Method for establishing digital twin of formation geological attributes based on well logging data

Abstract

The present invention relates to a method for establishing a digital twin of formation geological attributes based on well logging data, and belongs to the field of electronic digital data processing technology. By preprocessing single well logging data; selecting a representative single well logging curve or performing dimensionality reduction processing on multiple well logging curves, and combining the formation division method, the top boundary depth and bottom boundary depth of the small layer can be obtained more accurately, thereby improving the accuracy of formation division; and a three-dimensional visualization display of formation geological attributes is realized. This intuitive visualization method helps engineers to understand the formation structure and geological attribute distribution more clearly, and provides strong support for the formulation of oil and gas reservoir development plans; based on accurate formation geological attribute digital twins, engineers can more effectively analyze the distribution characteristics of formation geological attributes, and then formulate a more scientific oil and gas reservoir development plan. This helps to increase the production of oil and gas wells, reduce development costs, and improve overall economic benefits.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic digital data processing, and in particular relates to a method for establishing a digital twin of formation geological attributes based on well logging data. Background Art

[0002] In the process of oil and gas exploration and development, well logging data is an important means to obtain the geological properties of formations. However, traditional well logging data analysis methods can only provide limited geological information based on the area around a single well, which is difficult to fully and accurately reflect the geological characteristics of the formation. Well logging data is an important basis for understanding the geological characteristics of formations in the process of oil and gas exploration and development. Its data interpretation results can contain a variety of formation geological information, covering lithology, porosity, permeability, fluid properties and other aspects. However, the current mining and utilization of well logging data is still relatively low, and the potential of this data as a means of understanding formations has not been fully explored and utilized. With the rise of digital twin technology, it provides a new idea for real-time monitoring and prediction of formation geological properties. Digital twin technology establishes a virtual model of physical entities to achieve real-time interaction and mapping between physical entities and virtual models, providing a new means for monitoring and predicting formation geological properties in the process of oil and gas exploration and development. However, there are still some deficiencies in the existing technology when establishing digital twins, such as the division of formations based on well logging data, the establishment of formation geological property twins, and the timely update of formation geological properties. These problems limit the application effect of digital twin technology in the field of oil and gas exploration and development. Therefore, how to effectively apply digital twin technology in the field of oil and gas exploration and development, realize the digitization and visualization of formation geological properties, and solve the problems existing in existing technologies has become an important issue that needs to be solved urgently.

[0003] Therefore, at this stage, it is necessary to design a method, system and storage medium for establishing a digital twin of the geological properties of formations based on well logging data to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system and storage medium for establishing a digital twin of formation geological properties based on logging data, so as to solve the technical problems existing in the above-mentioned prior art and improve the analysis accuracy and visualization of formation geological properties, so that engineers can present the results based on the digital twin, obtain effective distribution characteristics of formation geological properties, and then formulate effective oil and gas reservoir development plans to increase oil and gas well production.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] The method for establishing a digital twin of formation geological attributes based on well logging data includes the following steps:

[0007] S1: Perform cleaning, denoising, interval extraction, and normalization preprocessing operations on single well logging data;

[0008] S2: Based on a single well logging curve that can represent the formation attribute characteristics or based on multiple well logging curves after dimensionality reduction, fine formation division is carried out to obtain the top and bottom boundary depths of the small layer;

[0009] S3: The same single-well logging data operation is performed on multiple wells to obtain the top boundary depth and bottom boundary depth of the small layer in the formation divided based on the logging data of each well;

[0010] S4: The top and bottom layers of the stratum are constructed based on the top and bottom depths of the corresponding small layers. Kriging interpolation is used to construct the top and bottom layers of the small layers covering the preset coordinate range. After the construction is completed, several layers are inserted between the top and bottom layers as needed. In this process, the grid distribution of each layer is obtained;

[0011] S5: On the basis of each layer of grids, the spatial grids with the same plane coordinates are connected by spatial hexahedrons to construct the framework of the stratigraphic twin;

[0012] S6: Assign the value reflecting the geological attribute of the formation to the spatial hexahedron to complete the establishment of the geological attribute twin of the formation. After the twin is established, the digital twin of the geological attribute of the formation is used to guide the next construction measures or the well location deployment of the new well. When new logging data is generated, the geological attribute information contained in the twin is updated according to the previous steps.

