Multi-source geophysical prospecting data fusion-based method and system for abnormity of contact zone of Zinzhite rock mass

Abstract

The invention provides a multi-source geophysical prospecting data fused Zenzenite rock mass contact zone anomaly method and system, and relates to the technical field of geophysical prospecting, the method comprises the following steps: executing joint denoising processing and time-frequency feature enhancement processing on a TFEM data set in a three-dimensional fused data volume, and outputting the optimized three-dimensional fused data volume; and based on the optimized three-dimensional fusion data body, by establishing a mapping relationship between magnetic, electric and polarization anomalies and a geologic body, quantifying the correlation between a multi-parameter combination and a mineralization index and screening an up-to-standard abnormal region, and finally delineating a magnetic anomaly gradient zone, a low-resistance region and a high-polarization anomaly co-occurrence ore prospecting target region three-dimensional distribution diagram. Through multi-source geophysical prospecting method collaboration, data deep fusion, quality optimization and accurate mapping, full-chain technology upgrading from data acquisition to target area delineation is realized, and the high efficiency of mineral exploration is improved.

Description

Technical Field

[0001] This invention relates to the field of geophysical exploration technology, and in particular to a method and system for fusion of multi-source geophysical data to identify contact zone anomalies in the Bangzhong zinc-copper ore mass. Background Technology

[0002] Integrated geophysical exploration technology, as a key means of mineral resource exploration, can effectively reveal the distribution characteristics of underground geological bodies by integrating multi-source physical parameters such as magnetic, electrical, and electromagnetic methods, providing important basis for the detection of concealed ore bodies, and playing an irreplaceable role, especially in mineralization exploration in complex geological environments such as rock mass contact zones.

[0003] Foreign countries have long been leading in this field, with technologies such as airborne geophysical exploration and borehole geophysical exploration being applied on a large scale. New technologies such as superconducting quantum interference devices and seismic scattering imaging, with their high resolution and high detection accuracy, have provided new means for the exploration of complex structures such as rock contact zones. They also emphasize the refinement of data processing and the quantitative interpretation of anomalies, thereby improving the success rate of mineral exploration.

[0004] Significant progress has been made in China in recent years. Conventional methods such as high-precision magnetic methods, induced polarization methods, and CSAMT are widely used in metal mineral exploration. The combination of multiple methods in complex mining areas has become a key focus, and big data and artificial intelligence technologies have begun to be integrated into data processing to improve mineral exploration efficiency.

[0005] However, in mineralization exploration of complex rock mass contact zones, single geophysical methods or conventional combinations of techniques may have limitations. Taking the Bangzhong zinc-copper mine in Linzhou County, Tibet Autonomous Region as an example, although time-frequency electromagnetic measurements in 2024 delineated multiple anomalies and inferred the existence of multiple favorable metallogenic zones, due to the complex geological conditions of the mining area (such as the development of folds and faults and the superposition of deep and shallow mineralization responses), existing technologies may find it difficult to accurately distinguish the differences in electromagnetic responses between the mineralized body and the surrounding rock, thus failing to achieve precise delineation of the ore body's extent and restricting exploration efficiency and the scientific nature of resource evaluation. Summary of the Invention

[0006] The technical problem to be solved by this invention is to provide a method and system for anomaly detection in the contact zone of the Bangzhong zinc-copper ore body by multi-source geophysical data fusion. Through the synergy of multi-source geophysical methods, deep data fusion, quality optimization and precise mapping, the invention achieves a full-chain technical upgrade from data acquisition to target area delineation, thereby improving the efficiency of mineral exploration.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, a method for analyzing contact zone anomalies in the Bangzhong zinc-copper ore mass through multi-source geophysical data fusion, the method comprising: Step S1: Based on the geological characteristics of the mining area, set the first sampling interval and the first measurement accuracy threshold, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set the first transmission power threshold and the first frequency range according to the deep electrical differences, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set the target transmission frequency sequence for the difference in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. Step S2: The high-resolution magnetic data set, CSAMT data set, TFEM data set, and the existing induced polarization data set in the mining area are combined by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting and fusion to allocate the material property sensitivity weights, to generate a three-dimensional fused data volume containing multi-attribute features. Step S3: Perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume; Step S4: Based on the optimized three-dimensional fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified and qualified anomaly areas are screened. Finally, a three-dimensional distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated.

[0008] Further, in step S1, the following steps are performed simultaneously in the contact zone area of ​​the zinc-copper ore rock mass in the Bangzhong mine: Based on the geological characteristics of the mining area, a first sampling interval and a first measurement accuracy threshold are set, and high-precision magnetic measurements are implemented to output a magnetic dataset containing magnetic field strength data; Based on the differences in deep electrical properties, a first transmission power threshold and a first frequency range are set, and CSAMT detection is implemented to output a CSAMT dataset containing resistivity data; For the differences in electromagnetic response between the mineralized body and the surrounding rock, a target transmission frequency sequence is set, and time-frequency electromagnetic measurements are implemented to output a TFEM dataset containing resistivity and polarizability parameters, including: Based on the fold and fault structure characteristics of the mining area, a proton magnetometer is used to set the first sampling interval to adapt to the local changes in the magnetic field caused by structural deformation, and a first measurement accuracy threshold is set to identify weak magnetic anomalies. A high-resolution magnetic dataset containing magnetic field strength is output. Based on the structural framework revealed by the high-resolution magnetic data set, the CSAMT device is used to set a first transmit power threshold according to the electrical differences of deep strata to ensure the deep signal penetration capability, and a first frequency range is set to match the contact electrical structural response of the rock mass, and the output is a CSAMT data set containing resistivity. Based on the deep electrical background provided by the CSAMT dataset, time-frequency electromagnetic measurements were performed. A target broadband transmission sequence was set to target the difference in electromagnetic response between the mineralized body and the surrounding rock. The first low-frequency band captured the slowly varying response of the deep mineralized body, and the second high-frequency band captured the transient response of the shallow contact zone. The output was a TFEM dataset containing resistivity and polarizability parameters.

[0009] Further, in step S2, the high-resolution magnetic dataset, CSAMT dataset, TFEM dataset, and existing induced polarization dataset from the mining area are processed through geographic coordinate unification to eliminate terrain offset, physical dimension standardization to transform multi-source parameters, and adaptive weighted fusion to allocate material property sensitivity weights, generating a three-dimensional fused data volume containing multi-attribute features, including: Using high-resolution magnetic data sets, CSAMT datasets, TFEM datasets, and mining area induced polarization datasets as input, the positioning coordinates of all datasets are extracted, and the horizontal projection deviation is calculated based on the terrain elevation gradient. Based on the projection deviation, the horizontal coordinate offset is corrected according to the elevation gradient distribution so that the corrected coordinate deviation does not exceed the set spatial accuracy threshold. The corrected coordinate data is resampled to a unified grid node, and the spatially aligned dataset is output. Based on the spatially aligned dataset, and taking the mean background magnetic field strength of the mining area as the benchmark, the first magnetic parameter value of each grid node is calculated. The value is the difference between the magnetic field strength of that point and the mean background value divided by the mean background value. The resistivity value of each node is logarithmically calculated to generate the second electrical parameter value. The probability distribution of the polarizability value of the entire mining area is statistically analyzed. The polarizability value of each node is normalized to the 0-1 interval by subtracting the minimum value of the entire mining area and dividing by the difference between the maximum value and the minimum value, generating the third induced polarization parameter value. Finally, a standardized matrix containing the first magnetic parameter, the second electrical parameter, and the third induced polarization parameter is output.

