Mineral prospecting prediction method and device based on geoscience big data fuzzy feature analysis

Abstract

The invention provides a prospecting prediction method and device based on geoscience big data fuzzy feature analysis. The method comprises the following steps: establishing a geoscience big data spatial database of a target prospecting area; dividing the target prospecting area to generate a plurality of grid units; selecting a plurality of model units from the plurality of grid units, and selecting an initial prediction variable; respectively determining a fuzzy membership function of any initial prediction variable in each model unit; obtaining a fuzzy membership degree of an initial prediction variable of the model unit according to the fuzzy membership degree function; calculating a fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership degree of the initial prediction variable of the model unit; determining a weight coefficient of each initial prediction variable according to the fuzzy matching coefficient; selecting a target prediction variable according to the weight coefficients of all the initial prediction variables; and calculating the metallogenic favorable degree of the target prospecting area according to the fuzzy membership function and the weight coefficient of the target prediction variable. The prediction precision can be improved.

Description

Technical Field

[0001] This application belongs to the technical field of big data prospecting, and more particularly, it relates to a prospecting prediction method and device based on fuzzy feature analysis of geoscience big data. Background Art

[0002] In recent years, big data technology has been applied to mineral exploration. By analyzing multi-source, heterogeneous, and massive geoscience big data, the distribution and enrichment laws of mineral resources can be predicted more accurately, reducing exploration risks. The feature analysis method is a multivariate statistical analysis method. It deeply analyzes geoscience big data to establish a feature model reflecting a certain type of ore deposit, aiming to predict this type of ore deposit. This method has a simple model structure, loose application conditions, and strong geological interpretability, so it was widely used by many geological and mineral personnel in the early days.

[0003] However, in recent years, this method has been neglected, mainly because it has an obvious defect: as a semi-quantitative method, it requires that the prediction variables must be logical variables. A logical variable represents logically opposite states, such as the presence or absence of geological factors, extremely high or low anomalies, etc. This is suitable for discrete variables such as the presence or absence of a certain geological factor, but not suitable for element content data obtained from geochemical exploration because they are continuous variables. A large amount of useful detailed information will be lost during the conversion into logical variables, ultimately reducing the accuracy of the prediction results.

[0004] Therefore, aiming at the deficiencies of the traditional feature analysis method, there is an urgent need to propose an improved prospecting prediction method to continue to maintain its technical advantages, and at the same time, while improving the prediction accuracy, it can also adapt to more mineral resource exploration scenarios. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a prospecting prediction method and device based on fuzzy feature analysis of geoscience big data, which can improve the prediction accuracy and at the same time adapt to more mineral resource exploration scenarios.

[0006] Embodiments of the present invention provide a prospecting prediction method based on fuzzy feature analysis of geoscience big data, including:

[0007] Establish a geoscience big data spatial database for the target prospecting area, where the geoscience big data spatial database includes various types of data;

[0008] Divide the target prospecting area to generate multiple grid cells;

[0009] Select multiple model cells from multiple grid cells, where the model cells are grid cells with a higher exploration degree and where one or more ore deposits have been discovered.

[0010] Select initial prediction variables from the geoscience big data spatial databases corresponding to each model unit, where the initial prediction variables are at least two types of data materials of different types;

[0011] Determine the fuzzy membership function of any one of the initial prediction variables in each model unit respectively;

[0012] Obtain the fuzzy membership degree of the initial prediction variable of the model unit according to the fuzzy membership function; calculate the fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership degree of the initial prediction variable of the model unit;

[0013] Determine the weight coefficient of each initial prediction variable according to the fuzzy matching coefficient;

[0014] Select the target prediction variable according to the weight coefficients of all the initial prediction variables;

[0015] Obtain the metallogenic favorability of all grid cells in the target ore prospecting area according to the fuzzy membership function and weight coefficient of the target prediction variable.

[0016] Furthermore, the ore prospecting prediction method further includes: establishing a geoscience big data spatial database for the target ore prospecting area, where the geoscience big data spatial database includes various types of data materials; the steps of establishing the geoscience big data spatial database for the target ore prospecting area include:

[0017] Obtain multiple different types of data materials of the target ore prospecting area, and the data scales or precisions of each data material are the same, and include at least two of geology, geophysical exploration, geochemical exploration, remote sensing, and ore occurrences;

[0018] Based on the same coordinate system, perform spatial coordinate transformation processing and data format conversion processing on various types of data materials to generate the geoscience big data spatial database.

[0019] Furthermore, calculate the area of the grid cell according to the following formula:

[0020]

[0021] Where d is the area of the grid cell, D is the preset data scale or precision, and β is the influence factor.

