Crack prediction method, electronic equipment, storage medium and device

Abstract

The invention discloses a crack prediction method and device, electronic equipment and a storage medium. The method comprises the steps of performing dip angle attribute extraction based on a seismic data volume to obtain a dip angle data volume; acquiring a diffusion tensor edge-preserving filtering data volume by adopting a diffusion tensor edge-preserving filtering algorithm based on the inclination angle data volume; calculating volume curvature attributes of different scales by adopting a Gaussian filter of a spatial wave number domain after Fourier series expansion based on the diffusion tensor edge-preserving filtering data volume; and crack prediction is carried out based on volume curvature attributes of different scales. The breakpoint information of the inclination angle data volume is calculated through the diffusion tensor edge-preserving filtering algorithm, the diffusion tensor edge-preserving filtering data volume is obtained, the Gaussian filter of the spatial wave number domain after Fourier series expansion is applied to the diffusion tensor edge-preserving filtering data volume, the geologic body edge and crack details can be protected, meanwhile, noise is suppressed, and the accuracy of the geologic body is improved. Edge features are highlighted, crack prediction by the volume curvature technology is further improved, and fine description of cracks is facilitated.

Description

Technical Field

[0001] This invention belongs to the field of geophysical exploration technology, and more specifically, relates to a fracture prediction method, electronic device, storage medium, and apparatus. Background Technology

[0002] Curvature attributes can improve the unclear fault characterization caused by inaccurate migrations, facilitating conventional structural interpretation (Ericsson, 1998). Furthermore, based on the relationship between stratum deformation and curvature, it can effectively characterize structural features such as faults and fractures in seismic data (Roberts, 2001). However, initially, curvature attributes relied on manually selecting seismic horizons for calculations, and the accuracy of these horizons severely limited the accuracy of curvature attributes, significantly reducing the reliability of the calculation results. The introduction of volume curvature attributes shifted curvature attribute research from two-dimensional to multi-dimensional, improving the accuracy of curvature in predicting fractures. This method does not require the involvement of horizons and employs a fractal wavenumber Fourier transform algorithm to achieve multi-scale volume curvature attribute prediction (A1-Dosssary and Marfurt, 2006). In recent years, maximum positive curvature and minimum negative curvature have been considered the best curvature attributes for characterizing structural geometry (Chopra and Marfurt, 2007). Wang Shixin (2012) proposed a correction formula for secondary scanning, combined with a multi-window edge-preserving calculation method to improve the accuracy of curvature calculation. Qin Honglei et al. (2016) demonstrated a positive correlation between the principal curvature of a structure and the fracture development zone. Chen Zhigang et al. (2020) used the rose diagram of the curvature attribute volume to calculate and predict the direction of fracture development. Zhang Wen et al. (2023) improved the computational efficiency of curvature attributes based on multiple recursive filters. Du Xin et al. (2024) used the high and low frequency end information extracted by the Contourlet transform technique to fuse curvature, Radon transform, and ant body attributes to achieve accurate prediction of multi-scale fracture zones. Currently, many scholars have improved the estimation of curvature attributes by improving the dip angle, but this is not accurate enough. There is a lack of spatial distribution of abnormal information of curvature at different scales, and the curvature attribute characterization for fracture prediction is relatively simple, resulting in low accuracy of fracture information description or even failure to identify fracture information.

[0003] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to propose a crack prediction method, electronic device, storage medium, and apparatus to solve the problem of low accuracy in crack information description or even inability to identify crack information, thereby improving the accuracy of crack prediction.

[0005] To achieve the above objectives, the present invention proposes a crack prediction method, an electronic device, a storage medium, and an apparatus.

[0006] According to a first aspect of the present invention, a crack prediction method is proposed, comprising:

[0007] Dip attribute extraction is performed on the seismic data volume to obtain the dip data volume;

[0008] Based on the tilt angle data volume, the diffusion tensor edge-preserving filtering algorithm is used to obtain the diffusion tensor edge-preserving filtered data volume.

[0009] Based on the diffusion tensor edge-preserving filtered data volume, the Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales.

[0010] Crack prediction is carried out based on the volume curvature properties at different scales.

[0011] Optionally, the dip angle attribute extraction based on the seismic data volume to obtain the dip angle data volume includes:

[0012] The instantaneous frequency of the seismic data volume and the instantaneous wavenumber of the seismic data volume along the main survey line are calculated based on the data volume.

