A method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis

Abstract

This invention relates to the field of marine mineral resource exploration and evaluation technology, specifically to a method for identifying shallow-surface zircon resource potential areas in marine areas based on zircon grade data simulation analysis. The method includes the following steps: S1: Collecting shallow-surface sediment samples from the seabed to obtain zircon grade and marine environmental parameters; S2: Constructing a multivariate regression model of zircon grade and environmental factors; S3: Establishing a spatial probability distribution model of zircon grade; S4: Establishing an index fusion system and evaluating the resource potential value of each region; S5: Iteratively optimizing the potential area classification threshold and correcting parameters; S6: Generating a potential distribution heatmap and outputting the boundary coordinates of high-potential areas and a resource estimation report. This invention, by constructing a discrimination system integrating data modeling, uncertainty quantification, and visualization delineation, achieves high-precision identification and quantitative assessment of shallow-surface zircon resource potential areas in marine areas.

Description

Technical Field

[0001] This invention relates to the field of marine mineral resource exploration technology, and in particular to a method for identifying shallow surface zircon resource potential areas in the sea based on zircon grade data simulation analysis. Background Technology

[0002] Zircon, as an important strategic mineral resource, is widely used in the metallurgical industry, refractory materials, and nuclear energy fields, and its distribution in shallow marine sedimentary environments shows significant resource potential. Currently, the exploration of marine zircon resources mainly relies on local sample data obtained from discrete sampling stations. By analyzing zircon grade and sediment characteristics, a preliminary assessment of the region's resource potential is made.

[0003] However, due to the complex and variable seabed sedimentary environment, existing methods are generally unable to fully characterize the impact of environmental parameter fluctuations on zircon grade distribution, making it difficult to accurately quantify the uncertainty of resource distribution. This results in limited accuracy in resource prediction and low efficiency in subsequent exploration. Therefore, there is an urgent need for a method for identifying shallow-surface zircon resource potential areas in marine environments based on zircon grade data simulation analysis to address these issues. Summary of the Invention

[0004] To achieve the above objectives, this invention provides a method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis.

[0005] A method for identifying shallow surface zircon resource potential zones in marine areas based on zircon grade data simulation analysis includes the following steps:

[0006] S1: Collect shallow seabed sediment samples, obtain zircon grade and marine environmental parameters, and standardize various data to construct a spatial database;

[0007] S2: Based on principal component analysis, sedimentary parameters related to zircon grade were screened, and a multiple regression model of zircon grade and environmental factors was constructed.

[0008] S3: Using the Monte Carlo method to generate multiple sets of simulated samples, and combining them with geostatistical interpolation, a spatial probability distribution model of zircon grade is established to quantify the impact of sedimentary environment uncertainty on grade distribution;

[0009] S4: Based on the three dimensions of zircon grade, sedimentary environment stability and mining economics, an index integration system is established, and a weighted scoring mechanism is used to evaluate the resource potential value of each region.

[0010] S5: Iteratively optimize the potential differentiation threshold through particle swarm optimization algorithm, and correct the parameters by combining existing drilling verification data, and perform supplementary sampling verification in the data boundary area;

[0011] S6: Generate a potential distribution heat map in the geographic information system platform, overlay seabed topography and sediment layers, execute an automatic delineation algorithm, and output the boundary coordinates of high-potential areas and a resource estimation report.

[0012] Optionally, S1 specifically includes:

[0013] S11: Using box samplers or gravity column samplers carried by marine geological survey vessels, 50 to 100 stations are set up in the sea area to vertically collect shallow surface sediment samples within a range of 0 to 30 centimeters on the seabed, and record the latitude and longitude coordinates, water depth and sampling time of the sampling points in real time.

[0014] S12: The zircon grade of the collected sediment samples was determined by X-ray fluorescence spectrometry, with the ZrO2 mass fraction as the determining index; at the same time, the particle size distribution parameters of the samples were determined by laser particle size analyzer, the bottom current velocity values ​​of the corresponding stations were obtained by acoustic Doppler current profiler, and the bottom turbidity values ​​were obtained by optical turbidimeter.

[0015] S13: Perform zero-mean normalization on the data obtained in S12 to unify the range of various data units to [-1, 1];

[0016] S14: The standardized zircon grades and various environmental parameters are spatially matched according to the station coordinates and imported into the PostgreSQL database to construct a structured spatial database containing station number, spatial location, zircon grade and various environmental factor fields.

[0017] Optionally, S2 specifically includes:

[0018] S21: Using the spatialized database constructed by S1, zircon grade data and environmental parameter data of all sample stations are extracted to form an analytical data matrix with zircon grade as the dependent variable and grain size distribution parameter, bottom current velocity value and bottom turbidity value as independent variables.

[0019] S22: Calculate the correlation coefficient of the analysis data matrix, remove independent variables with a correlation coefficient less than 0.3, retain environmental parameters with significant correlation, and perform principal component analysis on the retained environmental parameters. By calculating the eigenvalues ​​and cumulative variance contribution rate of the principal components, screen the principal components with a cumulative contribution rate of more than 85% and identify the corresponding deposition parameters.

