A multiscale fracture and fracture-cave prediction method, medium and device

Abstract

This invention relates to the field of petroleum exploration and development, specifically to a multi-scale fracture and fracture-cavity prediction method, medium, and equipment. The method includes: acquiring raw seismic data of the area to be predicted; performing interpretative enhancement processing on the raw seismic data to construct a multi-attribute discontinuity model for fracture identification with attitude constraints; based on step S3, integrating regional tectonic evolution and fracture formation mechanisms, analyzing the spatial distribution characteristics of fractures in the target area, and evaluating the potential for formation of fault-controlled fracture-cavity dominant reservoirs; further predicting the oil and gas accumulation potential within potential dominant reservoirs based on the evaluation of fault-controlled fracture-cavity; and performing three-dimensional geological visualization of the results. This invention addresses low signal-to-noise ratio seismic data by establishing a combined enhancement processing technology workflow, effectively removing random noise, enhancing seismic reflections at fracture sites, broadening the seismic frequency band, and restoring low-frequency and high-frequency energy, resulting in high signal-to-noise ratio, high resolution, and clearly defined fault points.

Description

Technical Field

[0001] This invention relates to the field of oil exploration and development, specifically to a method, medium, and equipment for predicting multi-scale fractures and fracture-cavities. Background Technology

[0002] Fault-controlled fractured-cavitary reservoirs are an important type of oil and gas reservoir, possessing not only high oil and gas storage capacity but also providing pathways for oil and gas migration, significantly influencing the formation and distribution of oil and gas reservoirs. Therefore, research on fault-controlled fractured-cavitary reservoirs is essential in oil and gas exploration and development. However, the genesis and distribution patterns of fault-controlled fractured-cavitary reservoirs are influenced by various factors, such as geological structure, stratigraphic lithology, formation pressure, and distance from the source oil source.

[0003] Chinese invention patent CN111273342A discloses a method for predicting carbonate rock fracture-cavities. This method involves performing vertical seismic profile (VSP) measurements at a predetermined distance from the initially acquired target fracture-cavity to obtain VSP velocity and lithological information of the drilled section. A pre-established underground velocity model is then updated, and the pre-acquired target well area is processed using migration imaging within a predetermined range to obtain seismic data volumes. Finally, the pre-established three-dimensional spatial model is re-delineated to obtain a new three-dimensional spatial model of the target fracture-cavity, further predicting its spatial location and improving the accuracy of predicting carbonate rock fracture-cavities.

[0004] Fractures and fractures / cavities in carbonate reservoirs provide excellent pathways for oil and gas migration and storage, easily forming fault-controlled fracture-cavity reservoirs, which are of great significance for oil and gas accumulation and development. However, when the near-surface strata of the oil and gas extraction area are weathered layers or desert-covered areas, the presence of low-velocity zones leads to significant energy attenuation of seismic wave propagation. Seismic data corresponding to ultra-deep target layers typically have low signal-to-noise ratios, resulting in unclear imaging of fractures and fractures / cavities. Furthermore, fractures / cavities are less common on main faults, but more developed in internal fault zones. Based on low signal-to-noise ratio seismic data, the identification of internal faults becomes more difficult, leading to challenges in the selection of target reservoirs.

[0005] Predicting fault-controlled fracture-cavitation reservoirs requires comprehensive consideration of multiple factors and necessitates the processing and utilization of low signal-to-noise ratio seismic data, presenting significant challenges. Therefore, a multi-scale method, medium, and equipment for predicting faults and fracture-cavitation reservoirs are urgently needed. Summary of the Invention

[0006] To avoid the aforementioned problems in the prior art, the present invention aims to provide a method, medium, and device for predicting multi-scale fractures and cavities.

[0007] The invention provides the following technical solution: a method for predicting multi-scale fractures and cavities, comprising the following steps:

[0008] S1: Obtain raw seismic data for the area to be predicted;

[0009] S2: Perform interpretive enhancement processing on the raw seismic data to improve data quality and signal-to-noise ratio;

[0010] S3: Construct a multi-scale fracture identification and multi-attribute discontinuity model with attitude constraints for fracture identification;

[0011] S4: Evaluation of fault-controlled fracture-cavity bodies. Based on step S3, which integrates regional tectonic evolution and fracture formation mechanisms, the spatial distribution characteristics of fractures in the target area are analyzed, and the potential for the formation of advantageous reservoirs in fault-controlled fracture-cavity bodies is evaluated.

