A method, system, and medium for low permeability reservoir geocandy prediction

Abstract

The present application relates to the technical field of geological data processing, and more particularly to a low-permeability reservoir geological dessert prediction method and system and a medium. The method comprises the following steps: obtaining low-permeability reservoir geological data; preprocessing the low-permeability reservoir geological data to obtain standard low-permeability reservoir geological data; performing multi-dimensional feature recognition on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; extracting key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information; and extracting low-permeability reservoir cores from the key low-permeability reservoir information to generate low-permeability reservoir core data. Through data processing technology, pattern recognition technology and machine learning technology, the present application realizes multi-dimensional analysis of low-permeability reservoir geological features and correlation analysis in combination with the seismic characteristics of low-permeability reservoirs, thereby improving the accuracy of low-permeability reservoir geological dessert prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geological data processing, and particularly relates to a low-permeability reservoir geological sweet spot prediction method, system and medium. BACKGROUND

[0002] The low-permeability reservoir geological sweet spot refers to a region with high exploration and development value in a low-permeability oil and gas reservoir due to good physical properties and high oil and gas content. These regions are usually characterized by high porosity, permeability and gas saturation, and low irreducible water saturation. Therefore, the prediction of the low-permeability reservoir geological sweet spot is of great significance to improving the success rate of oil and gas exploration and development efficiency. The traditional prediction of the low-permeability reservoir sweet spot mainly relies on the experience of geologists, and uses geological exploration technology to measure data of the low-permeability reservoir geological layer, such as rock layer measurement data and low-permeability reservoir downhole data, and performs prediction and analysis of the low-permeability reservoir geological sweet spot according to the data. However, the existing low-permeability reservoir geological sweet spot prediction method is relatively single in analyzing the characteristics of the low-permeability reservoir geological layer, and fails to correlate and analyze the seismic characteristics of the low-permeability reservoir, thereby resulting in low prediction accuracy of the low-permeability reservoir geological sweet spot. SUMMARY

[0003] Therefore, it is necessary to provide a low-permeability reservoir geological sweet spot prediction method, system and medium to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, a low-permeability reservoir geological sweet spot prediction method is provided, and the method comprises the following steps:

[0005] Step S1: obtaining low-permeability reservoir geological data; preprocessing the low-permeability reservoir geological data to obtain standard low-permeability reservoir geological data; performing multi-dimensional feature recognition on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; and extracting key low-permeability reservoir information from the multi-dimensional geological feature data to obtain the key low-permeability reservoir information.

[0006] Step S2: extracting low-permeability reservoir cores from the key low-permeability reservoir information to generate low-permeability reservoir core data; extracting core microstructure features from the multi-dimensional geological feature data according to the low-permeability reservoir core data to obtain the core microstructure features; and identifying low-permeability reservoir structure morphology from the key low-permeability reservoir information according to the core microstructure features to generate low-permeability reservoir morphology data.

[0007] Step S3: Based on the low-permeability reservoir morphology data, the key low-permeability reservoir information is determined in the range of the low-permeability reservoir to obtain low-permeability reservoir range data; the low-permeability reservoir range data is used to identify the seismic characteristics of the key low-permeability reservoir information to generate low-permeability reservoir seismic data; the low-permeability reservoir range data is used to extract shale reservoir characteristics to obtain shale reservoir characteristic data; the shale reservoir characteristic data is associated with the low-permeability reservoir seismic data for analysis to generate low-permeability reservoir seismic correlation data;

[0008] Step S4: According to the low-permeability reservoir seismic correlation data, the low-permeability reservoir geological data is measured to generate low-permeability reservoir geological indicators; the low-permeability reservoir geological correlation indicators are used to construct a low-permeability reservoir geological sweet spot prediction model to obtain the low-permeability reservoir geological sweet spot prediction model; the low-permeability reservoir geological sweet spot prediction model is used for pre-stack inversion to obtain model pre-stack inversion data; the low-permeability reservoir geological sweet spot prediction model is dynamically optimized through the model pre-stack inversion data to obtain a low-permeability reservoir geological sweet spot prediction optimization model; based on the low-permeability reservoir geological sweet spot prediction optimization model, the low-permeability reservoir geological data is used for geological sweet spot prediction to generate low-permeability reservoir geological sweet spot prediction data; the low-permeability reservoir geological sweet spot prediction data is visualized to generate a low-permeability reservoir geological sweet spot prediction report.

[0009] The present application can ensure the standardization of data by obtaining and preprocessing low-permeability reservoir geological data, and provide high-quality data support for subsequent analysis; multi-dimensional feature recognition can extract rich geological information from standard geological data; the extraction of key low-permeability reservoir information further focuses on factors that have an important influence on reservoir characteristics, improving the pertinence and accuracy of analysis. The low-permeability reservoir core information can be determined by extracting low-permeability reservoir cores from key low-permeability reservoir information; the understanding of the internal structure of the reservoir can be determined by extracting core microstructure features from multi-dimensional geological feature data according to low-permeability reservoir core data, which can reveal the micro characteristics of the reservoir; the spatial distribution characteristics of the reservoir can be determined by identifying the low-permeability reservoir structure morphology according to the core microstructure features of the key low-permeability reservoir information, which provides data basis for subsequent reservoir range distribution determination. The geographic distribution of the reservoir is determined by determining the low-permeability reservoir range distribution based on the low-permeability reservoir morphology data, so as to determine the range of the low-permeability reservoir; the seismic characteristics of the reservoir geological structure can be identified by identifying the seismic characteristics of the low-permeability reservoir range data, so as to determine the seismic situation of the low-permeability reservoir; the shale reservoir feature data can be obtained by extracting shale reservoir features from the low-permeability reservoir range data; the correlation between the low-permeability reservoir and the earthquake can be comprehensively determined by correlating and analyzing the shale reservoir parameters and the low-permeability reservoir seismic data. The quantitative reservoir index evaluation standard is provided by measuring and calculating the geological index of the low-permeability reservoir geological data according to the low-permeability reservoir seismic correlation data; the low-permeability reservoir geological sweet spot prediction model is obtained by constructing the low-permeability reservoir geological sweet spot prediction model using the low-permeability reservoir geological correlation index; the model pre-stack inversion data is obtained by pre-stack inversion of the low-permeability reservoir geological sweet spot prediction model; the accuracy and applicability of the prediction model are improved by dynamically optimizing the low-permeability reservoir geological sweet spot prediction model through the model pre-stack inversion data, so as to improve the accuracy of the geological sweet spot prediction; the low-permeability reservoir geological sweet spot prediction data is generated by predicting the geological sweet spot of the low-permeability reservoir geological data based on the low-permeability reservoir geological sweet spot prediction optimization model; the prediction report can be intuitively displayed by visualizing the low-permeability reservoir geological sweet spot prediction data. Therefore, the present application realizes multi-dimensional analysis of the low-permeability reservoir geological characteristics by data processing technology, pattern recognition technology and machine learning technology, and combines the correlation analysis of the seismic characteristics of the low-permeability reservoir, so as to improve the accuracy of the low-permeability reservoir geological sweet spot prediction.

[0010] Preferably, step S2 comprises the following steps:

[0011] Step S21: identifying the low-permeability reservoir core features by identifying the low-permeability reservoir core features of the key low-permeability reservoir information; generating the low-permeability reservoir core data by extracting the low-permeability reservoir core data from the key low-permeability reservoir information according to the low-permeability reservoir core features;

[0012] Step S22: Enhance the core features of the low-permeability reservoir core data to obtain enhanced core feature data; hierarchically process the enhanced core feature data to generate hierarchical core feature data; extract the hierarchical structure elements from the hierarchical core feature data to obtain core hierarchical structure element information.

[0013] Step S23: Mark the core layer structure of the multidimensional geological feature data using the core layer structure element information to obtain core layer structure marking data; identify the core microstructure features of the multidimensional geological feature data based on the core layer structure marking data to obtain core microstructure features;

[0014] Step S24: Based on the microstructural characteristics of the core, perform feature imaging processing on the key low-permeability reservoir information to obtain low-permeability reservoir structure images; perform structural image recognition on the low-permeability reservoir structure images to obtain low-permeability reservoir structure data; extract the low-permeability reservoir morphology from the low-permeability reservoir structure data to obtain low-permeability reservoir morphology data.

[0015] This invention identifies core features of key low-permeability reservoirs, extracting representative core features from these key reservoir information. Based on these core features, it extracts core data from key low-permeability reservoirs, clearly defining the core data. It enhances the core features, improving their identifiability and making their microscopic features more apparent. The enhanced core feature data is then hierarchically structured to clarify its multi-level structure. Hierarchical structural element extraction from this hierarchical data reveals the core's hierarchical structure. Finally, it analyzes the hierarchical structural elements to obtain core information about the core's hierarchical structure. Core hierarchical structure marking is performed on multidimensional geological feature data, combining the microstructure of the core with geological features to enhance data correlation. Based on the core hierarchical structure marking data, core microstructure feature identification is performed on multidimensional geological feature data, further refining the microstructure features of the core. Feature imaging processing is performed on key low-permeability reservoir information based on the core microstructure features, transforming abstract data into intuitive images for easier analysis and understanding. Structural image recognition is performed on low-permeability reservoir structure images to clarify key structural information, providing accurate data support for reservoir morphology extraction. Finally, low-permeability reservoir morphology extraction is performed on low-permeability reservoir structure data, clarifying the spatial morphological characteristics of the reservoir.

[0016] Preferably, step S3 includes the following steps:

[0017] Step S31: Extract geometric features from the low-permeability reservoir morphology data to obtain the low-permeability reservoir geometric morphology features; determine the spatial relationships of the low-permeability reservoir geometric morphology features to generate low-permeability reservoir spatial relationship data; classify the low-permeability reservoir spatial relationship data into relation types to obtain spatial relationship type data.

[0018] Step S32: Based on the spatial relationship type data, identify the spatial distribution of low-permeability reservoir morphology data to generate low-permeability reservoir spatial distribution information; determine the low-permeability reservoir boundary based on the low-permeability reservoir spatial distribution information to obtain low-permeability reservoir boundary data; determine the low-permeability reservoir range distribution based on the low-permeability reservoir spatial distribution information and low-permeability reservoir boundary data to obtain low-permeability reservoir range data.

