Method, equipment and medium for evaluating minerality of granite rock mass

Abstract

The present application provides a method for evaluating the mineralization of granite rock masses, which relates to the field of geochemical exploration, including: obtaining sample data and category label data of tungsten-bearing granite, tungsten-tin-bearing granite, and non-ore-bearing granite in a study area; determining a distance matrix based on the sample data; using the first sample of the sample data as the coordinate origin, i.e., the first center of a circle; initializing the positions of the first three centers of the circle using the distance matrix; performing iterative calculations using the distance matrix and the initialized center positions to determine the positions of the remaining n-1 centers of the circle; and using the category and element content as point attributes to evaluate the classification effect based on the position of the center of each sample and the category label data. The present invention improves the TSNE method, realizes the evaluation of the tungsten-tin mineralization capacity of granite, and solves the problem that traditional geochemical mapping cannot clearly distinguish the mineralization of rock masses in a data-driven environment.

Description

Technical Field

[0001] The present application relates to the field of geochemical exploration, and in particular to a method, equipment and medium for evaluating the mineralization of granite rock masses. Background Art

[0002] Granite, as a key player in magmatic hydrothermal mineralization, is a valuable tool for understanding the geochemical processes of tungsten-tin mineralization and distinguishing the mineralization properties of rock masses. Previous studies have examined granite and tungsten-tin deposits in the Nanling region from the perspectives of material sources, fluid sources, physicochemical conditions, and complex rock masses. A wealth of rock-geochemical data has been accumulated for both known mineralized rock masses and those with unknown mineralization potential. Efficiently utilizing this data to identify discriminant markers of tungsten-tin mineralization and rapidly and quantitatively evaluate the mineralization properties of numerous rock masses is a crucial task in mineral exploration. Machine learning is adaptable to the analysis of high-dimensional, multi-source, and large-scale data. It comprehensively considers all indicators of the research object and, based on data characteristics, uncovers the underlying connections behind geological problems. It is a highly effective geoscience data analysis technique. Researchers (Wang Huaitao et al., 2018; Xiao Yi et al., 2009) performed clustering and discriminant analysis on oxide indicators of igneous rock samples and found that known mineralized samples were almost all clustered together. They inferred that samples with similar characteristics to the mineralized samples were mineralized rock masses. Many researchers (Petrelli M, 2016; Ueki K, 2018; Wang Y, 2021) believe that using discriminant maps to infer the tectonic environment of igneous rocks has limited applicability and insufficient accuracy in the context of global igneous rock big data. Consequently, they have proposed intelligent discriminant models for tectonic environments of igneous rocks based on multivariate statistics and machine learning algorithms, with a proven accuracy rate exceeding 80%. Other researchers have applied machine learning to classify mineral deposits, reveal deposit stages, identify metamorphic protoliths, and determine the genesis of mineral deposits (Hasterok D, 2019; Gregory DD, 2019; Zhong RC, 2021). Major and trace elements are key characteristics that distinguish quartz vein-type tungsten-tin deposits. Geochemical mapping can use certain indicators to distinguish tungsten-bearing, tungsten-tin-bearing, and non-ore-bearing rock bodies in typical deposits. However, these differences can only be revealed through deviations in the distribution range of large amounts of data. Furthermore, if a rock feature falls into the overlapping area of the diagram, its mineralization cannot be determined. Samples of major and rare earth element compositions are high-dimensional data, so classifying the three types of rock bodies can be reduced to the problem of dimensionality reduction and visualization.

[0003] Common high-dimensional data classification methods include random forests (RF), support vector machines (SVM), principal component analysis (PCA), and manifold learning. Random forests and support vector machines are supervised machine learning algorithms. The random forest algorithm classifies unknown samples by establishing a mapping between input samples and corresponding labels; the support vector machine algorithm establishes a linear or bilinear classification surface with the largest interval in the feature space. Random forests and support vector machines are not suitable for visualizing classification results. Principal component analysis projects samples into a low-dimensional space through linear transformation, but it does not perform well for samples with nonlinear distributions. Manifold learning is a nonlinear dimensionality reduction algorithm. A manifold is a geometric structure in a high-dimensional space. It assumes that the distribution of data in a high-dimensional space lies on a lower-dimensional manifold, ensuring that the data after dimensionality reduction also satisfies the geometric constraints of the manifold in the high-dimensional space. Manifold dimensionality reduction algorithms include local linear mapping, t-distributed stochastic nearest neighbor embedding (TSNE), Laplace eigenmap, and isometric mapping. Local linear mapping approximates the reconstruction of each sample point through a linear combination of multiple adjacent points:

[0004]

[0005] Where x j is x i The surrounding samples, w ij is the weight; when mapping the sample Y in the D-dimensional space to the sample y in the d-dimensional space, the optimization objective function is:

[0006]

[0007] The Laplace eigenmap is a method based on graph theory. It constructs a weighted graph from sample points, then calculates the Laplace matrix and performs eigenvalue decomposition on the matrix to obtain the projection transformation matrix.

