Strike-slip fault identification method and device, electronic equipment and medium

Abstract

This application provides a method, apparatus, electronic device, and medium for identifying strike-slip faults. The method includes: determining the distribution range and waveform characteristics of strong reflection phase axes in stacked gather data; decomposing the stacked gather data into multiple wavelets, removing wavelets with waveform characteristics identical to those of the strong reflection phase axes; superimposing the remaining wavelets to obtain a reconstructed seismic signal; acquiring the time-spectrum signal corresponding to each seismic trace in the reconstructed seismic signal; enhancing the strike-slip fault reflection signal in the time-spectrum signal and attenuating the background reflection signal; converting the processed time-spectrum signal to the time domain to obtain a separated seismic signal; establishing a stratigraphic body based on the reconstructed seismic signal; determining profiles and planar slices on the stratigraphic body based on the reconstructed seismic signal, the reconstructed seismic signal, and the separated seismic signal, inputting them into a strike-slip fault identification library, and obtaining the strike-slip fault identification result. The method of this application can efficiently and accurately identify small-displacement strike-slip faults.

Description

Technical Field

[0001] This application relates to the field of seismic exploration, and in particular to a strike-slip fault identification method, device, electronic equipment, and medium. Background Technology

[0002] As a vital strategic energy resource for the nation, the exploration and development of petroleum is attracting increasing attention. Petroleum exploration and development results indicate that major oil and gas enrichment areas in carbonate rocks are mostly closely related to faults. Oil and gas are concentrated in karst development areas near fractured fault zones connecting oil sources, making this a crucial direction for exploring karst reservoirs. Identifying strike-slip faults is a necessary condition for finding these types of karst reservoirs. However, with the deepening of exploration and development in recent years, in some mature development areas, large-scale karst pore reservoirs have been almost completely drilled out, leaving mostly reservoirs with smaller reservoir spaces. These reservoirs are controlled by strike-slip faults with small displacements, exhibiting strong concealment. This makes it difficult to identify the internal structure of these fault-controlled karst reservoirs, which has consistently hindered the full exploitation of reserve potential, efficient reservoir development, and improved success rates of new well commissioning.

[0003] In existing technologies, strike-slip fault identification largely employs ant-tracking algorithms. The workflow involves enhancing the spatial discontinuities in seismic data using edge detection techniques and preprocessing the data arbitrarily through noise suppression. A large number of electronic "ants" are then deployed within the seismic body, each moving along potential fault planes. Upon encountering a predicted fault plane, a pheromone is used to clearly mark it, thereby automatically extracting fault groups or creating detailed maps of stratigraphic discontinuities. The extracted faults are then evaluated, corrected, and filtered through manual interaction. Ant-tracking algorithms can further enhance fault information and suppress non-fault information based on the fault detection attribute data, thus enabling the identification of strike-slip faults.

[0004] In the above scheme, when using the ant tracking algorithm to identify strike-slip faults, the effective signal of seismic data attenuates rapidly for deep target layers, making it difficult to identify small-displacement strike-slip faults. Furthermore, the efficiency and accuracy of manually interpreting strike-slip faults are low. Therefore, there is a problem that it is difficult to efficiently and accurately identify small-displacement strike-slip faults. Summary of the Invention

[0005] This application provides a strike-slip fracture identification method, device, electronic device, and medium to solve the problem of difficulty in efficiently and accurately identifying small-displacement strike-slip fractures.

[0006] On the one hand, this application provides a method for identifying strike-slip fractures, including:

[0007] Based on the unidentified strata data in the overlay data, the distribution range and waveform characteristics of strong reflection phase axes in the overlay data are determined; the overlay data is decomposed into multiple wavelets, and wavelets with waveform characteristics identical to those of the strong reflection phase axes are removed; the remaining wavelets are overlaid to obtain the reconstructed seismic signal.

[0008] The process involves: acquiring the time-spectrum seismic signals corresponding to each seismic trace in the reconstructed seismic signal; determining the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal; enhancing the strike-slip fault reflection signal and attenuating the background reflection signal to obtain the processed time-spectrum seismic signal; and converting the processed time-spectrum seismic signal into a time-domain signal to obtain the separated seismic signal.

[0009] Based on the reconstructed seismic signal, a stratigraphic body is established; based on the reconstructed seismic signal and the separated seismic signal, a profile and planar slice are determined on the stratigraphic body; the profile and planar slice are input into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, the strike-slip fault identification result including the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value.

[0010] Optionally, the method further includes:

[0011] Based on the seismic common reflection point gather, principal component analysis is performed on the seismic trace data within the preset incident angle range to construct the principal component trace;

[0012] Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and take the seismic trace with the largest correlation coefficient as the model trace;

[0013] For each seismic trace, a preset time window is slid by a preset distance. The correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace.

[0014] Based on the remaining time difference of each seismic trace, the seismic trace data of that seismic trace is corrected to obtain the optimized seismic trace data.

[0015] The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

[0016] Optionally, determining the distribution range and waveform characteristics of the strong reflection in-phase axis includes:

[0017] Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data.

[0018] Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of the strong reflection in-phase axis in the overlay data of the gathers are determined. The waveform characteristics include the frequency, wavelength, amplitude and phase of the strong reflection in-phase axis.

[0019] Optionally, the step of decomposing the overlay data of the gather into multiple wavelets and removing the wavelets whose waveform characteristics are the same as those of the strong reflection in-phase axis includes:

[0020] Based on the seismic trace multi-wavelet decomposition method, each seismic trace signal in the stacked gather data is decomposed into multiple wavelets;

[0021] Remove the wavelets whose frequency, wavelength, amplitude, and phase are the same as those of the strongly reflecting phase axis from the multiple wavelets.

[0022] Optionally, determining the strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal includes:

[0023] In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals.

[0024] Signal segments with frequencies higher than a first preset value and lower than a second preset value, and which remain stable over time, are identified as slip-slip fracture reflection signals; wherein the first preset value is lower than the second preset value.

[0025] Optionally, the method further includes:

[0026] Based on the reconstructed seismic signal and the separated seismic signal, profiles and planar slices are determined on the stratigraphic body, wherein the profiles and planar slices contain strike-slip fault response characteristics; based on the profiles and planar slices, a strike-slip fault identification database is established;

[0027] The neural network is trained based on the separated seismic signals corresponding to the profile and planar slices.

[0028] Based on spatial correlation, other cross sections and planar slices are determined on the stratigraphic body according to the cross section and planar slice; the other cross sections and planar slices are input into the neural network to obtain the strike-slip fault prediction results output by the neural network;

[0029] Based on the strike-slip fracture prediction results, the strike-slip fracture identification library is modified until a well-trained strike-slip fracture identification library is obtained.