[0019] Step S33: Perform continuity determination on the low-permeability reservoir range data to generate continuity determination data; use the continuity determination data to determine the interlayer continuity of key low-permeability reservoir information to obtain interlayer continuity data; record the number of interlayer collisions and compressions on key low-permeability reservoir information based on the interlayer continuity data to obtain interlayer collision and compression data; perform seismic feature identification on the interlayer collision and compression data to generate low-permeability reservoir seismic data.

[0020] Step S34: Perform shape and structure identification on the low-permeability reservoir range data to obtain shale reservoir shape data; extract shale reservoir parameters from the shale reservoir shape data to generate shale reservoir parameter information; integrate the shale reservoir shape data and shale reservoir parameter information to obtain shale reservoir characteristic data;

[0021] Step S35: Perform correlation analysis between shale reservoir characteristic data and low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data.

[0022] This invention extracts geometric features from low-permeability reservoir morphology data, quantifying the physical properties of the reservoirs and specifically extracting their geometric morphological features. It determines the spatial relationships of these features, clarifying the spatial interactions between different reservoir structures. It also classifies the spatial relationship data of low-permeability reservoirs into different types, identifying their spatial distribution based on these relationship types. Furthermore, it determines the spatial distribution characteristics of low-permeability reservoir morphology data, providing boundary values ​​for subsequent determination of the reservoir's extent. Based on the spatial distribution and boundary data, it determines the extent of key low-permeability reservoirs, clarifying their location. Finally, it measures the continuity of the low-permeability reservoir extent data, aiding in subsequent... This study continuously predicts the flow and distribution of fluids in reservoirs; uses continuous measurement data to determine the interlayer continuity of key low-permeability reservoir information, revealing the interrelationships between different low-permeability reservoirs; records the number of interlayer collisions and compressions of key low-permeability reservoir information based on interlayer continuity data, clarifying reservoir activity; identifies seismic features from interlayer collision and compression data to obtain seismic data for low-permeability reservoirs; identifies the shape and structure of low-permeability reservoir extent data, clarifying the geometric morphological characteristics of shale reservoirs; extracts shale reservoir parameters from shale reservoir shape data, providing key physical parameters of shale reservoirs; integrates shale reservoir shape data and shale reservoir parameter information to obtain shale reservoir characteristic data; and performs correlation analysis between shale reservoir characteristic data and low-permeability reservoir seismic data, combining geological structure information with seismic data to clarify the correlation between low-permeability reservoirs and seismic features.

[0023] Preferably, step S35 includes the following steps:

[0024] Step S351: Identify porosity features in shale reservoir characteristic data to obtain porosity feature data; determine porosity distribution in porosity feature data to generate porosity distribution data; calculate porosity and pore size in combination with porosity distribution data to obtain porosity and pore size data.

[0025] Step S352: Extract the mineral composition of the shale reservoir characteristic data to obtain the mineral composition data of the shale reservoir; identify the fractures of the shale reservoir based on the porosity and pore size data to generate shale reservoir fracture information;

[0026] Step S353: Simulate the interlayer seismic response of shale reservoir fracture information and low-permeability reservoir seismic data to obtain interlayer seismic response data; determine the interlayer vulnerability of the interlayer seismic response data to generate interlayer vulnerability data; identify the surface texture of the fractured area of ​​the shale reservoir based on the interlayer vulnerability data to obtain regional surface texture information; compare the regional surface texture information with preset regional surface texture information; if the regional surface texture information is less than the preset regional surface texture information, mark the regional surface texture information as high seismic resistance layer information; if the regional surface texture information is greater than or equal to the preset regional surface texture information, mark the regional surface texture information as low seismic resistance layer information; integrate the high seismic resistance layer information and the low seismic resistance layer information to obtain the seismic resistance parameters of the low-permeability reservoir.

[0027] Step S354: Combine inter-layer seismic response data to calculate the seismic intensity of low-permeability reservoir seismic data and generate low-permeability reservoir seismic intensity data; perform inter-layer seismic characteristic correlation between low-permeability reservoir seismic intensity data and low-permeability reservoir seismic resistance parameters to generate low-permeability reservoir seismic correlation data.

[0028] This invention identifies porosity features in shale reservoir characteristic data, clarifying porosity characteristics to aid in subsequent reservoir permeability assessment; determines porosity distribution from porosity feature data, revealing the distribution characteristics of pores in the reservoir, which helps in understanding fluid distribution and flow within the reservoir; calculates porosity and pore size based on porosity distribution data, clarifying the specific size information of reservoir pores; extracts rock mineral composition from shale reservoir characteristic data, enabling further analysis of rock physical properties; identifies rock fractures based on porosity and pore size data from rock mineral composition data, clarifying the fracture situation in the reservoir; simulates inter-layer seismic response by combining shale reservoir fracture information with seismic data from low-permeability reservoirs, providing the reservoir's response characteristics under seismic loading; and determines inter-layer vulnerability from inter-layer seismic response data, assessing the reservoir's vulnerability. Yes; based on interlayer vulnerability data, surface texture identification is performed on fracture information in shale reservoirs to obtain regional surface texture information; the regional surface texture information is compared with preset regional surface texture information. If the regional surface texture information is less than the preset regional surface texture information, it is marked as a high seismic-resistant layer area; if the regional surface texture information is greater than or equal to the preset regional surface texture information, it is marked as a low seismic-resistant layer area; the high-seismic-resistant layer area information and the low-seismic-resistant layer area information are integrated to obtain the seismic resistance parameters of low-permeability reservoirs; the seismic intensity of low-permeability reservoir seismic data is calculated by combining inter-seismic response data, providing the strength characteristics of the reservoir under seismic action; the seismic intensity data of low-permeability reservoirs and the seismic resistance parameters of low-permeability reservoirs are correlated with inter-layer seismic characteristics to clarify the seismic correlation of low-permeability reservoirs.

[0029] Preferably, step S4 includes the following steps:

[0030] Step S41: Calculate geological indicators of low-permeability reservoirs based on seismic correlation data of low-permeability reservoirs to generate geological indicators of low-permeability reservoirs.

[0031] Step S42: Construct a prediction model for the geological sweet spot of low-permeability reservoirs based on the geological correlation index of low-permeability reservoirs to obtain a pre-model for predicting the geological sweet spot of low-permeability reservoirs. Then, train the pre-model for predicting the geological sweet spot of low-permeability reservoirs using the geological correlation index of low-permeability reservoirs to obtain a training model for predicting the geological sweet spot of low-permeability reservoirs.

[0032] Step S43: Perform cross-validation evaluation on the low-permeability reservoir geological sweet spot prediction training model to obtain model evaluation data; adjust the model parameters of the low-permeability reservoir geological sweet spot prediction training model using the model evaluation data to obtain the low-permeability reservoir geological sweet spot prediction model.

[0033] Step S44: Perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the model pre-stack inversion data to obtain the low-permeability reservoir geological sweet spot prediction optimization model.

[0034] Step S45: Based on the low-permeability reservoir geological sweet spot prediction optimization model, perform geological sweet spot prediction on the low-permeability reservoir geological data to generate low-permeability reservoir geological sweet spot prediction data; perform data visualization on the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0035] This invention calculates geological indicators of low-permeability reservoirs based on seismic correlation data, quantifying key parameters for reservoir evaluation. It constructs a low-permeability reservoir geological sweet spot prediction model based on the geological correlation indicators, obtaining a pre-model. This pre-model is then trained using the same indicators to obtain a trained low-permeability reservoir geological sweet spot prediction model, improving prediction accuracy. Cross-validation evaluation of the trained model provides quantitative indicators of model performance, reflecting its reliability and predictive ability. The model evaluation data is used to further evaluate the low-permeability reservoir geological sweet spot prediction model. The training model for predicting geological sweet spots in permeable reservoirs underwent parameter adjustments to further improve its prediction accuracy, ensuring its effectiveness and applicability. Pre-stack inversion was performed on the prediction model for geological sweet spots in low-permeability reservoirs, enhancing its accuracy in identifying reservoir characteristics. Dynamic optimization of the low-permeability reservoir geological sweet spot prediction model using pre-stack inversion data improved its stability and long-term prediction accuracy. Based on the optimized model, geological sweet spots were predicted from low-permeability reservoir geological data, identifying the specific location and characteristics of sweet spot areas. Data visualization of the low-permeability reservoir geological sweet spot prediction data enabled the generation of intuitive reports.

[0036] Preferably, step S41 includes the following steps:

[0037] Step S411: Measure the magnitude of seismic amplitude in the seismic correlation data of the low-permeability reservoir to obtain interlayer seismic amplitude data; identify the degree of interlayer deformation in the low-permeability reservoir morphology data based on the interlayer seismic amplitude data to generate low-permeability reservoir deformation data.

[0038] Step S412: Calculate the pore density of the geological data of the low-permeability reservoir based on the deformation data of the low-permeability reservoir to obtain the pore density of the low-permeability reservoir; identify the porosity of the pore density of the low-permeability reservoir to obtain the pore parameters of the low-permeability reservoir; determine the inter-layer permeability of the low-permeability reservoir based on the pore parameters of the low-permeability reservoir to generate the inter-layer permeability of the low-permeability reservoir.

[0039] Step S413: Combine the seismic parameters of the low-permeability reservoir to perform seismic coefficient permeability mapping on the permeability between low-permeability reservoirs to obtain the permeability data of the seismic-resistant layer; calculate the rock elastic modulus of the geological data of the low-permeability reservoir based on the permeability data of the seismic-resistant layer to generate the rock elastic modulus data.

[0040] Step S414: Extract oil and gas features from key low-permeability reservoir information to obtain low-permeability reservoir oil and gas data; calculate oil and gas saturation based on rock elastic modulus data to generate low-permeability reservoir oil and gas saturation data.

[0041] Step S415: Integrate low-permeability reservoir deformation data, inter-layer permeability data, and low-permeability reservoir oil and gas saturation data to obtain low-permeability reservoir geological indicators.

