Three-dimensional metallogenic prediction method and system for skarn type deposit

Abstract

The invention discloses a skarn type deposit three-dimensional metallogenic prediction method, system and equipment and a medium, and relates to the technical field of mineral exploration. The method comprises the following steps: collecting and sorting geological information of a to-be-studied area and a surrounding area; according to the collected information, performing gravity-magnetic three-dimensional physical property inversion and lithology mapping; carrying out gravity-magnetic three-dimensional comprehensive geological modeling by adopting a discrete body inversion method; on the basis of the constructed model, performing three-dimensional metallogenic favorable contrast prediction; constructing a three-dimensional mineralization comprehensive prediction model based on the mineralization favorable degree value; and carrying out target region delineation and verification by using the three-dimensional metallogenic comprehensive prediction model to complete three-dimensional metallogenic prediction under the support of regional gravity and magnetic data. The method has the advantages of wider prediction range, high prediction reliability and the like.

Description

Technical Field

[0001] This invention relates to the field of mineral exploration technology, and in particular to a three-dimensional mineralization prediction method, system, equipment and medium for skarn-type deposits. Background Technology

[0002] With the increasing difficulty of shallow mineral exploration, deep mineral exploration has become one of the main focuses of the geosciences. Three-dimensional metallogenic prediction is of great significance for the search of concealed ore bodies and the development of metallogenic prediction theory, and has become a major highlight in the field of mineral exploration in recent years. Currently, there are no reports, domestically or internationally, on metallogenic prediction of skarn-type deposits based on three-dimensional gravity and magnetic inversion modeling technology. Compared with traditional three-dimensional metallogenic prediction, this project has significant advantages in the field of three-dimensional metallogenic prediction technology for skarn-type deposits. Three-dimensional metallogenic prediction is mainly achieved using borehole data, exploration profiles, and geophysical profiles. However, due to limitations in the amount of data integrated (especially the number of geological and geophysical profiles) and borehole depth, the prediction range and reliability need to be improved. Summary of the Invention

[0003] The technical problem to be solved by this invention is how to provide a three-dimensional mineralization prediction method, system, equipment and medium for skarn-type deposits with a wider prediction range and higher prediction reliability.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a three-dimensional mineralization prediction method for skarn-type deposits, comprising the following steps: S1, collect and organize geological information on the area to be studied and its surrounding areas; S2. Based on the collected information, construct a surface physical property reference model and perform gravity and magnetic three-dimensional physical property inversion and lithological mapping. S3, using discrete volume inversion method for three-dimensional integrated geological modeling of gravity and magnetic fields; S4. Based on the constructed model and the determination and assignment of predictive variables, a mineralization favorability comparison model is established to conduct three-dimensional mineralization favorability comparison prediction. S5, based on the mineralization favorability value, constructs a three-dimensional comprehensive mineralization prediction model; S6 uses a three-dimensional metallogenic comprehensive prediction model to delineate and verify target areas, and completes three-dimensional metallogenic prediction supported by regional gravity and magnetic data.

[0005] This invention also discloses a three-dimensional mineralization prediction system for skarn-type deposits, comprising: The data collection module is used to collect and organize geological information about the area under study and its surrounding areas; The inversion and lithological mapping module is used to perform gravity and magnetic three-dimensional physical property inversion and lithological mapping based on the collected geological information. The gravity and magnetic 3D integrated geological modeling module is used for gravity and magnetic 3D integrated geological modeling using the discrete volume inversion method. The 3D mineralization prediction module is used to predict the 3D mineralization favorability contrast based on the constructed model. The three-dimensional comprehensive prediction model construction module is used to construct a three-dimensional comprehensive metallogenic prediction model based on the mineralization favorability value. The target area delineation and verification module is used to delineate and verify target areas using a three-dimensional comprehensive mineralization prediction model, and to complete three-dimensional mineralization prediction supported by regional gravity and magnetic data.

[0006] The present invention also discloses a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the described three-dimensional mineralization prediction method for skarn-type deposits.

[0007] The present invention also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is used to implement the aforementioned three-dimensional mineralization prediction method for skarn-type deposits.

[0008] The beneficial effects of adopting the above technical solution are as follows: The method described in this application is based on three-dimensional gravity and magnetic joint inversion modeling technology. Under the dual constraints of geology and geophysics, it uses discrete volume inversion methods to construct three-dimensional physical property inversion, comprehensive geological modeling, mineralization favorability correlation analysis, prediction model construction, target area delineation, and geological verification, forming a multi-level prediction of deep-edge mineralization in skarn-type deposits. By combining wavelet analysis, range contrast, and two-dimensional apparent magnetic susceptibility and density imaging analysis results with gravity and magnetic physical property inversion model slices, a large number of two-dimensional geological profiles that conform to geological and physical constraints can be obtained, improving prediction accuracy. By adjusting the size, shape, and physical property parameters of geological model units, the reliability of the model is improved, effectively making up for the shortcomings of conventional skarn-type deposit mineralization prediction, such as limited prediction range and low prediction reliability. It has the advantages of convenient operation, accuracy, and high efficiency. Attached Figure Description

[0009] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0010] Figure 1 This is an overall flowchart of the method described in the embodiments of the present invention; Figure 2a This is a technical roadmap for gravity and magnetic three-dimensional physical property inversion and lithological mapping under prior information constraints in the method described in Embodiment 1 of the present invention; Figure 2b yes Figure 2a Medium physical property-lithology relationship diagram; Figure 3 This is a schematic diagram of the gravity and magnetic three-dimensional integrated geological modeling workflow in the method described in Embodiment 1 of the present invention; Figure 4 This is a flowchart illustrating the process of constructing a three-dimensional mineralization prediction model in the method described in Embodiment 1 of the present invention. Figure 5 This is a diagram showing the three-dimensional gravity inversion result in the method described in Embodiment 2 of the present invention; Figure 6 This is the magnetic three-dimensional inversion result in the method described in Embodiment 2 of the present invention; Figure 7 This refers to the distribution map of the bedrock geological map and modeling profile of the study area in the method described in Embodiment 2 of the present invention; Figure 8 This is a schematic diagram of the construction of a two-dimensional geological profile of the study area in the method described in Embodiment 2 of the present invention; Figure 9 This is a schematic diagram of the 2.5D model of the study area in the method described in Embodiment 2 of the present invention; Figure 10 These are the initial profile model (a) and the modified model profile (b) of the study area in the method described in Embodiment 2 of the present invention. Figure 11 This is the 2.5D model integration diagram in the method described in Embodiment 2 of the present invention; Figure 12 This is a three-dimensional mineralization prediction model diagram of the skarn deposit in the study area in the method described in Embodiment 2 of the present invention; Figure 13 This is a schematic diagram of the system described in Embodiment 3 of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0012] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0013] Example 1 Overall, such as Figure 1 As shown in the figure, this invention discloses a three-dimensional mineralization prediction method for skarn-type deposits, comprising the following steps: S1, collect and organize geological information on the area to be studied and its surrounding areas; S2. Based on the collected information, construct a surface physical property reference model and perform gravity and magnetic three-dimensional physical property inversion and lithological mapping. S3, using discrete volume inversion method for three-dimensional integrated geological modeling of gravity and magnetic fields; S4. Based on the constructed model and the determination and assignment of predictive variables, a mineralization favorability comparison model is established to conduct three-dimensional mineralization favorability comparison prediction. S5, based on the mineralization favorability value, constructs a three-dimensional comprehensive mineralization prediction model; S6 uses a three-dimensional metallogenic comprehensive prediction model to delineate and verify target areas, and completes three-dimensional metallogenic prediction supported by regional gravity and magnetic data.