[0013] Calculate the first apparent dip angle of the seismic data volume along the main survey line based on the instantaneous frequency and the instantaneous wavenumber;

[0014] Calculate the second apparent tilt angle of the seismic data volume along the connecting lateral line based on the first apparent tilt angle;

[0015] The tilt angle data volume is obtained based on the first and second tilt angles.

[0016] Optionally, the calculation of volume curvature properties at different scales based on the diffusion tensor edge-preserving filtered data volume using a Gaussian filter in the spatial wavenumber domain after Fourier series expansion includes:

[0017] Based on the grid data of the diffusion tensor edge-preserving filter data volume, the least squares method is used to fit the trend surface, and a quadratic fitting equation is obtained.

[0018] A Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used. The wavelength factor is used to change the derivative operator of the quadratic surface coefficients of the quadratic fitting equation. In order to keep the phase spectrum unchanged, the wavenumber selectivity is improved by changing the amplitude spectrum of the complex wavenumber.

[0019] The coefficients of the quadratic surface are calculated based on the derivative operator, and then the volume curvature properties at different scales are calculated.

[0020] Optionally, the quadratic fitting equation is:

[0021] (x, y) = ax 2 +by 2 +cxy+dx+ey+f;

[0022] Where a, b, c, d, e, and f are the coefficients of the quadratic surface, x is the abscissa, and y is the ordinate.

[0023] Optionally, the expression for the derivative operator is:

[0024] E = F -1 [-ik a F(u)T(k)];

[0025] Where E is the derivative operator, T(K) is the Gaussian filter, α is the wavelength factor, k is the wavenumber, and i is the... F represents the Fourier series expansion, F(u) represents the Fourier transform of the seismic data, F -1 This is the inverse Fourier transform.

[0026] Optionally, the expression for the Gaussian filter is:

[0027]

[0028] Where η is the direction of the seismic data, β is the adaptive adjustment factor corresponding to the wave number k, and β = β(k).

[0029] Optionally, the volume curvature attribute includes:

[0030] Maximum curvature, minimum curvature, mean curvature, curvature, Gaussian curvature, maximum positive curvature, minimum negative curvature, and normal curvature.

[0031] According to a second aspect of the present invention, a crack prediction device is provided, comprising:

[0032] The extraction module is used to extract dip attributes from seismic data volumes to obtain dip data volumes;

[0033] The acquisition module is used to acquire the diffusion tensor edge-preserving filtering data volume based on the tilt angle data volume using the diffusion tensor edge-preserving filtering algorithm;

[0034] The calculation module is used to calculate the volume curvature properties at different scales based on the diffusion tensor edge-preserving filtered data volume after Fourier series expansion using a Gaussian filter in the spatial wavenumber domain.

[0035] The prediction module is used to predict cracks based on volume curvature properties at different scales.

[0036] According to a third aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0037] At least one processor; and,

[0038] A memory communicatively connected to the at least one processor; wherein,

[0039] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the crack prediction methods described in the first aspect.

[0040] According to a fourth aspect of the invention, a non-transitory computer-readable storage medium is provided, which stores computer instructions for causing a computer to perform the crack prediction method described in any of the first aspects.

[0041] The beneficial effects of this invention are as follows: This invention calculates the breakpoint information of the dip angle data volume through the diffusion tensor edge-preserving filtering algorithm to obtain the diffusion tensor edge-preserving filtered data volume. By applying the Gaussian filter of the spatial wavenumber domain after Fourier series expansion to the diffusion tensor edge-preserving filtered data volume, it can protect the geological body edge and fracture details while suppressing noise, highlighting edge features, and further improving the volume curvature technology for fracture prediction, which is beneficial to the fine characterization of fractures.

[0042] The system of the present invention has other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description

[0043] The above and other objects, features and advantages of the present invention will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.

[0044] Figure 1 A flowchart illustrating the steps of a crack prediction method according to the present invention is shown.

[0045] Figure 2 A flowchart illustrating the steps of a crack prediction method according to Embodiment 2 of the present invention is shown.

[0046] Figure 3 A schematic diagram of the original seismic data profile according to Embodiment 2 of the present invention is shown.

[0047] Figure 4 A schematic diagram of the tilt data volume cross-section according to Embodiment 2 of the present invention is shown.

[0048] Figure 5A schematic diagram of a diffusion tensor edge-preserving filter data volume slice according to Embodiment 2 of the present invention is shown.

