A fracture azimuth fitting method and device based on seismic data

Abstract

The application discloses a fracture azimuth fitting method and device based on seismic data, and relates to the technical field of petroleum geophysical exploration. Ant body attribute data is acquired, grid cell division is performed on the ant body attribute data, the size of a local window is defined, grid quality evaluation is performed in the local window through a grid quality evaluation function, and the data quality of the grid cell is acquired; the size of a fitting window is selected, the spatial weight of each grid cell is acquired based on the spatial position of data, the quality weight of each grid cell is acquired based on the data quality of the grid cell, and a comprehensive weight is acquired; a double-target optimization function containing fitting accuracy and spatial smoothness is constructed, a constraint condition is designed, the fitting parameter vector in the double-target optimization function is solved, the optimal fitting parameter vector is acquired, and the fracture azimuth of the center of each fitting window is acquired. The problem of how to realize high-precision and high-robustness extraction of the fracture azimuth is solved.

Description

Technical Field

[0001] This application relates to the field of petroleum geophysical exploration technology, and in particular to a method and apparatus for fitting fracture azimuth angles based on seismic data. Background Technology

[0002] In unconventional oil and gas reservoir development, accurate characterization of natural fracture systems is crucial for improving extraction efficiency. Fracture azimuth, as a core parameter describing fracture geometry, directly impacts reservoir permeability, fracturing effectiveness, and well network deployment. Ant body technology, based on seismic coherence analysis, can effectively identify subsurface fractures and faults; however, accurately extracting fracture azimuth from ant body properties remains challenging, primarily due to data noise interference, lack of spatial continuity, insufficient geological constraints, and difficulties in fusing multi-scale features.

[0003] In existing technologies, the traditional gradient method derives the fracture azimuth by calculating the horizontal and vertical gradient values ​​of seismic attributes. However, it is sensitive to noise, prone to distortion in low signal-to-noise ratio regions, and assumes linear and continuous geological changes, making it difficult to adapt to complex structures. The point-to-point calculation mode also leads to insufficient spatial continuity. The simple least squares fitting method extracts the azimuth through linear or quadratic equations, but the equal-weighted processing of data points makes it susceptible to outliers, lacks effective constraints, and the fitting results may deviate from geological reality. Furthermore, it is sensitive to noise and data distribution, resulting in poor stability.

[0004] Therefore, how to achieve high-precision and robust extraction of crack azimuth angle has become an urgent technical problem to be solved. Summary of the Invention

[0005] In this embodiment of the application, a crack azimuth angle fitting method based on seismic data is provided, which solves the problem of how to achieve high-precision and high-robust extraction of crack azimuth angle.

[0006] In a first aspect, embodiments of this application provide a method for fitting crack azimuth angles based on seismic data. This method includes: acquiring ant body attribute data; dividing the ant body attribute data into grid cells; defining the size of a local window; and evaluating the grid quality within the local window using a grid quality evaluation function to obtain the data quality of the grid cells; selecting the size of the fitting window; obtaining the spatial weight of each grid cell based on its spatial location; obtaining the quality weight of each grid cell based on its data quality; obtaining a comprehensive weight based on the spatial weight and quality weight of each grid cell; constructing a bi-objective optimization function that includes fitting accuracy and spatial smoothness; designing constraints; using a constraint optimization algorithm to solve for the fitting parameter vector in the bi-objective optimization function to obtain the optimal fitting parameter vector; and obtaining the crack azimuth angle at the center of each fitting window based on the fitting mode and the optimal fitting parameter vector; wherein the fitting mode is either elliptical fitting or linear fitting.

[0007] In one possible implementation, the step of evaluating the grid quality within a local window using a grid quality evaluation function to obtain the data quality of the grid cells includes: the expression for the grid quality evaluation is: ;in, For grid cells Data quality, , and To evaluate the weights, For grid cells The signal-to-noise ratio, For grid cells The coherence index, For grid cells The gradient consistency index.

[0008] In one possible implementation, the selected fitting window size, the spatial weight of each grid cell obtained based on the data spatial location, and the quality weight of each grid cell obtained based on the data quality of the grid cell, include: the expression for the spatial weight of the grid cell is: ;in, Spatial weights of grid cells, The distance from the center point of the grid cell to the center of the fitting window. The spatial attenuation parameter is defined as the size of the fitting window. , It is an exponential function; the expression for the mass weight of the grid cell is: ;in, The mass weight of the grid cell. For grid cells Data quality, To fit the size of the window, To fit all grid cells within the window range The maximum value of data quality.

