A method, apparatus and device for modeling a multi-mineral rock structure

Abstract

The application discloses a multi-mineral rock structure modeling method, device and equipment, relates to the field of rock mechanics, and can model large-scale rock materials with multiple minerals. The method comprises the following steps: detecting the mineral composition of a target rock to obtain the component proportion of multiple mineral components contained in the target rock, determining the anisotropy representation parameter corresponding to the target rock, and the target fractal dimension and target hourglass parameter corresponding to the multiple mineral components; constructing a rock model of the target rock by adopting a bubbling method according to the anisotropy representation parameter and the component proportion; calculating the specified fractal dimension and specified hourglass parameter corresponding to the multiple mineral components in the rock model, and calculating the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter; and obtaining the target model when the first difference hits a first preset range and the second difference hits a second preset range.

Description

Technical Field

[0001] This application relates to the field of rock mechanics, and in particular to a method, apparatus and equipment for modeling multi-mineral rock structures. Background Technology

[0002] The detailed characterization and reconstruction of rock structure is an important prerequisite for studying rock mechanics and its hydraulic properties. Among them, the microscopic pore structure of rock core is an important factor affecting the macroscopic properties of reservoirs and the exploitation of oil and gas resources.

[0003] In related technologies, to construct rock models, DIP (Digital Image Processing) can be used. This involves using digital photographs of fresh rock sections, combined with image segmentation algorithms, to obtain digital images of different mineral compositions, and then constructing a model based on these images. Alternatively, X-ray CT (Computed Tomography) imaging technology can be used to obtain information about the internal structure of the rock material, and then the model can be constructed based on this information.

[0004] In the process of developing this application, the applicant discovered that the relevant technology has at least the following problems:

[0005] The modeling methods of related technologies are mainly aimed at sedimentary rocks such as sandstone. However, due to the fine structure but small scale of sedimentary rocks such as sandstone, they are usually not suitable for the study of rock strength characteristics. Furthermore, DIP technology is limited to a two-dimensional plane and cannot see the internal information of the rock. In practice, X-ray CT is difficult to distinguish different minerals and often requires processing a large number of CT images. The operation is complicated and cumbersome, and it can only generate models one-to-one based on rock samples, making it impossible to build a large number of similar rock models. Therefore, the industry urgently needs a modeling method for rock materials with large scale and many minerals. Summary of the Invention

[0006] In view of this, this application provides a method, apparatus and equipment for modeling multi-mineral rock structures, the main purpose of which is to solve the current problem of the urgent need for a modeling method for large-scale rock materials with many minerals.

[0007] According to the first aspect of this application, a method for modeling the structure of multi-mineral rocks is provided, the method comprising:

[0008] The target rock is subjected to mineral composition detection to obtain the composition ratio of various mineral components contained in the target rock. Based on the composition ratio and the two-dimensional image corresponding to the target rock, the anisotropic characterization parameters corresponding to the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components are determined.

[0009] According to the anisotropy characterization parameters and the component ratio, the initial control parameters of the bubbling method are set, and the bubbling method is used to construct a rock model of the target rock.

[0010] Calculate the specified fractal dimension and specified hourglass parameter corresponding to the various mineral components in the rock model, and calculate the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter, respectively;

[0011] When the first difference hits the first preset range and the second difference hits the second preset range, the target model is obtained.

[0012] Optionally, determining the anisotropic characterization parameters corresponding to the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components, based on the component ratio and the two-dimensional image corresponding to the target rock, includes:

[0013] According to the stated component ratio, add component labels to each pixel in the two-dimensional image;

[0014] Image recognition is performed on the two-dimensional image with added component labels, and the anisotropy characterization parameters are obtained by calculating the ratio of the major and minor axes of the mineral particles corresponding to each mineral component among the various mineral components.

[0015] Image recognition is performed on the two-dimensional image with added component labels to determine the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components, resulting in multiple fractal dimensions and multiple hourglass parameters. The mean of the multiple fractal dimensions is calculated as the target fractal dimension, and the mean of the multiple hourglass parameters is calculated as the target hourglass parameter.

[0016] Optionally, determining the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components includes:

[0017] The fractal dimension of each mineral component in the plurality of mineral components is calculated using a preset fractal dimension calculation method, thereby obtaining multiple fractal dimensions corresponding to the plurality of mineral components. The fractal dimension is used to describe the shape of the mineral particles corresponding to the mineral component.

[0018] According to the image recognition method, the composition distribution of the multiple mineral components is obtained, and according to the composition distribution, multiple mineral pixels corresponding to any mineral component are identified. The nearest distance between each mineral pixel and other mineral pixels is assigned to the mineral pixel, where the other mineral pixels are the mineral pixels corresponding to mineral components other than the mineral component in the multiple mineral components.

[0019] The values ​​of the multiple mineral pixels are summed to obtain a first parameter, and the multiple mineral pixels are aggregated to obtain an aggregated graphic. The nearest distance of each mineral element to the edge of the aggregated graphic is assigned to the corresponding mineral element to determine a second parameter. The ratio of the first parameter and the second parameter is used as the hourglass parameter of the mineral composition. The hourglass parameter is used to describe the mineral particle size of the mineral composition.

[0020] Calculate the hourglass parameters corresponding to each mineral component to obtain multiple hourglass parameters.

[0021] Optionally, constructing the rock model of the target rock using the bubbling method includes:

[0022] The mineral components are ranked in descending order of their proportions, and the distribution areas corresponding to each mineral component are selected one by one according to the ranking order to obtain the rock model.

[0023] The step of selecting the distribution area corresponding to each of the mineral components to obtain the rock model includes:

[0024] If a two-dimensional rock model is constructed, then for any mineral component, according to the initial control parameters, a preset number of circular or elliptical bubbles are generated upward or downward within a preset plane range, and an elevation value is assigned to the points contained in each of the preset number of circular or elliptical bubbles.

[0025] The upward-generated circular or elliptical bubbles have positive elevation values;

[0026] The downward-generated circular or elliptical bubbles have negative elevation values.

[0027] For each point within the preset plane range, determine all the elevation values ​​assigned to the point, and obtain the target elevation value associated with the point by adding up all the elevation values ​​corresponding to the point, so as to achieve the overlap of the multiple bubbles within the preset plane range;

[0028] Within the preset plane range, delete the point selected by the previous mineral component, rank the target elevation values ​​of the remaining points in descending order, and determine the optional quantity value corresponding to the mineral component based on the component ratio corresponding to the mineral component.

[0029] According to the ranking order of the target elevation values, multiple target points are selected one by one, and the number of the multiple target points is consistent with the selectable quantity value;

[0030] Selected target points are used as the distribution areas of the mineral components;

[0031] The distribution area corresponding to each of the multiple mineral components is determined to obtain a two-dimensional rock model of the target rock.

[0032] Optionally, the rock model is obtained by selecting the distribution area corresponding to each of the mineral components one by one, and further includes:

[0033] If a three-dimensional rock model is constructed, for any mineral component, according to the initial control parameters, a preset number of spherical or ellipsoidal bubbles are generated in a preset space, and positive or negative attribute values ​​are assigned to the points contained in each of the preset number of spherical or ellipsoidal bubbles.

[0034] For each point within the preset space, determine all attribute values ​​assigned to the point, and obtain the target attribute value associated with the point by adding all the attribute values ​​corresponding to the point, so as to achieve the overlap of the multiple bubbles within the preset plane range;

[0035] Within the preset space, delete the point selected by the previous mineral component, and rank the target attribute values ​​of the remaining points in descending order. Determine the optional quantity value corresponding to the mineral component based on the component ratio corresponding to the mineral component.

[0036] According to the ranking order of the target attribute values, multiple target points are selected one by one, and the number of the multiple target points is consistent with the selectable quantity value;

[0037] Selected target points are used as the distribution areas of the mineral components;

[0038] The distribution area corresponding to each of the multiple mineral components is determined to obtain a three-dimensional rock model of the target rock.

[0039] Optionally, calculating the specified fractal dimension and specified hourglass parameters corresponding to the various mineral components in the rock model includes:

[0040] If the rock model is a two-dimensional model, then based on the distribution area corresponding to each of the mineral components, the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components are determined, resulting in multiple fractal dimensions and multiple hourglass parameters. The mean of the multiple fractal dimensions is calculated as the specified fractal dimension, and the mean of the multiple hourglass parameters is calculated as the specified hourglass parameter.

[0041] If the rock model is a three-dimensional model, then multiple specified cross sections are cut at equal intervals in the rock model. Based on the distribution area of ​​each mineral component associated with each specified cross section, multiple fractal dimensions and multiple hourglass parameters corresponding to the multiple specified cross sections are calculated. The mean of the multiple fractal dimensions is taken as the specified fractal dimension, and the mean of the multiple hourglass parameters is taken as the specified hourglass parameter.

