A copper deposit big data three-dimensional prospecting model construction system and application method

Abstract

The present application relates to the technical field of mineral resources exploration, in particular to a copper deposit big data three-dimensional prospecting model construction system and application method, the construction system comprises: a three-dimensional geological modeling module, a three-dimensional geological modeling technology is adopted, a three-dimensional geological model containing a granite-basalt contact zone structure, alteration zoning and fault structure is constructed, and contact zone three-dimensional structure data is output; an ore-forming prediction module is used for receiving contact zone three-dimensional structure data and multi-source fusion data, using a structure-alteration-mineralization trinity analysis method and a machine learning algorithm to analyze the ore-forming potential, and outputting ore-forming probability distribution data. The present application is used for solving the technical defects of low ore-forming prediction accuracy caused by only single geological body description in traditional copper deposit exploration, finally improving the exploration efficiency and prospecting breakthrough probability of copper deposit under complex ore-forming environment, and providing systematic technical support for deep copper mineral resources exploration.

Description

Technical Field

[0001] This invention relates to the field of mineral resource exploration technology, specifically to a big data three-dimensional prospecting model construction system and application method for copper deposits. Background Technology

[0002] The copper deposit big data 3D prospecting model construction system and application method is a technical system that relies on big data technology and geological exploration theory to realize the reconstruction of the spatial pattern of copper deposit metallogenic deposits, prediction of metallogenic potential, location of concealed ore bodies, and interactive display of exploration results. Its core purpose is to solve problems in traditional copper deposit exploration such as data fragmentation, unintuitive interpretation of metallogenic regularities, and ambiguous location of deep target areas. Especially for complex metallogenic tectonic units such as granite-basalt contact zones, it provides accurate geological model support and scientific decision-making basis for deep copper deposit exploration, helping to improve the exploration efficiency and prospecting breakthrough probability of strategic copper mineral resources, and ensuring the national mineral resource security supply.

[0003] In copper deposit exploration practice, existing technologies have multi-dimensional defects and are difficult to adapt to the exploration needs of complex metallogenic systems. Due to the complexity of contact zone metallogenic systems, relying solely on conventional methods for single anomaly identification and two-dimensional geological volume mapping can lead to problems such as low efficiency, inaccurate mineralization prediction, and inaccurate ore body location when identifying deep ore bodies. There is an urgent need to establish a multidisciplinary exploration model and technical method. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a three-dimensional mineral exploration model construction system and application method for copper deposits based on big data. This system overcomes the technical shortcomings of traditional copper deposit exploration, which relies solely on the characterization of a single geological body, resulting in low accuracy in mineralization prediction. Ultimately, it improves the exploration efficiency and the probability of breakthroughs in copper deposits under complex mineralization environments, providing systematic technical support for the exploration of deep copper mineral resources.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A copper deposit big data three-dimensional prospecting model construction system, comprising a multi-source data integration module, a three-dimensional geological modeling module, a mineralization prediction module, a deep ore body positioning module, and a visualization analysis module, wherein the multi-source data integration module is signal-connected to an external acquisition device, wherein: The multi-source data integration module is used to receive raw geological, geophysical and geochemical data transmitted from external acquisition equipment, process the data using data standardization and fusion algorithms, and output multi-source fused data to the 3D geological modeling module and the deep ore body positioning module. The 3D geological modeling module is used to receive multi-source fusion data, use 3D geological modeling technology to construct a 3D geological model including the granite-basalt contact zone structure, alteration zoning and fault structure, and output the 3D structure data of the contact zone to the mineralization prediction module. The mineralization prediction module is used to receive three-dimensional structural data of the contact zone and multi-source fusion data. It uses a three-in-one analysis method of tectonics-alteration-mineralization and machine learning algorithms to analyze the mineralization potential and outputs the mineralization probability distribution data to the visualization analysis module. The deep ore body positioning module is used to receive geophysical data, use the combined inversion technology of micro-electric sounding and AMT to identify deep anomalies, and output the spatial location data of the concealed ore body to the visualization analysis module. The visualization analysis module receives contact zone 3D structural data, mineralization probability distribution data, and concealed ore body spatial location data. It integrates and displays these data using 3D rendering and interactive visualization technologies, outputting a comprehensive 3D scene for mineralization prediction.

[0006] Furthermore, the multi-source data integration module includes: The geological data receiving unit is used to receive regional geological maps, remote sensing images and geological profile data from external acquisition equipment, and to process them with coordinate unification and format standardization, and output standardized geological data to the data fusion unit. The geophysical data receiving unit is used to receive induced polarization bathymetry data and AMT data from external acquisition equipment. It employs anomaly extraction and filtering to output high polarizability anomaly and low resistivity anomaly data to the data fusion unit. The geochemical data receiving unit is used to receive copper, gold, silver and molybdenum element content data from external acquisition equipment, calculates the cumulative element anomaly index, and outputs geochemical anomaly data to the data fusion unit. The data fusion unit receives standardized geological data, high polarizability anomaly and low resistivity anomaly data, and geochemical anomaly data. It uses multi-source data registration and feature-level fusion algorithms to output multi-source fused data to the 3D geological modeling module and the deep ore body positioning module.

