Reservoir pore structure full-scale characterization method, device and equipment based on cumulative fractal and medium

Abstract

The invention discloses a reservoir pore structure full-scale characterization method and device based on cumulative fractal, equipment and a medium, and relates to the technical field of computers. Comprising the following steps: acquiring multi-source measurement data of a to-be-measured rock sample; correcting the high-pressure mercury injection experimental data by utilizing the pore geometric information of macropores in the rock sample to be measured so as to construct a capillary pressure-saturation curve; a cumulative fractal equation is established, and a fractal dimension spectrum is obtained through inversion solution by using a preset algorithm; performing variable exponential conversion on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on a nonlinear mapping relation by using a fractal dimension spectrum to obtain a full-scale aperture distribution curve; and carrying out data fusion on the three-dimensional pore topology network structure parameters extracted based on CT scanning image data, the multi-dimensional fractal geometric features and the full-scale pore size distribution curve so as to construct a comprehensive representation model. Therefore, multi-source information can be organically fused, and full-scale accurate characterization of continuous heterogeneity of the reservoir pore structure is realized.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, equipment and medium for full-scale characterization of reservoir pore structure based on cumulative fractals. Background Technology

[0002] As global oil and gas exploration and development continues to expand into deeper, deeper water, and unconventional areas, reservoir evaluation faces unprecedented challenges. Unlike conventional high-permeability reservoirs, unconventional reservoir rocks have undergone intense compaction and diagenesis, exhibiting pore systems ranging from nanoscale matrix pores to micron-scale microfractures, as well as strong heterogeneity and complex connectivity. The occurrence and migration mechanisms of fluids within these reservoirs are significantly controlled by the microscopic pore-throat structure, exhibiting complex non-Darcy flow and slippage effects. Therefore, traditional macroscopic physical properties are insufficient to meet the accuracy requirements of reservoir analysis.

[0003] Currently, in the field of characterizing the microscopic pore structure and physical properties of rocks, numerous technical solutions have emerged that combine multiple experimental methods to characterize pore structures. However, these existing methods still have significant limitations. First, the data fusion methods are simplistic and fail to achieve deep fusion at the mechanistic level: Most existing techniques simply perform linear splicing or piecewise fitting of data from Mercury Injection Capillary Pressure (MICP) and Nuclear Magnetic Resonance (NMR). Second, they fail to effectively address the inherent defects of key experiments: the "pore shielding effect" in MIP experiments easily leads to distortion in the characterization of large pores, while CT scans, limited by resolution, struggle to capture nanoscale micropore information. Existing techniques lack effective mechanisms to correct these systematic errors introduced by the experimental principles themselves. Third, the description of pore structure heterogeneity is overly simplistic: most methods are based on single fractal or simple piecewise fractal assumptions, failing to accurately characterize the continuous and heterogeneous variations of pore structure across the entire scale. Fourth, the representation dimension is singular and there is a lack of comprehensive structural models: Existing methods mostly focus on obtaining pore size distribution curves, but fail to organically integrate pore size information with the three-dimensional topological connection relationship of pore space, surface geometry, etc., and therefore cannot construct a comprehensive structural model that can fully reflect the physical nature of reservoir seepage.

[0004] As can be seen from the above, how to organically integrate multi-source information to achieve accurate full-scale characterization of the continuous heterogeneity of reservoir pore structure is an urgent problem to be solved. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a method, apparatus, device, and medium for full-scale characterization of reservoir pore structure based on cumulative fractals, which can organically fuse multi-source information to achieve accurate full-scale characterization of continuous heterogeneous reservoir pore structure. The specific solution is as follows: Firstly, this application provides a full-scale characterization method for reservoir pore structure based on cumulative fractals, including: Acquire multi-source measurement data of the rock sample to be tested; the multi-source measurement data includes target high-pressure mercury intrusion test data, nuclear magnetic resonance transverse relaxation time T2 spectrum data, and CT scan three-dimensional image data; Based on the CT scan three-dimensional image data, the pore geometry information of the target pores in the rock sample to be tested is extracted; the target pores are pores whose size meets the preset large pore determination conditions, and the pore geometry information includes pore radius distribution and topological parameters; The pore geometry information of the target pores is used to correct the target large-pore section data in the high-pressure mercury injection experiment data caused by the pore shielding effect, so as to construct the capillary pressure-saturation curve using the corrected data; Based on the capillary pressure-saturation curve, a cumulative fractal equation is established, and the cumulative fractal equation is inverted and solved using a preset iterative truncation singular value decomposition algorithm to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested. The fractal dimension spectrum is used to establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample to be tested, and the nuclear magnetic resonance transverse relaxation time T2 spectrum data is subjected to a variable exponential transformation based on the nonlinear mapping relationship to obtain the full-scale pore size distribution curve. Based on the CT scan 3D image data, the 3D pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test are extracted. The 3D pore topology network structure parameters, the multidimensional fractal geometric features, and the full-scale pore size distribution curve are then fused across scales to construct a comprehensive characterization model for representing the full-scale pore structure of the rock. The multidimensional fractal geometric features include a sequence of 3D fractal dimension and 2D fractal dimension.

[0006] Optionally, the cumulative fractal equation is used to describe the fractal characteristics of the pore structure in the capillary pressure-saturation curve, and its expression is a weighted sum of multiple fractal components; wherein each fractal component is determined by the fractal dimension and the weighting coefficient corresponding to the fractal dimension.

[0007] Optionally, the step of using a preset iterative truncated singular value decomposition algorithm to invert and solve the cumulative fractal equation to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested includes: The cumulative fractal equation is transformed into a system of linear equations, and a sparse kernel matrix corresponding to the system of linear equations is constructed; the solution vector of the system of linear equations is composed of the weight coefficients corresponding to each fractal component. Singular value decomposition is performed on the sparse kernel matrix to obtain the corresponding singular value decomposition result. A singular value truncation parameter is determined based on a preset information criterion formula so that the singular value decomposition result can be truncated using the singular value truncation parameter to obtain the truncated result. An initial solution vector is generated based on the truncation result, and iterative optimization and negative value constraint processing are performed based on the initial solution vector to obtain the fractal dimension spectrum.

[0008] Optionally, the step of establishing a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum, and performing a variable exponential transformation on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on the nonlinear mapping relationship to obtain a full-scale pore size distribution curve, includes: A variable exponential transformation model between pore radius and transverse relaxation time is established based on the fractal dimension values ​​corresponding to different pore intervals in the fractal dimension spectrum; wherein, the transformation exponent of the variable exponential transformation model is related to the fractal dimension value of the corresponding pore interval. The cumulative pore size distribution corresponding to the capillary pressure-saturation curve is used as the target benchmark, and the surface relaxation strength benchmark coefficient is determined based on the target benchmark using a preset optimization algorithm. The nuclear magnetic resonance transverse relaxation time T2 spectrum data is converted into the full-scale aperture distribution curve using the variable exponential conversion model and the surface relaxation intensity reference coefficient.

