Fracture enhanced filtering-based hidden fracture intelligent identification method and system
By combining guided filtering iteration and fracture enhancement filtering with a three-dimensional fault body model, the problem of identifying small-scale hidden fractures was solved, enabling accurate early warning and drilling cost control.
Patent Information
- Application Number
- CN202410609276.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-16
- Publication Date
- 2025-11-18
AI Technical Summary
Existing technologies are unable to effectively identify small-scale, hidden fractures, leading to severe well leakage and fracturing fluid loss during drilling, which increases drilling costs and complexity.
By combining guided filtering iteration, fracture enhancement filtering, and a three-dimensional fault body model, along with continuous wavelet transform, SVD matrix singular value decomposition, and multi-window coherence dip scanning, seismic data is optimized to identify small-scale faults.
It improves the identification accuracy of small-scale hidden fractures, reduces multiple solutions, enables accurate early warning of complex engineering events, and reduces drilling risks and costs.
Smart Images
Figure CN120972238A_ABST
Abstract
Description
Technical Field
[0001] This disclosure belongs to the field of geophysical exploration technology, and specifically relates to a method and system for intelligent identification of concealed fractures based on fracture enhancement filtering. Background Technology
[0002] In recent years, with the increasing development efforts in the Junggar Basin, horizontal wells have become the main method for increasing oil and gas reserves and production, leading to a year-on-year increase in the number of wells drilled. However, due to factors such as fractures and cracks, horizontal wells are frequently plagued by complex engineering problems such as well leakage and hydraulic fracturing interference, resulting in a significant increase in drilling cycles, severe fracturing fluid loss, and persistently high costs per well. Taking the Junggar Basin as an example, the leakage rate in the main development blocks of the basin is as high as 40%, and well leakage accounts for 70%-80% of complex well situations, severely restricting the development process.
[0003] Currently, mainstream fracture identification methods, based on the selection of earthquakes with sensitive azimuths, utilize eigenvalue algorithms to characterize fractures using seismic-sensitive attributes such as coherence and curvature. From a technical perspective, eigenvalue coherence primarily involves constructing a covariance matrix using seismic data within a coherence window after dip scanning. The coherence value is then estimated using the eigenvalues of the planned covariance matrix. This method shows significant identification effectiveness for large-scale fractures with obvious displacements, but it suffers from high ambiguity and difficulty in characterizing small-scale, hidden fractures. Curvature attributes mainly reflect the degree of curvature of an arc. Compared to coherence, it improves the accuracy of identifying mesoscale fractures, but it cannot effectively identify small-scale, hidden fractures. Therefore, how to tap the potential of the "two widths and one height" three-dimensional model, accurately characterize small-scale fractures, achieve early warning of well leakage sections, and reduce complex engineering events such as well leakage during drilling has become a pressing issue for efficient development. Summary of the Invention
[0004] To address the above problems, this disclosure provides a method for intelligent identification of concealed fractures based on fracture enhancement filtering, the method comprising:
[0005] Acquire several earthquake data points for the target area;
[0006] The seismic data is subjected to guided filtering iterative processing and fault enhancement filtering processing to obtain the target seismic data;
[0007] The target seismic data is identified by a pre-constructed three-dimensional fault body model to obtain prediction results, and the fault region of the target area is determined based on the prediction results.
[0008] In a preferred embodiment of this solution, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the following steps are included:
[0009] Determine the overlay scheme and perform azimuth overlay processing on the seismic data;
[0010] The seismic data are filtered based on the plate characteristics of the target region.
[0011] In a preferred embodiment of this solution, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the following steps are included:
[0012] The seismic data is frequency-upgraded using continuous wavelet transform.
[0013] In a preferred embodiment of this solution, the frequency upscaling process of the seismic data using continuous wavelet transform includes:
[0014] The seismic trace time series of the seismic data is decomposed by continuous wavelet transform, and the seismic data is converted from the time domain to the frequency domain.
[0015] The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data. In a preferred embodiment of this solution, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the following steps are included:
[0016] The seismic data is converted into matrix form, and the non-zero singular values of the seismic data are obtained using SVD matrix singular value decomposition. A non-zero singular value threshold is selected to remove noise from the seismic data.