[0009] In one possible implementation, obtaining the comprehensive weight based on the spatial weight and the mass weight of each grid cell includes: the expression for the comprehensive weight is: ;in, For comprehensive weighting, Spatial weights of grid cells, This represents the quality weight of the grid cell.

[0010] In one possible implementation, the expression for the bi-objective optimization function is: ;in, It is a bi-objective optimization function. and For adaptive weighting coefficients, For the fitting accuracy term, For spatial smoothing terms, The fitting parameter vector is used; the adaptive weight coefficients are obtained through the noise level, which is based on the signal-to-noise ratio normalized within the fitting window. ;in, For data points Error value under the fitted model , Let x be the x-coordinate of the data point on the horizontal plane. The vertical coordinate of the data point is the ordinate in the vertical direction. For vertical layers, For comprehensive weighting, For indexing, This represents the number of data points in the fitting window; when ellipse fitting is selected... , ;in, These are the parameters of the ellipse; when selecting a straight line fit, , ;in, The parameters of the line; ;in, For any grid cell The number, This refers to the grid number of the grid cell in the horizontal direction. This refers to the grid number of the grid cell in the vertical direction. For vertical layers, Let be the gradient of the crack azimuth angle in the horizontal direction. This represents the gradient of the crack azimuth angle in the vertical direction.

[0011] In one possible implementation, the constraints include elliptic validity and orientation angle continuity; the expression for the elliptic validity constraint is: ;in, Let be the elliptic validity constraint function. These are the first three parameters of the ellipse. The value is a pre-defined minimum positive number; the expression for the continuity of the direction angle is: ;in, It is a function of direction angle continuity. For data points The azimuth of the crack at that location. For data points The azimuth of the crack at that location. The maximum permissible difference in crack azimuth angle set for the cosine. For the neighborhood, The radius of the neighborhood is specifically twice the size of the grid cell.

[0012] In one possible implementation, when the fitting mode is elliptic fitting, the equation of the ellipse corresponding to the optimal fitting parameter vector is: The crack azimuth is the direction of the major axis of the fitted ellipse; based on Determine the radians of the principal axis of the ellipse relative to the horizontal x-axis; where, The radians between the principal axis of the ellipse and the x-axis in the horizontal direction; based on Convert the radians of the principal axis of the ellipse and the x-axis in the horizontal direction into degrees; where, The angle between the principal axis of the ellipse and the x-axis in the horizontal direction; based on Determine the crack azimuth angle; where, The azimuth angle of the crack. This is a modulo operation.

[0013] In one possible implementation, when the fitting mode is linear fitting, the equation of the line corresponding to the optimal fitting parameter vector is: The crack azimuth is the direction of a straight line; based on Determine the radians of the line relative to the horizontal x-axis; where, Let x be the radians of the line relative to the x-axis in the horizontal direction. The direction vector of the line; based on Convert the radians of the line and the horizontal x-axis into degrees; where, The angle between the straight line and the x-axis in the horizontal direction; based on Determine the crack azimuth angle; where, The azimuth angle of the crack. This is a modulo operation.

[0014] Secondly, embodiments of this application provide a crack azimuth fitting device based on seismic data. The device includes: an evaluation module for acquiring ant-body attribute data, dividing the ant-body attribute data into grid cells, defining the size of a local window, and evaluating the grid quality within the local window using a grid quality evaluation function to obtain the data quality of the grid cells; a comprehensive weight acquisition module for selecting the size of the fitting window, acquiring the spatial weight of each grid cell based on the data spatial location, acquiring the quality weight of each grid cell based on the data quality of the grid cells, and acquiring a comprehensive weight based on the spatial weight and quality weight of each grid cell; a solution module for constructing a bi-objective optimization function including fitting accuracy and spatial smoothness, designing constraints, and using a constraint optimization algorithm to solve for the fitting parameter vector in the bi-objective optimization function to obtain the optimal fitting parameter vector; and a crack azimuth acquisition module for acquiring the crack azimuth angle at the center of each fitting window based on the fitting mode and the optimal fitting parameter vector; wherein the fitting mode is elliptical fitting or linear fitting.