[0042] Optionally, after calculating the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter, the method further includes:

[0043] When the first difference does not hit the first preset range or the second difference does not hit the second preset range, the initial control parameters are modified to obtain updated control parameters;

[0044] According to the updated control parameters, the rock model of the target rock is constructed using the bubble method, and the specified fractal dimension and specified hourglass parameter corresponding to the rock model are calculated again, the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter;

[0045] When the first difference does not hit the first preset range or the second difference does not hit the second preset range, the update control parameters are modified, and the rock model is reconstructed according to the modified update control parameters until the first difference of the reconstructed rock model hits the first preset range and the corresponding second difference hits the second preset range.

[0046] Optionally, when the first difference does not fall within the first preset range or the second difference does not fall within the second preset range, modifying the initial control parameters to obtain updated control parameters includes:

[0047] When the first difference does not hit the first preset range, a first specified range and a second specified range associated with the first preset range are determined. If the first difference hits the first specified range, the standard deviation of the bubble radius distribution in the initial control parameters is reduced to obtain the updated control parameters. If the first difference hits the second specified range, the standard deviation of the bubble radius distribution in the initial control parameters is increased to obtain the updated control parameters.

[0048] When the second difference does not match the second preset range, a third specified range and a fourth specified range associated with the second preset range are determined. If the second difference matches the third specified range, the generated mesh size in the initial control parameters is reduced to obtain the updated control parameters. If the second difference matches the fourth specified range, the generated mesh size in the initial control parameters is increased to obtain the updated control parameters.

[0049] According to a second aspect of this application, a multi-mineral rock structure modeling apparatus is provided, the apparatus comprising:

[0050] The determination module is used to detect the mineral composition of the target rock, obtain the composition ratio of the various mineral components contained in the target rock, and determine the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components based on the composition ratio and the two-dimensional image corresponding to the target rock.

[0051] The setting module is used to set the initial control parameters of the bubbling method according to the anisotropy characterization parameters and the component ratio, and to construct a rock model of the target rock using the bubbling method.

[0052] The calculation module is used to calculate the specified fractal dimension and specified hourglass parameter corresponding to the various mineral components in the rock model, and compare the specified fractal dimension and specified hourglass parameter with the target fractal dimension and target hourglass parameter respectively;

[0053] The generation module is used to obtain the target model when the comparison result indicates that the specified fractal dimension is consistent with the target fractal dimension and the specified hourglass parameters are consistent with or differ from the target hourglass parameters by less than a set threshold.

[0054] Optionally, the determining module is configured to: add component labels to each pixel in the two-dimensional image according to the component proportions; perform image recognition on the two-dimensional image with added component labels; obtain the anisotropy characterization parameters by calculating the major and minor axis ratios of the mineral particles corresponding to each of the multiple mineral components; perform image recognition on the two-dimensional image with added component labels; determine the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components; obtain multiple fractal dimensions and multiple hourglass parameters; calculate the mean of the multiple fractal dimensions as the target fractal dimension; and calculate the mean of the multiple hourglass parameters as the target hourglass parameter.

[0055] Optionally, the determining module is configured to calculate the fractal dimension corresponding to each of the multiple mineral components using a preset fractal dimension calculation method, thereby obtaining multiple fractal dimensions corresponding to the multiple mineral components, wherein the fractal dimension is used to describe the shape of the mineral particles corresponding to the mineral components; according to the image recognition method, obtain the component distribution of the multiple mineral components, and according to the component distribution, identify multiple mineral pixels corresponding to any mineral component, assign the nearest distance of each mineral pixel to other mineral pixels to the mineral pixel, wherein the other mineral pixels are the mineral pixels corresponding to mineral components other than the mineral component in the multiple mineral components; sum the assigned values ​​of the multiple mineral pixels to obtain a first parameter, and aggregate the multiple mineral pixels to obtain an aggregated graphic, assign the nearest distance of each mineral element to the edge of the aggregated graphic to the corresponding mineral element, determine a second parameter, and use the ratio of the first parameter and the second parameter as the hourglass parameter of the mineral component, wherein the hourglass parameter is used to describe the size of the mineral particles of the mineral component; calculate the hourglass parameter corresponding to each mineral component to obtain multiple hourglass parameters.

[0056] Optionally, the setting module is used to rank the various mineral components in descending order of their proportions, and select the distribution area corresponding to each mineral component one by one according to the ranking order to obtain the rock model; wherein, the step of selecting the distribution area corresponding to each mineral component one by one to obtain the rock model includes: if a two-dimensional rock model is constructed, then for any mineral component, according to the initial control parameters, a preset number of circular or elliptical bubbles are generated upward or downward within a preset plane range, and an elevation value is assigned to the points contained in each of the preset number of circular or elliptical bubbles; the upward-generated circular or elliptical bubbles have positive elevation values; the downward-generated circular or elliptical bubbles have negative elevation values; for the preset plane... For each point within the plane range, all assigned elevation values ​​are determined. By summing all the elevation values ​​corresponding to the point, the target elevation value associated with the point is obtained, thereby enabling the multiple bubbles to overlap within the preset plane range. Within the preset plane range, the point with the highest selected mineral component is deleted, and the target elevation values ​​of the remaining points are ranked in descending order. Based on the component ratio corresponding to the mineral component, the optional quantity value corresponding to the mineral component is determined. According to the ranking order of the target elevation values, multiple target points are selected one by one, and the number of the multiple target points is consistent with the optional quantity value. The selected multiple target points are used as the distribution area of ​​the mineral component. The distribution area corresponding to each of the multiple mineral components is determined to obtain the two-dimensional rock model of the target rock.

[0057] Optionally, the setting module is further configured to, if constructing a three-dimensional rock model, generate a preset number of spherical or ellipsoidal bubbles within a preset space for any mineral component according to the initial control parameters; assign positive or negative attribute values ​​to the points contained in each of the preset number of spherical or ellipsoidal bubbles; for each point within the preset space, determine all attribute values ​​assigned to the point; obtain the target attribute value associated with the point by adding all attribute values ​​corresponding to the point, so as to achieve overlap of the multiple bubbles within the preset plane; delete the point selected by the previous mineral component within the preset space; rank the target attribute values ​​of the remaining points in descending order; determine the optional quantity value corresponding to the mineral component based on the component ratio corresponding to the mineral component; select multiple target points one by one according to the ranking order of the target attribute values, the number of the multiple target points being consistent with the optional quantity value; use the selected multiple target points as the distribution area of ​​the mineral component; determine the distribution area corresponding to each of the multiple mineral components to obtain the three-dimensional rock model of the target rock.

[0058] Optionally, the calculation module is configured to: if the rock model is a two-dimensional model, determine the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components based on the distribution area corresponding to each of the mineral components, obtain multiple fractal dimensions and multiple hourglass parameters, and calculate the mean of the multiple fractal dimensions as the specified fractal dimension and the mean of the multiple hourglass parameters as the specified hourglass parameter; if the rock model is a three-dimensional model, extract multiple specified cross sections at equal intervals from the rock model, calculate multiple fractal dimensions and multiple hourglass parameters corresponding to the multiple specified cross sections based on the distribution area of ​​each mineral component associated with each specified cross section, and use the mean of the multiple fractal dimensions as the specified fractal dimension and the mean of the multiple hourglass parameters as the specified hourglass parameter.

[0059] Optionally, the device further includes:

[0060] The modification module is used to modify the initial control parameters to obtain updated control parameters when the first difference does not hit the first preset range or the second difference does not hit the second preset range.

[0061] The comparison module is used to construct a rock model of the target rock using the bubble method according to the updated control parameters, and to recalculate the specified fractal dimension and specified hourglass parameter corresponding to the rock model, the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter;

[0062] The modification module is further configured to modify the update control parameters when the first difference does not hit the first preset range or the second difference does not hit the second preset range, and to reconstruct the rock model according to the modified update control parameters until the first difference corresponding to the reconstructed rock model hits the first preset range and the corresponding second difference hits the second preset range.

[0063] Optionally, the modification module is configured to: when the first difference does not fall within the first preset range, determine a first specified range and a second specified range associated with the first preset range; if the first difference falls within the first specified range, decrease the standard deviation of the bubble radius distribution in the initial control parameters to obtain the updated control parameters; if the first difference falls within the second specified range, increase the standard deviation of the bubble radius distribution in the initial control parameters to obtain the updated control parameters; when the second difference does not fall within the second preset range, determine a third specified range and a fourth specified range associated with the second preset range; if the second difference falls within the third specified range, decrease the generated mesh size in the initial control parameters to obtain the updated control parameters; if the second difference falls within the fourth specified range, increase the generated mesh size in the initial control parameters to obtain the updated control parameters.