[0017] In a preferred embodiment of this solution, the seismic data undergoes guided filtering iterative processing, which includes repeatedly performing the following steps:
[0018] Multi-window coherent dip scanning is performed on the seismic data to determine the dip angle and azimuth angle of the strata in the target area;
[0019] The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
[0020] In a preferred embodiment of this solution, the seismic data undergoes fault enhancement filtering processing, including:
[0021] Determine the dip guide body for the seismic data;
[0022] Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined.
[0023] Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing.
[0024] Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data;
[0025] The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
[0026] In a preferred embodiment of this solution, before identifying the target seismic data using a pre-constructed three-dimensional fault model, the following steps are included:
[0027] Frequency bands are divided based on the effective bandwidth of the target seismic data, and the target seismic data is then filtered based on the frequency band division results.
[0028] In a preferred embodiment of this solution, determining the fracture region of the target region based on the prediction result includes:
[0029] The prediction results are fused to obtain the final data volume;
[0030] Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
[0031] This disclosure also proposes a concealed fracture intelligent identification system based on fracture enhancement filtering, the system comprising:
[0032] The acquisition module is used to acquire several earthquake data points for the target area.
[0033] The processing module is used to perform guided filtering iterative processing and fault enhancement filtering processing on the seismic data to obtain the target seismic data;
[0034] The identification module is used to identify the target seismic data through a pre-constructed three-dimensional fault body model, obtain prediction results, and determine the fault region of the target area based on the prediction results.
[0035] In a preferred embodiment of this solution, the system further includes:
[0036] The overlay module is used to determine the overlay scheme and perform azimuth overlay processing on the seismic data;
[0037] The filtering module is used to filter the seismic data based on the plate characteristics of the target region.
[0038] In a preferred embodiment of this solution, the system further includes:
[0039] The frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform.
[0040] In a preferred embodiment of this solution, the frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform, including:
[0041] The frequency upscaling module is used to decompose the seismic trace time series of the seismic data through continuous wavelet transform, and convert the seismic data from the time domain to the frequency domain.
[0042] The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data.
[0043] In a preferred embodiment of this solution, the system further includes:
[0044] The noise reduction module is used to convert the seismic data into matrix form, use SVD matrix singular value decomposition to obtain the non-zero singular values of the seismic data, select a non-zero singular value threshold, and remove noise from the seismic data.
[0045] In a preferred embodiment of this solution, the processing module is used to perform guided filtering iterative processing on the seismic data, including repeatedly executing the following steps:
[0046] The processing module is used to perform multi-window coherent dip scanning on the seismic data to determine the dip angle and azimuth angle of the strata in the target area;
[0047] The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
[0048] In a preferred embodiment of this solution, the processing module is used to perform fault enhancement filtering processing on the seismic data, including:
[0049] The processing module is used to determine the dip angle guide body of the seismic data;
[0050] Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined.
[0051] Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing.
[0052] Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data;
[0053] The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
[0054] In a preferred embodiment of this solution, the system further includes:
[0055] The partitioning module is used to partition the target seismic data into frequency bands based on the effective bandwidth of the target seismic data, and to filter the target seismic data based on the frequency band partitioning results.
[0056] In a preferred embodiment of this solution, the identification module is used to determine the fracture region of the target region based on the prediction result, including:
[0057] The recognition module is used to perform feature fusion on the prediction results to obtain the result data volume;
[0058] Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
[0059] This disclosure also proposes a concealed fracture intelligent identification device based on fracture enhancement filtering, comprising:
[0060] Processor and memory;
[0061] The processor invokes the computer program stored in the memory to execute the concealed fracture intelligent identification method based on fracture enhancement filtering described above.
[0062] This disclosure also proposes a computer-readable storage medium.
[0063] The computer-readable storage medium stores a computer program that, when executed by a processor, enables the processor to perform any of the aforementioned methods for intelligent identification of concealed fractures based on fracture enhancement filtering.