[0009] Optionally, the step of extracting the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test based on the CT scan three-dimensional image data includes: The CT scan three-dimensional image is segmented and connectivity analysis is performed to obtain a three-dimensional connected pore space model. Based on the three-dimensional connected pore space model, the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test are extracted.

[0010] Optionally, the step of extracting the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test based on the three-dimensional connected pore space model includes: Based on the three-dimensional connected pore space model, a pore network model is identified and constructed using a preset pore network model extraction algorithm, and the three-dimensional pore topology network structure parameters of the rock sample to be tested are extracted based on the pore network model; the pore network model includes pore units and throat units connecting the pore units; the three-dimensional pore topology network structure parameters include coordination number, BiEuler number, pore shape factor, and centroid path tortuosity; The three-dimensional fractal dimension of the target pore system of the rock sample under test is calculated based on the three-dimensional connected pore space model. The two-dimensional fractal dimension is calculated layer by layer from consecutive two-dimensional slices of the CT scan three-dimensional image to generate the two-dimensional fractal dimension sequence.

[0011] Optionally, the step of correcting the target large-diameter segment data in the target high-pressure mercury intrusion test data caused by the pore shielding effect using the pore geometry information of the target pore includes: Based on a preset pore radius threshold, the target high-pressure mercury intrusion test data is stitched together with the pore radius distribution data extracted from the CT scan three-dimensional image data to obtain the corresponding corrected data.

[0012] Secondly, this application provides a full-scale characterization device for reservoir pore structure based on cumulative fractals, comprising: The data acquisition module is used to acquire multi-source measurement data of the rock sample to be tested; the multi-source measurement data includes target high-pressure mercury intrusion test data, nuclear magnetic resonance transverse relaxation time T2 spectrum data, and CT scan three-dimensional image data; The information extraction module is used to extract the pore geometry information of the target pores in the rock sample to be tested based on the CT scan three-dimensional image data; the target pores are pores whose size meets the preset large pore determination conditions, and the pore geometry information includes pore radius distribution and topological parameters; The data correction module is used to correct the target large-diameter segment data in the target high-pressure mercury injection experimental data caused by the pore shielding effect using the pore geometry information of the target pore, so as to construct the capillary pressure-saturation curve using the corrected data; The fractal inversion module is used to establish a cumulative fractal equation based on the capillary pressure-saturation curve, and to perform inversion solution on the cumulative fractal equation using a preset iterative truncation singular value decomposition algorithm to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested. The spectrum conversion module is used to establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum, and to perform a variable exponential conversion on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on the nonlinear mapping relationship to obtain the full-scale pore size distribution curve. The model building module is used to extract the three-dimensional pore topology network structure parameters and multi-dimensional fractal geometric features of the rock sample under test based on the CT scan three-dimensional image data, and to perform cross-scale data fusion of the three-dimensional pore topology network structure parameters, the multi-dimensional fractal geometric features and the full-scale pore size distribution curve, so as to construct a comprehensive characterization model for characterizing the full-scale pore structure of the rock; the multi-dimensional fractal geometric features include a sequence of three-dimensional fractal dimension and two-dimensional fractal dimension.

[0013] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned method for full-scale characterization of reservoir pore structure based on cumulative fractals.

[0014] Fourthly, this application provides a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned method for full-scale characterization of reservoir pore structure based on cumulative fractals.

[0015] This application provides a method for full-scale characterization of reservoir pore structure based on cumulative fractals to obtain multi-source measurement data of a rock sample. The multi-source measurement data includes target high-pressure mercury intrusion porosimetry (HIP) experimental data, nuclear magnetic resonance (NMR) transverse relaxation time (T2) spectrum data, and CT scan three-dimensional image data. Based on the CT scan three-dimensional image data, pore geometry information of target pores in the rock sample is extracted. This pore geometry information is then used to correct the target large-pore segment data in the target HIP experimental data caused by pore shielding effects, so as to construct a capillary pressure-saturation curve using the corrected data. The pore geometry information includes pore radius distribution and topological parameters. A cumulative fractal equation is established based on the capillary pressure-saturation curve, and a preset iterative truncated singular value decomposition algorithm is used to refine the cumulative fractal equation. Inversion calculations are performed to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample under test. A nonlinear mapping relationship is established between the nuclear magnetic resonance transverse relaxation time (T2) spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum. Based on this nonlinear mapping relationship, a variable exponential transformation is performed on the T2 spectrum data to obtain a full-scale pore size distribution curve. Three-dimensional pore topology network structural parameters and multidimensional fractal geometric features of the rock sample under test are extracted based on the CT scan three-dimensional image data. The three-dimensional pore topology network structural parameters, the multidimensional fractal geometric features, and the full-scale pore size distribution curve are then fused across scales to construct a comprehensive characterization model for representing the full-scale pore structure of the rock. The multidimensional fractal geometric features include a sequence of three-dimensional fractal dimensions and two-dimensional fractal dimensions.