[0012] S5: Sweet spot prediction of reservoirs, based on the evaluation of fault-controlled fracture caverns, further predicts the oil and gas accumulation potential within potential advantageous reservoirs;

[0013] S6: Based on fracture identification, fault-controlled fracture-cavity evaluation, and reservoir sweet spot prediction, a three-dimensional geological visualization is used to comprehensively evaluate the fault-controlled fracture-cavity fracture system, dominant reservoirs, and oil and gas potential.

[0014] The present invention is further configured such that the interpretability enhancement process includes mean filtering, frequency extension processing, and anisotropic diffusion filtering.

[0015] The present invention is further configured such that the mean filtering can effectively remove random noise and avoid its interference with fracture reflection; the frequency extension processing is then used to broaden the seismic frequency band, restore low-frequency and high-frequency energy, and improve seismic resolution; the anisotropic diffusion filtering determines the orientation of the seismic phase axis through the seismic tensor field, removes noise along the orientation direction of the phase axis, enhances seismic reflection at the fracture, and makes the fracture point clear.

[0016] The present invention is further configured such that step S3 specifically involves: accurately capturing the torsional deformation of the seismic axis in the same direction through high-precision discrete scanning to obtain accurate seismic dip and azimuth angles, providing attitude constraints for fracture detection; analyzing the multi-scale response characteristics of seismic data based on a multi-resolution time-frequency transformation algorithm, and extracting seismic responses at different depths and main scales through adaptive frequency selection technology; further constructing an attitude-constrained multi-scale fracture identification multi-attribute discontinuity model, automatically extracting fault planes through the multi-scale fracture identification multi-attribute discontinuity model, eliminating the artifacts of stratigraphic fractures caused by continuously dipping strata, and improving the resolution and signal-to-noise ratio of fracture detection.

[0017] The present invention is further configured such that step S4 specifically includes the following steps:

[0018] S41: Fault mode established; Based on the regional tectonic evolution law and fault formation mechanism, the target stratigraphic fault mode is determined, including normal faults, reverse faults and strike-slip faults.

[0019] S42: Characterization of main fractures; Based on the structural interpretation of stratigraphic horizons and fracture identification results, the detection of main fractures is analyzed. Following the fracture pattern, the trend surface method is used to establish the trend surface distribution map of each stratigraphic layer to characterize the spatial distribution of main fractures.

[0020] S43: Secondary fracture delineation; based on the secondary fracture detection results, multiple attributes are mutually verified to predict the distribution characteristics of secondary fractures and internal fracture zones associated with the main fracture.

[0021] S44: Establishment of the fracture system; Based on the spatial distribution of the main fractures and secondary fractures, the fracture structure and combination relationship are comprehensively characterized to establish the fracture system;

[0022] S45: Reservoir potential assessment; Based on oil and gas geology theory and according to the formation conditions of oil and gas reservoirs, assess the development location and scale of potential fault-controlled fracture-cavity dominant reservoirs.

[0023] The formation conditions of oil and gas reservoirs include "generating, reservoir, capping, trap, migration, and preservation"; specifically, "generating" refers to oil and gas-bearing strata with the conditions for oil generation; "reservoir" refers to rock strata that can store oil and natural gas and export oil and gas; "capping" refers to rock strata that cover the reservoir and prevent oil and gas from escaping upwards; "trap" refers to geological structures that can prevent the migration of oil and gas and enrich it; "migration" refers to the process of oil and gas migrating from the generation site to the reservoir; and "preservation" refers to the geological environment in which oil and gas are preserved for a long time in the trap.

[0024] The present invention is further configured such that step S5 specifically involves: analyzing the oil and gas reservoir state of the target reservoir based on the drilling conditions of a single well and the interpretation results of logging curves, collecting sweet spot indicator curves as sweet spot indicator targets; calculating multiple seismic attributes that conform to the geological model based on experience and extracting well-side paths; analyzing the contribution of different seismic attributes to the sweet spot indicator curve based on fuzzy mathematics theory, sorting them in descending order of contribution, and selecting a preset number of seismic attributes with high contribution according to the sorting; establishing a multi-attribute fusion sweet spot prediction model, thereby realizing sweet spot prediction for fault-controlled fracture-cavity reservoirs.