[0014] The steps described above will be explained in detail below with reference to specific content: Step S1, Data Collection: The system collects and organizes geological data, including geological, basic geological, geochemical, geophysical, geological exploration, and mining engineering data, as well as research data on ore deposits, tectonics, and petrology, for the study area and its surrounding regions. It systematically analyzes and summarizes the occurrence space, ore-controlling factors, metallogenic regularity, and metallogenic model of skarn deposits in the area, and establishes the theoretical basis for three-dimensional metallogenic prediction.

[0015] Step S2, Gravity and Magnetic 3D Property Inversion and Lithological Mapping: S2.1, Prior Information Collection: Collect physical property information, including direct physical property data (such as geophysical well logging data, surface physical property samples, borehole core samples, etc.) and indirect physical property data, that is, physical property information that needs to be converted to obtain, such as surface geological information, geological profiles, borehole data, seismic data, regional gravity and magnetic data, etc. Surface geological information characterizes the distribution of exposed geological bodies; geological profiles and borehole data reveal the distribution range and geometry of geological bodies underground, and the physical property parameters of underground geological bodies can be estimated by analyzing the physical property parameters of existing lithologies; seismic data requires the conversion of velocity into density. Statistical location, mean and extreme values ​​of physical properties constitute a spatial database of physical properties.

[0016] S2.2, Construction of the Surface Property Reference Model: Based on the physical property results obtained in step S2.1, a constrained reference model is constructed using the geological map of the area as a constraint. During modeling, the influence of topography is ignored, and the sizes of the gravity and magnetic meshes are reasonably selected and kept consistent for ease of study. Based on this, the geological map is simplified according to the physical property data, and then physical property units (common values ​​minus background field values) are used to replace surface geological units to construct a surface density difference reference model and a magnetic susceptibility reference model.

[0017] S2.3, Three-dimensional inversion of gravity and magnetic properties: UBC Mag3D and Grav3D were selected as the physical property inversion software. The inversion parameters were set as follows: for uncertain data, the chifact model was used to assign appropriate values ​​to balance the relationship between model smoothness and detail display; depth weighting was used to reduce the influence of the skin effect; and the length-width-depth ratio was set according to the length-width ratio and vertical extension of the study area.

[0018] Based on regional 1:50,000 surface Bouguer gravity and geomagnetic data, data preprocessing was completed. Various potential field separation methods were employed to separate the regional field from the residual field, and the residual anomaly that best matched the actual geological conditions was ultimately selected as the inversion data.

[0019] The surface property reference model established in step S2.2 is used as a constraint. Property values ​​are assigned to the subsurface half-space grid cells under the constraints of property averages, boundaries, and model smoothing factors. For grid cells with determinable properties, their density or magnetic susceptibility is kept constant during the inversion iteration process; for grid cells with determinable property extrema, the inverted properties are limited to not exceeding the upper and lower boundaries during the inversion iteration process; for grid cells with indeterminate properties and upper and lower boundaries, their property boundaries are extrapolated from the nearest grid cell with determinable properties; for grid cells with abrupt property changes such as faults, their smoothing coefficient is reduced. Through multiple iterative inversions, a three-dimensional gravity and magnetic property inversion model is obtained.

[0020] S2.4, Lithological Mapping: Based on the prior information collected in step S2.1, a lithological-physical property relationship study is conducted to establish the physical property combination of rocks (ores) in the area. Logical operations are performed on the gravity and magnetic three-dimensional inversion model obtained in step S2.3. Grid cells that satisfy a certain logical relationship are assigned corresponding lithological codes. Note that when there are regular differences between the density and magnetic susceptibility inverted in step S2.3 and the measured data, difference correction is performed. Finally, lithological mapping is achieved, such as... Figure 2a As shown, Figure 2b yes Figure 2a Enlarged diagram of the physical properties-lithology relationship.

[0021] Step S3, Three-dimensional integrated geological modeling of gravity and magnetic fields: The discrete volume inversion method is used for modeling. The overall idea is to construct a 3D model by assembling 2.5D geological cross-sections. The modeling process is as follows: Figure 3 As shown.

[0022] S3.1, Modeling Region Definition: First, determine the modeling plane range and depth; second, determine the spacing and position of the two-dimensional profiles, which are usually consistent with the spacing of the exploration profiles in the area, and the orientation is also consistent with the orientation of the exploration profiles, perpendicular to the direction of the main structural lines in the area.

[0023] S3.2, Prior Information Processing: This mainly includes simplifying surface geological units, analyzing the correspondence between lithology and physical properties, data preprocessing (such as editing, gridding, filtering, and local field separation), and interpreting seismic profiles. Appropriate simplification of geological units can reduce the difficulty of inversion simulation, especially in areas with complex geological structures. Borehole information provides the boundary depths of major deep stratigraphic units, which are generally kept unchanged as important constraints in gravity and magnetic inversion. Separation of regional fields and local anomalies is crucial in this stage; the separated local anomalies will serve as the basis for evaluating the model's rationality.

