Clastic rock clastic particle size quantitative characterization method based on CT scanning structure image

Abstract

The invention belongs to the technical field of shale oil and gas exploration and development, and discloses a method for quantitatively characterizing the particle size of clastic particles of clastic rock based on a CT scanning structure image. The method comprises the following steps: firstly, acquiring a geological environment of target layer rock deposition diagenesis, determining image acquisition parameters, and performing multi-scale image acquisition to obtain a grayscale image; then obtaining characterization parameters by using an equivalent sphere algorithm, and constructing a digital model according to a screening standard; building a standard model and a digital result model by means of machine learning; and finally determining a hydrocarbon reservoir dominant reservoir and a fractured target layer section. According to the method, by means of multi-scale image acquisition, machine learning and the like, the accuracy of the three-dimensional digital model of the size and structural characteristics of the clastic particles of the clastic rock is improved, the exploration and development conditions of oil and gas can be predicted more accurately, the exploration and development efficiency is improved, the cost is reduced, and the problem of inaccurate measurement in the prior art is effectively solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of shale oil and gas exploration and development, and in particular relates to a method for quantitatively characterizing particle sizes of clastic rock debris based on CT scanning structural images. Background Art

[0002] The quantitative characterization of particle size of clastic rocks has been a world-class technical problem that has long been needed but not fundamentally solved in oil and gas exploration and development projects.

[0003] Currently, the main technologies for characterizing the particle size of clastic rock debris are manual qualitative measurement and qualitative detection using a scanning electron microscope. Manual measurement can only identify particles larger than the millimeter scale and can only measure particles displayed on a two-dimensional cross-section. Scanning electron microscopes can only detect particles smaller than micrometers. Due to the limited field of view of the scanning electron microscope, it can only depict the particle development characteristics within a very small range of the two-dimensional slice, and cannot effectively depict the spatial variation of particles. The current method can only obtain random probability values and cannot effectively characterize the true three-dimensional spatial state of the particle size of clastic rock debris. This results in low accuracy in the obtained clastic rock debris particle development status, which in turn leads to technical problems with low accuracy in oil and gas exploration and development.

[0004] Therefore, there is an urgent need to provide a three-dimensional quantitative characterization method for the particle size development of clastic rocks to improve the accuracy and efficiency of oil and gas exploration and development. Summary of the Invention

[0005] In view of this, it is necessary to provide a three-dimensional quantitative characterization method for the particle size of clastic rock debris based on CT scanning structural images to solve the technical problem that the measurement inaccuracy in the existing technology leads to low accuracy of the development status of clastic rock debris particles, and thus leads to low accuracy of oil and gas exploration and development predictions.

[0006] In order to solve the above technical problems, the present invention is implemented through the following technical solutions:

[0007] A method for quantitatively characterizing particle size of clastic rock debris based on CT scanning structural images, comprising:

[0008] Step 1: Obtain the geological environment of the target layer's rock deposition and diagenesis, and determine the clastic rock image acquisition parameters based on the geological environment;

[0009] Step 2: Collect rocks based on the clastic rock image acquisition parameters, perform multi-scale image acquisition, and obtain the clastic rock debris particle size and structure grayscale image;

[0010] Step 3: Identify the grayscale image of the clastic rock debris particle size and structure, and apply the equivalent sphere algorithm to obtain the characterization parameters of the clastic rock debris particle size and structure;

[0011] Step 4: Obtain screening criteria for the particle size and structural characterization parameters of clastic rocks, determine different particle sizes and structural characterization parameters from the particle size and structural characterization parameters of clastic rocks based on the screening criteria, and then construct a digital model of the particle size and structural composition of clastic rocks;

[0012] Step 5: Based on machine learning of micron CT scan structural images, a standard machine learning model of clastic rock particle size and structure is constructed;

[0013] Step 6: Based on the standard machine learning model of clastic rock debris grain size and structure, obtain the digital result model of clastic rock debris grain size and structure through machine learning;

[0014] Step 7: Based on the digital model of the clastic rock particle size and structure, determine the dominant reservoir and target fracturing interval of the oil and gas layer.