[0007] Furthermore, the 3D geological modeling module includes: The contact zone structure modeling unit is used to receive borehole data and geological profile data from multi-source fusion data, and uses layer inference and solid modeling technology to construct a three-dimensional geological model of the contact zone. It outputs the contact surface morphology and attitude data to the alteration zoning modeling unit and the fault structure modeling unit. The alteration zoning modeling unit is used to receive the spectral data of altered minerals and the morphological data of the contact surface from the multi-source fusion data. It uses spectral discrimination and spatial interpolation methods to construct a three-dimensional distribution model of the silicification-sericite alteration zone and the chloritization-carbonatization zone, and outputs the alteration zoning distribution data to the mineralization prediction module. The fault structure modeling unit is used to receive structural geochemical data and joint statistics from multi-source fusion data, and to construct a three-dimensional geological model of the fault system by using structural surface simulation and intersection analysis. It outputs data of the fault intersections to the mineralization prediction module.

[0008] Furthermore, the mineralization prediction module includes: The structure-alteration-mineralization analysis unit is used to receive fracture intersection data and alteration zoning distribution data from the three-dimensional structural data of the contact zone. It uses the evidence weight method to calculate the structural ore-controlling weight and the correlation degree of alteration and mineralization. When the mineralization favorability exceeds the set threshold, it is determined to be a mineralization favorable area and outputs the mineralization control factor data to the machine learning training unit. The machine learning training unit is used to receive multi-source fused data and mineralization control factor data, use the random forest algorithm to train samples, generate a mineralization prediction model, and output model parameters to the mineralization probability calculation unit. The mineralization probability calculation unit is used to receive model parameters, apply the mineralization prediction model to calculate the mineralization probability, and output the mineralization probability distribution data to the visualization analysis module.

[0009] Furthermore, the structure-alteration-mineralization analysis unit is specifically used for: Receive data on fracture intersections and identify favorable structural locations; Receive alteration zoning distribution data to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Receive mineralization zoning data and determine the mineralization intensity by combining Cu and Au element content; The evidence weight method is used to integrate tectonic, alteration and mineralization factors. When the comprehensive favorable index exceeds the preset threshold, the mineralization control factor data are output.

[0010] Furthermore, the deep ore body positioning module includes: The micro-electric sounding data processing unit is used to receive high-power micro-electric sounding data, and uses the apparent polarizability calculation and anomaly extraction method to output high polarizability anomaly data to the joint inversion unit. The AMT data processing unit is used to receive AMT data, and uses resistivity inversion and low-resistivity anomaly identification methods to output low-resistivity anomaly volume data to the joint inversion unit. The joint inversion unit is used to receive high polarizability anomaly data and low resistivity anomaly data. It adopts a combination of micro-electrical constraint inversion and AMT three-dimensional imaging technology. When the two types of anomalies overlap in space, they are identified as the response of a concealed ore body. The three-dimensional geological model data of the concealed ore body is then output and transmitted to the visualization analysis module.

[0011] Furthermore, the visualization and analysis module includes: The 3D rendering unit is used to receive the 3D structural data of the contact zone, the mineralization probability distribution data and the spatial location data of the concealed ore body, and uses a 3D graphics engine to perform model rendering and lighting processing, and outputs preliminary 3D scene data to the data overlay unit. The data overlay unit is used to receive preliminary 3D scene data, and uses layer overlay and transparency adjustment technology to fuse the mineralization probability distribution and ore body location into the 3D geological model, generate fused scene data and transmit it to the user interaction unit. The user interaction unit is used to receive fused scene data, provide human-computer interaction functions such as querying, zooming, profile cutting and target area delineation, and output interactive commands and visualization results.

[0012] Furthermore, it also includes a model optimization module, which receives user feedback data and new exploration data, and uses incremental learning and model update algorithms. When the deviation between the new data and the original model exceeds the set tolerance, the model optimization process is triggered, and the optimized mineralization prediction model and three-dimensional geological model data are output to the mineralization prediction module and the three-dimensional geological modeling module.

[0013] The above approach has the following beneficial effects: 1. This solution standardizes and integrates geological, geophysical, and geochemical data through a multi-source data integration module. It is no longer limited to the independent analysis or simple overlay of traditional single data, but deeply explores the intrinsic correlation between different data related to mineralization, clearly presenting the coupling relationship of "geological body - physical property anomaly - elemental anomaly". This provides unified and comprehensive data support for subsequent 3D modeling, mineralization prediction, and deep positioning, meeting the needs of contact zone mineralization systems for multi-source information integration.

[0014] 2. This scheme, through the 3D geological modeling module, can fully construct a 3D geological model including the contact zone structure, alteration zoning (silicification-sericification zone, chloritization-carbonatization zone), and fault structures, breaking the limitations of traditional 2D profile modeling or single geological body characterization; the mineralization prediction module combines the three-in-one analysis of "structure-alteration-mineralization" with machine learning algorithms, overcoming the shortcomings of traditional experience-based judgment or single statistical models, more scientifically quantifying the weight of ore-controlling factors, identifying favorable mineralization areas, and making the interpretation and prediction of mineralization regularities more targeted, in line with the characteristics of the "structure-lithology" composite ore-controlling of the contact zone.