[0030] On the other hand, this application provides a strike-slip fracture identification device, comprising:

[0031] The reconstruction module is used to determine the distribution range and waveform characteristics of strong reflection phase axes in the stacked report data based on the stratigraphic data to be identified in the stacked report data; decompose the stacked report data into multiple wavelets, remove the wavelets whose waveform characteristics are the same as those of the strong reflection phase axes; and superimpose the remaining wavelets to obtain the reconstructed seismic signal.

[0032] The separation module is used to acquire the time-spectrum seismic signals corresponding to each seismic trace signal in the reconstructed seismic signal; determine the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal; enhance the strike-slip fault reflection signal and attenuate the background reflection signal to obtain the processed time-spectrum seismic signal; and convert the processed time-spectrum seismic signal into a time-domain signal to obtain the separated seismic signal.

[0033] The identification module is used to establish a stratigraphic body based on the reconstructed seismic signal; to determine a profile and planar slice on the stratigraphic body based on the reconstructed seismic signal and the separated seismic signal; and to input the profile and planar slice into a strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, wherein the strike-slip fault identification result includes the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value.

[0034] Optionally, the device further includes an optimization module for:

[0035] Based on the seismic common reflection point gather, principal component analysis is performed on the seismic trace data within the preset incident angle range to construct the principal component trace;

[0036] Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and take the seismic trace with the largest correlation coefficient as the model trace;

[0037] For each seismic trace, a preset time window is slid by a preset distance. The correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace.

[0038] Based on the remaining time difference of each seismic trace, the seismic trace data of that seismic trace is corrected to obtain the optimized seismic trace data.

[0039] The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

[0040] Optionally, the reconstruction module is specifically used for:

[0041] Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data.

[0042] Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of the strong reflection in-phase axis in the overlay data of the gathers are determined. The waveform characteristics include the frequency, wavelength, amplitude and phase of the strong reflection in-phase axis.

[0043] Optionally, the reconstruction module is further specifically used for:

[0044] Based on the seismic trace multi-wavelet decomposition device, each seismic trace signal in the gather stack data is decomposed into multiple wavelets.

[0045] Remove the wavelets whose frequency, wavelength, amplitude, and phase are the same as those of the strongly reflecting phase axis from the multiple wavelets.

[0046] Optionally, the separation module is specifically used for:

[0047] In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals.

[0048] Signal segments with frequencies higher than a first preset value and lower than a second preset value, and which remain stable over time, are identified as slip-slip fracture reflection signals; wherein the first preset value is lower than the second preset value.

[0049] Optionally, the device further includes a training module for:

[0050] Based on the reconstructed seismic signal and the separated seismic signal, profiles and planar slices are determined on the stratigraphic body, wherein the profiles and planar slices contain strike-slip fault response characteristics; based on the profiles and planar slices, a strike-slip fault identification database is established;

[0051] The neural network is trained based on the separated seismic signals corresponding to the profile and planar slices.

[0052] Based on spatial correlation, other cross sections and planar slices are determined on the stratigraphic body according to the cross section and planar slice; the other cross sections and planar slices are input into the neural network to obtain the strike-slip fault prediction results output by the neural network;

[0053] Based on the strike-slip fracture prediction results, the strike-slip fracture identification library is modified until a well-trained strike-slip fracture identification library is obtained.

[0054] In another aspect, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described above.

[0055] In another aspect, this application provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method described above.

[0056] The strike-slip fault identification method, apparatus, electronic device, and medium provided in this application identify strong reflection phase axes in the stacked gather data, decompose the stacked gather data into multiple wavelets, remove wavelets with waveform characteristics identical to those of the strong reflection phase axes, and stack the remaining wavelets to obtain a reconstructed seismic signal. This removes the masking effect of the strong energy reflection characteristics of the strong reflection phase axes on the response characteristics of the strike-slip fault, highlighting the characteristics of the strike-slip fault. The time-spectrum seismic signals corresponding to each seismic trace in the reconstructed seismic signal are obtained, and the strike-slip fault reflection signals and background reflection signals are determined to identify the strike-slip fault. The reflected signal is enhanced while the background reflected signal is attenuated. The processed time-spectrum seismic signal is converted into a time-domain signal to obtain a separated seismic signal. The difference in time-frequency characteristics between the strike-slip fault reflection and the background reflection is then used to separate the background reflection signal from the strike-slip fault reflection signal, removing the masking effect of surrounding reflection energy on the strike-slip fault. Based on the reconstructed and separated seismic signals, a stratigraphic body is established, and profiles and planar slices are determined on the stratigraphic body. The slices are input into a strike-slip fault identification database to obtain the strike-slip fault identification results for the strata to be identified, avoiding the problems of low efficiency and low accuracy of manual identification. By processing the seismic data, the masking effect of strong reflection phase axes and surrounding background reflections on the strike-slip fault is removed, highlighting the characteristics of the strike-slip fault. A model is then used to identify the strike-slip fault, solving the problem of difficulty in efficiently and accurately identifying small-offset strike-slip faults. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0058] Figure 1 The flowchart of the strike-slip fracture identification method provided in Embodiment 1 of this application is illustrated in the figure below;

[0059] Figure 2 The diagram below illustrates a flowchart of the process for obtaining overlay data of a gather, as provided in Embodiment 1 of this application.

[0060] Figure 3The image above shows an example diagram of the time profile of the raw seismic gather data provided in Embodiment 1 of this application;

[0061] Figure 4 An example diagram of the optimized seismic gather data time profile provided in Embodiment 1 of this application is shown in the figure below;

[0062] Figure 5 The figure above shows an example diagram of the time profile of the overlay data provided in Embodiment 1 of this application;

[0063] Figure 6 An example diagram of the reconstructed seismic data time profile provided in Embodiment 1 of this application is shown in the figure below;

[0064] Figure 7 The diagram above illustrates a flowchart of the process for determining background reflection signals and strike-slip fracture reflection signals according to Embodiment 1 of this application.

[0065] Figure 8 An example diagram of the separated seismic signal time profile provided in Embodiment 1 of this application is shown in the figure below;

[0066] Figure 9 The diagram above illustrates a flowchart of the training of the strike-slip fracture identification library provided in Embodiment 1 of this application.

[0067] Figure 10 The image above exemplifies an example slice of a seismic body depicting the geological features of a strike-slip fault, provided in Embodiment 1 of this application.

[0068] Figure 11 The diagram above exemplarily illustrates the structure of the slip-slip fracture identification device provided in Embodiment 2 of this application;

[0069] Figure 12 The diagram above illustrates the structure of the slip-slip fracture identification electronic device provided in Embodiment 3 of this application.