[0013] Furthermore, in step S1:

[0014] Remove the data that deviates from the normal value range or displays negative values ​​in the logging data;

[0015] According to the preset well depth step, the logging data within the step range is extracted by taking the average value, median value or weighted average value;

[0016] Extract the logging data of the corresponding step length according to the preset well depth step length;

[0017] The well logging data is normalized by dividing it by the maximum value in the well logging data to normalize it to between 0 and 1, including the endpoint values ​​0 and 1.

[0018] Furthermore, in step S2:

[0019] Fine division of the stratigraphic formations on the processed logging data:

[0020] Step S21: selecting a single well logging curve representing geological attribute characteristics of the formation;

[0021] Step S22: If a single well logging curve representing the geological attribute characteristics of the formation cannot be selected in step S21, a dimensionality reduction operation is performed on multiple well logging curves; a principal component analysis method is used to reduce the data dimension by extracting preset features while retaining the original information; the specific steps of the principal component analysis method include:

[0022] a. Standardization processing: standardize the original logging data so that the mean of the logging data is 0 and the variance is 1;

[0023] b. Calculate the covariance matrix: Calculate the covariance matrix between the standardized logging data to evaluate the correlation between different logging curves;

[0024] c. Solve for eigenvalues ​​and eigenvectors: Perform eigendecomposition on the covariance matrix to solve for the eigenvalues ​​and corresponding eigenvectors;

[0025] d. Select the principal component: According to the size of the eigenvalue, select the eigenvectors corresponding to the first N largest eigenvalues ​​as the principal component;

[0026] e. Calculate the principal component scores: project the original logging data onto the selected principal components to obtain the scores on each principal component;

[0027] Step S23: Based on the representative logging curve of a single well, fine division of the formation is carried out to obtain the top and bottom depths of the small layer; the fine division of the small layers in the formation is carried out by manually observing the changes in the logging data for annotation or by using an algorithm to extract a series of inflection points with obvious changes in the logging curve, and the inflection points serve as the boundary points where the logging data deviates from the existing trend; based on the inflection points, the vertical depth corresponding to the inflection points is calculated to obtain the top and bottom depths of a small layer in the formation traversed by the logging data.

[0028] Furthermore, in step S3:

[0029] The operation process of single well logging data in the previous steps is adopted for multiple wells to obtain the top boundary depth and bottom boundary depth of the small layer in the formation divided based on the logging data of each well, and the top boundary depth and bottom boundary depth of the same small layer in each well and the spatial coordinates of the corresponding well depth are recorded. When converting the depth, the depth is converted into altitude depth according to the wellhead ground altitude to form a discrete data point set of the top and bottom levels of the small layer.

[0030] Further, in step S4:

[0031] Based on the obtained discrete data point set of the small layer, the Kriging interpolation method is used to establish the small layer top layer and the small layer bottom layer covering the preset coordinate range, and the grid distribution of each layer is obtained at the same time;

[0032] Step S41: if the sub-layer depths of multiple wells and the corresponding spatial coordinates are not within a spatial coordinate range, then coordinate conversion is performed on the data points to ensure that all data points are in the same coordinate system;

[0033] Step S42: Selecting a Kriging interpolation model; applying the interpolation model to interpolate discrete data points of the small layer level to generate a stratigraphic level covering a certain coordinate range;

[0034] Step S43: inserting a number of layers between the top layer and the bottom layer as required;

[0035] Step S44: on the interpolated stratigraphic level, divide the grid according to a preset grid size to obtain the spatial coordinates of each grid boundary point in the grid.

[0036] Further, in step S5:

[0037] Based on the obtained stratigraphic sub-layer grid, the stratigraphic twin framework is established:

[0038] Step S51: Starting from the first small layer in the divided small layer level, identifying the spatial grids with the same plane coordinates above and below;

[0039] Step S52: using a spatial hexahedron connection method to connect spatial grids with the same plane coordinates to form a three-dimensional framework unit of the formation twin;

[0040] Step S53: Repeat the above steps S51 and S52 until all levels of grids are connected into spatial hexahedrons to construct a complete three-dimensional framework of the formation twin.