[0015] As a further optimization scheme of the present invention, in step 1, the geological environment includes: sedimentary phase type, hydrodynamic conditions of the sedimentary environment and diagenesis type; the determined clastic rock image acquisition parameters include: CT scanning resolution of 0.5-100 μm, scanning angle of 0-360°, and collecting an image every 0.5-1°.

[0016] As a further optimization scheme of the present invention, in step 2, when performing multi-scale image acquisition, three scales are divided: macro scale 1-10 mm, meso scale 100 μm-1 mm, and micro scale 1-100 μm; images are acquired in order from macro to micro. During the acquisition process, multi-scale image registration is achieved through three-dimensional coordinate calibration and SIFT feature matching to obtain the particle size and structural grayscale image of the clastic rock debris.

[0017] As a further optimization scheme of the present invention, in step 3, the obtained clastic rock debris particle size and structural grayscale image are binarized to segment the debris particle area; the volume V of each particle is calculated according to the sphere volume formula V= πr 3 , inversely deduce the equivalent sphere radius r= , which is used as the characterization parameter of particle size; the sphericity and flatness of the particles are calculated as structural characterization parameters, where: sphericity S=(4πr 2 ) / A, A is the surface area of the particle; flatness F=b / a, a and b are the lengths of the major and minor axes of the particle.

[0018] As a further optimization scheme of the present invention, in step 4, the screening criteria for the size of clastic rock debris particles are: excluding noise particles with a volume of less than 10 pixels and retaining particles with a particle size range of 5 μm-2 mm; the screening criteria for the structure of clastic rock debris particles are: analyzing particles with a sphericity greater than 0.4 and a flatness of 0.3-1.5.

[0019] As a further optimization scheme of the present invention, in step 5, the U-Net convolutional neural network algorithm is used to construct a standard machine learning model for the particle size and structure of clastic rock debris, which specifically includes: data preparation and preprocessing, U-Net model construction, model training, model evaluation and optimization.

[0020] As a further optimization scheme of the present invention, step 6 specifically includes: inputting the newly collected clastic rock debris particle size and structure grayscale image into the standard model, outputting the predicted particle size and structure characterization parameters, post-processing the output results, and finally constructing a digitized result model of the clastic rock debris particle size and structure.

[0021] As a further optimization solution of the present invention, in step 7, the determination of the dominant reservoir: the dominant reservoir index RQI=0.4D avg +0.3(1 / S o )+0.3P,S o =σg / Dg; where Davg is the average equivalent spherical radius, So is the particle size sorting coefficient, σg is the geometric standard deviation, Dg is the geometric mean particle size, and P is the porosity. When RQI>0.6, it is determined to be a dominant reservoir.

[0022] As a further optimization scheme of the present invention, in step 7, the target layer for fracturing is determined: combining rock mechanics parameters and ground stress data, a favorable reservoir layer with an average particle size of 100-300 μm, a particle size standard deviation of less than 50 μm, and a permeability of less than 1 mD is selected as the target layer for fracturing.

[0023] Compared with the existing technology, the quantitative characterization method of clastic rock particle size based on CT scanning structural images proposed in this invention has the following significant advantages and beneficial effects:

[0024] 1. Improving the comprehensiveness and accuracy of image acquisition: Image acquisition parameters are determined by understanding the geological environment of the target rock layer's sedimentary diagenesis. A high-resolution micron CT scanner is used, with a scanning resolution of 0.5-100μm, a scanning angle of 0-360°, and image acquisition intervals of 0.5-1°. This allows for comprehensive capture of information from different layers of the clastic rock. Image acquisition is performed at macro, meso, and micro scales, and multi-scale image registration is achieved using 3D coordinate calibration and SIFT feature matching. This allows for the acquisition of grayscale images of clastic rock particle size and structure, encompassing everything from the core structure to fine-grained structure, providing a rich and accurate data foundation for subsequent precise analysis.