[0015] 3. This solution improves the accuracy of locating deep concealed ore bodies and the dynamic adaptability of exploration work. The deep ore body location module adopts micro-electric sounding and AMT joint inversion technology, avoiding the problems of false anomalies and large location errors that are prone to occur in traditional single geophysical methods, and can more accurately locate deep concealed ore bodies. The visualization analysis module makes exploration information more intuitive and decision-making more efficient through integrated display and interactive functions. The model optimization module achieves dynamic updates through incremental learning, eliminating the need for traditional full-process remodeling. It can absorb new exploration data in a timely manner to correct the model and continuously adapt to the complexity of the contact zone metallogenic system and the progressive nature of exploration work.

[0016] Furthermore, an application method for a copper deposit big data three-dimensional prospecting model construction system, based on the aforementioned copper deposit big data three-dimensional prospecting model construction system, includes the following steps: Step S1: Data acquisition and integration. Geological data, geophysical data, and geochemical data are acquired, and preprocessed and fused through a multi-source data integration module to obtain multi-source fused data. Step S2: 3D geological modeling. Using multi-source fusion data, a 3D geological model of the granite-basalt contact zone is constructed through the 3D geological modeling module, including the contact surface morphology, alteration zoning, and fault structure. Step S3: Mineralization prediction and analysis. The mineralization prediction module applies a three-in-one analysis method of structure-alteration-mineralization and machine learning algorithms to analyze multi-source data and generate a mineralization probability distribution map. Step S4: Locating deep ore bodies. Based on geophysical exploration data, the deep ore body locating module uses micro-electric sounding and AMT joint inversion technology to determine the spatial location of concealed ore bodies. Step S5: Visualization and Analysis. The visualization and analysis module renders the 3D geological model, mineralization probability distribution map, and ore body location, providing an interactive analysis interface.

[0017] Furthermore, the integrated analysis method of structure-alteration-mineralization in step S3 includes: Step S31: Analyze the structural ore-controlling characteristics and identify fault intersections as favorable ore-forming structures; Step S32: Analyze the alteration zoning to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Step S33: Analyze the mineralization zoning and assess the mineralization intensity based on the Cu and Au elemental contents; Step S34: Integrate tectonic, alteration, and mineralization factors using the evidence weight method to calculate the mineralization favorability; Step S35: Apply the random forest algorithm to train the multi-source data and generate a mineralization probability distribution map.

[0018] Beneficial effects:

[0019] 1. Compared to the limitations of conventional two-dimensional geological mapping in presenting the three-dimensional structure of the contact zone, the three-dimensional geological model constructed in step S2 can intuitively reproduce the morphology of the granite-basalt contact surface (such as gently wavy or steeply dipping serrated), alteration zoning (potassic feldspar alteration zone, silicification-sericification zone, etc.), and the spatial distribution of fault structures. This three-dimensional presentation method can help explorers clearly identify the "tectonic-lithological" composite ore-controlling characteristics within the contact zone, avoid missing favorable ore-forming areas due to the limitations of the two-dimensional perspective, and provide an accurate spatial framework for subsequent mineralization analysis.

[0020] 2. Step S3's integrated "tectonic-alteration-mineralization" analysis method overcomes the limitations of traditional single-anomaly identification (such as relying solely on geochemical anomalies). It separately identifies fault intersection structures, alteration overlap areas, and high-mineralization areas, then quantifies and integrates various ore-controlling factors using the weight of evidence method, and finally combines the random forest algorithm to achieve intelligent training of multi-source data. This combination of "qualitative analysis + quantitative calculation + algorithm optimization" can more accurately generate mineralization probability distribution maps, effectively delineate high-potential mineral exploration target areas, significantly reduce the blindness of deep exploration, and improve the scientificity and reliability of mineralization prediction.

[0021] 3. Employing a combined micro-electric sounding and AMT (Automated Metal-Mechanical Transmission) inversion technique: Micro-electric sounding can accurately capture signals of sulfide ore bodies with high polarizability near the contact zone, while AMT can clearly invert underground resistivity distribution to reveal fault extension characteristics. The combination of these two techniques overcomes the limitations of single geophysical methods, accurately determining the spatial location of concealed ore bodies. This technological combination effectively solves the problem of ambiguous location of deep ore bodies using traditional methods, significantly improving the success rate of deep copper deposit exploration and providing crucial technical support for deep exploration after the depletion of shallow resources. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the system framework of the copper deposit big data three-dimensional prospecting model construction system and application method of the present invention. Figure 2 This is a schematic diagram illustrating the method steps of an embodiment of the copper deposit big data three-dimensional prospecting model construction system and application method of the present invention. Detailed Implementation

[0023] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0025] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0026] The following detailed description illustrates the specific implementation method: Example 1:

[0027] In traditional exploration models, geological, geophysical, and geochemical data are often processed independently, with "data-level fusion" achieved only through simple overlay. This fails to uncover the intrinsic connections between different data, resulting in the obscuring of the coupling relationship between tectonic ore control and alteration mineralization. 3D geological modeling is often limited to depicting single geological bodies, either constructing only the geometry of the contact zone or simulating alteration distribution separately. It cannot achieve a spatially integrated representation of the contact zone structure, alteration zoning, and fault system, making the spatial configuration relationship of ore-forming elements unclear.