[0041] Further, in step S6:

[0042] Based on the obtained three-dimensional grid of the formation twin, the values ​​reflecting the geological attributes of the formation are assigned to the spatial hexahedron to complete the establishment of the digital twin of the geological attributes of the formation; the digital twin is used to guide the next construction measures or the deployment location of the new well. When new logging data is generated, the geological attribute information contained in the twin is updated;

[0043] Step S61: selecting or calculating the data to be assigned to the formation twin from the well logging data that has been processed in the previous step;

[0044] Step S62: Associating the selected data with the three-dimensional grid of the formation twin to ensure that each spatial hexahedron can correspond to the corresponding geological attribute data;

[0045] Step S63: According to the content that needs to be presented by the digital twin of the geological attributes of the stratum, a grid assignment method is selected when the data in step S62 is associated, and the geological attribute value represented by the data in step S62 is assigned to each spatial hexahedron;

[0046] Step S64: Performing a three-dimensional visualization display on the optimized formation twin;

[0047] Step S65: guiding the next construction measures or the well location deployment of new wells according to the digital twin of the geological attributes of the formation;

[0048] Step S66: when new logging data is generated, timely collect and organize the logging data according to the processing method of the previous step;

[0049] Step S67: According to the method of the previous step, the new geological attribute data is assigned to the corresponding spatial hexahedron in the formation twin.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] By preprocessing the single well logging data, including cleaning and denoising, interval extraction, normalization and other operations, the quality and consistency of the data are significantly improved, laying a solid foundation for the subsequent fine division of the formation and geological attribute analysis, thereby improving the analysis accuracy of the formation geological attributes; by selecting representative single logging curves or reducing the dimension of multiple logging curves, and combining the formation division method, the top and bottom depths of the small layer can be obtained more accurately, thereby improving the accuracy of the formation division; the three-dimensional visualization of the formation geological attributes is realized. This intuitive visualization method helps engineers to understand the formation structure and geological attribute distribution more clearly, and provides strong support for the formulation of oil and gas reservoir development plans; based on accurate formation geological attribute digital twins, engineers can more effectively analyze the distribution characteristics of formation geological attributes, and then formulate a more scientific oil and gas reservoir development plan. This helps to increase the production of oil and gas wells, reduce development costs, and improve overall economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Schematic diagram of the analysis process involved in the method of the present invention.

[0053] Figure 2 It is a schematic diagram of a single well logging data curve involved in the method embodiment of the present invention.

[0054] Figure 3 It is a curve diagram of the single well logging data involved in the method embodiment of the present invention after taking the average value with the step length.

[0055] Figure 4The present invention is a schematic diagram of a curve after further interval extraction of the single well logging data involved in the method embodiment of the present invention.

[0056] Figure 5 The figure is a curve diagram of the single well logging data involved in the method embodiment of the present invention after further normalization operation.

[0057] Figure 6 The figure is a schematic diagram of a curve showing an inflection point (a black dot in the figure represents a top boundary well depth point or a bottom boundary well depth point of a small layer in the formation) analyzed from the single well logging data involved in the method embodiment of the present invention.

[0058] Figure 7 It is a schematic diagram of the top grid and the bottom grid of the sub-layer involved in the embodiment of the method of the present invention.

[0059] Figure 8 It is a schematic diagram of the top grid and the bottom grid of the formation sublayer involved in the embodiment of the method of the present invention after the grid is inserted.

[0060] Fig. 9 It is a schematic diagram of the formation sub-layer grids involved in the method embodiment of the present invention connected by spatial hexahedrons.

[0061] Fig.10 It is a schematic diagram of the digital twin of the geological properties of the sub-layer of the stratum involved in the embodiment of the method of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0063] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention. It should be noted that relational terms such as the terms "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0064] Moreover, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a list of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus. In the absence of more restrictions, an element defined by the phrase "comprises a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0065] Example:

[0066] Method for establishing digital twin of formation geological attributes based on well logging data. First, the single well logging data is preprocessed by cleaning, denoising, interval extraction, normalization and other preprocessing operations to improve the quality and consistency of the data. Then, the formation is finely divided based on a single well logging curve that can represent the formation attribute characteristics or based on multiple well logging curves after dimensionality reduction to obtain the top and bottom boundary depths of the small layer. Then, the same single well logging data operation is performed on multiple wells to obtain the top and bottom boundary depths of the small layers in the formation divided based on the logging data for each well. After that, the formation level is constructed according to the top boundary depth of the corresponding small layer, and the formation level covering a certain coordinate range is established by Kriging interpolation or similar interpolation methods. In this process, the grid distribution of each layer is obtained. Subsequently, on the basis of each layer of grids, the spatial grids with the same plane coordinates are connected by spatial hexahedrons to construct the framework of the formation twin. Finally, the values ​​reflecting the geological attributes of the formation are assigned to the spatial hexahedron to complete the establishment of the digital twin of the geological attributes of the formation. After the digital twin is established, it can be used to guide the next construction measures or the well location deployment of new wells. When new logging data is generated, the geological attribute information contained in the digital twin can be updated according to the previous steps. So that engineers can obtain effective distribution characteristics of geological attributes of the formation based on the results presented by the digital twin, and then formulate effective oil and gas reservoir development plans to increase oil and gas well production. The analysis process is as follows: Figure 1 shown.