[0024] S3.3, Construction of a two-dimensional geological model: Based on the profiles determined in step S3.1, and drawing upon the analysis of existing geological and borehole data, combined with wavelet analysis, range contrast analysis, and two-dimensional apparent magnetic susceptibility and density imaging analysis results, all two-dimensional geological profiles of the modeling area are sequentially inferred and drawn. The construction of these profiles in this step determines the accuracy of the work and the general geological structural framework within the area. It also provides an initial two-dimensional geological model for subsequent three-dimensional gravity and magnetic inversion, making it the most crucial part of the entire process.

[0025] Wavelet analysis is used to roughly delineate the locations of deep geological bodies with differences in physical properties and to determine their approximate distribution. Due to the numerous interference factors affecting magnetic methods, gravity wavelet profile analysis is the primary focus. The range contrast method can highlight weak anomalies at depth and weaken irrelevant anomalies at shallow depth, improving the ability to identify weak and slow-moving anomalies at deep sources and enabling the discovery of potentially hidden geological bodies at depth. Apparent magnetic susceptibility imaging and apparent density imaging methods mainly utilize their inversion functions to infer the location of deep, high-magnetic-density geological bodies. Gravity and magnetic inversion two-dimensional profiles can effectively display the approximate distribution characteristics of geological bodies within a certain subsurface area.

[0026] S3.4, 2.5D / 3D inversion modeling: Discrete modeling methods were used with Encom ModelVision Pro software.

[0027] 2.5D Gravity and Magnetic Fitting and Profile Correction: Based on the two-dimensional geological profile map formed in step S3.3, the basic parameters of the model, such as dip angle, azimuth, width, background density, background magnetic susceptibility, and depth, are first set. Then, different model bodies are used to represent different geological units, and each model body is assigned corresponding density and magnetic susceptibility values. For the same geological unit with significant differences in physical properties in different regions, it is divided into multiple parts and assigned values ​​separately. Forward modeling is then performed on the two-dimensional geological profile. The model is repeatedly modified until the curve fitting is satisfactory. The range of modification to the physical properties and spatial morphology of the model body during modification is determined by geological rationality.

[0028] 3D Inversion Simulation: The 2.5D models of all cross-sections are integrated into a 3D model in Modelvision software according to their spatial location. When the different cross-sections are combined into a whole, the theoretical anomalies of the model are calculated and compared with the actual anomalies. For areas with large fitting errors, the model is returned to the 2.5D cross-sections to modify the geological body morphology, etc. At this time, the fitting degree of gravity and magnetic signals of all cross-sections will change accordingly. The model is then compared with the actual gravity and magnetic anomalies, and adjustments are made to different cross-sections again until a satisfactory fitting anomaly signal is obtained.

[0029] S3.5, Geological Model Visualization: The inversion simulation results profile from step S3.4 are imported into GeoModeller modeling software, and a three-dimensional integrated geological model is formed through interpolation. This model is then displayed using visualization software, allowing for observation of each geological unit from different angles. This helps to better analyze the three-dimensional spatial morphology and distribution patterns of the geological units.

[0030] Step S4, Three-dimensional mineralization prediction: S4.1, Prediction Method: The Surpac 3D software is used to convert geological entity models into 3D block models. The feature analysis method is used for 3D mineralization prediction. That is, by establishing a feature model of the deposit, the purpose of predicting this type of deposit is achieved. The basic principle is the principle of similarity analogy, that is, similar geological conditions have similar mineral deposit distributions.

[0031] S4.2, Block Model Construction: The gravity and magnetic 3D integrated geological model is discretized into block units of the same size. The size of the block units meets the requirements of the feature analysis method and takes into account the scale of most deposits in the area and the completeness of the metallogenic geological elements. That is, each cell must contain at least the complete metallogenic geological conditions of typical deposits in the area. Then, interpolation calculations are performed based on the outline of the solid model or spatial data source to assign various attribute values ​​to the block units, such as grade, type, mineralization and alteration. Subsequently, digital methods are used to calculate the attribute values ​​of the block units, so that the key spatial information within all block units is quantified and assigned. These regular block unit sets constitute the block model.

[0032] S4.3, Selection of predictor variables and determination of weighting coefficients: Based on summarizing the metallogenic regularities of ore deposits and identifying ore-controlling factors, spatial analysis of ore-controlling factors such as strata, faults, magma, and geophysical and geochemical anomalies with known ore deposits (points) is conducted. Specifically, by analyzing the correlation between various ore-controlling factors and known ore deposits (points), the optimal predictive variables are ultimately selected. On this basis, the block cells representing the locations of ore deposits (points) within the area are used as the prospecting model units. The predictive variables within each model unit are binarized, with 1 representing the presence of a predictive variable and 0 representing its absence. Weight coefficients of the predictive variables are obtained using mathematical analysis methods such as geostatistics. The larger the weight coefficient of a predictive variable, the greater its contribution to mineralization.

[0033] S4.4, Establishment of the mineralization favorability comparison model and analysis of the mineralization favorability of the prediction unit: The mineralization favorability comparison model includes model units and prediction units. Model units are block units with known mineral resource characteristics, while prediction units have unknown resource characteristics. Generally, block units within blocks with high shallow exploration levels and relatively detailed prediction elements are selected as model units. Parameters of various prediction elements of the model units are extracted and calculated to establish the mineralization favorability comparison model.

[0034] Based on the calculation results of the predicted variables and weighting coefficients in step S4.3, the connection degree value of each unit is calculated using the unit connection degree (mineralization favorability) calculation formula. This value is a measure of the similarity between the predicted unit and the known model units, indicating the likelihood of mineral exploration in the unit. The connection degree value is used as the spatial data source for the mineralization favorability block model. Each block unit contains the predicted variable value and the connection degree value. The inverse power law of distance is used to estimate the mineralization favorability attribute of the predicted block. The larger the value, the higher the mineralization favorability of the predicted unit.

[0035] Step S5, construction of a three-dimensional comprehensive prediction model, such as Figure 4 As shown: Based on the mineralization favorability values ​​assigned to the block units in step S4.4, the mineralization favorability of each block in the 3D prediction block model is graded and color-coded. The numerical intervals are designed to meet the grading requirements without causing excessive data redundancy. The colors corresponding to the values ​​from smallest to largest are from blue to red, resulting in a color block map of the correlation level. The more red the color, the larger the value, and the greater the probability of mineralization in the corresponding block unit. This is used to construct the final 3D mineralization prediction model.

