Strike-slip fault breaking joint body earthquake prediction method, device and equipment based on difference algorithm and edge detection and storage medium

Abstract

The invention relates to the technical field of reservoir mining control, in particular to a strike-slip fault joint body earthquake prediction method, device and equipment based on a difference algorithm and edge detection and a storage medium, and the method comprises the steps: carrying out the intrinsic coherence calculation of three-dimensional earthquake data under each frequency after frequency extraction processing, and extracting fault data; processing the three-dimensional seismic data by using a difference algorithm and edge detection, and extracting crack data; and performing weighted average fusion on the fault data and the crack data based on the weight to obtain a breaking joint body index. According to the method, the identification precision of the strike-slip fault is improved through introduction of the eigenvalue coherent processing technology, the identification precision of the strike-slip fault is remarkably improved especially for a small fault conjugated with the strike-slip fault, and after gradient data is obtained from three-dimensional seismic data, crack information is extracted by using a Canny edge detection algorithm, so that the identification precision of the strike-slip fault is improved. Wherein the Prewitt algorithm is used for calculating the gradient edge, so that the precision of the Canny algorithm is enhanced, and the robustness of the Canny algorithm is improved.

Description

Technical Field

[0001] This invention relates to the field of reservoir exploitation control technology, and is a method, device, equipment and storage medium for seismic prediction of strike-slip fault fractures based on differential algorithms and edge detection. Background Technology

[0002] In the complex geological systems of tight sandstone reservoirs, fracture bodies play a crucial role, and their study can significantly enhance the oil-bearing properties of the reservoir. Existing research on fracture bodies includes deep learning-based seismic characterization techniques for carbonate fracture-cavity formations, seismic characterization schemes for carbonate fracture-cavity structures based on the same principles, techniques for constructing geological models of fracture-type reservoirs, and new strategies for comprehensively determining fracture body development zones using multi-dimensional seismic attributes. Furthermore, a series of quantitative analyses have been conducted on the spatial distribution of karst fracture-cavity reservoirs and the connectivity of fracture-cavity reservoirs at various scales. However, existing technologies are inevitably limited in their accuracy in predicting fracture bodies in the study area.

[0003] Current research on fracture characterization related to strike-slip faults still needs further development, and the prediction of strike-slip fault fractures remains unresolved. The difficulties in predicting strike-slip fault fractures can be divided into two main aspects: First, accurate identification of strike-slip faults is extremely challenging, primarily due to the significant horizontal movement of the fault blocks and the weak vertical displacement. Furthermore, the complex and varied morphology and characteristics of strike-slip faults themselves mean that different faults can exhibit vastly different appearances and properties, significantly reducing the accuracy of identification. Second, according to the classic Riedel's conjugate shear rupture criterion, the fracture system induced by strike-slip faults can generally be classified into two groups: one roughly parallel to the strike-slip fault, and the other forming a conjugate angle. While predicting fractures parallel to the strike-slip fault is relatively straightforward, the characteristics of smaller faults forming a conjugate angle with it are often blurred in 3D seismic data, making identification exceptionally difficult. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and storage medium for earthquake prediction of strike-slip fault fracture bodies based on differential algorithms and edge detection, overcoming the shortcomings of the prior art. It can effectively solve the problem that existing strike-slip fault fracture body prediction methods cannot effectively identify small faults conjugate with the strike-slip fault.

[0005] One of the technical solutions of this invention is achieved through the following measures: a method for earthquake prediction of strike-slip fault fractures based on differential algorithms and edge detection, comprising:

[0006] The 3D seismic data is up-frequency processed, and after up-frequency processing, intrinsic coherence calculation is performed on the 3D seismic data at each frequency to extract fault data.

[0007] Crack data is extracted by processing 3D seismic data using differential algorithms and edge detection.

[0008] The weights of fault data and fracture data are determined, and the fault data and fracture data are weighted and averaged based on the weights to obtain the fracture body index.

[0009] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0010] The above-mentioned frequency upscaling processing of 3D seismic data using the spectral blueing algorithm includes:

[0011] The fast Fourier transform is used to convert 3D seismic data from the time domain to the frequency domain;

[0012] Filtering of three-dimensional seismic data in the frequency domain;

[0013] The frequency upsampling process of filtered 3D seismic data is performed using the spectral blueing algorithm;

[0014] The frequency-upgraded 3D seismic data is converted from the frequency domain to the time domain using the inverse Fourier transform.