[0028] To address the aforementioned problems, the inventors propose an integrated solution based on multi-source data feature-level fusion and intelligent modeling, as shown in the attached document. Figure 1 The system described is a three-dimensional copper deposit exploration model construction system based on big data. It includes a multi-source data integration module, a three-dimensional geological modeling module, a mineralization prediction module, a deep ore body location module, and a visualization analysis module. The multi-source data integration module is connected to external acquisition equipment. Taking a copper-gold deposit exploration area in a granite-basalt contact zone in southern Hunan as an example, this area covers approximately 5 km², with the surface covered by loose Quaternary sediments. Traditional exploration has discovered three small copper ore bodies, but their deep extension and enrichment patterns are unclear. The implementation process after introducing this system is as follows: In the field, geologists used external acquisition equipment such as drilling rigs, induced polarization (IP) sounders, AMT magnetotellurics, and shortwave infrared spectrometers to simultaneously collect regional geological maps, core data from 50 boreholes, IPI and AMT data from 200 measuring points, and spectral and geochemical data from 500 rock samples. In the initial exploration phase, a 1:50,000 regional geological map and remote sensing imagery were acquired using a drone-borne LiDAR. A high-power IPI sounder (5A power supply) was used to collect IPI data at depths of 0-500m. An AMT instrument (frequency range 1-1000Hz) was used to complete measurements of 20 profiles. Simultaneously, Cu, Au, Ag, and Mo elemental analyses were conducted at 1000 soil sampling points. This data was transmitted in real-time to a multi-source data integration module via an IoT terminal.

[0029] The multi-source data integration module receives raw geological, geophysical, and geochemical data transmitted from external acquisition equipment, processes it using data standardization and fusion algorithms, and outputs multi-source fused data to the 3D geological modeling module and the deep ore body positioning module. The multi-source data integration module includes: The geological data receiving unit receives regional geological maps, remote sensing images, and geological profile data from external acquisition equipment. It uses coordinate unification and format standardization processing to output standardized geological data to the data fusion unit. The geological data receiving unit first processes the input remote sensing images (Landsat8) and geological profile data from 5 exploration lines, converting data from different coordinate systems into the 2000 National Geodetic Coordinate System and standardizing the format to Shapefile format. This solves the splicing error problem caused by the chaotic coordinates and incompatible formats of traditional data.

[0030] The geophysical data receiving unit receives induced polarization (IP) bathymetry (IP) data and analog-to-mass spectrometry (AMT) data from external acquisition equipment. It employs anomaly extraction and filtering to output high-polarization anomalies and low-resistivity anomalies to the data fusion unit. The geophysical data receiving unit uses a moving average method to denoise the IPT data, removes power frequency interference through high-pass filtering, and extracts high-polarization anomalies with an apparent polarization >5%. Occam inversion is performed on the AMT data to identify low-resistivity anomaly regions with resistivity <15 Ω·m. Compared to traditional manual mapping analysis, anomaly extraction efficiency is improved by 3 times and accuracy by 40%.

[0031] The geochemical data receiving unit receives copper, gold, silver, and molybdenum elemental content data from external acquisition equipment. It calculates the cumulative elemental anomaly index and outputs the geochemical anomaly data to the data fusion unit. The geochemical data receiving unit calculates the cumulative elemental anomaly index (Cu×0.4+Au×0.3+Ag×0.2+Mo×0.1), sets the threshold to 8.0, and automatically delineates geochemical anomaly areas, avoiding the subjective bias of traditional manual delineation.

[0032] The final data fusion unit establishes a spatial index based on borehole coordinates. This unit receives standardized geological data, high polarizability and low resistivity anomaly data, and geochemical anomaly data. It employs multi-source data registration and feature-level fusion algorithms to output multi-source fused data to the 3D geological modeling module and the deep ore body positioning module. This is achieved by correlating and matching contact zone outcrop information from standardized geological data, geophysical high polarizability-low resistivity composite anomalies, and geochemical Cu-Au elemental anomalies.

[0033] After the multi-source fused data enters the 3D geological modeling module, this module receives the data, employs 3D geological modeling technology to construct a 3D geological model including the granite-basalt contact zone structure, alteration zoning, and fault structures, and outputs the 3D structure data of the contact zone to the mineralization prediction module. Specifically, the 3D geological modeling module includes: The contact zone structure modeling unit receives borehole data and geological profile data from multi-source fusion data. It uses bedding plane inference and solid modeling techniques to construct a three-dimensional geological model of the contact zone and outputs contact surface morphology and attitude data to the alteration zoning modeling unit and the fault structure modeling unit. The contact zone structure modeling unit extracts core logging data from 20 boreholes (recording the contact zone burial depth of 200-350m) and geological profile information. It uses Kriging interpolation for bedding plane inference and combines boundary representation to carry out solid modeling. The constructed three-dimensional geological model of the contact zone clearly presents the undulating contact surface morphology with dip angles varying between 30-45°. Compared with traditional two-dimensional profile modeling, it achieves a complete restoration of the spatial morphology of the contact zone.

