcharts in the embodiments;

[0021] Figure 4b This is the second flowchart of the CT scale upgrade implementation.

[0022] Figure 4c This is the third flowchart of the CT scale upgrade process in the embodiment;

[0023] Figure 4d The fourth flowchart of the CT scale upgrade process in the embodiment;

[0024] Figure 4e The fifth flowchart of the CT scale upgrade implementation example;

[0025] Figure 5a The porous network before scale-up in this embodiment;

[0026] Figure 5b The pores generated for the scale upgrade of the embodiment;

[0027] Figure 5c The scaled-up porous network is shown in the example. Detailed Implementation

[0028] The specific technical solutions of the present invention will be described with reference to the embodiments.

[0029] A method for scaling up the pore structure of low-permeability, dense sandstone includes the following steps:

[0030] Step 1: Drill a small plunger with a diameter of 2 mm and perform a micron-scale CT scan on it. After scanning, cut a piece from one end of the sample to prepare a field emission scanning electron microscope (FEM) sample, such as... Figure 1 .

[0031] Step 2: Register and match the corresponding micron-sized CT slices in the field emission scanning electron microscope (FE-SEM) image, such as... Figure 2a , Figure 2b An automated pore extraction technique was used to extract pores from scanning electron microscope images and micron-sized CT slices, such as... Figure 2c , Figure 2d .

[0032] Step 3: Compare the pore distribution maps of the scanning electron microscope (SEM) images and their corresponding micron-sized CT slices to determine the location and type of unidentified pores in the micron-sized CT slices, and statistically analyze the size, pore diameter, and other characteristics of the unidentified pores. The results show that the porosity of the SEM slices is 9.65%, while that of the micron-sized CT slices is 2.61%, a significant difference. The pores extracted by FE-SEM are significantly more numerous than those extracted by the micron-sized CT slices. The main difference lies in the identification of pores at the particle edges. Figure 2c , Figure 2d Statistical analysis was performed on the pore size distribution of field emission scanning electron microscopy and micron-sized CT slices, such as... Figure 3 The results show that micron-CT is not advantageous in identifying pores smaller than 0.5 μm, and micron-CT mainly identifies large pores. Small pores connecting large pores are easily ignored, resulting in a large difference in porosity between micron-CT and field emission scanning electron microscopy results for pores with radii in the range of 12.5-25 μm. Therefore, when using micron-CT to characterize pore structures, it is necessary to supplement this part of the pores.

[0033] Step 4: Since the CT method cannot obtain micropores below its resolution, the random growth method will be used to fill the missing micropores into the digital core constructed by the CT method.

[0034] Establishing a three-dimensional pore model requires upgrading pores and throats smaller than 0.5 μm. The specific steps are as follows: Extract a 250×250×250 (X×Y×Z) data volume from the CT data, such as... Figure 4a , will be Figure 4b 16-bit CT grayscale images converted to such Figure 4c The 8-bit grayscale image shown first undergoes data volume filtering. Different thresholds are set corresponding to the EDS energy dispersive spectroscopy (EDS) image. Taking sample S6 as an example, the cutoff thresholds are set to 57 for porosity, 80 for quartz, 102 for feldspar, and 180 for calcite. Pixels with a grayscale value less than or equal to 57 are defined as pores; pixels in the 57-80 range are defined as quartz; pixels in the 80-102 range are defined as feldspar; pixels in the 102-180 range are defined as calcite; and pixels with a grayscale value greater than 180 are defined as heavy minerals. Figure 4dAnalysis of field emission scanning electron microscopy (SEM) images revealed that the pores between calcite and feldspar, and between calcite and quartz grains, are predominantly larger than 0.5 μm, with pores smaller than 0.5 μm being largely undeveloped. Pores associated with calcite do not require scale-up. This embodiment primarily focuses on scale-up of the quartz-feldspar grain pores. The boundaries between grains and pores are difficult to classify as either solid or pore; therefore, a random modeling method is used to supplement micropores that are difficult to identify with CT. Quartz grain voxels are expanded (by one voxel outward), and the expanded quartz voxels are assigned a value of 1; feldspar voxels are also expanded, and the expanded feldspar voxels are assigned a value of 1; the two are added together, and the voxels with a statistical value of 2 represent the edges of quartz and feldspar grains. Figure 4e Expand the 250×250×250 data volume to a 1250×1250×1250 data volume, that is, upgrade the data volume scale from a resolution of 1.7μm to a data volume of 0.34μm.