[0008] The isometric mapping algorithm calculates the distance between any two samples to construct a distance matrix, and then calculates the shortest path length between any two points i and j:

[0009] D G ={d G (i,j)}

[0010] The optimization goal is to make the shortest distance between points i and j before dimensionality reduction as close as possible to the shortest distance between the two points after dimensionality reduction:

[0011]

[0012] The Tsne method improves the sne method. Both methods convert the Euclidean distance into conditional probability to characterize the similarity between samples, then randomly generate two-dimensional or three-dimensional samples, and use the KL divergence to measure the probability distribution of the data after dimensionality reduction (Gaussian distribution for the sne method and t-distribution for the tsne method) and the probability distribution of the original data as the objective function:

[0013]

[0014] Find the minimum value of this function using gradient descent.

[0015] The general idea of manifold learning is to approximate the distribution characteristics of original high-dimensional data by randomly generated low-dimensional samples. This can achieve good results in image recognition and classification. However, since low-dimensional samples are randomly generated, manifold learning cannot explain the practical significance of low-dimensional samples in geological applications. Summary of the Invention

[0016] The purpose of the present invention is to solve the problems mentioned in the above background technology, improve the TSNE method, and provide a granite rock mass mineralization evaluation method, equipment and medium that can reduce the dimension of high-dimensional data for visual classification.

[0017] The above-mentioned purpose of this application is achieved through the following technical solutions:

[0018] S1: Obtain sample data and category label data of tungsten-bearing granite, tungsten-tin-bearing granite and non-ore-bearing granite in the study area;

[0019] S2: Determine an n×n distance matrix based on the sample data, where n is the number of samples in the sample data;

[0020] S3: Take the first sample of the sample data as the coordinate origin, that is, the first circle center; initialize the positions of the first three circle centers through the distance matrix; perform iterative calculations using the distance matrix and the initialized circle center positions to determine the positions of the remaining n-1 circle centers;

[0021] S4: Based on the position of the circle center of each sample and the category label data, the category and element content are used as point attributes to project the map to evaluate the classification effect.

[0022] Step S2 includes: for element data of the same dimension, the distance calculated is the Euclidean distance, for element data of different dimensions, the Mahalanobis distance is calculated, and the calculated distance is converted into a symmetric matrix, as follows:

[0023] The sample of each sampling point is taken as a row vector, and each sample is composed into a sample matrix;

[0024] The distance between every two samples in the sample matrix is calculated using the pdist function in the MATLAB software, and the calculated distance is transformed into a symmetric matrix using the squareform function in the MATLAB software. The symmetric matrix is the distance matrix.

[0025] Step S31 includes:

[0026] For a data set with n samples, center is the matrix of known sample center positions. First, create an n×n center reserved position matrix wzjz; transpose wzjz and take the upper right triangular matrix to obtain wzjz1; flip wzjz1 by column to obtain wzjz2; transpose wzjz2 to obtain wzjz3; add the wzjz2 and wzjz3 matrices to obtain wzjz4;

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034] Index center through wzjz4 and wzjz2 to create center_ii and center_jj;

[0035] center_ii is x i The reserved variable matrix and center_jj are x j The reserved variable matrix of x is replaced by the reserved variable matrix i and x j Perform calculations;

[0036] Calculate the distances between all known points and the required unknown points to form the radius matrix center_r, index center_r through wzjz4 and wzjz2, and create a reserved radius matrix.

[0037] The objective function is the minimum side length L of the area enclosed by the coordinate matrix of the intersection point, which is expressed as follows:

[0038]

[0039] Where k is the total number of intersection points.

[0040] Step S4 includes: projecting the category of each sample as a point attribute to obtain the classification result; projecting the element content of each sample as a point attribute to obtain the change trend of the element value; and evaluating the classification effect based on the classification result, change trend and category label data.

[0041] An electronic device includes a processor, a memory, a user interface and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs a method for evaluating the mineralization of granite rock mass.