[0034] The alteration zoning modeling unit receives spectral data of alteration minerals and contact surface morphology data from multi-source fusion data. Using spectral discrimination and spatial interpolation methods, it constructs three-dimensional distribution models of silicification-sericite alteration zones and chloritization-carbonatization alteration zones, and outputs the alteration zoning distribution data to the mineralization prediction module. The alteration zoning modeling unit receives sericite and chlorite spectral index data retrieved from Hyperion hyperspectral images. It performs spectral discrimination using support vector machines and then performs spatial interpolation using inverse distance weighting. It successfully constructs the silicification-sericite alteration zone (thickness 20-50m) distributed on the hanging wall and the chloritization-carbonatization zone (thickness 30-60m) on the footwall of the contact zone, and accurately locates the overlapping area of ​​the two alteration zones (width approximately 30-80m). This overlapping area has a 90% agreement with known ore bodies.

[0035] The fault structure modeling unit receives structural geochemical data and joint statistics from multi-source fusion data. It uses structural surface simulation and intersection analysis to construct a three-dimensional geological model of the fault system and outputs data from fault intersection locations to the mineralization prediction module. Based on structural geochemical data showing Au element anomalies within the fault zone, combined with three sets of joint occurrence data measured from borehole cores, the fault structure modeling unit uses B-spline surface fitting technology to simulate structural surfaces. Through intersection analysis, it was discovered that two NNE-trending main faults converge at a depth of 280m in the contact zone. This structural location is a key channel for hydrothermal migration, solving the problem of traditional structural analysis's difficulty in quantifying the location of deep fault intersections.

[0036] The mineralization prediction module receives three-dimensional structural data of the contact zone and multi-source fused data. It employs a three-dimensional analysis method combining tectonics, alteration, and mineralization, along with machine learning algorithms, to analyze mineralization potential. The resulting mineralization probability distribution data is then transmitted to the visualization analysis module. The mineralization prediction module includes: The structure-alteration-mineralization analysis unit is used to receive fracture intersection data and alteration zoning distribution data from the three-dimensional structural data of the contact zone. It uses the evidence weight method to calculate the structural ore-controlling weight and the correlation degree of alteration and mineralization. When the mineralization favorability exceeds the set threshold, it is determined to be a mineralization favorable area and outputs the mineralization control factor data to the machine learning training unit. The structure-alteration-mineralization analysis unit is specifically used for: Receive data on fracture intersections and identify favorable structural locations; Receive alteration zoning distribution data to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Receive mineralization zoning data and determine the mineralization intensity by combining Cu and Au element content; After the mineralization prediction module is started, the structure-alteration-mineralization analysis unit first identifies the fault intersection as a favorable structure (weight assigned 0.35), determines the alteration overlap area as a favorable alteration area (weight 0.4), and combines the Cu content of the borehole core (>2.0% indicates strong mineralization, weight 0.25) to calculate the comprehensive favorable index using the evidence weight method, sets the threshold to 0.7, screens out 3 favorable mineralization areas and outputs the control factor data.

[0037] The evidence weight method is used to integrate tectonic, alteration and mineralization factors. When the comprehensive favorable index exceeds the preset threshold, the mineralization control factor data are output.

[0038] The machine learning training unit is used to receive multi-source fused data and mineralization control factor data, use the random forest algorithm to train samples, generate a mineralization prediction model, and output model parameters to the mineralization probability calculation unit. The machine learning training unit selected 150 samples (50 known mineral deposits and 100 background points) and trained them using a random forest algorithm (100 decision trees, maximum depth 15). The generated mineralization prediction model showed that contact zone distance, alteration overlap, and fault intersection density were the top three mineralization-controlling factors, with a cumulative feature importance of 78%. Data from 100 newly collected sampling points were input into the mineralization probability calculation unit to obtain mineralization probability distribution data. The high-probability area (>0.8) covered approximately 0.5 km², with an 88% match rate with known ore bodies.

[0039] The mineralization probability calculation unit is used to receive model parameters, apply the mineralization prediction model to calculate the mineralization probability, and output the mineralization probability distribution data to the visualization analysis module.

[0040] The deep ore body positioning module receives geophysical data, uses a combined inversion technique of micro-electric sounding and AMT to identify deep anomalies, and outputs the spatial location data of the concealed ore body to the visualization analysis module. The deep ore body positioning module includes: The micro-electric sounding data processing unit is used to receive high-power micro-electric sounding data, and uses the apparent polarizability calculation and anomaly extraction method to output high polarizability anomaly data to the joint inversion unit. The AMT data processing unit is used to receive AMT data, and uses resistivity inversion and low-resistivity anomaly identification methods to output low-resistivity anomaly volume data to the joint inversion unit. The joint inversion unit is used to receive high polarizability anomaly data and low resistivity anomaly data. It adopts a combination of micro-electrical constraint inversion and AMT three-dimensional imaging technology. When the two types of anomalies overlap in space, they are identified as the response of a concealed ore body. The three-dimensional geological model data of the concealed ore body is then output and transmitted to the visualization analysis module.

