Method for determining shale in-situ heat treatment conditions based on SEM and image processing technology

Abstract

The invention provides a method for determining shale in-situ heat treatment conditions based on an SEM and an image processing technology. The method comprises the following steps: acquiring shale samples subjected to heat treatment under different conditions by adopting a high-temperature and high-pressure reaction kettle; obtaining an SEM image of the shale sample after heat treatment, and performing gray processing on the SEM image to obtain a gray image; carrying out super-resolution enhancement on the grayscale image by adopting an enhanced super-resolution generative adversarial network to obtain an enhanced image; extracting a pore structure of the shale sample after heat treatment from the enhanced image; and calculating the porosity of the shale surface of the pore structure subjected to heat treatment under different conditions, and optimizing the heat treatment conditions of the shale through the change of the porosity of the shale surface. According to the invention, the deep learning algorithm is adopted to process the rock SEM image, so that the image with clearer details and contours and higher resolution can be obtained; shale structure parameter changes after heat treatment are quantified through the SEM and image processing technology, reasonable heat treatment conditions are determined through the structure parameter changes, and the operation method is convenient.

Description

Technical Field

[0001] The present invention relates to the technical field of shale heat treatment, and in particular to a method for determining shale in-situ heat treatment conditions based on SEM and image processing technology. Background Art

[0002] Shale oil and gas resources are abundant and have become a realistic alternative energy source, but many regions started late in developing such reservoirs and are still unable to develop such reservoirs efficiently. Shale reservoirs are dense, with low porosity and permeability, and are difficult to develop through conventional development methods. In recent years, domestic and foreign countries have used in-situ heat treatment technology (Shell's electric heating technology, Taiyuan University of Technology's steam heating, LLNL's radio frequency heating, Phoenix Wyoming's microwave heating technology, etc.) to develop such reservoirs. Under the action of thermal energy, the micropores and microcracks of shale are connected and expanded, which increases the permeability of the reservoir and fluid fluidity, thereby increasing the recovery rate of shale oil and gas. The invention patent with publication number CN118194614A proposes a method for evaluating the improvement of shale reservoir permeability based on ultra-high temperature heating. This method applies the principle of clay and clay structural transformation at high temperature to shale reservoirs, and improves the reservoir permeability by transforming the properties and structure of clay minerals at high temperature (1200°C), thereby enhancing the shale recovery rate. However, the mechanism of shale pore expansion and permeability under the action of thermal energy is complex, and the expansion of shale pores under thermal action is not clear. Cui Jingwei et al. studied the expansion of shale samples under overburden pressure and found that with the increase of temperature, the shale pore permeability increased significantly, but when the temperature increased to a certain threshold, the shale pore permeability began to decrease. It can be seen that the heterogeneity of shale leads to the unknown heating threshold conditions (temperature, heating time, etc.); in addition, the heterogeneity of shale also leads to uneven heating of shale. If shale is heated at high temperature blindly, on the one hand, a large amount of heat energy will be wasted; on the other hand, the degree of change in shale pore permeability during heat treatment under different conditions is different. If the heat treatment threshold is exceeded, excessive thermal cracking of rock minerals may block thermal pores and cracks. Therefore, it is still necessary to establish an effective and reasonable method to quickly determine the shale heat treatment plan.

[0003] In recent years, SEM combined with image processing technology has been applied to the field of shale characterization. The invention patent with publication number CN116740089A proposes a method for segmentation optimization and quantitative parameter extraction of geotechnical material SEM scanning images. This method mainly uses Matlab to process SEM images to extract shale microstructure parameters. However, conventional image processing methods, such as median filtering for noise reduction, have obvious limitations, such as image details and contour blur after image processing. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology. A deep learning algorithm is used to process rock SEM images. Compared with common image enhancement methods, images with clearer details and contours and higher resolution can be obtained. SEM and image processing technology are used to quantify the changes in shale structural parameters after heat treatment, and reasonable heat treatment conditions are determined through the changes in structural parameters. The operation method is convenient, which provides a feasible method for determining the in-situ heat treatment scheme of shale.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for determining in-situ heat treatment conditions of shale based on SEM and image processing technology, comprising:

[0007] High temperature and high pressure reactors were used to obtain shale samples after heat treatment under different conditions;

[0008] Acquiring a SEM image of the heat-treated shale sample, and performing grayscale processing on the SEM image to obtain a grayscale image;

[0009] Performing super-resolution enhancement on the grayscale image using an enhanced super-resolution generative adversarial network to obtain an enhanced image;

[0010] extracting the pore structure of the shale sample after heat treatment in the enhanced image;

[0011] The porosity of shale surfaces with pore structures after heat treatment under different conditions is calculated, and the shale heat treatment conditions are optimized through the structural parameter changes of the porosity of the shale surfaces.