[0042] A computer-readable storage medium stores instructions. When the instructions are executed, a method for evaluating the mineralization of a granite rock mass is performed.

[0043] The beneficial effects of the technical solution provided by this application are:

[0044] Based on the t-SNE method, the distances between all sample points are calculated to form a distance matrix, and the properties of the distance matrix are directly studied. In a two-dimensional plane, the first sample position is initialized as the coordinate origin, and the elements in the distance matrix are used as the radius. A series of circle equations are combined to write the general formula for the coordinates of the first i-1 circle centers. The path enclosed by the intersection is then used as the objective function, and the minimum value of the objective function is calculated. The position of the i-th circle center is determined by the distance between the first i-1 circle centers. The center positions of n samples are determined through continuous iterative calculations. The sample points are colored according to the labels, and the three types of rock mass can be visually classified. By directly considering the properties of the distance matrix, the classification result retains the spatial structure of the sample data. The numerical value of the coordinate in the two-dimensional plane is the distance between the sample point to be classified and the sample point of the known category, which has practical significance in geological applications. This invention realizes the evaluation of the tungsten-tin mineralization capacity of granite, solving the problem that traditional geochemical projection in a data-driven environment cannot clearly distinguish the mineralization of rock masses. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present application will be further described below with reference to the accompanying drawings and embodiments, in which:

[0046] Figure 1 This is a flow chart of the method for evaluating the minerality of a granite rock mass in an embodiment of the present application;

[0047] Figure 2 It is the rock geochemical element projection of the granite rock mass mineralization evaluation method in the embodiment of the present application;

[0048] Figure 3 Schematic diagram of the initialization sample point locations of the granite rock mass mineralization evaluation method in the embodiment of the present application;

[0049] Figure 4 Schematic diagram of the next point position determined by the simulated annealing algorithm in the method for evaluating the mineralization of granite rock mass in the embodiment of the present application;

[0050] Figure 5 This is a comparison chart of the classification results of the granite rock mass mineralization evaluation method in the embodiment of this application;

[0051] Figure 6 This is a comparison diagram of the first classification effect of the granite rock mass mineralization evaluation method in the embodiment of the present application;

[0052] Figure 7 This is a comparison diagram of the second classification effect of the granite rock mass mineralization evaluation method in the embodiment of the present application;

[0053] Figure 8 This is a comparison diagram of the third classification effect of the granite rock mass mineralization evaluation method in the embodiment of this application;

[0054] Figure 9 It is a schematic diagram of the electronic device structure of the method for evaluating the mineralization properties of granite rock mass in the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to have a clearer understanding of the technical features, purposes and effects of this application, the specific implementation methods of this application are now described in detail with reference to the accompanying drawings.

[0056] The embodiments of the present application provide a method for evaluating the mineralization properties of a granite rock mass.

[0057] Please refer to Figure 1 , Figure 1 This is a flow chart of a method for evaluating the mineralization of a granite rock mass in an embodiment of the present application, comprising:

[0058] S1: Obtain sample data and category label data of tungsten-bearing granite, tungsten-tin-bearing granite and non-ore-bearing granite in the study area;

[0059] S2: Determine an n×n distance matrix based on the sample data, where n is the number of samples in the sample data;

[0060] S3: Take the first sample of the sample data as the coordinate origin, that is, the first circle center; initialize the positions of the first three circle centers through the distance matrix; perform iterative calculations using the distance matrix and the initialized circle center positions to determine the positions of the remaining n-1 circle centers;

[0061] S4: Based on the position of the circle center of each sample and the category label data, the category and element content are used as point attributes to project the map to evaluate the classification effect.

[0062] Specifically, elements beneficial to mineralization, such as light and heavy rare earth elements, negative europium anomalies, etc., are selected as point attributes to study the changing characteristics of these elements with rock mass types.

[0063] Specifically, the sample data includes major element data and rare earth element data. There are 10 major element data and 14 rare earth element data, resulting in a 24-dimensional sample data set. Elements beneficial to mineralization, such as light and heavy rare earth elements and negative europium anomalies, are selected as point attributes to study how these elements vary with rock mass type.

[0064] Specifically, the rock geochemical data were mapped according to common indicators, such as AR-SiO_2, A / NK-A / CNK, to study the connections and differences between the three types of rock masses.

