Mineral resource exploration data analysis method and device based on three-dimensional space modeling

Abstract

The embodiment of the invention discloses a mineral resource exploration data analysis method and device based on three-dimensional space modeling. The method comprises the steps that back-flushing pushing-covering structure feature information is extracted through structure interface occurrence measurement, lithology combination analysis and structure movement trace recognition and converted into geologic structure feature parameters; stratigraphic age data and lithology distribution information are integrated, and a three-dimensional stratigraphic model is constructed with a stratigraphic superposition relation as a constraint; geophysical data are imported to extract geophysical response characteristics of stratums with different depths, and lithologic boundary correction is combined to obtain deep fracture characteristic parameters; sedimentary system division is carried out on lithofacies paleogeographic information, and deep fracture characteristic parameters are fused to establish a fracture system three-dimensional model; the multi-feature association relationship is identified, the ore body prediction model is established, the spatial data is input to obtain the three-dimensional visual prediction result of the hidden ore body spatial distribution, and the exploration efficiency and accuracy in the deep phosphorite exploration process are improved.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of computer data analysis, in particular to a mineral resource exploration data analysis method and device based on three-dimensional space modeling. BACKGROUND

[0002] In the field of mineral resource exploration, the extraction of geological structure characteristics is an important basis for analyzing regional geological evolution and mineral resource distribution. As a kind of geological structure, the extraction of characteristic information of thrust nappe structure is usually realized by means of structure interface occurrence measurement, lithological combination analysis and structure movement trace identification. The characteristic information is crucial for understanding the formation and evolution of regional geological structure. The construction of stratigraphic model is a key link in mineral resource exploration. In the prior art, stratigraphic age data and lithological distribution information are usually integrated based on geological structure characteristic parameters, and a stratigraphic model is constructed based on the constraint condition of stratigraphic superimposition relationship, which is used to reflect the spatial distribution of strata in the target exploration area. Geophysical data is widely used in mineral resource exploration. By processing and analyzing geophysical data, the geophysical response characteristics of different depth strata can be extracted, which can be used to infer deep geological structure and mineral resource distribution. The analysis of lithofacies paleogeographic information plays an important role in understanding the depositional environment and the formation of mineral resources. By dividing the sedimentary system based on lithofacies paleogeographic information, the spatial correlation between faults and sedimentary facies belts is analyzed based on fault characteristic parameters, and a fault system model is established. The establishment of ore body prediction model is one of the core goals of mineral resource exploration. Based on geological structure, strata, geophysical and lithofacies paleogeographic data, the correlation between multiple characteristics is identified, and an ore body prediction model is established to predict the spatial distribution of concealed ore bodies.

[0003] However, the existing technology has problems such as unclear relationship between sedimentary microfacies and phosphorus enrichment mineralization, unclear spatial distribution rule of deep phosphorus mine, and no systematic establishment of deep phosphorus mine exploration technology, which makes it difficult to ensure the exploration efficiency and accuracy in the process of deep phosphorus mine exploration. SUMMARY

[0004] The embodiment of the present application provides a mineral resource exploration data analysis method and device based on three-dimensional space modeling.

[0005] The embodiment of the present application provides a mineral resource exploration data analysis method based on three-dimensional space modeling, which comprises: By means of structure interface occurrence measurement, lithological combination analysis and structure movement trace identification, the characteristic information of the thrust nappe structure of the target exploration area is extracted, and the characteristic information is converted into geological structure characteristic parameters; Integrate the stratum age data and the lithology distribution information of the target exploration area based on the geological structure feature parameter, to construct a three-dimensional stratum model reflecting the stratum spatial distribution of the target exploration area by taking the stratum superimposed relationship determined by the tectonic sequence division as a constraint condition; Based on the three-dimensional stratum model, import the geophysical data of the target exploration area, perform geophysical response feature extraction of strata at different depths on the geophysical data, and correct the extraction result in combination with the lithology boundary in the three-dimensional stratum model to obtain deep fault feature parameters; Perform sedimentary system division on the lithofacies paleogeographic information of the target exploration area to obtain a sedimentary system division result, integrate the deep fault feature parameters into the sedimentary system division result, and establish a fault system three-dimensional model reflecting the spatial distribution of faults and the relationship with deposition by spatial correlation analysis of faults and sedimentary facies belts; According to the three-dimensional stratum model and the fault system three-dimensional model, identify the correlation relationship among the lithology combination features corresponding to the ore body deposition environment, the tectonic ore-controlling regularity features corresponding to the geological structure feature parameter, and the physical property anomaly features corresponding to the geophysical data, establish an ore body prediction model by taking the correlation relationship as a constraint, input the spatial data of the target exploration area into the ore body prediction model, and obtain a three-dimensional visual prediction result for the spatial distribution of concealed ore bodies.

[0006] The embodiment of the present application also provides a computer device, comprising: a processor; a storage device having a computer program stored thereon, When the computer program is executed by the processor, the processor implements the three-dimensional space modeling-based mineral resource exploration data analysis method.

[0007] The embodiment of the present application also provides a readable storage medium having a program or instruction stored thereon, and the program or instruction is executed by a processor to implement the steps of the three-dimensional space modeling-based mineral resource exploration data analysis method.

[0008] Therefore, the embodiment of the present application has the following beneficial effects: the embodiment of the present application extracts and converts the feature information of the thrust nappe structure of the target exploration area into a geological structure feature parameter, integrates stratum age data and lithology distribution information to construct a three-dimensional stratum model, imports geophysical data to extract geophysical response features of strata at different depths and corrects the extraction result in combination with the lithology boundary to obtain deep fault feature parameters, performs sedimentary system division on lithofacies paleogeographic information and integrates deep fault feature parameters to establish a fault system three-dimensional model, and finally identifies multiple feature correlation relationships and establishes an ore body prediction model to obtain a three-dimensional visual prediction result for the spatial distribution of concealed ore bodies.

[0009] Thus, the multi-dimensional data fusion and analysis of geological structure, stratum spatial distribution, geophysical response, lithofacies paleogeography and ore body prediction are achieved as a whole, effectively solving the scientific problems of unclear sedimentary microfacies of phosphorus-bearing strata, unclear relationship between sedimentary microfacies and phosphorus enrichment mineralization, and unclear spatial distribution rule caused by large spatial distribution variation of deep phosphorus ore, and establishing deep phosphorus ore exploration technology. The fine reconstruction technology of lithofacies paleogeography based on big data, the multi-source data three-dimensional prediction technology of deep phosphorus ore in ore concentration area, the green and efficient ore prediction technology of the fusion of geology, geophysical prospecting, geochemical prospecting and remote sensing big data and artificial intelligence, and the spatial precise positioning technology of deep phosphorus ore are broken through, which provides comprehensive and accurate technical support for the exploration and prediction of deep phosphorus ore, and improves the efficiency and accuracy of deep phosphorus ore exploration. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 A flowchart of a mineral resource exploration data analysis method based on three-dimensional space modeling provided by an embodiment of the present application.

[0011] Figure 2 A schematic diagram of the basic structure of a computer device provided by an embodiment of the present application.

[0012] Figure 3 A function module block diagram of a mineral resource exploration data analysis device based on three-dimensional space modeling provided by an embodiment of the present application. DETAILED DESCRIPTION

[0013] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the embodiments of the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0014] Referring to Figure 1 As shown in the figure, the figure is a flowchart of a mineral resource exploration data analysis method based on three-dimensional space modeling provided by an embodiment of the present application, which is executed by a computer device. As shown in Figure 1 The method includes steps 110-150.

[0015] The embodiment of the present application is applied to the scene of mineral resource exploration data analysis based on three-dimensional space modeling, and the target exploration area is a certain phosphorus exploration area, aiming to realize accurate prediction of concealed phosphorus ore body through multi-dimensional geological data integration and analysis.

[0016] Firstly, the existing mineral resource exploration data analysis relies on single-dimensional data or experience-driven, while the embodiment of the application innovatively constructs a multi-dimensional technical system of "three-dimensional stratigraphic model-three-dimensional fracture system model-ore body prediction model", realizing the deep integration of multi-source data such as geological structure, stratum spatial distribution, geophysical response, lithofacies paleogeography. The traditional method of extracting the characteristics of the thrust nappe structure only stays at the qualitative description level, and the embodiment of the application converts the characteristics such as the occurrence of the structural interface, the lithological combination and the tectonic movement trace into quantitative parameters, and builds a three-dimensional stratigraphic model with the stratigraphic superimposed relationship of the tectonic sequence division as the constraint, solving the problem of quantitative characterization of stratum spatial distribution.

[0017] Secondly, in the aspect of deep fracture feature extraction, the existing technology often analyzes geophysical data separately, which is easily disturbed by lithological boundaries, leading to inaccurate anomaly identification. The embodiment of the application divides the geophysical data into subsets according to the depth gradient, corrects them in combination with the lithological boundaries of the three-dimensional stratigraphic model, and then extracts deep fracture feature parameters through signal feature decomposition and correlation comparison, realizing accurate identification of fracture characteristics. This process breaks through the technical bottleneck of confusing lithology and structure anomalies in traditional geophysical data processing.

[0018] Thirdly, in the construction of the three-dimensional fracture system model, the existing technology rarely considers the spatial correlation between fractures and sedimentary facies belts. The embodiment of the application integrates deep fracture feature parameters into the sedimentary system division, analyzes the spatial correlation rules between fractures and sedimentary facies belts, and establishes a three-dimensional model containing fracture spatial coordinates, extension form and sedimentary facies belt correlation attributes, revealing the control of fractures on sedimentary facies belts. This innovative idea provides a new technical approach for understanding the coupling relationship between structure and sedimentation.

[0019] Finally, in the construction of the ore body prediction model, the traditional method relies on single feature or simple superposition. The embodiment of the application identifies the correlation between lithological combination features, structure ore-controlling law features and physical property anomaly features, and constructs a prediction model containing feature input layer, correlation analysis layer and result output layer, realizing the collaborative analysis and quantitative calculation of multiple features, and solving the problem of multi-factor coupling analysis in concealed ore body prediction.

[0020] Step 110: Extracting feature information of the thrust nappe structure in the target exploration area by measuring the occurrence of the structural interface, analyzing the lithological combination and identifying the tectonic movement trace, and converting the feature information into geological structure feature parameters.

[0021] The computer device firstly calls geological survey system data of a target exploration area, measures an interface occurrence of a thrust nappe structure, and obtains strike, tendency and dip angle data of the interface; then performs lithology combination analysis on core sample data of the target exploration area, identifies a lithology combination mode of a phosphorus-bearing rock system, such as an interactive combination of carbonate rock and phosphate rock; then identifies structural movement traces, including fault scratches and fold axial surfaces; and then converts the above feature information into geological structure feature parameters, including interface occurrence parameters, lithology combination parameters and structural movement trace parameters, wherein the interface occurrence parameters include strike angle range, tendency angle range and dip angle range, the lithology combination parameters include thickness proportion and combination sequence of different lithologies, and the structural movement trace parameters include length range of fault scratches and morphological parameters of fold axial surfaces.

