A rock pore water-bearing identification method, device, equipment and memory

Abstract

The present application relates to a kind of rock pore water identification method and device, including the three-dimensional CT data of rock acquisition, the three-dimensional CT data of rock is handled, obtains the two-dimensional distribution map of three-dimensional density;According to the two-dimensional distribution map of three-dimensional density, the water-bearing region boundary of the rock is determined.The present application can obtain the distribution of water in rock and the non-uniform three-dimensional distribution of water under different water saturation state.The present application also relates to a kind of equipment and memory.

Description

Technical Field

[0001] This invention relates to the field of geosciences, and in particular to a method, apparatus, device, and memory for identifying water content in rock pores. Background Technology

[0002] Currently, identifying water content in rock pores is crucial for understanding rock damage and fracture characteristics. However, pore identification can only be achieved using relatively fine X-rays (micron-level CT, SEM, etc.), which cannot capture macroscopic fracture features. Therefore, larger-scale rocks are needed, but these require more intense X-rays to penetrate, making it impossible to obtain pore distribution information and directly observe water migration within the rock. Some studies use colored water to represent water movement, but this only reveals surface water migration, not internal water distribution. Furthermore, most studies are based on extreme conditions of saturation and dryness, and the effects of non-uniform water distribution remain unclear. Therefore, there is an urgent need for a technique to identify pore water distribution at resolutions lower than CT. Summary of the Invention

[0003] To address the technical problem that current methods cannot simultaneously identify both macroscopic fracture scale and microscopic pore size in rocks, this invention provides a method, apparatus, device, and memory for identifying water content in rock pores.

[0004] In a first aspect, the present invention provides a method for identifying water content in rock pores, the method comprising:

[0005] Acquire projection data and 3D CT data of dry rocks;

[0006] The feature points of the projected data of the dry rock are obtained by scanning the projected data of the dry rock, and the position of the feature points of the projected data is marked on the three-dimensional CT data of the dry rock to obtain the sample feature marking position.

[0007] Three-dimensional CT data of rocks with different water distributions are acquired, and the positions of the three-dimensional CT data of rocks with different water distributions are calibrated according to the sample features.

[0008] The three-dimensional CT data of the dried rock were sliced ​​to obtain a two-dimensional distribution map of the three-dimensional density of the dried rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample.

[0009] Three-dimensional CT data of rocks with different water distributions were sliced ​​to obtain two-dimensional distribution maps of the three-dimensional density of each rock with different water distributions.

[0010] Based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions, the fully water-bearing areas and other areas of rocks with different water distributions are determined.

[0011] Based on the above technical solution, further, the step of slicing the three-dimensional CT data of the dry rock to obtain a two-dimensional distribution map of the three-dimensional density of the dry rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample, specifically includes:

[0012] Based on the location of the reference sample features, the three-dimensional CT data of the dry rock is sliced ​​to obtain several slice data of the three-dimensional CT data of the dry rock.

[0013] Based on the radial average density values ​​of each slice of the calculated three-dimensional CT data of the dry rock, a two-dimensional distribution map of the three-dimensional density of the sample is obtained.

[0014] Based on the above technical solution, further, obtaining the two-dimensional distribution map of the sample's three-dimensional density from the radial average density values ​​of each slice of the calculated three-dimensional CT data of the dry rock specifically includes:

[0015] Voxel values ​​were selected at the same radius positions in each slice of the 3D CT data of the dry rock.

[0016] Based on the voxel values ​​of each slice of the 3D CT data of the dry rock, the circumferential average density value of each slice is calculated.

[0017] The circumferential average density value of each slice data is converted into a line density profile;

[0018] All the linear density profiles obtained by converting the circumferential average density values ​​of all slice data are sequentially plotted onto a two-dimensional plane along the axis to obtain a two-dimensional distribution map of the three-dimensional density of the sample.

[0019] Based on the above technical solution, further, the feature points of the projected data of the scanned dry rock are obtained, and the feature points of the projected data are located on the three-dimensional CT data of the dry rock to obtain the sample feature calibration position, specifically including:

[0020] The projection data of the dry rock is input into the convolutional layer of the neural network for processing to obtain the first feature location data;

[0021] The first feature location data is input into the normalization layer of the neural network to obtain the second feature location data;

[0022] The projection data of the dry rock is input into a deep separable convolutional layer of a neural network for processing, and the channel dimension and feature size of the projection data of the dry rock are transformed to obtain the third feature position data.