[0067] A method for establishing a digital twin of formation geological attributes based on well logging data is provided. The specific steps are as follows:

[0068] Step 1: Perform preprocessing operations such as cleaning and denoising, interval extraction, and normalization on the single well logging data to improve the quality and consistency of the data and lay a solid foundation for subsequent analysis.

[0069] (1) Remove the data that obviously deviates from the normal value range or displays negative values ​​in the logging data;

[0070] (2) According to the preset well depth step, the well logging data within the step range are averaged, median or weighted averaged to extract representative well logging data within the step range, thereby reducing the data density;

[0071] (3) Extract logging data with the corresponding step length according to the preset well depth step length, and further reduce the data density on the basis of ensuring the representativeness of the logging data;

[0072] (4) Normalizing the logging data. For example, the logging data can be normalized to between 0 and 1 by dividing it by the maximum value in the logging data.

[0073] Step 2: Based on a single well logging curve that can represent the formation attribute characteristics or multiple well logging curves after dimensionality reduction, carry out fine formation division and accurately obtain the top boundary depth of the small layer.

[0074] (1) Select a single well logging curve that can represent the geological attributes of the formation;

[0075] (2) If a representative single well logging curve cannot be selected in step 1, it is necessary to perform dimensionality reduction operations on multiple well logging curves. For example, methods such as principal component analysis can be used to reduce the data dimension by extracting the main features while retaining as much original information as possible. The specific steps of principal component analysis include:

[0076] 1) Standardization processing: Standardize the original logging data.

[0077] 2) Calculate the covariance matrix: Calculate the covariance matrix between the standardized logging data to evaluate the correlation between different logging curves.

[0078] 3) Solve for eigenvalues ​​and eigenvectors: Perform eigendecomposition on the covariance matrix to solve for the eigenvalues ​​and corresponding eigenvectors.

[0079] 4) Select principal components: According to the size of the eigenvalues, select the eigenvectors corresponding to the first N largest eigenvalues ​​as the principal components.

[0080] 5) Calculate the principal component scores: Project the original logging data onto the selected principal components to obtain the scores on each principal component.

[0081] (3) Based on the representative logging curves of a single well, a detailed division of small and medium layers in the formation is carried out to obtain the depths of the top and bottom boundaries of the small layers.

[0082] 1) Fine division can be achieved by manually observing the changes in logging data and marking them, or by using algorithms to extract a series of inflection points where the logging curve has obvious changes. These inflection points serve as boundary points where the logging data deviates from the existing trend, indicating that a new small layer may have been entered.

[0083] 2) When using the algorithm to extract a series of inflection points with obvious changes in the logging curve, the following method can be used:

[0084] a. Derivative method: First-order derivative: When the first-order derivative changes from positive to negative or from negative to positive, it means that the curve has turned at this point, that is, the inflection point. Second-order derivative: The second-order derivative of the logging curve reflects the concavity of the curve. When the second-order derivative changes from positive to negative or from negative to positive, it means that the curve changes from convex to concave or from concave to convex at this point, that is, this point is the inflection point.

[0085] b. Curve fitting and residual analysis: Perform polynomial or other forms of curve fitting on the logging curve, and then calculate the residual between the fitting curve and the actual logging curve. The point with a large residual may be an inflection point.

[0086] 3) Based on these inflection points, the vertical depth corresponding to the inflection points is calculated, and the top boundary depth and bottom boundary depth of a small layer in the formation passed by the well logging data are obtained.

[0087] Step 3: For multiple wells, the operation process of the single well logging data in the previous step is adopted to obtain the top boundary depth and bottom boundary depth of the small layer in the formation divided based on the logging data of each well, and the top boundary depth, bottom boundary depth and spatial coordinates of the same small layer in each well are recorded to form a set of discrete data points at the small layer level. The depth can be converted into altitude depth according to the wellhead altitude.