[0064] This disclosure has the following beneficial effects:
[0065] This method has already been industrially applied in the early warning of complex engineering events in the Mahu Depression on the northwestern edge of the Junggar Basin in Xinjiang, achieving good results. The accuracy rate of well leakage early warning reached 70%, and the accuracy rate of pressure string early warning reached 63%, providing geophysical support for drilling acceleration and efficient development. This method fully taps into the potential of "two widths and one height" three-dimensional micro-fracture information, highlighting the combined application of superimposed interpretation processing technology and multi-directional, multi-angle fracture intelligent identification and fusion technology, effectively reducing the ambiguity in the identification of small-scale hidden fractures. Moreover, this method is a complete fracture identification method with strong overall practicality.
[0066] Other features and advantages of this disclosure will be set forth in the following description and will be apparent in part from the description or may be learned by practicing the disclosure. The objects and other advantages of this disclosure may be realized and obtained by means of the structures pointed out in the description and the accompanying drawings. Attached Figure Description
[0067] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0068] Figure 1 A diagram illustrating the concealed fracture intelligent identification method based on fracture enhancement filtering in an embodiment of this disclosure is shown.
[0069] Figure 2 This diagram shows a comparison of the identification effects of data from various locations on the Dajuragou Fault in the embodiments of this disclosure;
[0070] Figure 3 This diagram shows a comparison of iterative experiments using guided filtering in embodiments of this disclosure.
[0071] Figure 4 This invention discloses a cross-sectional view comparing the tomographic enhancement filtering before and after the process in an embodiment of this invention.
[0072] Figure 5 This diagram shows a planar image of the Triassic Baikouquan Formation with multi-data intelligent feature fusion of planar fractures according to an embodiment of this disclosure.
[0073] Figure 6 This illustration shows a cross-section of a complex engineering early warning system in an embodiment of this disclosure.
[0074] Figure 7 This diagram illustrates a concealed fracture intelligent identification system based on fracture enhancement filtering in an embodiment of this disclosure.
[0075] Figure 8 A diagram of a concealed fracture intelligent identification device based on fracture enhancement filtering is shown in an embodiment of this disclosure. Detailed Implementation
[0076] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.
[0077] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware units or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.
[0078] The flowchart shown in the attached diagram is merely an illustrative example and does not necessarily include all steps. For example, some steps may be broken down, while others may be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.
[0079] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein.
[0080] Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such that a process, method, system, product, or device that includes a series of steps or sub-modules is not necessarily limited to those steps or sub-modules that are explicitly listed, but may include other steps or sub-modules that are not explicitly listed or that are inherent to such process, method, product, or device.
[0081] This disclosure proposes a method for identifying concealed fractures based on fracture enhancement filtering. This method fully utilizes the "two widths and one height" three-dimensional wide-azimuth and wide-frequency information, and enhances the original seismic signal-to-noise ratio by constructing a combination of guided filtering, median filtering, and introducing a fracture enhancement filter, thus highlighting fracture boundaries. Based on this, it employs AI fracture identification methods and innovative multi-attribute fusion methods to achieve fine characterization of minute fractures. This method can improve the accuracy of identifying small-scale concealed fractures while reducing ambiguity, enabling precise early warning for complex engineering events.
[0082] like Figure 1 As shown, this disclosure proposes a method for intelligent identification of concealed fractures based on fracture enhancement filtering, the method comprising:
[0083] Acquire several earthquake data points for the target area;
[0084] The seismic data is subjected to guided filtering iterative processing and fault enhancement filtering processing to obtain the target seismic data;
[0085] The target seismic data is identified by a pre-constructed three-dimensional fault body model to obtain prediction results, and the fault region of the target area is determined based on the prediction results.
[0086] Specifically, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the process includes:
[0087] Determine the overlay scheme and perform azimuth overlay processing on the seismic data;
[0088] The seismic data are filtered based on the plate characteristics of the target region.
[0089] Specifically, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the process includes:
[0090] The seismic data is frequency-upgraded using continuous wavelet transform.
[0091] Specifically, the frequency upscaling process of the seismic data using continuous wavelet transform includes:
[0092] The seismic trace time series of the seismic data is decomposed by continuous wavelet transform, and the seismic data is converted from the time domain to the frequency domain.