[0036] Step S6, Target delineation and verification: In step S5, blocks with a reddish tint in the color map are those with a higher probability of mineralization. The prediction threshold needs to be repeatedly adjusted to delineate the target area. When a range greater than a certain mineralization favorability value contains the vast majority (over 80%) of known mineral deposits (points) and has the smallest predicted volume, that mineralization favorability value is used as the prediction threshold. Blocks greater than or equal to this threshold are initially selected as target area blocks. Special attention must be paid to the consistency between the pre-delineated target area and the actual geological conditions, and the integrity of the target area's morphology should be maintained as much as possible. The target area is delineated through multiple comparative analyses. The target areas are then classified according to a relative classification method, and the final target area delineation is completed.

[0037] The delineated target areas were verified using methods such as field geological surveys, UAV geomagnetic surveys, induced polarization gradient profile measurements, high-density electrical resistivity profile measurements, and sample testing and analysis. Key target areas were selected for verification work such as trenching and drilling.

[0038] Through steps S1 to S6, three-dimensional mineralization prediction based on three-dimensional gravity and magnetic inversion modeling technology supported by regional gravity and magnetic data was completed, achieving the expected goal of "exploring the deep part, exploring the edge part, connecting the parts, and cultivating new areas". A multi-level three-dimensional mineralization prediction method system for skarn-type deposits was constructed, which includes gravity and magnetic three-dimensional physical property inversion, gravity and magnetic comprehensive geological modeling, prediction model construction and target area delineation, and geological verification.

[0039] The method described in this invention is based on three-dimensional gravity and magnetic joint inversion modeling technology. Under the dual constraints of geology and geophysics, it employs discrete volume inversion methods to construct a multi-level mineralization prediction system for the deep edges of skarn-type deposits, including three-dimensional physical property inversion, comprehensive geological modeling, mineralization favorability correlation analysis, prediction model construction, target area delineation, and geological verification. By combining wavelet analysis, range contrast analysis, and two-dimensional apparent magnetic susceptibility and density imaging analysis results with gravity and magnetic physical property inversion model slices, a large number of two-dimensional geological profiles conforming to geological and physical constraints can be obtained, improving prediction accuracy. The reliability of the model is improved by adjusting the size, shape, and physical property parameters of the geological model units, effectively overcoming the shortcomings of conventional skarn-type deposit mineralization prediction, such as limited prediction range and low prediction reliability. It has advantages such as being environmentally friendly, easy to operate, accurate, and efficient.

[0040] Example 2 Taking a skarn-type copper-gold polymetallic deposit as an example, the three-dimensional mineralization prediction method for skarn-type deposits of this invention is used for three-dimensional mineralization prediction. The specific steps are as follows: Step S1 involves systematically collecting and organizing previous geological data, including regional geology, mining area geology, geochemistry, geophysics, geological exploration, and mining engineering data, as well as research data in ore deposit geology, tectonics, and petrology. Based on various original data, maps, engineering data, test data, and geological reports, different types of data are categorized and organized, and the data is vectorized and standardized to achieve multi-source data acquisition. The system analyzes and summarizes the occurrence space, ore-controlling factors, metallogenic regularities, and metallogenic models of skarn deposits within the area, establishing a theoretical basis for three-dimensional metallogenic prediction.

[0041] Step S2 involves conducting mapping experiments for gravity and magnetic three-dimensional constrained inversion, following the approach of "prior information collection - construction of surface physical property reference model - construction of gravity and magnetic three-dimensional physical property inversion model - lithological mapping".

[0042] S2.1, Prior Information Collection The collected prior geological information mainly includes the following types: surface geological information, borehole data, geological profiles, and physical property data. Data from 134 boreholes and 884 sets of geological body magnetic susceptibility, density, and other physical property parameters (339 sets for the study area and 545 sets for the region) were collected as important prior information and were fully utilized in the gravity and magnetic property inversion work, providing constraints and a foundation for subsequent reference model construction.

[0043] S2.2 Construction of Surface Reference Model Based on the physical property data from step S2.1, statistical analysis of physical properties was performed. The statistical results show that the physical properties of different lithologies in the study area are as follows: sedimentary rocks such as sandstone, shale, limestone, and dolomite have relatively stable densities and magnetic susceptibility, with an average density of 2.64-2.73 g / cm³. 3 The average magnetic susceptibility is generally close to 100×4π×10. -6 SI, maximum magnetic susceptibility is less than 300 × 4π × 10 -6 SI rocks, overall, exhibit medium density and low magnetic characteristics; the Neoarchean Taishan Group metamorphic rocks, such as amphibolite and tremolite schist, exhibit high density and strong magnetic properties, with densities ranging from 2.831 to 3.134 g / cm³. 3 Between these values, the average was 3.02 g / cm³. 3 2.94 g / cm 3 The average magnetic susceptibility is 348×4π×10. -6 SI, 1299×4π×10 -6 SI can reach a maximum of 1370×4π×10 -6 SI, 3084×4π×10 -6 SI; Mesozoic quartz diorite porphyry exhibits low density and low magnetic properties, with a density ranging from 2.588 to 2.605 g / cm³.3 Between these values, the average magnetic susceptibility is only 194 × 4π × 10⁻⁶. -6 SI; Mesozoic diorite density 2.793-2.877 g / cm³ 3 The average value was 2.84 g / cm³. 3 The magnetic susceptibility is between 1053 and 3883 × 4π × 10 -6 Between SI, it belongs to a medium-to-high density, medium-to-high magnetic geological body; the Neoarchean monzogranite is a low-density, low-magnetic body, with an average density of only 2.56 g / cm³. 3， The average total magnetic susceptibility is 118 × 4π × 10⁻⁶. -6 SI; iron-bearing rocks such as magnetite quartzite and magnetite exhibit ultra-high density (extremely) strong magnetic characteristics, with an average density greater than 3.4 g / cm³. 3 The average magnetic susceptibility of magnetite quartzite is 89311×4π×10. -6 SI, the average magnetic susceptibility of magnetite is greater than 100000 × 4π × 10 -6 SI.

[0044] A constrained reference model was constructed using geological information constrained by a 1:50,000 geological map of the area. The density and magnetic susceptibility mesh size for the study area was 250m×250m×100m. First, the geological map was simplified based on physical property data. Then, physical property units (common values ​​minus background field values) were used to replace surface geological units, and the corresponding physical property parameters were assigned to the established mesh to create a reference model for surface density and magnetic susceptibility, and a reference model for surface physical properties (density difference, magnetic susceptibility).