[0022] Furthermore, when the initial prediction variable is a discrete variable, determine the fuzzy membership function of the initial prediction variable according to the following formula:

[0023]

[0024] where, μ A (x ij ) is the fuzzy membership degree that the j-th predictive variable of the i-th model unit belongs to the fuzzy set A, x ij is the value of the j-th predictive variable of the i-th model unit, m is the number of initial predictive variables, and n is the number of model units;

[0025] When the initial predictive variables are continuous variables, the following formula is used to determine the fuzzy membership degree function of the initial predictive variables:

[0026]

[0027] where, a and b are constants, and a < b, and a and b are determined based on the cumulative distribution function curves of the respective initial predictive variables.

[0028] Furthermore, the following formula is used to calculate the fuzzy matching coefficient:

[0029]

[0030] where, r jk is the fuzzy matching coefficient between the j-th initial predictive variable and the k-th initial predictive variable, μ A (x ij ) is the fuzzy membership degree of the j-th initial predictive variable of the i-th model unit, μ A (x ik ) is the fuzzy membership degree of the k-th initial predictive variable of the i-th model unit, m is the number of initial predictive variables, and n is the number of model units.

[0031] Furthermore, the following formula is used to determine the weight coefficients of the respective initial predictive variables:

[0032]

[0033] where, a j is the weight coefficient of the j-th initial predictive variable.

[0034] Furthermore, the step of selecting the target predictive variable according to the weight coefficients of all the initial predictive variables includes:

[0035] Based on the weight coefficients of all the initial predictive variables, a weight coefficient change curve is plotted in descending order, and according to the weight coefficient drop turning points on the weight coefficient change curve, all the initial predictive variables before the drop turning points are used as the selected target predictive variables.

[0036] Furthermore, the calculation formula of the metallogenic favorability is:

[0037]

[0038] Among them, f is the metallogenic favorability of the target prospecting area, μ A (x1), μ A (x2),......, μ A (x p ) is the fuzzy membership degree of each target prediction variable, a1, a2,......, a p are the weight coefficients of each target prediction variable, and p is the number of target prediction variables.

[0039] Furthermore, the prospecting prediction method further includes: dividing the types of each prospecting far - reaching area in the target prospecting area according to the metallogenic favorability of the target prospecting area.

[0040] An embodiment of the present invention also provides a prospecting prediction device based on the fuzzy feature analysis of geoscience big data, including:

[0041] A construction unit, configured to establish a geoscience big data spatial database of the target prospecting area, and the geoscience big data spatial database includes various types of data materials;

[0042] A division unit, configured to divide the target prospecting area to generate a plurality of grid units;

[0043] A first selection unit, configured to select a plurality of model units from the plurality of grid units, where the model units are grid units with a higher exploration degree and where one or more ore - producing areas have been discovered; and select initial prediction variables from the geoscience big data spatial databases corresponding to each model unit, where the initial prediction variables are at least two of the various types of data materials;

[0044] A processing unit, configured to respectively determine the fuzzy membership degree function of any one of the initial prediction variables in each model unit; and obtain the fuzzy membership degree of the initial prediction variable of the model unit according to the fuzzy membership degree function, calculate the fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership degree of the initial prediction variable of the model unit, and determine the weight coefficient of each initial prediction variable according to the fuzzy matching coefficient;

[0045] A second selection unit, configured to select target prediction variables according to the weight coefficients of all the initial prediction variables;

[0046] A determination unit, configured to calculate the metallogenic favorability of all grid units in the target prospecting area according to the fuzzy membership degree function and the weight coefficients of the target prediction variables.

[0047] Compared with the prior art, the prospecting prediction method based on geoscience big data fuzzy feature analysis provided by the embodiments of the present invention can obtain various types of data materials for characterizing the ore content in the target prospecting area by establishing a geoscience big data spatial database of the target prospecting area. According to various types of data materials, the target prospecting area can be divided to generate multiple grid units, which reduces the computational complexity. In addition, at least one model unit is selected from multiple grid units. Since the model unit is a grid unit with a relatively high exploration degree and one or more ore-producing areas have been discovered, initial prediction variables can be selected therefrom, and the initial prediction variables can reflect the favorable degree of mineralization. Then, the fuzzy membership function of any initial prediction variable in each model unit can be determined respectively. Since the fuzzy membership function characterizes the relationship between different initial prediction variables and makes full use of the data in the geoscience big data spatial database, avoiding the loss of detailed information, compared with the traditional prediction scheme based on logical variables, the accuracy of the fuzzy matching coefficient can be further improved, and then the weight coefficient of the initial prediction variable is increased, so that the selected target prediction variable can more accurately reflect the mineral distribution in the current target prospecting area, improving the calculation accuracy of the favorable degree of mineralization. Moreover, this scheme is based on the traditional feature analysis method, so that while improving the prediction accuracy, it can also adapt to more mineral resource exploration scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0049] Figure 1 The flowchart of a prospecting prediction method based on geoscience big data fuzzy feature analysis according to an embodiment of the present invention is shown;

[0050] Figure 2 The schematic diagram of the principle of a regional division according to an embodiment of the present invention is shown;

