31 results about "Tectonic model" patented technology

The invention discloses a sedimentary facies model construction method and device, equipment and a storage medium. The method comprises the following steps: acquiring regional geological parameters of a target construction region; and determining structure interpretation data according to the regional geological parameters, and constructing a three-dimensional structure model according to the structure interpretation data. And analyzing distribution information of each lithofacies in the three-dimensional structure model according to regional logging data, and constructing a lithofacies framework model. A target sedimentary facies model corresponding to the target construction area is constructed according to the three-dimensional sedimentary point cloud and the lithofacies framework model based on a multi-point geostatistics algorithm, and the three-dimensional sedimentary point cloud is obtained by performing actual sedimentary simulation on the target sedimentary construction model obtained through pre-training based on the area geological parameters of the target construction area. According to the technical scheme, the three-dimensional sedimentary point cloud of the research area can be accurately extracted, high geological objectivity and physical authenticity are achieved, and therefore the objectivity and geological reliability of the sedimentary facies model are remarkably improved.

The invention discloses a multi-source data fusion-oriented comprehensive geological cause analysis method and system. The method comprises the following steps of 1, generating semantic mapping data; 2, constructing a cause characteristic pedigree tree, and outputting cause main chain data; step 3, generating geological space-time interlacing data; 4, limiting a time window and a space window, and generating spectrum fusion modeling data through a spectrum-guided fusion converter; 5, constructing a three-dimensional voxel grid, and carrying out dynamic clustering to generate a three-dimensional construction model; step 6, generating geological evolution field data; and 7, inputting the geological evolution field data into the improved CrossViT model, setting an interlayer cross attention unit, reconstructing a Token interaction mode, introducing a cause field constraint attention modulation mechanism, and outputting a geological cause analysis result. According to the method, high-precision analysis of complex geological causes is realized through multi-source data semantic mapping, cause pedigree reasoning and CrossViT model improvement.

The invention belongs to the field of oil and gas field exploration and development, and discloses a numerical reservoir model optimization and history fitting method and related device.The method comprises the steps that firstly, geological exploration data are utilized, a spatial clustering algorithm is adopted to automatically determine the local updating range of a structural model, and structural layers and stratums are optimized through spatial interpolation; outputting an optimized construction model and uncertainty measurement; based on the model and logging lithofacies attribute data, adaptively identifying a lithofacies updating key area and variation function parameters by applying a machine learning algorithm, and performing lithofacies simulation to output an optimized lithofacies model; combining the porosity, permeability and saturation attribute data, adaptively determining an attribute optimization range and variation function parameters by adopting a probabilistic neural network, and outputting an optimized attribute model through phase control attribute modeling; a three-dimensional oil-water two-phase black oil seepage model is constructed, a global optimization algorithm is adopted to automatically adjust parameters based on a target function, simulation data is calculated until convergence, optimal parameters are output to complete historical fitting, and efficient and reliable intelligent support is provided for complex oil reservoir development.

The invention discloses a fault sample generation method based on oil and gas field seismic data, and belongs to geophysical exploration and machine learning data engineering.The technical scheme includes the steps that seismic exploration data, well data and actual geological data of an actual work area are obtained; establishing an initial structure model based on seismic exploration data, well data and actual geological data; iteratively adjusting the initial structure model according to a forward modeling result to obtain a calibration structure model; performing process simulation based on the calibration structure model, and outputting a plurality of geologic bodies according to time steps; and converting the geologic body into a forward modeling parameter model, carrying out seismic forward modeling to generate a seismic data body, and carrying out data enhancement and sample amplification. The method has the advantages that compared with other sample generation methods, the generated samples better conform to the actual geological law, so that the sample quality is improved, and meanwhile, the number of the samples can be greatly increased by simulating the forward modeling process of the actual body.