[0070] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0072] As a vital strategic energy resource for the nation, the exploration and development of petroleum is attracting increasing attention. Petroleum exploration and development results indicate that major oil and gas enrichment areas in carbonate rocks are mostly closely related to faults. Oil and gas are concentrated in karst development areas near fractured zones connecting oil sources, making this a crucial direction for exploring karst reservoirs. Identifying strike-slip faults is a necessary condition for finding these types of karst reservoirs. However, with the deepening of exploration and development in recent years, in some mature development areas, large-scale karst pore reservoirs have been almost completely drilled out. What remains are mostly reservoirs with smaller reservoir spaces, primarily composed of fractures and dissolution pores. The difference in wave impedance between reservoirs and non-reservoirs is small, exhibiting weak-energy seismic reflection characteristics. These reservoirs are controlled by small-displacement strike-slip faults, possessing strong concealment. This makes it difficult to identify the internal structure of these fault-controlled karst reservoirs, consistently hindering the full exploitation of reserve potential, efficient reservoir development, and improved success rates of new well commissioning. Therefore, identifying small-displacement strike-slip faults has become an urgent problem to be solved.

[0073] The technical solutions of this application are illustrated below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0074] Example 1

[0075] Figure 1 This is a schematic flowchart illustrating a strike-slip fracture identification method provided in an embodiment of this application. Figure 1 As shown, the strike-slip fracture identification method provided in this embodiment may include:

[0076] S101. Based on the unidentified strata data in the overlay data, determine the distribution range and waveform characteristics of strong reflection phase axes in the overlay data; decompose the overlay data into multiple wavelets, remove the wavelets whose waveform characteristics are the same as those of the strong reflection phase axes; overlay the remaining wavelets to obtain the reconstructed seismic signal;

[0077] S102. Obtain the time-spectrum seismic signal corresponding to each seismic trace signal in the reconstructed seismic signal; determine the strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal; enhance the strike-slip fault reflection signal and attenuate the background reflection signal to obtain the processed time-spectrum seismic signal; convert the processed time-spectrum seismic signal into a time-domain signal to obtain the separated seismic signal.

[0078] S103. Based on the reconstructed seismic signal, establish a stratigraphic body; based on the reconstructed seismic signal and the separated seismic signal, determine a profile and a planar slice on the stratigraphic body; input the profile and planar slice into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, the strike-slip fault identification result including the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value.

[0079] In practical applications, the execution subject of this embodiment can be a slip-slip fracture identification device, which can be implemented by a computer program, such as application software; or it can be implemented by a medium storing the relevant computer program, such as a USB flash drive or cloud drive; or it can be implemented by a physical device that integrates or installs the relevant computer program, such as a chip or server.

[0080] Specifically, in step 101, the gather stacking data is obtained by optimizing and stacking seismic gather data. The waveform characteristics of the strong reflection phase axis in the gather stacking data include frequency, wavelength, amplitude, and phase. The strata at the top of the layer to be identified in the gather stacking data have strong reflection characteristics, which affect the seismic response characteristics of the fault. By decomposing the gather stacking data into multiple wavelets and removing wavelets whose waveform characteristics are the same as those of the strong reflection phase axis, the influence of the strong reflection layer on the seismic response data is removed, highlighting the data characteristics corresponding to the strike-slip fault structure. The remaining wavelets after removing some wavelets are stacked to obtain the reconstructed seismic signal with the influence of the strong reflection layer removed, which helps in the identification and interpretation of small-displacement strike-slip faults.

[0081] In step 102, there are various ways to obtain the time-spectrum seismic signals corresponding to each seismic trace in the reconstructed seismic signal, including but not limited to performing an S-transform on the data of each seismic trace in the reconstructed seismic signal. Based on the difference in time-frequency characteristics between the strike-slip fault signal and the background signal in the time-spectrum seismic signal, the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal are determined. The strike-slip fault reflection signal is enhanced, and the background reflection signal is attenuated, thereby separating the strike-slip fault reflection features from the stratum background reflection. Correspondingly, converting the processed time-spectrum seismic signal into a time-domain signal can include performing an inverse S-transform on the processed time-spectrum seismic signal to obtain a separated seismic signal that separates the strike-slip fault features from the background strata.

[0082] In step 103, the established stratigraphic body characterizes the three-dimensional structure of each stratum corresponding to the reconstructed seismic signal. Based on the reconstructed and separated seismic signals, profiles and planar slices are determined on the stratigraphic body. The profiles and planar slices are slices containing strike-slip fault features to be identified. The profiles and planar slices are input into the strike-slip fault identification library to obtain the strike-slip fault identification results of the strata to be identified. The strike-slip fault identification results are reflected as three-dimensional strike-slip fault seismic bodies containing strike-slip fault structures and their corresponding probability values.

[0083] By processing seismic data, the strong reflection in-phase axis and surrounding background reflections are removed from the strike-slip faults, highlighting their characteristics. A model is then used to identify the strike-slip faults, solving the problem of efficiently and accurately identifying small-displacement strike-slip faults.

[0084] Before identifying strike-slip faults, seismic data preprocessing is required. In one example, Figure 2 This is a schematic diagram of the process for obtaining overlay data from a gather according to an embodiment of this application, as shown below. Figure 2 As shown, the method may further include:

[0085] S201. Based on the seismic common reflection point gather, perform principal component analysis on the seismic trace data within the preset incident angle range to construct the principal component trace;

[0086] S202. Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and take the seismic trace with the largest correlation coefficient as the model trace.

[0087] S203. For each seismic trace, a preset time window is slid by a preset distance, and the correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace.

[0088] S204. Based on the remaining time difference of each seismic trace, correct the seismic trace data of that seismic trace to obtain the optimized seismic trace data.

[0089] S205. The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

[0090] Specifically, in step 201, the preset incident angle range can be 0-10 degrees, or it can be selected according to the actual production needs, and there is no restriction on it here; principal component analysis can be used to calculate the average value of seismic data. For example, after acquiring the seismic common reflection point gather, the data of each seismic trace within the 0-10 degree incident angle range are obtained and the average value is calculated to obtain the principal component trace.

[0091] In step 202, the correlation coefficient r between each seismic trace and the principal component trace within a preset time window T is calculated. j for

[0092]

[0093] Where j represents the j-th time window, n represents the number of sample points selected within time window T, and i represents the i-th sample point; x i y i These represent the coordinates of the sample points, for example, x. i Indicates the earthquake track number, y i Indicates the time value of the sample point; These are the coordinates of the sample points of the principal component traces within a preset time window T. The time window T can be 200-300 ms, or selected according to actual production needs; no restriction is imposed here. The maximum correlation coefficient within each time window of each seismic trace is taken as the correlation coefficient of the seismic trace, and the seismic trace with the highest correlation coefficient is taken as the model trace. For example, suppose that for a certain seismic trace, the correlation coefficients with the principal component traces within the time window T = 200 ms are calculated as r1, r2, r3, r4, r5, r6, r7, r8, r9, r1, r1, r9, r1, r2, r1, r2, r3, r1, r2, r3, r4, r1, r2, r3, r1, r2, r3, r4, r1, r2, r3, r4, r1, r2, r3, r4, r1, r2, r3, r4, r1, r2, r3, r4, r1, r2, r3, r4, r1, r2, r3, r4, r2, r3, r4, r2, r3, r4, r3, r4, r5, r6, r7, r8, r1, r1, r2, r3, r2, r3, r4, r1 ... 2，…， r J , where r k If the maximum value is found, then the correlation coefficient of the seismic trace is r. k Assume the correlation coefficient r of this seismic trace k If the correlation coefficient is greater than that of other seismic traces, then the seismic trace is determined to be a model trace.

