A method for depicting superdeep carbonate rock hillock beach body based on frequency division attribute

Abstract

The application discloses a kind of based on the superdeep layer carbonate rock hillock beach body delineation method of frequency division attribute, the present technical comprehensive application of innovation is used in this method, by the full combination of geological model and seismic prediction method, using the constraint of geological prediction model, reduce the multi-solution of seismic prediction means, solve the limitation of single geology or seismic prediction method, improve the prediction accuracy of hillock beach body plane distribution, can be better used in the exploration and development of hillock beach body reservoir;Traditional method focuses on the conventional attribute delineation in full-band data body, and the prediction accuracy of hillock beach body is limited, and the frequency division attribute based on high-frequency and low-frequency information of seismic data can finely delineate the boundary range of hillock beach body, and also can effectively improve the delineation accuracy of hillock beach body details.

Description

Technical Field

[0001] This invention belongs to the field of petroleum exploration and development, and specifically relates to a method for characterizing ultra-deep carbonate hills and shoals based on frequency division properties. Background Technology

[0002] Microbial mound-shoal complexes (or mound-shoal complexes) are carbonate rock formations composed of microbial rocks and related grain rocks formed by microorganisms capturing and binding clastic sediments. As an important new type of oil and gas resource carrier, they have shown great exploration potential in global oil and gas exploration practices and have become a current research hotspot due to their special sedimentary characteristics and reservoir formation process.

[0003] Current methods for earthquake prediction of carbonate hills and shoals mainly rely on conventional seismic data, employing techniques such as seismic reflection geometry, seismic attributes, and seismic inversion. However, these methods are highly dependent on the quality of the seismic data. Furthermore, due to the generally deep burial of carbonate hills and shoals, the poor quality of seismic data, and the lack of suitable geological models, the reliability of interpreting hill and shoal distribution using seismic data is low, thus limiting the accuracy of hill and shoal distribution prediction results. To address these issues, we have developed a frequency-based method for characterizing ultra-deep carbonate hills and shoals, resolving the aforementioned technical problems. Summary of the Invention

[0004] In order to solve the problems existing in the prior art, the present invention aims to provide a method for characterizing ultra-deep carbonate hills and shoals based on frequency division properties, thereby improving the existing problems in the prior art.

[0005] To solve the technical problem, the technical solution of the present invention is as follows:

[0006] A method for characterizing ultra-deep carbonate hills and shoals based on frequency division properties, the method comprising:

[0007] S1: By loading drilling and logging data into the seismic interpretation area, combined with core and thin section data and field outcrop data into the seismic interpretation area, single-well sequence stratigraphy is performed to establish a sequence stratigraphic framework.

[0008] S2. After calculating and synthesizing seismic records from drilling logging data using acoustic waves and density curves, fine well-seismic calibration is performed using stratigraphic data to obtain time-depth relationships. After establishing a regional seismic framework profile using time-depth relationships, the three-dimensional seismic horizons of the entire region are precisely tracked and interpreted.

[0009] S3. Based on drilling and logging data, field outcrop data, core data, and the established sequence stratigraphic framework, determine the scale, petrophysical parameters, and longitudinal and lateral distribution patterns of the hill and shoal bodies in the study area, and establish a geological prediction model for the hill and shoal bodies.

[0010] S4. Based on the established hill and beach geological prediction model, conduct earthquake forward modeling and obtain earthquake forward modeling results;

[0011] S5. Perform earthquake frequency division processing on the earthquake forward modeling results to determine the earthquake reflection characteristics of hills and shoals of different sizes in different frequency bands. Based on the earthquake reflection characteristics, establish the earthquake response model of the hills and shoals in the study area. Based on the earthquake response model, match the optimal identification frequency corresponding to the hills and shoals of different sizes.

[0012] S6. Obtain the dominant frequency and effective bandwidth of the seismic data in the study area, perform seismic frequency division processing on the three-dimensional post-stack seismic data, and obtain multiple single-band data volumes based on the optimal identification frequency obtained in step S5.

[0013] S7. Based on the three-dimensional seismic horizon of the whole region, attribute extraction and algorithm optimization are performed on each single-frequency band data volume at different frequencies to obtain the optimized frequency-division attributes and sequence stratigraphic thickness.