[0065] Step S2 includes: for element data of the same dimension, the distance calculated is the Euclidean distance, for element data of different dimensions, the Mahalanobis distance is calculated, and the calculated distance is converted into a symmetric matrix, as follows:

[0066] The sample of each sampling point is taken as a row vector, and each sample is composed into a sample matrix;

[0067] The distance between every two samples in the sample matrix is calculated using the pdist function in the MATLAB software, and the calculated distance is transformed into a symmetric matrix using the squareform function in the MATLAB software. The symmetric matrix is the distance matrix.

[0068] Specifically, the Euclidean distance is calculated for samples with little difference in feature magnitude, and the Mahalanobis distance is calculated for samples with large difference in feature magnitude.

[0069] Step S3 includes:

[0070] S31: pre-assign variables in the programming environment, where the variables are the coordinates of any two circle centers and the corresponding radii at known locations;

[0071] Write the general formula for the intersection coordinates, transform the center coordinates and the corresponding radius into four matrices through matrix transformation, and substitute them into the general formula to speed up the calculation.

[0072] S32: Solve the general equations of the i-1 circle centers simultaneously, using an iterative method, to obtain the intersection point of the i-th circle center. The coordinates (x, y) of the intersection point are expressed as follows:

[0073]

[0074] Among them, (x i ,y i ) and (x j ,y j) is the center of the two known position circles; r i,n+1 and r j,n+1 is the distance between the centers of any two circles at the known positions and the center of the unknown circle required;

[0075] S33: Selecting a preset ratio of intersection points to form a path matrix;

[0076] S34: determining the objective function of the path matrix;

[0077] S35: Obtain the minimum value of the objective function through a simulated annealing algorithm and obtain the path to the minimum value;

[0078] S36: averaging the coordinates of the intersection points of the paths to obtain the coordinates of the center of the i-th circle;

[0079] S37: Repeat steps S32 to S36 to predict the center of the remaining samples.

[0080] Step S31 includes:

[0081] For a data set with n samples, center is the matrix of known sample center positions. First, create an n×n center reserved position matrix wzjz; transpose wzjz and take the upper right triangular matrix to obtain wzjz1; flip wzjz1 by column to obtain wzjz2; transpose wzjz2 to obtain wzjz3; add the wzjz2 and wzjz3 matrices to obtain wzjz4;

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] Index center through wzjz4 and wzjz2 to create center_ii and center_jj;

[0090] center_ii is x i The reserved variable matrix and center_jj are x j The reserved variable matrix of x is replaced by the reserved variable matrix i and x j Perform calculations;

[0091] Calculate the distances between all known points and the required unknown points to form the radius matrix center_r, index center_r through wzjz4 and wzjz2, and create a reserved radius matrix.

[0092] The objective function is the minimum side length L of the area enclosed by the coordinate matrix of the intersection point, which is expressed as follows:

[0093]

[0094] Where k is the total number of intersection points.

[0095] Specifically, in this application, the number of elements in the path matrix is set to half the number of intersections; the number of randomly selected points is generally half the number of all intersections, the objective function is the path distance enclosed by the intersections, and the Metropolis criterion is used to determine whether to accept a new solution in the simulated annealing algorithm.

[0096] Step S4 includes: projecting the category of each sample as a point attribute to obtain the classification result; projecting the element content of each sample as a point attribute to obtain the change trend of the element value; and evaluating the classification effect based on the classification result, change trend and category label data.

[0097] An embodiment of the present application is as follows:

[0098] Step 1: Collection and mapping of geochemical data.

[0099] refer to Figure 2 In this example, 316 samples of tungsten-bearing granite, tungsten-tin-bearing granite, and ore-free granite were collected from the eastern and central sections of the Nanling Mountains. Each sample is composed of major elements and rare earth elements, totaling 24 dimensions. AR-SiO2, ANK-ACNK, and rare earth element diagrams were created to illustrate the differences and connections between the three types of rock masses. It can be seen that the major element diagrams for the three types of rock masses largely overlap, while some samples show some differences in the rare earth element diagrams.

[0100] The present invention also provides an example data set consisting of 50 sets of three-dimensional data corresponding to five categories, for comparison between the method of the present invention and the TSNE method.

[0101] Step 2: Calculate the distance matrix.

[0102] In this embodiment, corresponding to n samples, the distance between any two samples is calculated to establish an n×n symmetric matrix, and the distance matrix is shown in Table 1. For features of the same dimension, the Euclidean distance is used, and for features of different dimensions, the Mahalanobis distance is used.

[0103] Initialize the positions of the first three points through the distance matrix, and set the first sample to the coordinate origin, such as Figure 3 .