[0051] Figure 3 The schematic diagram of the principle of determining a constant value according to an embodiment of the present invention is shown;

[0052] Figure 4 The schematic diagram of the curve change of the weight coefficient of an initial prediction variable according to an embodiment of the present invention is shown;

[0053] Figure 5 The schematic diagram of the structure of a prospecting prediction device based on geoscience big data fuzzy feature analysis according to an embodiment of the present invention is shown. Specific Embodiments

[0054] As described in the background art, while the feature analysis method is introduced, there are also some new and intractable problems in the field of mineral resources exploration. Among them, a relatively serious problem is that as a semi - quantitative method, the predictive variables must be logical variables. A logical variable refers to a variable that represents logically opposite states, such as the presence or absence of geological factors, the abnormally high or low values, etc. The most common is the binary representation method, which uses 1 and 0 to represent two opposite states respectively.

[0055] This is suitable for discrete variables such as the presence or absence of a certain geological factor, but not suitable for the element content data obtained from geochemical exploration because they are continuous variables. During the process of converting them into logical variables, a large amount of useful detailed information will be lost, and ultimately the accuracy of the prediction results will be reduced.

[0056] To solve the above - mentioned technical problems, the embodiments of the present invention provide a prospecting prediction scheme based on fuzzy feature analysis of geoscience big data. By establishing a geoscience big data spatial database for the target prospecting area, a large amount of data of various different types, which are used to characterize the ore - bearing amount of the target prospecting area, can be obtained. According to the data of various different types, the target prospecting area can be divided to generate multiple grid units, which reduces the computational complexity. And this scheme also selects multiple model units from multiple grid units. Since the model units are grid units with a relatively high exploration degree and where one or more ore - producing areas have been discovered, initial predictive variables can be selected from them, and the initial predictive variables can reflect the favorable degree of mineralization. Then, the fuzzy membership function of any one of the initial predictive variables in the model unit can be determined. Since the fuzzy membership function makes full use of the data in the geoscience big data spatial database, it avoids the loss of detailed information. Compared with the traditional prediction scheme based on logical variables, it can further improve the accuracy of the fuzzy matching coefficient, and then improve the accuracy of the weight coefficient of the initial predictive variable, so that the selected target predictive variable can more accurately reflect the mineral distribution in the current target prospecting area, improve the calculation accuracy of the favorable degree of mineralization, and this scheme inherits the advantages of the traditional feature analysis method. Therefore, while adapting to more mineral resources exploration scenarios, it can also improve the prediction accuracy.

[0057] In other words, by improving the deficiencies of the traditional feature analysis method, it is expected to continue to maintain its technical advantages, and while adapting to more mineral resources exploration scenarios, it can also improve the prediction accuracy.

[0058] To enable those skilled in the art to better understand the inventive concept, working principle and advantages of the embodiments of the present invention, the following will describe in detail the prospecting scheme based on fuzzy feature analysis of geoscience big data in the embodiments of the present invention.

[0059] See Figure 1 the flowchart of a prospecting prediction method based on the fuzzy feature analysis of geological big data in an embodiment of the present invention as shown in Figure 1 shown, the following steps can be executed:

[0060] S10. Establish a geological big data spatial database for the target prospecting area, and the geological big data spatial database includes various different types of data materials.

[0061] In this embodiment, based on typical ore deposits and regional metallogenic laws (for example, time, space, genesis, and ore types), and combined with existing regional geological and mineral resources data, select areas favorable for mineralization as the target prospecting area, which can improve the possibility of potential resources detected.

[0062] Based on typical ore deposits and regional metallogenic laws (time, space, genesis, and ore types), and combined with existing regional geological and mineral resources data, select areas favorable for the formation of the target ore deposit type as the target prospecting area.

[0063] For example, select skarn-type copper polymetallic ore as the target ore deposit type.

[0064] In this embodiment, the screening criteria for the target prospecting area favorable for mineralization include: 1) The target prospecting area is located within the scope of a known large metallogenic belt or ore concentration area; 2) There are known ore deposits (for example, skarn-type copper polymetallic ore deposits) around the target prospecting area.

[0065] It should be noted that the selection criteria listed in the above examples are only for illustrative purposes and are used to represent the requirements met by the target prospecting area, and should not be construed as a limitation to the present invention.

[0066] In this embodiment, various different types of data materials refer to parameters for characterizing different parameter information of the target prospecting area. By obtaining multiple different types of data materials, the topography and geological parameters of the target prospecting area can be more truly reflected.

[0067] In this embodiment, step S10 may include: obtaining geological big data materials for the target prospecting area; performing spatial coordinate conversion processing and data format conversion processing on various types of geological big data materials based on the same coordinate system to generate the geological big data spatial database.

[0068] Among them, multiple different types of geological big data materials may include at least two of geological data, geophysical exploration data, geochemical exploration data, remote sensing data, and ore field data.

[0069] More specifically, geological data mainly involves regional geological maps.