This disclosure relates to the field of seismic exploration technology, and particularly to a method, equipment, and medium for depth domain velocity model analysis based on well-seismic co-processing. The method includes: extracting stratigraphic and fault information from seismic data to construct a three-dimensional depth domain structural model of the target three-dimensional work area; smoothing well logging velocities and extrapolating the smoothed well logging velocities across the entire work area based on the three-dimensional depth domain structural model; inverting an initial velocity model based on the three-dimensional depth domain structural model and the overall work area well logging velocity model; iterating the initial depth domain velocity model based on the structural properties of the three-dimensional depth domain structural model to obtain an isotropic depth domain velocity model; analyzing the anisotropic parameters and anisotropic velocities of the target three-dimensional work area, and iterating the isotropic velocity model based on the anisotropic parameters and anisotropic velocities to obtain the depth domain velocity model. This invention can improve the accuracy of depth domain velocity model analysis.

The invention provides a carbon dioxide sequestration potential assessment method and device based on grid attribute assignment, and relates to the technical field of carbon dioxide capture, utilization and sequestration, and the method comprises the steps: respectively constructing a three-dimensional structure model and a sedimentary facies model of a carbon dioxide sequestration region through structure characteristic analysis and sedimentary facies analysis; further establishing a coarsening model of the carbon dioxide sequestration geologic body; multi-source and multi-scale data are obtained, carbon sequestration attribute information in the multi-source and multi-scale data is mined, a carbon sequestration attribute prediction model is constructed, and the prediction model is used for predicting the carbon sequestration attribute information corresponding to the grids; according to the carbon sequestration attribute information predicted by the carbon sequestration attribute prediction model, assigning values to each grid in the coarsening model of the carbon dioxide sequestration geologic body to obtain a three-dimensional carbon sequestration attribute model of the carbon dioxide sequestration region; and performing carbon dioxide sequestration potential evaluation by using the three-dimensional carbon sequestration attribute model of the carbon dioxide sequestration area to obtain sequestration potential evaluation data.

The invention belongs to the technical field of reservoir geological modeling, and provides a clastic rock oil and gas reservoir key interface characterization modeling method, device, equipment and medium, and the method comprises the steps: constructing a geological modeling basic database; key interface identification and representation are carried out based on the constructed geological modeling basic database; constructing a construction model based on the recognition and representation results of the key interface, the seismic data volume and the well data; checking whether a geological interface in the tectonic model is consistent with well data and seismic interpretation, and if not, optimizing the tectonic model; based on the sedimentary microfacies data, a deterministic assignment method is adopted to construct a sedimentary microfacies model; based on the sedimentary microfacies model, a sequential indication method is adopted for simulation, and a lithologic model is constructed; constructing an attribute model based on the lithologic model; according to the method, the structure model is optimized, so that the structure model restores the structure characteristics as much as possible, and reliable spatial constraints are provided for coarsening and modeling of attributes such as lithology, porosity, permeability and saturation.

The present specification relates to the field of oil and gas field exploration and development, and particularly relates to a structure modeling method and device based on horizontal well layering. The method comprises: determining logging characteristics and logging identification marks of a straight well section according to logging data of the straight well section in a study area; determining layer-penetrating point information of a horizontal well according to logging curves of the horizontal well and comparing with the logging characteristics; constructing a well trajectory one section of the horizontal well according to seismic interpretation data and the logging identification marks; extracting corresponding isostatic elevation data of the layer-penetrating point on the well trajectory one section according to the logging identification marks; determining a correction amount according to the elevation of the layer-penetrating point and the isostatic elevation data; and correcting structure surfaces of each small layer according to the correction amount to obtain a structure model. The present scheme solves the problem that it is difficult to establish an accurate, reliable and fine structure model due to the small amount of small layer interface data in structure modeling of an oil and gas reservoir developed mainly by horizontal wells.

The invention relates to the technical field of oil field development, in particular to a high-precision three-dimensional velocity field establishment method and device, equipment and a medium, and the method comprises the steps: obtaining a time domain structure interpretation result according to the corresponding relation between an earthquake and a geologic horizon, and obtaining a high-precision three-dimensional velocity field according to time domain geologic stratified data by taking the time domain structure interpretation result as a constraint; establishing a time domain three-dimensional construction model; obtaining a depth domain structure interpretation result through time-depth conversion, and building a depth domain three-dimensional structure model by taking the depth domain structure interpretation result as a constraint according to the depth domain geological stratified data; according to the three-dimensional structure model, the layer velocity of seismic waves propagating in the geologic horizon is determined, and according to the layer velocity, a three-dimensional space velocity field is established by adopting a stacked velocity field modeling method. The problems that an existing velocity field establishing method is large in longitudinal error, the geologic body layer penetrating phenomenon is prone to occurring, and the requirements for three-dimensional seismic information body high-precision time-depth conversion and small-layer-scale well-seismic combined reservoir characterization cannot be met are solved.