[0042] This invention measures the magnitude of seismic amplitude in seismic correlation data of low-permeability reservoirs, providing a quantitative index of reservoir seismic response; identifies the degree of inter-layer deformation in low-permeability reservoir morphology data based on inter-layer seismic amplitude data, revealing the deformation characteristics of the reservoir under geological processes; calculates the pore density of low-permeability reservoir geological data based on low-permeability reservoir deformation data, providing a quantitative description of reservoir pore density; identifies porosity in low-permeability reservoirs based on pore density, clarifying pore parameters; measures the inter-layer permeability of low-permeability reservoir geological data based on pore parameters, obtaining information on the inter-layer permeability of low-permeability reservoirs; and combines low-permeability reservoir seismic resistance parameters to improve the permeability of low-permeability reservoirs. Mapping the permeability between reservoir layers to the seismic coefficient permeability provides information on the reservoir's permeability under seismic loading, which helps assess the reservoir's behavior during geological dynamics. Calculating the elastic modulus of strata in low-permeability reservoir geological data based on seismic-resistant layer permeability data clarifies the elastic properties of the reservoir rocks. Extracting hydrocarbon characteristics from key low-permeability reservoir information yields hydrocarbon data. Calculating hydrocarbon saturation in low-permeability reservoir data based on strata elastic modulus data quantifies hydrocarbon saturation in low-permeability reservoirs. Integrating low-permeability reservoir deformation data, inter-layer permeability, and hydrocarbon saturation data yields comprehensive geological indicators for low-permeability reservoirs.

[0043] Preferably, step S44 includes the following steps:

[0044] Step S441: Perform pre-stack seismic inversion processing on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data;

[0045] Step S442: Extract deformation parameters of low-permeability reservoirs from the pre-stack inversion data of the model to generate low-permeability reservoir deformation inversion parameters; detect the permeability of the pre-stack seismic inversion data based on the low-permeability reservoir deformation parameters to obtain low-permeability reservoir permeability detection data; evaluate the hydrocarbon saturation of the pre-stack seismic inversion data using the low-permeability reservoir permeability detection data to generate hydrocarbon saturation evaluation data.

[0046] Step S443: Identify oil and gas geological sweet spots in the oil and gas saturation assessment data to obtain geological sweet spot data; determine the geological sweet spots in the geological sweet spots based on the low-permeability reservoir deformation inversion parameters, low-permeability reservoir permeability detection data and oil and gas saturation assessment data to generate low-permeability reservoir geological sweet spot determination data.

[0047] Step S444: Adjust the weights of geological indicators of low-permeability reservoirs in the low-permeability reservoir geological sweet spot prediction model based on the geological sweet spot measurement data of low-permeability reservoirs to obtain geological indicator weight adjustment data; use the geological indicator weight adjustment data to perform dynamic weight optimization on the low-permeability reservoir geological sweet spot prediction model to obtain the low-permeability reservoir geological sweet spot prediction optimization model.

[0048] This invention performs pre-stack seismic inversion processing on a low-permeability reservoir geological sweet spot prediction model, providing detailed information on the reservoir's seismic response and clarifying its physical characteristics. It extracts low-permeability reservoir deformation parameters from the pre-stack inversion data, obtaining these parameters. Based on these deformation parameters, it detects permeability in the pre-stack seismic inversion data, providing a quantitative indicator of reservoir permeability. It assesses hydrocarbon saturation in the pre-stack seismic inversion data using the low-permeability permeability detection data, clarifying the hydrocarbon saturation assessment status. Finally, it identifies hydrocarbon geological sweet spots in the reservoir using the hydrocarbon saturation assessment data, identifying relatively high-yield areas within the reservoir. The study focuses on the geological sweet spots in low-permeability reservoirs. Based on deformation inversion parameters, permeability detection data, and oil and gas saturation assessment data, the study determines the geological sweet spots in these areas, enabling a more specific identification of their geological conditions. Furthermore, by adjusting the weights of geological indicators in the low-permeability reservoir geological sweet spot prediction model based on the measured data, the model's parameter configuration is optimized, making it more closely aligned with actual geological conditions and thus improving prediction accuracy. Finally, dynamic weight optimization of the low-permeability reservoir geological sweet spot prediction model using the adjusted geological indicator weights further enhances the model's adaptability and predictive capabilities.

[0049] Preferably, step S45 includes the following steps:

[0050] Step S451: Based on the geological sweet spot prediction optimization model for low-permeability reservoirs, predict the porosity attribute of the geological sweet spot in the geological data of low-permeability reservoirs to obtain porosity attribute prediction data; based on the porosity attribute prediction data, spatially locate the sweet spot area in the geological data of low-permeability reservoirs to generate spatial prediction data of the sweet spot area.

[0051] Step S452: Perform permeability prediction on the spatial prediction data of the sweet spot area for low-permeability reservoirs to obtain regional permeability prediction data; use the regional permeability prediction data to predict oil and gas flow from the geological data of low-permeability reservoirs to generate oil and gas flow prediction data.

[0052] Step S453: Based on porosity attribute prediction data, regional permeability prediction data, and oil and gas flow prediction data, the geological data of low-permeability reservoirs are classified into sweet spot levels to obtain geological sweet spot level data; the geological sweet spot prediction optimization model for low-permeability reservoirs is used to predict the geological sweet spot level data to generate geological sweet spot prediction data for low-permeability reservoirs.

[0053] Step S454: Perform infographic processing on the low-permeability reservoir geological sweet spot prediction data to obtain a low-permeability reservoir geological sweet spot prediction chart; compile a visualization report on the low-permeability reservoir geological sweet spot prediction chart to obtain a low-permeability reservoir geological sweet spot prediction report.

[0054] This invention uses a low-permeability reservoir geological sweet spot prediction optimization model to predict the porosity attributes of geological sweet spots in low-permeability reservoir geological data, providing a quantitative description of the reservoir pore structure. Based on the porosity attribute prediction data, it spatially locates sweet spot regions in low-permeability reservoir geological data, clearly identifying the areas of sweet spots within the reservoir. It then uses the spatial prediction data of sweet spot regions to predict the permeability of low-permeability reservoirs in the region, predicting regional permeability characteristics. Finally, it uses the regional permeability prediction data to predict oil and gas flow in low-permeability reservoir geological data, enabling the prediction of oil and gas flow data. Based on porosity attribute prediction data and regional permeability... The prediction data of low-permeability reservoirs are classified into sweet spot grades based on the prediction data of oil and gas flow, thus providing a grading evaluation standard for reservoir sweet spot areas. The geological sweet spot prediction optimization model of low-permeability reservoirs is used to predict geological sweet spots based on the geological sweet spot grade data, providing a comprehensive description of reservoir sweet spot areas. The geological sweet spot prediction data of low-permeability reservoirs is processed into infographics to obtain low-permeability reservoir geological sweet spot prediction charts. The low-permeability reservoir geological sweet spot prediction charts are then used to compile a visualization report, resulting in a low-permeability reservoir geological sweet spot prediction report, which provides an intuitive and easy-to-understand graphical report.

[0055] This specification provides a low-permeability reservoir geological sweet spot prediction system for performing the aforementioned low-permeability reservoir geological sweet spot prediction method. The low-permeability reservoir geological sweet spot prediction system includes:

[0056] The low-permeability reservoir geological data acquisition module is used to acquire low-permeability reservoir geological data; preprocess the low-permeability reservoir geological data to obtain standard low-permeability reservoir geological data; perform multi-dimensional feature identification on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; and extract key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information.

[0057] The low-permeability reservoir structure and morphology identification module is used to extract low-permeability reservoir core data from key low-permeability reservoir information; extract core microstructure features from multidimensional geological feature data based on the low-permeability reservoir core data to obtain core microstructure features; and identify the low-permeability reservoir structure and morphology from key low-permeability reservoir information based on the core microstructure features to generate low-permeability reservoir morphology data.

[0058] The low-permeability reservoir seismic correlation analysis module is used to determine the distribution of low-permeability reservoirs based on low-permeability reservoir morphology data, thereby obtaining low-permeability reservoir range data; to identify seismic features of key low-permeability reservoir information using the low-permeability reservoir range data, generating low-permeability reservoir seismic data; to extract shale reservoir features from the low-permeability reservoir range data, thereby obtaining shale reservoir feature data; and to perform correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data, generating low-permeability reservoir seismic correlation data.

[0059] The low-permeability reservoir geological sweet spot prediction model module is used to calculate geological indicators of low-permeability reservoir geological data based on seismic correlation data, generating low-permeability reservoir geological indicators; construct a low-permeability reservoir geological sweet spot prediction model using the low-permeability reservoir geological correlation indicators, obtaining the low-permeability reservoir geological sweet spot prediction model; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots in low-permeability reservoir geological data based on the optimized low-permeability reservoir geological sweet spot prediction model, generating low-permeability reservoir geological sweet spot prediction data; and visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0060] This invention acquires and preprocesses geological data of low-permeability reservoirs through a low-permeability reservoir geological data acquisition module, ensuring data standardization and providing high-quality data support for subsequent analysis. Multidimensional feature recognition enables the extraction of rich geological information from standard geological data. The extraction of key low-permeability reservoir information further focuses on factors that significantly influence reservoir characteristics, improving the relevance and accuracy of the analysis. Through a low-permeability reservoir structure and morphology identification module, core samples of key low-permeability reservoirs are extracted, clearly defining the core information. Based on the low-permeability reservoir core data, multidimensional geological feature data are extracted to reveal the microstructure of the core, clarifying the understanding of the reservoir's internal structure and revealing its microscopic characteristics. Based on the core microstructure characteristics, the low-permeability reservoir structure and morphology are identified, clarifying the spatial distribution characteristics of the reservoir and providing data for subsequent determination of reservoir extent. The low-permeability reservoir seismic correlation analysis module determines the distribution of key low-permeability reservoirs based on low-permeability reservoir morphology data, thus clarifying the geographical distribution of the reservoirs and defining their extent. Seismic feature identification of key low-permeability reservoir information using low-permeability reservoir extent data reveals the seismic characteristics of the reservoir's geological structure, further clarifying the seismic situation of the low-permeability reservoirs. Shale reservoir feature extraction is performed on the low-permeability reservoir extent data to obtain shale reservoir characteristic data. Finally, correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data provides a comprehensive understanding of the correlation between low-permeability reservoir seismic data. This invention utilizes a low-permeability reservoir geological sweet spot prediction model module to calculate geological indicators of low-permeability reservoir geological data based on seismic correlation data, providing a quantitative reservoir indicator evaluation standard. A low-permeability reservoir geological sweet spot prediction model is constructed using these indicators. Pre-stack inversion is performed on the model to obtain pre-stack inversion data. The model is then dynamically optimized using this data, improving its accuracy and applicability, thereby enhancing the precision of geological sweet spot prediction. Based on the optimized model, geological sweet spot predictions are generated for low-permeability reservoir geological data. Finally, the prediction data is visualized to provide an intuitive prediction report. Therefore, this invention employs data processing, pattern recognition, and machine learning technologies to achieve multi-dimensional analysis of low-permeability reservoir geological characteristics and combines this with seismic characteristics for correlation analysis, thereby improving the accuracy of low-permeability reservoir geological sweet spot prediction.