[0025] The present invention is further configured to establish a multivariate nonlinear relationship between various seismic attributes and sweet spot indicator targets by applying machine learning algorithms based on the constraints of the fracture system; or to establish a multi-attribute fusion sweet spot prediction model by conducting phase control inversion based on the preferred seismic attributes and sweet spot indicator curves.

[0026] The present invention is further configured such that step S6 specifically involves using three-dimensional geological modeling technology to depict the spatial distribution of the main trunk and internal fractures and cavities based on stratigraphic geological modeling, simulating the location and distribution of potential reservoirs, rendering the potential level of advantageous oil and gas sweet spots, and assisting geologists in delineating favorable reservoir development areas and deploying well locations.

[0027] This invention proposes a multi-scale fault-controlled fracture-cavitation reservoir prediction method, which integrates interpretive enhancement processing, fracture identification, fault-controlled fracture-cavitation evaluation, reservoir sweet spot prediction, and three-dimensional geological visualization technology. Based on improving the quality of seismic data, it identifies and establishes a multi-scale fracture system, finely describes the reservoir scale, predicts the sweet spot potential of fault-controlled fracture-cavitation reservoirs, and provides three-dimensional spatial characterization and visualization, providing strong support for the exploration of fault-controlled fracture-cavitation oil and gas reservoirs.

[0028] The present invention also includes an electronic device, the electronic device comprising:

[0029] Memory, which stores executable instructions;

[0030] A processor that executes the executable instructions in the memory to implement the above-described method for predicting multi-scale fractures and cavities.

[0031] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for predicting multi-scale fractures and cavities.

[0032] In summary, the beneficial effects of the above-mentioned technical solution of the present invention are as follows:

[0033] 1. This invention targets low signal-to-noise ratio seismic data and serves multi-scale fault prediction. It establishes a combined process of enhancement processing techniques to effectively remove random noise, enhance seismic reflections at fault locations, broaden the seismic frequency band, and recover low- and high-frequency energy, resulting in high signal-to-noise ratio, high resolution, and clear fault points.

[0034] 2. Through multi-scale fracture detection, a multi-attribute discontinuity model for fracture identification constrained by attitude was established, improving the accuracy of fracture prediction and effectively identifying fractures of different scales, such as controlling main fractures, secondary fractures, and internal fracture zones, characterizing their spatial distribution and development scale. Integrating hydrocarbon geology theory and seismic data processing, a fault-controlled fracture-cavity system was established, characterizing fracture structures and their combinations, and assessing the development location and scale of potential fault-controlled fracture-cavity reservoirs. A multi-attribute fusion sweet spot prediction model constrained by the fracture system was established, combined with fine calibration of drilling reservoirs, achieving high-precision sweet spot prediction for fault-controlled fracture-cavity reservoirs and assessing the intra-reservoir hydrocarbon accumulation potential of potential advantageous reservoirs.

[0035] 3. By utilizing three-dimensional geological modeling technology, the three-dimensional spatial characterization and visualization of the fault-controlled fracture-cavity fracture system, dominant reservoirs, and sweet spot distribution were realized, providing intuitive guidance for the delineation of favorable reservoir development areas and well location design. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of a multi-scale fracture and cavity prediction method.

[0038] Figure 2 This is the original pure wave seismic profile.

[0039] Figure 3 This is a seismic profile after interpretive enhancement.

[0040] Figure 4 This is the result of fracture identification.

[0041] Figure 5 This is a schematic diagram showing the spatial distribution characteristics of fractures with different tectonic origins.

[0042] Figure 6 Predict cross-sectional views for desserts.

[0043] Figure 7 Predict the planar distribution of desserts.

[0044] Figure 8 Comparison chart of single wells for predicting sweet spots in fractured cavern reservoirs.

[0045] Figure 9 This is a three-dimensional geological visualization diagram of a fault-controlled fractured cavern reservoir. Detailed Implementation

[0046] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments of the present invention, other similar embodiments obtained by those skilled in the art without creative effort should all fall within the scope of protection of the present invention.

[0047] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0048] Example 1:

[0049] This embodiment takes the fault-controlled reservoir of the Yingshan Formation of the Ordovician as an example. The reservoir prediction faces two main challenges: (1) The acquisition area is located in the desert-covered area on the surface. The signal-to-noise ratio of the seismic data of the ultra-deep target layer is low, and the imaging of the string of beads and faults is not clear enough; (2) There are few strings of beads on the main faults and many strings of beads in the inner zone. The identification of the inner faults is not implemented, and it is difficult to select the target of the favorable reservoir area.