[0012] Preferably, the different conditions of heat treatment include: different heating temperatures and different heating times.

[0013] Preferably, the method for constructing the enhanced super-resolution generative adversarial network includes:

[0014] Collect shale SEM images as sample data sets;

[0015] Constructing a Real-ESRGAN neural network model; the Real-ESRGAN neural network model includes: a generator and a discriminator;

[0016] The generator is initialized, and the Real-ESRGAN neural network model is trained by a combination of L1 loss, perceptual loss and GAN loss to obtain the trained enhanced super-resolution generative adversarial network.

[0017] Preferably, the number of residual blocks of the generator before upsampling is 23; the discriminator adopts a U-Net discriminator with spectral normalization to enhance adversarial learning on image details; the input and output sizes of the U-Net discriminator are the same, and the U-Net discriminator provides pixel information to the generator while outputting the true value of the pixel.

[0018] Preferably, extracting the pore structure of the heat-treated shale sample in the enhanced image comprises:

[0019] Using the TrainableWeak Segmentation image segmentation algorithm to segment the enhanced image to obtain a segmentation result;

[0020] The segmentation result is corrected by repeatedly modifying the selected area to obtain the pore structure of the shale sample after heat treatment.

[0021] Preferably, the calculation formula for the shale surface porosity is:

[0022]

[0023] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0024] The present invention provides a method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology, including: using a high-temperature and high-pressure reactor to obtain shale samples after heat treatment under different conditions; obtaining SEM images of the heat-treated shale samples, and gray-processing the SEM images to obtain gray-scale images; using an enhanced super-resolution generative adversarial network to super-resolution enhance the gray-scale images to obtain enhanced images; extracting the pore structure of the heat-treated shale samples in the enhanced images; calculating the porosity of the shale surfaces after heat treatment under different conditions, and optimizing the shale heat treatment conditions through the changes in the porosity of the shale surfaces. The present invention uses a deep learning algorithm to process rock SEM images, which can obtain images with clearer details and contours and higher resolution than common image enhancement methods; quantifying the changes in shale structural parameters after heat treatment using SEM and image processing technology, and determining reasonable heat treatment conditions through changes in structural parameters. The operation method is convenient, and provides a feasible method for determining the in-situ heat treatment scheme of shale. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0026] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0027] Figure 2 A schematic diagram showing the comparison between the original SEM image and the SEM image enhanced by Real-ESRGAN provided in an embodiment of the present invention;

[0028] Figure 3 A schematic diagram of the original shale processing results provided by an embodiment of the present invention;

[0029] Figure 4 A schematic diagram of the shale results (36h) after treatment at 200°C provided in an embodiment of the present invention;

[0030] Figure 5 A schematic diagram of the shale results (36h) after treatment at 400°C provided in an embodiment of the present invention;

[0031] Figure 6 A schematic diagram of the shale results (36h) after 600°C treatment provided by an embodiment of the present invention;

[0032] Figure 7 A schematic diagram of the shale results (400° C.) after 12 hours of treatment provided by an embodiment of the present invention;

[0033] Figure 8 A schematic diagram of the shale results (400° C.) after 24 hours of treatment provided by an embodiment of the present invention;

[0034] Fig. 9 A schematic diagram of the shale results (400°C) after 48h treatment provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] The purpose of the present invention is to provide a method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology. A deep learning algorithm is used to process rock SEM images. Compared with common image enhancement methods, images with clearer details and contours and higher resolution can be obtained. SEM and image processing technology are used to quantify the changes in shale structural parameters after heat treatment, and reasonable heat treatment conditions are determined through the changes in structural parameters. The operation method is convenient, which provides a feasible method for determining the in-situ heat treatment scheme of shale.