[0015] One or more technical solutions provided in this application embodiment have at least the following technical effects: This application embodiment provides a crack azimuth fitting method based on seismic data. It obtains ant body attribute data, divides the ant body attribute data into grid cells, defines the size of a local window, and evaluates the grid quality within the local window using a grid quality evaluation function to obtain the data quality of the grid cells. The size of the fitting window is selected, and the spatial weight of each grid cell is obtained based on the spatial location of the data. The quality weight of each grid cell is also obtained based on its data quality. A comprehensive weight is obtained based on the spatial weight and quality weight of each grid cell. This comprehensive weight determination method, which considers both spatial location and data quality, can more comprehensively reflect the importance of each grid cell in the fitting process, making the fitting process more closely resemble the actual data characteristics. A bi-objective optimization function, including fitting accuracy and spatial smoothness, is constructed. Constraints are designed, and a constrained optimization algorithm is used to solve for the fitting parameter vector in the bi-objective optimization function to obtain the optimal fitting parameter vector. This method, while pursuing high fitting accuracy, also considers spatial smoothness, avoiding overfitting or discontinuities in the fitting results, and can obtain an optimal fitting parameter vector that better reflects the actual situation. Based on the fitting mode and the optimal fitting parameter vector, the fracture azimuth angle at the center of each fitting window is obtained; the fitting mode can be either elliptical fitting or linear fitting. According to the data distribution characteristics and prior geological knowledge, elliptical or linear fitting is automatically selected. The selection of multiple fitting modes increases the flexibility of the method, enabling it to adapt to the diversity of fracture morphology under different geological conditions, thereby obtaining fracture azimuth angle information more accurately and providing more precise data support for geological research and resource exploration. This solves the problem of how to achieve high-precision and robust extraction of fracture azimuth angles. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a crack azimuth fitting method based on seismic data, provided as an embodiment of this application;

[0018] Figure 2 A comparison diagram of azimuth field maps and final crack effects based on different azimuth fitting methods provided in the embodiments of this application;

[0019] Figure 3 This application provides a field diagram that divides ant body attribute data into grid cells, as shown in the embodiments of the present application.

[0020] Figure 4 A comparison diagram of well logging data and ant body identification results provided in the embodiments of this application;

[0021] Figure 5 A schematic diagram of a crack azimuth angle fitting device based on seismic data provided in this application embodiment;

[0022] Figure 6 This is a schematic diagram of a crack azimuth angle fitting server based on seismic data, provided in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0024] The following description of some technologies involved in the embodiments of this application is provided to aid understanding and should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, some descriptions of well-known functions and structures are omitted in the following description.

[0025] This application provides a method for fitting crack azimuth angles based on seismic data, such as... Figure 1As shown, the method includes steps S101 to S105. Wherein, Figure 1 This is merely one execution order shown in the embodiments of this application and does not represent the only execution order for a crack azimuth fitting method based on seismic data. Where the final result can be achieved, Figure 1 The steps shown can be performed in parallel or in reverse order.

[0026] S101: Obtain ant body attribute data, divide the ant body attribute data into grid cells, define the size of the local window, and evaluate the grid quality within the local window using the grid quality evaluation function to obtain the data quality of the grid cells.

[0027] Specifically, crack strength attribute (ant body attribute) data obtained based on seismic interpretation is used as the main input data, and the ant body attribute values ​​can be used... or It means that, among them Represents the coordinates of the center point of the grid cell. For grid cells The number, among which, For any grid cell Grid number in the horizontal direction, For grid cells Grid number in the vertical direction, This refers to a vertical layer.

[0028] Specifically, in the vertical layer Within, for each grid cell Define a local range centered on it. This is used to calculate the statistical properties of signals and noise. The size of the local window is typically chosen as [size not specified]. common A value of 5 or 7 indicates that the mesh cells within the local window will participate in subsequent calculations. For example, when... When the value is 5, It is actually 5 grid cells multiplied by 5 grid cells.

[0029] The grid quality is evaluated within a local window using a grid quality evaluation function to obtain the data quality of the grid cells, including the following:

[0030] The expression for grid quality evaluation is: .in, For grid cells Data quality, , and To evaluate the weights, For grid cells The signal-to-noise ratio, For grid cells The coherence index, For grid cells The gradient consistency index.

[0031] It is necessary to explain the evaluation weights. , and The values ​​are all .

[0032] Specifically, within a defined local window, the average value of all mesh properties is calculated to obtain the local average value. The calculation formula is: .in, Indicates local range any grid cell in Its number is . Represents any grid cell Ant body attribute values. Signal energy based on local mean. Defined as ,Right now .

[0033] Within the same local window, calculate all ant attribute values ​​relative to the local mean. Local variance The specific formula is as follows: Noise energy It is equivalent to local variance The value, .