[0050] like Figure 1 As shown in the preferred embodiment of the present invention, a method for predicting multi-scale fractures and cavities includes the following steps:

[0051] S1: Obtain the raw seismic data for the area to be predicted; the raw seismic data includes raw pure wave seismic profiles, such as... Figure 2 As shown.

[0052] S2: Perform interpretive enhancement processing on the raw seismic data to improve data quality and signal-to-noise ratio;

[0053] The interpretative enhancement process includes mean filtering, frequency extension processing, and anisotropic diffusion filtering. Mean filtering effectively removes random noise, preventing it from interfering with fracture reflections. Frequency extension processing broadens the seismic frequency band, recovering low- and high-frequency energy and improving seismic resolution. Anisotropic diffusion filtering determines the azimuth of the seismic phase axis using the seismic tensor field, eliminating noise along the phase axis's orientation, enhancing seismic reflections at the fracture site, and making the fracture points clearer. Figure 2 As indicated by the arrow in Figure 3.

[0054] S3: Construct a multi-scale fracture identification and multi-attribute discontinuity model with attitude constraints for fracture identification;

[0055] High-precision discrete scanning accurately captures the torsional deformation of seismic axes in the same direction, obtaining precise seismic dip and azimuth angles, providing attitude constraints for fracture detection. Based on a multi-resolution time-frequency transformation algorithm, the multi-scale response characteristics of seismic data are analyzed, and adaptive frequency selection technology is used to extract seismic responses at different depths and major scales. Furthermore, an attitude-constrained multi-scale fracture identification multi-attribute discontinuity model is constructed. This model automatically extracts fault planes, eliminating stratigraphic fracture artifacts caused by continuously dipping strata, and improving the resolution and signal-to-noise ratio of fracture detection. Figure 4 As shown, Figure 4 Figure a in the image shows a planar slice of the fracture detection results in a two-dimensional line-based display mode. Figure 4 In the image, 'b' represents a planar slice of the fracture detection results in the 3D relief display mode. The main fracture structure and combination relationship are clear in the detection results, and the deep internal fractures and secondary fractures have a certain response, which is consistent with the fracture formation mechanism of the fault-controlled fracture-cavity-reservoir collective.

[0056] S4: Evaluation of fracture-controlled fracture cavities, assessing the development of fracture-controlled fracture cavities based on fracture identification results; specifically including the following steps:

[0057] S41: Fault mode established; Based on the regional tectonic evolution law and fault formation mechanism, the target stratigraphic fault mode is determined, including normal faults, reverse faults and strike-slip faults.

[0058] S42: Characterization of main fractures; Based on the structural interpretation of stratigraphic horizons and fracture identification results, the detection of main fractures is analyzed. Following the fracture pattern, the trend surface method is used to establish the trend surface distribution map of each stratigraphic layer to characterize the spatial distribution of main fractures.

[0059] S43: Secondary fracture delineation; based on the secondary fracture detection results, multiple attributes are mutually verified to predict the distribution characteristics of secondary fractures and internal fracture zones associated with the main fracture.

[0060] S44: Fracture System Establishment; Based on the spatial distribution of the main and secondary fractures, the fracture structure and combination relationships are comprehensively characterized to establish the fracture system; results are as follows. Figure 5 As shown, Figure 5 Figure A in the diagram shows the spatial distribution characteristics of fractures of different tectonic origins at multiple scales. Figure 5 Figure ① in the text is Figure 5 The cross-sectional view at point ① in Figure A;

[0061] Figure 5 Figure ② in the middle is Figure 5 The cross-sectional view at point ② in Figure A; Figure 5 Figure ③ in the middle is Figure 6-7 The cross-sectional view at point ③ in Figure A.

[0062] It is evident that the main faults are well-developed and have good inheritance, serving as the main oil source channels. Overall, the faults are stronger in the south and weaker in the north, and the profile features are clear.

[0063] S45: Reservoir Potential Assessment; Based on oil and gas geology theory, assess the location and scale of potential fault-controlled fracture-cavity dominant reservoirs according to the formation conditions of oil and gas reservoirs, namely, "generation, storage, cap, enclosure, migration, and protection".