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology, comprising:

[0039] Step 100: using a high temperature and high pressure reactor to obtain shale samples after heat treatment under different conditions;

[0040] Step 200: obtaining a SEM image of the heat-treated shale sample, and gray-processing the SEM image to obtain a gray-scale image;

[0041] Step 300: using an enhanced super-resolution generative adversarial network to perform super-resolution enhancement on the grayscale image to obtain an enhanced image;

[0042] Step 400: extracting the pore structure of the shale sample after heat treatment in the enhanced image;

[0043] Step 500: Calculate the porosity of shale surfaces with pore structures after heat treatment under different conditions, and optimize the shale heat treatment conditions through the change of the porosity of the shale surfaces.

[0044] Specifically, this embodiment has two screening modes:

[0045] 1. Screening heating temperature:

[0046] (1) A shale heat treatment experiment was carried out using a high-temperature and high-pressure reactor, and shale samples were heat treated at different temperatures for 36 hours;

[0047] (2) using a scanning electron microscope to obtain a SEM image of the shale after heat treatment, and then gray-processing the shale SEM image to obtain a gray-scale image;

[0048] (3) The improved enhanced super-resolution generative adversarial network (Real-ESRGAN) is used to achieve super-resolution reconstruction of the image, and a high-resolution image with clear boundaries is obtained. Comparing the original SEM image and the processed image ( Figure 2 ), it was found that the edge contours of the pore structure of the shale SEM image after treatment were clearer, indicating that the image processing effect was better;

[0049] The specific steps include:

[0050] 1) Prepare a data set and collect shale SEM images as the data set. The degradation process adopts a high-order degradation model after extending the first-order degradation model, as shown in formula (1):

[0051] y=D(x)=[(x*k)↓ r +n] JPEG (1)

[0052] In the formula, D represents the degradation process; * represents convolution; ↓ represents downsampling; n represents noise; JPEG represents lossy digital compression technology.

[0053] 2) Construct a Real-ESRGAN neural network model, which mainly includes two parts: the generator and the discriminator. The structure of the generator is consistent with that of SRGAN, but the residual block (Residual inResidualDenseBlock) before upsampling (Unsample) is changed from 16 to 23, which greatly improves the feature extraction capability; in addition, each RRDB consists of 3 residual blocks; upsampling uses network interpolation to improve the stability of training. The discriminator uses a U-Net discriminator with spectral normalization to enhance adversarial learning on image details. The input and output sizes of U-Net are the same, and while outputting the true value of the pixel, it provides pixel information to the generator. Therefore, compared with VCG, it is better suitable for training in complex situations;

[0054] 3) Training phase. The training is divided into two parts. In the first phase, a generative model with an L1 loss function is used. In the second phase, the generator is initialized and Real-ESRGAN is trained by a combination of L1 loss, perceptual loss, and GAN loss.

[0055] (4) The Trainable Weak Segmentation (TWS) image segmentation algorithm is used to extract the shale skeleton model. By repeatedly observing the segmentation results and modifying the selected pore structure area, the segmentation results are continuously corrected to achieve a better segmentation effect. After multiple modifications, the area with a gray value below 40 is divided into pore structure and fracture structure, and the area with a gray value above 40 is divided into matrix and rock minerals. Figure 3 As shown; demonstratively, this method adopts a point-based global threshold image segmentation algorithm and uses a machine learning algorithm to quickly perform two-dimensional image segmentation, mainly including the following steps: 1) input a picture and perform segmentation learning by manually annotating specific features; 2) observe the segmentation results, and correct the segmentation results by repeatedly modifying the selected area to obtain a better segmentation effect.

[0056] (4) The pore area in the SEM images of shale after different temperature treatments was counted by Fiji, and the porosity was calculated (Table 1), as shown in Figures 4 to 6As shown, it can be seen that with the increase of temperature, the porosity of shale gradually increases; but when the heating temperature is increased to 400℃, the increase in porosity decreases; and from the SEM image of the shale after heating at 600℃, it can be seen that the rock minerals have undergone interlayer detachment under the action of heat, and the detached rock minerals may block the thermal pores / cracks, so it is considered that the heating temperature is more appropriate at 400-600℃.