[0010] Further, in step S2, the high-resolution magnetic dataset, CSAMT dataset, TFEM dataset, and existing induced polarization dataset from the mining area are processed through geographic coordinate unification to eliminate terrain offset, physical dimension standardization to transform multi-source parameters, and adaptive weighted fusion to allocate material property sensitivity weights, generating a three-dimensional fused data volume containing multi-attribute features. This also includes: Based on the standardized matrix, each grid node is traversed, and the contact zone boundary area, mineralization anomaly area and ordinary area are marked according to the threshold conditions of the first magnetic parameter, the second electrical parameter and the third induced polarization parameter, respectively, to generate a partition marking result set; Based on the partitioning labeling result set, for nodes labeled as contact zone boundary areas, the first magnetic weight coefficient assigned to the node is output; for nodes labeled as mineralization anomaly areas, the second electrical weight coefficient and the third polarization weight coefficient assigned to the node are output; for nodes labeled as ordinary areas, the equal weight coefficient assigned to the node is output; based on the first magnetic weight coefficient, the second electrical weight coefficient, the third polarization weight coefficient, and the equal weight coefficient, a weight allocation result set is generated. Based on the weighting results and the standardization matrix, the first magnetic parameter value, the second electrical parameter value, and the third induced polarization parameter value of the node are extracted. The magnetic weighting coefficient is multiplied by the first magnetic parameter value to obtain the magnetic weighted value. The electrical weighting coefficient is multiplied by the second electrical parameter value to obtain the electrical weighted value. The polarization weighting coefficient is multiplied by the third induced polarization parameter value to obtain the polarization weighted value. The magnetic weighted value, electrical weighted value, and polarization weighted value are added together to output the fusion value of the node and generate a set of node fusion values. Based on the node fusion value set, the grid node planar coordinates are used as the X and Y axis positioning references, and the exploration depth is used as the Z axis layering basis. The node fusion values ​​are extended to a continuous three-dimensional space through spatial interpolation to generate a three-dimensional fusion data volume containing multiple attribute features.

[0011] Further, in step S3, joint denoising and time-frequency feature enhancement processing are performed on the TFEM dataset in the 3D fused data volume, and the optimized 3D fused data volume is output, including: Based on the TFEM dataset in the 3D fused data volume, wavelet transform threshold denoising is performed on the high-frequency components above 100Hz to filter out random noise with amplitude less than twice the standard deviation of background noise. Empirical mode decomposition is performed on the low-frequency components of 0.1-10Hz to remove baseline drift components and output the denoised and reconstructed TFEM dataset. Based on the denoised and reconstructed TFEM dataset, within the sensitive frequency band of the rock mass contact zone, a first low-frequency sub-segment is captured using a longer time window to capture deep, slowly varying responses, and a second high-frequency sub-segment is captured using a shorter time window to capture shallow, transient anomalies, generating the time spectrum of the sensitive frequency band. Based on the time spectrum, the mean amplitude of the frequency point over the entire time period is calculated as a background benchmark, and the instantaneous amplitude is divided by the background benchmark to generate a relative energy coefficient. Based on the relative energy coefficient, the time spectra of anomalies exceeding a set anomaly threshold are extracted, and a high-resolution time-frequency anomaly spectrum set is output. Based on the high-resolution time-frequency anomaly spectrum set, the time-frequency anomaly spectrum set is mapped back to the original TFEM data nodes, replacing the corresponding node values ​​of the TFEM dataset in the original 3D fusion data volume, and outputting the optimized 3D fusion data volume.

[0012] Further, in step S4, based on the optimized 3D fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified, and qualified anomaly areas are screened. Finally, a 3D distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated, including: Based on the optimized 3D fused data volume, the horizontal gradient of the magnetic field intensity at each depth layer is calculated. Continuous regions with gradient values ​​exceeding a set magnetic gradient threshold are selected to obtain magnetic anomaly gradient zones, which are marked as skarn boundaries. Closed regions with resistivity values ​​below a set low resistivity threshold are extracted and marked as low resistivity zones. Closed regions with polarizability values ​​above a set high polarizability threshold are extracted and marked as high polarizability anomaly zones. The magnetic anomaly gradient zones are associated with skarn boundaries, and spatially overlapping low resistivity zones and high polarizability anomaly zones are associated as zinc-copper mineralization bodies, generating anomaly-geological body mapping models. Based on the anomaly-geological body mapping model, the magnetic susceptibility, resistivity, and polarizability values ​​of each node within the anomaly region are extracted to generate a set of node physical property parameters. Based on this set, the geometric mean of the three parameters is calculated as the combined parameter value for each node, generating a combined parameter set. Based on this combined parameter set, using borehole core analysis data as a benchmark, the Pearson correlation coefficient between the combined parameter values ​​and the mineralization grade is calculated to generate a preliminary set of related regions. Based on this preliminary set of related regions, anomaly regions with correlation coefficients exceeding a set correlation threshold are selected, and a set of qualified anomaly regions is output. Further, in step S4, based on the optimized 3D fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified, and qualified anomaly areas are screened. Finally, a 3D distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated, which also includes: Based on the set of qualified anomaly regions, three-dimensional spatial units that simultaneously contain magnetic anomaly gradient bands, low resistivity regions and high polarization anomaly regions are extracted to generate ternary anomaly co-occurrence regions. Based on the ternary anomaly co-occurrence region, internal holes are filled to eliminate data discontinuity areas, and boundaries are smoothed to regularize the target region outline, outputting an optimized target region unit set. Based on the optimized target area unit set, the combined parameter values ​​of each target area unit are extracted, and the target areas are divided into Grade A, Grade B, and Grade C according to the parameter value, generating a three-dimensional distribution map of graded mineral exploration target areas.

[0013] Secondly, a multi-source geophysical data fusion-based contact zone anomaly system for the Bangzhong zinc-copper ore mass includes: The acquisition module is used to set a first sampling interval and a first measurement accuracy threshold based on the geological characteristics of the mining area, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set a first transmission power threshold and a first frequency range based on the differences in deep electrical properties, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set a target transmission frequency sequence based on the differences in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. The fusion module is used to unify the high-resolution magnetic data set, CSAMT data set, TFEM data set and the existing induced polarization data set in the mining area by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting the fusion to allocate the property sensitivity weights, so as to generate a three-dimensional fused data volume containing multi-attribute features. The optimization module is used to perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume. The target area delineation module is used to establish a mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies based on the optimized 3D fusion data volume. It quantifies the correlation between multi-parameter combinations and mineralization indicators, screens qualified anomaly areas, and finally delineates a 3D distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist.

[0014] Thirdly, a computing device, comprising: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0015] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0016] The above-described solution of the present invention has at least the following beneficial effects: Simultaneously conducting high-precision magnetics, CSAMT, and time-frequency electromagnetic measurements, combined with existing excitation polarization data from the mining area, covering key physical property parameters such as magnetic field strength, resistivity, and polarizability, a multi-dimensional physical property feature matrix was formed. By selectively setting parameters such as sampling interval, transmission power, and frequency sequence, the data acquisition was ensured to be highly compatible with geological features, avoiding the limitations of single geophysical methods. Geographic coordinate unification eliminated terrain offset, ensuring spatial alignment of multi-source data. Physical dimension standardization addressed the issue of dimensional differences among multiple parameters. Adaptive weighted fusion highlighted the physical property sensitivity of key areas. The final generated three-dimensional fused data volume enabled the magnetic, electrical, and polarization parameters to form a synergistic correlation in three-dimensional space, improving the interpretability of the data. The study demonstrates the value of comprehensive analysis. Joint denoising of the TFEM dataset, employing high-frequency wavelet thresholding and low-frequency empirical mode decomposition to effectively filter out noise interference and improve the signal-to-noise ratio of the original data. Time-frequency feature enhancement further amplifies sensitive frequency band signals related to mineralization, making weak anomaly features more prominent. By establishing mapping relationships between magnetic anomaly gradient zones and skarn boundaries, and between low-resistivity-high-polarization overlap zones and zinc-copper mineralization bodies, abstract geophysical anomalies are transformed into concrete geological body associations. Quantifying the correlation between multi-parameter combinations and mineralization grades ensures a high degree of matching between anomaly areas and mineralization indicators. Finally, the study delineates target areas where magnetic anomaly gradient zones, low-resistivity zones, and high-polarization anomalies coexist, and classifies them according to parameter values, achieving precise target area location and clear priority. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass through multi-source geophysical data fusion, as provided in an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of an anomaly system of the contact zone of the Bangzhong zinc-copper ore rock mass, provided by an embodiment of the present invention, based on multi-source geophysical data fusion. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0020] like Figure 1 As shown, an embodiment of the present invention proposes a method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass through multi-source geophysical data fusion. The method includes the following steps: Step S1: Based on the geological characteristics of the mining area, set the first sampling interval and the first measurement accuracy threshold, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set the first transmission power threshold and the first frequency range according to the deep electrical differences, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set the target transmission frequency sequence for the difference in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. Step S2: The high-resolution magnetic data set, CSAMT data set, TFEM data set, and the existing induced polarization data set in the mining area are combined by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting and fusion to allocate the material property sensitivity weights, to generate a three-dimensional fused data volume containing multi-attribute features. Step S3: Perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume; Step S4: Based on the optimized three-dimensional fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified and qualified anomaly areas are screened. Finally, a three-dimensional distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated.