[0025] 2. Accurately characterize particle features: After binarizing the grayscale image, the equivalent sphere algorithm is applied. Based on the sphere volume formula, the equivalent sphere radius is inferred as a parameter representing particle size. Sphericity and flatness are also calculated as structural parameters. This method accurately quantifies the characteristics of debris particles in terms of both size and structure. Compared with traditional manual qualitative measurement and scanning electron microscopy qualitative testing, it can more comprehensively and accurately reflect the true condition of the particles.

[0026] 3. Construct a reliable digital model: Establish clear screening criteria to exclude noise particles with a volume of less than 10 pixels, retain particles in the 5μm-2mm particle size range, analyze particles with a sphericity greater than 0.4 and a flatness of 0.3-1.5, and construct a digital model of the particle size and structural composition of clastic rocks based on this. This effectively removes interference information, allowing the model to more realistically reflect the actual composition and structure of the clastic rocks, thereby improving the reliability and accuracy of the model.

[0027] 4. Enhanced model prediction capabilities: A standard machine learning model was constructed using the U-Net convolutional neural network algorithm. Through steps including data preparation and preprocessing, model construction, training, evaluation, and optimization, the model's ability to learn and predict the characteristics of clastic rock particles was enhanced. Newly acquired images were fed into the standard model and post-processed to create a digital model, enabling more accurate predictions of clastic rock characteristics under different conditions.

[0028] 5. Accurately determine the key layer: Based on the digital result model, the dominant reservoir index formula (RQI=0.4Davg+0.3(1 / S o )+0.3P) to identify the dominant reservoir, and combined with rock mechanics parameters and in-situ stress data, select the dominant reservoir section with an average particle size of 100-300μm, a particle size standard deviation of less than 50μm, and a permeability of less than 1mD as the target fracturing section, providing precise target positioning for oil and gas exploration and development, improving exploration and development efficiency, and reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 It is based on the CT scan structural image and the grayscale image of the clastic rock debris particles and structural development.

[0031] Figure 2 To obtain a digital quantitative model of clastic rock particle size and structural development through machine learning.

[0032] Figure 3 To obtain digital quantitative distribution results of clastic rock debris particle size and structural development through machine learning. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] Figure 1 The grayscale images of clastic grains and structural development of clastic rocks obtained by CT scanning are shown; Figure 2 and Figure 3 The digital quantitative model diagram and distribution results of clastic rock debris particle size and structural development obtained through machine learning are respectively shown.

[0035] The present invention provides a method for quantitatively characterizing particle size of clastic rock debris based on CT scan structural images, comprising:

[0036] Step 1: Obtain the geological environment of the target rock layer's sedimentary diagenesis and determine the clastic rock image acquisition parameters based on the geological environment.

[0037] Specifically, the geological environment includes information such as sedimentary facies type (e.g., fluvial, lacustrine, or marine), hydrodynamic conditions (e.g., flow velocity and wave intensity), and diagenetic process type (e.g., compaction, cementation, or dissolution). The established parameters for clastic rock image acquisition include the use of a high-resolution micron CT scanner (e.g., the German Zeiss Xradia 620 Versa), with a CT scan resolution of 0.5-100 μm, a scanning angle of 0-360°, and acquisition of an image every 0.5-1°.

[0038] Step 2: Collect rocks based on the clastic rock image acquisition parameters, perform multi-scale image acquisition, and obtain the clastic rock debris particle size and structure grayscale image.

[0039] Specifically, a German Zeiss Xradia620Versa micron CT scanner was used, equipped with a 110 kV microfocus X-ray source, a detector resolution of 2048 × 2048 pixels, a scanning angle of 0–360°, a step size of 0.5°, an exposure time of 500 ms / frame, and grayscale image quantization of 16 bits (0–65535, corresponding to a density of 0.2–3.0 g / cm 3). Multi-scale scanning of clastic rocks: macroscale (1-10mm, whole core scanning, resolution 50-100μm), mesoscale (100μm-1mm, local area magnification, resolution 5-20μm), microscale (1-100μm, fine particle scanning, resolution 0.5-5μm). Images are collected in order from macro to micro. During the collection process, multi-scale image registration is achieved through three-dimensional coordinate calibration (accuracy ±1μm) and SIFT feature matching to obtain the grayscale image of clastic rock particle size and structure, such as Figure 1 shown.