[0022] Step 120: integrating stratum age data and lithology distribution information of the target exploration area based on the geological structure feature parameters, to construct a three-dimensional stratum model reflecting spatial distribution of strata of the target exploration area as a constraint condition of a determined stratum superimposition relationship by structural sequence division.

[0023] The computer device firstly obtains stratum age data of the target exploration area, including absolute age range and relative age sequence of each era stratum, and then integrates lithology distribution information, including distribution range of different lithologies in a plane and in a vertical direction; then performs structural sequence division according to interface occurrence and structural movement trace parameters in the geological structure feature parameters, to determine a stratum superimposition relationship, such as integrated contact and unconformable contact; then constructs a three-dimensional stratum model as a constraint condition of the stratum superimposition relationship, using a three-dimensional modeling software, a spatial coordinate system of the model adopts a local coordinate system of the target exploration area, and the model includes spatial coordinate range, lithology attribute and age attribute of a stratum unit, wherein the spatial coordinate range of the stratum unit is represented by a three-dimensional coordinate point set, the lithology attribute includes lithology type and physical and chemical properties, and the age attribute includes absolute age and relative age of the stratum.

[0024] Step 130: importing geophysical data of the target exploration area based on the three-dimensional stratum model, performing geophysical response feature extraction of strata at different depths on the geophysical data, and combining lithology boundary correction of the three-dimensional stratum model to obtain deep fracture feature parameters.

[0025] The computer device imports geophysical data of a target exploration area first, including gravity data, magnetic data and seismic data; then delimits a processing interval of the geophysical data with reference to the spatial distribution range and lithological interface coordinates of the strata in the three-dimensional strata model; then performs geophysical response feature extraction of different depth strata on the geophysical data, and obtains response features of different depth strata, such as gravity anomaly features, magnetic anomaly features and seismic reflection features, through a signal processing algorithm; then corrects the extraction results in combination with the lithological boundaries in the three-dimensional strata model, and removes false anomalies caused by the lithological boundaries; finally, deep fracture feature parameters are obtained, which include the spatial position coordinates, extension direction, dip angle and fracture zone width of the fracture.

[0026] Step 131: delimiting a processing interval of the geophysical data with reference to the spatial distribution range and lithological interface coordinates of the strata in the three-dimensional strata model, and dividing the geophysical data into a plurality of data subsets according to depth gradients, each data subset corresponding to a stratum unit in the three-dimensional strata model.

[0027] In the method, the computer device extracts the spatial distribution range of the strata from the three-dimensional strata model first, including the maximum and minimum plane coordinates and the vertical depth range of the strata, and then extracts the lithological interface coordinates, including the three-dimensional coordinate point set of each lithological interface; then delimits the processing interval of the geophysical data according to the spatial distribution range and the lithological interface coordinates of the strata, the plane range of the processing interval being consistent with the plane distribution range of the strata, and the vertical range being consistent with the vertical depth range of the strata; then divides the geophysical data into a plurality of data subsets according to depth gradients, the depth gradients being determined according to the thickness of the strata, each data subset corresponding to a stratum unit in the three-dimensional strata model, such as dividing the vertical depth range into a plurality of depth segments, each depth segment corresponding to a stratum unit, and each data subset containing the geophysical data in the depth segment.

[0028] Step 132: performing geophysical response feature extraction on each data subset, and obtaining an initial response feature set containing amplitude features, phase features and continuity features through signal feature decomposition, and synchronously extracting the lithological boundary coordinates and lithological change attributes of the corresponding stratum unit in the three-dimensional strata model.

[0029] In the method, the computer device performs geophysical response feature extraction on each data subset by using a signal decomposition algorithm, the signal decomposition algorithm including Fourier transform, wavelet transform and the like, and obtains the initial response feature set through signal decomposition, wherein the amplitude features represent the intensity of the geophysical signal, the phase features represent the phase change of the signal, and the continuity features represent the continuity of the signal in space; and the lithological boundary coordinates of the corresponding stratum unit are extracted from the three-dimensional strata model, including the three-dimensional coordinate point set of the lithological boundary, and the lithological change attributes, including the conversion mode and conversion frequency of the lithological type.

[0030] Step 133: Spatially match the initial response feature set with the lithological boundary coordinates, determine whether there is an abnormal feature point in the initial response feature set based on the lithological change attribute, and if so, mark the spatial position corresponding to the abnormal feature point and analyze its spatial relationship with the lithological boundary.

[0031] Wherein, the computer device matches the initial response feature set with the lithological boundary coordinates by using a spatial matching algorithm, the spatial matching algorithm includes nearest neighbor matching, regional matching, etc.; then determines whether there is an abnormal feature point in the initial response feature set based on the lithological change attribute, such as sudden conversion of lithological type, abnormal change of lithological thickness, etc., the judgment basis of the abnormal feature point is that the feature value exceeds the normal range; if there is an abnormal feature point, mark the spatial position corresponding to the abnormal feature point, including three-dimensional coordinates, and then analyze its spatial relationship with the lithological boundary, such as the abnormal feature point being above, below or at the lithological boundary.

[0032] Step 134: For the marked abnormal feature point, backtrack to the corresponding geophysical data subset, re-execute signal feature decomposition and adjust the decomposition parameters to obtain a corrected response feature set, and associate and compare the corrected response feature set with the response feature set of the adjacent depth stratum unit.

[0033] Wherein, the computer device determines the geophysical data subset corresponding to the marked abnormal feature point; then re-executes signal feature decomposition and adjusts the decomposition parameters, such as the decomposition frequency range, wavelet basis function, etc., to obtain a corrected response feature set; then associate and compare the corrected response feature set with the response feature set of the adjacent depth stratum unit, the contents of association and comparison include the change trend of amplitude feature, the consistency of phase feature and the spatial extension of continuity feature.

[0034] Step 135: Extract the region with feature mutation in the association and comparison process, determine the spatial range of the region in combination with the lithological boundary coordinates, analyze the feature mutation law and spatial form of the region, and generate deep fracture feature parameters including mutation position, extension direction and feature change amplitude.

[0035] Wherein, the computer device extracts the region with feature mutation in the association and comparison process by using a feature mutation detection algorithm, the feature mutation detection algorithm includes gradient detection, difference detection, etc.; then determines the spatial range of the region in combination with the lithological boundary coordinates, including the planar range and the vertical depth range; then analyzes the feature mutation law of the region, such as mutation frequency, intensity change, etc., and the spatial form, such as linearity, arc shape, etc.; finally generates deep fracture feature parameters, the parameters include three-dimensional coordinates of mutation position, angle range of extension direction and numerical range of feature change amplitude.

[0036] Step 140: sedimentary system division is performed on the lithofacies paleogeographic information of the target exploration area to obtain a sedimentary system division result, the deep fault feature parameter is integrated into the sedimentary system division result, and a fault system three-dimensional model reflecting the spatial distribution of faults and the relationship with deposition is established through spatial correlation analysis of faults and sedimentary facies belts.

[0037] The computer device first acquires lithofacies paleogeographic information of a target exploration area, including sedimentary facies types, paleocurrent directions, and sedimentary environments; then performs sedimentary system division on the lithofacies paleogeographic information, the division basis including the combination of sedimentary facies types, the consistency of paleocurrent directions, and the similarity of sedimentary environments, to obtain a sedimentary system division result, which contains multiple sedimentary system units, each of which contains sedimentary facies types, distribution ranges, and sedimentary characteristics; then the deep fault feature parameter is integrated into the sedimentary system division result, including superimposing the spatial position of the fault on the distribution range of the sedimentary system unit; then the control of the fault on the sedimentary facies belt is analyzed through spatial correlation analysis of the fault and the sedimentary facies belt, such as the boundary limitation of the fault on the sedimentary facies belt, the consistency of the extension direction of the fault and the distribution direction of the sedimentary facies belt, etc.; finally, a fault system three-dimensional model is established, which contains the spatial distribution, extension direction, and correlation relationship with the sedimentary facies belt of the fault.

[0038] Step 141: collect lithofacies paleogeographic information covering sedimentary facies types, paleocurrent directions, and lithological facies change characteristics of the target exploration area, and perform sedimentary system division on the target exploration area based on the spatial distribution rule of the sedimentary facies types and the gradient change of the lithological facies change characteristics to obtain multiple initial sedimentary system units.

[0039] Among them, the computer device collects lithofacies paleogeographic information of the target exploration area, including sedimentary facies type data obtained through field geological investigation, such as shallow marine facies, coastal facies, and delta facies, paleocurrent direction data obtained through paleocurrent measurement, such as the flow direction angle range of the water flow, and lithological facies change characteristic data obtained through core observation, such as the conversion of lithological types and the change of thickness; then analyze the spatial distribution rule of the sedimentary facies types, such as the shallow marine facies mainly distributed in the middle of the target exploration area and the coastal facies distributed on the edge; then analyze the gradient change of the lithological facies change characteristics, such as the gradual change of the lithological types from carbonate rock to clastic rock and the gradual thinning of the thickness from the shallow marine facies to the coastal facies; then perform sedimentary system division on the target exploration area based on the spatial distribution rule of the sedimentary facies types and the gradient change of the lithological facies change characteristics, the division method including cluster analysis, regional division, etc., to obtain multiple initial sedimentary system units, each of which contains sedimentary facies types, distribution ranges, and lithological facies change characteristics.

[0040] Step 1411: Collect lithofacies paleogeographic information of the target exploration area, and classify the lithofacies paleogeographic information to determine sedimentary facies type, paleocurrent direction data and lithology facies change characteristic description.

[0041] Among them, the computer device collects the lithofacies paleogeographic information of the target exploration area, and the information sources include field geological report, core analysis data, logging data, etc.; then the lithofacies paleogeographic information is classified and sorted, and the information is divided into sedimentary facies type data, paleocurrent direction data and lithology facies change characteristic description, wherein the sedimentary facies type data includes the names and distribution positions of shallow marine facies, coastal facies and delta facies, the paleocurrent direction data includes the flow direction angle and measurement position of the water flow, and the lithology facies change characteristic description includes the conversion process of lithology type, the change of thickness, etc.

[0042] Step 1412: classify the sedimentary facies type, convert the paleocurrent direction data into spatial vector form, and quantitatively process the lithology facies change characteristic description to obtain lithology facies change gradient data.

[0043] Among them, the computer device classifies the sedimentary facies type, classifies similar sedimentary facies types into one category, determines the main sedimentary facies type, such as shallow marine facies, coastal facies and delta facies, and the typical lithology marks of each main sedimentary facies type include carbonate rock of shallow marine facies, clastic rock of coastal facies and sand-shale interbedding of delta facies; then the paleocurrent direction data is converted into spatial vector form, the direction of the spatial vector is consistent with the paleocurrent direction, and the length represents the intensity of the water flow; then the lithology facies change characteristic description is quantitatively processed, and the conversion frequency of lithology type and the change amplitude of thickness are converted into numerical form to obtain lithology facies change gradient data.

[0044] Step 1413: based on the plane coordinates of the target exploration area, construct a two-dimensional analysis grid, and map the classified sedimentary facies type, spatial vector form of paleocurrent direction and quantified lithology facies change gradient data to the corresponding nodes of the two-dimensional analysis grid, respectively.