This invention belongs to the field of seismic exploration technology, specifically a near-surface node seismic exploration system and its data collaborative acquisition method; it includes miniaturized intelligent node units, adaptive mesh network base stations, and a cloud-based collaborative processing platform. Each node integrates a three-component sensor, a hybrid timing module, an edge computing unit, and a multimodal communication interface, employing an adaptive coupling mechanism to adapt to complex terrain. The acquisition method includes intelligent deployment planning, adaptive network construction, multimodal data acquisition, edge-cloud collaborative processing, and dynamic quality control. The nodes are buried shallowly below the surface, receiving reflected seismic wave signals from deep underground targets without being affected by the complex shallow surface strata above. The node units can also receive direct signals from ground-based seismic sources, used to calibrate and detect near-surface velocities and tectonic models. This invention offers advantages such as high density and low cost, high-precision synchronization across all scenarios, efficient data transmission, and multi-physics field fusion.

The invention relates to a tectonic evolution method and device for recovering denudation strata through trend-structure extrapolation of an extended fractured basin, and the method comprises the steps: collecting geology, earthquake and logging information of the extended fractured basin, and building a three-dimensional geologic model of the extended fractured basin; carrying out horizon interpretation on a two-dimensional earthquake and a three-dimensional earthquake in a research area through seismic event stratum tracking by combining seismic reflection characteristics and drilling data on the basis of a synthetic seismic record calibration stratum; carrying out unconformity interface and fracture interpretation on the stretching fault basin, establishing a construction model, dividing stratums corresponding to the unconformity interface of the stretching fault basin, and determining formation time, distribution and evolution of the stratums; establishing a typical structure style, and selecting a typical profile; the denudated syngenetic fault and / or boundary fault and the stratum are / is recovered; and extrapolating and recovering the denudation fault and the stratum through a trend-structure external technology, and recovering a structure evolution history diagram. According to the method, the denudated stratum can be reliably recovered.

The application provides a large-scale coupling network simulation generation method and system based on three-dimensional geological structure, and belongs to the field of geological modeling and numerical simulation. The method comprises the following steps: integrating multi-source exploration data and constructing a structured geological database; performing geological interpretation based on the database to generate a stratum contour line; reconstructing three-dimensional entity models of strata, main faults and roadways respectively by using a triangular mesh modeling method based on the contour line, and coupling to generate a comprehensive three-dimensional hydrogeological structure model; converting model data into a standardized format compatible with multi-physical field simulation software through a special data interface and importing the model data into the simulation environment, giving the model physical properties, setting boundary conditions and coupling mechanisms, and completing large-scale coupling network simulation. The application solves the problems of insufficient multi-source data fusion, model distortion caused by artificial generalization and the fragmentation of modeling and simulation processes in traditional methods, and significantly improves the accuracy of the geological model and the reliability of the simulation analysis.

The invention provides a large-scale coupling network simulation generation method and system based on a three-dimensional geological structure, and belongs to the field of geological modeling and numerical simulation. Comprising the steps of integrating multi-source exploration data and constructing a structured geological database; geological interpretation is carried out based on the database, and a stratum contour line is generated; on the basis of the contour lines, through a triangular mesh modeling method, three-dimensional entity models of the stratum, the main fracture and the roadway are reconstructed respectively, and a comprehensive three-dimensional hydrogeological structure model is generated through coupling; model data is converted into a standardized format compatible with multi-physics field simulation software through a special data interface and imported, physical attributes are given to the model in a simulation environment, boundary conditions and a coupling mechanism are set, and large-scale coupling network simulation is completed. According to the method, the problems of model distortion and modeling and simulation flow separation caused by insufficient multi-source data fusion and manual generalization in a traditional method are solved, and the precision of a geologic model and the reliability of simulation analysis are remarkably improved.