[0014] S8. After matching and statistically correlating the actual drilling encounters of hill and shoal bodies and the data related to the development of hill and shoal bodies with the selected frequency division attributes and sequence formation thickness, establish the conversion relationship between frequency division attributes and hill and shoal bodies.

[0015] S9. Remove outliers from the frequency division attributes. Based on the conversion relationship between the frequency division attributes and the hills and beaches established in step S8, fuse the frequency division attributes to characterize the planar distribution of the hills and beaches in the study area.

[0016] Furthermore, prior to step S1, the method further includes:

[0017] Acquire basic data within the study area; the basic data includes: field outcrop data, core sampling data, drilling logging data, stratigraphic data, and three-dimensional post-stack seismic data.

[0018] The basic data within the study area were loaded into the seismic interpretation work area, and spectral analysis was performed on the three-dimensional post-stack seismic data loaded into the seismic interpretation work area to obtain the dominant frequency and effective bandwidth of the seismic data in the study area.

[0019] Furthermore, in step S4, during forward modeling, it is ensured that the excitation parameters are consistent with the parameters of the 3D seismic data volume, mainly including: dominant frequency and effective bandwidth information; at the same time, the forward modeling needs to use a bandpass wavelet, and the dominant frequency and bandwidth of the wavelet must be consistent with the parameters of the seismic data volume. The forward modeling results, i.e., the simulated profile, need to be compared with the actual drilling seismic profile to analyze the effect of the forward modeling and optimize and adjust the model parameters.

[0020] Furthermore, in steps S5 and S6, the seismic frequency division method employs a frequency division technique based on the matching pursuit algorithm. The matching pursuit algorithm decomposes any signal into a linear expansion of a set of basis functions, which is a technique for processing highly non-stationary signals. The wavelet signal that best approximates the residual signal is iteratively searched in the constructed ultra-complete wavelet library D, and the original seismic signal is gradually decomposed into the optimal linear superposition of multiple matching wavelets. The selected frequency f0 is used as the center frequency, and the seismic data is reconstructed using the time spectrum with a frequency band range of [(f0-Δf), (f0+Δf)] to obtain the frequency-divided seismic data volume.

[0021] Furthermore, in step S7, the low-frequency volume is used to characterize the phase band boundary of large-scale hills and beaches, while the high-frequency volume is used to sculpt small-scale hills and beaches and their internal details. The root mean square amplitude attribute is used for characterizing the phase band boundary of large-scale hills and beaches, while the waveform clustering attribute is used for sculpting small-scale hills and beaches and their internal details.

[0022] Furthermore, in S8, determining the sedimentary facies type of the hill-shoal body based on seismic attributes is an underdetermined problem. In practical applications, constraints need to be added, using actual drilling data as constraints. For core wells that encounter the hill-shoal body, the corresponding attribute types are determined using geological research results and corresponding logging facies, establishing a correspondence between the two and extrapolating to the un-well area, converting seismic facies into sedimentary facies, thereby achieving the goal of finely characterizing the planar distribution features of the hill-shoal body in the area.

[0023] Furthermore, in S9, outlier values ​​caused by data issues are removed. For areas where the 3D data is not fully covered or where there are fractures or breaks, the data quality is poor, and the extracted attribute values ​​often contain outliers. Based on the actual situation, outliers are removed to create the map, and finally, the fine carving of the hill body is completed.

[0024] Beneficial effects:

[0025] (1) This method is an innovative application of existing technologies. By fully combining geological models and earthquake prediction methods, and utilizing the constraints of geological prediction models, the ambiguity of earthquake prediction methods is reduced. This solves the limitations of single geological or earthquake prediction methods, improves the accuracy of hill and shoal body planar distribution prediction, and can be better applied to the exploration and development of hill and shoal body reservoirs.