[0049] Figure 6 a and Figure 6 b shows a schematic diagram of a conventional curvature property slice and a multi-scale volume curvature property slice with Fourier expansion according to Embodiment 2 of the present invention. Detailed Implementation

[0050] The invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0051] like Figure 1 As shown, a crack prediction method according to the present invention includes:

[0052] Dip attribute extraction is performed on the seismic data volume to obtain the dip data volume;

[0053] The diffusion tensor edge-preserving filtering algorithm is used to obtain the diffusion tensor edge-preserving filtering data volume based on the tilt angle data volume;

[0054] Based on the diffusion tensor edge-preserving filter data volume, the Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales.

[0055] Crack prediction is carried out based on the volume curvature properties at different scales.

[0056] Specifically, firstly, the original 3D seismic data is input and the dip volume is calculated using the complex trace analysis method to enhance the consistency and continuity of seismic reflection phase axes and highlight fault boundary characteristics. Then, based on the dip volume, a diffusion tensor edge-preserving filtering algorithm is used to obtain a diffusion tensor edge-preserving filtered volume. Under precise direction field driving, accurate guide horizons are calculated point-by-point, and the diffusion tensor edge-preserving filtering algorithm is applied to highlight fault information. Next, based on the diffusion tensor edge-preserving filtered volume, a Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales. Using an adaptive adjustment factor, while maintaining the phase spectrum, the amplitude spectrum of the complex wavenumber is changed to enhance the wave... This invention utilizes the characteristic of number selection to effectively preserve the linear structure of strata. By inputting dip components from 3D seismic data in different directions, volume curvature attributes at different scales can be calculated. Finally, fracture prediction is performed based on the volume curvature attributes at different scales. The innovation of this invention lies in the diffusion tensor edge-preserving filtering algorithm for calculating breakpoint information. At the same time, a Gaussian filter in the spatial wavenumber domain after Fourier series expansion is applied to the dip data volume. This can protect the geological body edges and fracture details while suppressing noise, highlighting edge features, and facilitating fine fracture characterization. This invention employs a Gaussian filter in the spatial wavenumber domain after Fourier series expansion and introduces a spatial wavenumber domain differential operator to further improve the volume curvature technology for fracture detection and identification, achieving high-resolution fracture characterization.

[0057] In one example, dip attribute extraction is performed based on seismic data volume, resulting in a dip data volume including:

[0058] The instantaneous frequency of the seismic data volume and the instantaneous wavenumber of the seismic data volume along the main survey line are calculated based on the data volume.

[0059] The first apparent dip angle of the seismic data volume along the main survey line is calculated based on the instantaneous frequency and instantaneous wavenumber;

[0060] The second apparent dip angle of the seismic data volume along the connecting lateral line is calculated based on the first apparent dip angle.

[0061] The tilt angle data volume is obtained based on the first and second tilt angles.

[0062] Specifically, the dip data volume is calculated using the complex trace analysis method to enhance the consistency and continuity of seismic reflection phase axes and highlight fault boundary characteristics. First, the instantaneous frequency of the seismic data is calculated:

[0063]

[0064] Where φ represents the instantaneous phase; H[u] represents the Hilbert transform of the seismic data volume; and t represents time. and It can be obtained through finite difference or Fourier transform. Further calculation of the instantaneous wavenumber k along the main seismic line for the seismic data volume u(t) is also possible. x :

[0065]

[0066] The apparent dip angle p of the seismic data volume u(t) along the main survey line:

[0067] p = k x / ω(t) (3)

[0068] Similarly, the apparent dip angle q in the direction of the connecting lateral line can be obtained, and the strike γ and true dip angle θ can be solved simultaneously by substituting these values:

[0069] γ=arctan(q / p) (4)

[0070] p = arctan(tanθ × sinδ) (5)

[0071] q=arctan(tanθ×sin(90+δ) (6)

[0072] The dip data volume is obtained based on the apparent dip angles p and q, where δ is the angle between the strike of the strata and the observation line.

[0073] In one example, the volume curvature properties at different scales are calculated using a Gaussian filter in the spatial wavenumber domain after Fourier series expansion based on the diffusion tensor edge-preserving filter data volume, including:

[0074] The least squares method is used to fit the trend surface of the data volume based on the edge-preserving filter data volume of the diffusion tensor to obtain the quadratic fitting equation.