[0020] S23: Using the sedimentary parameters obtained from S22 as independent variables and zircon grade as dependent variable, a multiple regression model between zircon grade and key sedimentary parameters is constructed using the least squares method. After significance test and goodness-of-fit test, independent variables with significance levels higher than 0.05 are removed to determine the final multiple regression model equation.

[0021] Optionally, S22 specifically includes:

[0022] S221: Based on the analytical data matrix constructed in S21, calculate the Pearson correlation coefficient between each environmental parameter and cobalt grade. ;

[0023] S222: Based on the calculated correlation coefficients of various environmental parameters Eliminate those that meet the requirements The independent variables are selected, and environmental parameters with an absolute correlation coefficient of not less than 0.3 are retained to form a subset of environmental parameter data;

[0024] S223: Perform principal component analysis on the retained subset of environmental parameter data, construct the covariance matrix, calculate the eigenvalues ​​and corresponding eigenvectors of the covariance matrix, and calculate the variance contribution rate of each principal component. ;

[0025] S224: Sort the principal components in descending order of eigenvalues ​​and calculate the cumulative variance contribution rate. The top K principal components with a cumulative variance contribution rate of over 85% are selected, and the original sedimentation parameters with the largest absolute weight among the principal components are identified based on their corresponding eigenvector coefficients, which are then used as the final selected sedimentation parameters.

[0026] Optionally, S23 specifically includes:

[0027] S231: Using the sedimentary parameters obtained from S22 as independent variables and zircon grade as the dependent variable, construct the expression for the multiple regression equation: In the formula, Y is the predicted zircon grade. As key sedimentary parameters, For the intercept term, Let be the regression coefficients to be determined. This is the random error term;

[0028] S232: Solve for the regression coefficients using the least squares method, with the objective function being to minimize the sum of squared residuals;

[0029] S233: Perform a significance test on the obtained regression coefficients, calculate the level values ​​of the regression coefficients, and remove independent variables with a significance level value greater than 0.05;

[0030] S234: Calculate the goodness of fit of the model; if the goodness of fit reaches the preset standard and the significance level of all regression coefficients is less than or equal to 0.05, then determine the final multiple regression model equation; otherwise, repeat the above steps until the model meets the conditions.

[0031] Optionally, S3 specifically includes:

[0032] S31: Based on the multivariate regression model constructed by S2, the actual observed mean and standard deviation of each key sedimentation parameter are used as the basis, and it is set to follow a normal distribution. Random sampling is performed using the Monte Carlo method to generate 1,000 to 5,000 sets of virtual environmental parameter sample sets.

[0033] S32: Substitute each set of virtual environmental parameter samples into the multivariate regression model to calculate the corresponding zircon grade prediction value, obtain the virtual sample dataset, and form a simulated sample library containing sample coordinates, predicted grade and environmental parameters.

[0034] S33: Using ordinary kriging interpolation, spatial modeling is performed on all predicted zircon grade values ​​in the virtual sample library to construct a spatial distribution function of zircon grade based on location coordinates;

[0035] S34: Based on the Kriging interpolation prediction results, and combined with the prediction variance between simulated samples, construct the zircon grade probability distribution function for each grid cell, which is used to statistically analyze the probability density of the predicted value of each cell within a set interval.

[0036] S35: Integrate the probability distribution results of all grid cells to form a spatial probability distribution model of zircon grade, the expression of which is: ,in, This represents the spatial probability distribution model of the entire study area; S is the set of all grid cells within the study area. Let be the zircon grade probability density function of unit s; U represents the probability characteristics of each unit integrated according to geographic coordinates.

[0037] Optionally, S4 specifically includes:

[0038] S41: Based on the zircon grade spatial probability distribution model constructed in S3, calculate the expected value of the zircon grade prediction in each grid cell, which serves as the zircon grade index.

[0039] S42: Based on the probability density function of S3, calculate the reciprocal of the predicted variance of zircon grade for each grid unit, and use it as an indicator of sedimentary environment stability.

[0040] S43: Using the water depth data corresponding to the grid cells, calculate the mining economic indicators based on the pre-established linear relationship between water depth and mining cost;

[0041] S44: The above three indicators are uniformly standardized using the minimum-maximum normalization method to ensure that the indicator values ​​are between 0 and 1.

[0042] S45: The resource potential score of each grid cell is calculated by weighting the zircon grade index (0.4), sedimentary environment stability index (0.3), and mining economics index (0.3). .

[0043] Optionally, S5 specifically includes:

[0044] S51: The resource potential score results of each grid cell calculated in S4 are used as the optimization target input. The number of resource potential scores and the initial set of scores thresholds are preset. An initial population containing multiple particles is constructed, and each particle represents a set of potential scores thresholds.

[0045] S52: Set the maximum number of iterations, inertia weight, learning factor and convergence criterion for the particle swarm optimization algorithm, and update the position and adjust the velocity of each particle in the search space to minimize the objective function, that is, the inconsistency measure between the predicted potential classification results and the existing drilling data distribution.

[0046] S53: In each iteration, using the grading threshold represented by the current particle, all grid cells are divided into potential regions of different grades, and compared with the actual grade of the drilling verification area, and the matching accuracy is recorded as the fitness value.

[0047] S54: Based on the fitness values ​​of all particles, update the individual extreme values ​​and the global optimal particle, and continuously update the potential classification threshold set through iterative optimization until the fitness converges or the maximum number of iterations is reached, and output the final optimal potential classification threshold.