[0045] S2.3, Construction of a three-dimensional gravity and magnetic property inversion model: The reference model established in step S2.2 was used as the physical property constraint. UBC Mag3D and Grav3D were selected as the physical property inversion software. The inversion parameters were: for uncertain data, the chifact model was used, with the chifact value set to 1; the aspect ratio was 2:2:1. Based on the regional 1:50,000 surface Bouguer gravity data and geomagnetic data, by comparing the original gravity (magnetic) anomaly, matched-filter residual gravity (magnetic) anomaly, third-order trend residual gravity (magnetic) anomaly, and moving average residual gravity (magnetic) anomaly, it was found that the residual anomaly obtained by moving average was in good agreement with the actual geological conditions. Therefore, the residual anomaly obtained by moving average was used as the constraint inversion data.

[0046] Under the constraints of average values, boundaries, and model smoothing factors, physical property values ​​are assigned to the subsurface half-space grid cells. During the inversion iteration process, the density or magnetic susceptibility of grid cells with determinable properties is kept constant; for grid cells with determinable property extrema, the inverted properties are restricted to not exceeding the upper and lower boundaries; for grid cells with indeterminate properties, their property boundaries are extrapolated from the nearest grid cell with determinable properties. Through multiple iterations, a three-dimensional gravity and magnetic property inversion model is obtained. The property models are colored, and surface gravity and magnetic anomalies and geological maps are inserted into the gravity and magnetic three-dimensional property inversion data at different depths for convenient analysis. The gravity three-dimensional inversion results are shown below. Figure 5 As shown, the three-dimensional inversion results of the magnetic method are as follows: Figure 6 As shown.

[0047] S2.4, Lithological Mapping: By combining the lithology-physical property correspondence, logical operations are performed on the inverted 3D data volume, and grid cells that satisfy a certain logical relationship are assigned corresponding lithology codes, thereby achieving lithology mapping. For example, the operational expression for the high magnetic susceptibility and high density metamorphic rock series of the Taishan Group is as follows: Old strata=if (((grav>=0.05)&(0.025>mag>=0.0025)), 4, 0) In the formula: Old strata represents the data volume of the Taishan rock group, and grav and mag represent the density difference and magnetic susceptibility obtained from the inversion, respectively.

[0048] According to this expression, when both grav and mag satisfy the "if" condition, the code 4 is returned to Old strata; otherwise, the value 0 is returned. After the logical operation is completed, the portion of the data body mafic with a value equal to 4 represents the Taishan rock formation.

[0049] Similarly, by applying a similar expression to perform logical judgments on each lithological combination, three-dimensional data volumes of different lithologies can be obtained, enabling three-dimensional lithological mapping. It is worth noting that, due to limitations of the inversion method, the inverted density and magnetic susceptibility are often lower than the measured data. Therefore, it is necessary to compare the inversion results with known profiles to determine the difference, and then add this difference to the inversion results. This allows for lithological identification and mapping using the aforementioned expression.

[0050] The lithological mapping results of the study area are as follows: low magnetic density mainly corresponds to Neoarchean monzogranite and Mesozoic quartz diorite porphyry; low magnetic medium density mainly represents sedimentary rocks such as limestone, sandstone, shale, and dolomite; medium-high magnetic medium-high density bodies mainly identify metamorphic rocks of the Taishan Group such as amphibolite and tremolite schist, as well as Mesozoic diorite, which are mainly distributed around high magnetic high density bodies, indicating a close relationship with ore bodies; high magnetic high density mainly corresponds to magnetite and magnetite quartzite.

[0051] The extracted physical property model, based on logical operations, fits well with actual geological conditions, and the lithological mapping is relatively accurate. Therefore, the obtained three-dimensional physical property model can serve as an important basis for subsequent gravity and magnetic three-dimensional joint inversion modeling and three-dimensional mineralization prediction.

[0052] Step S3: Based on the data collected in step S1 and the lithological mapping test results obtained from the gravity and magnetic three-dimensional constraint inversion in step S2, construct a comprehensive three-dimensional geological model of gravity and magnetic fields.

[0053] S3.1, Modeling Region Definition: The modeling plane extends over the study area, which is approximately 890 km². 2 The depth ranges from the surface to 2.7 km below ground. Most geological bodies and main faults in the area strike approximately 130°. Thirty-six modeling profiles are proposed, with an azimuth of 50°NE and lengths ranging from 13.8 to 52.5 km. The spacing between profiles is mostly 1 km, with some line spacing slightly adjusted according to actual geological conditions for more accurate anomaly fitting. The bedrock geological map and modeling profile distribution of the study area are shown below. Figure 7 As shown.

[0054] S3.2, Prior Information Processing: This mainly includes simplifying surface geological units, integrating and unifying geological boundaries, geological codes, and legends of various map sheets according to standards; recording and editing borehole locations, occurrences, lithologies, and grades for existing borehole data such as borehole columnar sections and exploration line profiles, and compiling four types of statistical tables: Assay, Collar, Lithology, and Survey. The deep mineral exploration borehole database is then constructed in Surpac software to achieve three-dimensional visualization; establishing lithology-physical property correspondences (see steps S2.2 and S2.4); and preprocessing gravity and magnetic data (see step S2.3).

[0055] S3.3, Construction of Two-Dimensional Geological Model Based on the profile determined in step S3.1, and under the constraints of existing geological and borehole data, using the 3D profile of the gravity and magnetic three-dimensional inversion property model constructed above, combined with wavelet analysis, range contrast, and two-dimensional apparent magnetic susceptibility and density imaging analysis results, all two-dimensional geological profiles of the modeling area are sequentially inferred and drawn, such as... Figure 8 As shown.

[0056] S3.4, 2.5D / 3D Inversion Modeling Based on the two-dimensional geological profile map formed in step S3.3, modeling was performed using Encom ModelVision Pro software.

[0057] The basic parameters of the 2.5D model are as follows: model tilt angle 0°, azimuth angle 50°NE, the initial width of the 36 model lines involved in the inversion is set at 5000 m, and the background density is 2.67 g / cm³. 3 The background magnetic susceptibility is 0 SI, and the depth is from the surface to 2700 m underground. The existing gravity and magnetic data range is used as the boundary (see [link]). Figure 9 Then, based on the physical properties corresponding to different lithologies, each model body is assigned corresponding density and magnetic susceptibility parameters. Forward modeling calculations are performed on the two-dimensional geological profiles, and the models are repeatedly modified until the curve fitting is satisfactory, thus completing the 2.5D model construction (see...). Figure 10 In the figure: 1. Cover layer excluding Mantou Formation; 2. Mantou Formation; 3. Taishan Group; 4. Mesozoic quartz diorite porphyry; 5. Mesozoic diorite; 6. Neoarchean intrusive rock mass; 7. Ore body; 8. Surface measured signal; 9. Gravity fitting signal; 10. Magnetic fitting signal.