[0075] A Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used. The wavelength factor is used to change the derivative operator of the quadratic surface coefficients of the quadratic fitting equation. In order to keep the phase spectrum unchanged, the wavenumber selectivity is improved by changing the amplitude spectrum of the complex wavenumber.

[0076] The coefficients of the quadratic surface are calculated based on the derivative operator, and then the volume curvature properties at different scales are calculated.

[0077] Specifically, first, the grid data is fitted with a trend surface using the least squares method:

[0078] (x, y) = ax 2 +by 2 +cxy+dx+ey+f (12)

[0079] Formulas for obtaining various curvatures using the quadratic surface coefficients (a, b, c, d, e, f) in equation (12), such as the maximum curvature K. max Minimal curvature Kmin Mean curvature K mean curvature K m Gaussian curvature K g and the maximum positive curvature K + and minimum negative curvature K - (Equations 13-19);

[0080]

[0081] The curvatures along the (x,y) direction are orthogonal and are all normal curvatures (K). x K y ),Right now

[0082]

[0083] In the formula, p′ and q′ are the components of the apparent tilt angle, and the formula for calculating the coefficients of the quadratic fitting equation in the volume curvature attribute is:

[0084]

[0085] Replace the operator of the first derivative in the above ae with the fractional derivative. and

[0086]

[0087] In the formula, F represents the Fourier series expansion; kx represents the wavenumber along the x-axis; and i is... α is the wavelength factor. The larger the value of α, the shorter the corresponding earthquake wavelength, and the smaller the value of α, the longer the corresponding earthquake wavelength.

[0088] A Gaussian filter in the spatial wavenumber domain, derived from Fourier series expansion, is employed. By utilizing an adaptive adjustment factor, the phase spectrum remains constant while the amplitude spectrum of the complex wavenumber is altered, thereby enhancing wavenumber selectivity and effectively preserving the linear structure of the formation.

[0089] E = F -1 [-ik a F(u)T(k)] (22)

[0090] In the formula, E is the derivative operator, and T(k) is the Gaussian filter. When the time window is ε, Where β=β(k), it can also be expressed as:

[0091]

[0092] By inputting the dip components of 3D seismic data in different directions, the volume curvature properties at different scales can be calculated.

[0093] In one example, the quadratic fitting equation is:

[0094] (x, y) = ax 2 +by 2 +cxy+dx+ey+f;

[0095] Where a, b, c, d, e, and f are the coefficients of the quadratic surface, x is the abscissa, and y is the ordinate.

[0096] In one example, the expression for the derivative operator is:

[0097] E = F -1 [-ik a F(u)T(k)];

[0098] Where E is the derivative operator, T(k) is the Gaussian filter, α is the wavelength factor, k is the wavenumber, and i is the... F represents the Fourier series expansion, F(u) represents the Fourier transform of the seismic data, F -1 This is the inverse Fourier transform.

[0099] In one example, the expression for a Gaussian filter is:

[0100]

[0101] Where η is the direction of the seismic data, β is the adaptive adjustment factor corresponding to the wave number k, and β = β(k).

[0102] In one example, the volume curvature property includes:

[0103] Maximum curvature, minimum curvature, mean curvature, curvature, Gaussian curvature, maximum positive curvature, minimum negative curvature, and normal curvature.

[0104] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the invention. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present invention can be combined with each other.

[0105] Example 1

[0106] This embodiment provides a crack prediction method, including:

[0107] Dip attribute extraction is performed on the seismic data volume to obtain the dip data volume;

[0108] The diffusion tensor edge-preserving filtering algorithm is used to obtain the diffusion tensor edge-preserving filtering data volume based on the tilt angle data volume;

[0109] Based on the diffusion tensor edge-preserving filter data volume, the Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales.

[0110] Crack prediction is carried out based on the volume curvature properties at different scales.

[0111] The diffusion tensor edge-preserving filtering data volume is obtained based on the tilt angle data volume using the diffusion tensor edge-preserving filtering algorithm, including:

[0112] The instantaneous frequency of the seismic data volume and the instantaneous wavenumber of the seismic data volume along the main survey line are calculated based on the data volume.

[0113] The first apparent dip angle of the seismic data volume along the main survey line is calculated based on the instantaneous frequency and instantaneous wavenumber;

[0114] The second apparent dip angle of the seismic data volume along the connecting lateral line is calculated based on the first apparent dip angle.

[0115] The tilt angle data volume is obtained based on the first and second tilt angles.