[0088] Step 4: Based on the obtained discrete data point set of the small layer, use the Kriging interpolation method or a similar interpolation method to establish the stratigraphic layer covering a certain coordinate range, and obtain the grid distribution of each layer.

[0089] (1) Perform coordinate transformation on the data points to ensure that all data points are in the same coordinate system.

[0090] (2) Select an appropriate Kriging interpolation model or similar interpolation method. Apply the interpolation model to interpolate discrete data points at the sub-layer level to generate the top and bottom levels of the stratigraphic sub-layer within a certain coordinate range.

[0091] (3) Insert several more layers between the top layer and the bottom layer of the sublayer as needed.

[0092] (4) At the stratigraphic level generated by interpolation, the grid is divided according to a certain grid size to obtain the spatial coordinates of each grid boundary point in the grid.

[0093] Step 5: Based on the grid of each layer in the small layer, establish the formation twin framework.

[0094] (1) Starting from the first level in the small layer, identify the spatial grids with the same plane coordinates above and below.

[0095] (2) These spatial grids with the same plane coordinates are connected by using a spatial hexahedron connection method to form the basic three-dimensional framework unit of the stratigraphic twin.

[0096] (3) Repeat the above steps until the grids at all levels are connected into spatial hexahedrons, thus constructing a complete three-dimensional framework of the stratigraphic twin.

[0097] Step 6: Based on the obtained three-dimensional grid of the formation twin, the values ​​reflecting the formation geological attributes are assigned to the spatial hexahedron to complete the establishment of the formation geological attribute twin, which can be used to guide the next construction measures or the well location deployment of new wells. When new logging data is generated, the geological attribute information contained in the twin can be updated according to the previous steps to ensure the timeliness and accuracy of the twin.

[0098] (1) From the well logging data that has been processed in the previous step, select or calculate the data that needs to be assigned to the formation twin.

[0099] (2) Associate the selected data with the three-dimensional grid of the stratigraphic twin to ensure that each spatial hexahedron can correspond to the corresponding geological attribute data.

[0100] (3) According to the content that the digital twin of the geological attributes of the formation needs to present, when performing the data association in step 2, select an appropriate grid assignment method, such as grid six-point interpolation, average value, median, and other different methods, and assign the geological attribute value represented by the data in step 2 to each spatial hexahedron.

[0101] (4) Perform a three-dimensional visualization of the optimized formation twin to intuitively display the geological properties and structural characteristics of the formation.

[0102] (5) The digital twin of the geological properties of the formation is used to guide the next construction measures or the deployment of new wells.

[0103] (6) When new logging data is generated, collect and organize the data in a timely manner according to the processing method of the logging data in the previous steps.

[0104] (7) Following the method of the previous steps, the new geological attribute data is assigned to the corresponding spatial hexahedron in the stratigraphic twin to achieve real-time updating of the twin.

[0105] The embodiments are further described below in conjunction with specific examples.

[0106] The first step is to process the single well logging data, such as Figure 2 As shown in FIG. 1 , a set of natural gamma data from a single well logging data is shown. The well is a vertical well, so the well depth is equivalent to the vertical depth. Figure 2The data in the table have no negative values ​​or values ​​that are significantly out of the normal range. Figure 2 The data in the interval is taken as a step of 10m, and the data in this interval are averaged to obtain the following Figure 3 The logging data distribution is shown in Figure 1, and then a data is extracted every 2m step, as shown in Figure 1. Figure 4 As shown in Figure 1, the distribution of logging data after the data density is finally reduced. Finally, the data is normalized and divided by the maximum value in the data. The resulting curve is shown in Figure 1. Figure 5 shown.

[0107] Step 2: Calculate the inflection point, take the derivative method, use the difference to calculate the first-order derivative, and then find the inflection point of the curve. Put the inflection point and the original logging data on the same graph, and you can see the inflection point position of the original data calculated based on the algorithm. At this time, you can manually select the appropriate inflection point of the stratification as the final stratification point, such as Figure 6 shown.

[0108] Figure 6 The figure is a schematic diagram of a curve showing an inflection point (a black dot in the figure represents a top boundary well depth point or a bottom boundary well depth point of a small layer in the formation) analyzed from the single well logging data involved in the method embodiment of the present invention.