[0015] The above-mentioned process involves performing intrinsic coherence calculations on the 3D seismic data at each frequency after frequency upscaling to extract fault data, including:

[0016] The intrinsic coherence functions of 3D seismic data at different frequencies are determined to obtain intrinsic coherence slices of the 3D seismic data;

[0017] Set a fault extraction threshold value, and use intrinsic coherence slices of 3D seismic data to extract fault data that meet the fault extraction threshold value.

[0018] The above-mentioned processing of 3D seismic data using differential algorithms and edge detection to extract crack data includes:

[0019] Gradient data is obtained by calculating 3D seismic data using a differential algorithm, including:

[0020] (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction.

[0021] (2) Based on the center difference value in each direction, determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point;

[0022] The gradient data was processed using the Canny edge detection algorithm with double thresholding to obtain crack data.

[0023] The above-mentioned determination of the weights for fault and fracture data, followed by a weighted average fusion of the fault and fracture data based on these weights, yields the fracture-fracture index, including:

[0024] The fault and fracture data are normalized using the following formula:

[0025]

[0026] Among them, f n The normalized fault data is given by f, where f is the specific value of the fault data. max f is the maximum value of the fault data. min c is the minimum value of the fault data. n Here are the normalized crack data, and c is the specific value of the crack data. max c represents the maximum value of the crack data. min This represents the minimum value of the crack data;

[0027] Determine the weights of fault data and fracture data, and perform a weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index;

[0028] fc = k1 × f n +k2×c n

[0029] Where fc is the fracture body index, k1 is the weight of the fault data, and k2 is the weight of the fracture data.

[0030] Before extracting fault and fracture data from the 3D seismic data, the 3D seismic data is subjected to Gaussian linear smoothing filtering.

[0031] The second technical solution of the present invention is achieved through the following measures: a strike-slip fault fracture body seismic prediction device based on differential algorithm and edge detection, comprising:

[0032] The first extraction unit performs frequency up-conversion processing on the 3D seismic data, and after frequency up-conversion processing, performs intrinsic coherence calculation on the 3D seismic data at each frequency to extract fault data.

[0033] The second extraction unit processes the 3D seismic data using differential algorithms and edge detection to extract crack data;

[0034] The prediction unit determines the weights of fault data and fracture data, and performs a weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0035] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0036] The aforementioned first extraction unit includes:

[0037] The frequency upscaling module uses Fast Fourier Transform to convert 3D seismic data from the time domain to the frequency domain, filters the frequency domain 3D seismic data, and uses a spectral blueing algorithm to upscale the filtered 3D seismic data; then it uses Inverse Fourier Transform to convert the frequency upscaled 3D seismic data from the frequency domain back to the time domain.

[0038] The fault data extraction module includes: determining the intrinsic coherence function of 3D seismic data at different frequencies, obtaining intrinsic coherence slices of 3D seismic data, setting fault extraction threshold values, and extracting fault data that meets the fault extraction threshold values ​​using the intrinsic coherence slices of 3D seismic data.

[0039] The aforementioned second extraction unit includes:

[0040] The gradient data extraction module uses a difference algorithm to calculate gradient data from 3D seismic data, including:

[0041] (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction.

[0042] (2) Based on the center difference value in each direction, determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point;

[0043] The crack data extraction module uses the Canny edge detection algorithm to perform double thresholding on the gradient data to obtain crack data.

[0044] The aforementioned preprocessing unit performs Gaussian linear smoothing filtering on the 3D seismic data.

[0045] This invention applies Gaussian linear smoothing filtering to 3D seismic data to remove high-frequency noise and impurities while retaining effective information. The introduction of eigenvalue coherence processing improves the accuracy of strike-slip fault identification, particularly for small faults conjugate to strike-slip faults. After obtaining gradient data from the 3D seismic data, the Canny edge detection algorithm is used to extract fracture information. The Prewitt algorithm is used to calculate gradient edges, enhancing both the accuracy and robustness of the Canny algorithm. Furthermore, a fracture body index is constructed, which fuses fault and fracture data. This allows for the acquisition of fracture body data, thus providing a clearer depiction of the fracture body distribution characteristics of strike-slip faults. Attached Figure Description

[0046] Appendix Figure 1 This is a schematic diagram of the earthquake prediction method for strike-slip fault fracture bodies according to the present invention.