[0116] Based on the diffusion tensor edge-preserving filter, the Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales, including:

[0117] The least squares method is used to fit the trend surface of the data volume based on the edge-preserving filter data volume of the diffusion tensor to obtain the quadratic fitting equation.

[0118] A Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used. The derivative operator of the quadratic surface coefficients of the quadratic fitting equation is changed by the wavelength factor to keep the phase spectrum unchanged while improving the wavenumber selectivity by changing the amplitude spectrum of the complex wavenumber. The quadratic surface coefficients are calculated based on the derivative operator, and then the volume curvature properties at different scales are calculated.

[0119] The quadratic fitting equation is: (x, y) = ax 2 +by 2 +cxy+dx+ey+f; where a, b, c, d, e, and f are the coefficients of the quadratic surface, x is the abscissa, and y is the ordinate.

[0120] The expression for the derivative operator is: E = F -1 [-ik a F(u)T(k)]; where E is the derivative operator, T(k) is the Gaussian filter, α is the wavelength factor, k is the wavenumber, and i is the wavelength factor. F represents the Fourier series expansion, F(u) represents the Fourier transform of the seismic data, and F- 1 Inverse Fourier transform;

[0121] The expression for a Gaussian filter is:

[0122]

[0123] Where η is the direction of the seismic data, β is the adaptive adjustment factor corresponding to the wave number k, and β = β(k).

[0124] Body curvature properties include:

[0125] Maximum curvature, minimum curvature, mean curvature, curvature, Gaussian curvature, maximum positive curvature, minimum negative curvature, and normal curvature.

[0126] Example 2

[0127] like Figure 2 As shown, this embodiment provides a crack prediction method, including:

[0128] Curvature technology can identify discontinuities such as faults, fractures, and ridges based on the degree of structural curvature. However, conventional curvature methods are relatively simple and lack accurate spatial descriptions of fractures at different scales. This paper fully utilizes a boundary-preserving filtering algorithm driven by a precise direction field to calculate discontinuity information and extracts three-dimensional multi-scale volume curvature based on dip-guided filtering data. This effectively preserves the linear structure of the strata and provides a new approach and solution for the accurate characterization of fractures.

[0129] Step 1: Input the raw 3D seismic data u. This embodiment uses post-stack seismic data as the initial data, such as... Figure 3 As shown.

[0130] Step 2: Using the complex trace analysis method, extract the dip angle attribute from the seismic data volume of the work area to obtain the dip angle data volume, such as... Figure 4 As shown.

[0131] The dip data volume is calculated using the complex trace analysis method, which enhances the consistency and continuity of seismic reflection phase axes and highlights fault boundary characteristics.

[0132] Calculate the instantaneous frequency of seismic data:

[0133]

[0134] Where φ represents the instantaneous phase; H[u] represents the Hilbert transform of the seismic data volume; and t represents time. and It can be obtained through finite difference or Fourier transform.

[0135] Further calculation of the instantaneous wavenumber k of the seismic data volume u(t) along the main survey line x :

[0136]

[0137] The apparent dip angle p of the seismic data volume u(t) along the main survey line:

[0138] p = kx / ω(t) (3)

[0139] Similarly, the apparent dip angle q in the direction of the connecting lateral line can be obtained, and the strike γ and true dip angle θ can be solved simultaneously by substituting these values:

[0140] γ=arctan(q / p) (4)

[0141] p = arctan(tanθ × sinδ) (5)

[0142] q=arctan(tanθ×sin(90+δ) (6)

[0143] Extract the tilt data volume based on the apparent tilt angles p and q.

[0144] Step 3: Substitute the dip data volume into the equation to obtain the diffusion tensor D, and then substitute D as the diffusion coefficient into the nonlinear anisotropic diffusion equation. Use finite differences to replace differentials, and filter the anisotropic direction information and parameters to obtain the edge-preserving filtered data volume seismic profile of the diffusion tensor, as shown below. Figure 5 As shown. Driven by a precise direction field, the accurate guiding layer position is calculated point by point, and the diffusion tensor edge-preserving filtering algorithm is applied to highlight the breakpoint information.