[0040] Step 3: Identify the grayscale image of the clastic rock debris particle size and structure, and apply the equivalent sphere algorithm to obtain the characterization parameters of the clastic rock debris particle size and structure.

[0041] Specifically, the obtained clastic rock debris particle size and structural grayscale image are binarized to segment the debris particle area; the volume V of each particle is calculated according to the sphere volume formula V= πr 3 , inversely deduce the equivalent sphere radius r= , which is used as the characterization parameter of particle size; the sphericity and flatness of the particles are calculated as structural characterization parameters, where: sphericity S=(4πr 2 ) / A, A is the surface area of the particle; flatness F=b / a, a and b are the lengths of the major and minor axes of the particle.

[0042] Step 4: Obtain the screening criteria for the particle size and structural characterization parameters of clastic rocks, determine different particle sizes and structural characterization parameters from the particle size and structural characterization parameters of clastic rocks based on the screening criteria, and then construct a digital model of the particle size and structural composition of clastic rocks.

[0043] Specifically, the screening criteria for the size of clastic rock debris are: exclusion volume < 10 pixels (corresponding to actual volume < (resolution) 3 For example, when the resolution is 1 μm, the corresponding actual volume is less than 10 μm. 3 ) noise particles, retaining the particle size range of 5μm-2mm (covering the main particle size of clastic rocks). Clastic rock particle structure screening criteria: Analyze particles with a sphericity greater than 0.4 and a flatness of 0.3-1.5, and exclude extremely irregular or flaky particles.

[0044] Step 5: Based on machine learning of micron CT scan structural images, a standard machine learning model of clastic rock debris particle size and structure is constructed.

[0045] Specifically, a U-Net convolutional neural network was used to input a single-channel grayscale image block of 128×128×128 voxels. After 5 layers of downsampling (3×3 convolution kernel, stride 2) and 5 layers of upsampling (deconvolution + skip connection), the particle boundary probability map was output. The training data contained 100 rock core samples, which were divided into training set, validation set and test set according to the ratio of 7:2:1. The preprocessing steps included grayscale normalization, random rotation (±15°) and Gaussian noise addition (σ=0.05). The Adam optimizer was used with a learning rate of 0.001, weight decay of 1e-5, and a batch size of 16. The Dice loss and cross entropy loss were jointly optimized. The Dice coefficient of the validation set was ≥0.95 as the convergence condition. A standard machine learning model for the particle size and structure of clastic rocks was constructed.

[0046] Step 6: Based on the standard machine learning model of clastic rock debris particle size and structure, obtain the digital result model of clastic rock debris particle size and structure through machine learning.

[0047] Specifically, the newly collected grayscale images of clastic rock particle morphology and structure are input into the standard model. The model extracts the abstract features of the image through the encoder, and the decoder maps the features back to the original image size and outputs the predicted particle size and structural characterization parameters. The output results are post-processed and a cluster analysis algorithm is used to classify particles with similar characteristics according to their particle size, sphericity, flatness and other characteristics. Finally, a complete digital result model is constructed, such as Figure 2 and Figure 3 shown.

[0048] Step 7: Based on the digital model of the clastic rock particle size and structure, determine the dominant reservoir and target fracturing interval of the oil and gas layer.

[0049] Specifically, the determination of the dominant reservoir: the dominant reservoir index RQI=0.4D avg +0.3(1 / S o )+0.3P,S o =σg / Dg; where Davg is the average equivalent spherical radius, So is the particle size sorting coefficient, σg is the geometric standard deviation, Dg is the geometric mean particle size, and P is the porosity. When RQI>0.6, it is determined to be a dominant reservoir.

[0050] ;

[0051] ;

[0052] Among them, d i is the particle equivalent radius, n i is the number of particles, and N is the total number of particles.

[0053] Determination of target fracturing intervals: Based on rock mechanics parameters (elastic modulus 60-80 GPa, Poisson's ratio 0.2-0.3) and in-situ stress data (horizontal principal stress difference <5 MPa), a favorable reservoir interval with an average particle size of 100-300 μm, a particle size standard deviation <50 μm, and a permeability <1 mD was selected as the target fracturing interval.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0055] The contents not described in detail in the specification of the present invention belong to the prior art known to those skilled in the art.