[0048] S55: In areas where the matching degree between the grading results and the actual drilling data is lower than the preset threshold, the corresponding spatial location is identified as a disputed area. 10 to 20 supplementary sampling stations are set up in the disputed area to collect shallow sediments and determine zircon grade and environmental parameters. The samples are then included in the database to backfill the model input and re-optimize the threshold.

[0049] S56: Merge the supplementary sampling data with the original verification data, and repeat the optimization process in S51 to S54 until the final potential classification result meets the set requirements for the actual drilling results.

[0050] Optionally, S6 specifically includes:

[0051] S61: Apply the potential differentiation threshold determined by S5 to the resource potential scoring results of each grid cell calculated by S4, and generate a heat map of zircon resource potential distribution in the geographic information system platform based on the spatial coordinate data of the grid cells.

[0052] S62: Import the acquired seabed topographic elevation data and spatial distribution vector data of sediment types into the geographic information system platform, align them with the potential distribution heat map using spatial location coordinates and overlay them to form a complete spatial visualization analysis map of resource potential.

[0053] S63: Using a raster-based automatic region growth algorithm, based on the highest potential differentiation threshold determined in S5, identify and label adjacent raster cell sets with potential scores higher than the threshold, and then spatially aggregate the raster cell sets to generate closed polygons for the boundaries of multiple potential regions.

[0054] S64: Extract the spatial boundary coordinates of each closed polygon and convert them into geographic coordinates. Calculate the estimated zircon resource quantity for each region.

[0055] S65: Based on the boundary coordinates, area, and estimated resource quantity of the generated closed polygon, automatically generate a structured resource potential area delineation report in the geographic information system platform and output it in electronic file form.

[0056] Optionally, S64 specifically includes:

[0057] S641: Based on the closed polygon region generated in S63, extract the set of boundary nodes for each polygon, and convert the projected coordinates of the boundary nodes into geographic coordinates using the geographic projection inverse calculation method, and use the WGS84 coordinate system for unified expression.

[0058] S642: Construct a polygonal object using a set of latitude and longitude boundary points, and calculate the actual area A of the closed polygonal region using the Gauss-Green formula;

[0059] S643: Obtain the resource potential score of all grid cells within each polygonal region, and calculate the expected value of the region's potential score. Then multiply it by the area of ​​the corresponding polygon to obtain the estimated zircon resource quantity Q of the area.

[0060] The beneficial effects of this invention are:

[0061] This invention systematically extracts key sedimentary parameters affecting zircon enrichment by introducing principal component analysis and multiple regression models; and constructs a spatial probability distribution of grade by combining Monte Carlo simulation and Kriging interpolation, effectively characterizing the impact of sedimentary environment uncertainty on grade prediction.

[0062] This invention achieves dynamic optimization of potential region classification through the fusion of multiple indicators of resource potential and particle swarm optimization algorithm, and generates a visual heat map based on GIS platform to automatically delineate the boundaries of high-potential areas and complete resource estimation. This method has the characteristics of high accuracy, high interpretability and strong scalability, and can significantly improve the scientificity of zircon resource potential identification in complex sea areas. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of the zircon resource potential area identification method according to an embodiment of the present invention;

[0065] Figure 2 This is a schematic diagram of the iterative optimization and parameter correction process in an embodiment of the present invention. Detailed Implementation

[0066] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0067] like Figures 1-2 As shown, a method for identifying shallow surface zircon resource potential areas in marine areas based on zircon grade data simulation analysis includes the following steps:

[0068] S1: Collect shallow seabed sediment samples, obtain zircon grade and marine environmental parameters, and standardize various data to construct a spatial database;

[0069] S2: Based on principal component analysis, sedimentary parameters related to zircon grade were screened, a multiple regression model of zircon grade and environmental factors was constructed, and significance verification and goodness-of-fit evaluation were performed.

[0070] S3: Using the Monte Carlo method to generate multiple sets of simulated samples, and combining them with geostatistical interpolation, a spatial probability distribution model of zircon grade is established to quantify the impact of sedimentary environment uncertainty on grade distribution;

[0071] S4: Based on the three dimensions of zircon grade, sedimentary environment stability and mining economics, an index integration system is established, and a weighted scoring mechanism is used to evaluate the resource potential value of each region.

[0072] S5: Iteratively optimize the potential differentiation threshold through particle swarm optimization algorithm, and correct the parameters by combining existing drilling verification data, and perform supplementary sampling verification in the data boundary area;

[0073] S6: Generate a potential distribution heat map in the geographic information system platform, overlay seabed topography and sediment layers, execute an automatic delineation algorithm, and output the boundary coordinates of high-potential areas and a resource estimation report.

[0074] S1 specifically includes:

[0075] S11: Using box samplers or gravity column samplers carried by marine geological survey vessels, 50 to 100 stations are set up in the sea area to vertically collect shallow surface sediment samples within a range of 0 to 30 centimeters on the seabed, and record the latitude and longitude coordinates, water depth and sampling time of the sampling points in real time.