[0064] According to a third aspect of this application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the first aspects above.

[0065] According to a fourth aspect of this application, a computer-readable storage medium is provided, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0066] Using the above technical solution, this application provides a method, apparatus, device, and storage medium for modeling multi-mineral rock structures. First, the mineral composition of the target rock is detected to obtain the proportions of various mineral components contained in the target rock. Based on the proportions and the corresponding two-dimensional image of the target rock, the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components, are determined. Then, according to the anisotropic characterization parameters and component proportions, the initial control parameters for the bubble bubbling method are set, and a rock model of the target rock is constructed using the bubble bubbling method. Next, the specified fractal dimension and specified hourglass parameters corresponding to the various mineral components in the rock model are calculated. The first difference between the specified fractal dimension and the target fractal dimension, and the second difference between the specified hourglass parameters and the target hourglass parameters are calculated respectively. When the first difference hits a first preset range and the second difference hits a second preset range, the target model is obtained. This invention obtains anisotropic characterization parameters of target rocks through laboratory mineral composition detection and numerical image processing of target rock surface images. Utilizing the correlation between these anisotropic characterization parameters and relevant parameters from the bubble bubbling method, a rock model of the target rock is constructed using the bubble bubbling method. Compared to other current methods, this invention is simpler and more efficient. It only requires image processing and analysis of the rock sample surface to batch construct three-dimensional rock models of various sizes, significantly reducing the excessive acquisition and processing of rock sample data, lowering costs while improving efficiency. Furthermore, it is completely unrestricted by size; theoretically, arbitrarily large-scale three-dimensional rock structure models can be built as long as computational power allows. In addition, by changing spherical bubbles to ellipsoidal bubbles, rock models with anisotropic mineral particle distribution characteristics can be generated. Whether the mineral composition has multiple irregular distributions or the pore structure is complex, two-dimensional or three-dimensional models can be built using this method, demonstrating great potential in physical experiments and numerical simulations of rock mechanics.

[0067] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0068] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0069] Figure 1 This paper illustrates a flowchart of a multi-mineral rock structure modeling method provided in an embodiment of this application.

[0070] Figure 2A This paper illustrates a flowchart of a multi-mineral rock structure modeling method provided in an embodiment of this application.

[0071] Figure 2B This illustration shows an image recognition diagram of a multi-mineral rock structure modeling method provided in an embodiment of this application;

[0072] Figure 2C A schematic diagram of parameters for a multi-mineral rock structure modeling method provided in an embodiment of this application is shown;

[0073] Figure 2D This illustration shows a schematic diagram of mineral distribution in a multi-mineral rock structure modeling method provided in an embodiment of this application;

[0074] Figure 2E This illustration shows a single mineral numerical image and an initial model cross-section schematic diagram of a multi-mineral rock structure modeling method provided in an embodiment of this application.

[0075] Figure 2F This illustration shows a schematic diagram of mineral distribution in a multi-mineral rock structure modeling method provided in an embodiment of this application;

[0076] Figure 2G This paper illustrates a flowchart of a multi-mineral rock structure modeling method provided in an embodiment of this application.

[0077] Figure 3A This paper shows a schematic diagram of a multi-mineral rock structure modeling device provided in an embodiment of this application.

[0078] Figure 3B This paper shows a schematic diagram of a multi-mineral rock structure modeling device provided in an embodiment of this application.

[0079] Figure 4 A schematic diagram of the device structure of a computer device provided in an embodiment of this application is shown. Detailed Implementation

[0080] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0081] The detailed characterization and reconstruction of rock structure is an important prerequisite for studying rock mechanics and its hydraulic properties. Among them, the microscopic pore structure of rock core is an important factor affecting the macroscopic properties of reservoirs and the exploitation of oil and gas resources.

[0082] In rock modeling, based on the heterogeneity of rocks, numerical models often assume a non-uniformity for the rock material and directly assign values ​​to relevant parameters, such as the Voronoi mosaic technique. While this produces a relatively uniform grain size distribution, it clearly does not accurately reflect the internal structure of the rock. Therefore, more sophisticated detection methods are needed to establish the non-uniform structure of rocks. Digital Image Processing (DIP) can be used, utilizing digital photographs of fresh rock sections combined with image segmentation algorithms to obtain digital images of different mineral components, which are then used to construct the model. Alternatively, X-ray computed tomography (CT) imaging can be employed to obtain information about the internal structure of the rock material, which is then used to construct the model. While DIP, using digital photographs of fresh rock sections combined with image segmentation algorithms, can obtain digital images of different mineral components, it is limited to a two-dimensional plane and struggles to achieve three-dimensional modeling. CT technology can obtain information about the internal structure of rock materials and is adept at identifying rock pore structures or cracks; however, CT research is generally small-scale, typically less than a few centimeters, and detailed differentiation of mineral components is difficult to achieve, making it challenging to establish multi-mineral rock models. In addition, rock microscopes can be used to detect rocks at the micrometer scale, providing more detailed information about rock structure and minerals. However, due to their smaller scale, they are not of research or engineering significance.

[0083] Regarding the specific steps of 3D modeling, building a 3D rock model based on CT images involves placing a series of adjacent rock slices in space to establish the spatial relationships of pores or rock materials. This requires calculating a large number of images, resulting in low efficiency and poor economics. Following the same principle, DIP (Digital In-Place) technology is used to build a 3D cuboid microstructure of granite samples based on a series of surface images, assuming that the rock surface represents internal heterogeneity and has a relatively shallow depth. This method requires an iterative milling scanning system, causing irreversible damage to the rock sample. Furthermore, both methods produce models that are one-to-one replicas of the rock sample, making it impossible to generate a large number of similar models. A single, isolated model has limited ability to describe the macroscopic differences in random rock structures. Unfortunately, due to environmental limitations or high costs, only one or a very limited number of samples can generally be obtained, making it difficult to meet the needs of studying the macroscopic mechanical responses of complex, random, heterogeneous mesostructures.

[0084] Therefore, various modeling methods based on statistical parameter strategies, stochastic theory, or related algorithms have enabled the batch creation of 3D rock models. However, to date, most 3D modeling methods for porous rock materials only include the two-phase medium of rock and pores, and are mainly aimed at sedimentary rocks such as sandstone. Due to their fine structure but small scale, they are generally unsuitable for studying rock strength characteristics. Therefore, a reconstruction method is still lacking for large-scale, mineral-rich rock materials. Based on Markov random field theory, a method for generating similar images of rock materials with different heterogeneous characteristics has been proposed. However, the technical approach based on two-dimensional image analysis inevitably faces difficulties in 3D rock model reconstruction. To establish multi-scale, multi-component rock models, it is usually essential to scan rock samples using microcomponent computed tomography at different resolutions. This is far from the goal of rapidly generating rock samples with the same or similar microstructure or composition, and is not a true "model reconstruction" in the practical sense. Therefore, this application provides a method for modeling multi-mineral rock structures. First, the mineral composition of the target rock is detected to obtain the proportions of various mineral components contained within the target rock. Based on the component proportions and the corresponding two-dimensional image of the target rock, the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components, are determined. Then, according to the anisotropic characterization parameters and component proportions, the initial control parameters of the bubble bubbling method are set, and a rock model of the target rock is constructed using the bubble bubbling method. Next, the specified fractal dimension and specified hourglass parameters corresponding to the various mineral components in the rock model are calculated, and the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameters and the target hourglass parameters are calculated respectively. When the first difference hits a first preset range and the second difference hits a second preset range, the target model is obtained. This invention obtains the anisotropic characterization parameters of the target rock through laboratory mineral composition detection and numerical image processing of the target rock surface image. Utilizing the correlation between the anisotropic characterization parameters and the relevant parameters of the bubble bubbling method, a rock model of the target rock is constructed using the bubble bubbling method. Compared to other current methods, the method of this invention is simpler and more efficient. It only requires image processing and analysis of the rock sample surface to batch construct three-dimensional rock models of various sizes, greatly reducing the excessive acquisition and processing of rock sample data, thus lowering costs and improving efficiency. Furthermore, it is completely unrestricted by size; theoretically, any size three-dimensional rock structure model can be built as long as computing power allows. In addition, by changing spherical bubbles to ellipsoidal bubbles, rock models with anisotropic mineral particle distribution characteristics can be generated. Whether it has multiple irregularly distributed mineral components or pore structures, two-dimensional or three-dimensional models can be built using this method, demonstrating great potential in physical experiments and numerical simulations of rock mechanics.