[0034] Signal-to-noise ratio of grid cells Defined as signal energy With noise energy The logarithmic representation of the ratio: .

[0035] Specifically, in the vertical layer Within, assuming any grid cell The number is .

[0036] For the number is Grid cells Calculate the ant body attribute values The property differences from the 8 adjacent grid cells. The 8 adjacent grid cells are distributed among the grid cells. The coordinates of the top, bottom, left, right, and diagonal directions are: the coordinates of the top, bottom, left, right, and neighboring regions. and diagonal neighborhood For any number Adjacent grids ( Representation and grid cells Adjacent attributes (the difference in attributes) can be calculated using the following formula: .in, For the number The ant body attribute values ​​of adjacent grid cells, Represents grid cells The numbers of adjacent grid cells.

[0037] grid cells The cumulative difference is obtained by summing the differences in ant body attributes between all eight adjacent grid cells. The calculation formula is: .

[0038] Calculate the grid cell based on the cumulative difference value. coherence index The coherence index is defined as follows: .in, There are a total of 9 ant body attribute values ​​(i.e., grid cells). The maximum value of the ant body attribute values ​​(and its 8 adjacent grid cells) is used to normalize the coherence index. This formula normalizes the coherence index. Normalize to interval When the difference is small, A value close to 1 indicates high coherence in the local area, meaning that the ant-like properties of these grid cells are quite similar; conversely, when the differences are large, A value close to 0 indicates that the local areas are incoherent, meaning that the ant body attributes of these grid cells differ significantly.

[0039] Specifically, for local areas The Chinese number is any grid cell The gradient along the horizontal x-axis is calculated as follows: .in, Let x be the gradient of any grid cell in the horizontal x-axis direction. For the number The ant body attribute values ​​of adjacent grid cells, and adjacent grid cells in the positive horizontal x-axis direction. For the number The ant body attribute values ​​of adjacent grid cells and adjacent grid cells in the negative x-axis direction.

[0040] Specifically, for local areas The Chinese number is any grid cell The gradient in the vertical y-axis direction is calculated as follows: .in, Let be the gradient of any grid cell in the vertical y-axis direction. For the number The ant body attribute values ​​of adjacent grid cells, and adjacent grid cells in the positive y-axis direction vertically. For the number The ant body attribute values ​​of adjacent grid cells and adjacent grid cells in the negative y-axis direction.

[0041] based on Calculate the local gradient magnitude. Wherein, This represents the local gradient magnitude.

[0042] based on Calculate the local gradient direction. Wherein, This represents the local gradient direction.

[0043] based on Calculate the local average gradient direction. Wherein, The direction of the local average gradient. For local area The cumulative value of the gradient in the horizontal x-axis direction for all grid cells within the area. Local area The cumulative value of the gradient in the vertical y-axis direction for all grid cells within the cell.

[0044] based on Calculate the maximum value of the local gradient magnitude. Wherein, This represents the maximum value of the local gradient magnitude.

[0045] For local area The Chinese number is any grid cell Its gradient direction is related to the local range The difference in direction between local average gradients is defined as: .in, For local area The Chinese number is The local gradient direction of the mesh element and the local range The difference between the directions of the local mean gradient within the interior. For local area The Chinese number is The local gradient direction of the grid cells.

[0046] Calculate the directional consistency coefficient based on the differences in direction between local average gradients. : . Directional consistency coefficient This describes the overall consistency of gradient directions within a local window. A value closer to 1 indicates greater consistency in local gradient directions within the window; a value closer to 0 indicates less consistency. The size of the local window can be adjusted to suit different scales of analysis based on the characteristics of the actual data.

[0047] based on Computational grid cells The gradient consistency index. Among them, This represents the normalization of the local gradient magnitude of the grid cell relative to the maximum local gradient magnitude. This design considers both the local gradient direction and magnitude information, and the calculation result is normalized to the [0,1] interval, greatly facilitating the comparison and analysis between different data. In addition, this index has a certain resistance to noise; even in noisy environments, the consistency of gradient direction and the significance of magnitude can still play an effective role.

[0048] S102: Select the size of the fitting window, obtain the spatial weight of each grid cell based on the spatial location of the data, and obtain the quality weight of each grid cell based on the data quality of the grid cell.

[0049] Select the size of the fitting window, obtain the spatial weight of each grid cell based on the spatial location of the data, and obtain the quality weight of each grid cell based on the data quality of the grid cell, including the following:

[0050] Specifically, the size of the fitting window can be That is, 7 grid cells multiplied by 7 grid cells.

