Gold mine air-ground-in-well rapid collaborative exploration method

Abstract

The invention relates to a gold mine air-ground-in-well rapid collaborative exploration method, belongs to the field of geological exploration, and aims to solve the technical problems of data splitting, low target area positioning precision, insufficient deep prediction capability and the like in traditional gold mine exploration. By integrating air remote sensing, surface geological survey and underground detection data, a multi-source data standardization processing flow is established, and a progressive exploration system from wide-area screening to accurate positioning is realized by combining geological, geophysical and geochemical comprehensive analysis methods. The system adopts a dynamic feedback mechanism to optimize an exploration process, forms closed-loop analysis and prediction, improves exploration efficiency and accuracy, and reduces resource waste and exploration cost. The method is suitable for gold ore resource exploration under complex geological conditions, has remarkable economic benefits and application value, and can be widely applied to the field of mineral resource development.

Description

Technical Field

[0001] The present invention belongs to the field of geological exploration, and more particularly relates to a method for rapid collaborative exploration of gold mines in the air, on the ground, and in wells. Background Art

[0002] Mineral resources are an essential material foundation for the development of human society. Gold, as a strategic mineral resource, plays an irreplaceable role in the economy, industry, science, and technology. With the increasing depletion of global gold resources and the increasing demand for deep mineral exploration, traditional gold exploration methods face numerous technical challenges. Current exploration technologies primarily include remote sensing, geochemical exploration, geophysical exploration, and drilling. However, these methods suffer from practical problems such as data fragmentation, inaccurate target location, and difficulty in deep prediction. An efficient and coordinated exploration technology system is urgently needed to improve the efficiency and accuracy of gold exploration.

[0003] Traditional gold exploration techniques typically rely on a single data source or exploration method, making it difficult to fully reveal the mineralization patterns of gold deposits under complex geological conditions. Remote sensing technology is primarily used to identify geological structures over large areas, but its spatial resolution and depth detection capabilities are limited. Geochemical exploration primarily identifies potential areas through elemental content anomalies, but cannot directly provide spatial information about ore bodies. Geophysical exploration can detect underground physical anomalies, but its accuracy is insufficient for predicting deep ore bodies. Drilling technology can directly obtain ore body information, but its cost is high, and target selection relies on the accuracy of previous exploration results. Due to the limitations of these technologies, the traditional gold exploration process often lacks systematicity and coordination, resulting in low exploration efficiency and large deviations in target positioning.

[0004] In gold mine exploration, the heterogeneity of multi-source data, including remote sensing, geochemical, geophysical, and drilling data, makes data integration and collaborative analysis difficult. Different data sources vary significantly in spatial resolution, measurement scale, and physical meaning. For example, remote sensing data primarily reflects surface information, while geophysical data focuses on subsurface physical anomalies, and geochemical data focuses on the distribution of elemental content. Traditional methods typically process these data separately, lacking unified spatial registration and systematic integration, leading to information loss or interpretation bias. Furthermore, a dynamic feedback mechanism between data has not yet been established, making it impossible to optimize early prediction models based on subsequent verification results during the exploration process, further limiting exploration accuracy. Summary of the Invention

[0005] The present invention addresses the problems of data fragmentation, low target positioning accuracy, insufficient deep prediction capability, and lack of dynamic feedback mechanism in traditional gold mine exploration methods. This invention proposes an air-ground-well rapid collaborative exploration method. Through the standardized integration of multi-source data, comprehensive analysis of geology, geophysics, and geochemistry, and a closed-loop optimization process, an efficient exploration system from wide-area screening to precise positioning is realized, which improves the positioning accuracy of gold mine target areas and deep prediction capability, while reducing exploration costs and risks.

[0006] In order to achieve the above object, the present invention is implemented by adopting the following technical solutions: the method comprises:

[0007] Data collection phase: aerial survey, ground survey and well survey are carried out in sequence to obtain remote sensing image data, aerial geophysical data, ground geophysical data, geochemical samples and drill core data;

[0008] Aerial surveys use satellite remote sensing technology and airborne geophysical surveys to delineate wide-area potential areas; ground surveys deploy high-precision geophysical networks and geochemical sampling points within aerial anomaly areas to further narrow the target area.