[0070] The collection of geophysical exploration data mainly involves the regional gravity, magnetic method, and radioactive measurement data obtained from geophysical reconnaissance work.

[0071] The collection of geochemical exploration data mainly involves the regional stream sediment, rock debris, and soil measurement data obtained from geochemical reconnaissance work. The media vary for different scales and different geochemical landscapes.

[0072] The collection of remote sensing data involves a wide variety of remote sensing data types. Which specific types of data to collect need to be determined according to the target ore deposit type and target task. Generally, it can be roughly divided into remote sensing data for alteration information extraction and remote sensing data for ore-controlling structure and ore-bearing geological body information extraction.

[0073] The collection of ore occurrence data mainly involves the discovered ore deposits (points) in the working area, including information such as coordinates and the cumulative identified resource volume.

[0074] It should be noted that: Generally, for the convenience of similar analogy, when collecting geological, geophysical, geochemical, and remote sensing data in the working area, the principle of scale equivalence should be adhered to, that is, the same type of data in the working area needs to achieve full coverage, the scales need to be as consistent as possible, and it is best to select data with a larger scale.

[0075] In other words, the data scales or precisions of each data are the same.

[0076] In this embodiment, the current data is often stored in vector files, raster files, or table files, and some old data exists in paper media and needs to be vectorized first.

[0077] Specifically, due to the different types, ages, and sources of the above-mentioned geological, mineral, geophysical, geochemical, and remote sensing data collected, there are also significant differences in the geographic and projection coordinate systems. Among them, the common geographic coordinate systems corresponding to the projection coordinate system include Beijing 54 Geodetic Coordinate System, Xi'an 80 Geodetic Coordinate System, CGCS2000 Geodetic Coordinate System, and WGS84 Geodetic Coordinate System; the projection coordinate systems include Gauss-Krüger (3-degree zone or 6-degree zone), Universal Transverse Mercator, and Lambert projection.

[0078] In order to perform data analysis within a unified spatial coordinate system, the geographic coordinate system will uniformly adopt the CGCS2000 Geodetic Coordinate System with the centroid coordinate system as the core in China. For a relatively small area, the Gauss-Krüger (3-degree zone or 6-degree zone) projection is used for the projection coordinate system, and for a very large area, the Lambert projection is used. The conversion of different spatial coordinate systems to a unified coordinate system can be carried out by using three-parameter or seven-parameter projection transformation methods in GIS software, and finally a unified geoscience big data spatial database is established.

[0079] In other words, based on the same coordinate system, performing spatial coordinate transformation processing on various types of data can improve the format consistency among various types of data, which is conducive to reducing the difficulty of subsequent data processing in the geoscience big data spatial database.

[0080] In this embodiment, while unifying the coordinate system, data conversion operations can also be performed.

[0081] For example, rasterize all data such as geology, ore deposits (points), geophysical exploration, geochemical exploration, and remote sensing. Among them, for vector files, the "polygon to raster" or "point to raster" method can be directly used for rasterization; for tabular files, the "inverse distance weighted" method is used for interpolation to form rasterized data; for raster files, the data format remains unchanged.

[0082] It should be noted that for some special geological units in geological data, such as faults, outer contact zones of rock bodies, and tectonic intersection parts, generally, the "buffer analysis" method needs to be used first to determine the influence range, and then the "polygon to raster" method is used for rasterization processing. Finally, a unified geoscience big data spatial database is established.

[0083] It should also be noted that the multiple different types of data in the above examples are only for illustrative purposes and are used to represent obtaining multiple data, and should not be construed as a limitation of the present invention. In some other embodiments, the data in the acquisition area may further include: mask data.

[0084] S20. Divide the target ore prospecting area to generate a plurality of grid units.

[0085] Specifically, the grid unit is the most commonly used basic unit for prospecting prediction in this solution. Dividing grid units in the target working area is the basic task of carrying out prospecting prediction. By dividing the target ore prospecting area into a plurality of grid units, it is conducive to reducing the execution difficulty of subsequent steps.

[0086] In this embodiment, when performing step S20, the most critical issue is to select the size of the grid unit, because different divisions of the grid unit size will have a greater impact on the prediction results.

[0087] In this embodiment, the data limits the topography and geological parameters of the target ore prospecting area, and the target ore prospecting area already includes the ore-bearing area. Therefore, based on the obtained data, multiple model units can be selected from them, and the selected model units are used as a reference basis for the ore prospecting prediction process.

[0088] In addition, when conducting prospecting prediction, the data accuracy or scale of geology, geophysical exploration, geochemical exploration, and remote sensing should be as consistent as possible, and the corresponding accuracy or scale should be used as the basis for dividing the size of grid units.

[0089] Furthermore, the scales of different target ore deposit types vary greatly. For example, for general endogenous ore deposits, the distribution range of the ore deposit is approximately from a few tenths of a square kilometer to several square kilometers, while the scale of exogenous ore deposits is often much larger, generally from several square kilometers to dozens of square kilometers. Therefore, the spatial distribution of the target ore deposit type is another important basis for dividing the size of grid units.