[0093] The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data.
[0094] Specifically, before performing guided filtering iteration processing and fault enhancement filtering processing on the seismic data, the process includes:
[0095] The seismic data is converted into matrix form, and the non-zero singular values of the seismic data are obtained using SVD matrix singular value decomposition. A non-zero singular value threshold is selected to remove noise from the seismic data.
[0096] Specifically, the seismic data undergoes guided filtering iterative processing, including repeatedly performing the following steps:
[0097] Multi-window coherent dip scanning is performed on the seismic data to determine the dip angle and azimuth angle of the strata in the target area;
[0098] The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
[0099] Specifically, the seismic data undergoes fault enhancement filtering processing, including:
[0100] Determine the dip guide body for the seismic data;
[0101] Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined.
[0102] Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing.
[0103] Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data;
[0104] The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
[0105] Specifically, before identifying the target seismic data using a pre-constructed three-dimensional fault body model, the process includes:
[0106] Frequency bands are divided based on the effective bandwidth of the target seismic data, and the target seismic data is then filtered based on the frequency band division results.
[0107] Specifically, determining the fracture region of the target region based on the prediction results includes:
[0108] The prediction results are fused to obtain the final data volume;
[0109] Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
[0110] This disclosure presents a method for identifying small-scale, concealed fractures based on fracture enhancement filtering, providing crucial support for improving the accuracy of identifying minute fractures and achieving precise early warning of complex engineering events. Taking the Mahu-1 well area in the northwestern margin of the Junggar Basin as an example, this disclosure elaborates on the specific steps in detail:
[0111] S1: Preferred azimuth overlay scheme
[0112] The azimuth stacking scheme primarily considers the stacked imaging effect, optimizing the effective offset by discarding near and far paths while preserving the mid-path. Based on the optimized offset, the main considerations are signal-to-noise ratio, computational speed, and the adequacy of ellipse fitting.
[0113] The optimal azimuth subdivision scheme for the fine three-dimensional test of Mahu-1 well was selected. The ellipse fitting effect of 4 azimuth and 6 azimuth was the best. The 6 azimuth can better highlight the characteristics of fractures with different strikes. Therefore, under the existing gather data, the data were superimposed into 6 azimuths.
[0114] Based on this, stress analysis was conducted in the Mahu-1 well area. This area is influenced by the collision of the Kazakhstani Plate to the northwest and the compression of the Siberian Plate to the northeast. Four sets of faults are developed in the Baikouquan Formation, exhibiting a concentrated, en echelon-like distribution along the Dajuragou fault. Comparative analysis of historical fracturing tests reveals that areas with well-developed R-shear exhibit concentrated, unidirectional microseismic events, significantly interfering with adjacent wells and resulting in poor fracturing performance. Areas with well-developed R'-shear tend to form an artificial fracture network structure after fracturing, leading to relatively better fracturing results. Therefore, data perpendicular to the R-shear and the main fault orientation from the six azimuth data points were prioritized for subsequent filtering analysis.
[0115] Figure 2 This is a north-south seismic geological interpretation profile of the Dajuragou Fault in the Guomahu 1 well area. AZ1 is located on the left, AZ2 is in the middle, and AZ3 is on the right.
[0116] S2: Frequency upsampling based on continuous wavelet transform seismic data
[0117] The three-dimensional acquisition parameters of the Mahu-1 well area are relatively weak, with a dominant frequency of only 30Hz. The data quality is insufficient to meet the requirements for characterizing small faults. It is necessary to use the continuous wavelet transform method to increase the frequency of the seismic data in each azimuth.
[0118] The specific implementation steps are as follows:
[0119] ① The seismic trace time series is decomposed using continuous wavelet transform, and the seismic data is converted from the time domain to the frequency domain;
[0120] ② The seismic data is frequency-spread at both the low-frequency and high-frequency ends, and time-frequency conversion is performed.
[0121] By using continuous wavelet transform, the original seismic frequency was increased, raising the three-dimensional target layer frequency of the Mahu-1 well from 30Hz to 35Hz.