[0064] Specifically, "generating" refers to oil- and gas-bearing strata with the conditions for oil generation; "reservoir" refers to rock strata that can store oil and natural gas and export oil and gas; "cap" refers to rock strata that cover the reservoir and prevent oil and gas from escaping upwards; "enclosure" refers to geological structures that can prevent the migration of oil and gas and enrich it; "transport" refers to the process of oil and gas migrating from the generating site to the reservoir; and "preservation" refers to the geological environment in which oil and gas are preserved for a long time within the trap.

[0065] S5: Sweet spot prediction of reservoirs, based on the evaluation of fault-controlled fracture caverns, further predicts the oil and gas accumulation potential within potential advantageous reservoirs;

[0066] Based on single-well drilling data and logging curve interpretation results, the oil and gas reservoir status of the target reservoir is analyzed, and sweet spot indicator curves are collected as sweet spot indicators. Multiple seismic attributes conforming to the geological model are calculated empirically and extracted from the wellbore. The contribution of different seismic attributes to the sweet spot indicator curve is analyzed using fuzzy mathematics theory, and the attributes are sorted in descending order of contribution. A specific number of high-contribution seismic attributes are selected. Based on the constraints of the fracture system, machine learning algorithms are applied to establish multivariate nonlinear relationships between various seismic attributes and the sweet spot indicator target; or, based on the optimized seismic attributes and the sweet spot indicator curve, phase-controlled inversion is performed to establish a multi-attribute fusion sweet spot prediction model. This ultimately enables sweet spot prediction for fault-controlled fracture-cavity reservoirs.

[0067] The results are as follows Figure 6 As shown, Figure 6 The upper and middle sections of the attached diagram show the original seismic well profile of the reservoir. Figure 8 The lower half of the attached figure shows the corresponding sweet spot prediction profile. It can be seen that the prediction results have eliminated the strong surface reflection characteristics, highlighted the internal structure of the reservoir, and clearly showed the spatial structure characteristics of the fault-controlled fracture cavity, which is consistent with the geological mechanism of strike-slip fault-controlled reservoir.

[0068] like Figure 8 As shown, Figure 8 Figure A in the diagram shows the original seismic profile of a single well in the reservoir. Figure 9 Figure B in the diagram shows the corresponding sweet spot prediction profile. Combining drilling findings and well logging interpretation, the location of the lost circulation encountered in the single well is at a low impedance value. The location of oil and gas development in the well logging interpretation matches the sweet spot prediction results well, with a single well matching rate of over 85%. The overall internal structure is clear.

[0069] S6: Based on fracture identification, fault-controlled fracture-cavity evaluation, and reservoir sweet spot prediction, a three-dimensional geological visualization is used to comprehensively evaluate the fault-controlled fracture-cavity fracture system, dominant reservoirs, and oil and gas potential.

[0070] Using 3D geological modeling technology, based on stratigraphic geological modeling, the spatial distribution of main and internal faults and fracture-cavities is depicted, the location and distribution of potential reservoirs are simulated, and the potential level of advantageous oil and gas sweet spots is rendered. This assists geologists in delineating favorable reservoir development areas and deploying well locations. The results are as follows: ​ As shown.

[0071] Example 2:

[0072] The present invention also includes an electronic device, the electronic device comprising:

[0073] Memory, which stores executable instructions;

[0074] A processor that executes the executable instructions in the memory to implement the multi-scale fracture and cavity prediction method described in Embodiment 1.

[0075] Example 3:

[0076] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the multi-scale fracture and cavity prediction method described in Example 1.

[0077] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for predicting multi-scale fractures and fissures, characterized in that, Includes the following steps: S1: Obtain raw seismic data for the area to be predicted; S2: Interpretive enhancement processing of raw seismic data; S3: Construct a multi-scale fracture identification and multi-attribute discontinuity model with attitude constraints for fracture identification; S4: Evaluation of fault-controlled fracture-cavity bodies. Based on step S3, which integrates regional tectonic evolution and fracture formation mechanisms, the spatial distribution characteristics of fractures in the target area are analyzed, and the potential for the formation of advantageous reservoirs in fault-controlled fracture-cavity bodies is evaluated. S5: Sweet spot prediction of reservoirs, based on the evaluation of fault-controlled fracture caverns, further predicts the oil and gas accumulation potential within potential advantageous reservoirs; S6: Based on fracture identification, fault-controlled fracture-cavity evaluation, and reservoir sweet spot prediction, a three-dimensional geological visualization is used to comprehensively evaluate the fault-controlled fracture-cavity fracture system, dominant reservoirs, and oil and gas potential.