[0016] As can be seen from the above, this application effectively overcomes the technical bottlenecks of single or traditional combined methods by integrating three experimental techniques: high-pressure mercury intrusion porosimetry (MICP), nuclear magnetic resonance (NMR), and X-ray CT scanning, and introducing an innovative cumulative fractal theory model. Using the target pore geometry information extracted by CT scanning, the data distortion problem in the large-pore segment caused by the "pore shielding effect" in the MIP experiment is corrected. Furthermore, based on the corrected capillary pressure curve, the fractal dimension spectrum that continuously varies with the pore throat radius is derived through the cumulative fractal equation, achieving a precise and quantitative characterization of the pore structure heterogeneity. On this basis, the fractal dimension spectrum is used to perform a variable-exponential nonlinear transformation on the NMR T2 spectrum, solving the pore size distribution error and multiple solutions problem caused by neglecting structural heterogeneity in traditional fixed transformation models, ultimately obtaining a full-scale, high-precision pore size distribution curve. By fusing microscopic pore size information with macroscopic three-dimensional topological network parameters and multidimensional fractal geometric features extracted by CT across scales, a comprehensive pore structure model that combines pore geometry, connectivity, and heterogeneity attributes is constructed. This enables the organic fusion of multi-source information, achieving accurate full-scale characterization of the continuous heterogeneity of reservoir pore structure. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 This is a flowchart of a full-scale characterization method for reservoir pore structure based on cumulative fractals disclosed in this invention. Figure 2 This is a flowchart of a specific method for full-scale characterization of reservoir pore structure based on cumulative fractals disclosed in this invention; Figure 3 This is a flowchart of an iterative optimization process for inverting and solving cumulative fractal equations, as disclosed in this invention. Figure 4 This invention discloses a fractal dimension spectrum curve of a rock sample under test that continuously varies with the pore throat radius. Figure 5 This invention discloses a full-scale aperture distribution curve based on fractal dimension spectral exponential transformation. Figure 6 This is a statistical diagram of the coordination number distribution of a pore network model constructed based on CT scan three-dimensional images, as disclosed in this invention. Figure 7 This is a diagram showing the Euler number analysis of a pore network model constructed based on CT scan three-dimensional images disclosed in this invention. Figure 8 This invention discloses a statistical diagram of the pore shape factor distribution in a pore network model constructed based on CT scan three-dimensional images. Figure 9 This is a comparison chart of the variation trends of two-dimensional fractal dimension and porosity based on layer-by-layer calculation of CT scan three-dimensional images disclosed in this invention; Figure 10 This is a schematic diagram of a reservoir pore structure full-scale characterization device based on cumulative fractals disclosed in this invention; Figure 11 This is a structural diagram of an electronic device disclosed in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Currently, in the field of characterizing the microscopic pore structure and physical properties of rocks, numerous technical solutions have emerged that combine multiple experimental methods to characterize pore structure. However, these existing methods still have significant limitations. First, the data fusion methods are simplistic and fail to achieve deep fusion at the mechanistic level: most existing techniques simply perform linear splicing or piecewise fitting of data from high-pressure mercury intrusion porosimetry (MICP), nuclear magnetic resonance (NMR), etc. Second, they fail to effectively address the inherent defects of key experiments: the "pore shielding effect" in MIP experiments easily leads to distortion in the characterization of large pores, while CT scans, limited by resolution, struggle to capture nanoscale micropore information. Existing techniques lack effective mechanisms to correct these systematic errors introduced by the experimental principles themselves. Third, the description of the heterogeneity of pore structure is overly simplistic: most methods are based on single fractal or simple piecewise fractal assumptions, failing to accurately characterize the continuous and heterogeneous variations of pore structure across the entire scale. Fourth, the existing methods focus on obtaining pore size distribution curves, failing to organically integrate pore size information with the three-dimensional topological connections of the pore space and surface geometry. Therefore, they cannot construct a comprehensive structural model that fully reflects the physical nature of reservoir seepage. To address this, this application provides a full-scale characterization scheme for reservoir pore structure based on cumulative fractals, which can organically fuse multi-source information to achieve accurate full-scale characterization of the continuous heterogeneity of reservoir pore structure.

[0021] See Figure 1 As shown in the embodiments of this application, a full-scale characterization method for reservoir pore structure based on cumulative fractals is disclosed, including: Step S11: Obtain multi-source measurement data of the rock sample to be tested.

[0022] In this embodiment, the multi-source measurement data of the rock sample to be tested includes high-pressure mercury intrusion (MICP) experimental data, nuclear magnetic resonance (NMR) T2 spectrum data, and X-ray CT scan three-dimensional image data; the MICP data includes capillary pressure curves and mercury intrusion saturation.

[0023] Step S12: Extract the pore geometry information of the target pores in the rock sample to be tested based on the CT scan three-dimensional image data.

[0024] In this embodiment, the target pores are pores whose size meets the preset macropore determination criteria, and the pore geometry information includes pore radius distribution and topological parameters. Digital core reconstruction is performed on the CT scan 3D image to extract the pore radius distribution and topological parameters of the macropores.

[0025] Step S13: Use the pore geometry information of the target pores to correct the target large-diameter segment data in the target high-pressure mercury injection experimental data caused by the pore shielding effect, so as to use the corrected data to construct a capillary pressure-saturation curve.

[0026] In this embodiment, the high-pressure mercury intrusion porosimetry experiment will attribute the large pore volume to the connected narrow throat, resulting in a pore shielding effect. To reduce the impact of the pore shielding effect, it is necessary to correct the target high-pressure mercury intrusion porosimetry experimental data using the pore geometry information of the target pores and construct a corrected capillary pressure-saturation curve. Specifically, the correction of the target large-pore segment data in the target high-pressure mercury intrusion porosimetry experimental data caused by the pore shielding effect using the pore geometry information of the target pores may include: based on a preset pore radius threshold, stitching the target high-pressure mercury intrusion porosimetry experimental data with pore radius distribution data extracted from the CT scan three-dimensional image data to obtain the corresponding corrected data. That is, based on a preset pore radius threshold, the high-pressure mercury intrusion porosimetry experimental data is stitched with pore radius distribution data extracted from the CT scan three-dimensional image data; for pore diameter ranges larger than the preset threshold, the pore radius distribution data extracted from the CT scan is used; for pore diameter ranges smaller than or equal to the preset threshold, the high-pressure mercury intrusion porosimetry experimental data is retained.

[0027] Step S14: Establish a cumulative fractal equation based on the capillary pressure-saturation curve, and use a preset iterative truncation singular value decomposition algorithm to invert and solve the cumulative fractal equation to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested.

[0028] In this embodiment, based on the modified capillary pressure-saturation curve, cumulative fractal theory is introduced. Considering the segmented and continuous heterogeneous characteristics of the pore structure, a cumulative fractal equation is established. This cumulative fractal equation describes the fractal characteristics of the pore structure in the capillary pressure-saturation curve, and its expression is a weighted sum of multiple fractal components. Each fractal component is determined by its fractal dimension and the corresponding weighting coefficient. Specifically, considering the heterogeneity of the rock pore structure, the cumulative pore volume and capillary pressure follow a cumulative fractal distribution law, as shown in the following cumulative fractal equation: ; in, is the cumulative saturation of pores with a radius less than r, used to reflect the proportion of pore volume below a preset pore size threshold r to the total volume; n is the total number of fractal dimensions; The weighting coefficients for the i-th fractal dimension component; Capillary pressure; Let be the local porosity fractal dimension of the i-th interval.

[0029] Furthermore, the aforementioned cumulative fractal equation is essentially a linear summation of multiple fractal dimensions Di and their weights ai, which can be transformed into an exponential form as follows: ; Furthermore, the above summation form equation is transformed into vector and matrix forms adapted to the matrix operation framework, thus transforming it into solving the optimal solution of the following least squares problem, as shown in the following formula: ; Where S is the cumulative regularized pore volume under a specific capillary pressure; A is the sparse kernel matrix, whose elements are calculated based on the capillary pressure and the preset candidate fractal dimension; X is the fractal dimension distribution vector, which consists of the weight coefficients corresponding to each fractal component. The composition is the unknown quantity to be solved; Let L be the L2 norm of the vector, also known as the Euclidean norm, which measures the overall magnitude of the residuals. This formula is an objective function established after transforming the cumulative fractal equation into an inversion problem, aiming to find the optimal fractal dimension distribution vector by minimizing the L2 norm of the residuals.