[0009] Well exploration obtains core samples and well geophysical data through controlled drilling to determine the spatial distribution of ore bodies and the depth of mineralization;

[0010] Data processing stage: standardization and anomaly extraction of multi-source data from air, ground and well, including remote sensing image correction, geochemical element normalization, geophysical background field stripping and well logging data normalization;

[0011] Comprehensive analysis stage: Based on the correlation modeling of geological, geophysical and geochemical data, a three-dimensional prediction model is constructed to quantify the mineralization probability field.

[0012] Target area determination stage: The mineralization target area is graded by combining geological structure, geophysical anomalies and geochemical element combinations through a multi-factor weighted evaluation method.

[0013] Verification and evaluation stage: Verify the target area prediction results through drilling, combine them with the three-dimensional model, and estimate the resource volume and economically recoverable reserves; at the same time, improve the prediction accuracy through iterative optimization of the model.

[0014] In one embodiment, the aerial survey uses satellite remote sensing technology to obtain multispectral, thermal infrared and radar image data, and combines aerial magnetic and electromagnetic measurements to identify geological structures and physical anomaly areas.

[0015] In one plan, induced polarization and transient electromagnetic methods are used for geophysical measurements during the ground exploration phase, and geochemical sampling points are increased to analyze the content of elements such as Au, As, and Sb. At the same time, geological mapping is carried out at a scale of 1:10,000 to 1:2,000.

[0016] In one plan, during the downhole exploration phase, a diamond drill rig is used to construct controlled boreholes, and downhole magnetic surveys, gamma ray spectrum logging, and high-density three-dimensional seismic surveys are conducted in the boreholes. At the same time, core samples are collected for rock and mineral identification and elemental geochemical analysis.

[0017] In one solution, the comprehensive analysis stage quantifies the correlation between stratum lithology, geophysical parameters and geochemical elements through covariance matrix analysis and principal component coupling analysis, and screens key mineral-controlling factors.

[0018] In one scheme, the entropy weight-hierarchy analysis method is used in the target area determination stage to conduct weighted evaluation of geological structure complexity, geophysical anomaly intensity, geochemical element combination entropy and alteration mineral aggregation to generate a mineralization target area distribution map.

[0019] In one scheme, during the verification and evaluation phase, a three-dimensional solid model of the mineralized body was constructed using the Kriging interpolation algorithm, and the resource volume was estimated using the inverse distance power method. At the same time, the prediction model was dynamically updated in combination with the while-drilling logging data.

[0020] In one solution, the entire exploration process relies on the GIS platform to achieve real-time integration and spatial registration of air, ground and well data, and optimize the exploration path through a dynamic feedback mechanism to improve the adaptability and prediction accuracy of the regional prospecting model.

[0021] Beneficial effects of the present invention:

[0022] By constructing a three-level, progressive, rapid, collaborative exploration system—air, ground, and well—this invention achieves standardized integration and comprehensive analysis of multi-source data, significantly improving the accuracy of gold target positioning and deep prediction capabilities. By optimizing the exploration process through a dynamic feedback mechanism, a closed-loop optimization system is formed, reducing deviations and resource waste during the exploration process, and lowering the cost and risk of drilling verification. Furthermore, this invention can efficiently identify mineralization patterns under complex geological conditions, providing technical support for the scientific development and rational utilization of mineral resources, with high economic benefits and broad application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0024] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate exemplary embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those understood by those skilled in the art to which the present invention pertains. The terms used in the present specification are for the purpose of describing specific embodiments only and are not intended to limit the present invention. To facilitate understanding of the present invention, a more comprehensive description of the present invention will be provided below with reference to the accompanying drawings. Typical embodiments of the present invention are shown in the drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0026] like Figure 1 The figure shows a rapid collaborative air-surface-in-hole exploration method for gold mines. This method systematically integrates and collaboratively analyzes multi-source data from remote sensing, geochemical exploration, geophysical exploration, and drilling. This three-level progressive air-surface-in-hole exploration method achieves a complete process from large-scale screening to precise positioning.