[0094] In step 203, for each seismic trace, a preset time window T is slid by a preset distance t to calculate the correlation coefficient r between the seismic trace and the model trace under different slip numbers. j The calculation formula is the same as above. The slip number corresponding to the largest correlation coefficient of each seismic trace is taken as the remaining time difference of the current seismic trace.

[0095] In step 204, the correction of the seismic trace data includes leveling the time data of the seismic trace based on the remaining time difference of the seismic trace to obtain optimized seismic trace data, thereby flattening the in-phase axis in the trace data and improving the quality of the seismic wave waveform.

[0096] Figure 3 An example diagram of the time profile of raw seismic gather data provided in an embodiment of this application. Figure 4 An example diagram of an optimized seismic gather data time profile provided in an embodiment of this application, as shown below. Figure 3 , Figure 4 As shown, the original seismic gather data has the problem of uneven gathers, while the in-phase axes of the optimized seismic gather data obtained after correction processing are basically parallel.

[0097] In step 205, the optimized seismic trace data obtained from the leveling are superimposed to obtain the trace aggregation data.

[0098] Figure 5 This is an example diagram of a time profile of gather overlay data provided in an embodiment of this application. Figure 4 Multiple sets of optimized seismic gather data from the example are overlaid to obtain the overlaid gather data as follows: Figure 5 As shown.

[0099] Optimizing gather data based on seismic waveform matching improves the quality and signal-to-noise ratio of seismic waves, which is beneficial for subsequent identification and analysis of strike-slip faults.

[0100] In practical applications, there can be multiple methods to determine the strong reflection in-phase axis. In one example, determining the distribution range and waveform characteristics of the strong reflection in-phase axis includes:

[0101] Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data.

[0102] Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of the strong reflection in-phase axis in the overlay data of the gathers are determined. The waveform characteristics include the frequency, wavelength, amplitude and phase of the strong reflection in-phase axis.

[0103] Specifically, based on the overlay data of the trace gathers, the target layer data that may contain strike-slip fault features is determined; extracting the root mean square amplitude attribute from the target layer data may include: calculating the square root of the average of the squares of the amplitudes of each sample point in each preset time window of the seismic trace data in the target layer data, and then obtaining the root mean square amplitude of each seismic trace data in each time window.

[0104] The median amplitude of the target layer data includes the median of the root mean square amplitudes in the target layer data. The phase axis is the line connecting the extreme values ​​(peaks or troughs) of the vibration phase of each seismic trace in the seismic record. Determining the distribution range of strong reflection phase axes in the stacked gather data can include: connecting the peaks of the vibration phase of each seismic trace in the stacked gather data to obtain the phase axis; obtaining the amplitude of each phase axis, and taking the phase axis with an amplitude greater than the median root mean square amplitude of the target layer data as the strong reflection phase axis.

[0105] For example, after determining the target layer data from the gather stacking data, the root mean square amplitude attribute is extracted from the target layer data, and the root mean square amplitude (RMS) of each seismic trace within the time window is calculated as follows:

[0106]

[0107] Where N is the number of sample points within the time window, and A i Let be the amplitude value of the i-th sample point within the time window.

[0108] With RMS Md This represents the root mean square amplitude median. Assume that the current gather overlay data includes four in-phase axes, namely To1, To2, To3, and To4, where the amplitude of To3 is greater than the RMS amplitude. Md Then To3 is determined to be a strong reflection in-phase axis, and the frequency f, wavelength λ, amplitude A, and phase of To3 are obtained.

[0109] To remove the influence of strong seismic reflection energy characteristics on strike-slip faults, in one example, the step of decomposing the gather stacked data into multiple wavelets and removing the wavelets whose waveform characteristics are the same as those of the strong reflection phase axis may include:

[0110] Based on the seismic trace multi-wavelet decomposition method, each seismic trace signal in the stacked gather data is decomposed into multiple wavelets;

[0111] Remove the wavelets whose frequency, wavelength, amplitude, and phase are the same as those of the strongly reflecting phase axis from the multiple wavelets.

[0112] Specifically, based on multi-wavelet seismic trace decomposition technology, each seismic trace signal in the gather stack data is decomposed into a set of multiple wavelets, which have different dominant frequencies, wavelet widths, and time positions. From the decomposed wavelet sets, wavelets with waveform characteristics identical to those of the strong reflection phase axis are removed. In practical applications, when obtaining wavelets with waveform characteristics identical to those of the strong reflection phase axis from the wavelet set, the interpreted strong reflection layer can be selected as a constraint, the layer time can be used as the initial time delay of the wavelet, and the dominant frequency and phase of the strong reflection layer wavelet can be used as the initial scanning parameters. This can significantly reduce the running time when processing data.

[0113] For example, suppose we obtain the strong reflection in-phase axis To3 and its frequency f, wavelength λ, amplitude A, and phase from the overlay data of the gathers. Each seismic trace in the overlay data is decomposed into a wavelet set. Within each wavelet set, a wavelet with frequency f, wavelength λ, amplitude A, and phase f is selected. The wavelets are removed to eliminate the reflection characteristics of strong reflection phase axes in the gather stacked data. The wavelets with frequencies between 4Hz and 60Hz from the remaining wavelets corresponding to each seismic trace are stacked to obtain the reconstructed seismic signal.

[0114] Figure 6 This is an example diagram of a reconstructed seismic data time profile provided in an embodiment of this application. Figure 5The gather stacking data shown above is processed to remove the influence of strong reflection phase axes, resulting in the reconstructed seismic data as follows. Figure 6 As shown, the strong reflection characteristics of the strata are removed, thus highlighting the seismic characteristics of the strike-slip fault.

[0115] In practical applications, there can be various methods to determine the strike-slip fault reflection signal and the background reflection signal. In one example, determining the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal can include:

[0116] In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which remain stable with time, are identified as strike-slip fault reflection signals; wherein the first preset value is lower than the second preset value.

[0117] Specifically, in the time-frequency spectrum, the strike-slip fault reflection signal and the background reflection signal have a significant time-frequency difference, with the frequency of the strike-slip fault reflection signal being higher than that of the background reflection signal. Therefore, in the time-frequency seismic signal, based on the time-frequency characteristics of the strike-slip fault reflection signal and the background reflection signal, a first preset value and a second preset value are preset to separate the strike-slip fault reflection signal and the background reflection signal.