[0090] Based on this, step S20 may include: dividing the target prospecting area according to the data scale or accuracy of the data and the type of the target prospecting area to generate multiple grid units, and each grid unit has its own area.

[0091] It should be noted that the preset data scale or accuracy can be obtained from a Geographic Information System (GIS) platform; or it can be understood that the data is obtained at this data scale or accuracy.

[0092] Specifically, the data scale or accuracy limits the detection accuracy when obtaining data, and the type of the target prospecting area limits the influence of the ore deposit type on the grid unit. In this way, after obtaining these two parameters, the grid division operation can be performed to form multiple grid units, so that each grid unit has its own area.

[0093] More specifically, the area of each grid unit can be determined in the following way:

[0094]

[0095] Where d is the area of the grid unit, D is the preset data scale or accuracy, and β is the influence factor.

[0096] It should be noted that, first, the value of β is generally taken as 1 in the prediction of endogenous ore deposits, while a larger value is generally taken in exogenous ore deposits, and the specific value should be determined according to the actual distribution of the target ore deposit type; second, the value of D is determined according to the data scale or accuracy, and generally represents the line spacing of route surveys or the side length of the sampling area of a single sample. Among them, the value of D for prospecting prediction at a scale of 1:50,000 is generally taken as 0.5 km, and the value of D for prospecting prediction at a scale of 1:200,000 is generally taken as 2 km; third, is a correction to the value of D, mainly to prevent the loss of local or detailed information.

[0097] Moreover, the area of the grid unit determined by the above formula can accurately divide the size of the grid unit. On the one hand, it avoids the problem of data thinning and loss of useful information due to the grid unit being too large, thereby improving the prediction accuracy; on the other hand, it avoids the problem of increasing the amount of calculation and affecting the operation efficiency due to the grid unit being too small, so that the calculation results will not produce the "salt and pepper" effect.

[0098] As an example, see Figure 2 A schematic diagram of a principle of area division in an embodiment of the present invention is shown in FIG. Figure 2 As shown, the target prospecting area 10 can be divided according to the data scale or accuracy of the data and the type of the target prospecting area to generate multiple grid units (e.g. Figure 2 Schematic grid unit 12).

[0099] In some embodiments, the areas of the grid cells 12 may be the same.

[0100] S30, selecting a plurality of model units from a plurality of grid units, wherein the model units are grid units with a high degree of exploration and in which one or more mineral deposits have been discovered.

[0101] Specifically, a plurality of grid units together constitute a target prospecting area, and different areas in the target prospecting area correspond to different parameter characteristics, so a plurality of model units can be selected based on different types of data.

[0102] It should be noted that this step is performed on a geographic information system (GIS) platform, and sufficient evidence is required in the process of obtaining the model unit.

[0103] First, there is at least one mineral deposit within the grid, that is, after exploration, a mineral deposit with a small or larger scale of resources has been submitted; second, the exploration level of the mineral deposit should be high, and it is believed that the resource volume within the grid unit has been basically ascertained, so the exploration should at least reach the level of a census, and it is best to reach the level of detailed investigation or exploration.

[0104] For example, see Figure 2 , by performing the partitioning operation, 3 model units can be selected from multiple grid units.

[0105] In other words, it should be noted that the relationship between the target prospecting area, grid unit and model unit is that the grid unit covers the entire target prospecting area with a fixed size, while the model unit is the ore-bearing unit in the grid unit.

[0106] It should be specifically noted that: the implementation of the feature analysis method determines the initial prediction variables by summarizing the ore-controlling factors and ore-prospecting indicators of multiple model units. Therefore, the number of model units is preferably multiple, and the more the better.

[0107] S40, select initial prediction variables from the geo-big data spatial databases corresponding to each model unit, where the initial prediction variables are at least two of multiple different types of data materials.

[0108] Specifically, the model unit contains multiple different types of data materials, and the roles played by these data materials in the prediction process are different. Therefore, initial prediction variables can be selected based on the understanding of ore-forming geology, and these initial prediction variables are data materials of ore-controlling factors and ore-prospecting indicators that are favorable for ore formation.

[0109] In this embodiment, there are at least two ways to select ore-controlling factors and ore-prospecting indicators that are favorable for ore formation as preliminary prediction variables. Among them: one is to select based on the understanding of ore-forming geology. For example, for magmatic copper-nickel sulfide deposits, through the study of the ore-forming laws of a large number of typical deposits, it is found that such deposits require mantle-derived magma to invade and emplace through super-crustal major faults to finally form. Therefore, the major faults in the geological map can be selected as one of the initial prediction variables.

[0110] The other is to select prediction variables by comparing whether a certain prediction variable has an indicative effect on the location of the ore deposit on the map. For example, for low-temperature hydrothermal gold deposits, gold deposits often occur at medium to high values of the contents of elements such as Au, As, Bi, and Hg, while less occur at low values. Therefore, the contents of elements Au, As, Bi, and Hg in the geochemical exploration data can be selected as part of the initial prediction variables.