[0045] Among them, the computer device sets the size of the grid based on the plane coordinates of the target exploration area, constructs a two-dimensional analysis grid, and the node coordinates of the grid correspond one-to-one to the plane coordinates of the target exploration area; then the classified sedimentary facies type is mapped to the corresponding nodes of the two-dimensional analysis grid, such as shallow marine facies mapped to the middle nodes of the grid and coastal facies mapped to the edge nodes; then the spatial vector form of the paleocurrent direction is mapped to the corresponding nodes, the starting point of the vector is the node coordinate, and the direction is the paleocurrent direction; then the quantified lithology facies change gradient data is mapped to the corresponding nodes, and the data value represents the lithology facies change gradient at the node.

[0046] Step 1414: Calculate the similarity of sedimentary facies types, the angle between paleocurrent direction vectors, and the gradient difference of lithofacies change of adjacent nodes in the two-dimensional analysis grid, set the similarity threshold, the angle threshold, and the difference threshold, and classify the adjacent nodes that meet the conditions of the similarity of sedimentary facies types being higher than the similarity threshold, the angle between paleocurrent direction vectors being smaller than the angle threshold, and the gradient difference of lithofacies change being smaller than the difference threshold into the same unit; wherein the similarity of sedimentary facies types is calculated based on type matching degree, the angle between paleocurrent direction vectors is calculated based on spatial vector angle measurement, and the gradient difference of lithofacies change is quantified based on lithofacies change intensity.

[0047] In the method, the computer device calculates the similarity of sedimentary facies types of adjacent nodes in the two-dimensional analysis grid, the calculation of the similarity is based on type matching degree, such as the similarity being 1 if the sedimentary facies types of two nodes are the same, and 0 if the sedimentary facies types of two nodes are different; then calculates the angle between paleocurrent direction vectors, the calculation of the angle is based on spatial vector angle measurement, such as the angle between paleocurrent direction vectors of two nodes being 30 degrees; then calculates the gradient difference of lithofacies change, the calculation of the difference is based on lithofacies change intensity quantification, such as the gradient difference of lithofacies change of two nodes being 0.5; then sets the similarity threshold as 0.8, the angle threshold as 45 degrees, and the difference threshold as 0.6; and finally classifies the adjacent nodes that meet the conditions of the similarity of sedimentary facies types being higher than 0.8, the angle between paleocurrent direction vectors being smaller than 45 degrees, and the gradient difference of lithofacies change being smaller than 0.6 into the same unit.

[0048] Step 1415: Traverse the entire two-dimensional analysis grid, repeat the node merging process, form initial sedimentary units on multiple planes, combine the stratigraphic thickness information and vertical lithofacies change characteristics in the three-dimensional stratigraphic model, extend the initial sedimentary units on the plane along the vertical direction to obtain multiple initial sedimentary system units containing vertical dimension information, and extract the vertical boundary coordinates, planar distribution range, and core characteristic parameters of each initial sedimentary system unit, wherein the core characteristic parameters include the dominant sedimentary facies type, the dominant paleocurrent direction, and the average lithofacies change gradient.

[0049] In the method, the computer device traverses the entire two-dimensional analysis grid, repeats the node merging process, classifies the adjacent nodes that meet the conditions into the same unit to form multiple initial sedimentary units on multiple planes; then extracts the stratigraphic thickness information and vertical lithofacies change characteristics from the three-dimensional stratigraphic model, wherein the stratigraphic thickness information includes the thickness of each stratigraphic unit, and the vertical lithofacies change characteristics include the vertical conversion sequence of lithofacies types; then extends the initial sedimentary units on the plane along the vertical direction to obtain multiple initial sedimentary system units containing vertical dimension information, wherein the thickness of the extension is determined according to the stratigraphic thickness information; and finally extracts the vertical boundary coordinates, planar distribution range, and core characteristic parameters of each initial sedimentary system unit, wherein the vertical boundary coordinates include the maximum and minimum vertical depths of the unit, the planar distribution range includes the maximum and minimum planar coordinates of the unit, and the core characteristic parameters include the dominant sedimentary facies type, the dominant paleocurrent direction, and the average lithofacies change gradient.

[0050] Wherein, the dominant paleocurrent direction is the dominant paleocurrent flow direction of the target exploration area in the target sedimentary period, and is determined by multi-source data statistics and spatial clustering, which is a key indicator reflecting the depositional environment, provenance supply and depositional system distribution. From the perspective of depositional environment reconstruction, the dominant paleocurrent direction can be used as a wind vane. A stable single direction may indicate a directional hydrodynamic environment such as fluvial facies and delta facies, a two-way alternation may correspond to a tidal facies, and a scattered but concentrated trend may reflect a shallow sea wave dominated environment. As an indicator of provenance location, the dominant paleocurrent direction usually points to the opposite direction of the provenance area, and combined with lithological assemblage, the provenance area can be accurately located. At the same time, the dominant paleocurrent direction is also the control axis of the depositional system distribution, which determines the extension direction and spatial distribution of the depositional facies belt, such as fluvial channel sand bodies and delta distributary channel.

[0051] Step 142: Extract the spatial boundary coordinates, core depositional characteristics and lithological composition information of each initial depositional system unit, superimpose the fault position coordinates in the deep fault feature parameters and the spatial boundary coordinates of the initial depositional system unit, and determine the initial depositional system unit and the passing position through which the fault passes.

[0052] Wherein, the computer device extracts the spatial boundary coordinates of each initial depositional system unit, including the maximum and minimum plane coordinates and vertical depth range of the unit, core depositional characteristics including sedimentary facies type, depositional environment, etc., and lithological composition information including the types of lithology contained in the unit and the thickness proportion of each lithology; then extract the fault position coordinates from the deep fault feature parameters, including the three-dimensional coordinate point set of the fault; then superimpose the fault position coordinates and the spatial boundary coordinates of the initial depositional system unit, use the coordinate matching method to determine the initial depositional system unit and the passing position through which the fault passes, and the passing position includes the three-dimensional coordinate points of the fault entering and leaving the initial depositional system unit.

[0053] Step 143: For the initial depositional system unit through which the fault passes, analyze the spatial relationship between the extension direction of the fault in the unit and the paleocurrent direction, judge the influence range of the fault on the depositional process combined with the characteristic variation amplitude of the fault, and adjust the spatial boundary of the initial depositional system unit according to the influence range to obtain the modified depositional system division result.

[0054] The computer device extracts extension direction data of the fracture in the initial depositional system unit, including the angle range of the extension direction, and extracts paleocurrent direction data in the unit, including the angle range of the paleocurrent direction; then analyzes the spatial relationship between the fracture extension direction and the paleocurrent direction, such as whether the two are parallel, perpendicular or at a certain angle; then judges the influence range of the fracture on the depositional process in combination with the characteristic variation range of the fracture, the characteristic variation range including the width variation and intensity variation of the fracture, and the influence range is determined according to the characteristic variation range of the fracture and the lithology composition of the depositional system unit; finally, the spatial boundary of the initial depositional system unit is adjusted according to the influence range, and the adjustment method includes expanding or reducing the spatial range of the unit, to obtain the modified depositional system division result.

[0055] Step 144: based on the modified depositional system division result, a depositional facies belt spatial distribution network is constructed, the fracture extension direction and the mutation position in the deep fracture characteristic parameter are integrated into the depositional facies belt spatial distribution network, and the intersection position and intersection mode of each fracture with different depositional facies belts are marked.

[0056] The computer device constructs a depositional facies belt spatial distribution network based on the modified depositional system division result, and the distribution network includes the spatial distribution, boundary coordinates and depositional characteristics of each depositional facies belt; then the fracture extension direction and the mutation position are extracted from the deep fracture characteristic parameter; then the above information is integrated into the depositional facies belt spatial distribution network, the extension direction of the fracture is associated with the distribution direction of the depositional facies belt, and the mutation position of the fracture is matched with the boundary of the depositional facies belt; finally, the intersection position of each fracture with different depositional facies belts is marked, including the three-dimensional coordinates of the intersection position, and the intersection mode such as cutting and tangency.

[0057] Step 1441: based on the spatial boundary coordinates and core depositional characteristics of each depositional system unit in the modified depositional system division result, the spatial range and vertical distribution level of each depositional facies belt are determined, and a three-dimensional depositional facies belt spatial distribution network is constructed, and the depositional facies belt spatial distribution network includes the unique identification and characteristic attribute index of each depositional facies belt.

[0058] The computer device determines the spatial range of each depositional facies belt based on the spatial boundary coordinates and core depositional characteristics of each depositional system unit in the modified depositional system division result, including the planar range and vertical depth range, and the vertical distribution level is determined according to the formation time of the depositional facies belt, such as the depositional facies belt formed earlier located in the lower layer and the depositional facies belt formed later located in the upper layer; then a three-dimensional depositional facies belt spatial distribution network is constructed, and the node coordinates of the distribution network correspond to the spatial coordinates of the depositional facies belt; finally, the unique identification and characteristic attribute index of each depositional facies belt are set in the depositional facies belt spatial distribution network, the unique identification is used to distinguish different depositional facies belts, and the characteristic attribute index is used to quickly query the characteristic attributes of the depositional facies belt.

[0059] Step 1442: Extract the fracture ID, spatial coordinate sequence, extension direction vector and characteristic mutation position coordinate of each fracture in the deep fracture feature parameter, and convert the spatial coordinate sequence of each fracture into the coordinate in the coordinate system consistent with the sedimentary facies belt spatial distribution network.

[0060] Wherein, the computer device extracts the fracture ID, spatial coordinate sequence, extension direction vector and characteristic mutation position coordinate of each fracture from the deep fracture feature parameter, the fracture ID is the unique identifier of each fracture, the spatial coordinate sequence is the three-dimensional coordinate point set of the fracture, the extension direction vector is the spatial vector representing the extension direction of the fracture, and the characteristic mutation position coordinate is the three-dimensional coordinate of the characteristic mutation of the fracture; then the spatial coordinate sequence of each fracture is converted into the coordinate in the coordinate system consistent with the sedimentary facies belt spatial distribution network, the conversion method includes coordinate translation, rotation, etc., to ensure the consistency of the coordinates.

[0061] Step 1443: Traverse the spatial coordinate sequence of each fracture, judge the sedimentary facies belt to which each coordinate point in the sequence belongs, when the coordinate point enters the coordinate range of another sedimentary facies belt from the coordinate range of one sedimentary facies belt, record the coordinate point as the intersection position of the fracture and the sedimentary facies belt, and record the unique identification of the sedimentary facies belt before and after the intersection; for each intersection position, determine the intersection mode of the fracture and the sedimentary facies belt in combination with the extension direction vector of the fracture and the distribution characteristics of the adjacent coordinate points.