[0145] The specific algorithm is as follows:

[0146] c1=ω,λ1=λ2 (7)

[0147]

[0148] Where c1 and c2 are diffusion tensor values, c1 represents the direction of greatest change in diffusion degree, and c2 represents the direction of least change in diffusion degree; λ1 and λ2 are structure tensor values, representing the directions of greatest and least change, respectively; 0 < ω < 10, α is the diffusion intensity along the image gradient direction with the greatest change, and is a controllable parameter; k is a consistency parameter, a threshold adjusted according to the actual data. The tilt data volume u calculated in step two is used to construct a diffusion tensor D with structural characteristics, and D is substituted as the diffusion coefficient to obtain the anisotropic diffusion equation:

[0149]

[0150] In the formula, w1 and w2 represent the gradient changes, and are the directions of maximum and minimum gradient changes, respectively; τ represents the diffusion time. Combining the calculated anisotropic direction information and parameters for filtering yields the diffusion tensor edge-preserving filtered data volume.

[0151] Step 4: Curvature Attribute Extraction Based on Tilt-Guided Filtering. Multiple curvature attributes are obtained based on the trend surface equation method. The coefficients in the quadratic fitting equation are replaced with fractional derivatives. A Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to change the amplitude spectrum of the complex wavenumber, thereby improving wavenumber selectivity and effectively preserving the linear structure of the strata. This yields three-dimensional volumetric curvature attribute slices at different scales, such as... Figure 6 As shown in b. Figure 6 Figure a shows a slice of traditional curvature properties. Figure 6 Figure b shows a multi-scale volume curvature property slice obtained using the method of this embodiment. It can be seen that the multi-scale volume curvature property slice obtained using the method of this embodiment has better resolution, achieving high-resolution characterization of cracks.

[0152] Crack prediction based on volume curvature is achieved using Fourier series expansion. First, a trend surface is fitted to the mesh data using the least squares method:

[0153] (x, y) = ax 2 +by 2 +cxy+dx+ey+f (12)

[0154] Formulas for obtaining various curvatures using the quadratic surface coefficients (a, b, c, d, e, f) in equation (12), such as the maximum curvature K. max Minimal curvature K min Mean curvature K mean curvature K m Gaussian curvature K g and the maximum positive curvature K + and minimum negative curvature K - (Equations 13-19).

[0155]

[0156] The curvatures along the (x,y) direction are orthogonal and are all normal curvatures (K). x K y ),Right now

[0157]

[0158] In the formula, p′ and q′ are the components of the apparent tilt angle, and the formula for calculating the coefficients of the quadratic fitting equation in the volume curvature attribute is:

[0159]

[0160] Replace the operator of the first derivative in the above ae with the fractional derivative. and

[0161]

[0162] In the formula, F represents the Fourier series expansion; k x Represents the wave number along the x-axis; i is α is the wavelength factor. The larger the value of α, the shorter the corresponding earthquake wavelength, and the smaller the value of α, the longer the corresponding earthquake wavelength.

[0163] A Gaussian filter in the spatial wavenumber domain, derived from Fourier series expansion, is employed. By utilizing an adaptive adjustment factor, the phase spectrum remains constant while the amplitude spectrum of the complex wavenumber is altered, thereby enhancing wavenumber selectivity and effectively preserving the linear structure of the formation.

[0164] E = F -1 [-ik a F(u)T(k)] (22)

[0165] In the formula, E is the derivative operator, and T(k) is the Gaussian filter. When the time window is ε, Where β=β(k), it can also be expressed as:

[0166]

[0167] Where η is the direction of the seismic data, and β is the adaptive adjustment factor corresponding to the wave number k;

[0168] By inputting the dip components of 3D seismic data in different directions, the volume curvature properties at different scales can be calculated.

[0169] This embodiment calculates breakpoint information using a diffusion tensor edge-preserving filtering algorithm. Simultaneously, by applying a Gaussian filter in the spatial wavenumber domain (derived from Fourier series expansion) to the dip angle data volume, it can preserve the geological body edges and fracture details while suppressing noise, highlighting edge features, and facilitating fine fracture characterization. This embodiment employs a Gaussian filter in the spatial wavenumber domain (derived from Fourier series expansion) and introduces a spatial wavenumber domain differential operator to further improve the volume curvature technique for fracture detection and identification, exhibiting good resolution and noise resistance, achieving high-resolution fracture characterization.

[0170] Example 3

[0171] This embodiment provides a crack prediction device, including:

[0172] The extraction module is used to extract dip attributes from seismic data volumes to obtain dip data volumes;

[0173] The acquisition module is used to acquire the diffusion tensor edge-preserving filtering data volume based on the tilt angle data volume using the diffusion tensor edge-preserving filtering algorithm;

[0174] The calculation module is used to calculate the volume curvature properties at different scales based on the Gaussian filter in the spatial wavenumber domain after Fourier series expansion of the diffusion tensor edge-preserving filtered data volume.