[0030] Furthermore, the system of equations is solved using the Iterative Truncation Singular Value Decomposition (TSVD) algorithm to obtain a non-negative fractal dimension distribution vector X0. The result is then iteratively optimized to obtain the corresponding fractal dimension spectrum D(r). Specifically, the step of using a preset iterative truncated singular value decomposition algorithm to invert and solve the cumulative fractal equations to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested can include: transforming the cumulative fractal equations into a system of linear equations and constructing a sparse kernel matrix corresponding to the linear equations; the solution vector of the linear equations consists of the weight coefficients corresponding to each fractal component; performing singular value decomposition on the sparse kernel matrix to obtain the corresponding singular value decomposition results, and determining the singular value truncation parameter based on a preset information criterion formula, so as to truncate the singular value decomposition results using the singular value truncation parameter to obtain the truncated result; generating an initial solution vector based on the truncated result, and performing iterative optimization and negative value constraint processing based on the initial solution vector to obtain the fractal dimension spectrum. .in, Surface roughness characteristics representing different pore sizes This represents the pore volume weight corresponding to this feature.

[0031] Step S15: Establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample to be tested using the fractal dimension spectrum, and perform a variable exponential transformation on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on the nonlinear mapping relationship to obtain the full-scale pore size distribution curve.

[0032] In this embodiment, a nonlinear mapping based on the fractal geometric characteristics of the rock pore structure is used to achieve a variable exponential transformation of the nuclear magnetic resonance transverse relaxation time (T2) spectrum data. Specifically, establishing a nonlinear mapping relationship between the nuclear magnetic resonance T2 spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum, and performing a variable exponential transformation on the nuclear magnetic resonance T2 spectrum data based on the nonlinear mapping relationship to obtain a full-scale pore size distribution curve, may include: establishing a variable exponential transformation model between pore radius and transverse relaxation time based on the fractal dimension values ​​corresponding to different pore intervals in the fractal dimension spectrum; wherein the transformation exponent of the variable exponential transformation model is related to the fractal dimension value of the corresponding pore interval; using the cumulative pore size distribution corresponding to the capillary pressure-saturation curve as the target benchmark, and determining the surface relaxation intensity benchmark coefficient based on the target benchmark using a preset optimization algorithm; and converting the nuclear magnetic resonance T2 spectrum data into the full-scale pore size distribution curve using the variable exponential transformation model and the surface relaxation intensity benchmark coefficient. That is, the cumulative saturation S is essentially a weighted power-law function of the pore radius r, and the capillary pressure is inversely proportional to the pore throat radius r. Based on the nuclear magnetic resonance relaxation mechanism and fractal geometry theory, the relationship between the transverse relaxation time T2 and the pore radius r is controlled by the fractal dimension of the pore surface. Using the extracted geometric features, a piecewise variable exponential transformation model is constructed, and the specific model function expression is shown below: ; in, is the pore radius corresponding to the i-th pore interval after transformation; C is the surface relaxation strength reference coefficient to be determined. This represents the local geometric fingerprint value corresponding to this saturation range. This step corrects the traditional fixed-index conversion to a variable-index conversion that reflects the true heterogeneity of the rock.

[0033] Furthermore, using the cumulative pore size distribution curve corresponding to the CT-corrected high-pressure mercury intrusion porosimetry data as a target reference, the baseline coefficient C is adjusted to optimize the overlap between the corrected NMR pore size distribution curve and the target curve across the entire scale range, thereby determining the final conversion coefficient C. Using the determined optimal coefficient C and the fractal dimension spectrum, the original NMR T2 spectrum is completely converted into a full-scale pore size distribution curve, realizing the full-scale coverage capability of NMR constrained by the geometric precision of MICP.

[0034] Step S16: Extract the three-dimensional pore topology network structure parameters and multi-dimensional fractal geometric features of the rock sample under test based on the CT scan three-dimensional image data, and perform cross-scale data fusion of the three-dimensional pore topology network structure parameters, the multi-dimensional fractal geometric features and the full-scale pore size distribution curve to construct a comprehensive characterization model for characterizing the full-scale pore structure of the rock.

[0035] In this embodiment, the CT scan three-dimensional image is segmented and connectivity analyzed to obtain a three-dimensional connected pore space model. Specifically, in one implementation, representative volume elements can be determined based on the trend of porosity changing with voxel volume; median filtering is performed on the grayscale image acquired by X-ray CT scan for noise reduction; an interactive threshold segmentation algorithm is used to binarize the image into pore space and skeleton space; and a connectivity algorithm is used to remove isolated pores to obtain a connected pore space model.

[0036] Furthermore, based on the three-dimensional connected pore space model, the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test are extracted. Specifically, the extraction of the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test based on the three-dimensional connected pore space model may include: identifying and constructing a pore network model using a preset pore network model extraction algorithm based on the three-dimensional connected pore space model, and extracting the three-dimensional pore topology network structure parameters of the rock sample under test based on the pore network model; the pore network model includes pore units and throat units connecting the pore units; calculating the three-dimensional fractal dimension of the target pore system of the rock sample under test based on the three-dimensional connected pore space model; calculating the two-dimensional fractal dimension layer by layer for continuous two-dimensional slices of the CT scan three-dimensional image to generate the two-dimensional fractal dimension sequence. For example, in some specific embodiments, the maximum sphere algorithm can be used to extract the network from the obtained connected pore space, identify pores and throats, and calculate the three-dimensional pore topology network structure parameters.

[0037] The structural parameters of the three-dimensional porous topology network include coordination number, Euler number, pore shape factor, and centroid path tortuosity.

[0038] The coordination number is the number of throats connected to a single pore, used to quantitatively evaluate the connectivity of a pore network.

[0039] The BiEuler number is used to evaluate the topological properties of pore structures per unit volume, and its calculation formula is shown below: ; in, It is a number greater than Euler's number; is the Euler number; V is the reference volume.

[0040] The pore shape factor is used to characterize the irregularity of the pore cross-section, and the calculation formula is as follows: ; Where G is the pore shape factor; P is the pore perimeter; and A is the pore cross-sectional area.

[0041] The centroid path tortuosity is the ratio of the actual flow path length of the fluid within the pore space to the straight-line distance between two points, used to characterize the tortuosity of the fluid's seepage path in a porous medium.

[0042] Furthermore, the box-counting method was used to calculate the three-dimensional fractal dimension of the CT digital core, characterizing the roughness of the macroporous surface. The two-dimensional fractal dimension of the slice images was calculated layer by layer, and the fluctuation trend of the two-dimensional fractal dimension with the stratigraphic sequence was used to characterize the vertical depositional heterogeneity of the rock. The generated full-scale pore size distribution curve containing micropore information was mapped onto the constructed three-dimensional macroporous skeleton network. Using the obtained cumulative fractal dimension spectrum as the basis for self-similarity filling, a comprehensive rock pore structure model was constructed, including full-scale pore size distribution, three-dimensional connectivity, and surface roughness.