[0023] The second feature position data and the third feature position data are added together to obtain the feature point positions of the projection data;

[0024] Using the feature point locations of the projection data, a neural network is used to perform position calibration on the three-dimensional CT data of the dry rock to obtain the sample feature calibration location.

[0025] Based on the above technical solution, further, the step of determining the fully water-bearing regions and other regions of rocks with different water distributions based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions specifically includes:

[0026] The CT difference between the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and the two-dimensional distribution map of the three-dimensional density of the sample is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions.

[0027] By accumulating the CT differences in the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions, a cumulative histogram of the radial average density distribution of CT for rocks with different water distributions is obtained.

[0028] Based on the cumulative histogram of water content of rocks with different water distributions, the fully water-bearing regions and other regions of rocks with different water distributions are determined.

[0029] Based on the above technical solution, further, the subtraction of the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions and the two-dimensional distribution map of the sample's three-dimensional density to obtain the CT difference value of the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions specifically includes:

[0030] Acquire feature points from CT slices to correct spatial position errors of samples caused by differences in the physical location of the scan.

[0031] Two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and two-dimensional distribution maps of the three-dimensional density of dry samples were obtained using the CT numerical accumulation method.

[0032] The relative mass migration of water within the rock is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distribution from the two-dimensional distribution image of the three-dimensional density of the sample.

[0033] Based on the above technical solution, further, determining the fully aquifer region and other regions of rocks with different water distributions based on the cumulative histogram of the radial average density distribution of rocks with different water distributions specifically includes:

[0034] The range of parameter t in the cumulative histogram is obtained as t∈(a,b), where a is the minimum value, b is the maximum value, and t is the time when the rock is immersed in water, which represents the water content of the rock under different water distributions.

[0035] The cumulative histogram is divided into two parts.

[0036] Will Using regression lines respectively Represent the values ​​and calculate the errors Err1 and Err2 of parameter t;

[0037] Final error calculation

[0038] The value corresponding to the minimum error in the final error calculation formula is the threshold.

[0039] Thus, the regions with CT values ​​greater than the threshold in the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions are taken as the regions with the highest water content of the rocks, that is, the boundaries of the water-bearing regions of the rocks are obtained.

[0040] Secondly, the present invention also provides a rock pore water content identification device, the device comprising:

[0041] The first processing module is used to acquire projection data and 3D CT data of the dry rock;

[0042] The second processing module is used to scan the projection data of the dry rock to obtain the feature points of the projection data of the dry rock, and to mark the position of the feature points of the projection data on the three-dimensional CT data of the dry rock to obtain the sample feature marking position.

[0043] The third processing module is used to acquire three-dimensional CT data of rocks with different water distributions and to perform position calibration on the three-dimensional CT data of rocks with different water distributions according to the sample features.

[0044] The fourth processing module is used to slice the three-dimensional CT data of the dry rock to obtain a two-dimensional distribution map of the three-dimensional density of the dry rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample.

[0045] The fifth processing module is used to slice the three-dimensional CT data of rocks with different water distributions to obtain two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions.

[0046] The sixth processing module is used to determine the fully water-bearing areas and other areas of rocks with different water distributions based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions.

[0047] Thirdly, the present invention also provides a memory that stores multiple instructions for implementing a rock pore water content identification method as described in any of the above technical solutions.

[0048] Fourthly, the present invention also provides an electronic device, including a processor and a memory connected to the processor, the memory storing a plurality of instructions which can be loaded and executed by the processor to enable the processor to perform a rock pore water identification method as described in any of the above technical solutions.