[0021] In this embodiment of the invention, multiple methods such as magnetics, CSAMT, and time-frequency electromagnetics are simultaneously performed to measure key physical properties such as magnetic field strength, resistivity, and polarizability. Combined with existing excitation polarization data, a multi-dimensional physical property feature matrix is ​​formed, avoiding the limitations of a single method. By unifying coordinates to eliminate terrain offset, standardizing physical dimensions to resolve differences in multi-source parameters, and adaptively weighting and fusion to allocate sensitivity weights, the scattered multi-source data is integrated into a three-dimensional fused data volume. This enables magnetic, electrical, and polarization parameters to achieve synergistic correlation in spatial and physical dimensions, improving data interpretability. Joint denoising and time-frequency feature enhancement are performed on the TFEM dataset to effectively suppress noise interference and strengthen signals in sensitive frequency bands, reduce data errors, and highlight the differences in electromagnetic response between mineralized bodies and surrounding rocks, providing a high-quality data foundation for subsequent anomaly identification. By establishing a mapping relationship between anomalies and geological bodies, the correlation between multiple parameters and mineralization indicators is quantified. Focusing on the co-occurrence areas of magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies, the accurate conversion from multi-source data to the three-dimensional distribution of the target area is achieved, reducing blind exploration and improving the hit rate and exploration efficiency of the mineral exploration target area.

[0022] In a preferred embodiment of the present invention, step S1 is performed synchronously in the contact zone area of ​​the zinc-copper ore rock mass in the blast furnace: a first sampling interval and a first measurement accuracy threshold are set based on the geological characteristics of the mining area; high-precision magnetic measurement is performed, and a magnetic dataset containing magnetic field strength data is output; a first transmission power threshold and a first frequency range are set according to the differences in deep electrical properties; CSAMT detection is performed, and a CSAMT dataset containing resistivity data is output; a target transmission frequency sequence is set for the differences in electromagnetic response between the mineralized body and the surrounding rock; time-frequency electromagnetic measurement is performed, and a TFEM dataset containing resistivity and polarizability parameters is output, which may include: Step S11: Based on the fold and fracture structure characteristics of the mining area, a proton magnetometer is used to set the first sampling interval to adapt to the local changes in the magnetic field caused by structural deformation, and a first measurement accuracy threshold is set to identify weak magnetic anomalies, and a high-resolution magnetic data set containing magnetic field strength is output. Step S12: Based on the structural framework revealed by the high-resolution magnetic data set, the CSAMT device is used to set a first transmit power threshold according to the electrical differences of deep strata to ensure the deep signal penetration capability, and a first frequency range is set to match the contact electrical structure response of the rock mass, and the CSAMT data set containing resistivity is output. Step S13: Based on the deep electrical background provided by the CSAMT dataset, perform time-frequency electromagnetic measurements. Set a target broadband transmission sequence for the difference in electromagnetic response between the mineralized body and the surrounding rock. The first low-frequency band captures the slowly varying response of the deep mineralized body, and the second high-frequency band captures the transient response of the shallow contact zone. Output a TFEM dataset containing resistivity and polarizability parameters.

[0023] In this embodiment of the invention, by matching the sampling interval and accuracy settings of fold and fault structural features, the ability of magnetic data to capture local magnetic field changes and weak magnetic anomalies is improved, providing high-precision basic data for subsequent structural framework analysis. Based on the previous structural framework, the transmission power and frequency of CSAMT are optimized, enhancing the penetration ability of deep signals and the matching degree with contact charged structures, thereby improving the resolution accuracy of resistivity data for deep strata. In combination with the deep electrical background, a broadband transmission sequence is designed to specifically capture the electromagnetic responses at both deep and shallow depths, enriching the physical parameters of TFEM data and improving the targeting and accuracy of identifying differences between mineralized bodies and surrounding rocks.

[0024] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S11 above, the distribution characteristics of folded and faulted structures in the mining area (such as fault strike, fold morphology, and structural deformation intensity) are analyzed to determine the first sampling interval that can reflect local changes in the magnetic field caused by structural deformation (i.e., adjusting the sampling point spacing according to the density of structures to ensure the capture of subtle changes in the local magnetic field); based on the need to identify weak magnetic anomalies, a first measurement accuracy threshold is set (i.e., defining the upper limit of the error for magnetic field strength measurement to ensure the ability to distinguish between weak magnetic bodies and the background magnetic field); using a proton magnetometer, magnetic field strength measurements are carried out in the mining area according to the first sampling interval and the first measurement accuracy threshold, and magnetic field strength data at each measuring point are collected; the measurement data are processed (e.g., removing obvious outliers and sorting by spatial location), and a high-resolution magnetic dataset containing continuous magnetic field strength data is output.

[0025] In step S12 above, based on a high-resolution magnetic dataset, the structural framework of the mining area is analyzed (such as the location of major fault zones and the distribution boundaries of rock masses); the electrical differences in deep strata are analyzed (such as the resistivity difference between the rock mass and the surrounding rock, and the resistivity variation law of strata at different depths); a first transmission power threshold is set according to the degree of difference (i.e., the lower limit of power to ensure that the transmitted signal can penetrate to the deep target strata); combined with the electrical structural characteristics of the rock mass contact zone (such as the resistivity distribution pattern inside and outside the contact zone), a first frequency range matching its response is set (i.e., the frequency range that can highlight the resistivity difference of the contact zone is selected); using a CSAMT device, detection is performed according to the above first transmission power threshold and first frequency range to collect resistivity data at each measuring point; preliminary correction is performed on the resistivity data (such as terrain correction), and a CSAMT dataset containing continuous resistivity data is output.

[0026] In step S13 above, based on the CSAMT dataset, the electrical background of deep strata is clarified (e.g., regional average resistivity, resistivity variation trend with depth); the differences in electromagnetic response between mineralized bodies and surrounding rocks are analyzed (e.g., the numerical differences in resistivity and polarizability between mineralized bodies and surrounding rocks), and a target broadband transmission sequence is designed to address these differences: a first low-frequency band (used to capture the slowly changing electromagnetic response of deep mineralized bodies) and a second high-frequency band (used to capture the rapidly changing electromagnetic response of shallow contact zones); electromagnetic signals are transmitted using a time-frequency electromagnetic measurement device according to the above target broadband transmission sequence, and resistivity and polarizability data at each measuring point are collected simultaneously; the collected parameter data are filtered (e.g., noise interference data is removed), and a TFEM dataset containing resistivity and polarizability parameters is output.