A method for improving the accuracy of shale gas reservoir structural models, belonging to the field of unconventional and new energy, includes: conducting detailed interpretation of seismic data to predict small-scale fault development; establishing a fault model through well-seismic integration; identifying feature points on the horizontal sections of completed horizontal wells and converting them into virtual wells; compiling planar maps of different sub-layer thicknesses based on regional geological research results, extracting sub-layer thickness data at virtual wells and converting it into variable-thickness virtual well stratification point data; interpreting structural information of structural surfaces, appraisal wells, and pilot wells using seismic data, and establishing a bedding model under the control of variable-thickness virtual wells at feature points of horizontal wells; inserting small bedding planes into the fault model and bedding model using variable-thickness virtual well stratification point data and sub-layer thickness planar maps to form a shale gas reservoir structural model. The shale gas reservoir structural model established by this invention is more refined and can accurately describe the structural changes and sub-layer thickness variations of underground reservoirs, significantly improving the accuracy of faults and sub-layer bedding planes.

The invention belongs to the technical field of geological modeling, and provides a lithofacies model establishment method and device, a storage medium and electronic equipment, and the method comprises the steps: determining a plurality of target geological facies and a plurality of rock physical facies; the geological facies and the rock physical facies are coupled, and multiple rock types of the reservoir are determined; acquiring sampling data of multiple rock types of a reservoir in the target well; inputting the sampling data into the structural model, and outputting a static lithofacies model established in the structural model; and correcting and correcting the static lithofacies model to obtain a dynamic lithofacies model. Through the method, the static lithofacies model can be established based on the construction model, and the static lithofacies model is corrected to obtain the dynamic lithofacies model; the method for establishing the lithofacies model is wide in applicable reservoir range, and is particularly suitable for extremely thick and complex bioclastic limestone reservoirs.

The application discloses a comprehensive geological genesis analysis method and system for multi-source data fusion, and comprises the following steps: step one, generating semantic mapping data; step two, constructing a cause characteristic pedigree tree and outputting cause main chain data; step three, generating geological space-time interleaving data; step four, limiting a time window and a space window, and generating spectrum fusion modeling data through a spectrum guide fusion transformer; step five, constructing a three-dimensional voxel grid, and performing dynamic clustering to generate a three-dimensional structure model; step six, generating geological evolution field data; and step seven, inputting the geological evolution field data into an improved CrossViT model, setting an interlayer cross attention unit, reconstructing a Token interaction mode, introducing a cause field constraint attention modulation mechanism, and outputting a geological genesis analysis result. Through multi-source data semantic mapping, cause pedigree reasoning and the improved CrossViT model, high-precision analysis of complex geological genesis is realized.

The application provides a three-dimensional reservoir modeling method based on semantic rule guidance and multi-point geostatistics, relates to the geological field, and the method comprises the following steps: arranging strata in reverse order according to deposition time, iteratively executing a finite difference interpolation algorithm from the newest stratum, and taking the current stratum as a normal vector constraint of the next stratum; quantifying the constraint of fault kinematics on the surrounding deformation zone; under the support of an implicit structure model, adopting a two-dimensional parameter optimization strategy to dynamically configure the parameters of a fast sampling algorithm; adaptively matching the local structure features of a training image through a kernel function, and globally searching for an optimal combination in a grid; finally, taking a structural model as a spatial boundary constraint, performing collaborative simulation of lithofacies parameters, generating an attribute field according to the quantile sampling rule of Euclidean distance and Hamming distance, until all grid nodes are simulated and a three-dimensional reservoir model is output. The technical problem of collaborative fusion of multi-source heterogeneous data and geological semantic rules in a complex heterogeneous reservoir is effectively solved.