[0085] This application provides a method for modeling multi-mineral rock structures, such as... Figure 1 As shown, the method includes:

[0086] 101. Perform mineral composition analysis on the target rock to obtain the composition ratio of various mineral components contained in the target rock. Based on the composition ratio and the corresponding two-dimensional image of the target rock, determine the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to various mineral components.

[0087] In this embodiment, mineral composition detection is performed by relevant technicians in a laboratory. By detecting the mineral composition of the target rock, the various mineral components contained in the target rock and the proportion of each mineral component can be identified. Further, by photographing the surface of the target rock, a two-dimensional image of the target rock can be obtained. Based on the proportion of each mineral component, a component label is added to each pixel in the two-dimensional image. Based on image recognition technology, the anisotropic characterization parameters corresponding to the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to various mineral components, are determined. In actual operation, structural scanning technology can also be used to scan the target rock, capture a two-dimensional image of the target rock, and then determine the anisotropic characterization parameters corresponding to the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to various mineral components based on the two-dimensional image. The fractal dimension describes the shape of the mineral particles, and the hourglass parameter describes the size of the mineral particles, i.e., the distribution concentration. The larger the hourglass parameter, the more concentrated the mineral distribution, and vice versa.

[0088] 102. Set the initial control parameters for the bubbling method according to the anisotropy characterization parameters and composition ratio, and use the bubbling method to construct a rock model of the target rock.

[0089] The bubbling method involves generating a series of hemispherical or spherical bubbles within a certain plane or space. These bubbles randomly rise or fall, exhibiting typical heterogeneous characteristics. When enough bubbles overlap, a rough rock surface is obtained, which is the rock model of the target rock. It should be noted that the bubbling method is a method for reconstructing joint surfaces in rough rocks (and can also be extended to 3D). The radii of the bubbles generated by the bubbling method follow a normal distribution, and positive or negative attribute values ​​are randomly assigned within the spherical bubbles, with the size varying with the radius. The control parameters of the bubbling method include the total number of bubbles N; the normal distribution of bubble radii μ and σ (if the radius produces a negative value, its absolute value is used as the bubble radius); the elevation (attribute value) scaling factor δH; anisotropy can be achieved by using ellipsoidal (hemispherical) bubbles and setting the scale factors δX, δY, and δZ in the x, y, and z axes; the spatial proportion ω of various minerals; the generated model size S; and the total number of meshes NT. The control parameters are assigned values ​​according to the anisotropic characterization parameters corresponding to the target rock, namely the scale coefficient and the proportion of each mineral component, to obtain the initial control parameters.

[0090] 103. Calculate the specified fractal dimension and specified hourglass parameter corresponding to multiple mineral components in the rock model, and calculate the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter, respectively.

[0091] In this embodiment, the specified fractal dimension and specified hourglass parameters corresponding to various mineral components in the rock model are first calculated. Based on the specified fractal dimension and specified hourglass parameters, a first difference between the specified fractal dimension and the target fractal dimension and a second difference between the specified hourglass parameters and the target hourglass parameters are calculated. Then, by comparing the first difference with a first preset range and the second difference with a second preset range, it is determined whether the constructed rock model meets the requirements.

[0092] 104. When the first difference hits the first preset range and the second difference hits the second preset range, the target model is obtained.

[0093] In this embodiment of the application, when the first difference hits the first preset range and the second difference hits the second preset range, it indicates that the constructed rock model meets the requirements, and the constructed model can be used as the target model for relevant personnel to perform subsequent operation steps.

[0094] The method provided in this application obtains anisotropic characterization parameters of the target rock through laboratory mineral composition detection and numerical image processing of images of the target rock surface. Utilizing the correlation between these anisotropic characterization parameters and relevant parameters from the bubble-diffusing method, a rock model of the target rock is constructed using the bubble-diffusing method. Compared to other current methods, this invention is simpler and more efficient. It only requires image processing and analysis of the rock sample surface to batch construct three-dimensional rock models of various sizes, significantly reducing the excessive acquisition and processing of rock sample data, lowering costs while improving efficiency. Furthermore, it is completely unrestricted by size; theoretically, arbitrarily large-scale three-dimensional rock structure models can be established as long as computing power allows. In addition, by changing spherical bubbles to ellipsoidal bubbles, rock models with anisotropic mineral particle distribution characteristics can be generated. Whether the mineral composition has multiple irregular distributions or the pore structure is varied, two-dimensional or three-dimensional models can be established using this method, demonstrating great potential in physical experiments and numerical simulations of rock mechanics.

[0095] This application provides a method for modeling multi-mineral rock structures, such as... Figure 2A As shown, the method includes:

[0096] 201. Perform mineral composition analysis on the target rock to obtain the composition ratio of various mineral components contained in the target rock. Based on the composition ratio and the corresponding two-dimensional image of the target rock, determine the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to various mineral components.

[0097] The following explanation uses a cubic granite rock with a target size of 50mm×50mm×100mm as an example.

[0098] In this step, mineral distribution characteristic parameters corresponding to the cubic granite are determined using mineral composition detection and image recognition technologies. These parameters include the compositional proportions of various mineral components, anisotropy characterization parameters, target fractal dimensions for each mineral component, and target hourglass parameters. The specific process for obtaining the mineral distribution characteristic parameters of the cubic granite is as follows:

[0099] First, the composition and proportions of the granite were determined through laboratory mineral analysis, such as 30% quartz, 30% mica, and 40% feldspar. Next, RGB two-dimensional images of the rock surface were acquired and processed. It should be noted that the two-dimensional images can be obtained by photographing the granite surface or by scanning the granite using structural scanning technology and then extracting the two-dimensional image. The image processing procedure for the two-dimensional image is as follows... Figure 2B As shown, relevant technicians can manually select the feature area and calculate the average value to obtain the RGB feature values ​​of the three minerals.

[0100] Next, the RGB feature values ​​of the three minerals are converted into HSV values, and these are used as cluster centers. Cluster analysis is performed with component ratio restrictions to determine the mineral composition of each pixel. The mineral composition is then used as the label content to add a component label to the corresponding pixel.

[0101] Furthermore, image recognition is performed on the two-dimensional images with added component labels. Anisotropy characterization parameters are obtained by calculating the aspect ratio of the mineral particles corresponding to each mineral component in the multiple mineral components. The fractal dimension and hourglass parameter corresponding to each mineral component in the multiple mineral components are also determined, resulting in multiple fractal dimensions and multiple hourglass parameters. The fractal dimension describes the shape of the mineral particles, and the hourglass parameter describes the size of the mineral particles, i.e., the concentration of distribution. A larger hourglass parameter indicates a more concentrated mineral distribution, and vice versa. Specifically, for example... Figure 2C As shown, based on the mineral cross-sectional distribution obtained from image recognition, the fractal dimension is calculated using a preset fractal dimension calculation method. This method includes box counting, random walk, and frequency domain methods. Specifically, box counting can be used for calculation, as shown in Formula 1 below:

[0102] Formula 1:

[0103] Where ε is the side length of the grid, and N(ε) is the number of grids occupied by the current mineral. Based on the distribution of mineral components obtained from image recognition, multiple mineral pixels corresponding to any given mineral component are identified. The nearest distance between each mineral pixel and other mineral pixels is assigned to that mineral pixel. The assigned values ​​for each mineral pixel corresponding to the given mineral component are summed to obtain V*. sum The first parameter. It should be noted that other mineral pixels refer to the mineral pixels corresponding to mineral components other than the mineral component itself. Multiple mineral pixels are aggregated into a rectangle or circle, and the nearest distance from each mineral element to the edge of the aggregated shape is assigned to the corresponding mineral element, thus determining V. Total The second parameter. The ratio of the first and second parameters is ultimately used as the hourglass parameter for the mineral composition. Finally, the hourglass parameter corresponding to each mineral composition is calculated, resulting in multiple hourglass parameters. Taking mineral A as an example, the distance *d* from each A mineral pixel to the nearest non-A mineral pixel is assigned to the corresponding A mineral pixel, and the assigned values ​​for all A mineral pixels are summed to obtain V*. sum Similarly, assuming all pixels of mineral A are grouped together (or the total number of pixels in the image are used directly) to form a large square or circle, the parameter V is obtained in the same way as above. Total Calculate V* sum With V Total The ratio obtained is the hourglass parameter.

[0104] 202. According to the anisotropy characterization parameters and component ratio, set the initial control parameters of the bubbling method, and use the bubbling method to construct a rock model of the target rock.