[0061] A computer-readable storage medium storing a computer program, wherein the computer program is used to perform the aforementioned method for predicting geological sweet spots in low-permeability reservoirs. Attached Figure Description

[0062] Figure 1 A schematic diagram illustrating the steps of a method for predicting geological sweet spots in low-permeability reservoirs;

[0063] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3.

[0064] Figure 3 for Figure 2 A detailed flowchart illustrating the implementation steps of step S35.

[0065] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0066] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0067] Furthermore, the accompanying drawings are merely illustrative of the invention 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 can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0068] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0069] To achieve the above objectives, please refer to Figures 1 to 3 A method for predicting geological sweet spots in low-permeability reservoirs, the method comprising the following steps:

[0070] Step S1: Obtain geological data of low-permeability reservoirs; preprocess the geological data of low-permeability reservoirs to obtain standard low-permeability reservoir geological data; perform multi-dimensional feature identification on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; extract key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information.

[0071] Step S2: Extract core samples from key low-permeability reservoirs to generate low-permeability reservoir core data; extract core microstructure features from multidimensional geological characteristic data based on the low-permeability reservoir core data to obtain core microstructure features; identify the low-permeability reservoir structure morphology based on the core microstructure features to generate low-permeability reservoir morphology data.

[0072] Step S3: Based on the low-permeability reservoir morphology data, determine the distribution of low-permeability reservoir ranges for key low-permeability reservoir information to obtain low-permeability reservoir range data; use the low-permeability reservoir range data to identify seismic features of key low-permeability reservoir information to generate low-permeability reservoir seismic data; extract shale reservoir features from the low-permeability reservoir range data to obtain shale reservoir feature data; perform correlation analysis between the shale reservoir feature data and the low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data.

[0073] Step S4: Calculate geological indices for low-permeability reservoirs based on seismic correlation data, generating low-permeability reservoir geological indices; construct a low-permeability reservoir geological sweet spot prediction model using these indices; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots for low-permeability reservoirs based on the optimized model, generating low-permeability reservoir geological sweet spot prediction data; visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0074] This invention ensures data standardization by acquiring and preprocessing geological data of low-permeability reservoirs, providing high-quality data support for subsequent analysis. Multidimensional feature recognition enables the extraction of rich geological information from standard geological data. The extraction of key low-permeability reservoir information further focuses on factors that significantly influence reservoir characteristics, improving the relevance and accuracy of the analysis. Core extraction of key low-permeability reservoir information clarifies the core information of low-permeability reservoirs. Extraction of core microstructure features from multidimensional geological feature data based on low-permeability reservoir core data clarifies the understanding of the reservoir's internal structure and reveals its microscopic characteristics. Identification of the structural morphology of key low-permeability reservoirs based on core microstructure features clarifies the spatial distribution characteristics of the reservoir, providing data basis for subsequent determination of reservoir extent. Based on low-permeability reservoir morphology data, the distribution of key low-permeability reservoir information is determined, thus clarifying the geographical distribution of the reservoirs and defining their extent. Seismic feature identification is performed on key low-permeability reservoir information using low-permeability reservoir extent data, revealing the seismic characteristics of the reservoir's geological structure and thus clarifying the seismic situation of the low-permeability reservoirs. Shale reservoir feature extraction is performed on the low-permeability reservoir extent data to obtain shale reservoir characteristic data. Correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data provides a comprehensive understanding of the correlation between low-permeability reservoir seismic data. Based on seismic correlation data of low-permeability reservoirs, geological indicators are calculated from the geological data of low-permeability reservoirs, providing a quantitative evaluation standard for reservoir indicators. A prediction model for low-permeability reservoir geological sweet spots is constructed using these geological correlation indicators. Pre-stack inversion is performed on the prediction model to obtain pre-stack inversion data. The prediction model is then dynamically optimized using this pre-stack inversion data, improving its accuracy and applicability, thereby enhancing the precision of geological sweet spot prediction. Based on the optimized prediction model, geological sweet spots are predicted from the low-permeability reservoir geological data, generating predicted low-permeability reservoir geological sweet spot data. This data is then visualized to provide an intuitive prediction report. Therefore, this invention utilizes data processing technology, pattern recognition technology, and machine learning technology to achieve multi-dimensional analysis of the geological characteristics of low-permeability reservoirs and combines this with seismic characteristics for correlation analysis, thereby improving the accuracy of low-permeability reservoir geological sweet spot prediction.

[0075] In this embodiment of the invention, reference is made to Figure 1 The above is a schematic flowchart of the steps of a method for predicting geological sweet spots in low-permeability reservoirs according to the present invention. In this example, the method for predicting geological sweet spots in low-permeability reservoirs includes the following steps:

[0076] Step S1: Obtain geological data of low-permeability reservoirs; preprocess the geological data of low-permeability reservoirs to obtain standard low-permeability reservoir geological data; perform multi-dimensional feature identification on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; extract key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information.

[0077] In this embodiment of the invention, geological data of low-permeability reservoirs are obtained through geological exploration and geophysical logging techniques. Specifically, logging techniques, such as sonic transit time logging, are used to measure low-permeability reservoirs. The geological data of low-permeability reservoirs is preprocessed, including data cleaning, noise reduction, and format standardization, to obtain standard low-permeability reservoir geological data. Multidimensional feature recognition techniques, such as grey relational analysis, are used to analyze the standard low-permeability reservoir geological data. Specifically, porosity curve indices and fluid indices are identified to determine the multidimensional geological characteristics of low-permeability reservoirs. At the same time, nuclear magnetic resonance logging is used to identify the pore structure of the reservoir. Based on the multidimensional geological feature data, modern geological mathematical methods and signal analysis techniques, such as wavelet analysis, are applied to analyze the response characteristics of the low-permeability reservoir curves and identify the fluid types in the reservoir to extract key low-permeability reservoir information.

[0078] Step S2: Extract core samples from key low-permeability reservoirs to generate low-permeability reservoir core data; extract core microstructure features from multidimensional geological characteristic data based on the low-permeability reservoir core data to obtain core microstructure features; identify the low-permeability reservoir structure morphology based on the core microstructure features to generate low-permeability reservoir morphology data.

[0079] In this embodiment of the invention, feature extraction technology is used to extract key low-permeability reservoir information from low-permeability reservoir core samples. Specifically, scanning imaging technology is used to perform three-dimensional scanning of the low-permeability reservoir core to obtain the pore structure and fracture development characteristics of the core. Image processing software, such as Avizo, is used to perform grayscale analysis and image segmentation on the pore structure and fracture development characteristics of the core, thereby extracting the pore and fracture network structure of the core. Quantitative analysis is performed on the pore size distribution, pore throat radius, and pore structure connectivity of the core. Specifically, wavelet analysis technology is used to perform multi-scale refinement analysis on the microstructural features of the core to extract the microstructural features of the core. Based on the extracted microstructural features of the core, the support vector machine (SVM) algorithm is applied to identify the structural morphology of the low-permeability reservoir. By constructing different kernel functions, such as Gaussian kernel (RBF) or polynomial kernel, the nanoscale pore throat structure of the core is classified, thereby identifying the morphological characteristics of the low-permeability reservoir.

[0080] Step S3: Based on the low-permeability reservoir morphology data, determine the distribution of low-permeability reservoir ranges for key low-permeability reservoir information to obtain low-permeability reservoir range data; use the low-permeability reservoir range data to identify seismic features of key low-permeability reservoir information to generate low-permeability reservoir seismic data; extract shale reservoir features from the low-permeability reservoir range data to obtain shale reservoir feature data; perform correlation analysis between the shale reservoir feature data and the low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data.

[0081] In this embodiment of the invention, seismic attribute analysis is performed using three-dimensional seismic data volumes to specifically identify reservoir continuity and fault boundaries; mathematical morphology methods are used to detect and classify configurational boundaries in the seismic data, and different levels of reservoir configuration analysis are conducted to determine the spatial distribution range of low-permeability reservoirs; seismic inversion technology is applied, combined with low-permeability reservoir range data, to identify seismic features of key low-permeability reservoir information; velocity and density attributes in the seismic data volumes are used, and seismic inversion technology is used to specifically reflect the physical characteristics and fluid properties of the reservoirs; digital core technology, combined with image processing technology, is used to extract shale reservoir features from the low-permeability reservoir range data; a three-dimensional digital core model of the reservoir is constructed using scanning technology, specifically extracting connected pore structures and quantitatively characterizing the micropore structure features of rock samples, thereby obtaining shale reservoir feature data; correlation analysis is performed between shale reservoir parameters and low-permeability reservoir seismic data, and support vector machine (SVM) is used for pattern recognition and parameter optimization; specifically, the seepage mechanism of tight sandstone reservoirs is determined based on a dual-scale pore coupling seepage simulation method, thereby generating seismic correlation data for low-permeability reservoirs.

[0082] Step S4: Calculate geological indices for low-permeability reservoirs based on seismic correlation data, generating low-permeability reservoir geological indices; construct a low-permeability reservoir geological sweet spot prediction model using these indices; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots for low-permeability reservoirs based on the optimized model, generating low-permeability reservoir geological sweet spot prediction data; visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0083] In this embodiment of the invention, seismic rock physics analysis technology, combined with geostatistical methods, is used to calculate geological indicators of seismic correlation data for low-permeability reservoirs. Specifically, by analyzing seismic amplitude, frequency, and waveform characteristics, and combining core analysis data, key geological indicators such as reservoir porosity and permeability are calculated. Using these geological indicators, combined with machine learning algorithms, such as random forests or neural networks, a geological sweet spot prediction model for low-permeability reservoirs is constructed. Specifically, seismic attribute analysis combined with geostatistical inversion methods is used to determine high-quality areas of the reservoir and construct the prediction model. Pre-stack inversion techniques, such as AVO (amplitude variation and migration) technology, are applied to the geological data. The geological sweet spot prediction model is inverted before stacking to obtain pre-stack inversion data. Using this data, the geological sweet spot prediction model for low-permeability reservoirs is dynamically optimized. Specifically, parametric modeling techniques are employed, based on refined lithofacies unit constraints, to dynamically optimize the reservoir. Based on the optimized low-permeability reservoir geological sweet spot prediction model, geological sweet spots are predicted from the low-permeability reservoir geological data, generating prediction data. The predicted geological sweet spot data is then visualized using professional geological modeling software, such as Petrel, to create a 3D visualization of the data and generate a low-permeability reservoir geological sweet spot prediction report.