[0027] In a preferred embodiment of the present invention, step S2 above, which involves unifying the high-resolution magnetic dataset, CSAMT dataset, TFEM dataset, and existing induced polarization dataset in the mining area to eliminate terrain offset through geographic coordinate unification, standardizing and transforming multi-source parameters through physical dimension standardization, and adaptively weighted fusion to allocate material property sensitivity weights, to generate a three-dimensional fused data volume containing multi-attribute features, may include: Step S21: Take the high-resolution magnetic method dataset, CSAMT dataset, TFEM dataset and mining area induced polarization method dataset as input, extract the positioning coordinates of all datasets, calculate the horizontal projection deviation based on the terrain elevation gradient; based on the projection deviation, correct the horizontal coordinate offset according to the elevation gradient distribution, so that the corrected coordinate deviation does not exceed the set spatial accuracy threshold; resample the corrected coordinate data to a unified grid node, and output the spatially aligned dataset. Step S22: Based on the spatially aligned dataset, and taking the mean background magnetic field strength of the mining area as a benchmark, calculate the first magnetic parameter value of each grid node. The value is the difference between the magnetic field strength of that point and the mean background value divided by the mean background value. Perform a logarithmic operation with base 10 on the resistivity value of each node to generate the second electrical parameter value. Statistically calculate the probability distribution of the polarizability value of the entire mining area. Normalize the polarizability value of each node to the 0-1 interval by subtracting the minimum value of the entire mining area and dividing by the difference between the maximum and minimum values ​​to generate the third induced polarization parameter value. Finally, output a standardized matrix containing the first magnetic parameter, the second electrical parameter, and the third induced polarization parameter. In this embodiment of the invention, horizontal projection offset caused by terrain is eliminated by coordinate correction and resampled to a unified grid to achieve spatial alignment of multi-source data, break the positional differences of different datasets, and provide a consistent spatial benchmark for subsequent multi-parameter fusion; through standardization processing, magnetic, electric, and polarization parameters with large differences in dimensions are converted into uniform scale parameters that can be directly compared, eliminating the interference of physical dimensions on the fusion results and ensuring the comparability of multi-source parameters.

[0028] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S21 above, the high-resolution magnetic method dataset, CSAMT dataset, TFEM dataset, and mining area induced polarization method dataset are input, and the original positioning coordinates (including plane coordinates and terrain elevation information) of all datasets are extracted. Based on the terrain elevation differences of each point, the horizontal projection deviation caused by elevation changes (i.e., the offset value of the projected position of different elevation points on the horizontal plane) is calculated. According to the elevation gradient distribution law (e.g., the more drastic the elevation change, the greater the correction amount), the horizontal coordinates of each dataset are offset corrected to ensure that the coordinate deviation of all data after correction is controlled within the set spatial accuracy threshold. All corrected data are resampled according to a uniform grid node spacing (i.e., the data are matched to the same grid position through interpolation, etc.), and finally, a dataset with completely aligned spatial positions is output.

[0029] In step S22 above, based on the spatially aligned dataset, the average value of the background magnetic field strength in the mining area is calculated (as a reference value); for each grid node, the magnetic field strength at that point is subtracted from the background mean, and the difference is divided by the background mean to obtain the first magnetic parameter value; for the resistivity value of each grid node, a logarithmic operation with base 10 is performed one by one, and the result is used as the second electrical parameter value; the probability distribution of the polarizability value of the entire mining area is statistically analyzed to determine the minimum and maximum values ​​of the polarizability; for the polarizability value of each grid node, the minimum value of the entire mining area is subtracted from the value, and the result is divided by (the difference between the maximum and minimum values) to obtain the third induced polarization parameter value normalized to the 0-1 interval; the first magnetic parameter, second electrical parameter, and third induced polarization parameter of all nodes are integrated to form a standardized matrix and output.

[0030] In a preferred embodiment of the present invention, step S2 above, which involves unifying the high-resolution magnetic data set, CSAMT data set, TFEM data set, and existing induced polarization data set in the mining area to eliminate terrain offset through geographic coordinate unification, standardizing and transforming multi-source parameters through physical dimension standardization, and adaptively weighting and fusion to allocate property sensitivity weights, to generate a three-dimensional fused data volume containing multi-attribute features, may further include: Step S23: Based on the standardized matrix, traverse each grid node and mark the contact zone boundary area, mineralization anomaly area and ordinary area according to the threshold conditions of the first magnetic parameter, the second electrical parameter and the third induced polarization parameter, respectively, and generate a partition marking result set; Step S24: Based on the partitioning labeling result set, for nodes labeled as contact zone boundary areas, output the first magnetic weight coefficient assigned to the node; for nodes labeled as mineralization anomaly areas, output the second electrical weight coefficient and the third polarization weight coefficient assigned to the node; for nodes labeled as ordinary areas, output the equal weight coefficient assigned to the node; and generate a weight allocation result set based on the first magnetic weight coefficient, the second electrical weight coefficient, the third polarization weight coefficient, and the equal weight coefficient. Step S25: Based on the weight allocation result and the standardization matrix, extract the first magnetic parameter value, the second electrical parameter value, and the third induced polarization parameter value of the node. Multiply the magnetic weight coefficient by the first magnetic parameter value to obtain the magnetic weight value. Multiply the electrical weight coefficient by the second electrical parameter value to obtain the electrical weight value. Multiply the polarization weight coefficient by the third induced polarization parameter value to obtain the polarization weight value. Add the magnetic weight value, electrical weight value, and polarization weight value to output the fusion value of the node and generate a set of node fusion values. Step S26: Based on the node fusion value set, the grid node planar coordinates are used as the X and Y axis positioning references, and the exploration depth is used as the Z axis layering basis. The node fusion values ​​are extended to a continuous three-dimensional space through spatial interpolation to generate a three-dimensional fusion data volume containing multiple attribute features.

[0031] In this embodiment of the invention, the grid nodes are partitioned and marked with explicit threshold conditions, providing a clear classification basis for subsequent weight allocation and ensuring that the parameter weight allocation for different geological feature areas is more targeted. The weights are adaptively allocated based on the partitioning results, highlighting the feature weights of the contact zone boundary area and the mineralization anomaly area, and avoiding the obscuring of key anomalies by equal weights. The organic fusion of multiple parameters is achieved through weighted summation, which not only preserves the independent characteristics of each parameter, but also highlights the differences in physical properties of key areas through weights, enhancing the indicativeness of the fused value for mineralization features. The discrete fused value is extended into a continuous three-dimensional data volume through spatial interpolation, so that the multi-attribute features form a complete distribution in three-dimensional space, which facilitates intuitive observation and analysis of the spatial morphology and distribution patterns of mineralization anomalies.

[0032] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S23 above, based on the standardized matrix, which contains the first magnetic parameter value, second electrical parameter value, and third induced polarization parameter value of all grid nodes, each grid node in the matrix is ​​traversed one by one, and attribute determination is performed on each node: if the first magnetic parameter value of the node reaches or exceeds the first magnetic threshold, it is determined to be a contact zone boundary area and marked as "contact zone boundary area"; if the second electrical parameter value of the node does not exceed the second electrical threshold and the third induced polarization parameter value reaches or exceeds the third polarization threshold, it is determined to be a mineralization anomaly area and marked as "mineralization anomaly area"; if neither of the above two conditions is met, it is determined to be a normal area and marked as "normal area"; the marking results of all nodes are collected, organized according to the spatial location of the grid nodes, and a partition marking result set (containing the region category label of each node) is generated.

[0033] In step S24 above, the partition label result set is called to obtain the region category label of each grid node; weight coefficients are assigned according to the label type of the node: nodes labeled as "contact zone boundary area" are assigned the first magnetic weight coefficient; nodes labeled as "mineralization anomaly area" are assigned the second electrical weight coefficient and the third polarization weight coefficient; nodes labeled as "ordinary area" are assigned equal weight coefficients (i.e., the magnetic, electrical, and polarization weight coefficients are the same); according to the spatial location of the grid node, the weight coefficients corresponding to each node are organized into structured data to generate a weight allocation result set (containing the weight coefficients of each node).