[0062] Wherein, the computer device traverses the spatial coordinate sequence of each fracture, for each coordinate point in the sequence, queries the sedimentary facies belt spatial distribution network to determine the sedimentary facies belt to which the coordinate point belongs; when the coordinate point enters the coordinate range of another sedimentary facies belt from the coordinate range of one sedimentary facies belt, record the coordinate point as the intersection position of the fracture and the sedimentary facies belt, and record the unique identification of the sedimentary facies belt before and after the intersection; then for each intersection position, determine the intersection mode of the fracture and the sedimentary facies belt in combination with the extension direction vector of the fracture and the distribution characteristics of the adjacent coordinate points, such as the extension direction of the fracture being perpendicular to the boundary of the sedimentary facies belt, then the intersection mode is cutting, and parallel is tangent.

[0063] Step 1444: Store the fracture ID, extension direction vector, all intersection position coordinates, corresponding sedimentary facies belt identification and intersection mode of each fracture in association, form a fracture-sedimentary facies belt association data table, and integrate the association data table as attribute information into the sedimentary facies belt spatial distribution network.

[0064] The computer device stores the fracture ID, the extension direction vector, all intersection position coordinates, the corresponding sedimentary facies belt identifier and the intersection mode of each fracture in association, in the form of a data table, and the fields of the data table include the fracture ID, the extension direction vector, the intersection position coordinates, the sedimentary facies belt identifier, and the intersection mode.

[0065] Step 145: Analyzing the association between the fracture characteristics and the sedimentary facies belt characteristics at the intersection position, and statistically analyzing the distribution density and extension rule of the fractures in different sedimentary facies belts to establish the spatial association rule between the fractures and the sedimentary facies belts.

[0066] The computer device analyzes the association between the fracture characteristics and the sedimentary facies belt characteristics at the intersection position, such as whether the extension direction of the fracture is consistent with the distribution direction of the sedimentary facies belt, whether the mutation position of the fracture coincides with the boundary of the sedimentary facies belt, etc.; then statistically analyzes the distribution density of the fractures in different sedimentary facies belts, such as the distribution density of the fractures in the shallow marine facies sedimentary facies belt is higher, and the distribution density of the fractures in the coastal facies is lower; further statistically analyzes the extension rule of the fractures, such as the fractures mainly extend linearly in the shallow marine facies sedimentary facies belt and branch-like in the delta facies; and finally establishes the spatial association rule between the fractures and the sedimentary facies belts, including the association rule between the extension direction of the fracture and the distribution direction of the sedimentary facies belt, and the association rule between the distribution density of the fracture and the type of the sedimentary facies belt.

[0067] Step 146: Based on the sedimentary facies belt spatial distribution network, combining the spatial association rule and the complete deep fracture characteristic parameters, a three-dimensional model of the fracture system is constructed, which includes the spatial coordinates of the fractures, the extension mode, and the associated attributes of the sedimentary facies belt.

[0068] The computer device determines the spatial framework of the three-dimensional model of the fracture system based on the sedimentary facies belt spatial distribution network; then associates the extension direction and the distribution density of the fractures with the characteristics of the sedimentary facies belt based on the spatial association rule; further combines the complete deep fracture characteristic parameters, including the spatial coordinates, the extension mode, and the mutation position of the fractures; and finally constructs the three-dimensional model of the fracture system, which includes the spatial coordinate point set of the fractures, the three-dimensional geometric model of the extension mode, and the parameters of the associated attributes of the sedimentary facies belt.

[0069] Step 150: According to the three-dimensional stratigraphic model and the three-dimensional model of the fracture system, the association relationship between the lithological combination characteristics corresponding to the ore body deposition environment, the structural ore-controlling rule characteristics corresponding to the geological structure characteristic parameters, and the physical property anomaly characteristics corresponding to the geophysical data is identified, a ore body prediction model is established based on the association relationship as a constraint, spatial data of the target exploration area is input into the ore body prediction model, and a three-dimensional visual prediction result for the spatial distribution of the concealed ore body is obtained.

[0070] In this step, the computer equipment first extracts lithological combination information corresponding to the ore body's sedimentary environment from the 3D stratigraphic model, and obtains lithological combination characteristics through lithological type statistics and combination sequence analysis. Then, it filters ore-controlling structural attributes from geological structural characteristic parameters, and analyzes the control effect of structures on ore body formation by combining the fault distribution and extension characteristics in the 3D fault system model, thus obtaining structural ore-controlling characteristics. Next, it extracts ore body-related physical property information from geophysical data, and identifies the spatial distribution and intensity changes of physical property anomalies by combining anomalous characteristic areas in deep fault characteristic parameters, thus obtaining physical property anomaly characteristics. Then, it identifies the correlations between the above characteristics, such as the correlation between lithological combination characteristics and physical property anomaly characteristics, and the correlation between structural ore-controlling characteristics and ore body distribution. Then, it establishes an ore body prediction model based on these correlations, using machine learning algorithms such as random forests and support vector machines. Finally, it inputs spatial data of the target exploration area into the ore body prediction model, including stratigraphic data, structural data, and geophysical data. Finally, it obtains a 3D visualization prediction result for the spatial distribution of concealed ore bodies, including the spatial location, morphology, and distribution direction of the concealed ore bodies.

[0071] Step 151: Extract lithological combination information corresponding to the ore body depositional environment from the three-dimensional stratigraphic model, and obtain lithological combination characteristics through lithological type statistics and combination sequence analysis. The lithological combination characteristics include the proportion of lithological types, the combination arrangement order and the vertical variation law.

[0072] The computer equipment first delineates the spatial range of the ore body's sedimentary environment from a three-dimensional stratigraphic model. The determination of the spatial range is based on the known distribution location of the ore body and the characteristics of the sedimentary environment. Then, it extracts lithological assemblage information within this spatial range, including the name and thickness of each lithological type. Next, it statistically analyzes the lithological types, calculating the proportion of each lithological type's thickness to the total thickness, thus obtaining the lithological type percentage. Then, it performs sequence analysis on the lithological assemblage, analyzing the arrangement order of the lithological types, such as clastic rocks, carbonate rocks, and phosphate rocks from bottom to top. Finally, it analyzes the vertical variation patterns, such as the frequency of lithological type transitions and the trend of thickness changes, to obtain the lithological assemblage characteristics.

[0073] Step 1511: Based on the regional geological background and known geological clues, delineate the spatial range of the ore body depositional environment in the three-dimensional stratigraphic model, and extract the lithological information and spatial coordinate data of all stratigraphic units within the spatial range.

[0074] The computer device determines the spatial range of the ore body deposition environment in the three-dimensional stratum model based on the regional geological background, such as the geological structure evolution history of the target exploration area, the deposition environment of the known ore body, and the known geological clues, such as the lithological combination and distribution position of the discovered ore body, and determines the boundary coordinates of the spatial range according to the known geological clues; then extracts the lithological information of all stratum units in the spatial range, including the lithological type, color, structure, and spatial coordinate data, including the three-dimensional coordinate point set of the stratum unit.

[0075] Step 1512: Classify and count the extracted lithological information to determine the lithological types contained in the spatial range and the number of stratum units corresponding to each lithological type, calculate the proportion of the number of each lithological type in the total number of stratum units, and obtain the lithological type proportion data.

[0076] The computer device classifies and counts the extracted lithological information, classifies the lithological information by lithological type, determines the lithological types contained in the spatial range, such as carbonate rock, phosphate rock, and clastic rock, then counts the number of stratum units corresponding to each lithological type, and then calculates the proportion of the number of each lithological type in the total number of stratum units, which is obtained by dividing the number of stratum units of each lithological type by the total number of stratum units, to obtain the lithological type proportion data.

[0077] Step 1513: Sort the stratum units in the spatial range in the order of vertical depth in the spatial coordinate data to form a stratum sequence from shallow to deep, extract the lithological type of each stratum unit, record the appearance sequence of the lithological type in the order of sorting, and obtain the initial lithological combination sequence.

[0078] The computer device sorts the stratum units in the spatial range in the order of vertical depth in the spatial coordinate data, with the basis of the vertical depth of the stratum unit, with the stratum unit with shallow depth arranged in front and the stratum unit with deep depth arranged in back, to form a stratum sequence from shallow to deep, then extracts the lithological type of each stratum unit, and then records the appearance sequence of the lithological type in the order of sorting to obtain the initial lithological combination sequence, such as the initial lithological combination sequence being clastic rock, carbonate rock, phosphate rock, and carbonate rock.

[0079] Step 1514: Perform sequence analysis on the initial lithological combination sequence to identify the repeated lithological combination fragments and the appearance frequency of the fragments, determine the dominant lithological combination mode, and analyze the conversion mode and conversion frequency of the lithological types in the sequence to determine the combination arrangement order characteristics.

[0080] The computer device performs sequence analysis on the initial lithology combination sequence, adopts a sequence pattern mining algorithm, identifies repeatedly occurring lithology combination fragments in the sequence, such as a carbonate rock-phosphate rock as a repeatedly occurring fragment, and a frequency of occurrence of the fragment, determines a dominant lithology combination pattern, the dominant pattern is a lithology combination fragment with the highest frequency of occurrence, analyzes a conversion mode of the lithology types in the sequence, such as a gradual change or a sudden change from a clastic rock to a carbonate rock, and a conversion frequency is a number of times of conversion divided by a length of the sequence, and finally determines a combination arrangement order feature, the feature including an arrangement order of the lithology combination fragments, the conversion mode, and the conversion frequency.

[0081] Step 1515: constructing a lithology vertical distribution map with a vertical depth as a horizontal axis and a lithology type as a vertical axis, marking a depth interval and a thickness corresponding to each lithology type in the lithology vertical distribution map, identifying variation characteristics of the lithology types in different depth intervals, and calculating a conversion gradient of the lithology types with respect to the depth; and mining a variation law of the lithology combination with respect to the vertical depth in combination with the lithology vertical distribution map and the conversion gradient data, the variation law including a vertical zoning feature of the lithology types and dominant lithology types of different depth sections.

[0082] The computer device constructs a lithology vertical distribution map with a vertical depth as a horizontal axis and a lithology type as a vertical axis; then marks a depth interval and a thickness corresponding to each lithology type in the lithology vertical distribution map, the depth interval is a vertical depth range of the lithology type distribution, and the thickness is a vertical thickness of the lithology type; then identifies variation characteristics of the lithology types in different depth intervals, such as that the lithology types in a shallow depth interval are mainly clastic rocks, the lithology types in a middle depth interval are mainly carbonate rocks, and the lithology types in a deep depth interval are mainly phosphate rocks; then calculates a conversion gradient of the lithology types with respect to the depth, the conversion gradient is calculated by dividing a conversion number of the lithology types by a length of the depth interval; and finally mines a variation law of the lithology combination with respect to the vertical depth in combination with the lithology vertical distribution map and the conversion gradient data, the variation law including a vertical zoning feature of the lithology types, such as three zones of shallow, middle, and deep, and dominant lithology types of different depth sections, such as that the dominant lithology type of the shallow depth section is clastic rock, the dominant lithology type of the middle depth section is carbonate rock, and the dominant lithology type of the deep depth section is phosphate rock.