[0027] S1. Data collection phase: Collect multi-source data in the order of air-surface-well.

[0028] During the data acquisition phase (S1), the method strictly follows a three-level progression from airborne to surface to borehole, collecting multi-source data to form a systematic exploration system from broad to local, and from shallow to deep. First, aerial surveys are used to rapidly screen large areas. Satellite remote sensing technology is used to acquire regional multispectral, thermal infrared, and radar image data. Combined with aeromagnetic and electromagnetic measurements, these methods identify geological structures (fault zones, alteration zones) and areas of physical anomalies associated with gold mineralization, initially delineating potential areas spanning tens to hundreds of square kilometers.

[0029] Then it entered the ground exploration stage, where a high-precision ground geophysical network (induced polarization method, transient electromagnetic method) was arranged in the aerial anomaly area, and geochemical sampling points were encrypted (soil and rock samples were collected to analyze the content of elements such as Au, As, and Sb). Simultaneously, geological mapping with a scale of 1:10,000 to 1:2,000 was carried out to depict the strata, structure, and mineralized outcrop information in detail, further reducing the aerial anomaly area to a target area of several square kilometers to hundreds of meters.

[0030] Finally, in-well exploration was conducted in key target areas identified by ground exploration. Diamond drill rigs were used to construct controlled boreholes, and in-well geophysical exploration (in-well magnetic surveys and gamma spectroscopy logging) and high-density three-dimensional seismic exploration were conducted within the boreholes. Core samples were systematically collected for rock and mineral identification, elemental geochemical analysis, and isotope dating, to obtain key parameters such as the spatial distribution of ore bodies, alteration zoning, and mineralization depth. The entire data collection process relied on a GIS platform to achieve real-time integration and spatial registration of airborne, ground-based, and in-well data, ensuring seamless integration of data from different scales and sources within a unified coordinate system. A dynamic feedback mechanism (ground geophysical results reversely corrected aerial anomaly interpretations) enabled closed-loop optimization of the exploration path, ultimately constructing a multi-dimensional exploration data volume covering the surface to the kilometer-scale underground, laying the foundation for subsequent collaborative analysis.

[0031] S2. Data processing stage: standardize the data at each level and extract anomalies.

[0032] In the data processing stage (S2), in view of the heterogeneous characteristics of the multi-level data from air, ground and well, a systematic standardization process is first carried out to eliminate the differences in data scale and dimension. For remote sensing data, radiometric correction is used (formula: ) and geometric correction, affine transformation:

[0033] x'=a0+a1x+a2y

[0034] y'=b0+b1x+b2y

[0035] Atmospheric interference and terrain distortion were eliminated, and the spatial resolution was unified to meter-level grid by resampling. Geochemical data were normalized by element content (logarithmic transformation: C norm =log 10 (C raw +1) to weaken the influence of extreme values, and use Kriging interpolation (semivariogram function:

[0036]

[0037] Generate a continuous element concentration distribution map. Geophysical data is stripped by background field (polynomial fitting: ) Eliminate regional field interference and retain local abnormal signals (residual field: ΔF=F obs -B), and depth-thickness normalization (integral normalization: ) to match the ground data scale.

[0038] In the anomaly extraction process, multimodal algorithms are used to mine hidden mineral signals. For geochemical data, based on the element combination index Identify Au-As-Sb and other nesting anomalies, where w i is the element weight, The geophysical anomaly is segmented by adaptive threshold (T = μ + kσ, k∈[2,3]) to delineate the high gradient zone, and the edge detection operator (Sobel convolution kernel:

[0039]

[0040] Strengthen the structural boundary. To solve the problem of weak anomaly identification, the isolation forest algorithm is introduced (anomaly score: in H is the harmonic number) to detect low signal-to-noise ratio signals and perform principal component analysis (eigendecomposition of the covariance matrix: Σ = PDP -1 Compress redundant information and extract the first three principal components (cumulative variance contribution rate > 85%) as comprehensive abnormality indicators. Finally, use Bayesian probability fusion

[0041]

[0042] Integrate multi-source anomaly results to generate a probabilistic anomaly target map, providing quantitative input for subsequent model construction. All processing processes are implemented in a distributed computing framework through data cubes. Integrate information in time and space dimensions to support collaborative interpretation of multi-scale anomalies.