[0041] The deep ore body positioning module was activated simultaneously. The micro-electric sounding data processing unit calculated the apparent polarizability (ηs=ΔU2 / ΔU1×100%) from the high-power micro-electric sounding data, extracting high-polarization anomalies with an apparent polarizability >8% at depths of 300-400m. The AMT data processing unit identified low-resistivity anomalies with resistivity <10Ω·m in the same depth range through three-dimensional resistivity inversion. The joint inversion unit, using high-polarization anomalies as constraints, employed a combination of micro-electric constraint inversion and AMT three-dimensional imaging technology, discovering that the two types of anomalies completely overlapped at depths of 320-380m, identifying them as responses of a concealed ore body. The output three-dimensional geological model showed that the ore body was lenticular, with a size of approximately 50×80×60m. Subsequent drilling verification revealed a 22m thick copper ore body at a depth of 350m, with an average Cu grade of 1.8% and a positioning error of only 12m. In contrast, the traditional single AMT method had previously encountered two false anomalies in this area, leading to drilling failures.

[0042] The visualization analysis module receives contact zone 3D structural data, mineralization probability distribution data, and concealed orebody spatial location data. It integrates and displays this data using 3D rendering and interactive visualization technologies, outputting a comprehensive 3D scene for mineralization prediction. The visualization analysis module includes: The 3D rendering unit is used to receive the 3D structural data of the contact zone, the mineralization probability distribution data and the spatial location data of the concealed ore body, and uses a 3D graphics engine to perform model rendering and lighting processing, and outputs preliminary 3D scene data to the data overlay unit. The data overlay unit receives preliminary 3D scene data and uses layer overlay and transparency adjustment techniques to fuse the mineralization probability distribution and ore body location into the 3D geological model, generating fused scene data and transmitting it to the user interaction unit. After receiving various types of data, the visualization analysis module uses the OpenGL graphics engine for processing in the 3D rendering unit. Granite is rendered as gray-white, basalt as dark gray, silicification-sericeization zones as light yellow, chloritization zones as dark green, faults as red lines, and concealed ore bodies as translucent golden yellow. Combined with simulated natural light effects, an intuitive preliminary 3D scene is generated. The data overlay unit overlays a mineralization probability heatmap (50% transparency) onto the geological model, achieving the fusion of four layers of information: geological structure, alteration distribution, mineralization probability, and ore body location.

[0043] The user interaction unit is used to receive fused scene data, provide human-computer interaction functions such as querying, zooming, profile cutting and target area delineation, and output interactive commands and visualization results.

[0044] Through the user interaction unit, exploration personnel can zoom in and out of the scene using the scroll wheel, cut profiles along the exploration line, and clearly observe the morphological changes of the ore body at different depths. Using the polygon tool, high-probability target areas can be delineated, and the system automatically calculates estimated resource quantities for those areas. Compared to the traditional method of analyzing more than 10 paper maps, target area delineation time is reduced from 3 days to 4 hours. Furthermore, geological parameters, geophysical anomalies, and geochemical element content at any location can be queried in real time, achieving visualization and efficiency in exploration decision-making.

[0045] It also includes a model optimization module, which receives user feedback data and new exploration data. It uses incremental learning and model update algorithms. When the deviation between the new data and the original model exceeds the set tolerance, it triggers the model optimization process and outputs the optimized mineralization prediction model and 3D geological model data to the mineralization prediction module and the 3D geological modeling module.

[0046] Three months later, ten verification boreholes were completed in the high-probability area, with six encountering mineralization (Cu grade 0.9-2.3%) and four failing to find mineralization. This new data, along with the drilling verification results, was fed back to the model optimization module. The system calculations revealed that the original model's prediction error for deep (>380m) mineralization probability reached 18%, exceeding the 10% tolerance limit, automatically triggering the optimization process. An incremental learning algorithm was used to supplement the original model with ten new samples for training. This eliminated the need to retrain all data, completing the model update in just 24 hours. The optimized model adjusted the contact zone dip angle weight from 0.12 to 0.18, reducing the mineralization probability prediction error to 7%. Subsequently, based on the second target area delineated by the optimized model, drilling revealed an 18m thick copper ore body at a depth of 370m, verifying the effectiveness of the dynamic optimization. Compared to the traditional full-process remodeling (requiring two months), efficiency was improved by 95%.