[0122] S3: Noise Reduction Based on Singular Value Decomposition of SVD Matrix
[0123] The 3D seismic data is converted into matrix form, and non-zero singular values are obtained using SVD matrix singular value decomposition. A threshold value for non-zero singular values is selected to achieve denoising of the 3D seismic data.
[0124] S4: Iterative processing based on constructed guided filter
[0125] Constructing guided filtering can effectively enhance the signal-to-noise ratio of 3D seismic data, improve the continuity and discontinuity characteristics of co-axial data, while maintaining the original effective seismic frequency and longitudinal resolution, and highlighting the response characteristics of small fractures.
[0126] The specific implementation steps are as follows:
[0127] ① Conduct multi-window coherence dip scanning of 3D seismic data;
[0128] ② Utilize the dip angle and azimuth angle of the strata to perform directional analysis of the rock strata;
[0129] ③ Analyze the characteristics of discontinuous boundaries, retain meaningful geological boundaries, and smooth out noise and other interference features.
[0130] ④ Repeat the above operation process to perform three iterations of guided filtering on the data volume.
[0131] Figure 3 The image shows a comparison of iterative experiments using structurally guided filtering. The left side is the original seismic profile, the middle side is the seismic profile with primary structurally guided filtering, and the right side is the seismic profile with tertiary structurally guided filtering. As can be seen from the image, tertiary structurally guided filtering has the best effect on fracture identification.
[0132] S5: Fracture Enhancement Filter
[0133] Based on the construction of guided filtering iterative processing, we will fully tap the potential of seismic data and continue to carry out fault enhancement filtering processing to further improve the ability of data to identify small-scale faults.
[0134] ① Calculate the dip guide body for the current seismic data (calculate the seismic dip information at each seismic sampling point);
[0135] ② Under the control of the tilt guide body, small step tilt guide median filtering is carried out to perform small-scale denoising and smoothing on the data volume;
[0136] ③ Under the control of the tilt guide body, large step tilt guide median filtering is carried out to perform large-scale denoising and smoothing on the data volume;
[0137] ④ Based on the small step tilt angle guide body, calculate the minimum coherence volume of the data;
[0138] ⑤ Under the control of the tilt guide body, tilt guide diffusion filtering is carried out to enhance the display effect of fault edges;
[0139] ⑥ An adjustment algorithm is added to adjust the ratio of fractured to non-fractured components using the coherence threshold value, so as to obtain the final fracture-enhanced filter.
[0140] Figure 4 The images show a comparison of seismic profiles before and after fault enhancement filtering. The left side shows the original seismic profile, and the right side shows the seismic profile after fault enhancement filtering. The comparison shows that the fault identification capability of the seismic data is significantly improved after the fault enhancement filtering process.
[0141] S6: Effective Frequency Band Optimization
[0142] The target 3D model is a fine 3D model of the pilot test in the Mahu 1 well area. The effective bandwidth of the target layer is 5-65Hz. Low-frequency information can reflect the characteristics of large faults, while appropriate high-frequency information can highlight the distribution of small faults. Therefore, it is necessary to optimize the dominant frequency bands to improve the identification accuracy of faults at each scale.
[0143] Based on the analysis of this three-dimensional single-frequency volume, it is believed that 10Hz seismic data can better identify large-scale faults, 30Hz seismic data can better identify medium-scale faults, and 60Hz seismic data can better identify small-scale faults. Therefore, the full-band data was optimized into three frequency bands: 10Hz, 30Hz, and 60Hz for fault identification.
[0144] S7: AI-powered fracture detection
[0145] Conventional fault interpretation methods mainly rely on human-computer interaction, which can lead to multiple interpretations. This disclosure, based on the post-stack interpretive filtering of seismic data volumes mentioned above, utilizes AI-based deep learning algorithms for intelligent fault identification. This method constructs a series of U-net three-dimensional fault volume models and pre-trains the deep learning network of the big data model, thereby reducing human-induced multiple interpretations and improving the accuracy of fault identification.