[0043] S3, comprehensive analysis stage: Establishment of geological-geophysical-geochemical correlation model. In the comprehensive analysis stage (S3), based on the standardized data and abnormal characteristics obtained in the S1-S2 stages, a geological-geophysical-geochemical (3G) joint interpretation model is established through multimodal data coupling and correlation modeling technology. First, a spatial registration algorithm (affine transformation residual optimization: Where T is the transformation matrix, p i ,q i As control points, the spatial superposition of geological mapping units, geophysical anomaly areas and geochemical element enrichment zones is realized, and a three-dimensional data cube is constructed.

[0044] M(x,y,z)={G lith ,Φ geophy ,C geochem}

[0045] For correlation modeling, covariance matrix analysis was used Quantifying formation lithology (G lith ) and geophysical parameters (polarizability η, magnetic susceptibility κ) and geochemical element combinations (Au-As-Cu), and the associated variable group with a correlation coefficient |ρ|>0.6 was screened out.

[0046] Further analysis of principal component coupling (coupled principal component: PC fusion =α·PC geophy+β·PC geochem +γ·PC geol , where the weights α, β, and γ are determined by the proportion of eigenvalues) to construct a cross-scale correlation index to reveal the spatial response relationship between deep structure (fault zone occurrence θ) and surface element zoning (As / Sb ratio). For nonlinear correlation, random forest feature importance evaluation (Gini impurity reduction: ) quantify the contribution of geochemical anomalies (log(Au)>2σ) to geophysical inversion parameters (resistivity ρ<100Ω·m) and screen key mineral control factors. At the same time, based on Kriging collaborative simulation (covariogram function:

[0047]

[0048] Realize the drilling core data (alteration intensity A alter ) and ground geophysical profiles (gravity anomaly Δg) to generate a three-dimensional correlation field of mineralization alteration-physical property parameters.

[0049] In order to characterize the multi-factor coupling mechanism of the mineralization process, a structural equation model (SEM) was established: η = Γξ + ζ, where the latent variable ξ represents geological processes such as tectonic stress field and magmatic activity, and the observed variable η includes parameters such as geophysical anomaly intensity and element zoning index. The path coefficient Γ is estimated by maximum likelihood (likelihood function: L = log|Σ| + tr(SΣ -1 )) Solve. For deep ore body prediction, a joint inversion algorithm (objective function: Φ=‖W d (d obs -F(m))‖ 2 +λ‖W m (mm prior )‖ 2 The gravity, magnetic and electrical data are constrained synchronously, where the model parameters m include density ρ, magnetic susceptibility κ and polarizability η, and the regularization parameter λ is optimized by the L-curve method. Finally, the three-dimensional implicit function modeling (radial basis function: Generate the ore body boundary surface, whose confidence is determined by the Bayesian posterior probability Quantify and form an interpretable prospecting model that integrates geological cognition and physical and chemical exploration responses to support drilling verification and resource estimation.

[0050] S4. Target area determination stage: Determine the mineralization target area based on multi-factor weighted evaluation.

[0051] In the target area determination stage (S4), based on the 3G correlation model constructed in the S3 stage, the quantitative evaluation of mineralization probability is achieved through multi-source data fusion and spatial decision-making model. First, a hierarchical evaluation index system is constructed, covering the complexity of geological structure (fracture density ), where N fis the number of faults within the unit area A), geophysical anomaly intensity (normalized gradient amplitude ), geochemical element combination entropy ( p i is the standardized concentration ratio of element i) and the spatial aggregation of alteration minerals (Ripley's K function: ) and other core parameters. The improved entropy weight-analytic hierarchy process (AHP-EWM) is used to determine the indicator weights: the subjective weights are obtained by constructing a judgment matrix (scale 1-9, consistency ratio CR<0.1) through AHP Based on information entropy Calculating objective weights The final fusion weight is (α is the preference coefficient, the default value is 0.5).