[0111] It should be specifically noted that: the core of the implementation of the feature analysis method is to determine the importance of variables through the relationships between variables. Therefore, the selection of initial prediction variables is at least more than two, otherwise the feature analysis method will not be able to determine the weight coefficients of the variables, and the feature analysis method will not be applicable.

[0112] S50, respectively determine the fuzzy membership functions of any one of the initial prediction variables in each model unit.

[0113] In this embodiment, when carrying out ore-prospecting prediction for grid units, there are multiple initial prediction variables in the model unit. By introducing the fuzzy membership function, continuous variables are transformed into membership values, which can make full use of geo-big data, avoid the loss of detailed information, and thus improve the accuracy of the prediction results while maintaining the advantages of traditional feature analysis.

[0114] In this embodiment, according to the numerical types of the respective initial prediction variables, the corresponding fuzzy membership functions are determined.

[0115] More specifically, when the initial prediction variable is a discrete variable, the following formula is used to determine the fuzzy membership function of the initial prediction variable:

[0116]

[0117] where μ A (x ij ) is the fuzzy membership of the j-th prediction variable of the i-th model unit belonging to the fuzzy set A, x ij is the value of the j-th prediction variable of the i-th model unit, m is the number of initial prediction variables, and n is the number of model units.

[0118] where x ij is directly read on the GIS software platform.

[0119] In other words, when the initial prediction variable is a logical variable expressing existence or non-existence, based on the existence or non-existence of the initial prediction variable, 0 or 1 can be taken.

[0120] For example, whether a granite body exists or is absent in a certain place. If it exists, the value is 1; if it is absent, the value is 0. Another example is whether the magnetic anomaly in a certain place is high or low. If it is high, the value is 1; if it is low, the value is 0.

[0121] When the initial prediction variable is a continuous variable, in order to prevent the loss of information during the prediction process, another method can be used to determine the fuzzy membership function of the initial prediction variable.

[0122] Among them, a continuous variable refers to a variable whose values are continuous. For example, the Cu element content data obtained from the geochemical prospecting survey in a certain place is a continuous variable, and its value range can be any real number greater than 0.

[0123] Specifically, when the initial prediction variable is a continuous variable, the following formula is used to determine the fuzzy membership function of the initial prediction variable:

[0124]

[0125] where a and b are constants, and a < b. a and b are determined based on the inflection point values of the cumulative distribution function curves of the respective initial prediction variables.

[0126] As a specific example, refer to Figure 3 the schematic diagram of the principle of determining the constant value in an embodiment of the present invention shown in Figure 3As shown, it is a cumulative distribution function curve, where the abscissa represents the initial prediction variable and the ordinate represents the value of the cumulative distribution function of the initial prediction variable. By using the cumulative distribution function and accumulating the distribution probabilities of each initial prediction variable, the inflection points a and b of the required membership function curve can be determined therefrom.

[0127] More specifically, the cumulative distribution function is as follows:

[0128]

[0129] where F U (u) is the cumulative distribution function, u represents the upper limit of integration, x is the initial prediction variable, and f(x) is the probability density function of the prediction variable.

[0130] It should be noted that in the actual prediction process, a and b can also be obtained through other means.

[0131] Thus, based on the properties of the initial prediction variables, calculations are performed through different situations, improving the accuracy of the fuzzy membership function.

[0132] S60. Obtain the fuzzy membership of the initial prediction variable of the model unit according to the fuzzy membership function; calculate the fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership of the initial prediction variable of the model unit.

[0133] In this embodiment, the initial prediction variables refer to the ore-controlling factors and ore prospecting signs related to mineralization. The introduction of the fuzzy membership function prevents the loss of detailed information of continuous prediction variables. Therefore, based on the membership of the initial prediction variables, the fuzzy matching coefficient between any two initial prediction variables in each model unit can be determined through the matching coefficient method.

[0134] In a specific embodiment, assuming that there are n model units and m initial prediction variables in the target ore prospecting area, the fuzzy matching coefficient between each initial prediction variable and other initial prediction variables in any one model unit can be determined.

[0135] More specifically, step S60 may include: respectively calculating the product value between the fuzzy membership function of any one initial prediction variable and the fuzzy membership function of any other one; taking the sum of all product values of the any one initial prediction variable on the n model units as the fuzzy matching coefficient.

[0136] In this embodiment, the following formula is used to calculate the fuzzy matching coefficient between any one initial prediction variable and all other initial prediction variables:

[0137]

[0138] where r jk is the fuzzy matching coefficient between the k-th initial prediction variable of the j-th initial prediction variable, and μ A (x ij ) is the fuzzy membership degree of the j-th initial prediction variable of the i-th model unit, and μ A (x ik ) is the fuzzy membership degree of the k-th initial prediction variable of the i-th model unit. m is the number of initial prediction variables, and n is the number of model units.