[0146] S8: Multi-attribute adaptive intelligent fusion
[0147] This disclosure selects nine data volumes from three dominant azimuths and three sensitive single-frequency volumes in the Mahu 1 well area for post-stack interpretive processing and fracture enhancement filtering. In order to highlight the advantages of each azimuth and each dominant frequency band in identifying fractures of different azimuths and scales, it is necessary to adaptively and intelligently fuse the various data volumes.
[0148] Taking two 3D seismic data volumes to be fused as an example, this paper illustrates the steps and algorithms for implementing intelligent feature fusion of multiple data.
[0149] First, analyze the earthquake data S. M×N×O T M×N×O Amplitude equalization is performed, where M, N, and O represent the horizontal and vertical coordinates of the three-dimensional data volume, respectively, and the longitudinal time.
[0150] Based on this, the two data volumes S M×N×O T M×N×O It is split into several sub-modules s of size m×n×o. m×n×o , t m×n×o The larger the value of m×n×o, the smaller the computational cost of fusion and the coarser the result; conversely, the smaller the value, the larger the computational cost of fusion and the finer the details of the resulting data volume. For a single data volume A, the algorithm described below is used to calculate the feature values of each sub-data volume with respect to the entire data volume:
[0151]
[0152]
[0153]
[0154] In the formula, C, D, and E represent the results data S, which are sub-data volumes of size m×n×o. M×N×O Towards, T M×N×O The number of partitions after splitting, that is, the number in the XYZ directions; (c, d, e) represent the labels of the partitions; A CON(c，d，e) represents the feature value of the sub-data body located at (c, d, e) in relation to the result data body; A represents the feature value; f represents the contribution.
[0155] Similarity_V(c, d, e) of submodules:
[0156]
[0157] This algorithm is discussed in two cases:
[0158] If 0.8 < Similarity_V(c, d, e) < 1.2, then:
[0159]
[0160] If 0.8 > Similarity_V(c, d, e) and Similarity_V(c, d, e) > 1.2, then:
[0161]
[0162] S CON(c，d，e) ≥T CON(c，d，e)
[0163] S CON(c，d，e) <T CON(c，d，e) ;
[0164] in,
[0165] γ = 1 - β.
[0166] In the formula, S CON(c，d，e) This represents the feature value of the individual module S located in the (c, d, e) sub-data body relative to the result data body; T CON(c，d，e) This indicates the feature value of the individual module T located in the (c, d, e) sub-data body relative to the result data body; f m×n×o (m×n×o) represents the fused result; β and γ both represent weights; This represents an interval of similarity values for submodules.
[0167] Figure 5 This is a planar image of multi-data intelligent feature fusion of planar fractures in the Triassic Baikouquan Formation. The fused data fractures are more intuitive and prominent, which can effectively overcome the uncertainty caused by single attributes and improve the accuracy of fracture prediction.
[0168] Figure 6 To verify the complexity of engineering projects in the study area using the research findings, a comparison before and after reveals that the innovative results disclosed in this paper have a cleaner fracture attribute background than conventional methods, effectively reducing problems such as multiple solutions in fracture identification. The results also show better agreement with actual well leakage points, significantly improving the success rate of engineering complexity early warning. The well leakage point early warning compliance rate reaches 70%, and the pressure chain early warning compliance rate reaches 63%, indicating that the method is practical and reliable.
[0169] like Figure 7 As shown, this disclosure also proposes a concealed fracture intelligent identification system based on fracture enhancement filtering, the system comprising:
[0170] The acquisition module is used to acquire several earthquake data points for the target area.
[0171] The processing module is used to perform guided filtering iterative processing and fault enhancement filtering processing on the seismic data to obtain the target seismic data;
[0172] The identification module is used to identify the target seismic data through a pre-constructed three-dimensional fault body model, obtain prediction results, and determine the fault region of the target area based on the prediction results.
[0173] Specifically, the system also includes:
[0174] The overlay module is used to determine the overlay scheme and perform azimuth overlay processing on the seismic data;
[0175] The filtering module is used to filter the seismic data based on the plate characteristics of the target region.
[0176] Specifically, the system also includes:
[0177] The frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform.