[0083] Step 152: screening a structure attribute related to ore control from the geological structure feature parameters, analyzing a control action of the structure on the formation of the ore body in combination with a fracture distribution and extension feature in the fracture system three-dimensional model, obtaining a structure ore-controlling law feature, and the structure ore-controlling law feature including a fracture ore-controlling range and a correlation between a structure intersection relationship and the ore body.

[0084] Among them, the computer device screens the structural properties related to ore control from the geological structure feature parameters, such as the properties of faults, the shapes of folds, the occurrences of structural interfaces, etc.; then combines the fault distribution and extension characteristics in the three-dimensional model of the fault system to analyze the control of the faults on the ore body, such as whether the extension direction of the fault is consistent with the distribution direction of the ore body, whether the intersection position of the fault is the enrichment area of the ore body, etc.; then analyzes the relevance of the structural intersection relationship and the ore body, such as whether the intersection of faults of different directions is easy to form an ore body; and finally obtains the structural ore-controlling regularity characteristics, which include the three-dimensional coordinate range of the fault ore-controlling range, the correlation rules of the structural intersection relationship and the ore body distribution, etc.

[0085] Step 153: Extracting the physical property information related to the ore body from the geophysical data, combining the abnormal feature area in the deep fault feature parameters, identifying the spatial distribution and intensity variation of the physical property anomaly, and obtaining the physical property anomaly feature, which includes the abnormal spatial coordinates, the intensity gradient, and the spatial correlation with the fault.

[0086] Among them, the computer device extracts the physical property information related to the ore body from the geophysical data, such as density, magnetism, and electrical properties; then combines the abnormal feature area in the deep fault feature parameters, such as gravity anomaly area, magnetic anomaly area, etc.; then identifies the spatial distribution of the physical property anomaly, including the three-dimensional coordinate range of the abnormal area, and the intensity variation, including the intensity value and the variation trend of the anomaly; finally analyzes the spatial correlation of the physical property anomaly and the fault, such as whether the physical property anomaly area coincides with the distribution range of the fault, whether the intensity variation of the anomaly is consistent with the characteristic variation of the fault, etc., to obtain the physical property anomaly feature.

[0087] Step 154: Standardizing the lithological combination feature, the structural ore-controlling regularity feature, and the physical property anomaly feature respectively to eliminate dimensional differences and convert them into feature vectors of a unified dimension, identifying the interdependence relationship and influence weight through feature correlation analysis, and forming a correlation relationship set.

[0088] Among them, the computer device standardizes the lithological combination feature, the structural ore-controlling regularity feature, and the physical property anomaly feature respectively, the standardization methods include mean standardization, standard deviation standardization, etc., to eliminate dimensional differences; then converts the standardized features into feature vectors of a unified dimension, the dimension of the feature vector is determined according to the number of features; then identifies the interdependence relationship and influence weight between the features through feature correlation analysis, such as Pearson correlation coefficient analysis, Spearman rank correlation analysis, etc.; finally arranges the interdependence relationship and influence weight into a correlation relationship set.

[0089] Step 155: constructing a model structure of the ore body prediction model with the set of association relationships as constraint conditions, the model structure comprising a feature input layer, an association analysis layer and a result output layer; wherein the association analysis layer embeds the interdependence rules and influence weights in the set of association relationships.

[0090] The computer device determines a model structure of the ore body prediction model with the set of association relationships as constraint conditions, the model structure comprising a feature input layer, an association analysis layer and a result output layer; the feature input layer is used to receive input feature data, the association analysis layer is used to analyze the association relationships between features, and the result output layer is used to output prediction results; wherein the association analysis layer embeds the interdependence rules and influence weights in the set of association relationships, the interdependence rules are used to determine the dependency relationship between features, and the influence weights are used to determine the influence degree of features on prediction results.

[0091] Step 1551: obtaining input indicators and output indicators of the ore body prediction model, the input indicators being lithology combination feature data, structure ore-controlling regularity feature data and physical property anomaly feature data of the target exploration area, and the output indicators being spatial distribution probability and morphological parameters of the concealed ore body.

[0092] The computer device obtains input indicators and output indicators of the ore body prediction model, the input indicators being lithology combination feature data, structure ore-controlling regularity feature data and physical property anomaly feature data of the target exploration area, the lithology combination feature data including lithology type proportion, combination arrangement order and vertical variation law, the structure ore-controlling regularity feature data including fault ore-controlling range, structure intersection relationship and association with the ore body, and the physical property anomaly feature data including abnormal spatial coordinates, intensity gradient and spatial association with the fault; the output indicators being spatial distribution probability and morphological parameters of the concealed ore body, the spatial distribution probability indicating the probability of existence of the concealed ore body at each position in the target exploration area, and the morphological parameters indicating the shape and size of the concealed ore body.

[0093] Step 1552: constructing a feature input layer; wherein the feature input layer comprises a data preprocessing branch and a feature standardization branch, the data preprocessing branch is used to unify the format and remove noise of the input feature data, and the feature standardization branch is used to convert the preprocessed feature data into a standardized feature vector with uniform dimension.

[0094] The computer device constructs a feature input layer, and the feature input layer includes a data preprocessing branch and a feature standardization branch. The data preprocessing branch uniformly formats the input feature data, converts feature data in different formats into the same format, such as converting text format into numerical format, and simultaneously removes noise, removes abnormal values and missing values in the feature data, and the like. The feature standardization branch converts the preprocessed feature data into a standardized feature vector with a uniform dimension, and the standardization method includes mean standardization, standard deviation standardization, and the like.

[0095] Step 1553: constructing a correlation analysis layer; the correlation analysis layer includes a rule storage sub-branch, a weight distribution sub-branch, and a correlation calculation sub-branch. The rule storage sub-branch is used to store interdependent rules in the correlation relationship set, convert the rules into logical judgment statements executable by the model, the weight distribution sub-branch is used to distribute the influence weight in the correlation relationship set to the corresponding feature index, determine the contribution degree of each feature index to the prediction result, and the feature index with a high contribution degree is assigned a higher weight coefficient. The correlation calculation sub-branch embeds the logical judgment statement and the weight coefficient, is used to receive the standardized feature vector from the feature input layer, analyzes the matching of each feature index in the feature vector through the logical judgment statement, and calculates the correlation score of each feature index in combination with the weight coefficient.

[0096] The computer device constructs a correlation analysis layer, and the correlation analysis layer includes a rule storage sub-branch, a weight distribution sub-branch, and a correlation calculation sub-branch. The rule storage sub-branch stores interdependent rules in the correlation relationship set, converts the rules into logical judgment statements executable by the model, such as “if the phosphate rock proportion in the lithology type proportion is greater than 50%, and the fracture ore-controlling range in the structure ore-controlling law feature contains the position, then the position has a higher probability of containing a concealed ore body”. The weight distribution sub-branch distributes the influence weight in the correlation relationship set to the corresponding feature index, such as the influence weight of the phosphate rock proportion is 0.6, and the influence weight of the fracture ore-controlling range is 0.4. The correlation calculation sub-branch embeds the logical judgment statement and the weight coefficient, receives the standardized feature vector from the feature input layer, analyzes the matching of each feature index in the feature vector through the logical judgment statement, and calculates the correlation score of each feature index in combination with the weight coefficient. The calculation method of the correlation score is that the standardized value of the feature index is multiplied by the weight coefficient.

[0097] Step 1554: constructing a result output layer; the result output layer includes a result fusion sub-branch and a format conversion sub-branch. The result fusion sub-branch is used to comprehensively fuse the correlation scores output by the correlation calculation sub-branch to obtain the spatial distribution probability value of the concealed ore body. The format conversion sub-branch is used to convert the spatial distribution probability value and the corresponding morphological parameters into a standardized output format.

[0098] The computer device constructs a result output layer, and the result output layer includes a result fusion sub-branch and a format conversion sub-branch; the result fusion sub-branch comprehensively fuses the correlation scores output by the correlation analysis sub-branch, and the fusion method includes weighted summation, logistic regression, etc., to obtain the spatial distribution probability value of the concealed ore body; and the format conversion sub-branch converts the spatial distribution probability value and the corresponding morphological parameters into a standardized output format, such as a JSON format, an XML format, etc.

[0099] Step 1555: connecting the feature input layer, the correlation analysis layer and the result output layer in a data transmission order to form the model structure.

[0100] The computer device connects the feature input layer, the correlation analysis layer and the result output layer in a data transmission order, the output of the feature input layer is connected to the input of the correlation analysis layer, the output of the correlation analysis layer is connected to the input of the result output layer, and the model structure is formed.

[0101] Step 156: collecting geological data of known ore bodies as training samples, inputting the feature data in the training samples into the ore body prediction model, and adjusting the internal parameters of the ore body prediction model to complete the construction of the ore body prediction model.

[0102] The computer device collects geological data of known ore bodies as training samples, and the training samples include lithological combination feature data, structure ore-controlling regularity feature data, physical property anomaly feature data and corresponding spatial distribution probability and morphological parameters of the concealed ore body; then the feature data in the training samples is input into the ore body prediction model; then the internal parameters of the ore body prediction model are adjusted, such as the weight coefficients in the correlation analysis layer and the threshold values of the logical judgment statements, and the adjustment method includes the gradient descent method and the genetic algorithm; finally, the construction of the ore body prediction model is completed.

[0103] Step 157: inputting the spatial data of the target exploration area into the constructed ore body prediction model to obtain the spatial coordinates, morphological parameters and distribution trend of the concealed ore body through the correlation analysis layer.

[0104] The computer device collects spatial data of the target exploration area, and the spatial data includes stratum data, structure data and geophysical data; then the spatial data is input into the constructed ore body prediction model; then the correlation analysis layer is used to analyze the correlation between the features to obtain the spatial coordinates, morphological parameters and distribution trend of the concealed ore body; and finally, the processing result is output.

[0105] Step 1571: Collect the spatial data of the target exploration area covering stratum lithology, geological structure, geophysical exploration and spatial coordinates, and after preprocessing the spatial data, extract the lithology combination features, structure ore-controlling regularity features and physical property anomaly features related data according to the input indicators of the ore body prediction model to generate the original input data set.

[0106] Among them, the computer device collects the spatial data of the target exploration area, and the data covers stratum lithology, geological structure, geophysical exploration and spatial coordinates, and the data sources include field geological survey, geophysical exploration, well logging, etc.; then the spatial data is preprocessed, the preprocessing includes format unification, noise removal, etc.; then the lithology combination features, structure ore-controlling regularity features and physical property anomaly features related data are extracted according to the input indicators of the ore body prediction model; finally, the original input data set is generated.

[0107] Step 1572: calling the feature input layer of the ore body prediction model to complete data standardization to obtain a standardized feature vector, inputting the standardized feature vector into the correlation calculation sub-branch of the correlation analysis layer, and calling the interdependent rules stored in the rule storage sub-branch by the correlation calculation sub-branch to perform one-by-one logical judgment on the matching relationship between each feature index in the standardized feature vector and the ore body.

