Mineral resource prediction model construction method based on complex geological background

Abstract

The invention discloses a mineral resource prediction model construction method based on a complex geological background, and relates to the technical field of mineral resource prediction, a three-dimensional ideal geological background field without mineralization interference is constructed based on the geological background of an unknown mining area, and the three-dimensional ideal geological background field is used for displaying the mining field of a non-mineralization area; constructing a three-dimensional real geological field based on the exploration data, and comparing the three-dimensional real geological field with the three-dimensional ideal geological background field to obtain first difference distribution data; extracting geologic features of the known mineralization area, and generating interference factors based on the geologic features; introducing the interference factor into the ideal geological background field to generate an interference coupling background field, and extracting the difference between the interference coupling background field and the ideal geological background field as second difference distribution data; the mineral resource prediction model is constructed based on the first difference distribution data and the second difference distribution data, the method can adapt to prediction requirements of different geological backgrounds, and the intelligent level and application flexibility of a mineral resource prediction system are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mineral resource prediction, and in particular to a method for constructing a mineral resource prediction model based on a complex geological background. Background Art

[0002] The mainstream technical approaches in the field of mineral resource prediction currently include statistically based data-driven methods and expert experience models based on geological concepts. Data-driven methods, such as weight of evidence methods, random forests, support vector machines and other machine learning algorithms, achieve spatial distribution prediction of mineral resources through statistical analysis and feature mining of large amounts of exploration data and known mineral deposit information. Existing technologies perform well in areas with sufficient data and clear mining types, but they generally rely on historical data and statistical correlations, making it difficult to fully express the dynamic evolution of complex geological processes. They are also less adaptable to unknown or rare mineralization types and have limited interpretability. Meanwhile, expert knowledge models based on geological concepts rely on experts' deep understanding of mineral deposit types and geological ore-controlling factors, and use methods such as genetic models and mineral deposit models for prospecting and prediction. These methods are widely used in areas with mature understanding of mineralization geology, but are highly subjective. The modeling process relies on expert experience and qualitative analysis, making it difficult to systematically integrate complex multi-source data. Furthermore, the models lack generalizability and scalability, making it difficult to meet the needs of refined predictions in complex geological settings such as deep and concealed deposits. Therefore, it is of great significance to develop a method for constructing a mineral resource prediction model based on complex geological background. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for constructing a mineral resource prediction model based on a complex geological background to solve the problems in the background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for constructing a mineral resource prediction model based on a complex geological background, comprising: Based on the geological background of the unknown mining area, a three-dimensional ideal geological background field without mineralization interference is constructed. The three-dimensional ideal geological background field is used to display the mining area without mineralization; constructing a three-dimensional realistic geological field based on the exploration data, and comparing the three-dimensional realistic geological field with the three-dimensional ideal geological background field to obtain first difference distribution data; Extract the geological characteristics of known mineralized areas and generate interference factors based on the geological characteristics; introducing interference factors into the ideal geological background field to generate an interference coupling background field, and extracting the difference between the interference coupling background field and the ideal geological background field as second difference distribution data; A mineral resource prediction model is constructed based on the first difference distribution data and the second difference distribution data.

[0005] In a preferred embodiment, the steps of constructing a three-dimensional ideal geological background field without mineralization interference are: Collect geological background data of unknown mining areas and construct a three-dimensional ideal geological background field based on the ideal background field generation model; The ideal background field generation model includes a lithology distribution simulator, a structural grid generator, a physical property field calculator, and a voxel segmenter; The lithology distribution simulator is based on the Markov random field algorithm and uses geological background data to constrain the lithology transition probability; The structural grid generator simulates the structural generation environment based on the stress field fitted by B-spline surface and generates a fracture network that conforms to the laws of mechanics. The physical property calculator calculates the density, magnetic susceptibility and resistivity fields of unknown mining areas based on rock physics empirical formulas and potential field forward modeling; The voxel segmenter segments the three-dimensional ideal geological background field into multiple ideal voxels.

[0006] In a preferred embodiment, the steps of constructing a three-dimensional realistic geological field are: Obtain exploration data of unknown mining areas and use the exploration data to generate models based on the real background field to construct a three-dimensional real geological field; The realistic background field generation model includes a data converter, an anomaly converter, and a voxel segmenter; The data converter converts discrete exploration data into 3D raster data with the same resolution as the 3D ideal geological background field based on the Kriging interpolation algorithm; The anomaly converter extracts effective anomalies that conform to geological laws through statistical threshold analysis and morphological processing; The voxel segmenter segments the 3D realistic geological field into multiple realistic voxels.