[0084] Preferably, step S2 includes the following steps:

[0085] Step S21: Identify the core features of key low-permeability reservoirs to obtain the core features of low-permeability reservoirs; extract the core data of key low-permeability reservoirs based on the core features of low-permeability reservoirs to generate low-permeability reservoir core data.

[0086] Step S22: Enhance the core features of the low-permeability reservoir core data to obtain enhanced core feature data; hierarchically process the enhanced core feature data to generate hierarchical core feature data; extract the hierarchical structure elements from the hierarchical core feature data to obtain core hierarchical structure element information.

[0087] Step S23: Mark the core layer structure of the multidimensional geological feature data using the core layer structure element information to obtain core layer structure marking data; identify the core microstructure features of the multidimensional geological feature data based on the core layer structure marking data to obtain core microstructure features;

[0088] Step S24: Based on the microstructural characteristics of the core, perform feature imaging processing on the key low-permeability reservoir information to obtain low-permeability reservoir structure images; perform structural image recognition on the low-permeability reservoir structure images to obtain low-permeability reservoir structure data; extract the low-permeability reservoir morphology from the low-permeability reservoir structure data to obtain low-permeability reservoir morphology data.

[0089] In this embodiment of the invention, key low-permeability reservoir information is identified through core feature recognition. Specifically, the reservoir is divided into lithological facies units at different microscales, and nuclear magnetic resonance logging technology is used to finely characterize the pore structure of different lithological facies units. Key features such as porosity and permeability of the low-permeability reservoir core are identified, and low-permeability reservoir core data is generated. Image processing techniques, such as image denoising and enhancement, are used to improve the contrast and clarity of the core images, thereby obtaining enhanced core feature data. Furthermore, image segmentation technology based on the Unet++ network is used to perform hierarchical processing of the core images to achieve hierarchical core features. The hierarchical data is analyzed to extract hierarchical structural element information of the core, such as pore size. The data includes information on core layer structure elements such as pore connectivity; image segmentation technology is used to label multidimensional geological feature data, generating core layer structure label data; based on the core layer structure label data, the microstructural features of the multidimensional geological feature data are identified, specifically such as pore structure and pore size distribution; digital core technology, combined with high-resolution imaging technology, such as focused ion beam scanning electron microscopy, is used to image the microstructural features of the core to obtain low-permeability reservoir structure images; further, image recognition technology is applied to identify the structural images and extract the morphological features of low-permeability reservoirs; morphological analysis methods are used to extract the morphology of the identified structural data to obtain low-permeability reservoir morphology data.

[0090] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S2 includes:

[0091] Step S31: Extract geometric features from the low-permeability reservoir morphology data to obtain the low-permeability reservoir geometric morphology features; determine the spatial relationships of the low-permeability reservoir geometric morphology features to generate low-permeability reservoir spatial relationship data; classify the low-permeability reservoir spatial relationship data into relation types to obtain spatial relationship type data.

[0092] Step S32: Based on the spatial relationship type data, identify the spatial distribution of low-permeability reservoir morphology data to generate low-permeability reservoir spatial distribution information; determine the low-permeability reservoir boundary based on the low-permeability reservoir spatial distribution information to obtain low-permeability reservoir boundary data; determine the low-permeability reservoir range distribution based on the low-permeability reservoir spatial distribution information and low-permeability reservoir boundary data to obtain low-permeability reservoir range data.

[0093] Step S33: Perform continuity determination on the low-permeability reservoir range data to generate continuity determination data; use the continuity determination data to determine the interlayer continuity of key low-permeability reservoir information to obtain interlayer continuity data; record the number of interlayer collisions and compressions on key low-permeability reservoir information based on the interlayer continuity data to obtain interlayer collision and compression data; perform seismic feature identification on the interlayer collision and compression data to generate low-permeability reservoir seismic data.

[0094] Step S34: Perform shape and structure identification on the low-permeability reservoir range data to obtain shale reservoir shape data; extract shale reservoir parameters from the shale reservoir shape data to generate shale reservoir parameter information; integrate the shale reservoir shape data and shale reservoir parameter information to obtain shale reservoir characteristic data;

[0095] Step S35: Perform correlation analysis between shale reservoir characteristic data and low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data.

[0096] In this embodiment of the invention, digital core technology combined with Micro-CT scanning technology is used to perform three-dimensional scanning of low-permeability reservoir cores, specifically extracting the geometric morphological features of the pore structure, such as pore size, shape, and distribution. Image processing software is used to analyze the CT scan data to determine the spatial relationships of the pore structure, such as the connectivity between pores and the complexity of the pore network. Pattern recognition technology is used to classify the spatial relationship data, obtaining data on different types of spatial relationships. Geostatistical methods are applied, combined with the spatial relationship type data, to perform spatial distribution analysis of the morphological data of the low-permeability reservoir, specifically determining the spatial distribution characteristics of the reservoir. Geological modeling software, such as Petrel, is used to accurately determine the reservoir boundaries, obtaining low-permeability reservoir boundary data. Combining spatial distribution information and boundary data, multidimensional geological modeling technology is used to determine the range distribution of the low-permeability reservoir, generating low-permeability reservoir range data. Seismic interpretation techniques, such as AVO analysis, are employed to analyze the low-permeability reservoir. The process involves several steps: First, continuous measurement data is generated from the seismic data volume through continuity analysis. Second, interlayer continuity data is generated using seismic inversion techniques based on velocity and amplitude attributes within the seismic data volume, recording the number of interlayer collisions and compressions. Third, seismic characteristics of interlayer collisions and compressions are identified through feature analysis of the seismic data volume, generating seismic data for low-permeability reservoirs. Fourth, shape and structure identification of low-permeability reservoir range data is performed using digital core technology combined with high-resolution imaging techniques, such as focused ion beam scanning electron microscopy (FIE), to obtain the pore and fracture network structure of shale reservoirs. Fifth, pore structure parameters of shale reservoirs are extracted using image processing and analysis software, such as Avizo, to generate shale reservoir parameter information. Sixth, shape data and parameter information are integrated using data fusion techniques to obtain shale reservoir characteristic data. Seventh, data association techniques are used to perform correlation analysis between shale reservoir characteristic data and low-permeability reservoir seismic data. Finally, pattern recognition and parameter optimization are used to generate seismic association data for low-permeability reservoirs.

[0097] As an example of the present invention, reference is made to... Figure 3 As shown, step S35 in this example includes:

[0098] Step S351: Identify porosity features in shale reservoir characteristic data to obtain porosity feature data; determine porosity distribution in porosity feature data to generate porosity distribution data; calculate porosity and pore size in combination with porosity distribution data to obtain porosity and pore size data.

[0099] Step S352: Extract the mineral composition of the shale reservoir characteristic data to obtain the mineral composition data of the shale reservoir; identify the fractures of the shale reservoir based on the porosity and pore size data to generate shale reservoir fracture information;

[0100] Step S353: Simulate the interlayer seismic response of shale reservoir fracture information and low-permeability reservoir seismic data to obtain interlayer seismic response data; determine the interlayer vulnerability of the interlayer seismic response data to generate interlayer vulnerability data; identify the surface texture of the fractured area of ​​the shale reservoir based on the interlayer vulnerability data to obtain regional surface texture information; compare the regional surface texture information with preset regional surface texture information; if the regional surface texture information is less than the preset regional surface texture information, mark the regional surface texture information as high seismic resistance layer information; if the regional surface texture information is greater than or equal to the preset regional surface texture information, mark the regional surface texture information as low seismic resistance layer information; integrate the high seismic resistance layer information and the low seismic resistance layer information to obtain the seismic resistance parameters of the low-permeability reservoir.

[0101] Step S354: Combine inter-layer seismic response data to calculate the seismic intensity of low-permeability reservoir seismic data and generate low-permeability reservoir seismic intensity data; perform inter-layer seismic characteristic correlation between low-permeability reservoir seismic intensity data and low-permeability reservoir seismic resistance parameters to generate low-permeability reservoir seismic correlation data.

[0102] In this embodiment of the invention, porosity characteristic data of shale reservoirs are obtained through nuclear magnetic resonance (NMR) logging technology. The T2 spectrum distribution of NMR is used to determine the porosity distribution, generating porosity distribution data. Then, combining the porosity distribution data, image analysis technology is used to calculate the porosity and pore size, obtaining porosity and pore size data. A core spectral scanner is used to extract the mineral composition of the shale reservoir characteristic data, obtaining mineral composition data. Combining the porosity and pore size data with the mineral composition data, fracture identification is performed, generating shale reservoir fracture information. Combining seismic data and shale reservoir fracture information, a regional building seismic damage simulation method is used to simulate inter-layer seismic response, obtaining inter-layer seismic response data. Using the seismic response data, through... Earthquake hazard simulation analysis identifies inter-layer vulnerability data. Based on this data, computer vision technology is used to identify surface textures in the fracturing area, obtaining regional surface texture information. Further, this regional surface texture information is compared with pre-defined regional surface texture information. Based on the comparison results, information on high-seismic-resistant and low-seismic-resistant layers is labeled, and this information is integrated to obtain seismic resistance parameters for low-permeability reservoirs. Using inter-layer seismic response data, seismic intensity calculation methods, such as earthquake hazard simulation analysis, are employed to calculate the seismic intensity of the low-permeability reservoir seismic data, generating seismic intensity data for low-permeability reservoirs. Further, combining the seismic intensity data and seismic resistance parameters of low-permeability reservoirs, seismic data analysis techniques are used to correlate inter-layer seismic characteristics, generating seismic correlation data for low-permeability reservoirs.

[0103] Preferably, step S4 includes the following steps:

[0104] Step S41: Calculate geological indicators of low-permeability reservoirs based on seismic correlation data of low-permeability reservoirs to generate geological indicators of low-permeability reservoirs.

[0105] Step S42: Construct a prediction model for the geological sweet spot of low-permeability reservoirs based on the geological correlation index of low-permeability reservoirs to obtain a pre-model for predicting the geological sweet spot of low-permeability reservoirs. Then, train the pre-model for predicting the geological sweet spot of low-permeability reservoirs using the geological correlation index of low-permeability reservoirs to obtain a training model for predicting the geological sweet spot of low-permeability reservoirs.