[0118] For each seismic trace in the time-spectrum seismic signal, signal segments with frequencies below a first preset value are marked as background reflection signals, and signal segments with frequencies above a second preset value are marked as strike-slip fault reflection signals. Signal segments with frequencies above the first preset value and below the second preset value are boundary signals, which can be determined as either background reflection signals or strike-slip fault reflection signals through time-division and frequency-division calculation. The time-division and frequency-division calculation includes determining whether the frequency of a boundary signal segment remains stable over time. If the frequency of the segment remains stable over time, it is marked as a strike-slip fault reflection signal; if the frequency decreases over time, it is marked as a background reflection signal.

[0119] For example, Figure 7 This is a flowchart illustrating the process of determining background reflection signals and strike-slip fracture reflection signals according to an embodiment of this application, as shown below. Figure 7As shown, set the first preset value as f1 and the second preset value as f2. For the frequency f of each point in the time-frequency spectrum seismic signal, if f < f1, mark this point as a background reflection signal; if f > f2, mark this point as a strike-slip fault reflection signal; if f1 < f < f2, determine this point as a boundary signal for further analysis. If the frequency f decreases as time increases, mark this point as a background reflection signal; if the frequency f remains stable as time increases, mark this point as a strike-slip fault reflection signal, thereby separating the background reflection signal from the strike-slip fault reflection signal.

[0120] Figure 8 This is an example diagram of separating the time profile of seismic signals provided by an embodiment of the present application. As Figure 8 shown, the background reflection signal appears light-colored, the strike-slip fault reflection signal appears dark-colored, and the background reflection signal and the strike-slip fault reflection signal are separated.

[0121] In practical applications, before predicting and identifying a strike-slip fault, it is necessary to train a strike-slip fault identification library. In one example, Figure 9 This is a schematic flowchart of training a strike-slip fault identification library provided by an embodiment of the present application. As Figure 9 shown, the method may further include:

[0122] S901. Based on the reconstructed seismic signal and the separated seismic signal, determine profiles and plane slices on the formation body, where the profiles and plane slices contain strike-slip fault response characteristics; establish a strike-slip fault identification library based on the profiles and plane slices;

[0123] S902. Train a neural network based on the separated seismic signal corresponding to the profiles and plane slices;

[0124] S903. Based on the spatial association relationship, determine other profiles and plane slices on the formation body according to the profiles and plane slices; input the other profiles and plane slices into the neural network to obtain the strike-slip fault prediction result output by the neural network;

[0125] S904. Modify the strike-slip fault identification library according to the strike-slip fault prediction result until a trained strike-slip fault identification library is obtained.

[0126] Specifically, in step 901, according to the reconstructed seismic signal and the separated seismic signal, select profiles and plane slices containing strike-slip fault characteristics on the formation body as typical spatial positions, and establish labels for each slice, where the label is whether it contains a strike-slip fault. The set of all slices and their corresponding labels is used as the strike-slip fault identification library.

[0127] In step 902, the neural network is trained based on the separated seismic signals used when determining the slices. In one possible implementation, U-Net++ is chosen as the neural network, but other neural networks can also be chosen according to the actual production needs; there are no restrictions on this.

[0128] In step 903, based on the determined profiles and planar slices, other profiles and planar slices are determined on the formation body. These other profiles and planar slices have a spatial relationship with the determined profiles and planar slices. For example, for a given profile slice, based on the depth location containing strike-slip fracture features, a planar slice is created at the corresponding depth on the formation body; the resulting planar slice also contains strike-slip fracture features. The other profiles and planar slices are input into a trained neural network to obtain strike-slip fracture prediction results, where the strike-slip fracture prediction result indicates whether the slice contains a strike-slip fracture.

[0129] In step 904, based on the strike-slip fracture prediction results output by the neural network, the labels of the slices in the strike-slip fracture identification library are corrected to obtain a trained strike-slip fracture identification library. For example, assuming that a slice in the strike-slip fracture identification library is labeled as not containing a strike-slip fracture, other slices associated with the spatial location of that slice are input into the neural network, and the strike-slip fracture prediction result is that it contains a strike-slip fracture, then the label of that slice is modified to contain a strike-slip fracture.

[0130] Furthermore, the generated three-dimensional strike-slip fault seismic bodies can be verified and analyzed to obtain strike-slip fault geological feature seismic bodies. By slicing and color-coded the strike-slip fault geological feature seismic bodies, the results are output. Figure 10 An example image of a seismic body slice illustrating the geological features of a strike-slip fault provided in an embodiment of this application, such as... Figure 10 As shown, the dark areas represent the strong reflection in-phase axis and strike-slip fracture data, in order to further analyze and interpret the strike-slip fracture structure.

[0131] In the strike-slip fault identification method provided in this embodiment, a strong reflection phase axis is identified in the stacked gather data. The stacked gather data is decomposed into multiple wavelets, and wavelets with the same waveform characteristics as the strong reflection phase axis are removed. The remaining wavelets are then stacked to obtain a reconstructed seismic signal, thereby removing the masking effect of the strong energy reflection characteristics of the strong reflection phase axis on the strike-slip fault response characteristics and highlighting the characteristics of the strike-slip fault. The time-spectrum seismic signals corresponding to each seismic trace signal in the reconstructed seismic signal are obtained, and the strike-slip fault reflection signal and background reflection signal are identified. The strike-slip fault reflection signal is enhanced, and the background reflection signal is attenuated. The processed time-spectrum seismic signal is converted into a time-domain signal to obtain a separated seismic signal. The time-frequency characteristic difference between the strike-slip fault reflection and the background reflection is used to separate the background reflection signal from the strike-slip fault reflection signal, removing the masking effect of the surrounding reflection energy on the strike-slip fault. Based on the reconstructed seismic signal and the separated seismic signal, a stratigraphic body is established, and profiles and planar slices are determined on the stratigraphic body. The slices are input into the strike-slip fault identification library to obtain the strike-slip fault identification results of the strata to be identified, avoiding the problems of low efficiency and low accuracy of manual identification. By processing seismic data, the strong reflection in-phase axis and surrounding background reflections are removed from the strike-slip faults, highlighting their characteristics. A model is then used to identify the strike-slip faults, solving the problem of efficiently and accurately identifying small-displacement strike-slip faults.

[0132] Example 2

[0133] Figure 11 This is a schematic diagram of the structure of a slip-slip fracture identification device provided in one embodiment of this application. Figure 11 As shown, the slip-slip fracture identification device provided in this embodiment may include:

[0134] The reconstruction module 111 is used to determine the distribution range and waveform characteristics of strong reflection phase axes in the stacked data based on the stratigraphic data to be identified in the stacked data; decompose the stacked data into multiple wavelets, remove the wavelets whose waveform characteristics are the same as those of the strong reflection phase axes; and superimpose the remaining wavelets to obtain the reconstructed seismic signal.

[0135] The separation module 112 is used to acquire the time-spectrum seismic signal corresponding to each seismic trace signal in the reconstructed seismic signal; determine the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal; enhance the strike-slip fault reflection signal and attenuate the background reflection signal to obtain the processed time-spectrum seismic signal; and perform an inverse Laplace transform on the processed time-spectrum seismic signal to obtain the separated seismic signal.