This invention provides a method for predicting source rocks in deep, well-free areas, comprising: constructing a stratigraphic velocity model; obtaining a three-dimensional stratigraphic structure image based on the stratigraphic velocity model and seismic data from the deep, well-free area, and performing seismic interpretation to determine the potential distribution range of source rocks; constructing a paleogeographic model based on the sedimentary facies and paleogeographic conditions of the deep, well-free area; constructing a tectonic model based on tectonic feature data; constructing a three-dimensional geological model based on the paleogeographic model, the tectonic model, and the stratigraphic data of the deep, well-free area; determining the distribution range of source rocks in the three-dimensional geological model; and overlaying the potential distribution area of ​​source rocks obtained from seismic analysis with the distribution range of source rocks obtained from the three-dimensional geological model to determine areas favorable for source rock distribution. By comprehensively utilizing seismic imaging and geological modeling methods, the accuracy of predicting the distribution of source rocks in deep, well-free areas is improved.

The application provides a wellbore stability prediction method, device and medium before drilling, and relates to the technical field of petroleum engineering, and comprises the following steps: obtaining a three-dimensional seismic data body and a seismic interval velocity of a to-be-drilled well area, well logging interpretation data of an adjacent well, and determining simulated well logging data of the to-be-drilled well; the adjacent well is a well adjacent to the to-be-drilled well; a fracture structure model of the to-be-drilled well area is established according to the simulated well logging data and the three-dimensional seismic data body; a formation pressure model, a rock mechanics attribute model and a ground stress model of the to-be-drilled well area are established based on the fracture structure model, the well logging interpretation data and the seismic interval velocity; a finite element model of the to-be-drilled well area is established according to the fracture structure model, the formation pressure model, the rock mechanics attribute model and the ground stress model; finite element forward modeling is carried out based on the finite element model to obtain a collapse pressure, so as to predict the wellbore stability of the to-be-drilled well. In this way, the influences of faults and fractures, ground stress and formation pressure on the wellbore stability are fully considered.

This invention discloses a seismic prediction method and system for natural alkali deposit exploration. The method includes: identifying the natural alkali development intervals and individual layer thicknesses through well logging lithology interpretation combined with well logging and core data, and determining natural alkali sensitivity logging curves; applying well-seismic joint interpretation technology to establish an isochronous stratigraphic framework, tracing the ore-bearing strata, and determining the structural characteristics of the top and bottom boundaries of the alkali-bearing strata and the seismic facies characteristics of the ore body; based on the structural characteristics of the top and bottom boundaries of the alkali-bearing strata, creating a stratigraphically constrained structural model, selecting sensitivity logging curves to reconstruct characteristic curves as control waveforms for seismic waveform simulation inversion, and predicting the distribution range and ore layer thickness of the natural alkali ore body; based on the seismic waveform simulation inversion results, characterizing the spatial morphology of the natural alkali ore body, and combining with a comprehensive evaluation of the deposit's geological conditions to determine the exploration engineering spacing and delineate the initial mining area. This invention applies seismic prediction technology to natural alkali deposit exploration for the first time, improving exploration efficiency and drilling success rate.

The present application relates to a kind of big platform horizontal well design method based on three-dimensional geological model, comprising: establishing structure model, establishing lithofacies model, establishing attribute model, determining the coordinate of multiple horizontal well wellsite, establishing design profile, first well trajectory design, big platform multi-well trajectory design.This big platform horizontal well deployment method provided by the present application is based on three-dimensional geological modeling, the accurate calculation of reference well elevation offset in three-dimensional model, the optimization of horizontal well trajectory in three-dimensional space, and the three-dimensional visualization of horizontal well design is realized.The lithology change, physical property change are considered at the same time, the coordinate of multiple wells is designed under a platform, multiple well trajectory design, when the first well is deployed on the platform, wellsite optimization can be carried out, when multiple wells are designed, reference adjacent well is drilled, and the basis of new well trajectory design is more sufficient.The problems of fast reservoir change of shale oil horizontal well, wellsite screening and horizontal well visualization trajectory design are effectively solved.