[0034] In step S25 above, the weight allocation result set and the original normalized matrix are called simultaneously to locate the same grid node; the first magnetic parameter value, the second electrical parameter value, and the third induced polarization parameter value of the node are extracted from the normalized matrix; the magnetic weight coefficient, electrical weight coefficient, and polarization weight coefficient of the node are extracted from the weight allocation result set; the weighted values ​​of each parameter are calculated: the magnetic weight coefficient is multiplied by the first magnetic parameter value to obtain the magnetic weighted value, the electrical weight coefficient is multiplied by the second electrical parameter value to obtain the electrical weighted value, and the polarization weight coefficient is multiplied by the third induced polarization parameter value to obtain the polarization weighted value; the three weighted values ​​are added together to obtain the fusion value of the node; the fusion values ​​of all nodes are arranged according to the spatial order of the grid nodes to generate a set of node fusion values.

[0035] In step S26 above, the planar coordinates of the grid nodes are used as the X-axis (horizontal axis) and Y-axis (vertical axis) as the horizontal positioning reference in three-dimensional space; the exploration depth is used as the Z-axis (vertical axis) as the basis for layering in three-dimensional space (i.e., dividing different layers according to depth); the set of node fusion values ​​is read, and these values ​​correspond to discrete grid nodes; the discrete node fusion values ​​are extended to the entire three-dimensional space through spatial interpolation methods (such as estimating the value of the blank area based on the fusion values ​​of adjacent nodes) to fill the blank areas between nodes; the interpolated continuous data is integrated to generate a three-dimensional fusion data volume containing multiple attribute features (magnetic, electrical, and polarization fusion features).

[0036] In a preferred embodiment of the present invention, step S3, which involves performing joint denoising and time-frequency feature enhancement processing on the TFEM dataset in the 3D fused data volume and outputting the optimized 3D fused data volume, may include: Step S31: Based on the TFEM dataset in the three-dimensional fused data volume, perform wavelet transform threshold denoising on the high-frequency components above 100Hz to filter out random noise with amplitude less than twice the standard deviation of background noise, perform empirical mode decomposition on the low-frequency components of 0.1-10Hz to remove baseline drift components, and output the denoised and reconstructed TFEM dataset. Step S32: Based on the denoised and reconstructed TFEM dataset, within the sensitive frequency band of the rock mass contact zone, the first low-frequency sub-segment uses a longer time window to capture the deep, slowly varying response, and the second high-frequency sub-segment uses a shorter time window to capture the shallow, transient anomalies, generating the time spectrum of the sensitive frequency band; based on the time spectrum, the mean amplitude of the frequency point over the entire time period is calculated as a background reference, and the instantaneous amplitude is divided by the background reference to generate a relative energy coefficient; based on the relative energy coefficient, the time spectra of anomalies exceeding a set anomaly threshold are extracted, and a high-resolution time-frequency anomaly spectrum set is output; Step S33: Based on the high-resolution time-frequency anomaly spectrum set, map the time-frequency anomaly spectrum set back to the original TFEM data nodes, replace the corresponding node values ​​of the TFEM dataset in the original 3D fusion data volume, and output the optimized 3D fusion data volume.

[0037] In this embodiment of the invention, targeted denoising by frequency band effectively filters out random noise and baseline drift, reducing the impact of interference on TFEM data and improving the signal-to-noise ratio and reliability of the original data. By dynamically adjusting the time window length to match the deep and shallow responses and combining energy normalization to extract the anomalous time spectrum, the time-frequency characteristics of mineralization-related signals in the sensitive frequency band are enhanced, highlighting the differences and details of mineralization anomalies in the deep and shallow regions and improving the accuracy of anomaly identification. The optimized time-frequency anomaly features are then fed back into the three-dimensional fusion data volume, realizing iterative optimization of the TFEM dataset and enabling the three-dimensional fusion data volume to more accurately reflect the mineralization-related physical properties.

[0038] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S31 above, the TFEM dataset is extracted from the 3D fusion data volume. This dataset contains the original signal data of time-frequency electromagnetic measurements. The signals in the TFEM dataset are then subjected to frequency band separation: high-frequency components above 100Hz are separated, and wavelet transform threshold denoising is applied to these components—first, the standard deviation of the background noise is calculated, and then signal components with amplitudes lower than "twice the standard deviation of the background noise" are selected, and these components are identified as random noise and filtered out. Low-frequency components between 0.1-10Hz are separated, and empirical mode decomposition is applied to these components—the low-frequency signal is decomposed into multiple intrinsic mode components, and components belonging to baseline drift are identified and removed. The denoised high-frequency and low-frequency components are integrated to reconstruct the denoised TFEM dataset and output it.

[0039] In step S32 above, based on the denoised and reconstructed TFEM dataset, the focus is on the sensitive frequency band of the rock mass contact zone; time-frequency analysis is performed on the signal within the sensitive frequency band: in the first low-frequency sub-segment of the sensitive frequency band (corresponding to the deep signal), a longer time window is used to perform a short-time Fourier transform on the signal to capture the slow change response of the deep mineralized body; in the second high-frequency sub-segment of the sensitive frequency band (corresponding to the shallow signal), a shorter time window is used to perform a short-time Fourier transform on the signal to capture the rapid change response of the shallow contact zone; the above processing results are integrated to generate the complete time spectrum of the sensitive frequency band.

[0040] Energy focusing processing is performed on the time spectrum: For each frequency point in the time spectrum, the average amplitude of that frequency point over the entire time period is calculated and used as the background reference for that frequency point; for the instantaneous amplitude of each frequency point, the instantaneous amplitude is divided by the background reference to obtain the relative energy coefficient; an anomaly threshold is set, and time spectrum portions with relative energy coefficients exceeding the threshold are selected and identified as anomalous time spectra; all anomalous time spectra are integrated to generate and output a high-resolution time-frequency anomalous spectrum set.

[0041] In step S33 above, based on the high-resolution time-frequency anomaly spectrum set, which contains optimized time-frequency anomaly feature data, the correspondence between each data point in the time-frequency anomaly spectrum set and the grid nodes of the original TFEM dataset is determined (through spatial coordinate matching); the optimized data values ​​in the time-frequency anomaly spectrum set are used to replace the old values ​​of the corresponding grid nodes in the original 3D fusion data volume of the TFEM dataset; all the replaced data are integrated to generate and output the optimized 3D fusion data volume.

[0042] In a preferred embodiment of the present invention, step S4 above, based on the optimized three-dimensional fused data volume, establishes a mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, quantifies the correlation between multi-parameter combinations and mineralization indicators, and screens qualified anomaly areas, ultimately delineating a three-dimensional distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist, which may include: Step S41: Based on the optimized 3D fused data volume, calculate the horizontal gradient of the magnetic field strength at each depth layer, filter continuous regions with gradient values ​​exceeding the set magnetic gradient threshold to obtain magnetic anomaly gradient zones, and mark them as skarn boundaries; extract closed regions with resistivity values ​​below the set low resistivity threshold and mark them as low resistivity regions, extract closed regions with polarizability values ​​above the set high polarizability threshold and mark them as high polarizability anomaly regions; associate the magnetic anomaly gradient zones with skarn boundaries, associate spatially overlapping low resistivity regions and high polarizability anomaly regions with zinc-copper mineralization, and generate anomaly-geological body mapping models; Step S42: Based on the anomaly-geological body mapping model, extract the magnetic susceptibility, resistivity, and polarizability values ​​of each node within the anomaly region to generate a set of node physical property parameters; based on the set of node physical property parameters, calculate the geometric mean of the three parameters as the combined parameter value for each node to generate a set of combined parameters; based on the set of combined parameters, using borehole core analysis grade data as a benchmark, calculate the Pearson correlation coefficient between the combined parameter values ​​and the mineralization grade to generate a preliminary set of related regions; based on the preliminary set of related regions, screen anomaly regions with correlation coefficients exceeding a set correlation threshold and output a set of qualified anomaly regions. In this embodiment of the invention, by selectively extracting key physical property anomalies, namely magnetic gradient, low resistivity, and high polarization, and establishing a mapping relationship with geological bodies, discrete geophysical anomalies are transformed into models that can be directly associated with geological features. This improves the geological specificity of anomaly interpretation and provides a clear geophysical-geological correspondence for mineralization body identification. By combining parameters to quantify multiple physical property characteristics and combining borehole grade data to calculate correlations, anomaly areas highly correlated with mineralization indicators are screened out, reducing the influence of subjective experience on anomaly screening and improving the effectiveness of anomaly areas.