[0049] This invention provides a method for identifying water content in rock pores, comprising acquiring three-dimensional CT data of the rock, processing the three-dimensional CT data to obtain a two-dimensional distribution map of the three-dimensional density, and determining the boundary of the water-bearing region of the rock based on the two-dimensional distribution map of the three-dimensional density. This invention can obtain the non-uniform three-dimensional distribution of water content inside the rock. Attached Figure Description

[0050] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0051] In the attached diagram:

[0052] Figure 1 This is a schematic flowchart of a method for identifying water content in rock pores provided in an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of CT slice image subtraction provided in an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of water migration on a rock surface provided in an embodiment of the present invention;

[0055] Figure 4 This is a schematic diagram of water migration on a rock surface provided in an embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram illustrating the conversion of three-dimensional data into two-dimensional data according to an embodiment of the present invention;

[0057] Figure 6 This is a schematic diagram of the internal water distribution of rocks under different saturation levels provided in an embodiment of the present invention;

[0058] Figure 7 This is a schematic diagram of the radial average distribution inside the rock under different saturation levels provided in the embodiments of the present invention;

[0059] Figure 8 This is a schematic diagram of three-dimensional water distribution in rocks provided in an embodiment of the present invention;

[0060] Figure 9 This is a schematic diagram of the cumulative mean distribution of rock CT values ​​provided in an embodiment of the present invention;

[0061] Figure 10This is a schematic diagram of the final water-bearing region of the rock provided in an embodiment of the present invention;

[0062] Figure 11 This is a schematic diagram of a rock pore water content identification device provided in an embodiment of the present invention. Detailed Implementation

[0063] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0064] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0065] The core idea of ​​this application is based on the fact that water filling pores will change the local CT value, which will appear darker in the image. By calculating the grayscale difference of rocks with different water distributions, the difference can be used to characterize the local changes caused by water filling pores in the rock.

[0066] Identifying water content in rock pores involves comparing CT slice images of two rocks under different saturation conditions. First, alignment is achieved using a digital image correlation algorithm. Then, the two slices are subtracted to obtain the relative mass migration of water within the rock. The water distribution results are displayed using voxel values ​​of different colors, representing the density changes caused by water migration within the rock. Simultaneously, a three-dimensional cumulative identification of the water infiltration process is performed, providing a more accurate characterization and identification of the process and distribution area of ​​water infiltration into the rock. Although the impact of water entering the pores on rock strength and fracture is easily noticeable, it is difficult to obtain the water migration process within the rock during the experiment because the pore scale and the rock sample scale cannot be simultaneously considered.

[0067] like Figure 1 As shown, this invention proposes a method for identifying water content in rock pores, the method comprising the following steps:

[0068] 110. Collect projection data and 3D CT data of the dry rock;

[0069] 120. Scan the projection data of the dry rock to obtain the feature points of the projection data of the dry rock, and then mark the position of the feature points of the projection data on the three-dimensional CT data of the dry rock to obtain the sample feature marking position.

[0070] 130. Obtain three-dimensional CT data of rocks with different water distributions, and perform position calibration on the three-dimensional CT data of rocks with different water distributions according to the sample features.

[0071] 140. The three-dimensional CT data of the dry rock is sliced ​​to obtain a two-dimensional distribution map of the three-dimensional density of the dry rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample.

[0072] 150. The three-dimensional CT data of rocks with different water distributions are sliced ​​to obtain two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions.

[0073] 160. Based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions, determine the fully water-bearing areas and other areas of rocks with different water distributions.

[0074] Based on the above technical solution, further, the step of slicing the three-dimensional CT data of the dry rock to obtain a two-dimensional distribution map of the three-dimensional density of the dry rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample, specifically includes:

[0075] Based on the location of the reference sample features, the three-dimensional CT data of the dry rock is sliced ​​to obtain several slice data of the three-dimensional CT data of the dry rock.

[0076] Based on the radial average density values ​​of each slice of the calculated three-dimensional CT data of the dry rock, a two-dimensional distribution map of the three-dimensional density of the sample is obtained.

[0077] Based on the above technical solution, further, obtaining the two-dimensional distribution map of the sample's three-dimensional density from the radial average density values ​​of each slice of the calculated three-dimensional CT data of the dry rock specifically includes:

[0078] Voxel values ​​were selected at the same radius positions in each slice of the 3D CT data of the dry rock.

[0079] Based on the voxel values ​​of each slice of the 3D CT data of the dry rock, the circumferential average density value of each slice is calculated.

[0080] The circumferential average density value of each slice data is converted into a line density profile;

[0081] All the linear density profiles obtained by converting the circumferential average density values ​​of all slice data are sequentially plotted onto a two-dimensional plane along the axis to obtain a two-dimensional distribution map of the three-dimensional density of the sample.