[0076] S12: The zircon grade of the collected sediment samples was determined by X-ray fluorescence spectrometry, with the ZrO2 mass fraction as the determining index; at the same time, the particle size distribution parameters of the samples were determined by laser particle size analyzer, the bottom current velocity values ​​of the corresponding stations were obtained by acoustic Doppler current profiler, and the bottom turbidity values ​​were obtained by optical turbidimeter.

[0077] S13: Perform zero-mean normalization on the data obtained in S12 to unify the range of various data units to [-1, 1], so as to eliminate the impact of data scale differences on subsequent modeling results;

[0078] S14: The standardized zircon grades and various environmental parameters are spatially matched according to the station coordinates and imported into the PostgreSQL database to construct a structured spatial database containing station number, spatial location, zircon grade, and various environmental factor fields, providing a unified data foundation for subsequent discriminant analysis. The above steps, through clear sampling depth control, standardized detection procedures, and database structure design, not only achieve accurate acquisition and standardized processing of zircon grades and environmental factors in shallow seabed sediments, but also ensure the traceability of the original data in the spatial dimension and the consistency of the modeling foundation.

[0079] S2 specifically includes:

[0080] S21: Using the spatialized database constructed by S1, zircon grade data and environmental parameter data of all sample stations are extracted to form an analytical data matrix with zircon grade as the dependent variable and grain size distribution parameter, bottom current velocity value and bottom turbidity value as independent variables.

[0081] S22: Calculate the correlation coefficient of the analysis data matrix, remove independent variables with a correlation coefficient less than 0.3, retain environmental parameters with significant correlation, and perform principal component analysis on the retained environmental parameters. By calculating the eigenvalues ​​and cumulative variance contribution rate of the principal components, screen the principal components with a cumulative contribution rate of more than 85% and identify the corresponding deposition parameters.

[0082] S23: Using the sedimentary parameters selected in S22 as independent variables and zircon grade as the dependent variable, a multiple regression model between zircon grade and key sedimentary parameters is constructed using the least squares method. Independent variables with a significance level higher than 0.05 are removed through significance and goodness-of-fit tests to determine the final multiple regression model equation. The above steps, by employing principal component analysis to select key parameters and constructing a regression model based on significance tests, ensure that the selected environmental parameters maximize their explanatory power for zircon grade, effectively improving the accuracy and reliability of the zircon resource spatial prediction model.

[0083] S22 specifically includes:

[0084] S221: Based on the analytical data matrix constructed in S21, calculate the Pearson correlation coefficient between each environmental parameter and cobalt grade. The calculation formula is: ,in, This represents the value of the i-th environmental parameter in the j-th sample. Let represent the mean of the i-th environmental parameter. Let j be the zircon grade of the j-th sample. denoted as the average zircon grade, and n as the total number of samples;

[0085] S222: Based on the calculated correlation coefficients of various environmental parameters Eliminate those that meet the requirements The independent variables are selected, and environmental parameters with an absolute correlation coefficient of not less than 0.3 are retained to form a subset of environmental parameter data;

[0086] S223: Perform principal component analysis on the retained subset of environmental parameter data, construct the covariance matrix, and calculate the eigenvalues ​​of the covariance matrix. and corresponding feature vectors And calculate the variance contribution rate of each principal component. The formula is: ,in, Let m be the contribution rate of the k-th principal component, and m be the total number of principal components.

[0087] S224: Sort the principal components in descending order of eigenvalues ​​and calculate the cumulative variance contribution rate. Its expression is: The process involves selecting the top K principal components with a cumulative variance contribution rate of over 85%, and identifying the original sedimentary parameters with the largest absolute weight (weight greater than 0.5) among the principal components based on their corresponding eigenvector coefficients. These parameters are then used as the final sedimentary parameters for selection. By quantifying the linear correlation between environmental parameters and zircon grade, and combining a multi-level discrimination strategy of principal component eigenvalues ​​and loading weights, redundant information can be scientifically eliminated, and the sedimentary parameters that dominate zircon distribution can be selected. This provides a stable, concise, and physically meaningful data input foundation for constructing a high-precision regression model.

[0088] S23 specifically includes:

[0089] S231: Using the sedimentary parameters obtained from S22 as independent variables and zircon grade as the dependent variable, construct the expression for the multiple regression equation: In the formula, Y is the predicted zircon grade. As key sedimentary parameters, For the intercept term, Let be the regression coefficients to be determined. This is the random error term;

[0090] S232: The regression coefficients are solved using the least squares method. The objective function is to minimize the sum of squared residuals. The calculation formula is as follows: In the formula, Let be the actual luminous grade value of the i-th sample. The corresponding predicted values ​​are given, where N is the total number of samples. The regression coefficient vector is obtained through matrix operations. The formula is: In the formula, X is the matrix of independent variables and y is the vector of dependent variables;

[0091] S233: Perform a significance test on the solved regression coefficients and calculate the level values ​​of the regression coefficients using the following formula: In the formula, Let j be the level value of the j-th regression coefficient. Let j be the estimated value of the j-th regression coefficient. The standard error of this coefficient is used to remove independent variables with a significance level greater than 0.05.