[0043] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S41 above, the optimized three-dimensional fusion data volume is read. This data volume contains continuous distribution data of physical property parameters such as magnetic field strength, resistivity, and polarizability at different depths in the mining area.

[0044] Magnetic anomaly gradient zone extraction and labeling: For the magnetic field strength data of each depth layer (horizontal layer divided according to exploration depth) in the three-dimensional fused data volume, calculate its horizontal gradient value (i.e. the rate of change of magnetic field strength in horizontal space; the larger the gradient value, the more drastic the change of magnetic field strength).

[0045] Set a magnetic gradient threshold (determined based on the characteristics of the background magnetic field fluctuations in the mining area and the intensity of the magnetic field anomalies related to mineralization, used to distinguish significant anomalies from background noise), and screen out areas where the gradient value exceeds the threshold; at the same time, determine whether these areas are continuous areas (i.e., through spatial continuity analysis, exclude isolated and discrete noise points to ensure that the gradient anomalies within the area are continuous).

[0046] The continuous area that meets the criteria is marked as a "magnetic anomaly gradient zone". Based on the geological characteristics of the mining area (it is known that skarnification often develops in the contact zone of the rock mass, and the contact zone is prone to sudden changes in magnetic field strength due to the enrichment of magnetic minerals, forming a gradient zone), the magnetic anomaly gradient zone is further associated and marked as a "skarnification boundary".

[0047] Extraction and labeling of low resistivity and high polarization anomaly zones: From the resistivity data of the 3D fused data volume, a low resistivity threshold is set (determined based on the resistivity difference between the mineralized body and the surrounding rock; zinc-copper mineralized bodies usually exhibit low resistivity due to the presence of conductive minerals). "Closed regions" (i.e., continuous regions with complete boundaries and internal resistivity all below the threshold, ensuring that they are independent geological bodies with clear boundaries) with resistivity values ​​below this threshold are identified and labeled as "low resistivity zones".

[0048] From the polarizability data, a high polarization threshold is set (determined based on the polarization characteristics of the mineral body; zinc-copper mineral bodies usually exhibit high polarizability due to electrochemical effects). "Closed regions" with polarizability values ​​higher than this threshold are identified (also requiring complete boundaries and internal polarizability values ​​higher than the threshold), and these are marked as "high polarization anomaly regions".

[0049] The establishment of mapping relationships involves associating geophysical anomalies with specific geological body types. Based on the theoretical correspondence between physical properties and geological characteristics, this is achieved through spatial analysis and specifically includes two parts: The three-dimensional spatial location of the magnetic anomaly gradient zone (including the range of X, Y plane coordinates and Z depth coordinates) is clearly defined. Based on the geological background of the mining area, skarnification is a product of alteration at the contact zone between the rock mass and the surrounding rock. Significant differences in magnetic properties exist between the rocks on both sides of the contact zone (e.g., fewer magnetic minerals on the rock mass side and enriched magnetic minerals on the surrounding rock side due to alteration), easily leading to abrupt changes in magnetic field strength at the contact zone, forming a magnetic anomaly gradient zone. Through spatial coordinate matching, the distribution range of the magnetic anomaly gradient zone is verified to be consistent with the theoretical development location of the skarnification boundary (e.g., the inferred range of the rock mass contact zone), confirming a high degree of spatial overlap. Based on the correspondence between the above physical properties and geological characteristics, the geophysical anomaly of the "magnetic anomaly gradient zone" is directly mapped to the geological body of the "skarnification boundary," establishing the first mapping relationship.

[0050] The three-dimensional spatial ranges (X, Y, Z coordinate ranges) of the low-resistivity zone and the high-polarization anomaly zone are determined separately. Through spatial overlay analysis, the overlapping part of the two regions is calculated—that is, the region that belongs to both the low-resistivity zone (resistivity below the threshold) and the high-polarization anomaly zone (polarizability above the threshold) at the same spatial location. Combining the physical properties of zinc-copper mineralizations: zinc-copper mineralizations, due to the presence of conductive minerals such as sulfides, usually exhibit low resistivity (consistent with the characteristics of low-resistivity zones); at the same time, the electrochemical difference between sulfides and the surrounding rock leads to high polarizability (consistent with the characteristics of high-polarization anomalies), that is, zinc-copper mineralizations naturally possess a combination of "low resistance + high polarization". Based on the correspondence between the above physical property combination and the mineralization, the overlapping area of ​​the low-resistivity zone and the high-polarization anomaly zone is mapped to "zinc-copper mineralization", establishing a second mapping relationship.

[0051] Integrating the above two mapping relationships (magnetic anomaly gradient zone → skarn boundary, low resistivity-high polarization overlap zone → zinc-copper mineralization), an "anomaly-geological body mapping model" is formed, which includes the correspondence between geophysical anomalies and specific geological body types. In step S42 above, based on the anomaly-geological body mapping model, the magnetic susceptibility, resistivity, and polarizability values ​​of each grid node in all anomaly regions (magnetic anomaly gradient zone, low resistivity zone, high polarization anomaly zone, etc.) are extracted and organized into a set of node physical property parameters. For each node, the geometric mean of the three parameters of magnetic susceptibility, resistivity, and polarizability is calculated (comprehensively reflecting the overall characteristics of the three parameters), and the result is used as the combined parameter value of the node to generate a set of combined parameters.

[0052] Collect the chemical grade data of drill cores in the mining area (as a benchmark for mineralization indicators), calculate the Pearson correlation coefficient (measures the degree of linear correlation between the combined parameter value of each node in the combined parameter set and the mineralization grade of the corresponding location), and organize it into a preliminary set of related regions; set a correlation threshold, screen out abnormal regions with correlation coefficients exceeding the threshold from the preliminary set of related regions, and output a set of qualified abnormal regions (i.e., regions with high correlation between combined parameters and mineralization indicators).

[0053] In a preferred embodiment of the present invention, step S4 above, based on the optimized three-dimensional fused data volume, establishes a mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, quantifies the correlation between multi-parameter combinations and mineralization indicators, and screens qualified anomaly areas, ultimately delineating a three-dimensional distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist, and may further include: Step S43: Based on the set of qualified anomaly regions, extract three-dimensional spatial units that simultaneously contain magnetic anomaly gradient bands, low resistivity regions and high polarization anomaly regions to generate a ternary anomaly co-occurrence region. Step S44: Based on the ternary anomaly co-occurrence region, fill the internal holes to eliminate data discontinuity, smooth the boundaries to regularize the target region outline, and output the optimized target region unit set. Step S45: Based on the optimized target area unit set, extract the combined parameter values ​​of each target area unit, divide the target area into level A, level B, and level C according to the parameter value, and generate a three-dimensional distribution map of graded mineral exploration target areas.