2. The method for predicting multi-scale fractures and cavities according to claim 1, characterized in that, The interpretive enhancement process includes mean filtering, frequency extension processing, and anisotropic diffusion filtering.

3. The method for predicting multi-scale fractures and cavities according to claim 2, characterized in that, Random noise is removed by mean filtering; then the frequency band is widened by the frequency extension processing to restore low-frequency and high-frequency energy; the anisotropic diffusion filtering determines the orientation of the seismic phase axis through the seismic tensor field and removes noise along the orientation direction of the phase axis.

4. The method for predicting multi-scale fractures and cavities according to claim 3, characterized in that, Step S3 specifically involves capturing the torsional deformation of the seismic axis in the same direction through discrete scanning to obtain the seismic dip and azimuth angles, providing attitude constraints for fault detection; analyzing the multi-scale response characteristics of seismic data based on a multi-resolution time-frequency transformation algorithm, and extracting the seismic responses at different depths and main scales through adaptive frequency selection technology; further constructing an attitude-constrained multi-scale fault identification multi-attribute discontinuity model, and automatically extracting fault planes through the multi-scale fault identification multi-attribute discontinuity model.

5. The method for predicting multi-scale fractures and cavities according to claim 4, characterized in that, Step S4 specifically includes the following steps: S41: Fault mode established; Based on the regional tectonic evolution law and fault formation mechanism, the target stratigraphic fault mode is determined, including normal faults, reverse faults and strike-slip faults. S42: Characterization of main fractures; Based on the structural interpretation of stratigraphic horizons and fracture identification results, the detection of main fractures is analyzed. Following the fracture pattern, the trend surface method is used to establish the trend surface distribution map of each stratigraphic layer to characterize the spatial distribution of main fractures. S43: Secondary fracture delineation; based on the secondary fracture detection results, multiple attributes are mutually verified to predict the distribution characteristics of secondary fractures and internal fracture zones associated with the main fracture. S44: Establishment of the fracture system; Based on the spatial distribution of the main fractures and secondary fractures, the fracture structure and combination relationship are comprehensively characterized to establish the fracture system; S45: Reservoir potential assessment; based on oil and gas geology theory and according to the formation conditions of oil and gas reservoirs, assess the development location and scale of potential fault-controlled fracture-cavity dominant reservoirs.

6. The method for predicting multi-scale fractures and cavities according to claim 5, characterized in that, Step S5 specifically involves analyzing the oil and gas reservoir status of the target reservoir based on the drilling conditions of a single well and the interpretation results of logging curves, collecting sweet spot indicator curves as sweet spot indicator targets, calculating various seismic attributes that conform to the geological model based on experience and extracting well-side passages, analyzing the contribution of different seismic attributes to the sweet spot indicator curves based on fuzzy mathematics theory, sorting them in descending order of contribution, selecting a preset number of seismic attributes with high contribution according to the sorting, and establishing a multi-attribute fusion sweet spot prediction model to achieve sweet spot prediction for fault-controlled fracture-cavity reservoirs.

7. The method for predicting multi-scale fractures and cavities according to claim 6, characterized in that, By establishing a multi-attribute fusion sweet spot prediction model based on the constraints of the fracture system and applying machine learning algorithms to establish multiple nonlinear relationships between various seismic attributes and sweet spot indicator targets; or by conducting phase-controlled inversion based on the optimized seismic attributes and sweet spot indicator curves to establish a multi-attribute fusion sweet spot prediction model.

8. The method for predicting multi-scale fractures and cavities according to claim 1, characterized in that, Step S6 specifically involves using 3D geological modeling technology to depict the spatial distribution of the main trunk and internal fractures and cavities based on stratigraphic geological modeling, simulating the location and distribution of potential reservoirs, rendering the potential level of advantageous oil and gas sweet spots, and assisting geologists in delineating favorable reservoir development areas and deploying well locations.

9. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement a multi-scale fracture and cavity prediction method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements a multi-scale fracture and cavity prediction method as described in any one of claims 1-8.

Images