[0047] Example 2:

[0048] As attached Figure 2 As shown, the difference from Embodiment 1 lies in the application method of a copper deposit big data three-dimensional prospecting model construction system, which, based on the copper deposit big data three-dimensional prospecting model construction system described in Embodiment 1, includes the following steps: Step S1: Data acquisition and integration. Geological data, geophysical data, and geochemical data are acquired, and preprocessed and fused through a multi-source data integration module to obtain multi-source fused data. Step S2: 3D geological modeling. Using multi-source fusion data, a 3D geological model of the granite-basalt contact zone is constructed through the 3D geological modeling module, including the contact surface morphology, alteration zoning, and fault structure. Step S3: Mineralization Prediction Analysis. The mineralization prediction module applies a three-in-one analysis method of tectonics-alteration-mineralization and machine learning algorithms. The three-in-one analysis method of tectonics-alteration-mineralization includes: Step S31: Analyze the structural ore-controlling characteristics and identify fault intersections as favorable ore-forming structures; Step S32: Analyze the alteration zoning to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Step S33: Analyze the mineralization zoning and assess the mineralization intensity based on the Cu and Au elemental contents; Step S34: Integrate tectonic, alteration, and mineralization factors using the evidence weight method to calculate the mineralization favorability; Step S35: Apply the random forest algorithm to train the multi-source data and generate a mineralization probability distribution map.

[0049] By analyzing multi-source data, a mineralization probability distribution map is generated.

[0050] Step S4: Locating deep ore bodies. Based on geophysical exploration data, the deep ore body locating module uses micro-electric sounding and AMT joint inversion technology to determine the spatial location of concealed ore bodies. Step S5: Visualization and Analysis. The visualization and analysis module renders the 3D geological model, mineralization probability distribution map, and ore body location, providing an interactive analysis interface.

[0051] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A system for constructing a three-dimensional prospecting model for copper deposits based on big data, characterized in that, It includes a multi-source data integration module, a 3D geological modeling module, a mineralization prediction module, a deep ore body location module, and a visualization analysis module. The multi-source data integration module is connected to external acquisition equipment. The multi-source data integration module is used to receive raw geological, geophysical and geochemical data transmitted from external acquisition equipment, process the data using data standardization and fusion algorithms, and output multi-source fused data to the 3D geological modeling module and the deep ore body positioning module. The 3D geological modeling module is used to receive multi-source fusion data, use 3D geological modeling technology to construct a 3D geological model including the granite-basalt contact zone structure, alteration zoning and fault structure, and output the 3D structure data of the contact zone to the mineralization prediction module. The mineralization prediction module is used to receive three-dimensional structural data of the contact zone and multi-source fusion data. It uses a three-in-one analysis method of tectonics-alteration-mineralization and machine learning algorithms to analyze the mineralization potential and outputs the mineralization probability distribution data to the visualization analysis module. The deep ore body positioning module is used to receive geophysical data, use the combined inversion technology of micro-electric sounding and AMT to identify deep anomalies, and output the spatial location data of the concealed ore body to the visualization analysis module. The visualization analysis module receives contact zone 3D structural data, mineralization probability distribution data, and concealed ore body spatial location data. It integrates and displays these data using 3D rendering and interactive visualization technologies, outputting a comprehensive 3D scene for mineralization prediction.

2. The copper deposit big data three-dimensional prospecting model construction system according to claim 1, characterized in that, The multi-source data integration module includes: The geological data receiving unit is used to receive regional geological maps, remote sensing images and geological profile data from external acquisition equipment, and to process them with coordinate unification and format standardization, and output standardized geological data to the data fusion unit. The geophysical data receiving unit is used to receive induced polarization bathymetry data and AMT data from external acquisition equipment. It employs anomaly extraction and filtering to output high polarizability anomaly and low resistivity anomaly data to the data fusion unit. The geochemical data receiving unit is used to receive copper, gold, silver and molybdenum element content data from external acquisition equipment, calculates the cumulative element anomaly index, and outputs geochemical anomaly data to the data fusion unit. The data fusion unit receives standardized geological data, high polarizability anomaly and low resistivity anomaly data, and geochemical anomaly data. It uses multi-source data registration and feature-level fusion algorithms to output multi-source fused data to the 3D geological modeling module and the deep ore body positioning module.

3. The copper deposit big data three-dimensional prospecting model construction system according to claim 2, characterized in that, The 3D geological modeling module includes: The contact zone structure modeling unit is used to receive borehole data and geological profile data from multi-source fusion data, and uses layer inference and solid modeling technology to construct a three-dimensional geological model of the contact zone. It outputs the contact surface morphology and attitude data to the alteration zoning modeling unit and the fault structure modeling unit. The alteration zoning modeling unit is used to receive the spectral data of altered minerals and the morphological data of the contact surface from the multi-source fusion data. It uses spectral discrimination and spatial interpolation methods to construct a three-dimensional distribution model of the silicification-sericite alteration zone and the chloritization-carbonatization zone, and outputs the alteration zoning distribution data to the mineralization prediction module. The fault structure modeling unit is used to receive structural geochemical data and joint statistics from multi-source fusion data, and to construct a three-dimensional geological model of the fault system by using structural surface simulation and intersection analysis. It outputs data of the fault intersections to the mineralization prediction module.