[0058] All 2.5D models of the cross-sections were integrated into a 3D model in Modelvision software according to their spatial location. When the different cross-section sets were combined, the theoretical anomalies of the model were calculated and compared with the actual anomalies. Areas with large fitting errors were identified by returning to the 2.5D cross-sections to modify the geological body morphology, etc., until a satisfactory anomaly signal was obtained. The integrated 2.5D model diagram is shown below. Figure 11 As shown.

[0059] S3.5 geological model visualization: Step S3.4 generates a three-dimensional integrated geological model, which is then displayed using visualization software. This allows for the acquisition of model slices at different angles and levels, facilitating the observation of various geological units from different perspectives and aiding in the better analysis of the three-dimensional spatial morphology and distribution patterns of the geological units.

[0060] Step S4: Based on the 3D integrated geological model constructed in step S3, the feature analysis method is adopted to carry out three-dimensional mineralization prediction research according to the idea of ​​"constructing block model - selecting prediction variables and determining weight coefficients - establishing mineralization favorability comparison model - mineralization favorability analysis of prediction unit".

[0061] S4.1, Block Model Construction: Following the prediction method requirements and the principle of defining block unit size, the 3D integrated geological model was discretized into block units of the same size. The block grid size was 250 m × 250 m × 100 m, consistent with the gravity and magnetic grid subdivision size used in the 3D gravity and magnetic property inversion. Furthermore, the block size in the x and y directions was 1 / 4 of the exploration line spacing. Then, borehole and other attribute data were used to assign geological attributes to each block grid. For geological bodies with single attribute values ​​(such as stratigraphy and lithology), a direct assignment method was used; for geological bodies whose attribute values ​​varied with spatial location (such as grade), interpolation was used. A total of over 420,000 block units were generated across the entire area, completing the block model construction.

[0062] S4.2, Selection of predictor variables and determination of weighting coefficients: Based on summarizing the metallogenic regularities of the deposits and identifying the ore-controlling factors, six predictive variables were selected: the Cambrian interstratal structural zone, the unconformity between the carbonate caprock and the Taishan Group, the fault buffer zone, the Mesozoic magmatic rock buffer zone, the Au-Ag-Cu-Pb-Zn geochemical anomaly, and the density difference value from the gravity three-dimensional density difference inversion model. Furthermore, the cells containing the locations of gold polymetallic deposits (points) and ore-bearing boreholes within the area were used as prospecting model units. The predictive variables within each model unit were binarized using a "0 / 1" method to obtain the weighting coefficients of the predictive variables.

[0063] Table 1 Weighting coefficients of predictive variables for skarn deposits in the study area

[0064] S4.3 Establishment of the mineralization favorability comparison model and analysis of mineralization favorability of prediction units Select block units with high shallow exploration levels and existing mineral deposits (points) as model units, extract and calculate various predictive element parameters of them, and use them as the basis for mineralization favorability comparison to establish a mineralization favorability comparison model.

[0065] Based on the calculation results of the predicted variables and weighting coefficients in step S4.3, the connectivity value of each unit is calculated using the unit connectivity degree (ore-forming favorability) calculation formula. The mathematical formula is as follows: Let there be m variables xj (j=1,2,3,…,m), n model units, and the value of the j-th variable in the i-th unit is xij (i=1,2,3,…,n; j=1,2,3,…,m). The original data matrix is ​​X, and the matching coefficient matrix is ​​E.

[0066] The correlation between variables is measured using the matching coefficient between them. The matching coefficient r between variables k and j is... kj The calculation formula is:

[0067] Similarly, by calculating the unit connectivity y of unit i... i It is possible to evaluate the mineralization potential of a unit.

[0068]

[0069] Based on the above formula for calculating the unit connectivity, the connectivity of all predicted units is calculated, and then the mineralization favorability of the predicted units is ranked to complete the mineralization favorability analysis.

[0070] Step S5, Construction of the three-dimensional integrated prediction model: Based on step S4, the mineralization favorability of the block is assigned according to the magnitude of the connectivity value, with a numerical interval of 0.1. Then, the mineralization favorability values ​​are graded and color-coded, with the colors ranging from blue to red from smallest to largest, resulting in a connectivity level color block map. This is used to construct the final three-dimensional mineralization prediction model, as shown below. Figure 12 As shown.

[0071] Step S6, Target delineation and verification: By continuously trying to adjust the mineralization favorability value, it was found that when the mineralization favorability is 0.6, more than 80% of the known skarn deposits (points) can be included in the range greater than 0.6 and the predicted volume is the smallest. Therefore, the mineralization favorability of 0.6 is used as the mineralization prediction threshold, and the number of mineralized favorable blocks is 11,943, accounting for about 2.6% of the total.

[0072] Based on the above-mentioned favorable mineral exploration blocks, the target area should be delineated by comprehensively considering the following principles: ① It must conform to the metallogenic geological conditions and spatial distribution patterns of the skarn deposits within the designated area; ② It should pay attention to the degree of consistency between the preliminary predictions of the model and the actual geological conditions; ③ The outline of the mineral exploration target area should be determined according to the principle of minimum volume and maximum ore content; ④ The target area should reflect the principle of independence as much as possible; ⑤ The integrity of the target area morphology should be maintained as much as possible.

[0073] Based on a comprehensive comparative analysis of geological, geophysical, and geochemical information, four prospecting target areas were ultimately identified. Considering factors such as geology, existing mineral deposits, and geophysical and geochemical anomalies within the target areas, and taking into account the purpose of prediction, the prospecting target areas were further divided into one Class A target area and three Class B target areas.

[0074] The designated key target area (Category A) was verified through geological route surveys, 1:10,000 UAV geomagnetic measurements, induced polarization gradient profiling, high-density electrical resistivity tomography (EPR) profiling, and sample testing and analysis. Surface mineralization was found within the target area, and analysis of collected samples revealed high-grade Au, Ag, Zn, and Mn. Geophysical surveys also showed anomalies. Multiple methods confirmed that the target area possesses promising skarn mineral exploration potential.

[0075] Through steps 1 to 6, a three-dimensional mineralization prediction method system based on 3D gravity and magnetic joint inversion technology is finally established, achieving the expected goal of "exploring deep parts, exploring edges, connecting, and cultivating new areas", and demonstrating good mineral exploration results in the geological verification target area.