[0108] Among them, the computer device calls the feature input layer of the ore body prediction model to standardize the original input data set and obtain a standardized feature vector; then the standardized feature vector is input into the correlation calculation sub-branch of the correlation analysis layer; then the correlation calculation sub-branch calls the interdependent rules stored in the rule storage sub-branch to perform one-by-one logical judgment on the matching relationship between each feature index in the standardized feature vector and the ore body, such as whether the phosphate rock proportion in the lithology type proportion is greater than the threshold value, whether the fracture ore-controlling range in the structure ore-controlling regularity features contains the position, etc.

[0109] Step 1573: synchronously acquiring the weight coefficients assigned to each feature index by the weight distribution sub-branch through the correlation calculation sub-branch, calculating the correlation score corresponding to each standardized feature vector based on the logical matching result and combining the weight coefficients through the weighted summation algorithm; wherein the weighted summation algorithm is based on the standardized feature vector.

[0110] The computer device synchronously obtains the weight coefficients allocated to each feature index by the weight distribution sub-branch through association calculation; then, based on the logical matching result, such as the matching relationship between the feature index and the ore body being yes or no, and in combination with the weight coefficients, the association score corresponding to each standardized feature vector is calculated through a weighted summation algorithm, the calculation method of the weighted summation algorithm being that the standardized value of each feature index is multiplied by the corresponding weight coefficient, and then the products are added; wherein the weighted summation algorithm is based on the standardized feature vector, ensuring the accuracy of the calculation result.

[0111] Step 1574: The association score is transmitted to the result fusion sub-branch of the result output layer, the single-feature association score is complementarily verified through a multi-feature fusion algorithm, the distribution probability of the concealed ore body at each spatial position of the target exploration area is obtained, and a potential area with a concealed ore body existence probability higher than the distribution probability threshold is screened out according to a preset distribution probability threshold.

[0112] The computer device transmits the association score to the result fusion sub-branch of the result output layer; then, the single-feature association score is complementarily verified through a multi-feature fusion algorithm, the multi-feature fusion algorithm including Bayesian fusion, D-S evidence theory, etc., the distribution probability of the concealed ore body at each spatial position of the target exploration area is obtained; then, a potential area with a concealed ore body existence probability higher than 0.7 is screened out according to a preset distribution probability threshold, such as 0.7.

[0113] Step 1575: In combination with the spatial coordinate data carried in the standardized feature vector, the three-dimensional spatial coordinate range of the potential area is located, the boundary profile, extension length and thickness variation of the potential area are analyzed through a geological body morphology analysis algorithm to obtain the morphology parameters, and the spatial positional relationship and morphology correlation of adjacent potential areas are analyzed to obtain the distribution trend of the concealed ore body.

[0114] The computer device combines the spatial coordinate data carried in the standardized feature vector to locate the three-dimensional spatial coordinate range of the potential area, the coordinate range being determined according to the maximum and minimum coordinate values in the spatial coordinate data; then, the boundary profile, such as circular, elliptical, irregular, etc., of the potential area, the extension length, such as the length of the longest extension direction of the potential area, and the thickness variation, such as the vertical thickness variation of the potential area, are analyzed through a geological body morphology analysis algorithm to obtain the morphology parameters; then, the spatial positional relationship and morphology correlation of adjacent potential areas, such as whether the adjacent potential areas are connected and whether the morphologies are similar, etc., are analyzed to obtain the distribution trend of the concealed ore body, the distribution trend including the extension direction and distribution range of the concealed ore body, etc.

[0115] Step 158: Based on the spatial coordinates, morphological parameters and distribution trend of the concealed ore body, combined with the spatial structure network of the three-dimensional stratum model, a three-dimensional visualization prediction result for the spatial distribution of the concealed ore body is generated, which is used to present the spatial positional relationship between the concealed ore body and the stratum and the fault.

[0116] In the method, the computer device determines the position and morphology of the concealed ore body in the three-dimensional space based on the spatial coordinates, morphological parameters and distribution trend of the concealed ore body, combined with the spatial structure network of the three-dimensional stratum model; then generates the three-dimensional visualization prediction result for the spatial distribution of the concealed ore body, which is presented in the form of a three-dimensional graph and contains the spatial position, morphology, distribution trend of the concealed ore body, and the spatial positional relationship with the stratum and the fault; and finally outputs the three-dimensional visualization prediction result for guiding the exploration of mineral resources.

[0117] In another embodiment, the method further comprises: Step 161: Extract the spatial correlation features of the concealed ore body area and the surrounding stratum from the three-dimensional visualization prediction result, and call the original stratum attribute data of the concealed ore body area from the three-dimensional stratum model, to obtain attribute difference information of the ore body associated stratum through feature comparison, and generate a feature difference data set.

[0118] In the method, the computer device extracts the spatial correlation features of the concealed ore body area and the surrounding stratum from the three-dimensional visualization prediction result, including the positional relationship and contact mode of the concealed ore body area and the surrounding stratum; then calls the original stratum attribute data of the concealed ore body area from the three-dimensional stratum model, including the lithology, thickness and age of the stratum; then compares the attribute difference between the concealed ore body area and the surrounding stratum through feature comparison, to obtain the attribute difference information of the ore body associated stratum; and finally organizes the attribute difference information into a feature difference data set.

[0119] Step 162: Fuse the feature difference data set with the original geological structure feature parameters of the target exploration area, perform feature vector conversion on the fused data, establish a mapping relationship between the difference features and the structure parameters through semantic correlation analysis, and obtain a fused feature set.

[0120] In the method, the computer device fuses the feature difference data set with the original geological structure feature parameters of the target exploration area, and the fusion method includes feature splicing and feature weighting; then performs feature vector conversion on the fused data, converts the fused data into feature vectors; then establishes a mapping relationship between the difference features and the structure parameters through semantic correlation analysis, such as the mapping relationship between the lithology difference in the attribute difference information and the fault property in the geological structure feature parameters; and finally obtains the fused feature set.

[0121] Step 163: dynamically adjust the lithology combination attribute and spatial distribution parameter of the corresponding stratigraphic unit in the three-dimensional stratigraphic model based on the fusion feature set, and generate an initial updated model; wherein the adjustment process maintains the consistency of the overall structure of the strata based on the constraint condition of the stratigraphic superimposition relationship.

[0122] Wherein, the computer device determines the stratigraphic unit that needs to be adjusted based on the fusion feature set; then dynamically adjusts the lithology combination attribute and spatial distribution parameter of the corresponding stratigraphic unit in the three-dimensional stratigraphic model, the adjustment of the lithology combination attribute is based on the lithology difference information in the feature difference data set, and the adjustment of the spatial distribution parameter is based on the spatial position difference information in the feature difference data set; then generate an initial updated model; wherein the adjustment process maintains the consistency of the overall structure of the strata based on the constraint condition of the stratigraphic superimposition relationship, such as the unchanged superimposition order of the strata.

[0123] Step 164: spatially match and verify the initial updated model with the three-dimensional fracture system model, identify the spatial connection contradiction area of the strata and the fractures, and correct the boundary parameter of the initial updated model based on the feature difference data of the contradiction area to obtain the feature update result of the three-dimensional stratigraphic model.

[0124] Wherein, the computer device spatially matches and verifies the initial updated model with the three-dimensional fracture system model, the verification method includes coordinate matching, shape matching, etc.; then identifies the spatial connection contradiction area of the strata and the fractures, such as the mismatch of the boundary of the strata and the position of the fractures; then corrects the boundary parameter of the initial updated model based on the feature difference data of the contradiction area, the correction of the boundary parameter is based on the spatial position difference information in the feature difference data set; finally obtains the feature update result of the three-dimensional stratigraphic model.

[0125] In another embodiment, the method further comprises: Step 171: extract the spatial coordinate boundary of the concealed ore body from the three-dimensional visualization prediction result, combine the spatial distribution data of the fractures in the three-dimensional fracture system model, identify the intersection features of the ore body boundary and the fractures, and generate a fracture-ore body correlation feature vector.

[0126] Wherein, the computer device extracts the spatial coordinate boundary of the concealed ore body from the three-dimensional visualization prediction result, the boundary coordinates include the maximum and minimum three-dimensional coordinates of the concealed ore body; then combines the spatial distribution data of the fractures in the three-dimensional fracture system model, including the three-dimensional coordinate point set of the fractures; then identifies the intersection features of the ore body boundary and the fractures, including the three-dimensional coordinates of the intersection position, the intersection mode, etc.; finally generates a fracture-ore body correlation feature vector, the dimension of the feature vector is determined according to the number of intersection features.

[0127] Step 172: Collecting regional tectonic evolution time series data of the target exploration area, dividing the time series data into stages and converting it into a space-time vector form, aligning the space-time vector with the fault-ore body correlation feature vector in dimension, and obtaining a space-time fusion vector through feature fusion.

[0128] In the method, the computer device collects regional tectonic evolution time series data of the target exploration area, the data including time stages of tectonic evolution, tectonic features of each stage, etc.; then divides the tectonic evolution into stages according to the time series data; converts the tectonic features of each stage into a space-time vector form, the direction of the space-time vector representing the time sequence of tectonic evolution and the length representing the intensity of tectonic features; then aligns the space-time vector with the fault-ore body correlation feature vector in dimension to ensure the consistency of the dimensions of the vectors; and finally obtains a space-time fusion vector through feature fusion, the fusion method including vector splicing and vector weighting, etc.

[0129] Step 173: Taking the space-time fusion vector as the iteration basis, the extension direction and spatial form parameters of the corresponding fault in the fault system three-dimensional model are time-sequenced and corrected to generate an initial iteration model, and the correction process keeps the correlation properties of the fault and the sedimentary facies unchanged.

[0130] In the method, the computer device takes the space-time fusion vector as the iteration basis to determine the fault to be corrected; then time-sequences and corrects the extension direction and spatial form parameters of the corresponding fault in the fault system three-dimensional model, the correction of the extension direction being based on the time stage information in the space-time fusion vector and the correction of the spatial form parameters being based on the tectonic feature intensity information in the space-time fusion vector; then generates an initial iteration model; and the correction process keeps the correlation properties of the fault and the sedimentary facies unchanged, such as the intersection position and intersection mode of the fault and the sedimentary facies.

[0131] Step 174: Spatially correlating and verifying the initial iteration model with the three-dimensional stratum model with feature updates, identifying the spatial connection contradiction area of the stratum and the fault, correcting the boundary parameters of the initial iteration model based on the feature difference data of the contradiction area, obtaining a space-time vector iteration result of the fault system three-dimensional model, and the iteration result being used to improve the time sequence adaptability of the ore body prediction.

[0132] The computer device performs spatial correlation verification on the initial iteration model and the three-dimensional stratum model after feature updating, and the verification method includes coordinate matching, shape matching, and the like; then, a spatial connection contradiction area of the stratum and the fracture is identified, such as an extension direction of the fracture being inconsistent with a distribution direction of the stratum; then, a boundary parameter of the initial iteration model is corrected based on feature difference data of the contradiction area, and the correction of the boundary parameter is based on spatial position difference information in the feature difference data set; finally, a time-space vector iteration result of the fracture system three-dimensional model is obtained, and the iteration result is used to improve time sequence adaptability of ore body prediction, so that the ore body prediction result is more in line with the law of regional tectonic evolution.