[0026] (2) Traditional methods focus on the characterization of conventional attributes within the full-band data body, resulting in limited accuracy of predicted hill and shoal bodies. Frequency-division attributes based on high-frequency and low-frequency information of seismic data can finely characterize the boundary range of hill and shoal bodies and effectively improve the characterization accuracy of hill and shoal body details. Attached Figure Description

[0027] Figure 1 This is a flowchart of the method of the present invention;

[0028] Figure 2 This is a spectral analysis diagram of three-dimensional seismic data in the study area in this embodiment of the invention;

[0029] Figure 3 This is a sequence stratigraphic framework and seismic framework profile of the study area in this embodiment of the invention;

[0030] Figure 4 This is the synthetic record calibration diagram of well Pengtan 1 in this embodiment of the invention;

[0031] Figure 5 This is a comparative diagram of sedimentary deposits in the study area in this embodiment of the invention;

[0032] Figure 6 This is the geological prediction model for hilly areas in this embodiment of the invention;

[0033] Figure 7 The results of the forward modeling of the hill and beach body and its frequency-division narrowband amplitude profile are shown in the embodiments of the present invention.

[0034] Figure 8 This is the RMS amplitude plane diagram (20Hz frequency division amplitude body) of the upper sub-segment of the study area in this embodiment of the invention;

[0035] Figure 9 This is the clustering attribute map of the waveform of the upper sub-segment of lamp 2 in the study area of ​​the embodiment of the present invention (40 Hz frequency division amplitude body);

[0036] Figure 10 This is a stratigraphic thickness map of the upper sub-member of the second lamp in the study area of ​​this invention embodiment;

[0037] Figure 11 This is a planar distribution diagram of the hilly and shoal bodies in the upper sub-segment of the study area in this embodiment of the invention. Detailed Implementation

[0038] The specific implementation of the present invention is described below with reference to embodiments:

[0039] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to complement the content disclosed in the specification for those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0040] Furthermore, the terms such as "upper," "lower," "left," "right," "middle," and "one" used in this specification are merely for clarity of description and are not intended to limit the scope of the invention. Any changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

[0041] Example 1:

[0042] A method for characterizing ultra-deep carbonate hills and shoals based on frequency division properties includes the following steps:

[0043] S1. Basic Data Collection: Collect relevant data within the study area, including: regional geological survey data, field outcrop data, core sampling data, conventional well logging curves, well inclination data, stratigraphic layering data, three-dimensional post-stack seismic data, etc.

[0044] S2. Establish a seismic interpretation work area, load 3D seismic data, wellhead data, well deviation data, layered data, well logging curves, and well logging interpretation results data. At the same time, perform spectral analysis on the 3D seismic data of the study area to obtain the dominant frequency and effective bandwidth of the seismic data in the study area.

[0045] S3. Based on the field outcrops, cores, thin sections, well logging data in the study area, clarify the sequence boundary characteristics of the target interval, carry out single-well sequence division, and establish a sequence stratigraphic framework.

[0046] S4. Calculate and synthesize seismic records using acoustic waves and density curves, perform fine well-seismic calibration, establish a regional seismic framework profile, and perform fine three-dimensional seismic interpretation of the entire region.

[0047] S5. Based on actual drilling data, field outcrop data and core thin section data, conduct research on the characteristics of hill and shoal bodies in the target stratigraphic section of the study area, analyze and statistically analyze the scale, petrophysical parameters and longitudinal and transverse distribution patterns of hill and shoal bodies in the study area, and establish a geological prediction model for hill and shoal bodies.

[0048] S6. Based on the above theoretical model, perform earthquake forward modeling, wherein the forward modeling excitation parameters are set with reference to the parameters of the actual earthquake data volume;

[0049] S7. Analyze the seismic forward modeling results, summarize the seismic response characteristics of hilly areas, and at the same time, perform seismic frequency division processing on the forward modeling seismic data traces to explore the seismic reflection characteristics of hilly areas of different sizes under different frequency bands, establish the seismic response model of hilly areas in the study area, and match the optimal identification frequency corresponding to hilly areas of different sizes.

[0050] S8. Perform seismic frequency division processing on the seismic data volume within the effective bandwidth of the original seismic data volume to obtain multiple single-band data volumes based on the optimal response frequency obtained in step S7.

[0051] S9. Attribute extraction and algorithm optimization are performed for each frequency divider at different frequencies.

[0052] S10. Analyze the selected frequency division attributes and sequence formation thickness in step S9, match and statistically correlate the actual drilling encounter of hills and shoals and the data related to the development of hills and shoals with the well point attribute values, and establish the conversion relationship between frequency division attributes and hills and shoals.