[0105] In this embodiment of the application, the control parameters are assigned values ​​according to the anisotropic characterization parameters corresponding to the target rock, namely the scale coefficient and the proportion of each mineral component, to obtain the initial control parameters. Then, bubbling is performed according to the initial control parameters to construct a rock model of the target rock.

[0106] Specifically, the initial control parameters for the bubble method are set as follows: model size S = 50mm × 50mm × 100mm, grid / pixel count NT = 500 × 500 × 1000; mineral ratio ω = quartz:mica:feldspar = 3:3:4; initial values ​​of bubble radius normal distribution parameters μ = 0, σ = 8; number of bubbles N = 10000; isotropic spherical bubbles are used, and the directional scale factor ratio δX:δY:δZ = 1:1:1; attribute value scaling factor δH = 1.

[0107] Next, as Figure 2D and 2FAs shown, the various mineral components are ranked in descending order of their proportion, and the distribution areas corresponding to each mineral component are extracted one by one according to the ranking order to obtain the rock model.

[0108] Specifically, for example, when constructing a two-dimensional rock model, then... Figure 2B As shown, for any mineral composition, according to initial control parameters, a predetermined number of circular or elliptical bubbles are randomly generated upwards or downwards within a predetermined plane range, so that multiple bubbles overlap within the predetermined plane range, exhibiting typical heterogeneous characteristics. Specifically, within the predetermined plane range, a predetermined number of circular or elliptical bubbles are generated upwards or downwards, and an elevation value is assigned to the points contained in each of the predetermined number of circular or elliptical bubbles. It should be noted that the elevation value is positive for upward-generated circular or elliptical bubbles, and negative for downward-generated circular or elliptical bubbles. For each point within the predetermined plane range, all the elevation values ​​assigned to the point are determined. By adding all the elevation values ​​corresponding to the point, the target elevation value associated with the point is obtained, thereby achieving the overlap of multiple bubbles within the predetermined plane range. It should be noted that the specific value of the predetermined number can be the system default value, or it can be determined by relevant personnel according to the actual situation. This application does not limit the specific meaning of the predetermined number. When the number of bubbles reaches the predetermined number, that is, when there are enough overlapping bubbles, a rough rock surface is obtained, that is, the rock model of the target rock. It should be noted that the bubble radius generated by the bubbling method follows a normal distribution. Further, the distribution area corresponding to each mineral component is selected within a preset plane range. Specifically, the selected point for the previous mineral component is deleted within the preset plane range, and the target elevation values ​​of the remaining points are ranked in descending order. Based on the component proportions corresponding to the mineral components, the selectable quantity values ​​corresponding to the mineral components are determined. Subsequently, multiple target points are selected one by one according to the ranking order of the target elevation values, with the number of target points consistent with the selectable quantity values. Finally, as shown... Figure 2E As shown, multiple selected target points are used as the distribution areas of mineral components. The distribution area corresponding to each mineral component is determined, resulting in a two-dimensional rock model of the target rock.

[0109] For example, assuming there are 1000 points within a preset planar area, and the composition ratio of cubic granite is quartz:mica:feldspar = 3:3:4, based on the composition ratio, firstly, a distribution area for feldspar is selected within the preset planar area. Bubbling is then performed within this area, and an elevation value is assigned to each point contained in the generated bubble. The target elevation value for each point within the preset planar area is then calculated. These target elevation values ​​are ranked from largest to smallest, and 400 target elevation values ​​are selected according to this ranking. The points corresponding to these 400 target elevation values ​​are then used as target points, and these target points are ultimately designated as the distribution area for feldspar. Further, a distribution area is selected for quartz, and bubblering is performed again. An elevation value is assigned to each point contained in the generated bubble, and the target elevation value for each point within the preset planar area is calculated. At this point, the target points selected for feldspar within the preset plane range need to be deleted. The remaining target elevation values ​​are then ranked from largest to smallest, and 300 target elevation values ​​are selected according to this ranking. The points corresponding to these 300 target elevation values ​​are taken as target points, and these target points are considered the distribution areas for quartz. Finally, the remaining points within the preset plane range can be considered the distribution areas for mica. In actual operation, to ensure the accuracy of the selected distribution areas and avoid under-selection or omission, the bubble-bubbling process can be repeated and values ​​assigned when selecting the distribution area for the last mineral component. The distribution area for the last mineral component is then selected based on the target elevation value. That is, if the target rock has n components, at least n-1 bubble-bubbling operations are required, and the distribution area corresponding to one mineral component is selected after each bubble-bubbling operation.

[0110] For example, constructing a three-dimensional rock model, such as Figure 2FAs shown, for any mineral component, according to initial control parameters, a preset number of spherical or ellipsoidal bubbles are generated within a preset space. Each point within a spherical or ellipsoidal bubble is randomly assigned a positive or negative attribute value, the size of which varies with the radius. Specifically, linear assignment is used. The attribute value of a point within a bubble is ρ = Rr, where R is the bubble radius and r is the distance from the point to the center. By summing all the attribute values ​​corresponding to the point, the target attribute value associated with the point is obtained, thus achieving overlap of the multiple bubbles within the preset plane. It should be noted that subsequent mineral components cannot cover the space occupied by previous mineral components. That is, after the first mineral component extracts a region or space, this extracted region or space is removed, and the second mineral component performs the subsequent extraction operation in the removed region or space. Specifically, the points selected by the previous mineral component are deleted within the preset space, and the target attribute values ​​of the remaining points are ranked in descending order. Based on the component proportions corresponding to the mineral components, the selectable quantity value corresponding to each mineral component is determined. Multiple target points are selected sequentially according to the ranking of the target attribute values, with the number of target points matching the available quantity. These selected target points are then used as the distribution areas of the mineral components. By determining the distribution area corresponding to each of the various mineral components, a three-dimensional rock model of the target rock is obtained.

[0111] 203. Calculate the specified fractal dimension and specified hourglass parameter corresponding to multiple mineral components in the rock model, and calculate the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter, respectively.

[0112] In this step, the specified fractal dimension and specified hourglass parameters corresponding to various mineral components in the rock model are first calculated. Based on the specified fractal dimension and specified hourglass parameters, the first difference between the specified fractal dimension and the target fractal dimension, and the second difference between the specified hourglass parameters and the target hourglass parameters are calculated. Then, by comparing the first difference with a first preset range and the second difference with a second preset range, it is determined whether the constructed rock model meets the requirements.

[0113] Specifically, if the rock model is a two-dimensional model, the mineral cross-sectional distribution of the rock model is obtained through image recognition based on the distribution area corresponding to each mineral component. The fractal dimension of each mineral component is calculated using box counting based on the mineral cross-sectional distribution, resulting in multiple fractal dimensions. The average of these multiple fractal dimensions is used as the specified fractal dimension. Based on the mineral component distribution obtained through image recognition, multiple mineral pixels corresponding to any given mineral component are identified. The nearest distance of each mineral pixel to other mineral pixels is assigned to that mineral pixel. This assignment of values ​​to each mineral pixel corresponding to that mineral component is repeated, and the values ​​of multiple mineral pixels are summed to obtain V*. sumThe first parameter. This aggregates multiple mineral pixels into a rectangle or circle, assigning the nearest distance from each mineral element to the edge of the aggregated shape to that element, thus determining V. Total The second parameter. The ratio of the first and second parameters is used as the hourglass parameter for mineral components. Hourglass parameters are calculated for each mineral component, resulting in multiple hourglass parameters. The average of these multiple hourglass parameters is then used as the specified hourglass parameter. Furthermore, the first difference between the specified fractal dimension and the target fractal dimension, and the second difference between the specified hourglass parameter and the target hourglass parameter are calculated. By comparing the first difference with a first preset range and the second difference with a second preset range, it is determined whether the constructed rock model meets the requirements.

[0114] If the rock model is a three-dimensional model, multiple specified cross-sections are taken at equal intervals in the rock model, such as planes with z = d, 2d, 3d… at equal intervals as specified cross-sections. For each specified cross-section, the fractal dimension and hourglass parameter corresponding to the specified cross-section are calculated based on the distribution area of ​​each mineral component associated with the specified cross-section. Multiple fractal dimensions and multiple hourglass parameters corresponding to multiple specified cross-sections are calculated, and the mean of the multiple fractal dimensions is taken as the specified fractal dimension, and the mean of the multiple hourglass parameters is taken as the specified hourglass parameter.

[0115] 204. When the first difference hits the first preset range and the second difference hits the second preset range, the target model is obtained.