[0136] The identification module 113 is used to establish a stratigraphic body based on the reconstructed seismic signal; to determine a profile and planar slice on the stratigraphic body based on the reconstructed seismic signal and the separated seismic signal; to input the profile and planar slice into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, wherein the strike-slip fault identification result includes the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value.

[0137] In practical applications, the slip-slip fracture identification device can be implemented through a computer program, such as application software; or it can be implemented as a medium storing the relevant computer program, such as a USB flash drive or cloud drive; or it can be implemented through a physical device that integrates or installs the relevant computer program, such as a chip or server.

[0138] Specifically, the gather stacking data is obtained by optimizing and stacking seismic gather data. The waveform characteristics of strong reflection phase axes in the gather stacking data include frequency, wavelength, amplitude, and phase. The strata above the layer to be identified in the gather stacking data have strong reflection characteristics, which affect the seismic response characteristics of the fault. By decomposing the gather stacking data into multiple wavelets and removing wavelets whose waveform characteristics are the same as those of the strong reflection phase axes, the influence of strong reflection layers on the seismic response data is removed, highlighting the data characteristics corresponding to the strike-slip fault structure. The remaining wavelets after removing some wavelets are stacked to obtain the reconstructed seismic signal with the influence of strong reflection layers removed, which helps in the identification and interpretation of small-displacement strike-slip faults.

[0139] There are various methods to obtain the time-spectrum seismic signals corresponding to each seismic trace in the reconstructed seismic signal, including but not limited to performing an S-transform on the data of each seismic trace in the reconstructed seismic signal. Based on the difference in time-frequency characteristics between the strike-slip fault signal and the background signal in the time-spectrum seismic signal, the strike-slip fault reflection signal and the background reflection signal in the time-spectrum seismic signal are determined. The strike-slip fault reflection signal is enhanced, and the background reflection signal is attenuated, thereby separating the strike-slip fault reflection features from the stratigraphic background reflection. Correspondingly, the processed time-spectrum seismic signal is converted into a time-domain signal, which may include performing an inverse S-transform on the processed time-spectrum seismic signal to obtain the separated seismic signal that separates the strike-slip fault features from the background stratigraphy.

[0140] The established stratigraphic body characterizes the three-dimensional structure of each stratum corresponding to the reconstructed seismic signal. Based on the reconstructed and separated seismic signals, profiles and planar slices are determined on the stratigraphic body. The profiles and planar slices are slices containing strike-slip fault features to be identified. The profiles and planar slices are input into the strike-slip fault identification database to obtain the strike-slip fault identification results of the strata to be identified. The strike-slip fault identification results are reflected as three-dimensional strike-slip fault seismic bodies containing strike-slip fault structures and their corresponding probability values.

[0141] By processing seismic data, the strong reflection in-phase axis and surrounding background reflections are removed from the strike-slip faults, highlighting their characteristics. A model is then used to identify the strike-slip faults, solving the problem of efficiently and accurately identifying small-displacement strike-slip faults.

[0142] Before identifying strike-slip faults, seismic data needs to be preprocessed. In one example, the device may further include an optimization module for:

[0143] Based on the seismic common reflection point gather, principal component analysis is performed on the seismic trace data within the preset incident angle range to construct the principal component trace;

[0144] Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and take the seismic trace with the largest correlation coefficient as the model trace;

[0145] For each seismic trace, a preset time window is slid by a preset distance. The correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace.

[0146] Based on the remaining time difference of each seismic trace, the seismic trace data of that seismic trace is corrected to obtain the optimized seismic trace data.

[0147] The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

[0148] Specifically, the preset incident angle range can be 0-10 degrees, or it can be selected according to the actual production needs; there are no restrictions on this. Principal component analysis can be used to calculate the average value of seismic data. For example, after acquiring the seismic common reflection point gather, the data of each seismic trace within the 0-10 degree incident angle range are obtained and the average value is calculated to obtain the principal component trace.

[0149] Calculate the correlation coefficient r between each seismic trace and the principal component trace within a preset time window T. j for

[0150]

[0151] Where j represents the j-th time window, n represents the number of sample points selected within time window T, and i represents the i-th sample point; x i y i These represent the coordinates of the sample points, for example, x. i Indicates the earthquake track number, y i Indicates the time value of the sample point; These represent the coordinates of sample points within a preset time window T for each principal component trace. The time window T can range from 200 to 300 ms, and can be selected based on actual production needs; no restriction is imposed here. The maximum correlation coefficient within each time window of each seismic trace is taken as the seismic trace's correlation coefficient, and the seismic trace with the highest correlation coefficient is selected as the model trace.

[0152] For each seismic trace, a preset time window T is slid by a preset distance t, and the correlation coefficient r between the seismic trace and the model trace is calculated under different slip numbers. j The calculation formula is the same as above. The slip number corresponding to the largest correlation coefficient of each seismic trace is taken as the remaining time difference of the current seismic trace.

[0153] Correcting seismic trace data involves calibrating the time data of the seismic traces based on the remaining time difference, thereby flattening the in-phase axes in the trace data and improving the quality of the seismic wave waveform.

[0154] The optimized seismic trace data obtained from the leveling process are then superimposed to obtain the trace aggregation data.

[0155] Optimizing gather data based on seismic waveform matching improves the quality and signal-to-noise ratio of seismic waves, which is beneficial for subsequent identification and analysis of strike-slip faults.

[0156] In practical applications, there can be multiple methods to determine the strong reflection phase axis. In one example, the reconstruction module can specifically be used for:

[0157] Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data.

[0158] Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of the strong reflection in-phase axis in the overlay data of the gathers are determined. The waveform characteristics include the frequency, wavelength, amplitude and phase of the strong reflection in-phase axis.

[0159] Specifically, based on the overlay data of the trace gathers, the target layer data that may contain strike-slip fault features is determined; extracting the root mean square amplitude attribute from the target layer data may include: calculating the square root of the average of the squares of the amplitudes of each sample point in each preset time window of the seismic trace data in the target layer data, and then obtaining the root mean square amplitude of each seismic trace data in each time window.

[0160] The median amplitude of the target layer data includes the median of the root mean square amplitudes in the target layer data. The phase axis is the line connecting the extreme values ​​(peaks or troughs) of the vibration phase of each seismic trace in the seismic record. Determining the distribution range of strong reflection phase axes in the stacked gather data can include: connecting the peaks of the vibration phase of each seismic trace in the stacked gather data to obtain the phase axis; obtaining the amplitude of each phase axis, and taking the phase axis with an amplitude greater than the median root mean square amplitude of the target layer data as the strong reflection phase axis.