[0106] Step S43: Perform cross-validation evaluation on the low-permeability reservoir geological sweet spot prediction training model to obtain model evaluation data; adjust the model parameters of the low-permeability reservoir geological sweet spot prediction training model using the model evaluation data to obtain the low-permeability reservoir geological sweet spot prediction model.

[0107] Step S44: Perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the model pre-stack inversion data to obtain the low-permeability reservoir geological sweet spot prediction optimization model.

[0108] Step S45: Based on the low-permeability reservoir geological sweet spot prediction optimization model, perform geological sweet spot prediction on the low-permeability reservoir geological data to generate low-permeability reservoir geological sweet spot prediction data; perform data visualization on the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0109] In this embodiment of the invention, the amplitude, frequency, and waveform characteristics of seismic data are utilized. Specifically, key geological indicators such as reservoir porosity and permeability are calculated using seismic rock physics analysis techniques. Seismic amplitude and frequency attributes are analyzed, and combined with core analysis data, the geological indicators of the reservoir are calculated. A prediction model for low-permeability reservoir geological sweet spots is constructed based on the geological correlation indicators of low-permeability reservoirs. Specifically, the LightGBM regression algorithm is selected as the prediction model. The parameters are set as follows: learning rate (learning_rate) is 0.1, number of decision trees (num_iterations) is 200, maximum depth (max_depth) is 5, and minimum data amount required for feature splitting (min_data_in_leaf) is 20. Thus, a pre-model for predicting low-permeability reservoir geological sweet spots is constructed and trained using the geological correlation indicators of low-permeability reservoirs to obtain a prediction training model for low-permeability reservoir geological sweet spots. The low-permeability reservoir geological sweet spot prediction training model is evaluated through model cross-validation, specifically using the K-fold cross-validation method, where K is set to 5. In each iteration, 4 / 5 of the data is used. Training was performed, with the remaining 1 / 5 used for validation. Evaluation metrics included R² score (r²_score) and mean squared error (MSE), where a higher R² score (closer to 1) indicates a better model fit, and a lower MSE indicates a smaller prediction error. Model parameters were adjusted based on cross-validation results; for example, the learning rate was adjusted to 0.05 to avoid overfitting; the number of decision trees was increased to 300 to improve model complexity and predictive ability. Through these adjustments, a prediction model for low-permeability reservoir geological sweet spots was finally obtained. Pre-stack inversion techniques, such as... AVO (Amplitude Variation and Migration) technology is used to perform pre-stack inversion on geological sweet spot prediction models to obtain pre-stack inversion data. This data is then used to dynamically optimize the model to improve the accuracy of reservoir hydrocarbon prediction. Specifically, AVO technology is used to assess the rock properties of hydrocarbon reservoirs, including porosity, density, lithology, and fluid content. The optimized low-permeability reservoir geological sweet spot prediction model is then used to predict geological data, generating prediction data. Data visualization technology is used to present the prediction results in the form of graphs and charts, thus obtaining a geological sweet spot prediction report.

[0110] Preferably, step S41 includes the following steps:

[0111] Step S411: Measure the magnitude of seismic amplitude in the seismic correlation data of the low-permeability reservoir to obtain interlayer seismic amplitude data; identify the degree of interlayer deformation in the low-permeability reservoir morphology data based on the interlayer seismic amplitude data to generate low-permeability reservoir deformation data.

[0112] Step S412: Calculate the pore density of the geological data of the low-permeability reservoir based on the deformation data of the low-permeability reservoir to obtain the pore density of the low-permeability reservoir; identify the porosity of the pore density of the low-permeability reservoir to obtain the pore parameters of the low-permeability reservoir; determine the inter-layer permeability of the low-permeability reservoir based on the pore parameters of the low-permeability reservoir to generate the inter-layer permeability of the low-permeability reservoir.

[0113] Step S413: Combine the seismic parameters of the low-permeability reservoir to perform seismic coefficient permeability mapping on the permeability between low-permeability reservoirs to obtain the permeability data of the seismic-resistant layer; calculate the rock elastic modulus of the geological data of the low-permeability reservoir based on the permeability data of the seismic-resistant layer to generate the rock elastic modulus data.

[0114] Step S414: Extract oil and gas features from key low-permeability reservoir information to obtain low-permeability reservoir oil and gas data; calculate oil and gas saturation based on rock elastic modulus data to generate low-permeability reservoir oil and gas saturation data.

[0115] Step S415: Integrate low-permeability reservoir deformation data, inter-layer permeability data, and low-permeability reservoir oil and gas saturation data to obtain low-permeability reservoir geological indicators.

[0116] In this embodiment of the invention, seismic data processing software, such as RockWorks, is used to perform amplitude analysis on seismic data, specifically determining the magnitude of seismic amplitude to obtain inter-layer seismic amplitude data. Deformation degree identification technology, such as deformation analysis methods based on seismic amplitude changes, is applied to determine the deformation status of low-permeability reservoir morphology data, generating low-permeability reservoir deformation data. Nuclear magnetic resonance logging (Log-NMR) technology is used to determine pore structure and porosity by measuring the nuclear magnetic resonance signal of fluids in rock pores; for example, by analyzing the T2 spectrum of nuclear magnetic resonance, information on the distribution of different pore sizes is obtained, and then the pore density is calculated. High-resolution array induction logging technology is employed to specifically identify porosity by measuring resistivity changes at different depths. Combining the deformation data of low-permeability reservoirs, nuclear magnetic resonance logging (Log-NMR) and array induction logging technology are used to calculate permeability by analyzing pore structure and fluid distribution; for example, the T2 distribution and porosity data obtained through nuclear magnetic resonance logging are used. The following methods are employed: First, core analysis data is combined to calculate the permeability between low-permeability reservoirs. Second, seismic parameters of low-permeability reservoirs, such as seismic wave velocity, amplitude, and frequency, are used to map the seismic resistance coefficient to permeability, thus obtaining permeability data for the seismic-resistant layer. Third, downhole acoustic logging technology is used to calculate the elastic modulus of the rock by measuring the propagation time of acoustic waves at different depths. For example, vertical seismic profiling (VSP) technology is used to obtain the propagation characteristics of downhole acoustic waves and then calculate the elastic modulus. Fourth, downhole spectral logging technology is used to extract oil and gas characteristics by analyzing the absorption spectra of different elements in the rock. For example, the presence of oil and gas is identified by measuring the infrared absorption spectrum of downhole rocks. Fifth, oil and gas saturation data of low-permeability reservoirs are calculated by combining rock elastic modulus data with oil and gas data, thus obtaining oil and gas saturation data for low-permeability reservoirs. Sixth, data integration techniques, such as data fusion methods based on geological cloud platforms, are used to integrate low-permeability reservoir deformation data, permeability data, and oil and gas saturation data to finally obtain geological indicators of low-permeability reservoirs.

[0117] Preferably, step S44 includes the following steps:

[0118] Step S441: Perform pre-stack seismic inversion processing on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data;

[0119] Step S442: Extract deformation parameters of low-permeability reservoirs from the pre-stack inversion data of the model to generate low-permeability reservoir deformation inversion parameters; detect the permeability of the pre-stack seismic inversion data based on the low-permeability reservoir deformation parameters to obtain low-permeability reservoir permeability detection data; evaluate the hydrocarbon saturation of the pre-stack seismic inversion data using the low-permeability reservoir permeability detection data to generate hydrocarbon saturation evaluation data.

[0120] Step S443: Identify oil and gas geological sweet spots in the oil and gas saturation assessment data to obtain geological sweet spot data;

[0121] Based on the deformation inversion parameters of low-permeability reservoirs, the permeability detection data of low-permeability reservoirs, and the oil and gas saturation assessment data, the geological sweet spot data of low-permeability reservoirs is determined by analyzing the geological sweet spot data of the geological sweet spot area.

[0122] Step S444: Adjust the weights of geological indicators of low-permeability reservoirs in the low-permeability reservoir geological sweet spot prediction model based on the geological sweet spot measurement data of low-permeability reservoirs to obtain geological indicator weight adjustment data; use the geological indicator weight adjustment data to perform dynamic weight optimization on the low-permeability reservoir geological sweet spot prediction model to obtain the low-permeability reservoir geological sweet spot prediction optimization model.

[0123] In this embodiment of the invention, the region is set as a low-permeability reservoir area, and the input raw seismic data is a specific time-domain dataset with a frequency range of 5-60 Hz. Specifically, inversion algorithms, such as wave equation inversion and least squares, are used to perform pre-stack seismic inversion processing on the low-permeability reservoir geological sweet spot prediction model, ultimately obtaining the model's pre-stack inversion data. Deformation parameters of the low-permeability reservoir are extracted from the model's pre-stack inversion data. Specifically, nuclear magnetic resonance logging and sonic transit time logging are used to extract the low-permeability reservoir deformation inversion parameters. Based on the low-permeability reservoir deformation inversion parameters, permeability is detected in the pre-stack seismic inversion data, and the permeability of the low-permeability reservoir is detected using logging data and core analysis. The permeability detection data is then analyzed. Based on the assessment of oil and gas saturation, oil and gas saturation assessment data is generated. After obtaining the oil and gas saturation assessment data, oil and gas geological sweet spots are identified. The fuzzy comprehensive evaluation method and the independent weight coefficient method are used to identify geological sweet spot data. Based on the measurement data of geological sweet spots in low-permeability reservoirs, the geological index weights of the low-permeability reservoir geological sweet spot prediction model are adjusted. The independent weight coefficient method, combined with correlation coefficients and multiple correlation coefficients, is used to obtain the geological index weight adjustment data. Through dynamic weight optimization, an optimized prediction model for low-permeability reservoir geological sweet spots is generated.

[0124] Preferably, step S45 includes the following steps:

[0125] Step S451: Based on the geological sweet spot prediction optimization model for low-permeability reservoirs, predict the porosity attribute of the geological sweet spot in the geological data of low-permeability reservoirs to obtain porosity attribute prediction data; based on the porosity attribute prediction data, spatially locate the sweet spot area in the geological data of low-permeability reservoirs to generate spatial prediction data of the sweet spot area.

[0126] Step S452: Perform permeability prediction on the spatial prediction data of the sweet spot area for low-permeability reservoirs to obtain regional permeability prediction data; use the regional permeability prediction data to predict oil and gas flow from the geological data of low-permeability reservoirs to generate oil and gas flow prediction data.