[0139] S70. Determine the weight coefficient of each initial prediction variable according to the fuzzy matching coefficients between any one initial prediction variable and all other initial prediction variables.

[0140] In this embodiment, the weight coefficient also determines the importance degree of the initial prediction variable. Therefore, based on the weight coefficient, it can be used as a determining factor for the target prediction variable.

[0141] In one embodiment, step S70 may include: determining the sum value of the squares of the fuzzy matching coefficients corresponding to any one initial prediction variable; taking the square root value of the sum value as the weight coefficient of the initial prediction variable.

[0142] More specifically, the weight coefficients of each initial prediction variable can be determined in the following manner:

[0143]

[0144] where a j is the weight coefficient of the j-th initial prediction variable.

[0145] In this embodiment, the cumulative sum of the squares of the fuzzy matching coefficients is calculated and the square root value is taken to calculate the weight coefficient of the prediction factor. Among them, the ideological basis of the sum of squares method is that the stronger the correlation between the initial prediction variable and other initial prediction variables, the more important the initial prediction variable is, and the greater the impact on mineralization, which conforms to the characteristics of geological mineralization.

[0146] S80. Select the target prediction variable from all the initial prediction variables based on the magnitudes of the weight coefficients of each initial prediction variable within the model unit.

[0147] The weight coefficients of different initial prediction variables vary in magnitude, which reflects the importance of the initial prediction variables. Therefore, the initial prediction variables serving as the target prediction variables can be determined based on the magnitudes of the weight coefficients of the initial prediction variables. Specifically, according to the weight coefficients of each initial prediction variable, they are sorted in descending order, and a schematic diagram of the weight coefficient change curve is established; according to the waveform changes indicated in the schematic diagram of the weight coefficient change curve, the turning points of the weight coefficient decline are determined; the number of target prediction variables is determined according to the positions of the turning points of the weight coefficient decline. That is to say, based on the weight coefficients of all initial prediction variables, the weight coefficient change curve is plotted in descending order, and according to the turning points of the weight coefficient decline on the weight coefficient change curve, all the initial prediction variables before the turning points of the decline are used as the selected target prediction variables.

[0148] Since the turning points on the weight coefficient change curve represent a sudden decrease in the magnitude of the weight coefficient, the initial variables corresponding to such significantly decreased weight coefficients are no longer important and can be removed. Therefore, in this embodiment, the number of target prediction variables is determined by the turning points of the sudden significant decrease on the weight coefficient change curve, rather than arbitrarily deciding in advance how many variables to take. Exemplarily, referring to Figure 4 as shown, it is determined that the turning point on the weight coefficient curve is P. Among the m initial prediction variables, the first p initial prediction variables with weight coefficient values greater than the weight coefficient at the turning point P are used as the final prediction variables for constructing the model. In other words, the prediction variables before the turning points of the decline on the weight coefficient curve are retained, and the prediction variables after the turning points are discarded. Therefore, when there are m initial prediction variables, p target prediction variables can be selected from them.

[0149] S90. According to the fuzzy membership function and weight coefficient of the target prediction variable, obtain the metallogenic favorability of all grid cells in the target ore prospecting area. The metallogenic favorability is a comprehensive index indicating the possibility of mineral formation and enrichment in the geological units of the target ore prospecting area.

[0150] Exemplarily, step S90 includes:

[0151] S91: Obtain the fuzzy membership function of the target prediction variable, and take the fuzzy membership function of the target prediction variable as the target prediction function; obtain the fuzzy membership of each target prediction variable according to the target prediction function; calculate the fuzzy matching coefficient between any one target prediction variable and any other target prediction variable according to the fuzzy membership of the target prediction variable.

[0152] Specifically, the target prediction variable inherits the main characteristics of the initial prediction variable. In this way, after determining the target prediction function, the fuzzy matching coefficient between any one target prediction variable and any other target prediction variable can be re-determined.

[0153] S92: Determine the weight coefficient of any one of the target prediction variables according to the fuzzy matching coefficients between any one of the target prediction variables and all other target prediction variables.

[0154] S93. Calculate the sum W1 of the products of the fuzzy membership degrees of the respective target prediction variables obtained in step S91 and the weight coefficient of any one of the target prediction variables obtained in step S92; and calculate the sum W2 of the weight coefficients of all target prediction variables.

[0155] S94. Take the ratio of the sum W1 of the products of the fuzzy membership degrees of the respective target prediction variables and their corresponding weight coefficients to the sum W2 of the weight coefficients of all target prediction variables as the metallogenic favorability of the target ore prospecting area. Specifically, calculate the metallogenic favorability according to the following formula:

[0156]

[0157] where f is the metallogenic favorability of the target ore prospecting area, μ A (x1), μ A (x2),......, μ A (x p ) are the fuzzy membership degrees of the respective target prediction variables, a1, a2,......, a p are the weight coefficients of the respective target prediction variables, and p is the number of target prediction variables.