[0161] To remove the influence of strong seismic reflection energy characteristics on strike-slip faults, in one example, the separation module can be used to:

[0162] Based on the seismic trace multi-wavelet decomposition method, each seismic trace signal in the stacked gather data is decomposed into multiple wavelets;

[0163] Remove the wavelets whose frequency, wavelength, amplitude, and phase are the same as those of the strongly reflecting phase axis from the multiple wavelets.

[0164] Specifically, based on multi-wavelet seismic trace decomposition technology, each seismic trace signal in the gather stack data is decomposed into a set of multiple wavelets, which have different dominant frequencies, wavelet widths, and time positions. From the decomposed wavelet sets, wavelets with waveform characteristics identical to those of the strong reflection phase axis are removed. In practical applications, when obtaining wavelets with waveform characteristics identical to those of the strong reflection phase axis from the wavelet set, the interpreted strong reflection layer can be selected as a constraint, the layer time can be used as the initial time delay of the wavelet, and the dominant frequency and phase of the strong reflection layer wavelet can be used as the initial scanning parameters. This can significantly reduce the running time when processing data.

[0165] In practical applications, there can be multiple methods to determine the strike-slip fracture reflection signal and the background reflection signal. In one example, the separation module can also be used for:

[0166] In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which remain stable with time, are identified as strike-slip fault reflection signals; wherein the first preset value is lower than the second preset value.

[0167] Specifically, in the time-frequency spectrum, the strike-slip fault reflection signal and the background reflection signal have a significant time-frequency difference, with the frequency of the strike-slip fault reflection signal being higher than that of the background reflection signal. Therefore, in the time-frequency seismic signal, based on the time-frequency characteristics of the strike-slip fault reflection signal and the background reflection signal, a first preset value and a second preset value are preset to separate the strike-slip fault reflection signal and the background reflection signal.

[0168] For each seismic trace in the time-spectrum seismic signal, signal segments with frequencies below a first preset value are marked as background reflection signals, and signal segments with frequencies above a second preset value are marked as strike-slip fault reflection signals. Signal segments with frequencies above the first preset value and below the second preset value are boundary signals, which can be determined as either background reflection signals or strike-slip fault reflection signals through time-division and frequency-division calculation. The time-division and frequency-division calculation includes determining whether the frequency of a boundary signal segment remains stable over time. If the frequency of the segment remains stable over time, it is marked as a strike-slip fault reflection signal; if the frequency decreases over time, it is marked as a background reflection signal.

[0169] In practical applications, a strike-slip fracture identification database needs to be trained before predicting and identifying strike-slip fractures. In one example, the method may further include a training module for:

[0170] Based on the reconstructed seismic signal and the separated seismic signal, profiles and planar slices are determined on the stratigraphic body, wherein the profiles and planar slices contain strike-slip fault response characteristics; based on the profiles and planar slices, a strike-slip fault identification database is established;

[0171] The neural network is trained based on the separated seismic signals corresponding to the profile and planar slices.

[0172] Based on spatial correlation, other cross sections and planar slices are determined on the stratigraphic body according to the cross section and planar slice; the other cross sections and planar slices are input into the neural network to obtain the strike-slip fault prediction results output by the neural network;

[0173] Based on the strike-slip fracture prediction results, the strike-slip fracture identification library is modified until a well-trained strike-slip fracture identification library is obtained.

[0174] Specifically, based on the reconstructed and separated seismic signals, sections and planar slices containing strike-slip fault features are selected on the stratigraphic body as typical spatial locations, and a label is established for each slice, where the label indicates whether it contains a strike-slip fault. The set of all slices and their corresponding labels is used as a strike-slip fault identification library.

[0175] The neural network is trained based on the separated seismic signals used to determine the slices. In one possible implementation, U-Net++ is chosen as the neural network, but other neural networks can also be selected according to the actual production needs; there are no restrictions on this.

[0176] Based on the established cross-sections and planar slices, other cross-sections and planar slices are determined on the stratigraphic body. These other cross-sections and planar slices have a spatial relationship with the established cross-sections and planar slices. For example, for a given cross-section slice, based on the depth location where strike-slip fracture features are present, a planar slice is created at the corresponding depth on the stratigraphic body; the resulting planar slice also contains strike-slip fracture features. The other cross-sections and planar slices are then input into a trained neural network to obtain strike-slip fracture prediction results, where the strike-slip fracture prediction result indicates whether the slice contains a strike-slip fracture.

[0177] Based on the strike-slip fracture prediction results output by the neural network, the labels of the slices in the strike-slip fracture identification library are corrected to obtain a trained strike-slip fracture identification library.

[0178] In the strike-slip fault identification device provided in this embodiment, a strong reflection phase axis is determined in the stacked gather data. The stacked gather data is decomposed into multiple wavelets, and wavelets with the same waveform characteristics as the strong reflection phase axis are removed. The remaining wavelets are superimposed to obtain a reconstructed seismic signal, thereby removing the masking effect of the strong energy reflection characteristics of the strong reflection phase axis on the strike-slip fault response characteristics and highlighting the characteristics of the strike-slip fault. The time-spectrum seismic signal corresponding to each seismic trace signal in the reconstructed seismic signal is obtained, and the strike-slip fault reflection signal and background reflection signal are determined. The strike-slip fault reflection signal is enhanced, and the background reflection signal is attenuated. The processed time-spectrum seismic signal is converted into a time-domain signal to obtain a separated seismic signal. The difference in time-frequency characteristics between the strike-slip fault reflection and the background reflection is used to separate the background reflection signal from the strike-slip fault reflection signal, removing the masking effect of the surrounding reflection energy on the strike-slip fault. Based on the reconstructed seismic signal and the separated seismic signal, a stratigraphic body is established, and profiles and planar slices are determined on the stratigraphic body. The slices are input into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, avoiding the problems of low efficiency and low accuracy of manual identification. By processing seismic data, the strong reflection in-phase axis and surrounding background reflections are removed from the strike-slip faults, highlighting their characteristics. A model is then used to identify the strike-slip faults, solving the problem of efficiently and accurately identifying small-displacement strike-slip faults.

[0179] Example 3

[0180] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure, such as... Figure 12 As shown, the electronic device includes:

[0181] The electronic device includes a processor 291 and a memory 292; it may also include a communication interface 293 and a bus 294. The processor 291, memory 292, and communication interface 293 can communicate with each other via the bus 294. The communication interface 293 can be used for information transmission. The processor 291 can invoke logical instructions stored in the memory 292 to execute the methods of the above embodiments.

[0182] Furthermore, the logic instructions in the aforementioned memory 292 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium.

[0183] The memory 292, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor 291 executes functional applications and data processing by running the software programs, instructions, and modules stored in the memory 292, thereby implementing the methods in the above-described method embodiments.

[0184] The memory 292 may include a program storage area and a data storage area. The program storage area may store the operating system and application programs required for at least one function; the data storage area may store data created based on the use of the terminal device. Furthermore, the memory 292 may include high-speed random access memory and may also include non-volatile memory.