[0133] The embodiment of the application extracts and converts feature information of the thrust nappe structure of the target exploration area into geological structure feature parameters, integrates stratum age data and lithology distribution information to construct a three-dimensional stratum model, imports geophysical data to extract geophysical response features of strata at different depths and corrects to obtain deep fracture feature parameters, divides a sedimentary system based on facies paleogeographic information and integrates the deep fracture feature parameters to establish a fracture system three-dimensional model, and finally identifies a multi-feature correlation relationship and establishes an ore body prediction model to obtain a three-dimensional visual prediction result of spatial distribution of the concealed ore body.

[0134] In this way, multi-dimensional data fusion and analysis of geological structure, stratum spatial distribution, geophysical response, facies paleogeography, and ore body prediction are realized as a whole, scientific problems such as unclear sedimentary microfacies of the phosphorus-bearing strata, unclear relationship between the sedimentary microfacies and phosphorus enrichment mineralization, and unclear spatial distribution law caused by large spatial distribution changes of the deep phosphorus ore are effectively solved, deep phosphorus ore exploration technology is established, fine reconstruction technology of facies paleogeography based on big data, three-dimensional prediction technology of deep phosphorus ore in a mineral concentration area based on multi-source data, green and efficient ore prediction technology based on the fusion of geological, geophysical, geochemical, and remote sensing big data and artificial intelligence, and deep phosphorus ore spatial accurate positioning technology are broken through, comprehensive and accurate technical support is provided for exploration and prediction of deep phosphorus ore, and the efficiency and accuracy of deep phosphorus ore exploration are improved.

[0135] Referring to Figure 2 As shown in the figure, the figure is a schematic diagram of a basic structure of a computer device 20 provided by an embodiment of the application, and the computer device 20 includes: a processor 201; a storage device 202, on which a computer program 2020 is stored; When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the mineral resource exploration data analysis methods based on three-dimensional space modeling.

[0136] Based on the above, a readable storage medium is provided, and the readable storage medium stores programs or instructions, and the programs or instructions are executed by a processor to implement the steps of the above method.

[0137] Referring to Figure 3 As shown in the figure, the figure is a function module block diagram of a mineral resource exploration data analysis device based on three-dimensional space modeling provided by an embodiment of the application, and the mineral resource exploration data analysis device based on three-dimensional space modeling comprises: A feature information extraction module is configured to extract feature information of a thrust nappe structure of a target exploration area by interface occurrence measurement, lithological combination analysis and tectonic movement trace identification, and convert the feature information into geological structure feature parameters. A stratum model construction module is configured to integrate stratum age data and lithological distribution information of the target exploration area based on the geological structure feature parameters, and construct a three-dimensional stratum model reflecting spatial distribution of strata of the target exploration area by taking the determined stratum superimposition relationship by tectonic sequence division as a constraint condition. A fault feature determination module is configured to import geophysical data of the target exploration area based on the three-dimensional stratum model, extract geophysical response features of strata at different depths from the geophysical data, and correct the extraction results in combination with lithological boundary correction in the three-dimensional stratum model to obtain deep fault feature parameters. A spatial correlation analysis module is configured to divide a sedimentary system of the target exploration area to obtain a sedimentary system division result, integrate the deep fault feature parameters into the sedimentary system division result, and establish a fault system three-dimensional model reflecting spatial distribution of faults and a relationship with deposition by spatial correlation analysis of faults and sedimentary facies belts. A mineral body prediction processing module is configured to identify an association relationship between lithological combination features corresponding to a mineral body deposition environment, tectonic ore-controlling regularity features corresponding to geological structure feature parameters, and physical property anomaly features corresponding to geophysical data based on the three-dimensional stratum model and the fault system three-dimensional model, establish a mineral body prediction model by taking the association relationship as a constraint, input spatial data of the target exploration area into the mineral body prediction model, and obtain a three-dimensional visual prediction result for spatial distribution of a concealed mineral body.

[0138] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the described device embodiments are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0139] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0140] In addition, each functional unit in the embodiments of the present application can be integrated in a processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.

[0141] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this, the technical solutions of the present application essentially or the part that makes a contribution to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0142] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent replacements; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of analyzing mineral resource exploration data based on three-dimensional spatial modeling, characterized by, The method comprises the following steps: characteristic information of the thrust nappe structure in the target exploration area is extracted by measuring the interface occurrence, analyzing the lithological combination and identifying the tectonic movement traces, and the characteristic information is converted into geological structure characteristic parameters; stratigraphic age data and lithological distribution information of the target exploration area are integrated based on the geological structure characteristic parameters, so as to construct a three-dimensional stratigraphic model reflecting the spatial distribution of the strata in the target exploration area by taking the determined stratigraphic superimposition relationship by tectonic sequence division as a constraint condition; based on the three-dimensional stratigraphic model, the geophysical data of the target exploration area are imported, the geophysical response characteristic extraction of different depth strata is performed on the geophysical data, and the extraction result is corrected in combination with the lithological boundary in the three-dimensional stratigraphic model, so as to obtain deep fault characteristic parameters; the lithofacies paleogeographic information of the target exploration area is divided into sedimentary systems, so as to obtain a sedimentary system division result, the deep fault characteristic parameters are integrated into the sedimentary system division result, and a three-dimensional fault system model reflecting the spatial distribution of faults and the relationship with deposition is established through spatial correlation analysis of faults and sedimentary facies belts; according to the three-dimensional stratigraphic model and the three-dimensional fault system model, the correlation between the lithological combination characteristics corresponding to the ore body deposition environment, the tectonic ore-controlling regularity characteristics corresponding to the geological structure characteristic parameters and the physical property anomaly characteristics corresponding to the geophysical data is identified, a ore body prediction model is established by taking the correlation as a constraint, spatial data of the target exploration area are input into the ore body prediction model, and a three-dimensional visual prediction result for the spatial distribution of concealed ore bodies is obtained.

2. The method of claim 1, wherein, The method of obtaining the deep fault characteristic parameters based on the three-dimensional stratigraphic model, importing the geophysical data of the target exploration area, performing geophysical response characteristic extraction of different depth strata on the geophysical data and correcting the extraction result in combination with the lithological boundary in the three-dimensional stratigraphic model comprises the following steps: the spatial distribution range and lithological interface coordinates of the strata in the three-dimensional stratigraphic model are taken as references to delimit the processing interval of the geophysical data, the geophysical data are divided into a plurality of data subsets according to the depth gradient, and each data subset corresponds to a stratum unit in the three-dimensional stratigraphic model; geophysical response characteristic extraction is performed on each data subset, initial response characteristic sets containing amplitude characteristics, phase characteristics and continuity characteristics are obtained through signal feature decomposition, and the lithological boundary coordinates and lithological change attributes of the corresponding stratum unit in the three-dimensional stratigraphic model are synchronously extracted; the initial response characteristic set and the lithological boundary coordinates are spatially matched, whether there is an abnormal feature point in the initial response characteristic set is judged according to the lithological change attribute, if there is, the spatial position corresponding to the abnormal feature point is marked and its spatial relationship with the lithological boundary is analyzed; for the marked abnormal feature point, the corresponding geophysical data subset is traced back, signal feature decomposition is re-performed and decomposition parameters are adjusted, a corrected response characteristic set is obtained, and the corrected response characteristic set and the response characteristic set of the adjacent depth stratum unit are associated and compared. The region with feature mutation in the correlation matching process is extracted, the spatial range of the region is determined in combination with the lithological boundary coordinates, the feature mutation law and spatial form of the region are analyzed, and deep fault feature parameters including the mutation position, extension direction and feature change amplitude are generated.

3. The method of claim 1, wherein, The lithofacies paleogeographic information of the target exploration area is collected, the sedimentary system is divided based on the spatial distribution law of the sedimentary facies type and the gradient change of the lithofacies change feature, and a plurality of initial sedimentary system units are obtained, including: The lithofacies paleogeographic information of the target exploration area is collected, the sedimentary system is divided based on the spatial distribution law of the sedimentary facies type and the gradient change of the lithofacies change feature, and a plurality of initial sedimentary system units are obtained, including: The spatial boundary coordinates, core sedimentary features and lithological composition information of each initial sedimentary system unit are extracted, the fault position coordinates in the deep fault feature parameters are superimposed with the spatial boundary coordinates of the initial sedimentary system unit, and the initial sedimentary system unit and the passing position through which the fault passes are determined; For the initial sedimentary system unit through which the fault passes, the spatial relationship between the extension direction of the fault in the unit and the paleocurrent direction is analyzed, the influence range of the fault on the sedimentary process is judged in combination with the feature change amplitude of the fault, and the spatial boundary of the initial sedimentary system unit is adjusted according to the influence range to obtain a corrected sedimentary system division result; Based on the corrected sedimentary system division result, a sedimentary facies belt spatial distribution network is constructed, the fault extension direction and mutation position in the deep fault feature parameters are integrated into the sedimentary facies belt spatial distribution network, and the intersection position and intersection mode of each fault with different sedimentary facies belts are marked; The correlation of the fault features and the sedimentary facies belt features at the intersection position is analyzed, the distribution density and extension law of the faults in different sedimentary facies belts are counted, and the spatial correlation rule of the faults and the sedimentary facies belts is established; Based on the sedimentary facies belt spatial distribution network, the spatial correlation rule and the complete deep fault feature parameters, a fault system three-dimensional model including the fault spatial coordinates, extension form and sedimentary facies belt correlation attribute is constructed.