[0082] Based on the above technical solution, further, the feature points of the projected data of the scanned dry rock are obtained, and the feature points of the projected data are located on the three-dimensional CT data of the dry rock to obtain the sample feature location, specifically including:

[0083] The projection data of the dry rock is input into the convolutional layer of the neural network for processing to obtain the first feature location data;

[0084] The first feature location data is input into the normalization layer of the neural network to obtain the second feature location data;

[0085] The projection data of the dry rock is input into a deep separable convolutional layer of a neural network for processing, and the channel dimension and feature size of the projection data of the dry rock are transformed to obtain the third feature position data.

[0086] The second feature position data and the third feature position data are added together to obtain the feature point positions of the projection data;

[0087] Using the feature point locations of the projection data, a neural network is used to perform position calibration on the three-dimensional CT data of the dry rock to obtain the sample feature calibration location.

[0088] Based on the above technical solution, further, the step of determining the fully aquifer and partially aquifer regions of rocks with different water distributions based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions specifically includes:

[0089] The CT difference between the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and the two-dimensional distribution map of the three-dimensional density of the sample is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions.

[0090] By accumulating the CT differences in the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions, a cumulative histogram of the radial average density distribution of CT for rocks with different water distributions is obtained.

[0091] Based on the cumulative histogram of water content of rocks with different water distributions, the fully water-bearing regions and other regions of rocks with different water distributions are determined.

[0092] Based on the above technical solution, further, the step of subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions from the two-dimensional distribution map of the sample's three-dimensional density to obtain the CT difference value of the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions specifically includes:

[0093] Acquire feature points from CT slices to correct spatial position errors of samples caused by differences in the physical location of the scan.

[0094] Two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and two-dimensional distribution maps of the three-dimensional density of dry samples were obtained using the CT numerical accumulation method.

[0095] The relative mass migration of water within the rock is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distribution from the two-dimensional distribution image of the three-dimensional density of the sample.

[0096] Based on the above technical solution, further, determining the fully aquifer and partially aquifer regions of rocks with different water distributions based on the cumulative histogram of the radial average density distribution of rocks with different water distributions specifically includes:

[0097] The range of parameter t in the cumulative histogram is obtained, i.e., t∈(a,b), where a is the minimum value, b is the maximum value, and t is the time when the rock is soaked in water, which represents the water content of the rock under different water distributions.

[0098] The cumulative histogram is divided into two parts.

[0099] Will Using regression lines respectively Represent the values ​​and calculate the errors Err1 and Err2 of parameter t;

[0100] Final error calculation

[0101] The value corresponding to the minimum error in the final error calculation formula is the threshold.

[0102] Thus, the regions with CT values ​​greater than the threshold in the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions are taken as the regions with the highest water content of the rocks, that is, the boundaries of the water-bearing regions of the rocks are obtained.

[0103] It should be understood that the fully aquifer region and other regions of rocks with different water distributions can be obtained through the error calculation formula.

[0104] This embodiment provides a method for identifying water content in rock pores, including acquiring three-dimensional CT data of the rock, processing the three-dimensional CT data to obtain a two-dimensional distribution map of the three-dimensional density, and determining the boundary of the water-bearing region of the rock based on the two-dimensional distribution map of the three-dimensional density. This invention can obtain the non-uniform distribution state of water content within rocks.

[0105] It should be understood that, such as Figure 2This image displays CT slice images of two rocks under different water saturation conditions. The two images were aligned using a digital image correlation algorithm to eliminate the influence of global sample motion. The two slices were then subtracted to obtain the relative mass migration of water within the rocks. The subtraction results are displayed in different colors to represent the mass changes caused by water migration within the rock object.

[0106] By comparing images of the initial state and different water-saturated states, it becomes clear that the scanning precision cannot match the pore identification precision required for stronger X-rays to penetrate the sample. Therefore, it is difficult to directly observe water migration. However, water filling the pores alters the local CT value, causing the area to darken. For example... Figure 3 As shown, the CT count of the same sample under different moisture contents is statistically analyzed. The results show that the CT values ​​have a large difference, so it is reasonable to use grayscale difference to identify moisture distribution.

[0107] like Figure 4 As shown, the two-dimensional migration process of water is clearly visualized by grayscale difference and represented by a false-color image.