[0185] This disclosure provides a non-transitory computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the methods described in the foregoing embodiments.

[0186] Example 4

[0187] This disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the methods provided in any of the embodiments described above.

[0188] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0189] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A strike-slip fault identification method, characterized by, include: Based on the unidentified strata data in the overlay data, the distribution range and waveform characteristics of strong reflection phase axes in the overlay data are determined. The gathered stacked data is decomposed into multiple wavelets, each with a different dominant frequency, wavelet width, and time position. Wavelets whose waveform characteristics are identical to those of the strong reflection phase axis are removed. These waveform characteristics include frequency, wavelength, amplitude, and phase. The remaining wavelets are then stacked to obtain the reconstructed seismic signal. During the removal process, the interpreted strong reflection layer is used as a constraint, the layer time is used as the initial time delay of the wavelets, and the dominant frequency and phase of the strong reflection layer wavelets are used as initial scanning parameters. The time-spectrum seismic signal corresponding to each seismic trace in the reconstructed seismic signal is obtained by performing an S-transform on each seismic trace signal. The strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal are determined; the strike-slip fault reflection signal is enhanced and the background reflection signal is attenuated to obtain the processed time-spectrum seismic signal; the processed time-spectrum seismic signal is subjected to an inverse S-transform to obtain the separated seismic signal. Based on the reconstructed seismic signal, a stratigraphic body is established; based on the reconstructed seismic signal and the separated seismic signal, a profile and planar slice are determined on the stratigraphic body; the profile and planar slice are input into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, the strike-slip fault identification result including the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value. The determination of the strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal includes: In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals. Signal segments with frequencies higher than a first preset value and lower than a second preset value, and which remain stable over time, are identified as strike-slip fracture reflection signals; wherein the first preset value is lower than the second preset value. The determination of the distribution range and waveform characteristics of the strong reflection in-phase axis includes: Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data. Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of strong reflection in-phase axes in the overlay data of the gathers are determined.

2. The method of claim 1, wherein, The method further includes: Based on the seismic common reflection point gather, principal component analysis is performed on the seismic trace data within the preset incident angle range to construct the principal component trace; Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and take the seismic trace with the largest correlation coefficient as the model trace; For each seismic trace, a preset time window is slid by a preset distance. The correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace. Based on the remaining time difference of each seismic trace, the seismic trace data of that seismic trace is corrected to obtain the optimized seismic trace data. The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

3. The method of claim 1, wherein, The step of decomposing the overlay data of the gather into multiple wavelets and removing the wavelets whose waveform characteristics are the same as those of the strong reflection in-phase axis includes: Based on the seismic trace multi-wavelet decomposition method, each seismic trace signal in the stacked gather data is decomposed into multiple wavelets; Remove the wavelets whose frequency, wavelength, amplitude, and phase are the same as those of the strongly reflecting phase axis from the multiple wavelets.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Based on the reconstructed seismic signal and the separated seismic signal, profiles and planar slices are determined on the stratigraphic body, wherein the profiles and planar slices contain strike-slip fault response characteristics; based on the profiles and planar slices, a strike-slip fault identification database is established; The neural network is trained based on the separated seismic signals corresponding to the profile and planar slices. Based on spatial correlation, other cross sections and planar slices are determined on the stratigraphic body according to the cross section and planar slice; the other cross sections and planar slices are input into the neural network to obtain the strike-slip fault prediction results output by the neural network; Based on the strike-slip fracture prediction results, the strike-slip fracture identification library is modified until a well-trained strike-slip fracture identification library is obtained.

5. A strike-slip fault identification apparatus, characterized by, include: The reconstruction module is used to determine the distribution range and waveform characteristics of strong reflection phase axes in the overlay data based on the stratigraphic data to be identified in the overlay data. The gathered stacked data is decomposed into multiple wavelets, each with a different dominant frequency, wavelet width, and time position. Wavelets whose waveform characteristics are identical to those of the strong reflection phase axis are removed. These waveform characteristics include frequency, wavelength, amplitude, and phase. The remaining wavelets are then stacked to obtain the reconstructed seismic signal. During the removal process, the interpreted strong reflection layer is used as a constraint, the layer time is used as the initial time delay of the wavelets, and the dominant frequency and phase of the strong reflection layer wavelets are used as initial scanning parameters. The separation module is used to perform S-transform on each seismic trace signal in the reconstructed seismic signal to obtain the time-spectrum seismic signal corresponding to each seismic trace signal; The strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal are determined; the strike-slip fault reflection signal is enhanced and the background reflection signal is attenuated to obtain the processed time-spectrum seismic signal; the processed time-spectrum seismic signal is subjected to an inverse S-transform to obtain the separated seismic signal. The identification module is used to establish a stratigraphic body based on the reconstructed seismic signal; to determine a profile and planar slice on the stratigraphic body based on the reconstructed seismic signal and the separated seismic signal; and to input the profile and planar slice into the strike-slip fault identification library to obtain the strike-slip fault identification result of the stratum to be identified, wherein the strike-slip fault identification result includes the three-dimensional strike-slip fault seismic body of the stratum to be identified and its corresponding probability value. The determination of the strike-slip fault reflection signal and background reflection signal in the time-spectrum seismic signal includes: In the time-spectrum seismic signal, signal segments with frequencies lower than a first preset value are identified as background reflection signals; signal segments with frequencies higher than a second preset value are identified as strike-slip fault reflection signals; and signal segments with frequencies higher than the first preset value and lower than the second preset value, and which decrease with time, are identified as background reflection signals. Signal segments with frequencies higher than a first preset value and lower than a second preset value, and which remain stable over time, are identified as strike-slip fracture reflection signals; wherein the first preset value is lower than the second preset value. Specifically, the reconstruction module is used for: Based on the overlay data of the gathers, the target layer data is determined, and the root mean square amplitude attribute is extracted from the target layer data. Based on the data in the overlay data of the gathers whose amplitude is greater than the median root mean square amplitude of the target layer data, the distribution range and waveform characteristics of strong reflection in-phase axes in the overlay data of the gathers are determined.

6. The apparatus of claim 5, wherein, The device further includes an optimization module, used for: Based on the seismic common reflection point gather, principal component analysis is performed on the seismic trace data within the preset incident angle range to construct the principal component trace; Calculate the correlation coefficient between each seismic trace and the principal component trace within a preset time window, and select the seismic trace with the largest correlation coefficient as the model trace; For each seismic trace, a preset time window is slid by a preset distance. The correlation coefficient between the seismic trace and the model trace within the preset time window is calculated at different sliding numbers. The sliding number with the largest correlation coefficient is taken as the remaining time difference of the seismic trace. Based on the remaining time difference of each seismic trace, the seismic trace data of that seismic trace is corrected to obtain the optimized seismic trace data. The optimized seismic trace data of each seismic trace are superimposed to obtain the trace set superposition data.

7. An electronic device, comprising: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-4.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-4.

Images