[0051] The expression for the spatial weight of a grid cell is: .in, Spatial weights of grid cells, The distance from the center point of the grid cell to the center of the fitting window. The spatial attenuation parameter is defined as the size of the fitting window. , It is an exponential function.

[0052] Specifically, the spatial decay parameter limits the influence of spatial distance on spatial weights, ensuring that grid cells farther from the center of the fitting window have lower weights.

[0053] The expression for the mass weight of a mesh cell is: .in, The mass weight of the grid cell. For grid cells Data quality, To fit the size of the window, To fit all grid cells within the window range The maximum value of data quality.

[0054] S103: Obtain the comprehensive weight based on the spatial weight and the mass weight of each grid cell.

[0055] The overall weight is obtained based on the spatial weight and the mass weight of each grid cell, including...

[0056] The expression for the overall weight is: .in, For comprehensive weighting, Spatial weights of grid cells, This represents the quality weight of the grid cell.

[0057] S104: Construct a bi-objective optimization function that includes fitting accuracy and spatial smoothness, design constraints, use a constrained optimization algorithm to solve for the fitting parameter vector in the bi-objective optimization function, and obtain the optimal fitting parameter vector.

[0058] Specifically, in the process of seismic data processing to achieve crack azimuth estimation, although the preliminary steps effectively improve the robustness and reliability of data fusion by considering the spatial location and data quality of seismic data and adopting an adaptive weight allocation strategy for data points, i.e., the calculation of comprehensive weights, it is difficult to meet the comprehensive requirements of crack azimuth estimation in terms of multi-objective needs and actual constraints by optimizing only a single factor. Therefore, this application constructs a dual-objective optimization function that includes fitting accuracy and spatial smoothness, and designs constraints.

[0059] Constrained optimization algorithms (such as MATLAB's fmincon function or other mature optimization tools) can be used to solve for the fitting parameter vector in the biobjective optimization function, thereby obtaining the optimal fitting parameter vector. .

[0060] The expression for the bi-objective optimization function is: .in, It is a bi-objective optimization function. and For adaptive weighting coefficients, For the fitting accuracy term, For spatial smoothing terms, To fit the parameter vector.

[0061] The adaptive weighting coefficients are obtained through the noise level, which is based on the normalization of the signal-to-noise ratio within the fitting window.

[0062] Specifically, , .in, For noise level, , This represents the maximum signal-to-noise ratio within the fitting window. This represents the current signal-to-noise ratio.

[0063] .in, For data points Error value under the fitted model , Let x be the x-coordinate of the data point on the horizontal plane. The vertical coordinate of the data point is the ordinate in the vertical direction. For vertical layers, For comprehensive weighting, For indexing, This represents the number of data points in the fitting window.

[0064] When choosing ellipse fitting , .in, These are the parameters of the ellipse.

[0065] When choosing linear fitting , .in, These are the parameters of the line.

[0066] .in, For any grid cell The number, This refers to the grid number of the grid cell in the horizontal direction. This refers to the grid number of the grid cell in the vertical direction. For vertical layers, Let be the gradient of the crack azimuth angle in the horizontal direction. This represents the gradient of the crack azimuth angle in the vertical direction.

[0067] Specifically, , .in, For the number The crack azimuth attribute value of adjacent mesh cells, and adjacent mesh cells in the positive x-axis direction in the horizontal direction. For the number The crack azimuth attribute value of adjacent mesh cells, and adjacent mesh cells in the negative x-axis direction in the horizontal direction. For the number The crack azimuth attribute value of adjacent mesh cells, and adjacent mesh cells in the positive y-axis direction vertically. For the number The crack azimuth attribute value of adjacent mesh cells, and adjacent mesh cells in the negative y-axis direction vertically. The length of the grid cell. This represents the width of the grid cell.

[0068] The constraints include ellipse validity and orientation angle continuity.

[0069] The expression for the elliptic validity constraint is: .in, Let be the elliptic validity constraint function. These are the first three parameters of the ellipse. It is a pre-defined, extremely small positive number. , used to ensure that it is strictly an ellipse.

[0070] The expression for the continuity of the direction angle is: .in, It is a function of direction angle continuity. For data points The azimuth of the crack at that location. For data points The azimuth of the crack at that location. The maximum permissible difference in crack azimuth angle set for the cosine. For the neighborhood, The radius of the neighborhood is specifically twice the size of the grid cell. .