[0116] Specifically, when the first difference value matches the first preset range and the second difference value matches the second preset range, it indicates that the constructed rock model meets the requirements, and the constructed model can be used as the target model for relevant personnel to perform subsequent operation steps. It should be noted that the first preset range and the second preset range can be the same, and this application does not specifically limit the values ​​of the first preset range and the second preset range.

[0117] When the first difference does not fall within the first preset range or the second difference does not fall within the second preset range, the initial control parameters need to be modified to obtain updated control parameters. Specifically, when the first difference does not fall within the first preset range, the first specified range and the second specified range associated with the first preset range are determined. For example, if the first difference is less than 0.1 and does not fall within the first preset range (0-0.05), then the first specified range (-∞, 0) and the second specified range (0.05, +∞) associated with the first preset range are checked. If the first difference falls within the first specified range, the standard deviation of the bubble radius distribution in the initial control parameters is reduced to obtain updated control parameters; if the first difference falls within the second specified range, the standard deviation of the bubble radius distribution in the initial control parameters is increased to obtain updated control parameters. When the second difference does not hit the second preset range, the third and fourth specified ranges associated with the second preset range are determined. If the second difference hits the third specified range, the generated grid size in the initial control parameters is reduced to obtain the updated control parameters. If the second difference hits the fourth specified range, the generated grid size in the initial control parameters is increased to obtain the updated control parameters.

[0118] Repeat steps 202 and 203 above. Based on the updated control parameters, construct a rock model of the target rock using the bubble method, and recalculate the specified fractal dimension and specified hourglass parameters corresponding to the rock model, the first difference between the specified fractal dimension and the target fractal dimension, and the second difference between the specified hourglass parameters and the target hourglass parameters. If the first difference does not fall within the first preset range or the second difference does not fall within the second preset range, modify the updated control parameters and reconstruct the rock model according to the modified updated control parameters until the first difference of the reconstructed rock model falls within the first preset range and the corresponding second difference falls within the second preset range.

[0119] In summary, such as Figure 2GAs shown, the target rock is first subjected to compositional detection and image recognition to determine mineral distribution characteristic parameters. These parameters include the proportions of various mineral components, anisotropy characterization parameters, target fractal dimensions, and target hourglass parameters corresponding to each mineral component. Based on these parameters, initial control parameters for the bubble-diffusing method are set, and bubble-diffusing begins according to these parameters. The number of bubble-diffusing cycles is continuously counted. When the number of cycles reaches a target value, individual mineral components are selected until all mineral components are identified. Based on the selection results, the fractal dimension and hourglass parameters for each mineral component are calculated, resulting in multiple fractal dimensions and multiple hourglass parameters. The mean of these multiple fractal dimensions and the mean of these multiple hourglass parameters are used as the specified fractal dimension and the specified hourglass parameter, respectively. Next, the first difference between the specified fractal dimension and the target fractal dimension, and the second difference between the specified hourglass parameter and the target hourglass parameter are calculated. When the first difference falls within a first preset range and the second difference falls within a second preset range, the target model is obtained.

[0120] The method provided in this application obtains anisotropic characterization parameters of the target rock through laboratory mineral composition detection and numerical image processing of images of the target rock surface. Utilizing the correlation between these anisotropic characterization parameters and relevant parameters from the bubble-diffusing method, a rock model of the target rock is constructed using the bubble-diffusing method. Compared to other current methods, this invention is simpler and more efficient. It only requires image processing and analysis of the rock sample surface to batch construct three-dimensional rock models of various sizes, significantly reducing the excessive acquisition and processing of rock sample data, lowering costs while improving efficiency. Furthermore, it is completely unrestricted by size; theoretically, arbitrarily large-scale three-dimensional rock structure models can be established as long as computing power allows. In addition, by changing spherical bubbles to ellipsoidal bubbles, rock models with anisotropic mineral particle distribution characteristics can be generated. Whether the mineral composition has multiple irregular distributions or the pore structure is varied, two-dimensional or three-dimensional models can be established using this method, demonstrating great potential in physical experiments and numerical simulations of rock mechanics.

[0121] Furthermore, as Figure 1 To specifically implement the method, this application provides a multi-mineral rock structure modeling device, such as... Figure 3A As shown, the device includes: a determining module 301, a setting module 302, a calculation module 303, and a generating module 304.

[0122] The determining module 301 is used to detect the mineral composition of the target rock, obtain the composition ratio of the various mineral components contained in the target rock, and determine the anisotropic characterization parameters of the target rock, as well as the target fractal dimension and target hourglass parameters corresponding to the various mineral components based on the composition ratio and the two-dimensional image corresponding to the target rock.

[0123] The setting module 302 is used to set the initial control parameters of the bubbling method according to the anisotropy characterization parameters and the component ratio, and to construct a rock model of the target rock using the bubbling method.

[0124] The calculation module 303 is used to calculate the specified fractal dimension and specified hourglass parameter corresponding to the multiple mineral components in the rock model, and to calculate the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter, respectively.

[0125] The generation module 304 is used to obtain a target model when the first difference hits a first preset range and the second difference hits a second preset range.

[0126] In a specific application scenario, the determining module 301 is used to add component labels to each pixel in the two-dimensional image according to the component ratio; perform image recognition on the two-dimensional image with added component labels, and obtain the anisotropy characterization parameters by calculating the ratio of the major and minor axes of the mineral particles corresponding to each of the multiple mineral components; perform image recognition on the two-dimensional image with added component labels, determine the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components, obtain multiple fractal dimensions and multiple hourglass parameters, and calculate the mean of the multiple fractal dimensions as the target fractal dimension, and calculate the mean of the multiple hourglass parameters as the target hourglass parameter.

[0127] In a specific application scenario, the determining module 301 is used to calculate the fractal dimension corresponding to each of the multiple mineral components using a preset fractal dimension calculation method, thereby obtaining multiple fractal dimensions corresponding to the multiple mineral components. The fractal dimension is used to describe the shape of the mineral particles corresponding to the mineral components. Based on the image recognition, the component distribution of the multiple mineral components is obtained, and based on the component distribution, multiple mineral pixels corresponding to any mineral component are identified. The nearest distance of each mineral pixel to other mineral pixels is assigned to the mineral pixel, where other mineral pixels are mineral pixels corresponding to mineral components other than the mineral component in the multiple mineral components. The assigned values ​​of the multiple mineral pixels are summed to obtain a first parameter. The multiple mineral pixels are aggregated to obtain an aggregated graphic. The nearest distance of each mineral element to the edge of the aggregated graphic is assigned to the corresponding mineral element to determine a second parameter. The ratio of the first parameter and the second parameter is used as the hourglass parameter of the mineral component, which is used to describe the size of the mineral particles of the mineral component. The hourglass parameter corresponding to each mineral component is calculated to obtain multiple hourglass parameters.

[0128] In a specific application scenario, the setting module 302 is used to rank the various mineral components according to their proportion from largest to smallest, and select the distribution area corresponding to each mineral component one by one according to the ranking order to obtain the rock model; wherein, the step of selecting the distribution area corresponding to each mineral component one by one to obtain the rock model includes: if a two-dimensional rock model is constructed, then for any mineral component, according to the initial control parameters, a preset number of circular or elliptical bubbles are generated upward or downward within a preset plane range, and an elevation value is assigned to the points contained in each of the preset number of circular or elliptical bubbles; the elevation value of the upwardly generated circular or elliptical bubbles is positive; the elevation value of the downwardly generated circular or elliptical bubbles is negative; for the mineral component, the elevation value is positive; the elevation value of the downwardly generated circular or elliptical bubbles is negative; the elevation value of the mineral component is positive; the elevation value of the mineral component ... For each point within a preset plane range, determine all assigned elevation values ​​for that point. Summing all elevation values ​​for each point yields a target elevation value associated with that point, enabling the multiple bubbles to overlap within the preset plane range. Delete the point selected for the preceding mineral component within the preset plane range. Rank the remaining points by their target elevation values ​​from largest to smallest. Determine the optional quantity value corresponding to each mineral component based on its component proportion. Select multiple target points one by one according to the ranking order of the target elevation values, ensuring the number of target points matches the optional quantity value. Use the selected target points as the distribution area of ​​the mineral components. Determine the distribution area corresponding to each of the multiple mineral components to obtain a two-dimensional rock model of the target rock.