The application discloses a shale gas structural geological model correction method based on a wellbore trajectory and layered data, and comprises the following steps: constructing a stratum thickness model; obtaining small-layer apparent thickness plane data; establishing a passing mode of the wellbore trajectory in each small layer; calculating specific spatial positions of small-layer boundaries by the established passing mode of the wellbore trajectory in each small layer and in combination with the small-layer apparent thickness plane data; judging whether the spatial position result is continuous and smooth according to the calculated small-layer boundary spatial position result; and adjusting thickness data of the small-layer apparent thickness plane data and depth data of the specific spatial positions of the small-layer boundaries when the spatial position result is not smooth. The geological model correction method can better utilize horizontal well related information to quickly and efficiently process batch well data, thereby correcting seismic horizons, making the structural model well match the spatial position relationship of the horizontal well passing, and providing reliable data basis for the next attribute modeling and numerical simulation.

The invention provides an earthquake low-frequency information fluid prediction method. The seismic low-frequency information fluid prediction method comprises the following steps: S1, seismic data preprocessing and low-frequency signal amplitude preservation processing: preprocessing low-signal-to-noise-ratio seismic data collected from a target work area, the preprocessing comprising waveform distortion correction and low-frequency protective noise suppression based on a near-surface frozen soil model, to obtain amplitude-preserved seismic data, in particular low-frequency components; s2, low-frequency accompanying shadow anomaly extraction under the constraint of a construction model: performing time-frequency analysis on the data processed in the step S1 in a target reservoir section based on a backlash pushing construction model established on the basis of seismic interpretation, and quantitatively extracting pure low-frequency accompanying shadow anomaly below a target reservoir in a mode of stripping a regional background field; and S3, multi-attribute collaborative fluid comprehensive evaluation. The earthquake low-frequency information fluid prediction method provided by the invention has the advantage of being capable of effectively adapting to low signal-to-noise ratio, frozen soil layer influence and complex structure geological conditions.

The invention discloses a strike-slip fault control effective crack evaluation method based on tectonic stress field simulation, which comprises the following steps of: determining a target horizon and fault-crack distribution characteristics of a research area, determining tectonic stress field simulation boundary conditions, constructing a reasonable tectonic model and a mathematical model, carrying out tectonic stress field simulation, and then evaluating a strike-slip fault control effective crack according to a simulation result. The method comprises the following steps: calculating geological and engineering parameters for representing the fault-joint system, performing weight assignment by using an analytic hierarchy process, comprehensively constructing effectiveness evaluation parameters of the fault-joint system by multiple parameters, and establishing effectiveness evaluation standards of the fault-joint system by calibrating the effectiveness evaluation parameters of the fault-joint system. And finally, accurately determining the spatial distribution area of the effective cracks in the fracture-crack system. According to the method, on the basis of a rock fracture criterion, the development distribution rule of effective fractures in a reservoir is predicted by taking ground stress as an application condition, and an effective scheme is provided for professional technicians for researching the effectiveness distribution characteristics of a fracture-fracture system in a strike-slip fault zone and well location deployment.

The application discloses a mine structure detection method based on a multi-parameter imaging technology, and belongs to the technical field of mine structure detection. The method comprises the following steps: S1, a laser scanning mode is used to scan and model the inner side of a mine tunnel, and an inner surface model of the mine is obtained; S2, the mine is divided into a plurality of groups, each group comprising sequentially arranged detection sections and prediction sections; S3, a collection device is arranged in the detection sections, and sound logging curves and density logging curves of the detection sections are obtained; S4, a geological model of the detection sections is constructed according to the information collected in step S3; S5, the prediction sections are sampled according to the geological models of adjacent detection sections, and the geological model is completed; and S6, a mine structure model is obtained according to the geological models of the detection sections and the prediction sections. According to the method, the inner surface model of the mine is established by the surface scanning mode, the inner surface model of the mine is filled correspondingly, and finally, the mine structure model is formed, and the parameter quantization of the mine model is realized.

The application discloses a kind of fracture-cave carbonate gas reservoir modeling methods, it is related to carbonate rock oil and gas geology and development technical field, solve the simple, lack of systematicness of existing fracture-cave modeling method;The simulation result of cave and fracture is not reasonable enough, and then it is difficult to be applied to guide gas field exploration and development and the like problem, the method of the present application establishes fine structure model comprehensively stratum correlation and seismic interpretation result.On the basis of the fine structure model, matrix attribute model, fracture model and cave model are established, and finally a dual medium model is constructed through the matrix attribute model, the fracture model and the cave model.The method of the present application establishes a fine geological model to depict the distribution space of the preferred reservoir based on the integration of the distribution patterns of matrix, fractures and caves. By analyzing the controlling factors of fractures and caves and establishing a fracture-cave model, more accurate decision-making basis can be provided for gas field development, and the exploration and development of gas fields can be guided.