[0007] In a preferred embodiment, the step of comparing the actual geological background field with the three-dimensional ideal geological background field to obtain first difference distribution data is: Perform voxel registration between the real geological background field and the three-dimensional ideal geological background field; Calculate the voxel difference between the corresponding ideal voxels and the actual voxels. The voxel difference includes the lithology classification difference, geochemical relative deviation and geophysical absolute difference. Based on morphological opening operation and connected domain analysis, the real voxels with differences are connected into effective abnormal areas as the first difference distribution data.

[0008] In a preferred embodiment, the step of extracting geological characteristics of a known mineralized area and generating interference factors based on the geological characteristics is as follows: Based on the known geological data and background data of the existing mining area, a mining area geological knowledge map is constructed. The mining area geological knowledge map includes element nodes and element-related edges; Extract geological features of geological data based on feature extraction algorithms; Unified vectorization of feature nodes and geological features and projection into matching space; In the matching space, the matching element node with the highest matching degree is obtained based on the dual-tower attention matching model; Multiple interference factors are constructed based on the matching feature nodes. The interference factors include a voxel range constraint, an attribute assigner, and an attribute modifier.

[0009] In a preferred embodiment, the step of introducing the interference factor into the ideal geological background field to generate the interference coupling background field is: The action voxel range constraint converts the action ranges defined by multiple interference factors into voxel index coordinates of a three-dimensional ideal geological background field to obtain the action target voxel; The attribute allocator generates the target interference factor by injecting the corresponding attribute into the corresponding interference factor by analyzing the lithology code, structural position and physical property threshold of the target voxel; The target interference factor is placed into the target voxel, and the corresponding attribute stored in the target interference factor is released based on the attribute modifier; Modify the properties of the target voxel based on the corresponding properties to obtain the updated voxel; Check the attribute deviation between the updated voxel and its adjacent voxels, and define two voxels whose attribute deviation exceeds a preset safety threshold as conflicting voxels; Based on the conflict resolution mechanism, the properties of the conflicting voxels are modified until all target voxels are updated to generate an interference coupling background field.

[0010] In a preferred embodiment, the step of extracting the difference between the interference coupling background field and the ideal geological background field as the second difference distribution data is: Segmenting the interference coupling background field into a plurality of interference coupling voxels; Calculate voxel differences between the corresponding ideal voxels and the interference coupling voxels. The voxel differences include lithologic classification differences, geochemical relative deviations, and geophysical absolute differences. Based on morphological opening operation and connected domain analysis, the interfering coupling voxels with differences are connected into effective abnormal areas as the second difference distribution data.

[0011] In a preferred embodiment, the step of constructing a mineral resource prediction model based on the first difference distribution data and the second difference distribution data is: Performing voxel-level spatial coupling analysis on the first difference distribution data and the second difference distribution data to calculate the overlap rate and difference intensity of the overlapping abnormal area; Based on the overlap rate and difference intensity, the overlapping abnormal area is divided into three levels of A / B / C according to the preset threshold; generating a three-dimensional mineralization probability field by weighted fusion of the overlapping abnormal areas of the first difference distribution data and the second difference distribution data; Convert the three-dimensional mineralization probability field into voxel index coordinates of the three-dimensional ideal geological background field, and generate a mineral resource prediction model based on the mapping of the voxel index coordinates to the three-dimensional ideal geological background field; The mineral resource prediction model includes mineral distribution, mineral type identification, mineral probability value and confidence level.

[0012] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention achieves cross-validation of mineral prediction under complex geological backgrounds by constructing a dual difference comparison between an ideal geological background field and a real geological background field, combined with the theoretical response model of the interference coupling field. The first difference analysis comprehensively captures potential mineralization signals by comparing the spatial anomaly distribution of the real field and the ideal field; the second difference analysis verifies the geological rationality of the anomaly based on the theoretical interference model. This dual verification mechanism can effectively distinguish mineralization-related anomalies from non-mineralized geological noise, and adopts a voxel-level spatial coupling algorithm, which is suitable for the identification and evaluation of weak anomalies under complex geological backgrounds. It can effectively overcome the multi-solution problem commonly found in traditional single anomaly analysis methods and significantly improve the reliability of prediction results. 2. The present invention realizes a closed loop of knowledge conversion from geological features to prediction models by constructing a knowledge graph-driven interference factor system. It uses a multi-scale feature extraction algorithm to automatically identify key mineral-controlling factors, and realizes intelligent matching with unknown mining areas through a matching algorithm. Based on clearly defined voxel ranges, spatial constraints, propagation mechanisms, and parameter modification rules, interference factors can perform differential simulations for different geological actions and mineralization actions. Through the interference factor mechanism, data obtained by different exploration methods and expert geological knowledge can be deeply integrated to form a parameterized and spatialized interference model. This knowledge-driven method breaks through the limitations of traditional empirical models, can adapt to the prediction needs of different geological backgrounds, and significantly improves the intelligence level and application flexibility of the mineral resource prediction system. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0014] Figure 1 Flow chart of the method of the present invention.