4. The method of claim 3, wherein, The lithofacies paleogeographic information of the target exploration area is collected, the sedimentary system is divided based on the spatial distribution law of the sedimentary facies type and the gradient change of the lithofacies change feature, and a plurality of initial sedimentary system units are obtained, including: The lithofacies paleogeographic information of the target exploration area is collected, the sedimentary system is divided based on the spatial distribution law of the sedimentary facies type and the gradient change of the lithofacies change feature, and a plurality of initial sedimentary system units are obtained, including: The sedimentary facies type is classified, the main sedimentary facies type and the typical lithological markers of each type are determined, the paleocurrent direction data is converted into a spatial vector form, and the lithofacies change gradient data is obtained by quantifying the lithofacies change feature description; Based on the planar coordinates of the target exploration area, a two-dimensional analysis grid is constructed, and the classified sedimentary facies types, paleocurrent directions in the form of spatial vectors, and quantified lithofacies change gradient data are respectively mapped to the corresponding nodes of the two-dimensional analysis grid; The sedimentary facies type similarity, paleocurrent direction vector angle, and lithofacies change gradient difference of adjacent nodes in the two-dimensional analysis grid are calculated, and a similarity threshold, an angle threshold, and a difference threshold are set. Adjacent nodes that satisfy the conditions of the sedimentary facies type similarity being higher than the similarity threshold, the paleocurrent direction vector angle being smaller than the angle threshold, and the lithofacies change gradient difference being smaller than the difference threshold are classified into the same unit. The sedimentary facies type similarity is calculated based on the type matching degree, the paleocurrent direction vector angle is measured based on the spatial vector angle, and the lithofacies change gradient difference is quantified based on the lithofacies change intensity. The node merging process is repeated to form initial sedimentary units on multiple planes by traversing the entire two-dimensional analysis grid. Combined with the stratigraphic thickness information and vertical lithofacies change characteristics in the three-dimensional stratigraphic model, the initial sedimentary units on the plane are extended along the vertical direction to obtain multiple initial sedimentary system units containing vertical dimension information. The vertical boundary coordinates, planar distribution range, and core characteristic parameters of each initial sedimentary system unit are extracted. The core characteristic parameters include the dominant sedimentary facies type, the dominant paleocurrent direction, and the average lithofacies change gradient. The dominant paleocurrent direction is the dominant paleocurrent flow direction of the target exploration area during the target sedimentary period. Based on the corrected sedimentary system division result, a sedimentary facies belt spatial distribution grid is constructed. The fault extension direction and mutation position in the deep fault characteristic parameters are integrated into the sedimentary facies belt spatial distribution grid. The intersection position and intersection mode of each fault and different sedimentary facies belts are marked, including: Based on the spatial boundary coordinates and core sedimentary characteristics of each sedimentary system unit in the corrected sedimentary system division result, the spatial range and vertical distribution level of each sedimentary facies belt are determined, and a three-dimensional sedimentary facies belt spatial distribution grid is constructed. The sedimentary facies belt spatial distribution grid contains the unique identification and characteristic attribute index of each sedimentary facies belt. The fault ID, spatial coordinate sequence, extension direction vector, and characteristic mutation position coordinates of each fault in the deep fault characteristic parameters are extracted. The spatial coordinate sequence of each fault is converted to coordinates in the same coordinate system as the sedimentary facies belt spatial distribution grid. The spatial coordinate sequence of each fault is traversed, and the sedimentary facies belt to which each coordinate point in the sequence belongs is determined. When a coordinate point enters the coordinate range of another sedimentary facies belt from the coordinate range of a sedimentary facies belt, the coordinate point is recorded as the intersection position of the fault and the sedimentary facies belt, and the unique identification of the sedimentary facies belt before and after the intersection is recorded. For each intersection position, the intersection mode of the fault and the sedimentary facies belt is determined based on the extension direction vector of the fault and the distribution characteristics of the adjacent coordinate points. The fault ID, extension direction vector, all intersection position coordinates, corresponding sedimentary facies belt identification, and intersection mode of each fault are associated and stored to form a fault-sedimentary facies belt association data table. The association data table is integrated into the sedimentary facies belt spatial distribution grid as attribute information.

5. The method according to any one of claims 1 to 4, characterized in that, The correlation between the lithological combination characteristics corresponding to the ore body deposition environment, the structural ore-controlling regularity characteristics corresponding to the geological structure feature parameters, and the physical property anomaly characteristics corresponding to the geophysical data is identified according to the three-dimensional stratum model and the fracture system three-dimensional model, a model structure of the ore body prediction model is constructed by taking the correlation set as a constraint condition, the model structure includes a feature input layer, a correlation analysis layer and a result output layer, the correlation analysis layer embeds the interdependent rules and influence weights in the correlation set, the spatial data of the target exploration area is input into the constructed ore body prediction model, and the spatial coordinates, shape parameters and distribution trend of the concealed ore body are obtained through the correlation analysis layer. The lithological combination information corresponding to the ore body deposition environment is extracted from the three-dimensional stratum model, the lithological combination characteristics are obtained through lithological type statistics and combination sequence analysis, and the lithological combination characteristics include the proportion of lithological types, the arrangement order of combinations and the vertical variation law. The structural properties related to ore control are selected from the geological structure feature parameters, the control of structures on the formation of ore bodies is analyzed in combination with the fracture distribution and extension characteristics in the fracture system three-dimensional model, and the structural ore-controlling regularity characteristics are obtained, which include the fracture ore-controlling range, the correlation between structural intersection relationship and ore bodies. The physical property information related to the ore body is extracted from the geophysical data, the spatial distribution and intensity variation of the physical property anomaly are identified in combination with the abnormal feature region in the deep fracture feature parameters, and the physical property anomaly characteristics are obtained, which include the abnormal spatial coordinates, the intensity gradient and the spatial correlation with fractures. The lithological combination characteristics, the structural ore-controlling regularity characteristics and the physical property anomaly characteristics are standardized to eliminate dimensional differences and converted into feature vectors of a unified dimension, the interdependent relationship and influence weight are identified through feature correlation analysis, and a correlation set is formed. The model structure of the ore body prediction model is constructed by taking the correlation set as a constraint condition, the model structure includes a feature input layer, a correlation analysis layer and a result output layer; wherein the correlation analysis layer embeds the interdependent rules and influence weights in the correlation set; The geological data of known ore bodies are collected as training samples, the feature data in the training samples are input into the ore body prediction model, and the internal parameters of the ore body prediction model are adjusted to complete the construction of the ore body prediction model; The spatial data of the target exploration area is input into the constructed ore body prediction model, and the spatial coordinates, shape parameters and distribution trend of the concealed ore body are obtained through the correlation analysis layer; Based on the spatial coordinates, shape parameters and distribution trend of the concealed ore body, in combination with the spatial structure network of the three-dimensional stratum model, a three-dimensional visual prediction result for the spatial distribution of the concealed ore body is generated, and the three-dimensional visual prediction result is used to present the spatial positional relationship between the concealed ore body and the stratum and fractures.

6. The method of claim 5, wherein, The lithological combination information corresponding to the ore body deposition environment is extracted from the three-dimensional stratum model, the lithological combination characteristics are obtained through lithological type statistics and combination sequence analysis, and the lithological combination characteristics include the proportion of lithological types, the arrangement order of combinations and the vertical variation law, including: Based on the regional geological background and known geological clues, the spatial range of the ore body deposition environment is determined in the three-dimensional stratum model, and the lithological information and spatial coordinate data of all stratum units in the spatial range are extracted; Classify and count the extracted lithology information to determine the lithology types contained in the spatial range and the number of stratigraphic units corresponding to each lithology type, calculate the proportion of the number of each lithology type in the total number of stratigraphic units, and obtain the lithology type proportion data; Sort the stratigraphic units in the spatial range in the order of vertical depth in the spatial coordinate data to form a stratigraphic sequence from shallow to deep, extract the lithology types of each stratigraphic unit, and record the appearance sequence of the lithology types in the order of sorting to obtain an initial lithology combination sequence; Perform sequence analysis on the initial lithology combination sequence to identify the repeated lithology combination fragments and the appearance frequency of the fragments, determine the dominant lithology combination mode, and analyze the conversion mode and conversion frequency of the lithology types in the sequence to determine the combination arrangement sequence characteristics; Construct a lithology vertical distribution map with vertical depth as the horizontal axis and lithology type as the vertical axis, mark the depth interval and thickness corresponding to each lithology type in the lithology vertical distribution map, identify the change characteristics of the lithology types in different depth intervals, and calculate the conversion gradient of the lithology types with depth; Combine the lithology vertical distribution map and the conversion gradient data to mine the change rule of the lithology combination with vertical depth, which includes the vertical zonation characteristics of the lithology types and the dominant lithology types in different depth sections; Combine the lithology type proportion data, the combination arrangement sequence characteristics, and the vertical change rule to generate the lithology combination characteristics.

7. The method of claim 5, wherein, The model structure of the ore body prediction model is constructed by taking the association relationship set as a constraint condition, including: Obtain the input indicators and output indicators of the ore body prediction model, the input indicators are the lithology combination characteristic data, the structure ore-controlling regularity characteristic data, and the physical property anomaly characteristic data of the target exploration area, and the output indicators are the spatial distribution probability and the shape parameters of the concealed ore body; Construct a feature input layer; wherein the feature input layer includes a data preprocessing branch and a feature standardization branch, the data preprocessing branch is used for format unification and noise removal of the input feature data, and the feature standardization branch is used for converting the preprocessed feature data into a standardized feature vector with uniform dimensions; Construct an association analysis layer; wherein the association analysis layer includes a rule storage sub-branch, a weight distribution sub-branch, and an association calculation sub-branch, the rule storage sub-branch is used for storing the interdependent rules in the association relationship set, converting the rules into logical judgment statements executable by the model, the weight distribution sub-branch is used for distributing the influence weight in the association relationship set to the corresponding feature indicators to determine the contribution degree of each feature indicator to the prediction result, the feature indicators with high contribution degree are assigned higher weight coefficients, and the association calculation sub-branch is embedded with the logical judgment statements and the weight coefficients, used for receiving the standardized feature vector from the feature input layer, analyzing the matching of each feature indicator in the feature vector through the logical judgment statements, and calculating the association scores of each feature indicator in combination with the weight coefficients; construct a result output layer; wherein the result output layer comprises a result fusion sub-branch and a format conversion sub-branch, the result fusion sub-branch is used for comprehensively fusing the correlation scores output by the correlation calculation sub-branch to obtain the spatial distribution probability value of the concealed ore body, and the format conversion sub-branch is used for converting the spatial distribution probability value and the corresponding morphological parameters into a standardized output format; The feature input layer, the correlation analysis layer and the result output layer are connected in a data transmission sequence to form the model structure.

8. The method of claim 5, wherein, The spatial data of the target exploration area is input into the constructed ore body prediction model, and the spatial coordinates, morphological parameters and distribution trend of the concealed ore body are obtained through the correlation analysis layer, including: Collecting the spatial data of the target exploration area covering the stratum lithology, geological structure, geophysical exploration and spatial coordinates, and extracting the lithological combination features, structure ore-controlling regularity features and physical property anomaly feature related data according to the input indicators of the ore body prediction model after preprocessing the spatial data to generate the original input data set; The feature input layer of the ore body prediction model is called to complete data standardization to obtain a standardized feature vector, the standardized feature vector is input into the correlation calculation sub-branch of the correlation analysis layer, the interdependent rules stored in the rule storage sub-branch are called by the correlation calculation sub-branch, and the matching relationship between each feature indicator in the standardized feature vector and the ore body is logically judged one by one; The correlation scores are transmitted into the result fusion sub-branch of the result output layer, the single-feature correlation scores are complementarily verified by a multi-feature fusion algorithm to obtain the distribution probability of the concealed ore body at each spatial position of the target exploration area, and the potential area with a concealed ore body existence probability higher than the distribution probability threshold is selected according to the preset distribution probability threshold; The three-dimensional spatial coordinate range of the potential area is located in combination with the spatial coordinate data carried in the standardized feature vector, the morphological parameters are obtained by analyzing the boundary contour, extension length and thickness variation of the potential area through a geological body morphological analysis algorithm, and the spatial positional relationship and morphological correlation of adjacent potential areas are analyzed to obtain the distribution trend of the concealed ore body. comprise:

9. A computer device, comprising: a processor; a storage device having a computer program stored thereon, when the computer program is executed by the processor, the processor implements the three-dimensional space modeling based mineral resource exploration data analysis method according to any one of claims 1-8. The readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the three-dimensional space modeling based mineral resource exploration data analysis method according to any one of claims 1-8.

10. A readable storage medium, characterized by, ​

Images