[0053] S11. Remove outliers in the frequency division attributes caused by data issues, and fuse the frequency division attributes to characterize the planar distribution of hills and beaches in the study area.

[0054] Furthermore, in S4, the fine calibration process of well seismic data requires the use of seismic wavelets extracted from the wellside channel for calibration, in order to improve the accuracy of calibration correlation and time-depth relationship.

[0055] Furthermore, in S4, before refining the seismic horizon, the seismic reflection characteristics of each sequence boundary should be determined first. Based on the seismic reflection characteristics of the target layer, on the well-connected seismic calibration profile, the horizon interpretation scheme is first determined, and then the interpretation is densified sequentially according to the seismic grid density of 128→64→32→16→8→4→2, and finally the fine tracking of the top and bottom boundaries of the target layer is completed. The specific horizon interpretation principles are as follows: (1) The well-controlled area mainly adopts the well-seismic combined interpretation principle; (2) The seismic phase axis strong reflection area mainly adopts the seismic layer reflection continuity automatic tracking interpretation principle; (3) The seismic phase axis weak reflection area mainly adopts the upper and lower strata attitude relationship interpretation principle, and automatic and manual interpretation are alternated to ensure interpretation accuracy; (4) The local seismic reflection disordered area mainly adopts the time thickness coordination interpretation principle.

[0056] Furthermore, in S5, the geological prediction model for hill and shoal bodies needs to be designed based on actual drilling data. The rock physical parameters include the thickness, average velocity, and average density of each sequence stratum, as well as the thickness, velocity, and density of the hill and shoal bodies. At the same time, the stacking relationship and size of the hill and shoal bodies in the model must be consistent with the field outcrops and actual drilling measured parameters.

[0057] Furthermore, in step S6, the forward modeling should ensure that the excitation parameters are as consistent as possible with the parameters of the 3D seismic data volume, mainly including information such as the dominant frequency and effective bandwidth. Simultaneously, the forward modeling needs to use a bandwidth wavelet, and the wavelet's dominant frequency and bandwidth must be consistent with the seismic data volume parameters. The forward modeling results (simulated profiles) need to be compared with actual drilling seismic profiles to analyze the effectiveness of the forward modeling and optimize and adjust the model parameters.

[0058] Furthermore, in S7 and S8, the seismic frequency division method is recommended to use a frequency division technique based on the matching pursuit algorithm. The matching pursuit algorithm essentially decomposes any signal into a linear expansion of a set of basis functions, and is a means of processing highly non-stationary signals. The wavelet signal that best approximates the residual signal is iteratively searched in the constructed ultra-complete wavelet library D, and the original seismic signal is gradually decomposed into the optimal linear superposition of multiple matching wavelets. Using the selected frequency f0 as the center frequency, the seismic data is reconstructed using the time-frequency spectrum with a frequency band range of [(f0-Δf), (f0+Δf)] to obtain the frequency-divided seismic data volume.

[0059] Furthermore, in S9, during the frequency-based attribute extraction and algorithm optimization, it is recommended that low-frequency volumes be used for characterizing the facies boundaries of large-scale hill-shoal bodies, while high-frequency volumes are used for sculpting small-scale hill-shoal bodies and their internal details. For characterizing the facies boundaries of large-scale hill-shoal bodies, the root-mean-square amplitude attribute can be used; for sculpting small-scale hill-shoal bodies and their internal details, waveform clustering attributes can be used. In carbonate strata, due to the uplift of shoal facies, the strata exhibit alternating uplifts and depressions, resulting in relatively large variations in stratum thickness. In the sedimentary environment of continental sea carbonate platforms, under the isochronous framework, the deposition rate of hill-shoal bodies is higher than that of other microfacies zones within the platform. The uplift amplifies the differences in stratum thickness; therefore, stratum thickness can also be used as a means of sculpting the details of hill-shoal bodies. Combining this with waveform clustering attributes results in more accurate sculpting of hill-shoal body details. The time windows opened for all the above attributes should ideally be at the top and bottom of the hill-shoal body within the target layer.