[0035] Step 5: Present the pore model before scaling, the newly generated pores after scaling, and the pore model after scaling as 3D images. Before scaling, the pores are mainly larger in diameter, such as... Figure 5a Easily identifiable by CT scans, the pores created by scaling are primarily micropores, such as... Figure 5b The two are combined to obtain the upgraded pore model, such as Figure 5c .

[0036] The pore structure data before and after the upgrade were analyzed and compared. The results showed that after the scale upgrade, the average pore radius decreased, the throat radius decreased, the sphericity decreased, the coordination number increased, the throat length increased, and the tortuosity increased. In addition, the total porosity increased, the connected pores increased, and the connectivity increased, as shown in Table 1.

[0037] Table 1 Comparison of pore structure parameters before and after scale upgrade

[0038]

[0039] The throat radius changed significantly, decreasing by more than half, indicating that the portion smaller than 0.5 μm is mainly the throat portion connecting the pores; after the upgrade, the pore throat connectivity improved, the coordination number increased, the connectivity rate increased, the tortuosity increased, and the throat length increased.

Claims

1. A method for upgrading the pore structure scale of low-permeability-dense sandstone, characterized in that, Includes the following steps: Step 1: Drill a small plunger with a diameter of 2 mm and perform micron-CT scanning on it. After scanning, cut a piece from one end of the sample to make a field emission scanning electron microscope sample. Step 2: Register and match the micron-sized CT slices in the field emission scanning electron microscope (FE-SEM) image, and use automatic pore extraction technology to extract the pores in the FE-SEM image and the micron-sized CT slices; Step 3: Compare the pore distribution maps of the scanning electron microscope images and their corresponding micron-CT slices to determine the location and type of unidentified pores in the micron-CT, and statistically analyze the parameter characteristics of the unidentified pores. By comparing the pores in the same field of view micron-CT slices and field emission scanning electron microscope images, the pore diameter, pore development location, and pore type parameters of the pores not identified by micron-CT were clarified. Based on this, the scale upgrade in step 4 was carried out. The radius of the unidentified pores was less than 0.5 µm, and the pores were mainly intergranular pores, which developed between quartz and feldspar grains. Step 4: The pores that were not identified by micron CT in Step 3 are the pores that need to be upgraded. Using their location and type as boundary conditions, the random growth method is used to supplement the missing small pores into the digital core constructed by the CT method. To upgrade pores smaller than 0.5 µm and establish a three-dimensional pore model, the specific steps are as follows: Extract a 250×250×250 data volume from the CT data, convert the 16-bit CT grayscale image to an 8-bit grayscale image, first filter the data volume, and set different thresholds corresponding to the EDS energy dispersive spectroscopy (EDS) image to scale up the pores between quartz and feldspar grains; perform an expansion operation on the quartz grain voxels, i.e., expand them outward by one voxel, and assign a value of 1 to the expanded quartz voxels; perform an expansion operation on the feldspar voxels, and assign a value of 1 to the expanded feldspar voxels; Add the two together, and the voxels with a statistical value of 2 are the edges of quartz and feldspar grains. Expand the 250×250×250 data volume to a 1250×1250×1250 data volume. Step 5: Present the pore model before scale upgrade and the newly generated pores after scale upgrade in the form of three-dimensional images, and merge them to obtain the upgraded pore model.

Images