[0108] Comparisons between 2D slices can only show in-plane variations of water during rock migration. 3D CT data is further reprocessed and displayed in 2D maps, such as... Figure 5 As shown. The original 3D CT image consists of 700 slices, which are perpendicular to the principal axis (Z-axis) of the cylindrical sample. In cylindrical coordinates, the voxel values ​​at the same radius r of each slice are summed and averaged to transform each slice into a line profile, representing the average density along the tangential direction. In this way, the 3D-CT image can be converted into 700 line profiles and redrawn in a 2D planar profile, with grayscale values ​​representing the average density, as shown. Figure 5 As shown. Then, the spatial and temporal density variations in the 3-D data are presented through 2D density distribution maps of rocks at two different saturation states, as shown. Figure 5-6 As shown. In the experiment, the rock was completely covered by water, and water transport mainly occurred radially, using two-dimensional false color. Simultaneously, vertical averaging was performed on the 2D slices to represent the saturation level of the water's influence on the rock. Water has a lower CT number than air, therefore its CT value is smaller than that of pores in the water-bearing state. A larger difference indicates a greater local increase in water content. Figure 7 As shown, when water comes into contact with a rock, it migrates inward along the pores, and the rock becomes saturated in an order from the inside out.

[0109] Rock infiltration is a gradual process, but once the rock is fully saturated, its CT number is fixed. A larger difference in the linear density distribution map indicates a higher water content. Figure 8The three-dimensional identification of water-bearing areas in rocks can be represented as follows: Based on three-dimensional density distribution data, obtain the frequency distribution map and cumulative distribution map of CT number; use truncated regression estimation to obtain an accurate threshold for the water-bearing area, such as... Figure 9 The final result after segmentation is as follows Figure 10 .

[0110] like Figure 11 As shown, the present invention also provides a rock pore water content identification device, the device comprising:

[0111] The first processing module is used to acquire projection data and 3D CT data of the dry rock;

[0112] The second processing module is used to scan the projection data of the dry rock to obtain the feature points of the projection data of the dry rock, and to mark the position of the feature points of the projection data on the three-dimensional CT data of the dry rock to obtain the sample feature marking position.

[0113] The third processing module is used to acquire three-dimensional CT data of rocks with different water distributions and to perform position calibration on the three-dimensional CT data of rocks with different water distributions according to the sample features.

[0114] The fourth processing module is used to slice the three-dimensional CT data of the dry rock to obtain a two-dimensional distribution map of the three-dimensional density of the dry rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample.

[0115] The fifth processing module is used to slice the three-dimensional CT data of rocks with different water distributions to obtain two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions.

[0116] The sixth processing module is used to determine the fully water-bearing areas and other areas of rocks with different water distributions based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions.

[0117] This embodiment provides a rock pore water content identification device, which includes acquiring three-dimensional CT data of the rock, processing the three-dimensional CT data of the rock to obtain a two-dimensional distribution map of the three-dimensional density, and determining the water-bearing region boundary of the rock based on the two-dimensional distribution map of the three-dimensional density. This invention can obtain the distribution of water content inside the rock and its non-uniform three-dimensional distribution under different water saturation states.

[0118] The present invention also provides a memory that stores multiple instructions for implementing a rock pore water content identification method as described in any of the above technical solutions.

[0119] The present invention also provides an electronic device, including a processor and a memory connected to the processor, the memory storing a plurality of instructions which can be loaded and executed by the processor to enable the processor to perform a rock pore water content identification method as described in any of the above technical solutions.