[0178] Specifically, the frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform, including:
[0179] The frequency upscaling module is used to decompose the seismic trace time series of the seismic data through continuous wavelet transform, and convert the seismic data from the time domain to the frequency domain.
[0180] The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data.
[0181] Specifically, the system also includes:
[0182] The noise reduction module is used to convert the seismic data into matrix form, use SVD matrix singular value decomposition to obtain the non-zero singular values of the seismic data, select a non-zero singular value threshold, and remove noise from the seismic data.
[0183] Specifically, the processing module is used to perform guided filtering iterative processing on the seismic data, including repeatedly executing the following steps:
[0184] The processing module is used to perform multi-window coherent dip scanning on the seismic data to determine the dip angle and azimuth angle of the strata in the target area;
[0185] The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
[0186] Specifically, the processing module is used to perform fault enhancement filtering on the seismic data, including:
[0187] The processing module is used to determine the dip angle guide body of the seismic data;
[0188] Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined.
[0189] Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing.
[0190] Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data;
[0191] The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
[0192] Specifically, the system also includes:
[0193] The partitioning module is used to partition the target seismic data into frequency bands based on the effective bandwidth of the target seismic data, and to filter the target seismic data based on the frequency band partitioning results.
[0194] Specifically, the identification module is used to determine the fracture region of the target region based on the prediction result, including:
[0195] The recognition module is used to perform feature fusion on the prediction results to obtain the result data volume;
[0196] Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
[0197] like Figure 8 As shown, corresponding to the concealed fracture intelligent identification method based on fracture enhancement filtering provided above, this disclosure also provides a concealed fracture intelligent identification device based on fracture enhancement filtering. Since the embodiment of this device is similar to the above method embodiment, the description is relatively simple; relevant details can be found in the description of the above method embodiment section. The device described below is merely illustrative. This device may include: a processor 1, a memory 2, a communication bus (i.e., the aforementioned device bus), and a lookup engine. The processor 1 and memory 2 communicate with each other through the communication bus and communicate with external systems through a communication interface. The processor 1 can call logical instructions in the memory 2 to execute the concealed fracture intelligent identification method based on fracture enhancement filtering.
[0198] Furthermore, the logical instructions in the aforementioned memory 2 can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as memory chips, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0199] On the other hand, this disclosure also provides a processor-readable storage medium storing a computer program 3, which, when executed by a processor 1, is implemented to perform the concealed fracture intelligent identification method based on fracture enhancement filtering provided in the above embodiments.
[0200] The processor-readable storage medium can be any available medium or data storage device that the processor 1 can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).
[0201] Those skilled in the art should understand that, despite the detailed description of this disclosure with reference to the foregoing embodiments, modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.
Claims
1. A method for intelligent identification of concealed fractures based on fracture enhancement filtering, characterized in that, The method includes: Acquire several earthquake data points for the target area; The seismic data is subjected to guided filtering iterative processing and fault enhancement filtering processing to obtain the target seismic data; The target seismic data is identified by a pre-constructed three-dimensional fault body model to obtain prediction results, and the fault region of the target area is determined based on the prediction results.
2. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, Before performing guided filtering iteration and fault enhancement filtering on the seismic data, the process includes: Determine the overlay scheme and perform azimuth overlay processing on the seismic data; The seismic data are filtered based on the plate characteristics of the target region.
3. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, Before performing guided filtering iteration and fault enhancement filtering on the seismic data, the process includes: The seismic data is frequency-upgraded using continuous wavelet transform.
4. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 3, characterized in that, The frequency upscaling process of the seismic data using continuous wavelet transform includes: The seismic trace time series of the seismic data is decomposed by continuous wavelet transform, and the seismic data is converted from the time domain to the frequency domain. The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data.
5. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, Before performing guided filtering iteration and fault enhancement filtering on the seismic data, the process includes: The seismic data is converted into matrix form, and the non-zero singular values of the seismic data are obtained using SVD matrix singular value decomposition. A non-zero singular value threshold is selected to remove noise from the seismic data.
6. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, The seismic data is subjected to guided filtering iterative processing, including repeatedly performing the following steps: Multi-window coherent dip scanning is performed on the seismic data to determine the dip angle and azimuth angle of the strata in the target area; The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
7. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, The seismic data is subjected to fault enhancement filtering processing, including: Determine the dip guide body for the seismic data; Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined. Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing. Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data; The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
8. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, Before identifying the target seismic data using a pre-constructed three-dimensional fault body model, the process includes: Frequency bands are divided based on the effective bandwidth of the target seismic data, and the target seismic data is then filtered based on the frequency band division results.
9. The method for intelligent identification of concealed fractures based on fracture enhancement filtering according to claim 1, characterized in that, Determining the fracture region of the target area based on the prediction results includes: The prediction results are fused to obtain the final data volume; Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
10. A concealed fracture intelligent identification system based on fracture enhancement filtering, characterized in that, The system includes: The acquisition module is used to acquire several earthquake data points for the target area. The processing module is used to perform guided filtering iterative processing and fault enhancement filtering processing on the seismic data to obtain the target seismic data; The identification module is used to identify the target seismic data through a pre-constructed three-dimensional fault body model, obtain prediction results, and determine the fault region of the target area based on the prediction results.
11. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The system also includes: The overlay module is used to determine the overlay scheme and perform azimuth overlay processing on the seismic data; The filtering module is used to filter the seismic data based on the plate characteristics of the target region.
12. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The system also includes: The frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform.
13. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 12, characterized in that, The frequency upscaling module is used to perform frequency upscaling processing on the seismic data through continuous wavelet transform, including: The frequency upscaling module is used to decompose the seismic trace time series of the seismic data through continuous wavelet transform, and convert the seismic data from the time domain to the frequency domain. The seismic data is subjected to frequency upscaling processing at both the low-frequency and high-frequency ends to increase the dominant frequency of the seismic data.
14. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The system also includes: The noise reduction module is used to convert the seismic data into matrix form, use SVD matrix singular value decomposition to obtain the non-zero singular values of the seismic data, select a non-zero singular value threshold, and remove noise from the seismic data.
15. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The processing module is used to perform guided filtering iterative processing on the seismic data, including repeatedly executing the following steps: The processing module is used to perform multi-window coherent dip scanning on the seismic data to determine the dip angle and azimuth angle of the strata in the target area; The dip and azimuth angles of the seismic data are used to perform directional analysis of the rock strata, screen the geological boundaries of the seismic data, and optimize the interference features in the seismic data.
16. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The processing module is used to perform fault enhancement filtering on the seismic data, including: The processing module is used to determine the dip angle guide body of the seismic data; Based on the dip guide body, the seismic data is subjected to small-step dip guide median filtering, the seismic data is subjected to small-scale denoising and smoothing, and the minimum coherence volume of the seismic data is determined. Based on the tilt guide body, the seismic data is subjected to large-step tilt guide median filtering, and the seismic data is subjected to large-scale denoising and smoothing processing. Based on the tilt guide, the seismic data is subjected to tilt-guided diffusion filtering to enhance the edge display effect of the seismic data; The ratio of fractures to non-fractures in the target region is adjusted according to the threshold value of the minimum coherence volume.
17. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The system also includes: The partitioning module is used to partition the target seismic data into frequency bands based on the effective bandwidth of the target seismic data, and to filter the target seismic data based on the frequency band partitioning results.
18. The concealed fracture intelligent identification system based on fracture enhancement filtering according to claim 10, characterized in that, The identification module is used to determine the fracture region of the target region based on the prediction result, including: The recognition module is used to perform feature fusion on the prediction results to obtain the result data volume; Based on the aforementioned data, the fracture regions of the target area at different orientations and scales are determined.
19. A concealed fracture intelligent identification device based on fracture enhancement filtering, characterized in that, include: Processor and memory; The processor invokes the computer program stored in the memory to execute the concealed fracture intelligent identification method based on fracture enhancement filtering as described in any one of claims 1 to 9.
20. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the processor to perform the concealed fracture intelligent identification method based on fracture enhancement filtering as described in any one of claims 1 to 9.