[0015] Figure 2 It is a logic block diagram of the present invention. DETAILED DESCRIPTION

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 and Figure 2 As shown, the method for constructing a mineral resource prediction model based on a complex geological background described in this embodiment includes: S1. Construct a 3D ideal geological background field without mineralization interference based on the geological background of the unknown mining area. The 3D ideal geological background field is used to display the mining area without mineralization. S2. constructing a three-dimensional realistic geological field based on the exploration data, and comparing the three-dimensional realistic geological field with the three-dimensional ideal geological background field to obtain first difference distribution data; S3, extracting geological characteristics of known mineralized areas and generating interference factors based on the geological characteristics; S4, introducing the interference factor into the ideal geological background field to generate an interference coupling background field, and extracting the difference between the interference coupling background field and the ideal geological background field as second difference distribution data; S5. Constructing a mineral resource prediction model based on the first difference distribution data and the second difference distribution data; As described in steps S1-S5 above, the current mainstream technical routes in the field of mineral resource prediction mainly include data-driven methods based on statistics and expert experience models based on geological concepts. Data-driven methods such as weight of evidence method, random forest, support vector machine and other machine learning algorithms achieve spatial distribution prediction of mineral resources through statistical analysis and feature mining of large amounts of exploration data and known mineral deposit information; Existing technologies perform well in areas with sufficient data and clear mining types, but they generally rely on historical data and statistical correlations, making it difficult to fully express the dynamic evolution of complex geological processes. They are also less adaptable to unknown or rare mineralization types and have limited interpretability. Meanwhile, expert knowledge models based on geological concepts rely on experts' deep understanding of mineral deposit types and geological ore-controlling factors, and use methods such as genetic models and mineral deposit models for prospecting and prediction. These methods are widely used in areas with mature understanding of mineralization geology, but are highly subjective. The modeling process relies on expert experience and qualitative analysis, making it difficult to systematically integrate complex multi-source data. Furthermore, the models lack generalizability and scalability, making it difficult to meet the needs of refined predictions in complex geological settings such as deep and concealed deposits. The present invention achieves cross-validation of mineral prediction under complex geological backgrounds by constructing a dual difference comparison between an ideal geological background field and a real geological background field, combined with a theoretical response model of an interference coupling field. The first difference analysis comprehensively captures potential mineralization signals by comparing the spatial anomaly distribution of the real field and the ideal field. The second difference analysis verifies the geological rationality of the anomaly based on the theoretical interference model. This dual verification mechanism can effectively distinguish between mineralization-related anomalies and non-mineralized geological noise, and adopts a voxel-level spatial coupling algorithm, which is suitable for the identification and evaluation of weak anomalies under complex geological backgrounds. It can effectively overcome the multi-solution problem commonly found in traditional single anomaly analysis methods and significantly improve the reliability of prediction results. By constructing a knowledge graph-driven interference factor system, a closed loop of knowledge conversion from geological features to prediction models is achieved. A multi-scale feature extraction algorithm is used to automatically identify key mineral-controlling factors, and intelligent matching with unknown mining areas is achieved through a matching algorithm. Based on clearly defined voxel ranges, spatial constraints, propagation mechanisms, and parameter modification rules, interference factors can perform differential simulations for different geological actions and mineralization actions. Through the interference factor mechanism, data obtained from different exploration methods and expert geological knowledge can be deeply integrated to form a parameterized and spatialized interference model. This knowledge-driven approach breaks through the limitations of traditional empirical models, can adapt to the prediction needs of different geological backgrounds, and significantly improves the intelligence level and application flexibility of the mineral resource prediction system.