[0092] S234: Calculate the goodness of fit of the model The formula is: In the formula, The average value of zircon grade is used. If the goodness of fit meets the preset standard and the significance level of all regression coefficients is less than or equal to 0.05, the final multiple regression model equation is determined. Otherwise, the above steps are repeated until the model meets the conditions. The above steps accurately estimate the linear relationship between zircon grade and sedimentary parameters using the least squares method. With rigorous statistical significance tests and goodness of fit evaluation, this method can effectively ensure that the final model has statistical reliability and practical predictive ability, which helps to accurately and efficiently guide the assessment of marine zircon resource potential.

[0093] S3 specifically includes:

[0094] S31: Based on the multivariate regression model constructed by S2, the actual observed mean and standard deviation of each key sedimentation parameter are used as the basis, and it is set to follow a normal distribution. Random sampling is performed using the Monte Carlo method to generate 1,000 to 5,000 sets of virtual environmental parameter sample sets.

[0095] S32: Substitute each set of virtual environmental parameter samples into the multivariate regression model to calculate the corresponding zircon grade prediction value, obtain the virtual sample dataset, and form a simulated sample library containing sample coordinates, predicted grade and environmental parameters.

[0096] S33: Using ordinary kriging interpolation, spatial modeling is performed on all predicted zircon grade values ​​in the virtual sample library to construct a spatial distribution function of zircon grade based on location coordinates, specifically expressed as follows:

[0097] In the formula, Location to be predicted Zircon grade valuation at the location For known location The quality value at that location These are the weighting coefficients. The number of known sample points for interpolation;

[0098] S34: Based on the Kriging interpolation prediction results and combined with the prediction variance between simulated samples, construct the zircon grade probability distribution function for each grid cell. This function is used to statistically analyze the probability density of the predicted value of each cell within a set interval. The expression for the zircon grade probability distribution function is:

[0099] ,in, Represents the zircon grade probability density function of unit s; This is the mean of the predicted grade values ​​for all simulated samples in the corresponding unit. denoted as , where z is the variance of the predicted grade values ​​for all simulated samples in the corresponding unit; z is the target predicted grade value; this function is used to represent the uncertainty distribution of the zircon grade prediction results for each unit under fluctuating sedimentary environment.

[0100] S35: Integrate the probability distribution results of all grid cells to form a spatial probability distribution model of zircon grade, the expression of which is: ,in, This represents the spatial probability distribution model of the entire study area; S is the set of all grid cells within the study area. is the zircon grade probability density function for unit s; U represents the probability characteristics of each unit integrated according to geographic coordinates, forming a spatial probability layer.

[0101] S4 specifically includes:

[0102] S41: Based on the zircon grade spatial probability distribution model constructed in S3, calculate the expected value of the predicted zircon grade within each grid cell, which serves as the zircon grade index. The calculation formula is as follows: In the formula, Let be the expected value of the bauxite grade for grid cell s; Let z be the zircon grade probability density function of the grid cell s; z is the zircon grade value.

[0103] S42: Based on the probability density function in S3, calculate the reciprocal of the predicted variance of zircon grade for each grid cell. This reciprocal serves as an index of sedimentary environment stability. A higher index value indicates a smaller influence of the sedimentary environment on the prediction results. The calculation formula is as follows: In the formula, This is an index of the sedimentary environment stability of grid cell s; The variance of the predicted luminous grade in grid cell s;

[0104] S43: Using the water depth data corresponding to the grid cells, and based on the pre-established linear relationship between water depth and mining cost, calculate the mining economic index using the following formula: In the formula, For the economic indicators of mining grid cell s; The water depth value for grid cell s; The linear coefficients are determined by fitting historical exploration cost data, and This indicates that economic efficiency decreases as water depth increases;

[0105] S44: The above three indicators are uniformly standardized using the minimum-maximum normalization method to ensure that the indicator values ​​are between 0 and 1.

[0106] S45: The resource potential score of each grid cell is calculated by weighting the zircon grade index (0.4), sedimentary environment stability index (0.3), and mining economics index (0.3). The formula is: In the formula, Score the resource potential of the i-th grid cell; , , The values ​​are standardized values ​​for the zircon grade, sedimentary environment stability, and mining economics of the i-th grid cell, respectively.

[0107] S5 specifically includes:

[0108] S51: The resource potential score results of each grid cell calculated in S4 are used as the optimization target input. The number of resource potential scores and the initial set of scores thresholds are preset. An initial population containing multiple particles is constructed, and each particle represents a set of potential scores thresholds.

[0109] S52: Set the maximum number of iterations, inertia weight, learning factor and convergence criterion for the particle swarm optimization algorithm, and update the position and adjust the velocity of each particle in the search space to minimize the objective function, that is, the inconsistency measure between the predicted potential classification results and the existing drilling data distribution.

[0110] S53: In each iteration, using the grading threshold represented by the current particle, all grid cells are divided into potential regions of different grades, and compared with the actual grade of the drilling verification area, and the matching accuracy is recorded as the fitness value.

[0111] S54: Based on the fitness values ​​of all particles, update the individual extreme values ​​and the global optimal particle, and continuously update the potential classification threshold set through iterative optimization until the fitness converges or the maximum number of iterations is reached, and output the final optimal potential classification threshold.