[0109] In the third step, the analysis methods of the first and second steps are applied to multiple wells to obtain the top boundary well depth of each small layer. The well depth of the same small layer in multiple wells and the spatial coordinates under the well depth are recorded to form a discrete data point set of the same small layer level. The obtained discrete data set of the small layer level is in the form of {x, y, z}, where x and y are plane coordinates and z is the depth considering the wellhead altitude. For example, assuming that the obtained discrete data set of the small layer top level is {0, 0, -4100}, {500, 500, -4120}, {1000, 500, -4130}, {-500, -800, -4160}, {-1000,-1000, -4120}, and the discrete data set of the small layer bottom level is {0, 0, -4300}, {500, 500, -4330}, {1000,500, -4350}, {-1000, -800, -4310}, {-1000, -1000, -4320}, the above data unit is m.

[0110] In the fourth step, since the assumed data is located in the same coordinate, there is no need to convert it. In the x-coordinate range of -1000 to 1000, the step length is 50, and the y-coordinate range of -1000 to 1000, the step length is 50, and Kriging interpolation is used to find the coordinate depth at each step length. In this way, the coordinates of each node of the top grid and the bottom grid of the small layer can be obtained. The plotted graph is as follows Figure 7 shown.

[0111] The fifth step is to insert 8 layers evenly between the top grid and the bottom grid of the small layer to further refine the small layer, such as Figure 8 shown.

[0112] The sixth step is to use the spatial hexahedron to connect the upper and lower grids and establish the digital twin framework of the formation, such as Fig. 9 shown.

[0113] In the seventh step, the logging data corresponding to the well depth or the geological attribute value interpreted based on the logging data is assigned to each spatial hexahedron in the digital twin according to the well depth. For example, the natural gamma data corresponding to the logging data is assigned by taking the average value according to the well depth interval, and a certain color is assigned to each spatial hexahedron according to the value range. Then each hexahedron represents a natural gamma value, such as Fig.10 shown.

[0114] The eighth step is to guide the next construction measures or the deployment of new wells based on the established digital twin of the formation geological properties. When new logging data is generated, the formation sub-layer division or the mapping of the sub-layer geological properties in the digital twin can be re-performed according to the previous steps to ensure that the digital twin reflects the distribution of the formation geological properties.

[0115] The above are preferred embodiments of the present invention. Any changes made according to the technical solution of the present invention, as long as the resulting functions do not exceed the scope of the technical solution of the present invention, belong to the protection scope of the present invention.