[0047] Appendix Figure 2 This is a schematic diagram of the method for frequency upscaling of three-dimensional seismic data in this invention.

[0048] Appendix Figure 3 This is a schematic diagram of the method for extracting fault data according to the present invention.

[0049] Appendix Figure 4 This is a schematic diagram of the method for extracting crack data according to the present invention.

[0050] Appendix Figure 5 This is a schematic diagram of another earthquake prediction method for strike-slip fault fracture bodies according to the present invention.

[0051] Appendix Figure 6 This is a cross-sectional view of the three-dimensional seismic data before filtering in Embodiment 4 of the present invention.

[0052] Appendix Figure 7 This is a cross-sectional view of the filtered three-dimensional seismic data in Embodiment 4 of the present invention.

[0053] Appendix Figure 8 This is a cross-sectional view of three-dimensional seismic data before and after spectral blueing processing in Embodiment 4 of the present invention.

[0054] Appendix Figure 9 This is a schematic diagram of intrinsic coherence slices before and after spectral blueing treatment in Example 4 of the present invention.

[0055] Appendix Figure 10 This is a schematic diagram of the tomographic data extracted using intrinsic coherence slices in Embodiment 4 of the present invention.

[0056] Appendix Figure 11 This is a schematic diagram illustrating the overlay effect of gradient data and ladder data with three-dimensional seismic data in Embodiment 4 of the present invention.

[0057] Appendix Figure 12 This is a schematic diagram illustrating the edge detection effect of gradient data in Embodiment 4 of the present invention.

[0058] Appendix Figure 13 This is a schematic diagram of the crack data obtained by edge detection in Embodiment 4 of the present invention.

[0059] Appendix Figure 14 This is a schematic diagram of the fracture body data plane in Embodiment 4 of the present invention.

[0060] Appendix Figure 15 This is a schematic diagram of the device structure of the present invention. Detailed Implementation

[0061] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.

[0062] The present invention will be further described below with reference to embodiments and accompanying drawings:

[0063] Example 1: As shown in the attached document Figure 1 As shown, this invention discloses a method for earthquake prediction of strike-slip fault fractures based on differential algorithms and edge detection, comprising:

[0064] Step S110: Frequency up-up processing is performed on the three-dimensional seismic data, and after frequency up-up processing, intrinsic coherence calculation is performed on the three-dimensional seismic data at each frequency to extract fault data.

[0065] Step S120: Process the 3D seismic data using differential algorithms and edge detection to extract crack data;

[0066] Step S130: Determine the weights of fault data and fracture data, and perform weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0067] This invention discloses a method for upsampling three-dimensional seismic data, followed by intrinsic coherence calculation to extract fault data. The method then uses a differential algorithm and edge detection to process the three-dimensional seismic data to extract fracture data. Finally, the fault data and fracture data are fused to generate a fracture body index. This allows for the accurate depiction of fracture development zones by combining the fracture body index and fracture attribute map, resulting in a clearer characterization of the fracture body distribution characteristics of strike-slip faults.

[0068] Example 2: This embodiment of the invention discloses a method for earthquake prediction of strike-slip fault fractures based on differential algorithms and edge detection, including:

[0069] Step S210: Frequency upscaling is performed on the 3D seismic data, and after frequency upscaling, intrinsic coherence calculation is performed on the 3D seismic data at each frequency to extract fault data.

[0070] In the above steps, as shown in the appendix Figure 2 As shown, the spectral blueing algorithm is used to perform frequency upscaling on 3D seismic data, including:

[0071] Step S2111: Use Fast Fourier Transform to convert the three-dimensional seismic data from the time domain to the frequency domain;

[0072] Step S2112: Filter the three-dimensional seismic data in the frequency domain; the filtering can be achieved by a high-pass filter or other fine filtering methods, thereby removing low-frequency noise and interference and enhancing high-frequency components.

[0073] Step S2113: The filtered 3D seismic data is frequency-upgraded using the spectral blueing algorithm;

[0074] Step S2114: Use inverse Fourier transform to convert the frequency-upgraded 3D seismic data from the frequency domain to the time domain.