[0057] Table 1

[0058] sample Porosity Original shale 0.19 200℃ heating 1.59 400℃ heating 12.4 600℃ heating 14.9

[0059] 2. Screening heating time:

[0060] (1) A shale heat treatment experiment was carried out using a high temperature and high pressure reactor at 400°C for different time periods;

[0061] (2) using a scanning electron microscope to obtain a SEM image of the shale after heat treatment, and then gray-processing the shale SEM image to obtain a gray-scale image;

[0062] (3) Real-ESRGAN is used to enhance the SEM images of shale after different heating times;

[0063] (3) The Trainable Weak Segmentation (TWS) image segmentation algorithm is used to extract the shale skeleton model. By repeatedly observing the segmentation results and modifying the selected pore structure area, the segmentation results are continuously corrected to achieve better segmentation effects ( Figure 7-9 );

[0064] (4) The pore area in the SEM images of shale after different temperature treatments was counted by Fiji, and the porosity was calculated (Table 2), as shown in Figure 7 and Fig. 9 As shown in the figure, it can be seen that with the increase of heating time, the porosity of shale gradually increases. The increase of heating time makes the micropore structure of shale fully heated, so the pore structure expands. When the heating time increases to 48h, the increase of porosity decreases, so it is considered that the heating time is more appropriate at 36-48h.

[0065] Table 2

[0066] sample Porosity Original shale 0.19 12h heating 6.9 24h heating 9.76 36h heating 12.4 48h heating 13.9

[0067] The beneficial effects of the present invention are as follows:

[0068] (1) Deep learning algorithms are used to process rock SEM images, which can obtain images with clearer details and contours and higher resolution than common image enhancement methods;

[0069] (2) SEM and image processing technology are used to quantify the changes in shale structural parameters after heat treatment, and reasonable heat treatment conditions are determined through the changes in structural parameters. The operation method is convenient and provides a feasible method for determining the in-situ heat treatment scheme of shale.

[0070] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0071] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for determining in-situ heat treatment conditions of shale based on SEM and image processing technology, characterized in that: include: High temperature and high pressure reactors were used to obtain shale samples after heat treatment under different conditions; Acquiring a SEM image of the heat-treated shale sample, and performing grayscale processing on the SEM image to obtain a grayscale image; Performing super-resolution enhancement on the grayscale image using an enhanced super-resolution generative adversarial network to obtain an enhanced image; extracting the pore structure of the shale sample after heat treatment in the enhanced image; The porosity of shale surfaces with pore structures after heat treatment under different conditions is calculated, and the shale heat treatment conditions are optimized through the change of the porosity of the shale surfaces.

2. The method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology according to claim 1, characterized in that: Different conditions of heat treatment include: different heating temperatures and different heating times.

3. The method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology according to claim 1, characterized in that: The method for constructing the enhanced super-resolution generative adversarial network includes: Collect shale SEM images as sample data sets; Constructing a Real-ESRGAN neural network model; the Real-ESRGAN neural network model includes: a generator and a discriminator; The generator is initialized, and the Real-ESRGAN neural network model is trained by a combination of L1 loss, perceptual loss and GAN loss to obtain the trained enhanced super-resolution generative adversarial network.

4. The method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology according to claim 3, characterized in that: The number of residual blocks of the generator before upsampling is 23; the discriminator adopts a U-Net discriminator with spectral normalization to enhance adversarial learning on image details; the input and output sizes of the U-Net discriminator are the same, and the U-Net discriminator provides pixel information to the generator while outputting the true value of the pixel.

5. The method for determining the in-situ heat treatment conditions of shale based on SEM and image processing technology according to claim 1, characterized in that: Extracting the pore structure of the heat-treated shale sample in the enhanced image includes: Using the TrainableWeak Segmentation image segmentation algorithm to segment the enhanced image to obtain a segmentation result; The segmentation result is corrected by repeatedly modifying the selected area to obtain the pore structure of the shale sample after heat treatment.

6. The method for determining shale in-situ heat treatment conditions based on SEM and image processing technology according to claim 1, characterized in that: The calculation formula of the shale surface porosity is:

Images