Claims

1. A method for establishing a digital twin of formation geological attributes based on well logging data, characterized in that: The following steps are involved: S1: Perform cleaning, denoising, interval extraction, and normalization preprocessing operations on single well logging data; S2: Based on a single well logging curve that can represent the formation attribute characteristics or based on multiple well logging curves after dimensionality reduction, fine formation division is carried out to obtain the top and bottom boundary depths of the small layer; In step S2: Fine division of the stratigraphic formations on the processed logging data: Step S21: selecting a single well logging curve representing geological attribute characteristics of the formation; Step S22: If a single well logging curve representing the geological attribute characteristics of the formation cannot be selected in step S21, a dimensionality reduction operation is performed on multiple well logging curves; a principal component analysis method is used to reduce the data dimension by extracting preset features while retaining the original information; the specific steps of the principal component analysis method include: a. Standardization processing: standardize the original logging data so that the mean of the logging data is 0 and the variance is 1; b. Calculate the covariance matrix: Calculate the covariance matrix between the standardized logging data to evaluate the correlation between different logging curves; c. Solve for eigenvalues ​​and eigenvectors: Perform eigendecomposition on the covariance matrix to solve for the eigenvalues ​​and corresponding eigenvectors; d. Select the principal component: According to the size of the eigenvalue, select the eigenvectors corresponding to the first N largest eigenvalues ​​as the principal component; e. Calculate the principal component scores: project the original logging data onto the selected principal components to obtain the scores on each principal component; Step S23: Based on the representative logging curve of a single well, fine division of the formation is carried out to obtain the top and bottom depths of the small layer; the fine division of the small layers in the formation is carried out by manually observing the changes in the logging data for annotation or by using an algorithm to extract a series of inflection points with obvious changes in the logging curve, and the inflection points are used as the boundary points where the logging data deviates from the existing trend changes; based on the inflection points, the vertical depths corresponding to the inflection points are calculated to obtain the top and bottom depths of a small layer in the formation traversed by the logging data; S3: The same single-well logging data operation is performed on multiple wells to obtain the top boundary depth and bottom boundary depth of the small layer in the formation divided based on the logging data of each well; S4: The top and bottom layers of the stratum are constructed based on the top and bottom depths of the corresponding small layers. Kriging interpolation is used to construct the top and bottom layers of the small layers covering the preset coordinate range. After the construction is completed, several layers are inserted between the top and bottom layers as needed. In this process, the grid distribution of each layer is obtained; In step S4: Based on the obtained discrete data point set of the small layer, the Kriging interpolation method is used to establish the small layer top layer and the small layer bottom layer covering the preset coordinate range, and the grid distribution of each layer is obtained at the same time; Step S41: if the sub-layer depths of multiple wells and the corresponding spatial coordinates are not within a spatial coordinate range, then coordinate conversion is performed on the data points to ensure that all data points are in the same coordinate system; Step S42: Selecting a Kriging interpolation model; applying the interpolation model to interpolate discrete data points of the small layer level to generate a stratigraphic level covering a certain coordinate range; Step S43: inserting a number of layers between the top layer and the bottom layer as required; Step S44: on the interpolated stratigraphic level, dividing the grid according to a preset grid size to obtain the spatial coordinates of each grid boundary point in the grid; S5: On the basis of each layer of grids, the spatial grids with the same plane coordinates are connected by spatial hexahedrons to construct the framework of the stratigraphic twin; In step S5: Based on the obtained stratigraphic sub-layer grid, the stratigraphic twin framework is established: Step S51: Starting from the first small layer in the divided small layer level, identifying the spatial grids with the same plane coordinates above and below; Step S52: using a spatial hexahedron connection method to connect spatial grids with the same plane coordinates to form a three-dimensional framework unit of the formation twin; Step S53: repeat the above steps S51 and S52 until all levels of grids are connected into a spatial hexahedron, and a complete three-dimensional framework of the formation twin is constructed; S6: Assign the value reflecting the geological attribute of the formation to the spatial hexahedron to complete the establishment of the geological attribute twin of the formation. After the twin is established, the digital twin of the geological attribute of the formation is used to guide the next construction measures or the well location deployment of the new well. When new logging data is generated, the geological attribute information contained in the twin is updated according to the previous steps. In step S6: Based on the obtained three-dimensional grid of the formation twin, the values ​​reflecting the geological attributes of the formation are assigned to the spatial hexahedron to complete the establishment of the digital twin of the geological attributes of the formation; the digital twin is used to guide the next construction measures or the deployment location of the new well. When new logging data is generated, the geological attribute information contained in the twin is updated; Step S61: selecting or calculating the data to be assigned to the formation twin from the well logging data that has been processed in the previous step; Step S62: Associating the selected data with the three-dimensional grid of the formation twin to ensure that each spatial hexahedron can correspond to the corresponding geological attribute data; Step S63: According to the content that needs to be presented by the digital twin of the geological attributes of the stratum, a grid assignment method is selected when the data in step S62 is associated, and the geological attribute value represented by the data in step S62 is assigned to each spatial hexahedron; Step S64: Performing a three-dimensional visualization display on the optimized formation twin; Step S65: guiding the next construction measures or the well location deployment of new wells according to the digital twin of the geological attributes of the formation; Step S66: when new logging data is generated, timely collect and organize the logging data according to the processing method of the previous step; Step S67: According to the method of the previous step, the new geological attribute data is assigned to the corresponding spatial hexahedron in the formation twin.

2. The method for establishing a digital twin of formation geological attributes based on well logging data according to claim 1, characterized in that: In step S1: Remove the data that deviates from the normal value range or displays negative values ​​in the logging data; According to the preset well depth step, the logging data within the step range is extracted by taking the average value, median value or weighted average value; Extract the logging data of the corresponding step length according to the preset well depth step length; The well logging data is normalized by dividing it by the maximum value in the well logging data to normalize it to between 0 and 1, including the endpoint values ​​0 and 1.

3. The method for establishing a digital twin of formation geological attributes based on well logging data according to claim 2, characterized in that: In step S3: The operation process of single well logging data in the previous steps is adopted for multiple wells to obtain the top boundary depth and bottom boundary depth of the small layer in the formation divided based on the logging data of each well, and the top boundary depth and bottom boundary depth of the same small layer in each well and the spatial coordinates of the corresponding well depth are recorded. When converting the depth, the depth is converted into altitude depth according to the wellhead ground altitude to form a discrete data point set of the top and bottom levels of the small layer.

Images