[0075] The spectral blueing algorithm used here enhances high-frequency signals in seismic data, making the outlines of geological structures more distinct and three-dimensional. Processing with the spectral blueing algorithm can enhance the high-frequency components of 3D seismic data, improve its spatial resolution, while suppressing low-frequency noise, reducing noise interference, and enhancing the signal-to-noise ratio of the seismic signal.

[0076] In the above steps, as shown in the appendix Figure 3 As shown, after frequency upscaling, intrinsic coherence calculations are performed on the 3D seismic data at each frequency to extract fault data, including:

[0077] Step S2121: Determine the intrinsic coherence function of the 3D seismic data at different frequencies to obtain the intrinsic coherence slices of the 3D seismic data;

[0078] Step S2122: Set the fault extraction threshold value and use the intrinsic coherence slices of 3D seismic data to extract fault data that meets the fault extraction threshold value.

[0079] This study employs eigenvalue coherence (EVC) techniques to deeply analyze the eigencoherence functions of 3D seismic data at different frequencies, enabling precise identification of the location and orientation of subsurface faults. Specifically, for each frequency of 3D seismic data, EVC techniques are used to obtain the corresponding eigencoherence function. The eigencoherence function at each frequency is then binarized to obtain a fault detection image, which displays the location and orientation of the fault. By fusing the fault detection images from different frequencies, a final 3D seismic fault model is constructed, resulting in an eigencoherence slice of the 3D seismic data.

[0080] Step S220: The 3D seismic data is processed using differential algorithms and edge detection to extract crack data.

[0081] In the above steps, as shown in the appendix Figure 4 As shown, it specifically includes:

[0082] Step S221: Calculate the gradient data from the 3D seismic data using a differential algorithm, including:

[0083] (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction.

[0084] (2) Determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point based on the center difference value of each direction.

[0085] The above-mentioned difference algorithm can compare the changes between adjacent data points in seismic data to obtain gradient information. In specific applications, the central difference method is used to obtain gradient data for three-dimensional seismic data. For three-dimensional seismic data, it can be regarded as a three-dimensional grid, where each point has a value representing the amplitude of the seismic wave.

[0086] The process of obtaining gradient data described above is as follows:

[0087] (1) For each grid point, calculate the center difference values ​​in the x, y, and z directions respectively. Taking the x direction as an example, the center difference formula is:

[0088] dx=(f(i+1,j,k)-f(i-1,j,k)) / (2×dx) (1)

[0089] Where f(i,j,k) represents the value at grid point i,j,k, and dx represents the distance between grid points.

[0090] (2) Combine the central difference values ​​in the x, y, and z directions into a three-dimensional vector, i.e.:

[0091] grad(i,j,k)=(dx,dy.dz) (2)

[0092] (3) For each grid point, calculate its gradient magnitude, i.e.:

[0093]

[0094] (4) For each grid point, determine its gradient direction, i.e.:

[0095] dir(i,j,k)=(dx / mag,dy / mag,dz / mag) (4)

[0096] Through the above steps, gradient data of the three-dimensional seismic data were obtained.

[0097] Step S222: Use the Canny edge detection algorithm to perform double thresholding on the gradient data to obtain crack data.

[0098] Edge detection algorithms can extract crack contours from seismic gradient data, more accurately identifying cracks in underground rock layers. By performing edge detection on seismic data gradients, information such as the location, shape, and distribution of underground cracks can be obtained. In this embodiment, the Canny edge detection algorithm is selected to perform double-threshold processing on the gradient data, dividing the non-maximum suppressed gradient data into two threshold intervals: strong edges and weak edges, each accounting for 50%. For the gradient value G(x,y) of a pixel, if it falls within the weak edge region, then it is a crack signal. Connectivity analysis is performed on the weak edges, and crack information can be extracted from the weak edge pixels after edge connection.