[0127] Step S453: Based on porosity attribute prediction data, regional permeability prediction data, and oil and gas flow prediction data, the geological data of low-permeability reservoirs are classified into sweet spot levels to obtain geological sweet spot level data; the geological sweet spot prediction optimization model for low-permeability reservoirs is used to predict the geological sweet spot level data to generate geological sweet spot prediction data for low-permeability reservoirs.

[0128] Step S454: Perform infographic processing on the low-permeability reservoir geological sweet spot prediction data to obtain a low-permeability reservoir geological sweet spot prediction chart; compile a visualization report on the low-permeability reservoir geological sweet spot prediction chart to obtain a low-permeability reservoir geological sweet spot prediction report.

[0129] In this embodiment of the invention, a geological sweet spot prediction optimization model for low-permeability reservoirs is used. Based on seismic data and well logging data, such as P-wave velocity and density, pre-stack inversion is performed to predict the porosity attributes of low-permeability reservoir geological data. Sensitive parameters for geological "sweet spots" are determined through cross-analysis of well logging information, and parameter prediction is performed. Subsequently, based on the porosity attribute prediction data, the spatial location of sweet spot areas in the low-permeability reservoir geological data is specifically performed, generating spatial prediction data for sweet spot areas. For the located sweet spot area spatial prediction data, phase-controlled permeability detection technology based on pore structure parameters is used. Through multiple regression analysis, combined with parameters such as porosity, P-wave velocity, P-wave / S-wave velocity ratio, and shear compliance factor, the permeability of the regional low-permeability reservoir is predicted. Based on the regional permeability prediction data, combined with... Based on fluid properties and reservoir pressure data, hydrocarbon flow prediction is performed on low-permeability reservoir geological data, generating hydrocarbon flow prediction data. Based on porosity attribute prediction data, regional permeability prediction data, and hydrocarbon flow prediction data, a fuzzy comprehensive evaluation method is used to classify the sweet spot levels of low-permeability reservoir geological data. A low-permeability reservoir geological sweet spot prediction optimization model is used to predict geological sweet spots based on the geological sweet spot level data, generating low-permeability reservoir geological sweet spot prediction data. This data is then processed into information charts and graphs using data visualization technologies such as GIS software, presenting the prediction results in the form of charts, curves, and maps. Furthermore, a visualization report is compiled based on the low-permeability reservoir geological sweet spot prediction charts, integrating geological, well logging, seismic, and experimental analysis data to form a low-permeability reservoir geological sweet spot prediction report.

[0130] This specification provides a low-permeability reservoir geological sweet spot prediction system for performing the aforementioned low-permeability reservoir geological sweet spot prediction method. The low-permeability reservoir geological sweet spot prediction system includes:

[0131] The low-permeability reservoir geological data acquisition module is used to acquire low-permeability reservoir geological data; preprocess the low-permeability reservoir geological data to obtain standard low-permeability reservoir geological data; perform multi-dimensional feature identification on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; and extract key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information.

[0132] The low-permeability reservoir structure and morphology identification module is used to extract low-permeability reservoir core data from key low-permeability reservoir information; extract core microstructure features from multidimensional geological feature data based on the low-permeability reservoir core data to obtain core microstructure features; and identify the low-permeability reservoir structure and morphology from key low-permeability reservoir information based on the core microstructure features to generate low-permeability reservoir morphology data.

[0133] The low-permeability reservoir seismic correlation analysis module is used to determine the distribution of low-permeability reservoirs based on low-permeability reservoir morphology data, thereby obtaining low-permeability reservoir range data; to identify seismic features of key low-permeability reservoir information using the low-permeability reservoir range data, generating low-permeability reservoir seismic data; to extract shale reservoir features from the low-permeability reservoir range data, thereby obtaining shale reservoir feature data; and to perform correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data, generating low-permeability reservoir seismic correlation data.

[0134] The low-permeability reservoir geological sweet spot prediction model module is used to calculate geological indicators of low-permeability reservoir geological data based on seismic correlation data, generating low-permeability reservoir geological indicators; construct a low-permeability reservoir geological sweet spot prediction model using the low-permeability reservoir geological correlation indicators, obtaining the low-permeability reservoir geological sweet spot prediction model; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots in low-permeability reservoir geological data based on the optimized low-permeability reservoir geological sweet spot prediction model, generating low-permeability reservoir geological sweet spot prediction data; and visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

[0135] This invention acquires and preprocesses geological data of low-permeability reservoirs through a low-permeability reservoir geological data acquisition module, ensuring data standardization and providing high-quality data support for subsequent analysis. Multidimensional feature recognition enables the extraction of rich geological information from standard geological data. The extraction of key low-permeability reservoir information further focuses on factors that significantly influence reservoir characteristics, improving the relevance and accuracy of the analysis. Through a low-permeability reservoir structure and morphology identification module, core samples of key low-permeability reservoirs are extracted, clearly defining the core information. Based on the low-permeability reservoir core data, multidimensional geological feature data are extracted to reveal the microstructure of the core, clarifying the understanding of the reservoir's internal structure and revealing its microscopic characteristics. Based on the core microstructure characteristics, the low-permeability reservoir structure and morphology are identified, clarifying the spatial distribution characteristics of the reservoir and providing data for subsequent determination of reservoir extent. The low-permeability reservoir seismic correlation analysis module determines the distribution of key low-permeability reservoirs based on low-permeability reservoir morphology data, thus clarifying the geographical distribution of the reservoirs and defining their extent. Seismic feature identification of key low-permeability reservoir information using low-permeability reservoir extent data reveals the seismic characteristics of the reservoir's geological structure, further clarifying the seismic situation of the low-permeability reservoirs. Shale reservoir feature extraction is performed on the low-permeability reservoir extent data to obtain shale reservoir characteristic data. Finally, correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data provides a comprehensive understanding of the correlation between low-permeability reservoir seismic data. This invention utilizes a low-permeability reservoir geological sweet spot prediction model module to calculate geological indicators of low-permeability reservoir geological data based on seismic correlation data, providing a quantitative reservoir indicator evaluation standard. A low-permeability reservoir geological sweet spot prediction model is constructed using these indicators. Pre-stack inversion is performed on the model to obtain pre-stack inversion data. The model is then dynamically optimized using this data, improving its accuracy and applicability, thereby enhancing the precision of geological sweet spot prediction. Based on the optimized model, geological sweet spot predictions are generated for low-permeability reservoir geological data. Finally, the prediction data is visualized to provide an intuitive prediction report. Therefore, this invention employs data processing, pattern recognition, and machine learning technologies to achieve multi-dimensional analysis of low-permeability reservoir geological characteristics and combines this with seismic characteristics for correlation analysis, thereby improving the accuracy of low-permeability reservoir geological sweet spot prediction.

[0136] A computer-readable storage medium storing a computer program, wherein the computer program is used to perform the aforementioned method for predicting geological sweet spots in low-permeability reservoirs.

[0137] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0138] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for predicting geological sweet spots in low-permeability reservoirs, characterized in that, Includes the following steps: Step S1: Obtain geological data of low-permeability reservoirs; preprocess the geological data of low-permeability reservoirs to obtain standard geological data of low-permeability reservoirs; Multidimensional feature identification was performed on standard low-permeability reservoir geological data to obtain multidimensional geological feature data. Key low-permeability reservoir information is extracted from multidimensional geological feature data. Step S2: Extract core samples from key low-permeability reservoirs to generate low-permeability reservoir core data; extract core microstructure features from multidimensional geological characteristic data based on the low-permeability reservoir core data to obtain core microstructure features; identify the low-permeability reservoir structure morphology based on the core microstructure features to generate low-permeability reservoir morphology data. Step S3: Based on the low-permeability reservoir morphology data, determine the distribution of low-permeability reservoir ranges for key low-permeability reservoir information to obtain low-permeability reservoir range data; use the low-permeability reservoir range data to identify seismic features of key low-permeability reservoir information to generate low-permeability reservoir seismic data; extract shale reservoir features from the low-permeability reservoir range data to obtain shale reservoir feature data; perform correlation analysis between the shale reservoir feature data and the low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data. Step S4: Calculate geological indices for low-permeability reservoirs based on seismic correlation data, generating low-permeability reservoir geological indices; construct a low-permeability reservoir geological sweet spot prediction model using these indices; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots for low-permeability reservoirs based on the optimized model, generating low-permeability reservoir geological sweet spot prediction data; visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

2. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Identify the core features of key low-permeability reservoirs to obtain the core features of low-permeability reservoirs; extract the core data of key low-permeability reservoirs based on the core features of low-permeability reservoirs to generate low-permeability reservoir core data. Step S22: Enhance the core features of the low-permeability reservoir core data to obtain enhanced core feature data; hierarchically process the enhanced core feature data to generate hierarchical core feature data; extract the hierarchical structure elements from the hierarchical core feature data to obtain core hierarchical structure element information. Step S23: Mark the core layer structure of the multidimensional geological feature data using the core layer structure element information to obtain core layer structure marking data; identify the core microstructure features of the multidimensional geological feature data based on the core layer structure marking data to obtain core microstructure features; Step S24: Based on the microstructural characteristics of the core, perform feature imaging processing on the key low-permeability reservoir information to obtain low-permeability reservoir structure images; perform structural image recognition on the low-permeability reservoir structure images to obtain low-permeability reservoir structure data; extract the low-permeability reservoir morphology from the low-permeability reservoir structure data to obtain low-permeability reservoir morphology data.

3. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Extract geometric features from the low-permeability reservoir morphology data to obtain the low-permeability reservoir geometric morphology features; determine the spatial relationships of the low-permeability reservoir geometric morphology features to generate low-permeability reservoir spatial relationship data; classify the low-permeability reservoir spatial relationship data into relation types to obtain spatial relationship type data. Step S32: Based on the spatial relationship type data, identify the spatial distribution of low-permeability reservoir morphology data to generate low-permeability reservoir spatial distribution information; determine the low-permeability reservoir boundary based on the low-permeability reservoir spatial distribution information to obtain low-permeability reservoir boundary data; determine the low-permeability reservoir range distribution based on the low-permeability reservoir spatial distribution information and low-permeability reservoir boundary data to obtain low-permeability reservoir range data. Step S33: Perform continuity determination on the low-permeability reservoir range data to generate continuity determination data; use the continuity determination data to determine the interlayer continuity of key low-permeability reservoir information to obtain interlayer continuity data; record the number of interlayer collisions and compressions on key low-permeability reservoir information based on the interlayer continuity data to obtain interlayer collision and compression data; perform seismic feature identification on the interlayer collision and compression data to generate low-permeability reservoir seismic data. Step S34: Perform shape and structure identification on the low-permeability reservoir range data to obtain shale reservoir shape data; extract shale reservoir parameters from the shale reservoir shape data to generate shale reservoir parameter information; integrate the shale reservoir shape data and shale reservoir parameter information to obtain shale reservoir characteristic data; Step S35: Perform correlation analysis between shale reservoir characteristic data and low-permeability reservoir seismic data to generate low-permeability reservoir seismic correlation data.

4. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 3, characterized in that, Step S35 includes the following steps: Step S351: Identify porosity features in shale reservoir characteristic data to obtain porosity feature data; determine porosity distribution in porosity feature data to generate porosity distribution data; calculate porosity and pore size in combination with porosity distribution data to obtain porosity and pore size data. Step S352: Extract the mineral composition of the shale reservoir characteristic data to obtain the mineral composition data of the shale reservoir; identify the fractures of the shale reservoir based on the porosity and pore size data to generate shale reservoir fracture information; Step S353: Simulate the interlayer seismic response of shale reservoir fracture information and low-permeability reservoir seismic data to obtain interlayer seismic response data; determine the interlayer vulnerability of the interlayer seismic response data to generate interlayer vulnerability data; identify the surface texture of the fractured area of ​​the shale reservoir based on the interlayer vulnerability data to obtain regional surface texture information; compare the regional surface texture information with preset regional surface texture information; if the regional surface texture information is less than the preset regional surface texture information, mark the regional surface texture information as high seismic resistance layer information; if the regional surface texture information is greater than or equal to the preset regional surface texture information, mark the regional surface texture information as low seismic resistance layer information; integrate the high seismic resistance layer information and the low seismic resistance layer information to obtain the seismic resistance parameters of the low-permeability reservoir. Step S354: Combine inter-layer seismic response data to calculate the seismic intensity of low-permeability reservoir seismic data and generate low-permeability reservoir seismic intensity data; perform inter-layer seismic characteristic correlation between low-permeability reservoir seismic intensity data and low-permeability reservoir seismic resistance parameters to generate low-permeability reservoir seismic correlation data.

5. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Calculate geological indicators of low-permeability reservoirs based on seismic correlation data of low-permeability reservoirs to generate geological indicators of low-permeability reservoirs. Step S42: Construct a prediction model for the geological sweet spot of low-permeability reservoirs based on the geological correlation index of low-permeability reservoirs to obtain a pre-model for predicting the geological sweet spot of low-permeability reservoirs. Then, train the pre-model for predicting the geological sweet spot of low-permeability reservoirs using the geological correlation index of low-permeability reservoirs to obtain a training model for predicting the geological sweet spot of low-permeability reservoirs. Step S43: Perform cross-validation evaluation on the low-permeability reservoir geological sweet spot prediction training model to obtain model evaluation data; adjust the model parameters of the low-permeability reservoir geological sweet spot prediction training model using the model evaluation data to obtain the low-permeability reservoir geological sweet spot prediction model. Step S44: Perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the model pre-stack inversion data to obtain the low-permeability reservoir geological sweet spot prediction optimization model. Step S45: Based on the low-permeability reservoir geological sweet spot prediction optimization model, perform geological sweet spot prediction on the low-permeability reservoir geological data to generate low-permeability reservoir geological sweet spot prediction data; perform data visualization on the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

6. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 5, characterized in that, Step S41 includes the following steps: Step S411: Measure the magnitude of seismic amplitude in the seismic correlation data of the low-permeability reservoir to obtain interlayer seismic amplitude data; identify the degree of interlayer deformation in the low-permeability reservoir morphology data based on the interlayer seismic amplitude data to generate low-permeability reservoir deformation data. Step S412: Calculate the pore density of the geological data of the low-permeability reservoir based on the deformation data of the low-permeability reservoir to obtain the pore density of the low-permeability reservoir; identify the porosity of the pore density of the low-permeability reservoir to obtain the pore parameters of the low-permeability reservoir; determine the inter-layer permeability of the low-permeability reservoir based on the pore parameters of the low-permeability reservoir to generate the inter-layer permeability of the low-permeability reservoir. Step S413: Combine the seismic parameters of the low-permeability reservoir to perform seismic coefficient permeability mapping on the permeability between low-permeability reservoirs to obtain the permeability data of the seismic-resistant layer; calculate the rock elastic modulus of the geological data of the low-permeability reservoir based on the permeability data of the seismic-resistant layer to generate the rock elastic modulus data. Step S414: Extract oil and gas features from key low-permeability reservoir information to obtain low-permeability reservoir oil and gas data; calculate oil and gas saturation based on rock elastic modulus data to generate low-permeability reservoir oil and gas saturation data. Step S415: Integrate low-permeability reservoir deformation data, inter-layer permeability data, and low-permeability reservoir oil and gas saturation data to obtain low-permeability reservoir geological indicators.

7. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 5, characterized in that, Step S44 includes the following steps: Step S441: Perform pre-stack seismic inversion processing on the low-permeability reservoir geological sweet spot prediction model to obtain the model pre-stack inversion data; Step S442: Extract deformation parameters of low-permeability reservoirs from the pre-stack inversion data of the model to generate low-permeability reservoir deformation inversion parameters; detect the permeability of the pre-stack seismic inversion data based on the low-permeability reservoir deformation parameters to obtain low-permeability reservoir permeability detection data; evaluate the hydrocarbon saturation of the pre-stack seismic inversion data using the low-permeability reservoir permeability detection data to generate hydrocarbon saturation evaluation data. Step S443: Identify oil and gas geological sweet spots in the oil and gas saturation assessment data to obtain geological sweet spot data; Based on the deformation inversion parameters of low-permeability reservoirs, the permeability detection data of low-permeability reservoirs, and the oil and gas saturation assessment data, the geological sweet spot data of low-permeability reservoirs is determined by analyzing the geological sweet spot data of the geological sweet spot area. Step S444: Adjust the weights of geological indicators of low-permeability reservoirs in the low-permeability reservoir geological sweet spot prediction model based on the geological sweet spot measurement data of low-permeability reservoirs to obtain geological indicator weight adjustment data; use the geological indicator weight adjustment data to perform dynamic weight optimization on the low-permeability reservoir geological sweet spot prediction model to obtain the low-permeability reservoir geological sweet spot prediction optimization model.

8. The method for predicting geological sweet spots in low-permeability reservoirs according to claim 5, characterized in that, Step S45 includes the following steps: Step S451: Based on the geological sweet spot prediction optimization model for low-permeability reservoirs, predict the porosity attribute of the geological sweet spot in the geological data of low-permeability reservoirs to obtain porosity attribute prediction data; based on the porosity attribute prediction data, spatially locate the sweet spot area in the geological data of low-permeability reservoirs to generate spatial prediction data of the sweet spot area. Step S452: Perform permeability prediction on the spatial prediction data of the sweet spot area for low-permeability reservoirs to obtain regional permeability prediction data; use the regional permeability prediction data to predict oil and gas flow from the geological data of low-permeability reservoirs to generate oil and gas flow prediction data. Step S453: Based on porosity attribute prediction data, regional permeability prediction data, and oil and gas flow prediction data, the geological data of low-permeability reservoirs are classified into sweet spot levels to obtain geological sweet spot level data; the geological sweet spot prediction optimization model for low-permeability reservoirs is used to predict the geological sweet spot level data to generate geological sweet spot prediction data for low-permeability reservoirs. Step S454: Perform infographic processing on the low-permeability reservoir geological sweet spot prediction data to obtain a low-permeability reservoir geological sweet spot prediction chart; compile a visualization report on the low-permeability reservoir geological sweet spot prediction chart to obtain a low-permeability reservoir geological sweet spot prediction report.

9. A geological sweet spot prediction system for low-permeability reservoirs, characterized in that, For performing the method for predicting geological sweet spots in low-permeability reservoirs as described in claim 1, the system for predicting geological sweet spots in low-permeability reservoirs includes: The low-permeability reservoir geological data acquisition module is used to acquire low-permeability reservoir geological data; preprocess the low-permeability reservoir geological data to obtain standard low-permeability reservoir geological data; perform multi-dimensional feature identification on the standard low-permeability reservoir geological data to obtain multi-dimensional geological feature data; and extract key low-permeability reservoir information from the multi-dimensional geological feature data to obtain key low-permeability reservoir information. The low-permeability reservoir structure and morphology identification module is used to extract low-permeability reservoir core data from key low-permeability reservoir information; extract core microstructure features from multidimensional geological feature data based on the low-permeability reservoir core data to obtain core microstructure features; and identify the low-permeability reservoir structure and morphology from key low-permeability reservoir information based on the core microstructure features to generate low-permeability reservoir morphology data. The low-permeability reservoir seismic correlation analysis module is used to determine the distribution of low-permeability reservoirs based on low-permeability reservoir morphology data, thereby obtaining low-permeability reservoir range data; to identify seismic features of key low-permeability reservoir information using the low-permeability reservoir range data, generating low-permeability reservoir seismic data; to extract shale reservoir features from the low-permeability reservoir range data, thereby obtaining shale reservoir feature data; and to perform correlation analysis between shale reservoir parameters and low-permeability reservoir seismic data, generating low-permeability reservoir seismic correlation data. The low-permeability reservoir geological sweet spot prediction model module is used to calculate geological indicators of low-permeability reservoir geological data based on seismic correlation data, generating low-permeability reservoir geological indicators; construct a low-permeability reservoir geological sweet spot prediction model using the low-permeability reservoir geological correlation indicators, obtaining the low-permeability reservoir geological sweet spot prediction model; perform pre-stack inversion on the low-permeability reservoir geological sweet spot prediction model to obtain pre-stack inversion data; dynamically optimize the low-permeability reservoir geological sweet spot prediction model using the pre-stack inversion data to obtain an optimized low-permeability reservoir geological sweet spot prediction model; predict geological sweet spots in low-permeability reservoir geological data based on the optimized low-permeability reservoir geological sweet spot prediction model, generating low-permeability reservoir geological sweet spot prediction data; and visualize the low-permeability reservoir geological sweet spot prediction data to generate a low-permeability reservoir geological sweet spot prediction report.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed, it implements the method for predicting geological sweet spots in low-permeability reservoirs as described in any one of claims 1 to 8.

Images