[0018] In one embodiment, the step S1 of constructing a three-dimensional ideal geological background field without mineralization interference includes: S11, collecting geological background data of unknown mining areas, and constructing a three-dimensional ideal geological background field based on an ideal background field generation model; S12, ideal background field generation model including lithology distribution simulator, structural grid generator, physical property field calculator and voxel segmenter; S13, the lithology distribution simulator is based on the Markov random field algorithm and uses geological background data to constrain the lithology transition probability; S14, the structural grid generator simulates the structural generation environment based on the stress field fitting of the B-spline surface and generates a fracture network that conforms to the laws of mechanics; S15, physical property field calculator calculates density, magnetic susceptibility and resistivity fields of unknown mining areas based on rock physics empirical formulas and potential field forward modeling; S16, a voxel segmenter segments the three-dimensional ideal geological background field into a plurality of ideal voxels; As described in the above steps S11-S16, basic geological data of the unknown mining area are collected, including regional stratigraphic histograms, standard section lithologic sequences, regional structural outline maps, and a physical property parameter library. Based on this, an ideal background field generation model was invoked. This model, based on a geostatistical framework, defines the three-dimensional spatial extent through boundary conditions (such as basin outlines and unconformities) and initializes the grid resolution. The ideal background field generation model includes a lithologic distribution simulator, a structural grid generator, a physical property field calculator, and a voxel segmenter. The lithologic distribution simulator employs a Markov random field algorithm, constructing a conditional probability field based on prior knowledge from drill core data and surface outcrop lithologic maps. This algorithm simulates the spatial correlation of lithologic units. For example, when sandstone is present in an area, the probability that it is shale itself increases significantly. Through iterative calculations, a three-dimensional lithologic skeleton model conforming to geological laws (such as sedimentary phase transitions and intrusive contact zones) is generated. The structural grid generator is used for stress field reconstruction and fracture network generation. Stress field reconstruction uses the input of regional principal stress directions and strength parameters and B-spline surface technology to generate a smooth stress distribution field, simulating the stress state of the crust. Fracture network generation automatically generates fracture zones in stress concentration areas based on the Coulomb failure criterion (rocks fracture under specific compressive and shear stresses). The algorithm ensures that the fracture morphology conforms to the principles of geomechanics (such as the law of normal fault dip angle), and adds secondary cracks through fractal algorithm to enhance authenticity. The physical property field calculator includes a rock physics conversion module and a geophysical forward modeling module. The rock physics conversion module matches the empirical formula according to the lithology type (for example, density is calculated by seismic wave velocity, and resistivity is estimated by porosity). The geophysical forward modeling module is used to generate the density field and use the gravity forward modeling algorithm to infer the distribution of gravity anomalies; magnetic anomalies are simulated based on the magnetic susceptibility field, and multi-field data are mutually verified to ensure the self-consistency of physical property parameters. The voxel segmenter discretizes the continuous three-dimensional geological field into a regular cubic grid. Each voxel stores attribute labels including but not limited to lithology code, density value, magnetic susceptibility value, and resistivity. rate values ​​and structural markers (such as whether it is located in a fault zone). Adaptive grid technology can be used to optimize storage efficiency. The four core components work together to generate a three-dimensional ideal geological background field. The lithology simulator generates the original lithology framework. The structural grid generator adds a fracture / fold system (cuts and transforms the lithology unit). The physical property field calculator allocates physical property parameters according to the lithology-structure association rules (for example, the rock density in the fault zone is reduced). The voxel segmenter outputs a structured grid model. Any mineralization and alteration related parameter input (such as mineralization element content, alteration mineral combination) is disabled throughout the process. Only the regional background geological characteristics are retained. The final output three-dimensional model is the ideal geological background field without mineralization interference, providing a pure background for subsequent mineralization anomaly identification.

[0019] In one embodiment, the step S2 of constructing a three-dimensional realistic geological field includes: S21. Acquire exploration data of unknown mining areas, and use the exploration data to generate a model based on the real background field to construct a three-dimensional real geological field; S22, the realistic background field generation model includes a data converter, an anomaly converter, and a voxel segmenter; S23, the data converter converts the discrete exploration data into three-dimensional raster data with the same resolution as the three-dimensional ideal geological background field based on the Kriging interpolation algorithm; S24, anomaly converter extracts effective anomalies that conform to geological laws through statistical threshold analysis and morphological processing; S25, a voxel segmenter segments the three-dimensional real geological field into a plurality of real voxels; As described in the above steps S21-S25, multi-source exploration data of unknown mining areas are obtained, including but not limited to drill cores, geochemical sampling points, geophysical survey lines and remote sensing interpretation results, and input into a customized real background field generation model. The model adopts a modular pipeline design to ensure the full process conversion of original data to three-dimensional real geological fields. The real background field generation model contains three core processors, namely a data converter for solving the problem of spatial continuity of discrete data, an anomaly converter for extracting abnormal signals with clear geological significance, and a voxel segmenter for constructing computable three-dimensional grid voxels. The data converter is implemented by using the Kriging algorithm framework to automatically optimize the interpolation parameters according to the spatial distribution characteristics of the sampling points, and dynamically adjust the influence weights in different directions by analyzing the spatial autocorrelation of the exploration points. Geological boundary constraints (such as fault lines and lithologic contact zones) are introduced to prevent Prevent erroneous interpolation across geological units and output a 3D raster data volume that is fully aligned with the ideal geological background field. The resolution is forced to match the voxel size parameter. The anomaly converter processing logic includes a multi-level anomaly filtering mechanism. Statistical threshold stratification is used to independently calculate the background distribution for each geophysical / chemical parameter. The cumulative frequency curve inflection point method is used to determine the lower limit of the anomaly and distinguish between the background field and the anomaly field. Morphological optimization is performed by first performing an erosion operation to eliminate isolated noise points, and then performing an expansion operation to connect potential anomaly areas in spatial proximity. Geological law verification is carried out by integrating structural control models (such as the distribution of anomaly zones along fault strikes) and cross-validating the spatial coupling of multi-parameter anomalies (such as the overlap of high resistivity and Cu element anomalies). The voxel segmenter performs attribute inheritance segmentation, cutting the 3D raster data into a preset voxel size (such as 50m×50m×20m). Each real voxel inherits the multi-dimensional attribute label of the original raster.