[0043] As can be seen from the above, the embodiments of this application utilize CT data to constrain large pore segments, MIP data to constrain throat segments, and construct a continuously changing conversion coefficient spectrum through a cumulative fractal model, thereby achieving dynamic and nonlinear correction of the NMR T2 spectrum. This not only overcomes the pore shielding effect and resolution limitations of traditional methods, but also allows for quantitative evaluation of reservoir heterogeneity through the fractal dimension spectrum obtained by inversion. Thus, it enables the organic fusion of multi-source information, achieving accurate full-scale characterization of the continuous heterogeneity of reservoir pore structure.

[0044] See Figure 2 As shown in the embodiments, this application discloses a specific method for full-scale characterization of reservoir pore structure based on cumulative fractals, including: This embodiment selects a typical low-permeability, dense dolomite core sample for quantitative characterization of its pore structure across all scales. The sample is dense and highly heterogeneous, with a porosity of 10.53%. First, multi-source experimental data from mercury intrusion porosimetry (MIP), nuclear magnetic resonance (NMR), and CT scans were acquired. For the high-pressure mercury intrusion porosimetry (MICP) experiment, a dolomite core sample was used, and the experiment was conducted according to standard procedures using an AutoPore IV590 fully automated mercury intrusion porosimetry instrument. The maximum mercury intrusion pressure was 414 MPa (corresponding to a minimum pore throat radius of approximately 2 nm). Capillary pressure curves, pore throat radius distribution, and mercury intrusion saturation data were obtained. For NMR, the sample was dried at 95°C for 24 hours to remove pore fluid and then subjected to vacuum treatment. Afterward, formation water was saturated under pressure, and the transverse relaxation time (T2) spectrum was measured using a 2 MHz NMR spectrometer. The echo interval (TE) was set to 0.2 ms, and the waiting time (TW) was set to 3000 ms. The original T2 distribution spectrum was obtained. The X-ray CT scanning experiment used micron-level CT scans on samples with a processed diameter of 3 mm and a length of 1 cm, with a resolution accuracy set to 2.4 μm. 2020 two-dimensional slice images were acquired, and after image correction, ring artifact removal, and three-dimensional reconstruction, a three-dimensional grayscale image data volume of the rock was obtained.

[0045] Furthermore, this embodiment utilizes the macropore geometric information extracted from CT scans to correct the pore shielding effect of mercury intrusion porosimetry (MIP) data, constructing a full-pore-scale corrected capillary pressure curve. To extract the CT pore size distribution, the acquired CT images are subjected to median filtering for noise reduction, and an interactive threshold segmentation algorithm is used to binarize the images into pores and a skeleton. The pore volume distribution with an equivalent pore size greater than the resolution is statistically analyzed. The MICP experiment will attribute the large pore volume to the connected narrow throats, resulting in a pore shielding effect. The MICP pore size distribution is compared with the CT pore size distribution. It is found that for the large pore range with radius r > 5 μm, the pore volume ratio measured by MICP is significantly lower than that of CT results. Therefore, 5 μm is used as the stitching threshold to construct a corrected full-pore-scale capillary pressure-saturation curve. That is, in the range r ≤ 5 μm, the pore radius distribution data measured by MICP is retained; otherwise, the pore volume radius distribution data extracted from CT scans is used to replace the MICP data.

[0046] Furthermore, based on the obtained mercury intrusion pressure corrected capillary pressure data and corresponding mercury saturation data, this embodiment introduces cumulative fractal theory, considers the heterogeneity of the rock pore structure, and establishes a cumulative fractal equation. The mercury saturation is converted into a cumulative regularized pore volume S, and the fractal dimension is assumed to be... The matrix is ​​divided into 50 equal parts (step size 0.02), generating 50 candidate fractal dimensions. The matrix dimension is m×n, where m is the number of pressure measurement points and n=50 is the preset number of fractal dimensions to construct a sparse kernel matrix A. Each element... The calculation formula is as follows: ; in, For the j-th pressure point, Let be the i-th pre-defined fractal dimension. That is, the equation is transformed into a linear system of equations S=AX using the least squares method.

[0047] Furthermore, matrix A is decomposed into singular value matrices and orthogonal matrices to facilitate subsequent stability analysis and truncation. The singular value decomposition of matrix A is performed as follows: ; Where U is an m×m orthogonal matrix (left singular vector); Z is an m×n diagonal matrix (singular values ​​are arranged in descending order, z1≥z2≥...). ≥z l l = min(m, n)). V is an n × n orthogonal matrix (right singular vector). Singular value z i It decays rapidly with index i.

[0048] See Figure 3 As shown, based on the AIC (Akaike Information Criterion) method, the optimal solution stage position is determined to select the number of singular values ​​q to retain, in order to balance stability and information fidelity. The AIC criterion formula is as follows: ; Where p is the cutoff parameter to be optimized (retaining the first p singular values); l is the total number of non-zero singular values; and n is the number of pressure measurement points. Iterate through all possible p values, calculate the corresponding AIC value, and select the value that minimizes the AIC. and determine the cut-off position. Keep the first q singular values ​​and discard the rest. to The truncated matrix is ​​obtained. .in, These are the submatrices corresponding to the first q singular values. Based on the truncated matrices, the initial solution X0 is calculated as follows: ; Furthermore, the initial solution is iteratively optimized and corrected, with the specific steps as follows: Calculate the residual: residual matrix and mean square residual .

[0049] Termination condition judgment: If (Preset threshold, set to 1000) and If the maximum number of iterations I is reached, the iteration terminates and X is output. i .

[0050] Update the solution vector: Calculate the correction amount The corrected value is then superimposed with the initial solution, and the updated solution is forcibly set to zero. This process is repeated until a termination condition is met to obtain the weight of each element in X corresponding to a preset fractal dimension, representing the proportion of that fractal dimension in the porous structure. The final solution is then calculated as follows: Figure 4 The fractal dimension spectrum D(r) shown is continuously varying with the throat radius.

[0051] See Figure 4 As shown, the fractal dimension spectrum of this sample exhibits a significant high-value single-peak concentration. This suggests that the reservoir space of the sample is not dominated by regular large pores, but rather consists of a large number of complex micro- and nano-scale pores or throats, and the inner walls of the pores may be adhered with complex clay minerals, resulting in a large specific surface area and irregular structure. This continuous spectrum concentrated in the high-value range (2.65~2.90) accurately characterizes the microscopic features of this rock sample, which are dominated by micropores and have extremely strong heterogeneity, constituting the unique fractal geometry of this rock.