[0129] In specific application scenarios, the setting module 302 is also used to, if a three-dimensional rock model is constructed, generate a preset number of spherical or ellipsoidal bubbles within a preset space for any mineral component according to the initial control parameters; assign positive or negative attribute values ​​to the points contained in each of the preset number of spherical or ellipsoidal bubbles; for each point within the preset space, determine all attribute values ​​assigned to the point; obtain the target attribute value associated with the point by adding all attribute values ​​corresponding to the point, so as to achieve overlap of the multiple bubbles within the preset plane; delete the point selected by the previous mineral component within the preset space; rank the target attribute values ​​of the remaining points in descending order; determine the optional quantity value corresponding to the mineral component based on the component ratio corresponding to the mineral component; select multiple target points one by one according to the ranking order of the target attribute values, the number of the multiple target points being consistent with the optional quantity value; use the selected multiple target points as the distribution area of ​​the mineral component; determine the distribution area corresponding to each of the multiple mineral components to obtain the three-dimensional rock model of the target rock.

[0130] In specific application scenarios, the calculation module 303 is used to: if the rock model is a two-dimensional model, determine the fractal dimension and hourglass parameter corresponding to each of the multiple mineral components based on the distribution area corresponding to each of the mineral components, obtain multiple fractal dimensions and multiple hourglass parameters, and calculate the mean of the multiple fractal dimensions as the specified fractal dimension and the mean of the multiple hourglass parameters as the specified hourglass parameter; if the rock model is a three-dimensional model, extract multiple specified cross sections at equal intervals from the rock model, calculate multiple fractal dimensions and multiple hourglass parameters corresponding to the multiple specified cross sections based on the distribution area of ​​each mineral component associated with each specified cross section, use the mean of the multiple fractal dimensions as the specified fractal dimension, and use the mean of the multiple hourglass parameters as the specified hourglass parameter.

[0131] In specific application scenarios, such as Figure 3B As shown, the device also includes: a modification module 305 and a comparison module 306.

[0132] The modification module 305 is used to modify the initial control parameters to obtain updated control parameters when the first difference does not hit the first preset range or the second difference does not hit the second preset range.

[0133] The comparison module 306 is used to construct a rock model of the target rock using the bubble method according to the updated control parameters, and to recalculate the specified fractal dimension and specified hourglass parameter corresponding to the rock model, the first difference between the specified fractal dimension and the target fractal dimension and the second difference between the specified hourglass parameter and the target hourglass parameter;

[0134] The modification module 305 is further configured to modify the update control parameters when the first difference does not hit the first preset range or the second difference does not hit the second preset range, and to reconstruct the rock model according to the modified update control parameters until the first difference corresponding to the reconstructed rock model hits the first preset range and the corresponding second difference hits the second preset range.

[0135] In a specific application scenario, the modification module 305 is used to determine a first specified range and a second specified range associated with the first preset range when the first difference does not match the first preset range; if the first difference matches the first specified range, the standard deviation of the bubble radius distribution in the initial control parameters is reduced to obtain the updated control parameters; if the first difference matches the second specified range, the standard deviation of the bubble radius distribution in the initial control parameters is increased to obtain the updated control parameters; when the second difference does not match the second preset range, a third specified range and a fourth specified range associated with the second preset range are determined; if the second difference matches the third specified range, the generated mesh size in the initial control parameters is reduced to obtain the updated control parameters; if the second difference matches the fourth specified range, the generated mesh size in the initial control parameters is increased to obtain the updated control parameters.

[0136] The apparatus provided in this application obtains anisotropic characterization parameters of the target rock through laboratory mineral composition detection and numerical image processing of images of the target rock surface. Utilizing the correlation between these anisotropic characterization parameters and relevant parameters of the bubble-diffusing method, a rock model of the target rock is constructed using the bubble-diffusing method. Compared to other current methods, the method of this invention is simpler and more efficient. It only requires image processing and analysis of the rock sample surface to batch construct three-dimensional rock models of various sizes, greatly reducing the excessive acquisition and processing of rock sample data, lowering costs while improving efficiency. Furthermore, it is completely unrestricted by size; theoretically, arbitrarily large-scale three-dimensional rock structure models can be established as long as computing power allows. In addition, by changing spherical bubbles to ellipsoidal bubbles, rock models with anisotropic distribution characteristics of mineral particles can be generated. Whether the mineral composition has multiple irregular distributions or the pore structure is poor, two-dimensional or three-dimensional models can be established using this method, demonstrating great potential in physical experiments and numerical simulations of rock mechanics.

[0137] It should be noted that other corresponding descriptions of the functional units involved in the multi-mineral rock structure modeling device provided in this application embodiment can be found in the following references. Figure 1 and Figures 2A to 2G The corresponding descriptions in [the document] will not be repeated here.

[0138] In an exemplary embodiment, see Figure 4 Furthermore, a device is provided, comprising a communication bus, a processor, a memory, and a communication interface. It may also include input / output interfaces and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores a computer program, and the processor executes the program stored in the memory to perform the multi-mineral rock structure modeling method described in the above embodiments.

[0139] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented in hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0140] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0141] Those skilled in the art will understand that the modules in the apparatus of the implementation scenario can be distributed within the apparatus of the implementation scenario as described, or they can be located in one or more apparatuses different from this implementation scenario, with corresponding changes. The modules of the above-described implementation scenario can be combined into one module, or they can be further divided into multiple sub-modules.

[0142] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenario.

[0143] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method of modeling a multi-mineral rock structure, characterized by, The method comprises the following steps: mineral composition detection is performed on a target rock to obtain a component ratio of a plurality of mineral components contained in the target rock, and an anisotropy representation parameter corresponding to the target rock and a target fractal dimension and a target hourglass parameter corresponding to the plurality of mineral components are determined according to the component ratio and a two-dimensional image corresponding to the target rock; initial control parameters of a bubble method are set according to the anisotropy representation parameter and the component ratio, and a rock model of the target rock is constructed by using the bubble method; a specified fractal dimension and a specified hourglass parameter corresponding to the plurality of mineral components in the rock model are calculated, and a first difference between the specified fractal dimension and the target fractal dimension and a second difference between the specified hourglass parameter and the target hourglass parameter are calculated respectively; when the first difference hits a first preset range and the second difference hits a second preset range, a target model is obtained; wherein, the target hourglass parameter corresponding to the plurality of mineral components is determined according to the component ratio and the two-dimensional image corresponding to the target rock, comprising: adding a component label to each pixel in the two-dimensional image according to the component ratio; image recognition is performed on the two-dimensional image with the component label added, the hourglass parameter corresponding to each mineral component in the plurality of mineral components is determined, a plurality of hourglass parameters are obtained, and the mean value of the plurality of hourglass parameters is calculated as the target hourglass parameter; the determination of the hourglass parameter corresponding to each mineral component in the plurality of mineral components comprises: the component distribution of the plurality of mineral components is obtained according to the image recognition method, and a plurality of mineral pixels corresponding to any mineral component are confirmed according to the component distribution, the nearest distance of each mineral pixel to other mineral pixels is assigned to the mineral pixel, and the other mineral pixels are mineral pixels corresponding to mineral components other than the mineral component in the plurality of mineral components; the sum of the assignments of the plurality of mineral pixels is obtained to obtain a first parameter, and the plurality of mineral pixels are aggregated to obtain an aggregated graph, the nearest distance of each mineral element to the edge of the aggregated graph is assigned to the corresponding mineral element to determine a second parameter, and the ratio of the first parameter to the second parameter is taken as the hourglass parameter of the mineral component, which is used to describe the size of the mineral particles of the mineral component; a plurality of hourglass parameters are calculated by calculating the hourglass parameter corresponding to each mineral component; the construction of the rock model of the target rock by using the bubble method comprises: the plurality of mineral components are ranked in descending order of component ratio, and a distribution area corresponding to each mineral component is selected in the ranking order to obtain the rock model; wherein, the selection of the distribution area corresponding to each mineral component to obtain the rock model comprises: if a two-dimensional rock model is constructed, for any mineral component, a preset number of circular or elliptical bubbles are generated upward or downward within a preset plane range according to the initial control parameters, and a height value is assigned to the points contained in each circular or elliptical bubble in the preset number of circular or elliptical bubbles. The upward generated circular or elliptical bubble has a positive elevation value; The downward generated circular or elliptical bubble has a negative elevation value; For each point in the preset plane range, the total elevation value of the point is determined, and the target elevation value associated with the point is obtained by adding the total elevation value corresponding to the point, so as to realize the overlap of the multiple bubbles in the preset plane range. In the preset plane range, the selected points of the previous mineral component are deleted, the target elevation values of the remaining points are ranked in descending order, and the selectable number value corresponding to the mineral component is determined according to the component proportion corresponding to the mineral component. According to the ranking order of the target elevation values, multiple target points are selected one by one, and the number of the multiple target points is consistent with the selectable number value. The selected multiple target points are taken as the distribution area of the mineral component. The distribution area corresponding to each mineral component in the multiple mineral components is determined, and a two-dimensional rock model of the target rock is obtained.