[0020] In one embodiment, the step S2 of comparing the actual geological background field with the three-dimensional ideal geological background field to obtain first difference distribution data includes: S26, performing voxel registration of the actual geological background field and the three-dimensional ideal geological background field; S27, calculating voxel differences between the corresponding ideal voxels and the actual voxels, where the voxel differences include lithologic classification differences, geochemical relative deviations, and geophysical absolute differences; S28. Connecting the actual voxels with differences into valid abnormal regions based on morphological opening operation and connected domain analysis as first difference distribution data; As described in the above steps S26-S28, to ensure that the voxel spatial positions of the actual geological background field and the three-dimensional ideal geological background field are strictly aligned, it is necessary to force the unified use of the mining area's geodetic coordinate system and elevation benchmark, and assign globally unique three-dimensional index coordinates to all voxels. Furthermore, in the alignment process, structural displacement compensation and key layer calibration can be introduced. Among them, structural displacement compensation identifies the fault dislocation caused by tectonic movement in the actual field, establishes local displacement field models on both sides of the fault, and dynamically adjusts the voxel position to achieve structural alignment. Key layer calibration is to select landmark strata (such as those containing chemical elements). The thin plate spline interpolation algorithm is used to make the actual geological background field and the actual geological background field overlap in space. Differentiation processing is used to calculate the difference types, including lithologic classification differences, geochemical relative deviations, and geophysical absolute differences. Lithologic classification differences can be compared using semantic coding: when the main lithologic category is inconsistent, it is marked as 1, when the subcategory is inconsistent, it is marked as 0.5, and when it is completely consistent, it is marked as 0. This is used to identify lithologic variations caused by alteration and mineralization. Geochemical relative deviations are calculated by element by element to calculate the difference between the element values ​​in the actual geological background field and the three-dimensional ideal geological background field. Divide by the element standard deviation of the three-dimensional ideal geological background field to capture the enrichment intensity of ore-forming elements, geophysical absolute difference, take field type processing, take the absolute value difference by magnetic method, and retain the sign of gravity field to locate mineralization-related physical property anomalies; further, start independent calculation threads for cross-fault voxels to avoid the transmission of structural boundary errors; after identifying the real voxels with differences, use morphological opening operation to optimize the difference voxel space formed by the real voxels with differences, use spherical structure elements to scan the difference voxel space, and delete all isolated abnormal points that are not completely covered by the structure elements (such as <8 The connected voxel clusters are connected), and anisotropic ellipsoidal structural elements (with the major axis along the main structural direction of the region) are used to directionally expand the retained abnormal clusters and repair the abnormal interruptions caused by the lack of exploration data; the connected domain analysis uses a three-dimensional region growing algorithm to optimize the connectivity of each difference voxel space to make its boundary smoother, and finally generate effective abnormal areas as the first differential distribution data through geological rule constraints (including but not limited to deleting areas with an overlap of more than 70% with known non-mining interference sources, shielding geochemical false anomalies distributed along modern rivers, and retaining only abnormal areas within 2km of the ore-controlling structure).

[0021] In one embodiment, the step S3 of extracting geological features of a known mineralized area and generating interference factors based on the geological features includes: S31. Constructing a mining area geological knowledge map based on known geological data and known background data of the existing mining area, wherein the mining area geological knowledge map includes element nodes and element-related edges; S32, extracting geological features of the geological data based on a feature extraction algorithm; S33, uniformly vectorizing the element nodes and geological features and projecting them into the matching space; S34. In the matching space, the matching element node with the highest matching degree is obtained based on the dual-tower attention matching model; S35, constructing multiple interference factors based on the matching element nodes, the interference factors including an action voxel range constraint, an attribute assigner, and an attribute modifier; As described in the above steps S31-S35, a mining area geological knowledge map is constructed based on the existing mining area geological and background data, which includes feature nodes representing geological elements such as ore bodies, alteration zones, and ore-controlling structures, and feature-related edges describing the causes, space, and attribute associations; a feature extraction algorithm is used to extract multi-scale geological features from the exploration data, and the feature extraction algorithm includes using a convolutional neural network to process geophysical raster data and a graph neural network to analyze structural topological relationships; the feature node semantics are vectorized through word embedding technology, and the feature nodes and feature-related edges are converted into relationship vectors in combination with graph embedding technology. At the same time, an autoencoder is used to compress the geological features into feature vectors and project them into a unified matching space; a dual-tower attention matching model is deployed in this space The left tower inputs a feature node vector, and the right tower inputs a geological feature vector. The interaction weights between the features and the nodes are calculated through a multi-layer attention mechanism, and the feature node with the highest matching degree (such as the skarn-type copper mineralization system node) is finally output. The interference factor is constructed based on the matching nodes: the voxel range constraint generates an ellipsoid scope based on the spatial pattern of node association (such as the mineralization zoning radius), the attribute allocator parses the node attribute modification rules (such as reducing the resistivity of the marble contact zone voxels to 50-100Ω·m) to generate a sequence of operation instructions carrying parameters, and the attribute modifier encapsulates the atomic operation functions that execute the instructions (such as gradual interpolation of physical property values ​​and replacement of lithology codes). The three work together to form a structured interference factor that can be implanted in the geological background field.