[0112] S55: In areas where the matching degree between the grading results and the actual drilling data is lower than the preset threshold (e.g., 85%), the corresponding spatial location is identified as a disputed area. 10 to 20 supplementary sampling stations are set up in the disputed area to collect shallow sediments and determine zircon grade and environmental parameters. These are then included in the database for backfilling the model input and re-optimizing the threshold.

[0113] S56: Merge the supplementary sampling data with the original verification data, and repeat the optimization process in S51 to S54 until the final potential classification result meets the set requirements for the degree of agreement with the actual drilling. The above steps introduce the particle swarm optimization algorithm to globally optimize the potential classification threshold, and combine drilling data feedback to realize dynamic correction of the classification results and sampling reinforcement of disputed areas, which significantly improves the empirical agreement and spatial adaptability of the classification model, and ensures that the resource identification results have higher geological reliability and practicality.

[0114] S6 specifically includes:

[0115] S61: Apply the potential differentiation threshold determined by S5 to the resource potential scoring results of each grid cell calculated by S4, and generate a heat map of zircon resource potential distribution in the geographic information system platform based on the spatial coordinate data of the grid cells.

[0116] S62: Import the acquired seabed topographic elevation data and spatial distribution vector data of sediment types into the geographic information system platform, align them with the potential distribution heat map using spatial location coordinates and overlay them to form a complete spatial visualization analysis map of resource potential.

[0117] S63: Using a raster-based automatic region growth algorithm, based on the highest potential differentiation threshold determined in S5, identify and label adjacent raster cell sets with potential scores higher than the threshold, and then spatially aggregate the raster cell sets to generate closed polygons for the boundaries of multiple potential regions.

[0118] S64: Extract the spatial boundary coordinates of each closed polygon and convert them into geographic coordinates. Calculate the estimated zircon resource quantity for each region.

[0119] S65: Based on the boundary coordinates, area, and resource estimate of the generated closed polygon, a structured resource potential area delineation report is automatically generated in the geographic information system platform and output in electronic file form to guide subsequent marine exploration and resource evaluation. The above steps, by combining potential scoring data with geographic information system technology and adopting an automatic area delineation algorithm, can accurately realize the automatic identification of the spatial range of high-potential zircon resource areas and resource estimation. The visualization analysis effect is obvious, improving the efficiency and accuracy of resource survey and exploration decision-making.

[0120] S64 specifically includes:

[0121] S641: Based on the closed polygon region generated in S63, extract the set of boundary nodes for each polygon, and convert the projected coordinates (such as UTM coordinates) of the boundary nodes into geographic coordinates (latitude and longitude) using the geographic projection inverse calculation method, and express them uniformly using the WGS84 coordinate system.

[0122] S642: Construct a polygonal object using a set of latitude and longitude boundary points, and calculate the actual area A of the closed polygonal region using the Gauss-Green formula (applicable to approximate planar projections), as follows:

[0123] Where A represents the area of ​​the polygon; and These are the coordinates of two adjacent boundary points (represented in the projected coordinate system). The number of boundary points, ;

[0124] S643: Obtain the resource potential score of all grid cells within each polygonal region, and calculate the expected value of the region's potential score. Specifically, the arithmetic mean of the potential scores of the grid cells within each closed polygon region is used as the... Then, multiply this value by the area of ​​the corresponding polygon to obtain the estimated zircon resource quantity Q for the area. The calculation formula is as follows: Where Q represents the estimated amount of luminescein resources in the region; The above steps, through boundary coordinate transformation, area calculation and potential score integration, realize the quantitative estimation of resource quantity in high-potential areas, ensuring that the delineation results not only have spatial accuracy, but also have resource quantity indicators to support them, providing a reliable basis for subsequent resource exploitation and economic assessment.

[0125] I. Resource potential score is obtained as follows

[0126] According to the claims and specification, the resource potential score is set as follows: It is obtained by weighted summation of three normalized indicators: zircon grade, sedimentary environment stability, and mining economics; example data is as follows:

[0127] Grid numbering Grade index Xi1′ Stability index Xi2′ Economic indicators Xi3′ Resource potential score Ri 1 0.92 0.85 0.78 0.4×0.92+0.3×0.85+0.3×0.78=0.857 2 0.81 0.73 0.68 0.4×0.81+0.3×0.73+0.3×0.68=0.747 3 0.65 0.55 0.59 0.4×0.65+0.3×0.55+0.3×0.59=0.600 4 0.4 0.44 0.51 0.4×0.40+0.3×0.44+0.3×0.51=0.445

[0128] Resource potential score for each grid All values ​​are derived by weighting the normalized indicators mentioned above, reflecting the overall potential for zircon resource development in this grid area.

[0129] II. Potential Rating Expected Value Typically, the score of all grid cells within a high-potential delineation area is calculated as the average. Assuming a high-potential area includes the four grid cells mentioned above, its expected potential score is: .

[0130] III. Zircon resource estimates are as follows: Assuming the area of ​​the polygonal high-potential zone A = 1.2 square kilometers, the estimated gibbsite resource Q is calculated using the following formula: The estimated zircon resource quantity represents the resource development potential per unit area after comprehensively considering resource quality, environment, and economics. Combined with the area of ​​the delineated zone, it forms a quantitative value of the total regional resource potential, providing a basis for subsequent mining and investment decisions.