[0104] Table 1

[0105]

[0106] Step 3: Calculate the intersection coordinate matrix.

[0107] By making a circle with the coordinates of the three initialized points and the radius from these three points to the fourth point, we can get a series of intersection points. By path planning, we can get the best position of the fourth point. Then, we can make a circle with the coordinates of these four points and the radius from these four points to the fifth point to get the best position of the sixth point. And so on. At most n points can be obtained. An intersection.

[0108] set up

[0109] The intersection coordinates can be expressed as:

[0110]

[0111] or:

[0112]

[0113] The pseudo code for the loop iteration is as follows:

[0114] Center=[]; %center is used to store the intersection coordinates

[0115] For m=4:n

[0116] Center(m)=function(x,y); %function is the coordinate intersection formula of the above formula

[0117] M = m + 1;

[0118] End

[0119] In actual operation, the intersection coordinate formula is solved once every cycle, which is slow. Therefore, the present invention adopts the method of pre-allocating variables for the above-mentioned objective function to speed up the calculation process.

[0120] Step 4: Determine the optimal position of the center of the next sample through the simulated annealing algorithm.

[0121] The objective function is the minimum side length of the area enclosed by the intersection coordinate matrix, which is expressed as follows:

[0122]

[0123] The simulated annealing algorithm is used to find the minimum value of this function to obtain the corresponding intersection coordinates, and the average value of the intersection coordinates is the optimal position of the next point, such as Figure 4 .

[0124] Step 5: Verify the classification effect of the data using the actual data provided by this application and the collected rock geochemical data. Use the categories of the two samples as attributes and the positions of the predicted samples calculated in the above steps as coordinates to perform a projection. Compare the projection results with the known categories to evaluate the correctness of the classification. In this example, the classification effect of the method of this application and the t-SNE method are compared. Figure 5 The classification results of the example data provided by the present invention, Figure 5 (a) is the method of the present invention, Figure 5 (b) is the tsne method; Figure 6 、 Figure 7 and Figure 8 The left side shows the classification results of rock geochemical data after improvement, and the right side shows the results of tsne.

[0125] By directly studying the properties of the distance matrix, the positions of the samples to be classified in the two-dimensional plane are determined through a general formula for sample positions and path optimization. By then converting the attributes of the sample points into element concentrations, the changing trends of element values within the sample distribution range can be determined. The application of manifold learning to dimensionality reduction and visualization of high-dimensional data has been improved. To address the problem of uninterpretable and geologically meaningless results obtained by the TSNE and isometric mapping methods, by studying the properties of the distance matrix and employing iterative intersection point calculation and path optimization, the classification results are more holistic and have clear geological significance. Furthermore, compared to traditional geochemical mapping methods, this method can more accurately distinguish between tungsten-bearing granite, tungsten-tin-bearing granite, and non-ore-bearing granite. It can also reflect the changing trends of elements among the three types of rock masses, providing a basis for mineral exploration.

[0126] It should be noted that the above embodiments provide devices that implement their functions using only the division of the above functional modules as examples. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0127] This application also discloses an electronic device. Figure 9 , Figure 9Schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. The electronic device 500 may include: at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

[0128] The communication bus 502 is used to implement the connection and communication between these components.

[0129] The user interface 503 may include a display screen (Display), and the optional user interface 503 may also include a standard wired interface or a wireless interface.

[0130] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0131] The processor 501 may include one or more processing cores. The processor 501 utilizes various interfaces and lines to connect various parts of the entire server, and executes various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 505, as well as calling data stored in the memory 505. Optionally, the processor 501 may be implemented in the form of at least one hardware component selected from digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 501 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 501 and may be implemented separately on a single chip.

[0132] The memory 505 may include a random access memory (RAM) or a read-only memory (Read-Only Memory).

[0133] Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 optionally also includes but is not limited to at least one storage device located away from the aforementioned processor 501. Reference Figure 9 The memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program for a method for evaluating the mineralization properties of a granite rock mass.

[0134] exist Figure 9 In the electronic device 500 shown, the user interface 503 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 501 can be used to call an application program for a method for evaluating the mineralization of a granite rock mass stored in the memory 505. When executed by one or more processors 501, the electronic device 500 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for this application.

[0135] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0136] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed includes but is not limited to indirect coupling or communication connection of devices or units through some service interfaces, including but not limited to electrical or other forms.

[0137] Units described as separate components include, but are not limited to, or may not be physically separate. Components shown as units include, but are not limited to, or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of this embodiment.