[0022] In one embodiment, the step S4 of introducing the interference factor into the ideal geological background field to generate the interference coupling background field includes: S41, the action voxel range constraint converts the action ranges defined by the multiple interference factors into voxel index coordinates of the three-dimensional ideal geological background field to obtain the action target voxel; S42, the attribute allocator generates a target interference factor by injecting corresponding attributes into the corresponding interference factor by analyzing the lithology code, structural location and physical property threshold of the target voxel; S43, placing the target interference factor into the target voxel, and releasing the corresponding attribute stored in the target interference factor based on the attribute modifier; S44, modifying the attribute of the target voxel based on the corresponding attribute to obtain an updated voxel; S45, checking the attribute deviation between the updated voxel and the adjacent voxels, and defining two voxels whose attribute deviation exceeds a preset safety threshold as conflicting voxels; S46, modifying the attributes of the conflicting voxels based on the conflict resolution mechanism until all target voxels are updated to generate an interference coupling background field; As described in the above steps S41-S46, the fusion of the interference factor and the ideal geological background field is achieved through a fifth-order coupling process. First, the voxel range constraint is used to convert the three-dimensional scope of the interference factor (such as the fault buffer zone and the mineralization halo) into a voxel index coordinate. Furthermore, if there are multiple overlapping scopes of interference factors, a distance weighted algorithm can be used to generate a composite scope. The distance to the overlapping area of ​​the scope is calculated based on the central voxel of the scope, and the modification weight of the central voxel on the composite scope is calculated according to the distance. Then, the attribute allocator allocates the interference factor based on the lithology code, structural position and physical property threshold of the target voxel. The corresponding attributes are as follows: an interference factor attribute library is set in the attribute allocator; for the address data of the target voxel, the attribute allocator dynamically extracts the matching alteration rules, element enrichment equations and physical property adjustment amounts from the interference factor attribute library to generate a target interference factor carrying geological process parameters; then, the attribute modifier injects the target interference factor into the corresponding voxel: performs lithologic alteration conversion (for example, retains the original rock structure and superimposes 15% pyrite), reconstructs the geochemical field according to the equation, and adjusts the physical property parameters step by step; after the target interference factor is modified for the target voxel, a 26-neighborhood voxel verification unit is established to detect the updated voxel and the surrounding The voxel attribute mutation, the inspection dimensions include but are not limited to lithologic compatibility, physical property gradient and element zoning, the inspection indicators include but are not limited to coding difference, density change rate and element ratio gradient, if the inspection indicator exceeds the preset threshold, it is determined to be a conflict voxel, and the attributes of the conflict voxel are revised through the conflict resolution mechanism, and the conflict resolution mechanism includes: for lithologic mutation, inserting transition lithologic voxels in the contact zone (such as generating chlorite zone between andesite and sericite), for physical property jump, starting the thermal diffusion model smooth transition, triggering the smoothing filter to recalculate the physical property distribution, for element inverse zoning, reverse iteratively correcting the migration equation, adjusting the activity The parameters are adjusted until all conflicts are resolved to generate a geologically reasonable interference coupling background field. Furthermore, the boundaries of the updated voxels in the interference coupling background field can be smoothed by Gaussian. Based on the above analysis, this method realizes the triple precise coupling of space, attribute and genesis: the positioning accuracy is ensured by the spatial constraint transformation of the action voxels; the geological process adaptation is achieved based on the dynamic attribute injection mechanism of lithology, structure and physical property linkage; the multi-parameter neighborhood verification and genesis-driven conflict resolution mechanism are adopted to ensure the three-dimensional geological rationality of the updated voxels and the background field, and finally a quantifiable and verifiable interference coupling background field is generated, providing a high-fidelity simulation basis for mineral resource prediction.