[0060] Furthermore, in S10, determining the sedimentary facies type (shoal body) by seismic attribute analysis is an underdetermined problem. The actual analysis results are highly multifaceted, and constraints need to be added in practical applications. The inventors suggest using actual drilling data as constraints: for core wells that encounter shoal bodies, the corresponding attribute type (seismic facies) is determined by using the geological research results and the corresponding logging facies, establishing the correspondence between the two and extrapolating to the un-well area, converting the seismic facies into sedimentary facies, thereby achieving the purpose of finely characterizing the planar distribution characteristics of shoal bodies in the area.

[0061] Furthermore, in S11, outlier values ​​caused by data issues are removed. For areas where the 3D data is not fully covered or where there are fractures or breaks, the data quality is poor, and the extracted attribute values ​​often contain outliers. It is necessary to remove outliers to form an image based on the actual situation, and finally complete the fine carving of the hill body.

[0062] Example 2

[0063] This embodiment is a study on the distribution of the second segment of the Dengying Formation of the Sinian System in the Penglai area of ​​central Sichuan Basin.

[0064] like Figure 1 As shown, a method for characterizing ultra-deep carbonate hills and shoals based on frequency division properties includes the following steps:

[0065] S1. Basic Data Collection: Collect relevant data within the study area, including: regional geological survey data, field outcrop data, core sampling data, conventional well logging curves, well inclination data, stratigraphic layering data, three-dimensional post-stack seismic data, etc.

[0066] S2. Establish the seismic interpretation work area, load 3D seismic data, wellhead data, well deviation data, layered data, well logging curves, and well logging interpretation results. Simultaneously, perform spectral analysis on the 3D seismic data of the study area to obtain the dominant frequency and effective bandwidth of the seismic data in the study area. Figure 2 The results are from the seismic data spectral analysis. The dominant frequency of the seismic data is 30 Hz, and the effective frequency band is 5-60 Hz.

[0067] The seismic data in this study area are Gaomo 3D post-stack seismic data, and the wells include Pengtan 1, Pengtan 101, Pengtan 102 and Pengtan 103.

[0068] S3. Based on the field outcrops, cores, thin sections, well logging data in the study area, clarify the sequence boundary characteristics of the target interval, carry out single-well sequence division, and establish a sequence stratigraphic framework.

[0069] Specifically, based on well logging data from actual drilling in the area (including logging curves and imaging logging charts), combined with observations of core and thin section data, and field outcrop data, the Deng 2 Member in the study area is divided into two third-order sequence stratigraphy (i.e., the upper and lower Deng 2 sub-members). The boundary between SQ1 and SQ2 is a typical type II sequence boundary. Below the boundary are microbial clotted dolomite, cohesive grain dolomite, etc., while above the boundary, it often appears as micritic or argillaceous dolomite formed by rapid marine transgression. On conventional logging curves, GR transitions from box-shaped low values ​​to micro-toothed low values. Because the top of the high SQ1 region is often accompanied by reservoir development, resistivity often transitions from low to high values, and sonic transit time transitions from high to low values. Seismically, this boundary is often marked at the trough ( Figure 3 (Sequence stratigraphic framework and seismic framework profile of the study area).

[0070] S4. Calculate and synthesize seismic records using acoustic waves and density curves, perform fine well-seismic calibration, establish a regional seismic framework profile, and perform fine three-dimensional seismic interpretation of the entire region.

[0071] Specifically, fine well-seismic calibration within the area was carried out using the well-seismic calibration module of LandMark software. During the synthesis of the seismic record, acoustic and density curves were selected. Simultaneously, seismic wavelets were extracted from the well points with a radius of 10 seismic traces. The time window length for wavelet extraction was 1500 milliseconds to ensure that the time window included the target layer, and the correlation of the synthesized record was higher than 0.8. Figure 4(The synthetic record calibration results of Pengtan 1 well); Finally, the correspondence between sequence boundaries and seismic reflection phase axes was determined by well-seismic calibration. The top boundary of the second segment was calibrated at the wave crest, and the bottom boundary of the second segment and the bottom boundary of the upper sub-segment of the second segment were both calibrated at the wave trough. They are stable and can be continuously tracked within the study area, thus enabling further detailed interpretation of the three seismic horizons.