[0052] Furthermore, this embodiment utilizes a fractal dimension spectrum to construct a variable exponential nonlinear transformation model, inverting the NMR T2 spectrum into a high-precision full-scale pore size distribution curve. It abandons the linear transformation formula that assumes a constant fractal dimension in traditional methods, instead using the fractal dimension spectrum obtained in the preceding steps to construct a variable exponential model that conforms to the heterogeneous characteristics of the rock. The main peak of the original T2 spectrum is distributed within a narrow range of 0–5 ms. If a traditional linear transformation is used, the corresponding pore size distribution will be extremely uniform, failing to reflect the heterogeneity of the reservoir. This embodiment introduces fractal dimension spectrum constraints; the peak value of the rock fractal dimension spectrum extracted in this embodiment is located at D=2.80. According to the transformation model described herein, the transformation exponent n is as high as 5.0, successfully mapping the narrow signal range to cover a wider pore size distribution, truly restoring the complex matrix pore structure of the rock sample. During implementation, the full-pore size cumulative distribution curve after CT scan macropore geometric correction and MICP throat feature fusion in step S2 is used as the target calibration curve. A nonlinear least-squares objective function was constructed, and the NMR T2 spectrum data were substituted into a variable exponential transformation model constrained by the cumulative fractal dimension spectrum D(r). The optimal surface relaxation intensity reference coefficient C was calculated using a global iterative optimization algorithm. Finally, using the determined coefficient C and the fractal dimension spectrum D(r) that continuously varies with aperture, a point-by-point nonlinear mapping was performed on the original NMR T2 spectrum, and the results are shown below. Figure 5As shown, comparing the original data with the inversion results reveals a low-amplitude secondary peak in the relaxation time range of 60–100 ms in the original NMR T2 spectrum. By introducing the cumulative fractal dimension spectrum obtained from the joint inversion of MICP and CT for variable dimension constraint transformation, this T2 signal range is correctly nonlinearly stretched and mapped to the 100–300 μm macropore / microcrack range. This result is highly consistent with the actual macropore distribution characteristics observed by CT scans. This embodiment can effectively identify the complex dual-medium characteristics of micro / nanoscale matrix pores and microcracks coexisting, and through the geometric constraint of the fractal dimension spectrum, successfully overcomes the problems of underestimation of macropore volume and misjudgment of pore size caused by the pore shielding effect in the single MICP method, achieving accurate reconstruction of the full-scale pore structure of the reservoir.

[0053] Furthermore, this embodiment extracts three-dimensional topological parameters and multidimensional fractal heterogeneity features from CT images and integrates them with the full-scale pore size distribution to establish a comprehensive characterization model of the rock's full-scale pore structure, encompassing geometric, topological, and fractal attributes. Based on the principle that porosity tends to stabilize with increasing voxel volume, a 600×600×600 voxel cube region is extracted from the CT scan data of the rock sample as a representative volume element for analysis. An interactive threshold segmentation algorithm is used, setting a grayscale threshold to segment the image into pores and a framework. A connected component labeling algorithm is used to classify pores into connected pores and dead pores. Analysis shows that isolated pore space accounts for 3.16% of the total volume, and the pore space volume accounts for 41.20%, quantitatively describing the complex pore space topology. The maximum sphere algorithm is used to extract the network of connected pore space, constructing a pore network model (PNM) composed of spheres and cylinders. This model realistically recreates the three-dimensional spatial connection framework of medium-to-large pores and micro-cracks in the rock sample.

[0054] Furthermore, based on the constructed PNM model, the following core parameters were calculated to complete the quantitative characterization of the macroporous structure of rocks: (1) Coordination number: Defined as the number of throats connected by a single pore, directly reflecting the local connectivity of the pore network. For example... Figure 6 As shown, the statistical results show that the maximum coordination number of the rock sample reached 21, indicating that although the overall average coordination number is low, there are extremely complex hub pores in the rock. These key nodes play a crucial role in the convergence and diversion of fluid transport.

[0055] (2) Euler number: used to characterize the connectivity of pore space per unit volume. For example... Figure 7 As shown, the calculated Euler number is negative. The combination of the low average coordination number and the negative Euler number quantitatively confirms that the pore system of this rock sample is not a simple discrete distribution, but rather exhibits a complex network topology with interconnected paths in three-dimensional space.

[0056] (3) Pore shape factor: used to characterize the degree of irregularity of the pore cross-section, such as Figure 8 As shown, statistics show that the pore shape factor values ​​of most pores are concentrated between 0.035 and 0.055, indicating that the cross-section of the rock pores is mostly an extremely irregular triangle or polygon, and the inner wall surface of the pores is rough. This geometric feature is highly consistent with the high fractal dimension feature extracted using the cumulative fractal theory mentioned above, forming a mutual confirmation between geometry and fractal.

[0057] (4) Centroid path tortuosity: Characterizes the tortuosity of the fluid flow path in a porous medium, and is the ratio of the shortest connecting path length between two points in the pore space to the straight-line distance. Calculations show that the average tortuosity of this rock sample in the main flow direction is 3.28, which is significantly higher than that of conventional sandstone. This indicates that the fluid transport path in the pore network is extremely complex and tortuous, which is a key geometric factor leading to high flow resistance and low permeability in this type of low-permeability reservoir.

[0058] (5) Multidimensional fractal heterogeneity characterization based on CT images: Further revealing the surface roughness and spatial heterogeneity of rock pore structure at the macroscopic scale, and calculating the three-dimensional and two-dimensional fractal dimensions using X-ray CT grayscale images: The box-counting method was used to perform three-dimensional fractal dimension analysis of the pore surface of digital core samples. The calculation formula is shown below: ; in, The required size to cover the pore space is The number of cube boxes, The fractal dimension is 2.45. The calculation results show that the three-dimensional fractal dimension of the macroporous system of this rock sample is 2.45. This value not only quantifies the roughness of the macroscopic pore surface but also serves as an independent physical constraint, verifying the accuracy of the aforementioned cumulative fractal dimension spectrum in the large pore size range. For each layer of two-dimensional slice image from the CT scan, the fractal dimension of its pore boundaries was calculated layer by layer, constructing a fluctuation curve of the fractal dimension changing with the layer sequence. Statistical results are as follows: Figure 9 As shown, the two-dimensional fractal dimension and single-layer porosity exhibit a highly correlated and synchronous fluctuation trend. This layer-by-layer analysis reveals the vertical sedimentary heterogeneity of the rock. High-porosity layers are often accompanied by higher fractal dimensions, indicating that the pore boundary morphology in high-porosity and permeability regions is actually more fragmented and complex.