4. The copper deposit big data three-dimensional prospecting model construction system according to claim 3, characterized in that, The mineralization prediction module includes: The structure-alteration-mineralization analysis unit is used to receive fracture intersection data and alteration zoning distribution data from the three-dimensional structural data of the contact zone. It uses the evidence weight method to calculate the structural ore-controlling weight and the correlation degree of alteration and mineralization. When the mineralization favorability exceeds the set threshold, it is determined to be a mineralization favorable area and outputs the mineralization control factor data to the machine learning training unit. The machine learning training unit is used to receive multi-source fused data and mineralization control factor data, use the random forest algorithm to train samples, generate a mineralization prediction model, and output model parameters to the mineralization probability calculation unit. The mineralization probability calculation unit is used to receive model parameters, apply the mineralization prediction model to calculate the mineralization probability, and output the mineralization probability distribution data to the visualization analysis module.

5. The copper deposit big data three-dimensional prospecting model construction system according to claim 4, characterized in that, The structure-alteration-mineralization analysis unit is specifically used for: Receive data on fracture intersections and identify favorable structural locations; Receive alteration zoning distribution data to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Receive mineralization zoning data and determine the mineralization intensity by combining Cu and Au element content; The evidence weight method is used to integrate tectonic, alteration and mineralization factors. When the comprehensive favorable index exceeds the preset threshold, the mineralization control factor data are output.

6. The copper deposit big data three-dimensional prospecting model construction system according to claim 5, characterized in that, The deep ore body positioning module includes: The micro-electric sounding data processing unit is used to receive high-power micro-electric sounding data, and uses the apparent polarizability calculation and anomaly extraction method to output high polarizability anomaly data to the joint inversion unit. The AMT data processing unit is used to receive AMT data, and uses resistivity inversion and low-resistivity anomaly identification methods to output low-resistivity anomaly volume data to the joint inversion unit. The joint inversion unit is used to receive high polarizability anomaly data and low resistivity anomaly data. It adopts a combination of micro-electrical constraint inversion and AMT three-dimensional imaging technology. When the two types of anomalies overlap in space, they are identified as the response of a concealed ore body. The three-dimensional geological model data of the concealed ore body is then output and transmitted to the visualization analysis module.

7. The copper deposit big data three-dimensional prospecting model construction system according to claim 6, characterized in that, The visualization and analysis module includes: The 3D rendering unit is used to receive the 3D structural data of the contact zone, the mineralization probability distribution data and the spatial location data of the concealed ore body, and uses a 3D graphics engine to perform model rendering and lighting processing, and outputs preliminary 3D scene data to the data overlay unit. The data overlay unit is used to receive preliminary 3D scene data, and uses layer overlay and transparency adjustment technology to fuse the mineralization probability distribution and ore body location into the 3D geological model, generate fused scene data and transmit it to the user interaction unit. The user interaction unit is used to receive fused scene data, provide human-computer interaction functions such as querying, zooming, profile cutting and target area delineation, and output interactive commands and visualization results.

8. The copper deposit big data three-dimensional prospecting model construction system according to claim 7, characterized in that, Also includes: The model optimization module receives user feedback data and new exploration data. It uses incremental learning and model update algorithms. When the deviation between the new data and the original model exceeds the set tolerance, the model optimization process is triggered, and the optimized mineralization prediction model and 3D geological model data are output to the mineralization prediction module and the 3D geological modeling module.

9. An application method for a copper deposit big data three-dimensional prospecting model construction system, based on the copper deposit big data three-dimensional prospecting model construction system according to any one of claims 1 to 8, characterized in that, Includes the following steps: Step S1: Data acquisition and integration. Geological data, geophysical data, and geochemical data are acquired, and preprocessed and fused through a multi-source data integration module to obtain multi-source fused data. Step S2: 3D geological modeling. Using multi-source fusion data, a 3D geological model of the granite-basalt contact zone is constructed through the 3D geological modeling module, including the contact surface morphology, alteration zoning, and fault structure. Step S3: Mineralization prediction and analysis. The mineralization prediction module applies a three-in-one analysis method of structure-alteration-mineralization and machine learning algorithms to analyze multi-source data and generate a mineralization probability distribution map. Step S4: Locating deep ore bodies. Based on geophysical exploration data, the deep ore body locating module uses micro-electric sounding and AMT joint inversion technology to determine the spatial location of concealed ore bodies. Step S5: Visualization and Analysis. The visualization and analysis module renders the 3D geological model, mineralization probability distribution map, and ore body location, providing an interactive analysis interface.

10. The application method of the copper deposit big data three-dimensional prospecting model construction system according to claim 9, characterized in that, The integrated analysis method of structure-alteration-mineralization in step S3 includes: Step S31: Analyze the structural ore-controlling characteristics and identify fault intersections as favorable ore-forming structures; Step S32: Analyze the alteration zoning to determine the overlapping area of ​​the silicification-sericification zone and the chloritization-carbonatization zone; Step S33: Analyze the mineralization zoning and assess the mineralization intensity based on the Cu and Au elemental contents; Step S34: Integrate tectonic, alteration, and mineralization factors using the evidence weight method to calculate the mineralization favorability; Step S35: Apply the random forest algorithm to train the multi-source data and generate a mineralization probability distribution map.

Images