[0072] S5. Based on actual drilling data, field outcrop data, and core thin section data, conduct a study on the characteristics of the target formation, Deng 2 Member Qiutan Body, in the study area. Figure 5 To study the longitudinal and lateral distribution characteristics of hills and shoals in the study area, this study analyzes and statistically analyzes the scale, petrophysical parameters, and longitudinal and lateral distribution patterns of the hills and shoals in the study area, and establishes a geological prediction model for the hills and shoals. Figure 6 );

[0073] S6. Based on the above theoretical model ( Figure 6 Seismic forward modeling was performed, with the excitation parameters set according to the actual seismic data volume parameters. Specifically, the wave equation method was used for forward modeling, with a bandpass wavelet selected as the wavelet, a dominant frequency of 30 Hz, and a bandwidth of 5-60 Hz.

[0074] S7. Analyze the seismic forward modeling results, summarize the seismic response characteristics of hilly areas, and at the same time, perform seismic frequency division processing on the forward modeling seismic data traces to explore the seismic reflection characteristics of hilly areas of different sizes under different frequency bands, establish the seismic response model of hilly areas in the study area, and select the optimal identification frequency corresponding to hilly areas of different sizes.

[0075] Based on the dominant frequency and bandwidth range of the bandpass wavelet, the forward modeling results are divided into five narrowband amplitude bodies (10 Hz, 20 Hz, 30 Hz, 40 Hz, and 50 Hz) with a step size of 10 Hz. Combining geological knowledge and hill and shoal characteristics, the optimal identification frequencies corresponding to hill and shoal bodies of different sizes are selected as 20 Hz and 40 Hz, respectively. Figure 7 This is the result of the forward modeling and its narrowband amplitude profile.

[0076] S8. Perform seismic frequency division processing on the seismic data volume within the effective bandwidth of the original seismic data volume to obtain multiple single-band data volumes based on the optimal response frequency obtained in step S7 (specifically decomposed into 20 Hz and 40 Hz narrowband amplitude volumes).

[0077] S9. Attribute extraction and algorithm optimization are performed for each frequency divider at different frequencies.

[0078] Specifically, for the 20Hz frequency-division seismic data volume, the root mean square amplitude attribute of the target segment is calculated. Figure 8 For a 40Hz frequency-division seismic data volume, calculate the waveform clustering attributes of the target segment. Figure 9 Furthermore, in carbonate strata, the formation of shoal facies leads to alternating uplifts and depressions, resulting in relatively large variations in stratum thickness. In the sedimentary environment of continental sea carbonate platforms, under the isochronous framework, the deposition rate of shoal bodies is higher than that of other microfacies zones within the platform. The formation of shoals amplifies the differences in stratum thickness. Therefore, stratum thickness can also be used as a means of sculpting the details of shoal bodies; combining it with waveform clustering attributes to sculpt the details of shoal bodies is more accurate. Figure 10 This is a thickness map of the upper sub-section of the Deng 2 formation.

[0079] S10. Analyze the selected frequency division attributes and sequence formation thickness in step S9, match and statistically correlate the actual drilling encounter of hills and shoals and the data related to the development of hills and shoals with the well point attribute values, and establish the conversion relationship between frequency division attributes and hills and shoals.

[0080] S11. Remove outliers in the frequency division attributes caused by data issues, and fuse the frequency division attributes to characterize the planar distribution of hills and shoals in the study area. Figure 11 This is a planar distribution map of the hills and beaches in the study area.