[0059] As can be seen from the above, the embodiments of this application successfully corrected the pore shielding effect of the high-pressure mercury intrusion porosimetry experiment by utilizing the macropore geometric information extracted by CT scans, and constructed a more realistic full-pore capillary pressure-saturation curve. Secondly, based on the cumulative fractal theory and the iterative truncated singular value decomposition algorithm, the fractal dimension spectrum that continuously varies with the pore throat radius was obtained, realizing a quantitative characterization of the microstructure characteristics of strongly heterogeneous reservoirs. Thirdly, based on this fractal dimension spectrum, a variable exponential nonlinear transformation of the nuclear magnetic resonance T2 spectrum was performed, overcoming the fundamental deviation of the traditional fixed transformation model, and obtaining a full-scale pore size distribution that highly matches the actual situation. Finally, by fusing the micropore size information with the three-dimensional topological network parameters and multi-dimensional fractal geometric features extracted by CT across scales, a comprehensive characterization model covering pore geometry, connectivity, and heterogeneity properties was constructed. The method of this invention realizes continuous quantitative characterization of the entire scale from nanoscale matrix pores to micrometer-scale microcracks, significantly improving the accuracy and reliability of the evaluation of complex reservoir pore structures.

[0060] See Figure 10 As shown in the figure, this application discloses a full-scale characterization device for reservoir pore structure based on cumulative fractals, comprising: Data acquisition module 11 is used to acquire multi-source measurement data of the rock sample to be tested; the multi-source measurement data includes target high-pressure mercury intrusion test data, nuclear magnetic resonance transverse relaxation time T2 spectrum data and CT scan three-dimensional image data; Information extraction module 12 is used to extract pore geometric information of target pores in the rock sample to be tested based on the CT scan three-dimensional image data; the target pores are pores whose size meets the preset large pore determination conditions, and the pore geometric information includes pore radius distribution and topological parameters; Data correction module 13 is used to correct the target large-diameter segment data in the target high-pressure mercury injection experimental data caused by the pore shielding effect using the pore geometry information of the target pore, so as to construct a capillary pressure-saturation curve using the corrected data; The fractal inversion module 14 is used to establish a cumulative fractal equation based on the capillary pressure-saturation curve, and to perform inversion calculation on the cumulative fractal equation using a preset iterative truncated singular value decomposition algorithm to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample under test. The cumulative fractal equation is used to describe the fractal characteristics of the pore structure in the capillary pressure-saturation curve, and its expression is a weighted sum of multiple fractal components. Each fractal component is determined by the fractal dimension and the weight coefficient corresponding to the fractal dimension. The spectrum conversion module 15 is used to establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample to be tested using the fractal dimension spectrum, and to perform a variable exponential conversion on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on the nonlinear mapping relationship to obtain a full-scale pore size distribution curve. The model building module 16 is used to extract the three-dimensional pore topology network structure parameters and multi-dimensional fractal geometric features of the rock sample under test based on the CT scan three-dimensional image data, and to perform cross-scale data fusion of the three-dimensional pore topology network structure parameters, the multi-dimensional fractal geometric features and the full-scale pore size distribution curve, so as to construct a comprehensive characterization model for characterizing the full-scale pore structure of the rock; the multi-dimensional fractal geometric features include a sequence of three-dimensional fractal dimension and two-dimensional fractal dimension.

[0061] In some specific embodiments, the data correction module 13 may specifically include: The data correction unit is used to stitch together the target high-pressure mercury intrusion test data with the pore radius distribution data extracted from the CT scan three-dimensional image data based on a preset pore radius threshold, so as to obtain the corresponding corrected data.

[0062] In some specific embodiments, the fractal inversion module 14 may specifically include: The equation transformation unit is used to transform the cumulative fractal equation into a system of linear equations and construct a sparse kernel matrix corresponding to the system of linear equations; the solution vector of the system of linear equations is composed of the weight coefficients corresponding to each fractal component. The singular value decomposition unit is used to perform singular value decomposition on the sparse kernel matrix to obtain the corresponding singular value decomposition result, and to determine the singular value truncation parameter based on the preset information criterion formula, so as to use the singular value truncation parameter to truncate the singular value decomposition result to obtain the truncation result. The fractal dimension spectrum generation unit is used to generate an initial solution vector based on the truncation result, and to perform iterative optimization and negative value constraint processing based on the initial solution vector to obtain the fractal dimension spectrum.

[0063] In some specific embodiments, the spectrum conversion module 15 may specifically include: The conversion model establishment unit is used to establish a variable exponential conversion model between pore radius and transverse relaxation time based on the fractal dimension values ​​corresponding to different pore intervals in the fractal dimension spectrum; wherein, the conversion exponent of the variable exponential conversion model is related to the fractal dimension value of the corresponding pore interval; The reference coefficient determination unit is used to take the cumulative pore size distribution corresponding to the capillary pressure-saturation curve as the target reference, and determine the surface relaxation strength reference coefficient based on the target reference using a preset optimization algorithm. The spectrum conversion unit is used to convert the nuclear magnetic resonance transverse relaxation time T2 spectrum data into the full-scale aperture distribution curve using the variable exponential conversion model and the surface relaxation intensity reference coefficient.

[0064] In some specific embodiments, the model building module 16 may specifically include: The parameter extraction submodule is used to segment and perform connectivity analysis on the CT scan three-dimensional image to obtain a three-dimensional connected pore space model, and based on the three-dimensional connected pore space model, extract the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample to be tested. Accordingly, the model construction submodule may specifically include: The structural parameter extraction unit is used to identify and construct a pore network model based on the three-dimensional connected pore space model using a preset pore network model extraction algorithm, and to extract the three-dimensional pore topology network structural parameters of the rock sample to be tested based on the pore network model; the pore network model includes pore units and throat units connecting the pore units; the three-dimensional pore topology network structural parameters include coordination number, BiEuler number, pore shape factor, and centroid path tortuosity; The first multidimensional feature extraction unit is used to calculate the three-dimensional fractal dimension of the target pore system of the rock sample under test based on the three-dimensional connected pore space model. The second multidimensional feature extraction unit is used to calculate the two-dimensional fractal dimension layer by layer from the continuous two-dimensional slices of the CT scan three-dimensional image to generate the two-dimensional fractal dimension sequence.

[0065] Furthermore, embodiments of this application also disclose an electronic device, Figure 11 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the full-scale characterization method of reservoir pore structure based on cumulative fractals disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0066] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0067] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0068] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the full-scale characterization method of reservoir pore structure based on cumulative fractals disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0069] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned disclosed method for full-scale characterization of reservoir pore structure based on cumulative fractals. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0070] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0071] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0072] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0073] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0074] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for full-scale characterization of reservoir pore structure based on cumulative fractals, characterized in that, include: Acquire multi-source measurement data of the rock sample to be tested; the multi-source measurement data includes target high-pressure mercury intrusion test data, nuclear magnetic resonance transverse relaxation time T2 spectrum data, and CT scan three-dimensional image data; Based on the CT scan three-dimensional image data, the pore geometry information of the target pores in the rock sample to be tested is extracted; The target pore is a pore whose size meets the preset large pore determination conditions, and the pore geometric information includes pore radius distribution and topological parameters; The pore geometry information of the target pores is used to correct the target large-pore section data in the high-pressure mercury injection experiment data caused by the pore shielding effect, so as to construct the capillary pressure-saturation curve using the corrected data; Based on the capillary pressure-saturation curve, a cumulative fractal equation is established, and the cumulative fractal equation is inverted and solved using a preset iterative truncation singular value decomposition algorithm to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested. The fractal dimension spectrum is used to establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample to be tested, and the nuclear magnetic resonance transverse relaxation time T2 spectrum data is subjected to a variable exponential transformation based on the nonlinear mapping relationship to obtain the full-scale pore size distribution curve. Based on the CT scan 3D image data, the 3D pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test are extracted. The 3D pore topology network structure parameters, the multidimensional fractal geometric features, and the full-scale pore size distribution curve are then fused across scales to construct a comprehensive characterization model for representing the full-scale pore structure of the rock. The multidimensional fractal geometric features include a sequence of 3D fractal dimension and 2D fractal dimension.

2. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 1, characterized in that, The cumulative fractal equation is used to describe the fractal characteristics of the pore structure in the capillary pressure-saturation curve. Its expression is a weighted sum of multiple fractal components. Each fractal component is determined by the fractal dimension and the weighting coefficient corresponding to the fractal dimension.

3. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 2, characterized in that, The step of using a preset iterative truncated singular value decomposition algorithm to invert and solve the cumulative fractal equation to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested includes: The cumulative fractal equation is transformed into a system of linear equations, and a sparse kernel matrix corresponding to the system of linear equations is constructed; the solution vector of the system of linear equations is composed of the weight coefficients corresponding to each fractal component. Singular value decomposition is performed on the sparse kernel matrix to obtain the corresponding singular value decomposition result. A singular value truncation parameter is determined based on a preset information criterion formula so that the singular value decomposition result can be truncated using the singular value truncation parameter to obtain the truncated result. An initial solution vector is generated based on the truncation result, and iterative optimization and negative value constraint processing are performed based on the initial solution vector to obtain the fractal dimension spectrum.

4. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 1, characterized in that, The process involves establishing a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time (T2) spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum, and performing a variable exponential transformation on the T2 spectrum data based on the nonlinear mapping relationship to obtain a full-scale pore size distribution curve, including: A variable exponential transformation model between pore radius and transverse relaxation time is established based on the fractal dimension values ​​corresponding to different pore intervals in the fractal dimension spectrum; wherein, the transformation exponent of the variable exponential transformation model is related to the fractal dimension value of the corresponding pore interval. The cumulative pore size distribution corresponding to the capillary pressure-saturation curve is used as the target benchmark, and the surface relaxation strength benchmark coefficient is determined based on the target benchmark using a preset optimization algorithm. The nuclear magnetic resonance transverse relaxation time T2 spectrum data is converted into the full-scale aperture distribution curve using the variable exponential conversion model and the surface relaxation intensity reference coefficient.

5. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 1, characterized in that, The extraction of three-dimensional pore topology network structural parameters and multidimensional fractal geometric features of the rock sample based on the CT scan three-dimensional image data includes: The CT scan three-dimensional image is segmented and connectivity analysis is performed to obtain a three-dimensional connected pore space model. Based on the three-dimensional connected pore space model, the three-dimensional pore topology network structure parameters and multidimensional fractal geometric features of the rock sample under test are extracted.

6. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 5, characterized in that, The extraction of three-dimensional pore topology network parameters and multidimensional fractal geometric features from the rock sample under test, based on the three-dimensional connected pore space model, includes: Based on the three-dimensional connected pore space model, a pore network model is identified and constructed using a preset pore network model extraction algorithm, and the three-dimensional pore topology network structure parameters of the rock sample to be tested are extracted based on the pore network model; the pore network model includes pore units and throat units connecting the pore units; the three-dimensional pore topology network structure parameters include coordination number, BiEuler number, pore shape factor, and centroid path tortuosity; The three-dimensional fractal dimension of the target pore system of the rock sample under test is calculated based on the three-dimensional connected pore space model. The two-dimensional fractal dimension is calculated layer by layer from consecutive two-dimensional slices of the CT scan three-dimensional image to generate the two-dimensional fractal dimension sequence.

7. The method for full-scale characterization of reservoir pore structure based on cumulative fractals according to claim 1, characterized in that, The step of correcting the target large-diameter segment data in the target high-pressure mercury intrusion test data caused by the pore shielding effect using the pore geometry information of the target pore includes: Based on a preset pore radius threshold, the target high-pressure mercury intrusion test data is stitched together with the pore radius distribution data extracted from the CT scan three-dimensional image data to obtain the corresponding corrected data.

8. A full-scale characterization device for reservoir pore structure based on cumulative fractals, characterized in that, include: The data acquisition module is used to acquire multi-source measurement data of the rock sample to be tested; the multi-source measurement data includes target high-pressure mercury intrusion test data, nuclear magnetic resonance transverse relaxation time T2 spectrum data, and CT scan three-dimensional image data; The information extraction module is used to extract the pore geometry information of the target pores in the rock sample to be tested based on the CT scan three-dimensional image data; The target pore is a pore whose size meets the preset large pore determination conditions, and the pore geometric information includes pore radius distribution and topological parameters; The data correction module is used to correct the target large-diameter segment data in the target high-pressure mercury injection experimental data caused by the pore shielding effect using the pore geometry information of the target pore, so as to construct the capillary pressure-saturation curve using the corrected data; The fractal inversion module is used to establish a cumulative fractal equation based on the capillary pressure-saturation curve, and to perform inversion solution on the cumulative fractal equation using a preset iterative truncation singular value decomposition algorithm to obtain the fractal dimension spectrum related to the pore throat radius of the rock sample to be tested. The spectrum conversion module is used to establish a nonlinear mapping relationship between the nuclear magnetic resonance transverse relaxation time T2 spectrum data and the pore radius of the rock sample under test using the fractal dimension spectrum, and to perform a variable exponential conversion on the nuclear magnetic resonance transverse relaxation time T2 spectrum data based on the nonlinear mapping relationship to obtain the full-scale pore size distribution curve. The model building module is used to extract the three-dimensional pore topology network structure parameters and multi-dimensional fractal geometric features of the rock sample under test based on the CT scan three-dimensional image data, and to perform cross-scale data fusion of the three-dimensional pore topology network structure parameters, the multi-dimensional fractal geometric features and the full-scale pore size distribution curve, so as to construct a comprehensive characterization model for characterizing the full-scale pore structure of the rock; the multi-dimensional fractal geometric features include a sequence of three-dimensional fractal dimension and two-dimensional fractal dimension.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the full-scale characterization method of reservoir pore structure based on cumulative fractals as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the full-scale characterization method of reservoir pore structure based on cumulative fractals as described in any one of claims 1 to 7.

Images