[0138] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, including but not limited to each unit being physically separate, or two or more units being integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0139] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent commodity, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0140] The above are merely exemplary embodiments of the present disclosure and are not intended to limit the scope of the present disclosure. In other words, any equivalent variations and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the disclosure and the practical implications thereof.

[0141] This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not described herein. The description and examples are to be considered as exemplary only, and the scope and spirit of the present disclosure are to be defined by the claims.

Claims

1. A method for evaluating the minerality of a granite rock mass, characterized in that: The method comprises the following steps: S1: Obtain sample data and category label data of tungsten-bearing granite, tungsten-tin-bearing granite and non-ore-bearing granite in the study area; S2: Determine an n×n distance matrix based on the sample data, where n is the number of samples in the sample data; Step S2 includes: for element data of the same dimension, the distance calculated is the Euclidean distance, for element data of different dimensions, the Mahalanobis distance is calculated, and the calculated distance is converted into a symmetric matrix, as follows: The sample of each sampling point is taken as a row vector, and each sample is composed into a sample matrix; The distance between each two samples in the sample matrix is calculated using the pdist function in MATLAB software, and the calculated distance is transformed into a symmetric matrix using the squareform function in MATLAB software. The symmetric matrix is the distance matrix. S3: Take the first sample of the sample data as the coordinate origin, that is, the first circle center; initialize the positions of the first three circle centers through the distance matrix; perform iterative calculations using the distance matrix and the initialized circle center positions to determine the positions of the remaining n-1 circle centers; Step S3 includes: S31: pre-assign variables in the programming environment, where the variables are the coordinates of any two circle centers and the corresponding radii at known locations; Write the general formula for the intersection coordinates, and use matrix transformation to separate the center coordinates and the corresponding radius into four matrices; substitute them into the general formula to speed up the calculation; S32: Solve the general equations of the i-1 circle centers simultaneously using an iterative method to obtain the intersection point of the i-th circle center. The coordinates (x, y) of the intersection point are expressed as follows: Among them, (x i ,y i ) and (x j ,y j ) is the center of the two known position circles; r i+1 and r j+1 is the distance between the centers of any two circles at the known positions and the center of the unknown circle required; S33: Selecting a preset ratio of intersection points to form a path matrix; S34: determining the objective function of the path matrix; S35: Obtain the minimum value of the objective function through a simulated annealing algorithm and obtain the path to the minimum value; S36: averaging the coordinates of the intersection points of the paths to obtain the coordinates of the center of the i-th circle; S37: Repeat steps S32 to S36 to predict the center of the remaining samples; S4: Based on the position of the circle center of each sample and the category label data, the category and element content are used as point attributes to project the map to evaluate the classification effect.

2. A method for evaluating the mineralization of a granite rock mass according to claim 1, characterized in that: Step S31 includes: For a data set with n samples, center is the matrix of known sample center positions. First, create an n×n center reserved position matrix wzjz; transpose wzjz and take the upper right triangular matrix to obtain wzjz1; flip wzjz1 by column to obtain wzjz2; transpose wzjz2 to obtain wzjz3; add the wzjz2 and wzjz3 matrices to obtain wzjz4; Index center through wzjz4 and wzjz2 to create center_ii and center_jj; center_ii is x i The reserved variable matrix and center_jj are x j The reserved variable matrix of x is replaced by the reserved variable matrix i and x j Perform calculations; Calculate the distances between all known points and the required unknown points to form the radius matrix center_r, index center_r through wzjz4 and wzjz2, and create a reserved radius matrix.

3. A method for evaluating the mineralization of a granite rock mass according to claim 2, characterized in that: The objective function is the minimum side length L of the area enclosed by the coordinate matrix of the intersection point, which is expressed as follows: Where k is the total number of intersection points.

4. A method for evaluating the mineralization of a granite rock mass according to claim 1, characterized in that: Step S4 includes: projecting the category of each sample as a point attribute to obtain the classification result; projecting the element content of each sample as a point attribute to obtain the change trend of the element value; and evaluating the classification effect based on the classification result, change trend and category label data.

5. An electronic device, characterized in that: The electronic device (500) comprises a processor (501), a memory (505), a user interface (503) and a network interface (504), wherein the memory (505) is used to store instructions, the user interface (503) and the network interface (504) are used to communicate with other devices, and the processor (501) is used to execute the instructions stored in the memory (505) so that the electronic device (500) executes the method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed, the method steps according to any one of claims 1 to 4 are performed.

Images