[0081] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

[0082] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. A method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes, characterized in that, The method includes: S1: By loading drilling and logging data into the seismic interpretation area, combined with core and thin section data and field outcrop data into the seismic interpretation area, single-well sequence stratigraphy is performed to establish a sequence stratigraphic framework. S2. After calculating and synthesizing seismic records from drilling logging data using acoustic waves and density curves, fine well-seismic calibration is performed using stratigraphic data to obtain time-depth relationships. After establishing a regional seismic framework profile using time-depth relationships, the three-dimensional seismic horizons of the entire region are precisely tracked and interpreted. S3. Based on drilling and logging data, field outcrop data, core data, and the established sequence stratigraphic framework, determine the scale, petrophysical parameters, and longitudinal and lateral distribution patterns of the hill and shoal bodies in the study area, and establish a geological prediction model for the hill and shoal bodies. S4. Based on the established hill and beach geological prediction model, conduct earthquake forward modeling and obtain earthquake forward modeling results; S5. Perform earthquake frequency division processing on the earthquake forward modeling results to determine the earthquake reflection characteristics of hills and shoals of different sizes in different frequency bands. Based on the earthquake reflection characteristics, establish the earthquake response model of the hills and shoals in the study area. Based on the earthquake response model, match the optimal identification frequency corresponding to the hills and shoals of different sizes. S6. Obtain the dominant frequency and effective bandwidth of the seismic data in the study area, perform seismic frequency division processing on the three-dimensional post-stack seismic data, and obtain multiple single-band data volumes based on the optimal identification frequency obtained in step S5. S7. Based on the three-dimensional seismic horizon of the whole region, attribute extraction and algorithm optimization are performed on each single-frequency band data volume at different frequencies to obtain the optimized frequency-division attributes and sequence stratigraphic thickness. S8. After matching and statistically correlating the actual drilling encounters of hill and shoal bodies and the data related to the development of hill and shoal bodies with the selected frequency division attributes and sequence formation thickness, establish the conversion relationship between frequency division attributes and hill and shoal bodies. S9. Remove outliers from the frequency division attributes. Based on the conversion relationship between the frequency division attributes and the hills and beaches established in step S8, fuse the frequency division attributes to characterize the planar distribution of the hills and beaches in the study area.

2. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, Prior to step S1, the method further includes: Acquire basic data within the study area; the basic data includes: field outcrop data, core sampling data, drilling logging data, stratigraphic data, and three-dimensional post-stack seismic data. The basic data within the study area were loaded into the seismic interpretation work area, and spectral analysis was performed on the three-dimensional post-stack seismic data loaded into the seismic interpretation work area to obtain the dominant frequency and effective bandwidth of the seismic data in the study area.

3. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, In step S4, during forward modeling, ensure that the excitation parameters are consistent with the parameters of the 3D seismic data volume, including the dominant frequency and effective bandwidth information. At the same time, the forward modeling needs to use a bandpass wavelet, and the dominant frequency and bandwidth of the wavelet must be consistent with the parameters of the seismic data volume. The forward modeling results, i.e., the simulated profile, need to be compared with the actual drilling seismic profile to analyze the effect of the forward modeling and optimize and adjust the model parameters.

4. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, In steps S5 and S6, the seismic frequency division method employs a frequency division technique based on the matching pursuit algorithm. The matching pursuit algorithm decomposes any signal into a linear expansion of a set of basis functions, serving as a technique for processing highly non-stationary signals. Iteratively, the wavelet signal that best approximates the residual signal is searched within the constructed overcomplete wavelet library D, and the original seismic signal is progressively decomposed into an optimal linear superposition of multiple matching wavelets at a selected frequency. As the center frequency, the frequency band range is The seismic data is reconstructed from the time spectrum to obtain the frequency-divided seismic data volume.

5. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, In step S7, low-frequency volumes are used to characterize the phase band boundaries of large-scale hills and beaches, while high-frequency volumes are used to sculpt small-scale hills and beaches and their internal details. The root mean square amplitude attribute is used for characterizing the phase band boundaries of large-scale hills and beaches, while waveform clustering attributes are used for sculpting small-scale hills and beaches and their internal details.

6. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, In S8, determining the sedimentary facies type of the hill-shoal body based on seismic attributes is an underdetermined problem. In practical applications, constraints need to be added, using actual drilling data as constraints. For core wells that encounter the hill-shoal body, the corresponding attribute types are determined using geological research results and corresponding logging facies, establishing a correspondence between the two and extrapolating to the un-well area, converting seismic facies into sedimentary facies, thereby achieving the goal of finely characterizing the planar distribution features of the hill-shoal body in the area.

7. The method for characterizing ultra-deep carbonate hills and shoals based on frequency division attributes according to claim 1, characterized in that, In step S9, outlier values ​​caused by data issues are removed. For areas where the 3D data is not fully covered or where there are fractures or breaks, the data quality is poor, and the extracted attribute values ​​often contain outliers. Based on the actual situation, outliers are removed to create the map, and finally, the fine carving of the hill body is completed.

Images