[0131] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0132] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle 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 shallow-surface zircon resource potential zones in marine areas based on zircon grade data simulation analysis, characterized in that, Includes the following steps: S1: Collect shallow seabed sediment samples, obtain zircon grade and marine environmental parameters, and standardize various data to construct a spatial database; S2: Based on principal component analysis, sedimentary parameters related to zircon grade were screened, and a multiple regression model of zircon grade and environmental factors was constructed. S3: Using the Monte Carlo method to generate multiple sets of simulated samples, and combining them with geostatistical interpolation, a spatial probability distribution model of zircon grade is established to quantify the impact of sedimentary environment uncertainty on grade distribution; S3 specifically includes: S31: Based on the multivariate regression model constructed by S2, the actual observed mean and standard deviation of each key sedimentation parameter are used as the basis, and it is set to follow a normal distribution. Random sampling is performed using the Monte Carlo method to generate 1,000 to 5,000 sets of virtual environmental parameter sample sets. S32: Substitute each set of virtual environmental parameter samples into the multivariate regression model to calculate the corresponding zircon grade prediction value, obtain the virtual sample dataset, and form a simulated sample library containing sample coordinates, predicted grade and environmental parameters. S33: Using ordinary kriging interpolation, spatial modeling is performed on all predicted zircon grade values ​​in the virtual sample library to construct a spatial distribution function of zircon grade based on location coordinates; S34: Based on the Kriging interpolation prediction results, and combined with the prediction variance between simulated samples, construct the zircon grade probability distribution function for each grid cell, which is used to statistically analyze the probability density of the predicted value of each cell within a set interval. S35: Integrate the probability distribution results of all grid cells to form a spatial probability distribution model of zircon grade, the expression of which is: ,in, A spatial probability distribution model representing the entire study area; The set of all grid cells within the study area; For unit The zircon grade probability density function; U represents the probability characteristics of each unit integrated according to geographic coordinates; S4: Based on the three dimensions of zircon grade, sedimentary environment stability and mining economics, an index integration system is established, and a weighted scoring mechanism is used to evaluate the resource potential value of each region. S4 specifically includes: S41: Based on the zircon grade spatial probability distribution model constructed in S3, calculate the expected value of the zircon grade prediction in each grid cell, which serves as the zircon grade index. S42: Based on the probability density function of S3, calculate the reciprocal of the predicted variance of zircon grade for each grid unit, and use it as an indicator of sedimentary environment stability. S43: Using the water depth data corresponding to the grid cells, calculate the mining economic indicators based on the pre-established linear relationship between water depth and mining cost; S44: The above three indicators are uniformly standardized using the minimum-maximum normalization method to ensure that the indicator values ​​are between 0 and 1. S45: The resource potential score of each grid cell is calculated by weighting the zircon grade index (0.4), sedimentary environment stability index (0.3), and mining economics index (0.3). ; S5: Iteratively optimize the potential differentiation threshold through particle swarm optimization algorithm, and correct the parameters by combining existing drilling verification data, and perform supplementary sampling verification in the data boundary area; S6: Generate a potential distribution heat map in the geographic information system platform, overlay seabed topography and sediment layers, execute an automatic delineation algorithm, and output the boundary coordinates of high-potential areas and a resource estimation report.

2. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis as described in claim 1, characterized in that, S1 specifically includes: S11: Using box samplers or gravity column samplers carried by marine geological survey vessels, 50 to 100 stations are set up in the sea area to vertically collect shallow surface sediment samples within a range of 0 to 30 centimeters on the seabed, and record the latitude and longitude coordinates, water depth and sampling time of the sampling points in real time. S12: The zircon grade of the collected sediment samples was determined by X-ray fluorescence spectrometry, with the ZrO2 mass fraction as the determining index; at the same time, the particle size distribution parameters of the samples were determined by laser particle size analyzer, the bottom current velocity values ​​of the corresponding stations were obtained by acoustic Doppler current profiler, and the bottom turbidity values ​​were obtained by optical turbidimeter. S13: Perform zero-mean normalization on the data obtained in S12 to unify the range of various data units to [-1,1]. S14: The standardized zircon grades and various environmental parameters are spatially matched according to the station coordinates and imported into the PostgreSQL database to construct a structured spatial database containing station number, spatial location, zircon grade and various environmental factor fields.

3. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis as described in claim 1, characterized in that, S2 specifically includes: S21: Using the spatialized database constructed by S1, zircon grade data and environmental parameter data of all sample stations are extracted to form an analytical data matrix with zircon grade as the dependent variable and grain size distribution parameter, bottom current velocity value and bottom turbidity value as independent variables. S22: Calculate the correlation coefficient of the analysis data matrix, remove independent variables with a correlation coefficient less than 0.3, retain environmental parameters with significant correlation, and perform principal component analysis on the retained environmental parameters. By calculating the eigenvalues ​​and cumulative variance contribution rate of the principal components, screen the principal components with a cumulative contribution rate of more than 85% and identify the corresponding deposition parameters. S23: Using the sedimentary parameters obtained from S22 as independent variables and zircon grade as dependent variable, a multiple regression model between zircon grade and key sedimentary parameters is constructed using the least squares method. After significance test and goodness-of-fit test, independent variables with significance levels higher than 0.05 are removed to determine the final multiple regression model equation.

4. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis as described in claim 3, characterized in that, S22 specifically includes: S221: Based on the analytical data matrix constructed in S21, calculate the Pearson correlation coefficient between each environmental parameter and cobalt grade. ; S222: Based on the calculated correlation coefficients of various environmental parameters Eliminate those that meet the requirements The independent variables are selected, and environmental parameters with an absolute correlation coefficient of not less than 0.3 are retained to form a subset of environmental parameter data; S223: Perform principal component analysis on the retained subset of environmental parameter data, construct the covariance matrix, calculate the eigenvalues ​​and corresponding eigenvectors of the covariance matrix, and calculate the variance contribution rate of each principal component. ; S224: Sort the principal components in descending order of eigenvalues ​​and calculate the cumulative variance contribution rate. Selecting those with a cumulative variance contribution rate that reaches The above Each principal component is identified, and the original depositional parameter with the largest absolute weight among the principal components is identified based on its corresponding eigenvector coefficients, which is then used as the final selected depositional parameter.

5. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis according to claim 4, characterized in that, S23 specifically includes: S231: Using the sedimentary parameters obtained from S22 as independent variables and zircon grade as the dependent variable, construct the expression for the multiple regression equation: In the formula, This is the predicted zircon grade. As key sedimentary parameters, For the intercept term, Let be the regression coefficients to be determined. This is the random error term; S232: Solve for the regression coefficients using the least squares method, with the objective function being to minimize the sum of squared residuals; S233: Perform a significance test on the obtained regression coefficients, calculate the level values ​​of the regression coefficients, and remove independent variables with a significance level value greater than 0.05; S234: Calculate the goodness of fit of the model; if the goodness of fit reaches the preset standard and the significance level of all regression coefficients is less than or equal to 0.05, then determine the final multiple regression model equation; otherwise, repeat the above steps until the model meets the conditions.

6. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis as described in claim 1, characterized in that, S5 specifically includes: S51: The resource potential score results of each grid cell calculated in S4 are used as the optimization target input. The number of resource potential scores and the initial set of scores thresholds are preset. An initial population containing multiple particles is constructed, and each particle represents a set of potential scores thresholds. S52: Set the maximum number of iterations, inertia weight, learning factor and convergence criterion for the particle swarm optimization algorithm, and update the position and adjust the velocity of each particle in the search space to minimize the objective function, that is, the inconsistency measure between the predicted potential classification results and the existing drilling data distribution. S53: In each iteration, using the grading threshold represented by the current particle, all grid cells are divided into potential regions of different grades, and compared with the actual grade of the drilling verification area, and the matching accuracy is recorded as the fitness value. S54: Based on the fitness values ​​of all particles, update the individual extreme values ​​and the global optimal particle, and continuously update the potential classification threshold set through iterative optimization until the fitness converges or the maximum number of iterations is reached, and output the final optimal potential classification threshold. S55: In areas where the matching degree between the grading results and the actual drilling data is lower than the preset threshold, the corresponding spatial location is identified as a disputed area. 10 to 20 supplementary sampling stations are set up in the disputed area to collect shallow sediments and determine zircon grade and environmental parameters. The samples are then included in the database to backfill the model input and re-optimize the threshold. S56: Merge the supplementary sampling data with the original verification data, and repeat the optimization process in S51 to S54 until the final potential classification result meets the set requirements for the actual drilling results.

7. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis according to claim 1, characterized in that, S6 specifically includes: S61: Apply the potential differentiation threshold determined by S5 to the resource potential scoring results of each grid cell calculated by S4, and generate a heat map of zircon resource potential distribution in the geographic information system platform based on the spatial coordinate data of the grid cells. S62: Import the acquired seabed topographic elevation data and spatial distribution vector data of sediment types into the geographic information system platform, align them with the potential distribution heat map using spatial location coordinates and overlay them to form a complete spatial visualization analysis map of resource potential. S63: Using a raster-based automatic region growth algorithm, based on the highest potential differentiation threshold determined in S5, identify and label adjacent raster cell sets with potential scores higher than the threshold, and then spatially aggregate the raster cell sets to generate closed polygons for the boundaries of multiple potential regions. S64: Extract the spatial boundary coordinates of each closed polygon and convert them into geographic coordinates. Calculate the estimated zircon resource quantity for each region. S65: Based on the boundary coordinates, area, and estimated resource quantity of the generated closed polygon, automatically generate a structured resource potential area delineation report in the geographic information system platform and output it in electronic file form.

8. The method for identifying shallow marine zircon resource potential zones based on zircon grade data simulation analysis according to claim 1, characterized in that, S64 specifically includes: S641: Based on the closed polygon region generated in S63, extract the set of boundary nodes for each polygon, and convert the projected coordinates of the boundary nodes into geographic coordinates using the geographic projection inverse calculation method, and use the WGS84 coordinate system for unified expression. S642: Construct polygonal objects using latitude and longitude boundary point sets, and calculate the actual area of ​​closed polygonal regions using the Gauss-Green's formula. ; S643: Obtain the resource potential score of all grid cells within each polygonal region, and calculate the expected value of the region's potential score. Multiplying this value by the area of ​​the corresponding polygon yields an estimated zircon resource quantity for the region. .

Images