[0076] The method described in this application, based on the three-dimensional mineralization prediction approach and practice of 3D gravity and magnetic joint inversion physical property modeling, comprehensive geological modeling, three-dimensional mineralization prediction and model construction, target area delineation, and joint verification by multiple geological methods, establishes a multi-disciplinary and multi-level analytical method system for the identification and prediction of skarn-type concealed mineralization. It effectively compensates for the limitations of conventional three-dimensional prediction in terms of prediction range and accuracy for this type of deposit, and can provide an effective basis and solution for deep mineral exploration and prediction of skarn-type deposits.

[0077] Example 3 like Figure 13 As shown, correspondingly, this embodiment of the invention also discloses a three-dimensional mineralization prediction system for skarn-type deposits, including: The data collection module is used to collect and organize geological information about the area under study and its surrounding areas; The inversion and lithological mapping module is used to perform gravity and magnetic three-dimensional physical property inversion and lithological mapping based on the collected geological information. The gravity and magnetic 3D integrated geological modeling module is used for gravity and magnetic 3D integrated geological modeling using the discrete volume inversion method. The 3D mineralization prediction module is used to predict the 3D mineralization favorability contrast based on the constructed model. The three-dimensional comprehensive prediction model construction module is used to construct a three-dimensional comprehensive metallogenic prediction model based on the mineralization favorability value. The target area delineation and verification module is used to delineate and verify target areas using a three-dimensional comprehensive mineralization prediction model, and to complete three-dimensional mineralization prediction supported by regional gravity and magnetic data.

[0078] The specific implementation steps of the modules in the system described in this invention can be referred to the corresponding steps in Embodiment 1, and will not be repeated here.

[0079] Example 4 Accordingly, this invention also discloses a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the three-dimensional mineralization prediction method for skarn-type deposits.

[0080] Example 5 Accordingly, embodiments of the present invention also disclose a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is used to implement the three-dimensional mineralization prediction method for skarn-type deposits.

[0081] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A three-dimensional mineralization prediction method for skarn-type deposits, characterized in that... Includes the following steps: S1, collect and organize geological information on the area to be studied and its surrounding areas; S2. Based on the collected information, construct a surface physical property reference model and perform gravity and magnetic three-dimensional physical property inversion and lithological mapping. S3, using discrete volume inversion method for three-dimensional integrated geological modeling of gravity and magnetic fields; S4. Based on the constructed model and the determination and assignment of predictive variables, a mineralization favorability comparison model is established to conduct three-dimensional mineralization favorability comparison prediction. S5, based on the mineralization favorability value, constructs a three-dimensional comprehensive mineralization prediction model; S6 uses a three-dimensional metallogenic comprehensive prediction model to delineate and verify target areas, and completes three-dimensional metallogenic prediction supported by regional gravity and magnetic data.

2. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S1 specifically includes the following steps: The system collects and organizes basic geological, geochemical, geophysical, geological exploration, mining engineering, mineral deposit geology, tectonics, and petrology research data of the study area and its surrounding areas, and systematically analyzes and summarizes the occurrence space, ore-controlling factors, metallogenic regularity, and metallogenic model of skarn deposits in the area.

3. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S2 specifically includes the following steps: S2.1, Prior Information Collection Collect physical property information, including surface geological information, geological profiles, borehole data, seismic data, and regional gravity and magnetic data; statistically analyze locations, mean values, and extreme values ​​of physical properties to form a spatial database of physical properties; S2.2 Construction of Surface Property Reference Model Based on the physical property results obtained in step S2.1, a constrained reference model is constructed using the geological map of the area as the constraint geological information. When modeling, the influence of topography is ignored, and gravity and magnetic mesh sizes are selected. On this basis, the geological map is simplified according to the physical property data, and then the physical property units of the statistical results are used to replace the surface geological units to construct a surface density difference reference model and a magnetic susceptibility reference model. S2.3, Three-dimensional inversion of gravity and magnetic properties: The physical property inversion software was selected, and the inversion parameters were set as follows: For uncertain data, the chifact mode was used to assign values ​​to balance the relationship between the smoothness of the model and the display of details; depth weighting was used to reduce the influence of the skin effect; the length-width-depth ratio was set according to the length-width ratio and vertical extension of the study area. Based on regional 1:50,000 surface Bouguer gravity and geomagnetic data, data preprocessing was completed; regional field and residual field were separated from the data by using a variety of potential field separation methods, and finally the residual anomaly that best matches the actual geological conditions was selected as the inversion data. Using the reference model established in step S2.2 as constraints, and under the premise of limiting the average value of physical properties, boundaries, and model smoothing factor, physical property values ​​are assigned to the grid cells of the underground half-space. For grid cells with determinable physical properties, their density or magnetic susceptibility is kept constant during the inversion iteration process. For grid cells with determinable extreme values ​​of physical properties, the inverted physical properties are restricted from not exceeding the upper and lower boundaries during the inversion iteration process. For grid cells with indeterminate physical properties and upper and lower boundaries, their physical property boundaries are extrapolated by the nearest grid cell with determinable physical properties. For grid cells with abrupt changes in physical properties, their smoothing coefficient is reduced. Through multiple iterative inversions, a three-dimensional gravity and magnetic physical property inversion model is obtained. S2.4, Lithological Mapping Based on the prior information collected in step S2.1, lithology-physical property relationship analysis is performed to establish the rock physical property combination in the area; logical operation is performed on the gravity and magnetic three-dimensional inversion model obtained in step S2.3, and the corresponding lithology code is assigned to the grid cells that satisfy a certain logical relationship; when the density and magnetic susceptibility inverted in step S2.3 have regular differences from the measured data, difference correction is performed, and finally lithology mapping is achieved.

4. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S3 specifically includes the following steps: S3.1, Modeling Region Definition: First, determine the modeling plane range and depth. Then, determine the spacing and position of the two-dimensional profiles, which should be consistent with the spacing of the exploration profiles in the area, and the orientation should be consistent with the orientation of the exploration profiles, and perpendicular to the direction of the main structural lines in the area. S3.2, Prior Information Processing: Simplification of surface geological units, analysis of the correspondence between lithology and physical properties, data preprocessing, and interpretation of seismic profiles; S3.3, Construction of a two-dimensional geological model: Based on the profile determined in step S3.1, and on the basis of the analysis of existing geological and borehole data, combined with wavelet analysis, range contrast and two-dimensional apparent magnetic susceptibility and density imaging analysis results, all two-dimensional geological profiles of the modeling area are inferred and drawn in sequence. S3.4, 2.5D / 3D inversion modeling: 1) 2.5D Gravity and Magnetic Fitting and Profile Correction: Based on the two-dimensional geological profile map formed in step S3.3, firstly, the dip angle, azimuth, width, background density, background magnetization, and depth of the model are set. Then, different model bodies are used to represent different geological units, and each model body is assigned a corresponding density and magnetic susceptibility value. For the same geological unit with large differences in physical properties in different regions, it is divided into multiple parts and assigned values ​​separately. Forward modeling calculations are performed on the two-dimensional geological profile. The model is repeatedly modified until the curve fitting meets the requirements. The range of modification of the physical properties and spatial morphology of the model body during modification is determined by geological rationality. (ii) 3D Inversion Simulation: In Modelvision software, the 2.5D models of all profiles are integrated into a 3D model according to their spatial location. When different profiles are combined into a whole, the theoretical anomalies of the model are calculated and compared with the actual anomalies. Where the fitting error is greater than the set threshold, the model is returned to the 2.5D profile to modify the geological body morphology. At this time, the fitting degree of gravity and magnetic signals of all profiles will change accordingly. The model is compared with the actual gravity and magnetic anomalies, and the different profiles are adjusted accordingly again until a satisfactory fitting anomaly signal is obtained. S3.5, Geological Model Visualization: The inversion simulation results profile from step S3.4 are imported into GeoModeller modeling software, and a three-dimensional integrated geological model is formed through interpolation. Each geological unit is observed from different angles, and the three-dimensional spatial morphology and distribution pattern of the geological units are analyzed.

5. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S4 specifically includes the following steps: S4.1, Prediction Method The geological entity model is transformed into a three-dimensional block model using Surpac 3D software. The feature analysis method is used to predict the three-dimensional mineralization. The feature model of the deposit is established to predict this type of deposit. S4.2, Block Model Construction The gravity and magnetic three-dimensional integrated geological model is discretized into block units of the same size. The size of the block units meets the requirements of the feature analysis method and takes into account the scale of most mineral deposits in the area and the completeness of the metallogenic geological elements. Each cell must include the complete metallogenic geological conditions of at least the typical mineral deposits in the area. Then, interpolation calculations are performed based on the outline of the solid model or spatial data source to assign various attribute values ​​to the block units. Subsequently, digital methods are used to calculate the attribute values ​​of the block units so that the spatial key information in all block units is quantified and assigned. These regular block unit sets constitute the block model. S4.3, Selection of predictor variables and determination of weighting coefficients Spatial analysis of ore-controlling factors and known ore deposits was conducted, and the optimal predictive variables were finally selected by analyzing the overlap between various ore-controlling factors and known ore deposits. The block cells where the ore deposits are located in the area were used as the prospecting model units. The predictive variables in each model unit were binarized, with 1 representing the presence of predictive variables in the model unit and 0 representing the absence of predictive variables. The weight coefficients of the predictive variables were obtained through geostatistical methods. S4.4 Establishment of the mineralization favorability comparison model and mineralization favorability analysis of the prediction unit The mineralization favorability comparison model includes model units and prediction units. Model units are block units with known mineral resource characteristics, while prediction units have unknown resource characteristics. Block units within blocks with high shallow exploration levels and detailed prediction elements are selected as model units. Parameters of various prediction elements of the model units are extracted and calculated to establish the mineralization favorability comparison model. Based on the calculation results of the predicted variables and weight coefficients in step S4.3, the connection degree value of each unit is calculated using the unit connection degree calculation formula. This value is a measure of the similarity between the predicted unit and the known model unit, indicating the probability of mineral exploration in the unit. The connection degree value is used as the spatial data source of the mineralization favorable block model. Each block unit includes the predicted variable value and the connection degree value. The inverse power distance method is used to estimate the mineralization favorableness attribute of the predicted block. The larger the value, the higher the mineralization favorableness of the predicted unit.

6. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S5 specifically includes the following steps: Based on the mineralization favorability value assigned to the block unit, the mineralization favorability of each block in the three-dimensional prediction block model is graded and colored, with the colors ranging from blue to red from small to large, resulting in a correlation level color block map. The more red the color and the larger the value, the greater the mineralization probability of the corresponding block unit. In this way, the final three-dimensional mineralization comprehensive prediction model is constructed.

7. The three-dimensional mineralization prediction method for skarn-type deposits as described in claim 1, characterized in that, S6 specifically includes the following steps: In the graded color block map, blocks with a reddish tint are blocks with a higher probability of mineralization. It is necessary to repeatedly try and adjust the prediction threshold to delineate the mineral exploration target area. When the range above a certain mineralization favorability value can contain most of the known mineral deposits and has the smallest predicted volume, then the mineralization favorability value is used as the prediction threshold. Block units greater than or equal to the prediction threshold are initially screened as target area block units. The pre-delineation of the target area is consistent with the actual geological conditions, and the integrity of the target area shape is maintained as much as possible. After multiple comparative analyses, the mineral exploration target area is delineated. The target area is graded according to the relative classification method, and the target area delineation is finally completed. The delineated target areas were verified using methods such as field geological surveys, UAV geomagnetic surveys, induced polarization gradient profiling, high-density electrical resistivity profiling, and sample testing and analysis. Key target areas were selected for verification work.

8. A three-dimensional mineralization prediction system for skarn-type deposits, characterized in that... include: The data collection module is used to collect and organize geological information about the area under study and its surrounding areas; The inversion and lithological mapping module is used to perform gravity and magnetic three-dimensional physical property inversion and lithological mapping based on the collected geological information. The gravity and magnetic 3D integrated geological modeling module is used for gravity and magnetic 3D integrated geological modeling using the discrete volume inversion method. The 3D mineralization prediction module is used to predict the 3D mineralization favorability contrast based on the constructed model. The three-dimensional comprehensive prediction model construction module is used to construct a three-dimensional comprehensive metallogenic prediction model based on the mineralization favorability value. The target area delineation and verification module is used to delineate and verify target areas using a three-dimensional comprehensive mineralization prediction model, and to complete three-dimensional mineralization prediction supported by regional gravity and magnetic data.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the three-dimensional mineralization prediction method for skarn-type deposits as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program is used to implement the three-dimensional mineralization prediction method for skarn-type deposits as described in any one of claims 1-7.

Images