[0099] Step S230: Determine the weights of fault data and fracture data, and perform weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0100] The above steps specifically include:

[0101] (1) Normalize the fault data and fracture data. The normalization formula is shown below:

[0102]

[0103] Among them, f n The normalized fault data is given by f, where f is the specific value of the fault data. max f is the maximum value of the fault data. min c is the minimum value of the fault data. n Here are the normalized crack data, and c is the specific value of the crack data. max c represents the maximum value of the crack data. min This represents the minimum value of the crack data;

[0104] (2) Determine the weights of fault data and fracture data, and perform weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0105] fc = k1 × f n +k2×c n

[0106] Where fc is the fracture body index, k1 is the weight of the fault data, and k2 is the weight of the fracture data.

[0107] The above-described data normalization process for fault and fracture data ensures that both types of data are in the same coordinate system, thereby enabling their fusion within the same value range. The weights of the fault and fracture data can be set according to actual needs. In this embodiment, the weights of the fault and fracture data can each be set to 0.5.

[0108] Faults and fractures are two of the most common features in geological structures, and they have a decisive impact on tectonic evolution and hydrocarbon migration. Changes in faults affect hydrocarbon accumulation and migration, while the presence of fractures affects reservoir permeability and porosity. By organically fusing fault and fracture data through the above steps, the resulting fracture-fracture index (fracture-fracture data) is obtained. This allows for the creation of fracture attribute maps, revealing the distribution characteristics of strike-slip fault fractures. This more accurately depicts geological structures and reservoir characteristics, improving the efficiency and success rate of oil and gas exploration and development.

[0109] In summary, this embodiment of the invention improves the identification accuracy of strike-slip faults by introducing eigenvalue coherent processing technology, especially for small faults conjugate with strike-slip faults, where the identification accuracy is significantly improved. After obtaining gradient data from 3D seismic data, the Canny edge detection algorithm is used to extract fracture information, with the Prewitt algorithm used to calculate gradient edges, which not only enhances the accuracy of the Canny algorithm but also improves its robustness. Furthermore, this embodiment constructs a fracture body index, which fuses fault data and fracture data, thereby allowing the fracture body data to be obtained using the fracture body index, thus more clearly characterizing the distribution features of strike-slip fault fracture bodies.

[0110] Example 3: As shown in the attached document Figure 5 As shown, this invention discloses a method for earthquake prediction of strike-slip fault fractures based on differential algorithms and edge detection, comprising:

[0111] Step S310: Perform Gaussian linear smoothing filtering on the three-dimensional seismic data;

[0112] In this step, the 3D seismic data is subjected to Gaussian linear smoothing filtering. That is, the 3D seismic data is unfolded into a 3D image in the sequence of time axis, rows and columns, and then Gaussian linear smoothing filtering is applied in these three dimensions to remove high-frequency noise and impurities in the data and retain effective information.

[0113] Step S320: Frequency up-up processing is performed on the 3D seismic data, and after frequency up-up processing, intrinsic coherence calculation is performed on the 3D seismic data at each frequency to extract fault data.

[0114] Step S330: Process the 3D seismic data using differential algorithms and edge detection to extract crack data;

[0115] Step S340: Determine the weights of fault data and fracture data, and perform weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0116] Example 4: Obtaining as shown in the attached document Figure 6The three-dimensional seismic data profile shown is analyzed using the seismic prediction method for strike-slip fault fractures based on differential algorithm and edge detection of this invention. Specifically, the analysis includes:

[0117] (1) Gaussian linear smoothing filtering was applied to the 3D seismic data profile to obtain the following result: Figure 7 The filtered 3D seismic data profile shown is attached. Figure 7 As can be seen, the signal-to-noise ratio of the 3D seismic data is greatly improved after Gaussian linear smoothing filtering.

[0118] (2) The 3D seismic data after Gaussian linear smoothing filtering was subjected to spectral blueing. The 3D seismic data profiles before and after spectral blueing are shown in the attached figure. Figure 8 As shown in the figure, the left side of the image is before spectral blueing, and the right side is after spectral blueing.

[0119] From the appendix Figure 8 As can be seen, faults ① and ② are clearer on the cross-section after spectral blueing, which improves the accuracy of fault identification.

[0120] (3) Calculate the intrinsic coherence of the 3D seismic data after spectral blueing and extract fault data. (See appendix) Figure 9 The images show intrinsic coherence slices before and after spectral blueing. The black lines represent faults. Before blueing, only northeast-trending faults were identified in the intrinsic coherence slices, while after blueing, both northeast-trending and northwest-trending faults were clearly identified (as indicated by the arrows in the images). (See attached image.) Figure 10 The data is the tomographic data extracted from the intrinsic coherence slice after spectral blueing, with a tomographic data threshold of 0.8.