[0120] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for identifying water content in rock pores, characterized in that, The method includes: Acquire projection data and 3D CT data of dry rocks; The feature points of the projected data of the dry rock are obtained by scanning the projected data of the dry rock, and the position of the feature points of the projected data is marked on the three-dimensional CT data of the dry rock to obtain the sample feature marking position. Three-dimensional CT data of rocks with different water distributions are acquired, and the positions of the three-dimensional CT data of rocks with different water distributions are calibrated according to the sample features. The three-dimensional CT data of the dried rock were sliced ​​to obtain a two-dimensional distribution map of the three-dimensional density of the dried rock, i.e., a two-dimensional distribution map of the three-dimensional density of the sample. Three-dimensional CT data of rocks with different water distributions were sliced ​​to obtain two-dimensional distribution maps of the three-dimensional density of each rock with different water distributions. Based on the two-dimensional distribution map of the three-dimensional density of the sample and the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions, the fully water-bearing regions and other regions of rocks with different water distributions are determined. The process of slicing the 3D CT data of the dried rock to obtain a 2D distribution map of the 3D density of the dried rock, i.e., a 2D distribution map of the 3D density of the sample, specifically includes: Based on the location of the reference sample features, the three-dimensional CT data of the dry rock is sliced ​​to obtain several slice data of the three-dimensional CT data of the dry rock. Based on the radial average density values ​​of each slice of the three-dimensional CT data of the dry rock obtained by calculation, a two-dimensional distribution map of the three-dimensional density of the sample is obtained. The step of obtaining a two-dimensional distribution map of the three-dimensional density of the sample based on the radial average density values ​​of each slice of the calculated three-dimensional CT data of the dry rock specifically includes: Voxel values ​​were selected at the same radius positions in each slice of the 3D CT data of the dry rock. Based on the voxel values ​​of each slice of the 3D CT data of the dry rock, the circumferential average density value of each slice is calculated. The circumferential average density value of each slice data is converted into a line density profile; All the linear density profiles obtained by converting the circumferential average density values ​​of all slice data are sequentially plotted onto a two-dimensional plane along the axis to obtain a two-dimensional distribution map of the three-dimensional density of the sample. The process of determining the fully water-bearing regions and other regions of rocks with different water distributions based on the two-dimensional distribution maps of the three-dimensional density of the samples and the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions specifically includes: The CT difference between the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and the two-dimensional distribution map of the three-dimensional density of the sample is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions. By accumulating the CT differences in the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions, a cumulative histogram of the radial average density distribution of CT for rocks with different water distributions is obtained. Based on the cumulative histogram of water content of rocks with different water distributions, determine the fully water-bearing regions and other regions of rocks with different water distributions. The method of determining the fully aquifer region and other regions of rocks with different water distributions based on the cumulative histogram of the radial average density distribution of CT in rocks with different water distributions specifically includes: The range of parameter t in the cumulative histogram is obtained, i.e., t∈(a,b), where a is the minimum value, b is the maximum value, and t is the time when the rock is soaked in water, which represents the water content of the rock under different water distributions. The cumulative histogram is divided into two parts. and ; Will and Using regression lines respectively and Represent the expression and calculate the error of parameter t. and ; Final error calculation ; The value corresponding to the minimum error in the final error calculation formula is the threshold. Thus, the regions with CT values ​​greater than the threshold in the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions are taken as the regions with the highest water content of the rocks, that is, the boundaries of the water-bearing regions of the rocks are obtained.

2. The method according to claim 1, characterized in that, The projected data of the scanned dry rock is used to obtain feature points of the projected data of the dry rock. These feature points are then positioned on the 3D CT data of the dry rock to obtain the sample feature calibration location. Specifically, this includes: The projection data of the dry rock is input into the convolutional layer of the neural network for processing to obtain the first feature location data; The first feature location data is input into the normalization layer of the neural network to obtain the second feature location data; The projection data of the dry rock is input into a deep separable convolutional layer of a neural network for processing, and the channel dimension and feature size of the projection data of the dry rock are transformed to obtain the third feature position data. The second feature position data and the third feature position data are added together to obtain the feature point positions of the projection data; Using the feature point locations of the projection data, a neural network is used to perform position calibration on the three-dimensional CT data of the dry rock to obtain the sample feature calibration location.

3. The method according to claim 1, characterized in that, The step of subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distributions from the two-dimensional distribution map of the sample's three-dimensional density to obtain the CT difference value of the two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions specifically includes: Acquire feature points from CT slices to correct spatial position errors of samples caused by differences in the physical location of the scan. Two-dimensional distribution maps of the three-dimensional density of rocks with different water distributions and two-dimensional distribution maps of the three-dimensional density of dry samples were obtained using the CT numerical accumulation method. The relative mass migration of water within the rock is obtained by subtracting the two-dimensional distribution map of the three-dimensional density of rocks with different water distribution from the two-dimensional distribution image of the three-dimensional density of the sample.

4. A memory, characterized in that, The system stores multiple instructions for implementing the rock pore water content identification method as described in any one of claims 1-3.

5. An electronic device, characterized in that, The device includes a processor and a memory connected to the processor, the memory storing multiple instructions that can be loaded and executed by the processor to enable the processor to perform the rock pore water identification method as described in any one of claims 1-3.

Images