This invention discloses a method and system for risk assessment of small geological structures in mining compliance areas based on multi-source data fusion, relating to the field of coal mine safety technology. The method includes: identifying various small geological structures based on multi-source geological data; extracting quantitative characteristic parameters of each type of small structure and calibrating its influence radius to construct a three-dimensional structural model; spatially fusing the quantitative characteristic parameters, real-time monitoring data, and numerical simulation data of various small geological structures based on the three-dimensional structural model and influence radius to generate a multi-parameter three-dimensional fused data field; calculating the spatial distribution overlap of key risk factors and identifying high-risk spatial units based on the three-dimensional fused data field; inputting the three-dimensional fused data field into a robust gradient boosting decision tree model to output dynamic risk levels and dynamic weights of each risk dimension; and generating targeted remediation measures based on the dynamic risk level, dynamic weights of each risk dimension, and small geological structure type. This achieves accurate positioning, dynamic assessment, and targeted remediation of small geological structure risks.

The invention provides a quantitative evaluation method and device for dynamic packer of a low-permeability carbonate rock gas reservoir interlayer. The corresponding method comprises the steps that a reservoir model of a research area is generated according to a structural model, a facies model, an attribute model, a well type, a well pattern and a production system of the research area; traversing variable parameters influencing the packing property of the interlayers of the research area in the reservoir model to determine the reservoir outflowing quantity flowing out of the interlayers in different production periods of the research area; and evaluating the packing property of the interlayer in different production periods of the research area according to the reservoir outbound quantity. Based on core analysis data, a numerical simulation method is applied, through gas reservoir rock type division and total factor uncertainty simulation, the dynamic packer of the interlayer is quantitatively evaluated, and a geological basis is provided for gas reservoir development scheme compilation and accurate yield prediction.

This invention discloses an integrated evaluation method for geological modeling, fracture propagation, and fluid-structure interaction production prediction, involving multiple technical fields. The method includes: constructing a three-dimensional geological model of a deep shale gas reservoir; constructing a hydraulic fracture model based on a structural model, attribute model, three-dimensional geostress model, and rock mechanics parameter model; establishing a fluid-structure interaction production simulation mathematical model for the deep shale gas reservoir based on the hydraulic fracture model and the three-dimensional geological model; predicting the production of fractured horizontal wells in the shale gas reservoir based on the fluid-structure interaction production simulation mathematical model; and comprehensively evaluating and optimizing the prediction results. This invention integrates multiple stages such as geological modeling, fracture propagation simulation, and fluid-structure interaction production prediction, comprehensively considering the geological characteristics of the gas reservoir, fracture propagation patterns, and the interaction between fluid flow and solid deformation, thereby achieving a comprehensive evaluation and optimized development of deep shale gas reservoirs.

The application discloses a method and system for realizing integration of geological modeling and stress field simulation, and belongs to the technical field of geological exploration and development. The method comprises the following steps: project setting and data loading, seismic interpretation and attribute analysis, structural modeling, discrete fracture network modeling and stress field simulation. The application takes a three-dimensional structural model as a unified carrier, takes stratum boundaries, fracture distribution and stratum layering as basic constraints of mechanical modeling, realizes complete matching of a geological framework and a mechanical model grid and a spatial range, adopts a fracture distance constraint and a variogram hierarchical control coupling method, distinguishes a main fracture, a secondary fracture and a surrounding rock area, sets different interpolation rules, takes lithology and a tectonic deformation zone as partition constraints, and completes spatial simulation of mechanical parameters under geological recognition constraints. The structural model provides a spatial boundary for mechanical attribute modeling, and mechanical parameter inversion results feed back and correct the geological model, so that two-way data transmission and collaborative constraints are formed.