[0071] S105: Based on the fitting mode and the optimal fitting parameter vector, obtain the crack azimuth angle at the center of each fitting window. The fitting mode can be either ellipse fitting or line fitting.

[0072] When the fitting mode is elliptic fitting, the optimal fitting parameter vector The corresponding equation of the ellipse is: .

[0073] The crack azimuth is the direction of the major axis of the fitted ellipse.

[0074] based on Determine the radians of the principal axis of the ellipse relative to the horizontal x-axis. Wherein, It is the radian of the principal axis of the ellipse and the x-axis in the horizontal direction.

[0075] Specifically, this application uses Able to avoid using At that time, denominator Direct calculation This will cause a division by zero error. Through parameter separation design, automatic processing can be achieved. In this situation, ensure the results fall into place. Within the correct quadrant.

[0076] based on Convert the radians of the principal axis of the ellipse and the x-axis in the horizontal direction into degrees. Wherein, The angle between the principal axis of the ellipse and the x-axis in the horizontal direction.

[0077] Specifically, the mathematical angle is based on the horizontal x-axis (east-west direction), with counterclockwise rotation being the positive direction. The crack azimuth is based on true north (positive y-axis), with clockwise rotation being the positive direction.

[0078] based on Determine the crack azimuth. Among other things, The azimuth angle of the crack. This is a modulo operation.

[0079] When the fitting mode is linear fitting, the optimal fitting parameter vector The corresponding equation of the line is: .

[0080] The crack azimuth is in the direction of a straight line.

[0081] based on Determine the radians of the line relative to the horizontal x-axis. Wherein, Let x be the radians of the line relative to the x-axis in the horizontal direction. It is the direction vector of the line.

[0082] Specifically, the normal vector of the equation of the line is Its direction is perpendicular to the line itself. Therefore, its direction vector can be represented as... or .

[0083] The first parameter must be the perpendicular component of the direction vector. The second parameter is the horizontal component (axis). (axis). For direction vectors ,correspond .

[0084] based on Convert the radians of the line and the horizontal x-axis into degrees. Wherein, The angle between the straight line and the x-axis in the horizontal direction.

[0085] based on Determine the crack azimuth. Among other things, The azimuth angle of the crack. This is a modulo operation.

[0086] Figure 2 Comparison of azimuth field maps and final crack effects based on different azimuth fitting methods provided in this application embodiment. As can be seen from the figures, the azimuth field data obtained by the traditional method is relatively chaotic in color distribution. The transitions between different color regions are not smooth enough, and there are many local abrupt changes. The azimuth field data obtained by this application has a more orderly and continuous color distribution. The color transitions are more natural, without obvious local abrupt changes. This indicates that the method of this application can better suppress the influence of noise, more accurately reflect the actual azimuth information of the crack, and make the azimuth field data have better spatial consistency and continuity. Discrete cracks generated using the traditional method exhibit a relatively fragmented and discontinuous morphology. The crack direction and connection relationship are not clear enough, with many unreasonable breaks and branches, making it difficult to accurately characterize the actual structural features of the crack. Discrete cracks generated based on the method of this application are more complete and coherent. The crack morphology is more consistent with the actual geological conditions, the direction is clear, the connection relationship is reasonable, and it can better reflect the actual distribution and structural features of the crack.

[0087] Figure 3 This is a field map provided in this application embodiment, dividing ant body attribute data into grid cells. The background is a light blue grid, and crack areas are represented by dark blue. A black vertical line in the map is marked as a horizontal well, indicating the location of the horizontal well. Three red circles are marked in the map, labeled 1, 2, and 3, indicating the three locations that need to be compared.

[0088] Figure 4 This is a comparison chart of well logging data and ant body identification results provided in an embodiment of this application. The horizontal axis represents the position (1, 2, 3), corresponding to... Figure 3 The three marker points are shown. The vertical axis represents the fracture azimuth (in degrees). Each method is represented by a different colored bar: orange for the traditional method, green for well logging data, and purple for the method of this application. At positions 1, 2, and 3, the fracture azimuth interpreted by the method of this application is closer to the well logging data, while the interpretation results of the traditional method deviate significantly from the well logging data.

[0089] This application also provides a crack azimuth angle fitting device 500 based on seismic data, such as... Figure 5 As shown, the device includes: an evaluation module 501, a comprehensive weight acquisition module 502, a solution module 503, and a crack azimuth acquisition module 504.