[0121] (4) Gradient data were obtained by calculating the 3D seismic data after Gaussian linear smoothing filtering using the difference algorithm, as shown in the attached figure. Figure 11 As shown, a represents the gradient data calculated using 3D seismic data, and b represents the overlay effect of the gradient data and 3D seismic data.

[0122] (5) The gradient data is processed using the Canny edge detection algorithm with double thresholding to extract crack information, resulting in crack data, including the attached... Figure 12 The image shows the profile results of edge detection on gradient data. Figure 13 To obtain crack data using edge detection.

[0123] (6) The fault data and fracture data are organically integrated to form the fracture body index, thereby obtaining the attached... Figure 14 The data of the fracture structure shown in the figure clearly depicts the fracture development zones in the northeast and northwest directions, and these two sets of fractures are distributed in a conjugate pattern.

[0124] Example 5: As shown in the attached document Figure 15 As shown, this embodiment of the invention discloses an earthquake prediction device for strike-slip fault fractures based on differential algorithms and edge detection, comprising:

[0125] The preprocessing unit performs Gaussian linear smoothing filtering on the 3D seismic data;

[0126] The first extraction unit performs frequency up-conversion processing on the 3D seismic data, and after frequency up-conversion processing, performs intrinsic coherence calculation on the 3D seismic data at each frequency to extract fault data.

[0127] The second extraction unit processes the 3D seismic data using differential algorithms and edge detection to extract crack data;

[0128] The prediction unit determines the weights of fault data and fracture data, and performs a weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

[0129] The first extraction unit includes:

[0130] The frequency upscaling module uses Fast Fourier Transform to convert 3D seismic data from the time domain to the frequency domain, filters the frequency domain 3D seismic data, and uses a spectral blueing algorithm to upscale the filtered 3D seismic data; then it uses Inverse Fourier Transform to convert the frequency upscaled 3D seismic data from the frequency domain back to the time domain.

[0131] The fault data extraction module includes: determining the intrinsic coherence function of 3D seismic data at different frequencies, obtaining intrinsic coherence slices of 3D seismic data, setting fault extraction threshold values, and extracting fault data that meets the fault extraction threshold values ​​using the intrinsic coherence slices of 3D seismic data.

[0132] The second extraction unit includes:

[0133] The gradient data extraction module uses a difference algorithm to calculate gradient data from 3D seismic data, including:

[0134] (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction.

[0135] (2) Based on the center difference value in each direction, determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point;

[0136] The crack data extraction module uses the Canny edge detection algorithm to perform double thresholding on the gradient data to obtain crack data.

[0137] Example 6: This embodiment of the invention discloses an electronic device, including a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement an earthquake prediction method for strike-slip fault fracture bodies based on differential algorithms and edge detection.

[0138] The processor described above can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. It can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The memory can include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, portable hard drives, magnetic disks, or optical disks.

[0139] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0140] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0141] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] The above technical features constitute the preferred embodiment of the present invention, which has strong adaptability and optimal implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the requirements of different situations.

Claims

1. A method for earthquake prediction of strike-slip fault fractures based on differential algorithm and edge detection, characterized in that, include: The 3D seismic data is up-frequency processed, and after the up-frequency processing, the intrinsic coherence of the 3D seismic data at each frequency is calculated to extract fault data. Crack data is extracted by processing 3D seismic data using differential algorithms and edge detection. The weights of fault data and fracture data are determined, and the fault data and fracture data are weighted and averaged based on the weights to obtain the fracture body index.

2. The earthquake prediction method for strike-slip fault fractures based on differential algorithm and edge detection according to claim 1, characterized in that, Frequency upsampling of 3D seismic data is performed using the spectral blueing algorithm, including: The fast Fourier transform is used to convert 3D seismic data from the time domain to the frequency domain; Filtering of three-dimensional seismic data in the frequency domain; The frequency upsampling process of filtered 3D seismic data is performed using the spectral blueing algorithm; The frequency-upgraded 3D seismic data is converted from the frequency domain to the time domain using the inverse Fourier transform.