Claims

1. A method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images, characterized in that: include: Step 1: Obtain the geological environment of the target layer's rock deposition and diagenesis, and determine the clastic rock image acquisition parameters based on the geological environment; Step 2: Collect rocks based on the clastic rock image acquisition parameters, perform multi-scale image acquisition, and obtain the clastic rock debris particle size and structure grayscale image; Step 3: Identify the grayscale image of the clastic rock debris particle size and structure, and apply the equivalent sphere algorithm to obtain the characterization parameters of the clastic rock debris particle size and structure; Step 4: Obtain screening criteria for the particle size and structural characterization parameters of clastic rocks, determine different particle sizes and structural characterization parameters from the particle size and structural characterization parameters of clastic rocks based on the screening criteria, and then construct a digital model of the particle size and structural composition of clastic rocks; Step 5: Based on machine learning of micron CT scan structural images, a standard machine learning model of clastic rock particle size and structure is constructed; Step 6: Based on the standard machine learning model of clastic rock debris grain size and structure, obtain the digital result model of clastic rock debris grain size and structure through machine learning; Step 7: Based on the digital model of the clastic rock particle size and structure, determine the dominant reservoir and target fracturing interval of the oil and gas layer.

2. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 1, the geological environment includes: sedimentary phase type, hydrodynamic conditions of the sedimentary environment, and diagenesis type; the determined clastic rock image acquisition parameters include: CT scanning resolution of 0.5-100 μm, scanning angle of 0-360°, and acquisition of an image every 0.5-1°.

3. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 2, when performing multi-scale image acquisition, three scales are divided: macroscale 1-10 mm, mesoscale 100 μm-1 mm, and microscale 1-100 μm; images are acquired in order from macro to micro. During the acquisition process, multi-scale image registration is achieved through three-dimensional coordinate calibration and SIFT feature matching to obtain the particle size and structural grayscale image of the clastic rock.

4. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 3, the obtained clastic rock debris particle size and structural grayscale image are binarized to segment the debris particle area; the volume V of each particle is calculated according to the sphere volume formula V= πr 3 , inversely deduce the equivalent sphere radius r= , which is used as the characterization parameter of particle size; the sphericity and flatness of the particles are calculated as structural characterization parameters, where: sphericity S=(4πr 2 ) / A, A is the surface area of the particle; flatness F=b / a, a and b are the lengths of the major and minor axes of the particle.

5. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 4, the clastic rock debris particle size screening criteria include excluding noise particles with a volume of less than 10 pixels and retaining particles with a size range of 5 μm-2 mm. The clastic rock debris particle structure screening criteria include analyzing particles with a sphericity greater than 0.4 and a flatness of 0.3-1.

5.

6. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 5, the U-Net convolutional neural network algorithm is used to construct a standard machine learning model for the particle size and structure of clastic rocks, which specifically includes: data preparation and preprocessing, U-Net model construction, model training, model evaluation and optimization.

7. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 6, the newly collected grayscale images of the clastic rock debris particle size and structure are input into the standard model, the predicted particle size and structure characterization parameters are output, the output results are post-processed, and finally a digital result model of the clastic rock debris particle size and structure is constructed.

8. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 7, determination of the dominant reservoir: dominant reservoir index RQI=0.4D avg +0.3(1 / S o )+0.3P,S o =σg / Dg; Where Davg is the average equivalent spherical radius, So is the particle size sorting coefficient, σg is the geometric standard deviation, Dg is the geometric mean particle size, and P is the porosity. When RQI>0.6, it is determined to be a dominant reservoir.

9. The method for quantitatively characterizing particle size of clastic rock based on CT scanning structural images according to claim 1, characterized in that: In step 7, the target layer for fracturing is determined: combining rock mechanics parameters and ground stress data, a favorable reservoir segment with an average particle size of 100-300 μm, a particle size standard deviation of less than 50 μm, and a permeability of less than 1 mD is selected as the target layer for fracturing.

Images