[0090] The evaluation module 501 is used to acquire ant body attribute data, divide the ant body attribute data into grid cells, define the size of the local window, and evaluate the grid quality within the local window using the grid quality evaluation function to obtain the data quality of the grid cells.

[0091] The comprehensive weight acquisition module 502 is used to select the size of the fitting window, acquire the spatial weight of each grid cell based on the spatial location of the data, and acquire the quality weight of each grid cell based on the data quality of the grid cell. The comprehensive weight is then acquired based on the spatial weight and the quality weight of each grid cell.

[0092] The solver module 503 is used to construct a bi-objective optimization function that includes fitting accuracy and spatial smoothness, design constraints, and use a constraint optimization algorithm to solve for the fitting parameter vector in the bi-objective optimization function to obtain the optimal fitting parameter vector.

[0093] The crack azimuth angle acquisition module 504 is used to acquire the crack azimuth angle at the center of each fitting window based on the fitting mode and the optimal fitting parameter vector. The fitting mode is either elliptical fitting or linear fitting.

[0094] Some modules in the apparatus described in this application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0095] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. When implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0096] The methods, apparatus, or modules described in this application can be implemented in a computer-readable program code manner. The controller can be implemented in any suitable manner, for example, as a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of a memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code manner, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included within it for implementing various functions can also be considered as structures within the hardware component. Alternatively, the device used to implement various functions can be viewed as either a software module that implements the method or a structure within a hardware component.

[0097] like Figure 6 As shown in the figure, this application embodiment also provides a crack azimuth angle fitting server based on seismic data, including a memory 601 and a processor 602; the memory 601 is used to store computer-executable instructions; the processor 602 is used to execute computer-executable instructions to implement the crack azimuth angle fitting method based on seismic data described above in this application embodiment.

[0098] This application also provides a computer-readable storage medium storing executable instructions. When a computer executes the executable instructions, it can implement the crack azimuth fitting method based on seismic data described above in this application.

[0099] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary hardware. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, or it can be embodied in the process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in the embodiments of this application.

[0100] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this application can be used in numerous general-purpose or special-purpose computer system environments or configurations.

[0101] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A seismic data based fracture azimuth fitting method, characterized in that, The method comprises the following steps: acquiring ant body attribute data, dividing the ant body attribute data into grid cells, defining the size of a local window, and performing grid quality evaluation in the local window through a grid quality evaluation function to obtain data quality of the grid cells; selecting the size of a fitting window, acquiring the spatial weight of each grid cell based on the spatial position of the data, and acquiring the quality weight of each grid cell based on the data quality of the grid cell; acquiring a comprehensive weight based on the spatial weight of each grid cell and the quality weight of each grid cell; constructing a double-target optimization function containing fitting accuracy and spatial smoothness, designing a constraint condition, solving a fitting parameter vector in the double-target optimization function by using a constraint optimization algorithm, and obtaining an optimal fitting parameter vector; acquiring a fracture azimuth angle at the center of each fitting window based on a fitting mode and the optimal fitting parameter vector; wherein the fitting mode is elliptical fitting or linear fitting; the constraint condition includes elliptical validity and direction angle continuity; The expression of the ellipse validity constraint is: ; where, is the ellipse validity constraint function, are the first three parameters of the ellipse, is a pre-set minimum positive number; The expression for the continuity of the direction angle is: ;in, It is a function of direction angle continuity. For data points The azimuth of the crack at that location. For data points The azimuth of the crack at that location. The maximum permissible difference in crack azimuth angle set for the cosine. For the neighborhood, The radius of the neighborhood is specifically twice the size of the grid cell.

2. The seismic data based fracture azimuth fitting method of claim 1, wherein, the grid quality evaluation in the local window through the grid quality evaluation function to obtain the data quality of the grid cells comprises: The expression of the grid quality evaluation is: ; wherein, is the data quality of the grid unit , , and are evaluation weights, is the signal-to-noise ratio of the grid unit , is the coherence index of the grid unit , is the gradient consistency index of the grid unit .

3. The seismic data based fracture azimuth fitting method of claim 2, wherein, the selection of the size of the fitting window, the acquisition of the spatial weight of each grid cell based on the spatial position of the data, and the acquisition of the quality weight of each grid cell based on the data quality of the grid cell comprises: The expression for the spatial weight of a grid cell is: ; where, is the spatial weight of a grid cell, is the distance from the center point of the grid cell to the center of the fitting window, is a spatial decay parameter defined as , is an exponential function; The expression of the quality weight of the grid cell is: ; wherein is the quality weight of the grid cell, is the data quality of the grid cell is the data quality of the grid cell is the size of the fitting window, is the maximum value of the data quality of all grid cells within the fitting window range.