[0052] Further use of fuzzy membership function (trapezoidal function:

[0053]

[0054] After normalizing each indicator and eliminating the dimension difference, a weighted linear combination (WLC) was used. The comprehensive metallogenic favorableness score of each spatial unit (30m×30m grid) was calculated and the natural breakpoint method (Jenks optimization: minimizing the intra-class variance) was used to calculate the comprehensive metallogenic favorableness score of each spatial unit (30m×30m grid). ) The scores are divided into three target areas: A (>85% quantile), B (70%-85%), and C (50%-70%). In view of spatial uncertainty, Monte Carlo simulation is introduced (the weights are randomly perturbed in each iteration). and membership function parameters), generate 1000 sets of random fields and calculate the target area probability Screening P A >0.8 high confidence region. At the same time, based on the weight of evidence model, the posterior probability is:

[0055]

[0056] in Quantify the spatial correlation strength between known mineral points and each evidence layer, and perform Bayesian correction on WLC results Finally, a three-dimensional distribution map of mineralization target areas with dual constraints of probability and risk is generated, and the threshold is optimized through the receiver operating characteristic curve to provide a quantitative decision-making basis for exploration project deployment.

[0057] S5. Verification and evaluation stage: Drilling verification and resource potential evaluation.

[0058] In the verification and evaluation stage (S5), based on the mineralization target area delineated in the S4 stage, the exploration closed loop is achieved through engineering verification and dynamic evaluation of resource potential. First, according to the target area grade and probability distribution, an adaptive drilling grid design method is adopted to prioritize the deployment of verification drill holes in Class A target areas (high probability, low risk). The drilling trajectory and target layer are determined in combination with the three-dimensional geological-geophysical model. At the same time, the well logging technology is used to obtain core physical properties (density, resistivity) and alteration mineral combination data in real time, and dynamically compare them with the mineralization alteration-physical property correlation field in the prediction model. For the hole positions that encounter industrial ore bodies, the spatial morphology of the ore body is accurately calibrated through core sampling analysis (including grade testing and mineral paragenesis sequence identification). The Kriging interpolation algorithm is used to construct a three-dimensional solid model of the mineralized body, and the inverse distance power method is used to estimate the resource volume. The tonnage-grade distribution curve and the resource classification confidence level (proven, controlled, inferred) are simultaneously calculated. For drill holes that did not achieve the expected targets, the correspondence between geophysical parameters and geological interfaces was recalibrated through inversion correction technology (Bayesian model update), the weight distribution of structural ore-controlling elements in the 3G correlation model was optimized, and the mineralization probability field was iteratively updated.

[0059] Uncertainty propagation analysis is further introduced to comprehensively analyze the degree of fit between the drilling verification results and the original prediction model (area under the ROC curve AUC value, confusion matrix accuracy index) to quantify the credible interval of resource potential evaluation. For verified mineralized areas, the beneficiation recovery rate is determined through ore process mineralogy research, and the economically recoverable reserves are dynamically evaluated in combination with the market price fluctuation model (Monte Carlo simulation). Resource development priorities are divided based on mining technical conditions (burial depth, inclination, hydrogeology). Ultimately, a comprehensive evaluation report is formed that includes resource reserves, exploration risks, and economic feasibility. The drilling data, ore body model, and spatial distribution of prediction parameters are integrated through a three-dimensional visualization platform to support subsequent exploration engineering optimization and mine planning. The full-process data and conclusions of the verification stage are synchronously fed back to each link S1-S4, forming a dynamic exploration knowledge system of "prediction-verification-iteration" to improve the adaptability and prediction accuracy of regional prospecting models.