[0023] In one embodiment, the step S4 of extracting the difference between the interference coupling background field and the ideal geological background field as the second difference distribution data includes: S47, dividing the interference coupling background field into a plurality of interference coupling voxels; S48, calculating voxel differences between the corresponding ideal voxels and the interference coupling voxels, where the voxel differences include lithologic classification differences, geochemical relative deviations, and geophysical absolute differences; S49, connecting the interfering coupling voxels with differences into effective abnormal regions as second difference distribution data based on morphological opening operation and connected domain analysis; As described in steps S47-S49 above, the interfering coupling background field is divided into interfering coupling voxels that are completely aligned with the spatial index of the three-dimensional ideal geological background field by a voxel splitter; the registered voxels are subjected to difference analysis using the same parallel architecture as the first difference calculation: the lithology classification difference is based on the comparison between the modified lithology code and the ideal code, the geochemical relative deviation is calculated using the (coupled field value - ideal value) / ideal background standard deviation model, and the geophysical absolute difference is processed for the magnetic / gravity / resistivity fields respectively (e.g., the absolute value difference is taken for the magnetic method, and a positive difference in the gravity field indicates a high-density anomaly); then Start the anomaly optimization process guided by geological laws - first set the spherical structural element with the minimum volume of the mineralized domain as the diameter to perform an opening operation (erosion eliminates isolated noise points and then directional expansion repairs the anomaly boundary), then perform intelligent aggregation of anomaly areas based on 3D connected domain analysis, and use a regional growth algorithm constrained by stratigraphic dip to force the merging of adjacent anomalies with a spacing less than 2 times the voxel size. At the same time, activate the simulated anomaly exclusive filter to eliminate anomaly clusters with a spatial matching degree of less than 60% with the known mineralization pattern, strengthen the identification of alteration zoning ring structures, and finally output the effective anomaly area as the second difference distribution data.

[0024] In one embodiment, the step S5 of constructing a mineral resource prediction model based on the first difference distribution data and the second difference distribution data includes: S51, performing voxel-level spatial coupling analysis on the first difference distribution data and the second difference distribution data, and calculating the overlap rate and difference intensity of the overlapping abnormal area; S52, dividing the overlapping abnormal area into three levels of A / B / C according to the preset threshold value based on the overlap rate and the difference intensity; S53, generating a three-dimensional mineralization probability field by weighted fusion of the overlapping abnormal areas of the first difference distribution data and the second difference distribution data; S54, converting the three-dimensional mineralization probability field into voxel index coordinates of the three-dimensional ideal geological background field, and generating a mineral resource prediction model based on mapping the voxel index coordinates into the three-dimensional ideal geological background field; S55. The mineral resource prediction model includes mineral distribution, mineral species identification, mineral probability value and confidence level; As described in steps S51-S55 above, voxel-level spatial coupling analysis is performed on the first difference distribution data and the second difference distribution data. The overlapping abnormal areas are identified through three-dimensional spatial Boolean operations, and their overlap rate is calculated. The overlap rate is: number of overlapping voxels / total number of abnormal voxels × 100%. The difference intensity is calculated by taking the geometric mean of lithologic, geophysical, and geochemical differences. The overlapping areas are divided into three levels based on preset thresholds: Level A (for example, overlap rate ≥ 70% and difference intensity > 85 percentile), Level B (for example, overlap rate 40%-70% and intensity > 60 percentile), and Level C (for example, overlap rate < 40% or insufficient intensity). A three-dimensional mineralization probability field is generated through dynamic weighted fusion: the first difference data is given a weight of 70% in the Level A area (to enhance the contribution of actual anomalies), and the Level B / C areas are given a weight of 5 0% / 30% weighting (focusing on simulating anomaly guidance), probability attenuation is performed on non-overlapping areas (the probability value decreases by 20% for every 100m increase from the boundary); the probability field voxel index coordinates are strictly matched with the three-dimensional ideal geological background field, and bilinear interpolation is used to map them to the background field grid nodes; the final output mineral resource prediction model includes: 1. Mineral distribution triangulation entity (voxels with probability > 0.6 are aggregated into closed surfaces), 2. Mineral type identification (automatically labeled according to the source knowledge graph node of the matching interference factor), 3. Probability distribution (0-1 continuous distribution), 4. Confidence (confidence in Class A area = 0.9×data coverage coefficient, Class C area = 0.5×simulation matching degree); further, the mineral resource prediction model supports visualization and reserve estimation interface calls based on probability threshold levels.

[0025] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for constructing a mineral resource prediction model based on a complex geological background, characterized in that: Based on the geological background of the unknown mining area, a three-dimensional ideal geological background field without mineralization interference is constructed. The three-dimensional ideal geological background field is used to display the mining area without mineralization; constructing a three-dimensional realistic geological field based on the exploration data, and comparing the three-dimensional realistic geological field with the three-dimensional ideal geological background field to obtain first difference distribution data; Extract the geological characteristics of known mineralized areas and generate interference factors based on the geological characteristics; introducing interference factors into the ideal geological background field to generate an interference coupling background field, and extracting the difference between the interference coupling background field and the ideal geological background field as second difference distribution data; A mineral resource prediction model is constructed based on the first difference distribution data and the second difference distribution data.

2. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The steps of constructing a three-dimensional ideal geological background field without mineralization interference are: Collect geological background data of unknown mining areas and construct a three-dimensional ideal geological background field based on the ideal background field generation model; The ideal background field generation model includes a lithology distribution simulator, a structural grid generator, a physical property field calculator, and a voxel segmenter; The lithology distribution simulator is based on the Markov random field algorithm and uses geological background data to constrain the lithology transition probability; The structural grid generator simulates the structural generation environment based on the stress field fitted by B-spline surface and generates a fracture network that conforms to the laws of mechanics. The physical property calculator calculates the density, magnetic susceptibility and resistivity fields of unknown mining areas based on rock physics empirical formulas and potential field forward modeling; The voxel segmenter segments the three-dimensional ideal geological background field into multiple ideal voxels.

3. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The steps of constructing a three-dimensional realistic geological field based on exploration data are: Obtain exploration data of unknown mining areas and use the exploration data to generate models based on the real background field to construct a three-dimensional real geological field; The realistic background field generation model includes a data converter, an anomaly converter, and a voxel segmenter; The data converter converts discrete exploration data into 3D raster data with the same resolution as the 3D ideal geological background field based on the Kriging interpolation algorithm; The anomaly converter extracts effective anomalies that conform to geological laws through statistical threshold analysis and morphological processing; The voxel segmenter segments the 3D realistic geological field into multiple realistic voxels.

4. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The step of comparing the three-dimensional actual geological field with the three-dimensional ideal geological background field to obtain first difference distribution data is: Perform voxel registration between the real geological background field and the three-dimensional ideal geological background field; Calculate the voxel difference between the corresponding ideal voxels and the actual voxels. The voxel difference includes the lithology classification difference, geochemical relative deviation and geophysical absolute difference. Based on morphological opening operation and connected domain analysis, the real voxels with differences are connected into effective abnormal areas as the first difference distribution data.

5. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The steps of extracting geological features of a known mineralized area and generating interference factors based on the geological features are as follows: Based on the known geological data and background data of the existing mining area, a mining area geological knowledge map is constructed. The mining area geological knowledge map includes element nodes and element-related edges; Extract geological features of geological data based on feature extraction algorithms; Unified vectorization of feature nodes and geological features and projection into matching space; In the matching space, the matching element node with the highest matching degree is obtained based on the dual-tower attention matching model; Multiple interference factors are constructed based on the matching feature nodes. The interference factors include a voxel range constraint, an attribute assigner, and an attribute modifier.

6. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 5, characterized in that: The steps of introducing the interference factor into the ideal geological background field to generate the interference coupling background field are: The action voxel range constraint converts the action ranges defined by multiple interference factors into voxel index coordinates of a three-dimensional ideal geological background field to obtain the action target voxel; The attribute allocator generates the target interference factor by injecting the corresponding attribute into the corresponding interference factor by analyzing the lithology code, structural position and physical property threshold of the target voxel; The target interference factor is placed into the target voxel, and the corresponding attribute stored in the target interference factor is released based on the attribute modifier; Modify the properties of the target voxel based on the corresponding properties to obtain the updated voxel; Check the attribute deviation between the updated voxel and its adjacent voxels, and define two voxels whose attribute deviation exceeds a preset safety threshold as conflicting voxels; Based on the conflict resolution mechanism, the properties of the conflicting voxels are modified until all the target voxels are updated to generate an interference coupling background field.

7. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The step of extracting the difference between the interference coupling background field and the ideal geological background field as the second difference distribution data is: Segmenting the interference coupling background field into a plurality of interference coupling voxels; Calculate voxel differences between the corresponding ideal voxels and the interference coupling voxels. The voxel differences include lithologic classification differences, geochemical relative deviations, and geophysical absolute differences. Based on morphological opening operation and connected domain analysis, the interfering coupling voxels with differences are connected into effective abnormal areas as the second difference distribution data.

8. The method for constructing a mineral resource prediction model based on a complex geological background according to claim 1, characterized in that: The step of constructing a mineral resource prediction model based on the first difference distribution data and the second difference distribution data is: Performing voxel-level spatial coupling analysis on the first difference distribution data and the second difference distribution data to calculate the overlap rate and difference intensity of the overlapping abnormal area; Based on the overlap rate and difference intensity, the overlapping abnormal area is divided into three levels of A / B / C according to the preset threshold; generating a three-dimensional mineralization probability field by weighted fusion of the overlapping abnormal areas of the first difference distribution data and the second difference distribution data; Convert the three-dimensional mineralization probability field into voxel index coordinates of the three-dimensional ideal geological background field, and generate a mineral resource prediction model based on the mapping of the voxel index coordinates to the three-dimensional ideal geological background field; The mineral resource prediction model includes mineral distribution, mineral type identification, mineral probability value and confidence level.

Images