4. The seismic data based fracture azimuth fitting method of claim 3, wherein, the acquisition of the comprehensive weight based on the spatial weight of each grid cell and the quality weight of each grid cell comprises: The expression of the comprehensive weight is: ; wherein, is the comprehensive weight, is the spatial weight of the grid unit, is the quality weight of the grid unit.

5. The seismic data based fracture azimuth fitting method of claim 4, wherein, The expression of the dual-objective optimization function is: ; wherein, is a dual-objective optimization function, and is an adaptive weight coefficient, is a fitting accuracy term, is a spatial smoothing term, is a fitting parameter vector; the adaptive weight coefficient is acquired based on a noise level normalized based on the signal-to-noise ratio in the fitting window; ; wherein, is a data point is an error value of a data point under a fitted model, , is a horizontal coordinate of a data point on a horizontal plane, is a vertical coordinate of a data point on a vertical direction, is a vertical horizon, is a comprehensive weight, is an index, is a number of data points in a fitting window; When ellipse fit is selected, , ; wherein, are parameters of the ellipse; When a straight line fit is selected, , ; wherein, are parameters of a straight line; ; wherein is the number of any grid cell , is the grid number of the grid cell in the horizontal plane direction, is the grid number of the grid cell in the vertical direction, is the vertical level, is the gradient of the fracture azimuth angle in the horizontal direction, is the gradient of the fracture azimuth angle in the vertical direction.

6. The seismic data based fracture azimuth fitting method of claim 5, wherein, When the fitting mode is an ellipse fitting, the ellipse equation corresponding to the optimal fitting parameter vector is: ; the fracture azimuth angle is the direction of the major axis of the fitted ellipse; based on determining an arc of the major axis of the ellipse with the x-axis in the horizontal direction; wherein, is an arc of the major axis of the ellipse with the x-axis in the horizontal direction; based on converting the arc of the major axis of the ellipse to an angle with the horizontal x-axis; wherein, is the angle of the major axis of the ellipse with the horizontal x-axis; based on determining a fracture azimuth angle; wherein, is the fracture azimuth angle, is a modulo operation.

7. The seismic data based fracture azimuth fitting method of claim 5, wherein, When the fitting mode is a linear fitting, the linear equation corresponding to the optimal fitting parameter vector is: ; the fracture azimuth angle is the direction of the straight line; based on determining the radian of the straight line with the x-axis in the horizontal direction; wherein, is the radian of the straight line with the x-axis in the horizontal direction, is the direction vector of the straight line; based on converts the straight line and the radian of the x-axis in the horizontal direction into an angle; wherein, is the angle of the straight line and the x-axis in the horizontal direction; based on determining a fracture azimuth angle; wherein, is the fracture azimuth angle, is a modulo operation.

8. An apparatus for fitting fracture azimuths based on seismic data, characterized by, The method comprises the following steps: an evaluation module is configured to acquire ant body attribute data, divide the ant body attribute data into grid cells, define the size of a local window, and perform grid quality evaluation in the local window through a grid quality evaluation function to obtain data quality of the grid cells; a comprehensive weight acquisition module is configured to select the size of a fitting window, acquire the spatial weight of each grid cell based on the spatial position of the data, and acquire the quality weight of each grid cell based on the data quality of the grid cell; acquire a comprehensive weight based on the spatial weight of each grid cell and the quality weight of each grid cell; a solving module is configured to construct a double-target optimization function containing fitting accuracy and spatial smoothness, design a constraint condition, solve a fitting parameter vector in the double-target optimization function by using a constraint optimization algorithm, and obtain an optimal fitting parameter vector; a fracture azimuth angle acquisition module is configured to acquire a fracture azimuth angle at the center of each fitting window based on a fitting mode and the optimal fitting parameter vector; wherein the fitting mode is elliptical fitting or linear fitting; the constraint condition includes elliptical validity and direction angle continuity; The expression of the ellipse validity constraint is: ; where, is the ellipse validity constraint function, are the first three parameters of the ellipse, is a pre-set minimum positive number; The expression for the continuity of the direction angle is: ;in, It is a function of direction angle continuity. For data points The azimuth of the crack at that location. For data points The azimuth of the crack at that location. The maximum permissible difference in crack azimuth angle set for the cosine. For the neighborhood, The radius of the neighborhood is specifically twice the size of the grid cell.

Images