[0060] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0061] It should be understood that the detailed description of the technical solutions of the present invention using the preferred embodiments above is illustrative and not restrictive. A person skilled in the art, after reading the present specification, may modify the technical solutions described in the embodiments or replace some of the technical features therein with equivalents; such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rapid collaborative exploration method for gold mines in the air, on the ground, and in the well, characterized by: The method includes: Data collection phase: aerial survey, ground survey and well survey are carried out in sequence to obtain remote sensing image data, aerial geophysical data, ground geophysical data, geochemical samples and drill core data; Aerial surveys use satellite remote sensing technology and airborne geophysical surveys to delineate wide-area potential areas; For ground surveys, high-precision geophysical networks and geochemical sampling points are deployed within the aerial anomaly area to further narrow the target area; Well exploration obtains core samples and well geophysical data through controlled drilling to determine the spatial distribution of ore bodies and the depth of mineralization; Data processing stage: standardization and anomaly extraction of multi-source data from air, ground and well, including remote sensing image correction, geochemical element normalization, geophysical background field stripping and well logging data normalization; Comprehensive analysis stage: Based on the correlation modeling of geological, geophysical and geochemical data, a three-dimensional prediction model is constructed to quantify the mineralization probability field; Target area determination stage: The target area is graded by combining geological structure, geophysical anomalies and geochemical element combinations through a multi-factor weighted evaluation method; Verification and evaluation stage: Verify the target area prediction results through drilling, combine them with the three-dimensional model, and estimate the resource volume and economically recoverable reserves; at the same time, improve the prediction accuracy through iterative optimization of the model.

2. The method for rapid collaborative exploration of gold mines in the air, on the ground, and in the well according to claim 1, characterized in that: The aerial survey uses satellite remote sensing technology to obtain multi-spectral, thermal infrared and radar image data, and combines aerial magnetic and electromagnetic methods to measure and identify geological structures and physical property anomalies.

3. The method for rapid collaborative exploration of gold mines in the air, on the ground, and in the well according to claim 1, characterized in that: During the ground exploration phase, induced polarization and transient electromagnetic methods were used for geophysical measurements, and geochemical sampling points were increased to analyze the content of Au, As, and Sb elements. At the same time, geological mapping was carried out at a scale of 1:10,000 to 1:2,000.

4. The method for rapid collaborative exploration of gold mines in the air, on the ground, and in the well according to claim 1, characterized in that: During the downhole exploration stage, controlled drilling is carried out using a diamond drilling rig, and downhole magnetic measurement, gamma ray spectrum logging and high-density three-dimensional seismic exploration are carried out in the borehole. At the same time, core samples are collected for rock and mineral identification and elemental geochemical analysis.

5. The method for rapid collaborative exploration of gold mines in the air, on the ground, and in the well according to claim 1, characterized in that: The comprehensive analysis stage quantifies the correlation between stratum lithology, geophysical parameters and geochemical elements through covariance matrix analysis and principal component coupling analysis, and screens key ore-controlling factors.

6. The method for rapid collaborative exploration of gold mines in the air, on the ground, and in the well according to claim 1, characterized in that: During the target area determination stage, the entropy weight-analytic hierarchy process is used to perform weighted evaluation on the geological structure complexity, geophysical anomaly intensity, geochemical element combination entropy and alteration mineral aggregation to generate a mineralization target area distribution map.

7. The method for rapid aerial-surface-in-hole collaborative exploration of gold mines according to claim 1, characterized in that: During the verification and evaluation phase, a three-dimensional solid model of the mineralized body was constructed using the Kriging interpolation algorithm, and the inverse distance power method was used to estimate the resource volume. At the same time, the prediction model was dynamically updated in combination with the while-drilling logging data.

8. The method for rapid aerial-surface-in-hole collaborative exploration of gold mines according to claim 1, characterized in that: The entire exploration process relies on the GIS platform to achieve real-time integration and spatial registration of air, ground and well data, and optimizes the exploration path through a dynamic feedback mechanism to improve the adaptability and prediction accuracy of the regional prospecting model.

Images