[0054] In this embodiment of the invention, by extracting the co-occurrence zone of ternary anomalies, focusing on the superimposed area of ​​"magnetic anomaly gradient zone + low resistivity zone + high polarization anomaly zone", the synergistic indicative effect of the three anomalies significantly reduces false positive interference caused by a single anomaly, improving the accuracy of target area positioning. By filling voids and smoothing boundaries, the discontinuity of the target area caused by data acquisition errors or noise is eliminated, making the target area outline more regular and the internal data more complete, thus enhancing the reliability of the target area spatial distribution. By classifying the combined parameter values, the priority of the mineral exploration potential of different target areas is clarified, allowing exploration resources to be tilted towards Class A target areas, reducing inefficient exploration investment. At the same time, the three-dimensional distribution map intuitively displays the spatial location and level of the target area, providing clear and operable decision support for exploration engineering deployment and improving mineral exploration efficiency.

[0055] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step S43 above, the set of qualified anomaly regions is read. This set contains anomaly region data that has been filtered for correlation and is highly correlated with mineralization indicators. From the set of qualified anomaly regions, the three-dimensional spatial ranges of the magnetic anomaly gradient zone, low resistivity zone, and high polarization anomaly zone are extracted respectively (each region includes the X and Y plane coordinate ranges and the Z depth range). Through spatial overlay analysis, the intersection of the three regions is calculated: that is, the three-dimensional spatial units that simultaneously belong to the magnetic anomaly gradient zone, low resistivity zone, and high polarization anomaly zone are found (each unit is a spatial volume with a clear coordinate boundary). These spatial units that simultaneously contain the three types of anomalies are organized to generate the "ternary anomaly co-occurrence area" (i.e., the set of regions where the three types of anomalies co-occur in space).

[0056] In step S44 above, the data of the ternary anomaly co-occurrence region is read. This region may contain internal voids (i.e., locally uncovered blank areas within the co-occurrence region) and irregular boundaries (jagged or irregular edges caused by the discreteness of the original data) due to incomplete data acquisition or noise. Void regions within the co-occurrence region are identified (the judgment criterion is that the surrounding area is a co-occurrence region but there is no data inside). Based on the physical property parameters (such as combined parameter values) of the adjacent units around the void, the void is filled by interpolation or assignment to eliminate data discontinuities. The edge contour of the co-occurrence region is processed by calculating the average parameter values ​​of the boundary nodes and adjacent normal units, and adjusting the position or attributes of the boundary nodes to make the originally irregular boundaries smooth and continuous, and regularize the target area contour. The processed region is integrated, and the optimized target area unit set (a set of three-dimensional spatial units with continuous internal structure and regular boundaries) is output.

[0057] In step S45 above, the optimized target unit set is read, and the combined parameter values ​​of each target unit are extracted (i.e., the geometric mean of the three parameters, magnetic susceptibility, resistivity, and polarizability, calculated in step S42). The combined parameter values ​​of all target units are statistically sorted to determine the grading threshold (e.g., the first 30% are grade A, 30%-70% are grade B, and the last 30% are grade C, based on the parameter values ​​from high to low; the specific threshold can be adjusted according to the mineralization intensity distribution). Each target unit is divided into grade A (highest mineral exploration potential), grade B (medium mineral exploration potential), and grade C (lower mineral exploration potential) according to the magnitude of the combined parameter values. The grading results are associated with the three-dimensional spatial coordinates to generate a "grading mineral exploration target area three-dimensional distribution map" containing the location, range, and grade of each target unit.

[0058] like Figure 2 As shown, embodiments of the present invention also provide a multi-source geophysical data fusion-based contact zone anomaly system for the Bangzhong zinc-copper ore mass, comprising: The acquisition module is used to set a first sampling interval and a first measurement accuracy threshold based on the geological characteristics of the mining area, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set a first transmission power threshold and a first frequency range based on the differences in deep electrical properties, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set a target transmission frequency sequence based on the differences in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. The fusion module is used to unify the high-resolution magnetic data set, CSAMT data set, TFEM data set and the existing induced polarization data set in the mining area by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting the fusion to allocate the property sensitivity weights, so as to generate a three-dimensional fused data volume containing multi-attribute features. The optimization module is used to perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume. The target area delineation module is used to establish a mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies based on the optimized 3D fusion data volume. It quantifies the correlation between multi-parameter combinations and mineralization indicators, screens qualified anomaly areas, and finally delineates a 3D distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist.

[0059] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying contact zone anomalies in the Bangzhong zinc-copper ore mass through multi-source geophysical data fusion, characterized in that, The method includes: Step S1: Based on the geological characteristics of the mining area, set a first sampling interval and a first measurement accuracy threshold, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set a first transmission power threshold and a first frequency range according to the differences in deep electrical properties, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set a target transmission frequency sequence for the differences in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. Step S2: The high-resolution magnetic data set, CSAMT data set, TFEM data set, and the existing induced polarization data set in the mining area are combined by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting and fusion to allocate the material property sensitivity weights, to generate a three-dimensional fused data volume containing multi-attribute features. Step S3: Perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume; Step S4: Based on the optimized three-dimensional fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified and qualified anomaly areas are screened. Finally, a three-dimensional distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated.

2. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 1, characterized in that, Step S1, synchronously performed in the contact zone area of ​​the zinc-copper ore rock mass in the Bangzhong mine: based on the geological characteristics of the mining area, a first sampling interval and a first measurement accuracy threshold are set, high-precision magnetic measurement is implemented, and a magnetic dataset containing magnetic field strength data is output; based on the differences in deep electrical properties, a first transmission power threshold and a first frequency range are set, CSAMT detection is implemented, and a CSAMT dataset containing resistivity data is output. A target emission frequency sequence was set based on the difference in electromagnetic response between the mineralized body and the surrounding rock. Time-frequency electromagnetic measurements were performed, and a TFEM dataset containing resistivity and polarizability parameters was output, including: Based on the fold and fault structure characteristics of the mining area, a proton magnetometer is used to set the first sampling interval to adapt to the local changes in the magnetic field caused by structural deformation, and a first measurement accuracy threshold is set to identify weak magnetic anomalies. A high-resolution magnetic dataset containing magnetic field strength is output. Based on the structural framework revealed by the high-resolution magnetic data set, the CSAMT device is used to set a first transmit power threshold according to the electrical differences of deep strata to ensure the deep signal penetration capability, and a first frequency range is set to match the contact electrical structural response of the rock mass, and the output is a CSAMT data set containing resistivity. Based on the deep electrical background provided by the CSAMT dataset, time-frequency electromagnetic measurements were performed. A target broadband transmission sequence was set to target the difference in electromagnetic response between the mineralized body and the surrounding rock. The first low-frequency band captured the slowly varying response of the deep mineralized body, and the second high-frequency band captured the transient response of the shallow contact zone. The output was a TFEM dataset containing resistivity and polarizability parameters.

3. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 2, characterized in that, Step S2 involves unifying the high-resolution magnetic dataset, CSAMT dataset, TFEM dataset, and existing induced polarization dataset from the mining area using geographic coordinate unification to eliminate terrain offset, standardizing and transforming multi-source parameters using physical dimension normalization, and adaptively weighting and fusion to allocate material property sensitivity weights, generating a three-dimensional fused data volume containing multi-attribute features, including: Using high-resolution magnetic data sets, CSAMT datasets, TFEM datasets, and mining area induced polarization datasets as input, the positioning coordinates of all datasets are extracted, and the horizontal projection deviation is calculated based on the terrain elevation gradient. Based on the projection deviation, the horizontal coordinate offset is corrected according to the elevation gradient distribution so that the corrected coordinate deviation does not exceed the set spatial accuracy threshold. The corrected coordinate data is resampled to a unified grid node, and the spatially aligned dataset is output. Based on the spatially aligned dataset, and taking the mean background magnetic field strength of the mining area as the benchmark, the first magnetic parameter value of each grid node is calculated. The value is the difference between the magnetic field strength of that point and the mean background value divided by the mean background value. The resistivity value of each node is logarithmically calculated to generate the second electrical parameter value. The probability distribution of the polarizability value of the entire mining area is statistically analyzed. The polarizability value of each node is normalized to the 0-1 interval by subtracting the minimum value of the entire mining area and dividing by the difference between the maximum value and the minimum value, generating the third induced polarization parameter value. Finally, a standardized matrix containing the first magnetic parameter, the second electrical parameter, and the third induced polarization parameter is output.

4. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 3, characterized in that, Step S2 involves unifying the high-resolution magnetic dataset, CSAMT dataset, TFEM dataset, and existing induced polarization dataset from the mining area using geographic coordinate unification to eliminate terrain offset, standardizing and transforming multi-source parameters using physical dimension normalization, and adaptively weighting and fusion to allocate material property sensitivity weights, generating a three-dimensional fused data volume containing multi-attribute features. This also includes: Based on the standardized matrix, each grid node is traversed, and the contact zone boundary area, mineralization anomaly area and ordinary area are marked according to the threshold conditions of the first magnetic parameter, the second electrical parameter and the third induced polarization parameter, respectively, to generate a partition marking result set; Based on the partitioning labeling result set, for nodes labeled as contact zone boundary areas, the first magnetic weight coefficient assigned to the node is output; for nodes labeled as mineralization anomaly areas, the second electrical weight coefficient and the third polarization weight coefficient assigned to the node are output; for nodes labeled as ordinary areas, the equal weight coefficient assigned to the node is output; based on the first magnetic weight coefficient, the second electrical weight coefficient, the third polarization weight coefficient, and the equal weight coefficient, a weight allocation result set is generated. Based on the weighting results and the standardization matrix, the first magnetic parameter value, the second electrical parameter value, and the third induced polarization parameter value of the node are extracted. The magnetic weighting coefficient is multiplied by the first magnetic parameter value to obtain the magnetic weighted value. The electrical weighting coefficient is multiplied by the second electrical parameter value to obtain the electrical weighted value. The polarization weighting coefficient is multiplied by the third induced polarization parameter value to obtain the polarization weighted value. The magnetic weighted value, electrical weighted value, and polarization weighted value are added together to output the fusion value of the node and generate a set of node fusion values. Based on the node fusion value set, the grid node planar coordinates are used as the X and Y axis positioning references, and the exploration depth is used as the Z axis layering basis. The node fusion values ​​are extended to a continuous three-dimensional space through spatial interpolation to generate a three-dimensional fusion data volume containing multiple attribute features.

5. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 4, characterized in that, Step S3: Perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume, including: Based on the TFEM dataset in the 3D fused data volume, wavelet transform threshold denoising is performed on the high-frequency components above 100Hz to filter out random noise with amplitude less than twice the standard deviation of background noise. Empirical mode decomposition is performed on the low-frequency components of 0.1-10Hz to remove baseline drift components and output the denoised and reconstructed TFEM dataset. Based on the denoised and reconstructed TFEM dataset, within the sensitive frequency band of the rock mass contact zone, a first low-frequency sub-segment is captured using a longer time window to capture deep, slowly varying responses, and a second high-frequency sub-segment is captured using a shorter time window to capture shallow, transient anomalies, generating the time spectrum of the sensitive frequency band. Based on the time spectrum, the mean amplitude of the frequency point over the entire time period is calculated as a background benchmark, and the instantaneous amplitude is divided by the background benchmark to generate a relative energy coefficient. Based on the relative energy coefficient, the time spectra of anomalies exceeding a set anomaly threshold are extracted, and a high-resolution time-frequency anomaly spectrum set is output. Based on the high-resolution time-frequency anomaly spectrum set, the time-frequency anomaly spectrum set is mapped back to the original TFEM data nodes, replacing the corresponding node values ​​of the TFEM dataset in the original 3D fusion data volume, and outputting the optimized 3D fusion data volume.

6. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 5, characterized in that, Step S4: Based on the optimized 3D fused data volume, by establishing the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, the correlation between multi-parameter combinations and mineralization indicators is quantified, and qualified anomaly areas are screened. Finally, a 3D distribution map of the prospecting target area where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist is delineated, including: Based on the optimized 3D fused data volume, the horizontal gradient of the magnetic field intensity at each depth layer is calculated. Continuous regions with gradient values ​​exceeding a set magnetic gradient threshold are selected to obtain magnetic anomaly gradient zones, which are marked as skarn boundaries. Closed regions with resistivity values ​​below a set low resistivity threshold are extracted and marked as low resistivity zones. Closed regions with polarizability values ​​above a set high polarizability threshold are extracted and marked as high polarizability anomaly zones. The magnetic anomaly gradient zones are associated with skarn boundaries, and spatially overlapping low resistivity zones and high polarizability anomaly zones are associated as zinc-copper mineralization bodies, generating anomaly-geological body mapping models. Based on the anomaly-geological body mapping model, the magnetic susceptibility, resistivity, and polarizability values ​​of each node within the anomaly region are extracted to generate a set of node physical property parameters. Based on the set of node physical property parameters, the geometric mean of the three parameters is calculated as the combined parameter value for each node to generate a set of combined parameters. Based on the set of combined parameters, using borehole core analysis grade data as a benchmark, the Pearson correlation coefficient between the combined parameter values ​​and the mineralization grade is calculated to generate a preliminary set of related regions. Based on the preliminary set of related regions, anomaly regions with correlation coefficients exceeding a set correlation threshold are selected, and a set of qualified anomaly regions is output.

7. The method for detecting contact zone anomalies in the Bangzhong zinc-copper ore mass by fusing multi-source geophysical data according to claim 6, characterized in that, Step S4, based on the optimized 3D fused data volume, establishes the mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies, quantifies the correlation between multi-parameter combinations and mineralization indicators, and screens qualified anomaly areas. Finally, it delineates a 3D distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist. This also includes: Based on the set of qualified anomaly regions, three-dimensional spatial units that simultaneously contain magnetic anomaly gradient bands, low resistivity regions and high polarization anomaly regions are extracted to generate ternary anomaly co-occurrence regions. Based on the ternary anomaly co-occurrence region, internal holes are filled to eliminate data discontinuity areas, and boundaries are smoothed to regularize the target region outline, outputting an optimized target region unit set. Based on the optimized target area unit set, the combined parameter values ​​of each target area unit are extracted, and the target areas are divided into Grade A, Grade B, and Grade C according to the parameter value, generating a three-dimensional distribution map of graded mineral exploration target areas.

8. A multi-source geophysical data fusion system for contact zone anomalies in the Bangzhong zinc-copper ore mass, the system implementing the method as described in any one of claims 1 to 7, characterized in that, include: The acquisition module is used to set a first sampling interval and a first measurement accuracy threshold based on the geological characteristics of the mining area, implement high-precision magnetic measurement, and output a high-resolution magnetic dataset containing magnetic field strength data; set a first transmission power threshold and a first frequency range based on the differences in deep electrical properties, implement CSAMT detection, and output a CSAMT dataset containing resistivity data; set a target transmission frequency sequence based on the differences in electromagnetic response between the mineralized body and the surrounding rock, implement time-frequency electromagnetic measurement, and output a TFEM dataset containing resistivity and polarizability parameters. The fusion module is used to unify the high-resolution magnetic data set, CSAMT data set, TFEM data set and the existing induced polarization data set in the mining area by unifying the geographic coordinates to eliminate terrain offset, standardizing the physical dimensions to transform the multi-source parameters, and adaptively weighting the fusion to allocate the property sensitivity weights, so as to generate a three-dimensional fused data volume containing multi-attribute features. The optimization module is used to perform joint denoising and time-frequency feature enhancement on the TFEM dataset in the 3D fused data volume, and output the optimized 3D fused data volume. The target area delineation module is used to establish a mapping relationship between magnetic, electrical, and polarization anomalies and geological bodies based on the optimized 3D fusion data volume. It quantifies the correlation between multi-parameter combinations and mineralization indicators, screens qualified anomaly areas, and finally delineates a 3D distribution map of prospecting target areas where magnetic anomaly gradient zones, low resistivity zones, and high polarization anomalies coexist.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images