2. The method of claim 1, wherein, According to the component proportion and the two-dimensional image corresponding to the target rock, the anisotropy representation parameter corresponding to the target rock and the target fractal dimension and target hourglass parameter corresponding to the multiple mineral components are determined, including: According to the component proportion, an ingredient label is added to each pixel in the two-dimensional image; Image recognition is performed on the two-dimensional image with the ingredient label added, the length-to-short-axis ratio of the mineral particles corresponding to each mineral component in the multiple mineral components is calculated, and the anisotropy representation parameter is obtained. Image recognition is performed on the two-dimensional image with the ingredient label added, the fractal dimension corresponding to each mineral component in the multiple mineral components is determined, a plurality of fractal dimensions are obtained, and the mean value of the plurality of fractal dimensions is calculated as the target fractal dimension.

3. The method of claim 2, wherein, The determination of the fractal dimension corresponding to each mineral component in the multiple mineral components includes: A preset fractal dimension calculation method is used to calculate the fractal dimension corresponding to each mineral component in the multiple mineral components, a plurality of fractal dimensions corresponding to the multiple mineral components are obtained, and the fractal dimension is used to describe the shape of the mineral particles corresponding to the mineral component.

4. The method of claim 1, wherein, The distribution area corresponding to each mineral component is selected one by one, and the rock model is obtained, further including: If a three-dimensional rock model is constructed, for any mineral component, a preset number of spherical or ellipsoidal bubbles are generated in a preset space according to the initial control parameter, a positive attribute value or a negative attribute value is assigned to each point included in the preset number of spherical or ellipsoidal bubbles; For each point in the preset space, the total attribute value of the point is determined, and the target attribute value associated with the point is obtained by adding the total attribute value corresponding to the point, so as to realize the overlap of the multiple bubbles in the preset plane range. In the preset space, the selected points of the previous mineral component are deleted, the target attribute values of the remaining points are ranked in descending order, and the selectable number value corresponding to the mineral component is determined according to the component proportion corresponding to the mineral component. Select multiple target points one by one according to the ranking order of the target attribute values, the number of the multiple target points being consistent with the selectable number value; select the multiple target points as the distribution area of the mineral composition; determine the distribution area corresponding to each of the multiple mineral compositions to obtain a three-dimensional rock model of the target rock.

5. The method according to any one of claims 1-4, characterized in that, The calculation of the specified fractal dimension and the specified sand clock parameter corresponding to the multiple mineral compositions in the rock model comprises: If the rock model is a two-dimensional model, determine the fractal dimension and the sand clock parameter corresponding to each of the multiple mineral compositions according to the distribution area corresponding to each of the mineral compositions, obtain multiple fractal dimensions and multiple sand clock parameters, and calculate the mean value of the multiple fractal dimensions as the specified fractal dimension and the mean value of the multiple sand clock parameters as the specified sand clock parameter; If the rock model is a three-dimensional model, multiple specified cross sections are taken at equal intervals in the rock model, the multiple fractal dimensions and the multiple sand clock parameters corresponding to the multiple specified cross sections are calculated according to the distribution area of each mineral composition associated with each specified cross section, and the mean value of the multiple fractal dimensions is taken as the specified fractal dimension and the mean value of the multiple sand clock parameters is taken as the specified sand clock parameter.

6. The method of claim 1, wherein, After the first difference value between the specified fractal dimension and the target fractal dimension and the second difference value between the specified sand clock parameter and the target sand clock parameter are calculated, the method further comprises: When the first difference value does not hit the first preset range or the second difference value does not hit the second preset range, modify the initial control parameter to obtain an updated control parameter; According to the updated control parameter, the bubble method is used to construct the rock model of the target rock, and the specified fractal dimension and the specified sand clock parameter corresponding to the rock model are calculated again, the first difference value between the specified fractal dimension and the target fractal dimension and the second difference value between the specified sand clock parameter and the target sand clock parameter; When the first difference value does not hit the first preset range or the second difference value does not hit the second preset range, modify the updated control parameter, and reconstruct the rock model according to the modified updated control parameter until the first difference value corresponding to the reconstructed rock model hits the first preset range and the second difference value corresponding to the reconstructed rock model hits the second preset range.

7. The method of claim 6, wherein, When the first difference value does not hit the first preset range or the second difference value does not hit the second preset range, the initial control parameter is modified to obtain an updated control parameter, which comprises: When the first difference value does not hit the first preset range, determine the first specified range and the second specified range associated with the first preset range, if the first difference value hits the first specified range, reduce the bubble radius distribution standard deviation in the initial control parameter to obtain the updated control parameter, and if the first difference value hits the second specified range, increase the bubble radius distribution standard deviation in the initial control parameter to obtain the updated control parameter; When the second difference value does not hit the second preset range, a third specified range and a fourth specified range associated with the second preset range are determined, if the second difference value hits the third specified range, the generated grid size in the initial control parameter is reduced to obtain the updated control parameter, if the second difference value hits the fourth specified range, the generated grid size in the initial control parameter is increased to obtain the updated control parameter.

8. A multi-mineral rock structure modeling apparatus, characterized by, Comprise: The determination module is configured to perform mineral component detection on the target rock to obtain component proportions of a plurality of mineral components contained in the target rock, and determine an anisotropy representation parameter corresponding to the target rock and target fractal dimensions and target hourglass parameters corresponding to the plurality of mineral components according to the component proportions and a two-dimensional image corresponding to the target rock; The setting module is configured to set initial control parameters of a bubbling method according to the anisotropy representation parameter and the component proportions, and construct a rock model of the target rock by using the bubbling method; The calculation module is configured to calculate specified fractal dimensions and specified hourglass parameters corresponding to the plurality of mineral components in the rock model, and calculate a first difference value between the specified fractal dimensions and the target fractal dimensions and a second difference value between the specified hourglass parameters and the target hourglass parameters, respectively; The generation module is configured to obtain a target model when the first difference value hits a first preset range and the second difference value hits a second preset range. The determination module is configured to add a component label to each pixel in the two-dimensional image according to the component proportions when determining the target hourglass parameters corresponding to the plurality of mineral components according to the component proportions and the two-dimensional image corresponding to the target rock; and perform image recognition on the two-dimensional image with the component label added to determine hourglass parameters corresponding to each mineral component in the plurality of mineral components, obtain a plurality of hourglass parameters, and calculate a mean value of the plurality of hourglass parameters as the target hourglass parameters. When determining the hourglass parameters corresponding to each mineral component in the plurality of mineral components, the determination module is configured to obtain a component distribution of the plurality of mineral components according to the image recognition method, and confirm a plurality of mineral pixels corresponding to any mineral component according to the component distribution, assign a nearest distance of each mineral pixel to other mineral pixels to the mineral pixel, the other mineral pixels being mineral pixels corresponding to mineral components other than the mineral component in the plurality of mineral components; sum the assignments of the plurality of mineral pixels to obtain a first parameter, aggregate the plurality of mineral pixels to obtain an aggregated graph, assign a nearest distance of each mineral element to an edge of the aggregated graph to the corresponding mineral element, determine a second parameter, and take a ratio of the first parameter and the second parameter as the hourglass parameter of the mineral component, the hourglass parameter being used to describe a mineral particle size of the mineral component; and calculate the hourglass parameter corresponding to each mineral component to obtain a plurality of hourglass parameters. The setting module, when constructing the rock model of the target rock by using the bubble method, is configured to: rank the plurality of mineral components in descending order of component proportion, and select the distribution area corresponding to each of the mineral components in the ranked order to obtain the rock model; wherein the selecting the distribution area corresponding to each of the mineral components to obtain the rock model comprises: if a two-dimensional rock model is constructed, for any mineral component, a preset number of circular or elliptical bubbles are generated upward or downward in a preset plane range according to the initial control parameter, and a height value is assigned to each point included in each of the bubbles; the circular or elliptical bubble generated upward has a positive height value; the circular or elliptical bubble generated downward has a negative height value; for each point in the preset plane range, the target height value associated with the point is obtained by adding all the height values assigned to the point, so as to realize the overlap of the bubbles in the preset plane range; the points selected by the previous mineral component are deleted in the preset plane range, the target height values of the remaining points are ranked in descending order, the selectable number value corresponding to the mineral component is determined according to the component proportion corresponding to the mineral component, a plurality of target points are selected in the ranked order of the target height values, the number of the target points is consistent with the selectable number value, the selected plurality of target points are taken as the distribution area of the mineral component, and the distribution areas corresponding to each of the plurality of mineral components are determined to obtain the two-dimensional rock model of the target rock. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the method in any one of claims 1 to 7.

Images