3. The earthquake prediction method for strike-slip fault fractures based on differential algorithm and edge detection according to claim 1 or 2, characterized in that, After frequency upscaling, intrinsic coherence calculations were performed on the 3D seismic data at each frequency to extract fault data, including: The intrinsic coherence functions of 3D seismic data at different frequencies are determined to obtain intrinsic coherence slices of the 3D seismic data; Set a fault extraction threshold value, and use intrinsic coherence slices of 3D seismic data to extract fault data that meet the fault extraction threshold value.

4. The earthquake prediction method for strike-slip fault fractures based on differential algorithm and edge detection according to any one of claims 1 or 2, characterized in that, The 3D seismic data is processed using differential algorithms and edge detection to extract crack data, including: Gradient data is obtained by calculating 3D seismic data using a differential algorithm, including: (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction. (2) Based on the center difference value in each direction, determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point; The gradient data was processed using the Canny edge detection algorithm with double thresholding to obtain crack data.

5. The earthquake prediction method for strike-slip fault fractures based on differential algorithm and edge detection according to any one of claims 1 to 4, characterized in that, The weights of fault and fracture data are determined, and a weighted average fusion is performed on the fault and fracture data based on these weights to obtain the fracture-fracture index, which includes: The fault and fracture data are normalized using the following formula: Among them, f n The normalized fault data is given by f, where f is the specific value of the fault data. max f is the maximum value of the fault data. min c is the minimum value of the fault data. n Here are the normalized crack data, and c is the specific value of the crack data. max c represents the maximum value of the crack data. min This represents the minimum value of the crack data; Determine the weights of fault data and fracture data, and perform a weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index; fc=k1×f n +k2×c n Where fc is the fracture body index, k1 is the weight of the fault data, and k2 is the weight of the fracture data.

6. The earthquake prediction method for strike-slip fault fractures based on differential algorithm and edge detection according to any one of claims 1 to 5, characterized in that, Before extracting fault and fracture data from the 3D seismic data, Gaussian linear smoothing filtering is applied to the 3D seismic data.

7. A device for predicting earthquakes in strike-slip fault fractures based on differential algorithms and edge detection, using the method described in any one of claims 1 to 5, characterized in that, include: The first extraction unit performs frequency up-conversion processing on the 3D seismic data, and after frequency up-conversion processing, performs intrinsic coherence calculation on the 3D seismic data at each frequency to extract fault data. The second extraction unit uses differential algorithms and edge detection to process the 3D seismic data and extract crack data; The prediction unit determines the weights of fault data and fracture data, and performs a weighted average fusion of fault data and fracture data based on the weights to obtain the fracture body index.

8. The earthquake prediction device for strike-slip fault fractures based on differential algorithm and edge detection according to claim 7, characterized in that, The first extraction unit includes: The frequency upscaling module uses Fast Fourier Transform to convert 3D seismic data from the time domain to the frequency domain, filters the frequency domain 3D seismic data, uses the spectral blueing algorithm to upscale the filtered 3D seismic data, and uses Inverse Fourier Transform to convert the frequency upscaled 3D seismic data from the frequency domain back to the time domain. The fault data extraction module includes: determining the intrinsic coherence function of 3D seismic data at different frequencies, obtaining intrinsic coherence slices of 3D seismic data, setting fault extraction threshold values, and extracting fault data that meets the fault extraction threshold values ​​using the intrinsic coherence slices of 3D seismic data. or / and, The second extraction unit includes: The gradient data extraction module uses a difference algorithm to calculate gradient data from 3D seismic data, including: (1) Select a center point and calculate the center difference value in each direction based on the seismic wave amplitude values ​​of the neighboring points in each direction. (2) Based on the center difference value in each direction, determine the gradient magnitude and gradient direction of the gradient data corresponding to the center point; The crack data extraction module uses the Canny edge detection algorithm to perform double thresholding on the gradient data to obtain crack data.

9. The earthquake prediction device for strike-slip fault fractures based on differential algorithm and edge detection according to claim 7 or 8, characterized in that, It also includes a preprocessing unit that performs Gaussian linear smoothing filtering on the 3D seismic data.

10. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement the earthquake prediction method for strike-slip fault fracture bodies based on differential algorithms and edge detection as described in any one of claims 1 to 6.

Images