[0175] The prediction module is used to predict cracks based on volume curvature properties at different scales.

[0176] Example 4

[0177] This disclosure also provides an electronic device, which includes:

[0178] At least one processor; and,

[0179] A memory communicatively connected to the at least one processor; wherein,

[0180] The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the crack prediction method in Embodiment 1.

[0181] An electronic device according to embodiments of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0182] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory.

[0183] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0184] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0185] Example 5

[0186] This disclosure provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the crack prediction method in Embodiment 1.

[0187] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods described in the foregoing embodiments of the present disclosure are performed.

[0188] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0189] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A crack prediction method, characterized in that, include: Dip attribute extraction is performed on the seismic data volume to obtain the dip data volume; Based on the tilt angle data volume, the diffusion tensor edge-preserving filtering algorithm is used to obtain the diffusion tensor edge-preserving filtered data volume. Based on the diffusion tensor edge-preserving filtered data volume, the Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used to calculate the volume curvature properties at different scales. Crack prediction is carried out based on the volume curvature properties at different scales.

2. The crack prediction method according to claim 1, characterized in that, The dip angle attribute extraction based on the seismic data volume yields the following dip angle data volume: The instantaneous frequency of the seismic data volume and the instantaneous wavenumber of the seismic data volume along the main survey line are calculated based on the data volume. Calculate the first apparent dip angle of the seismic data volume along the main survey line based on the instantaneous frequency and the instantaneous wavenumber; Calculate the second apparent tilt angle of the seismic data volume along the connecting lateral line based on the first apparent tilt angle; The tilt angle data volume is obtained based on the first and second tilt angles.

3. The crack prediction method according to claim 1, characterized in that, The calculation of volume curvature properties at different scales based on the diffusion tensor edge-preserving filtered data volume using a Gaussian filter in the spatial wavenumber domain after Fourier series expansion includes: Based on the grid data of the diffusion tensor edge-preserving filter data volume, the least squares method is used to fit the trend surface, and a quadratic fitting equation is obtained. A Gaussian filter in the spatial wavenumber domain after Fourier series expansion is used. The wavelength factor is used to change the derivative operator of the quadratic surface coefficients of the quadratic fitting equation. In order to keep the phase spectrum unchanged, the wavenumber selectivity is improved by changing the amplitude spectrum of the complex wavenumber. The coefficients of the quadratic surface are calculated based on the derivative operator, and then the volume curvature properties at different scales are calculated.

4. The crack prediction method according to claim 3, characterized in that, The quadratic fitting equation is: (x,y)=ax 2 +by 2 +cxy+dx+ey+f; Where a, b, c, d, e, and f are the coefficients of the quadratic surface, x is the abscissa, and y is the ordinate.

5. The crack prediction method according to claim 4, characterized in that, The expression for the derivative operator is: E=F- 1 [-I a F(u)T(k)]; Where E is the derivative operator, T(K) is the Gaussian filter, α is the wavelength factor, k is the wavenumber, and i is the... F represents the Fourier series expansion, F(u) represents the Fourier transform of the seismic data, F -1 This is the inverse Fourier transform.

6. The crack prediction method according to claim 4, characterized in that, The expression for the Gaussian filter is: Where η is the direction of the seismic data, β is the adaptive adjustment factor corresponding to the wave number k, and β = β(k).

7. The crack prediction method according to claim 1, characterized in that, The body curvature attribute includes: Maximum curvature, minimum curvature, mean curvature, curvature, Gaussian curvature, maximum positive curvature, minimum negative curvature, and normal curvature.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the crack prediction method according to any one of claims 1-7.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the crack prediction method according to any one of claims 1-7.

10. A crack prediction device, characterized in that, include: The extraction module is used to extract dip attributes from seismic data volumes to obtain dip data volumes; The acquisition module is used to acquire the diffusion tensor edge-preserving filtering data volume based on the tilt angle data volume using the diffusion tensor edge-preserving filtering algorithm; The calculation module is used to calculate the volume curvature properties at different scales based on the diffusion tensor edge-preserving filtered data volume after Fourier series expansion using a Gaussian filter in the spatial wavenumber domain. The prediction module is used to predict cracks based on volume curvature properties at different scales.

Images