[0158] In step S90, for the specific calculation processes of the fuzzy membership degree, fuzzy matching coefficient, and weight coefficient of the target prediction variables, refer to the foregoing steps S50 - S70.

[0159] It should be noted that in this solution, after determining the target prediction variables, the reason for determining the weight coefficients of the target prediction variables again is as follows: The weight coefficients assigned to the prediction variables in the foregoing example are determined according to the matching relationships between all the prediction variables participating in the calculation, which reflects the mutual relationships between all the prediction variables; however, the weight coefficients of the finally obtained target prediction variables (p) should still be determined according to the matching relationships between these p prediction variables, so that the best model can be obtained.

[0160] In other words, through two calculations of the weight coefficients in this solution, the parameters of the target prediction variables used to calculate the metallogenic favorability can be determined.

[0161] In this embodiment, after determining the metallogenic favorability, the types of various prospecting districts in the target prospecting area can be divided according to the metallogenic favorability of the target prospecting area (or the magnitude of the metallogenic favorability), so as to further guide the prospecting process.

[0162] For example, within the target prospecting area, for all grid cells, the metallogenic favorability is calculated using the above formula, and based on the magnitude of the metallogenic favorability, the grid cells with a metallogenic favorability greater than 0.8 are classified as Class A prospecting districts; those with a metallogenic favorability less than or equal to 0.8 and greater than 0.65 are classified as Class B prospecting districts; and those with a metallogenic favorability less than or equal to 0.65 and greater than 0.5 are classified as Class C prospecting districts.

[0163] It should be particularly noted that there is no strict regulation for the threshold of the metallogenic favorability used to classify prospecting districts. Generally, it can be determined artificially, or the metallogenic favorabilities can be sorted, and the threshold for classifying prospecting districts can be determined according to the inflection points that appear.

[0164] Thus, using the prospecting prediction method based on the fuzzy feature analysis of geoscience big data in the above example, on the one hand, when dividing grid cells, the characteristics of the data scale and the target ore deposit type are comprehensively considered, making the prediction results have high accuracy while ensuring the prediction efficiency; on the other hand, the fuzzy membership function in fuzzy mathematics is introduced into the feature analysis to replace logical variables, and thus the geoscience big data can be fully utilized, avoiding the loss of detailed information, improving the accuracy of the prediction results while maintaining the advantages of traditional feature analysis.

[0165] The embodiment of the present invention also provides a device corresponding to the prospecting prediction method based on the fuzzy feature analysis of geoscience big data, such as Figure 4 shown in the structural schematic diagram of a potential resource volume estimation device in an embodiment of the present invention, as Figure 5 shown, the prospecting prediction device 100 based on the fuzzy feature analysis of geoscience big data may include:

[0166] A construction unit 110, configured to establish a geoscience big data spatial database for the target prospecting area, where the geoscience big data spatial database includes various types of data materials;

[0167] A division unit 120, configured to divide the target prospecting area to generate a plurality of grid cells;

[0168] A first selection unit 130, configured to select a plurality of model cells from the plurality of grid cells, where the model cells are grid cells with a relatively high exploration degree and where one or more ore occurrences have been discovered; and to select initial prediction variables from the geoscience big data spatial databases corresponding to the respective model cells, where the initial prediction variables are at least two of the various types of data materials;

[0169] A processing unit 140 is configured to respectively determine the fuzzy membership functions of any one initial prediction variable in each model unit; obtain the fuzzy membership degrees of the initial prediction variables of the model unit according to the fuzzy membership functions, calculate the fuzzy matching coefficients between any initial prediction variable and any other initial prediction variable according to the fuzzy membership degrees of the initial prediction variables of the model unit, and determine the weight coefficients of each initial prediction variable according to the fuzzy matching coefficients between any one initial prediction variable and all other initial prediction variables.

[0170] A second selection unit 150 is configured to select a target prediction variable according to the weight coefficients of all initial prediction variables.

[0171] A determination unit 160 is configured to determine the metallogenic favorability of all grid cells in the target prospecting area according to the fuzzy membership function and the weight coefficients of the target prediction variable.

[0172] Wherein, the specific working processes and principles of the construction unit 110, the division unit 120, the first selection unit 130, the processing unit 140, the second selection unit 150, and the determination unit 160 can refer to the relevant descriptions of the foregoing examples.

[0173] It can be understood that the division of the above units is only a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or physically separated. In addition, the above units can be implemented in the form of a processor calling software.

[0174] It should be noted that the so-called "one embodiment" or "embodiment" in the present invention refers to specific features, structures or characteristics that can be included in at least one implementation manner of the present invention. And in the description of the present invention, terms such as "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with terms such as "first" and "second" may explicitly or implicitly include one or more of such features. Moreover, terms such as "first" and "second" are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or indicate importance. It can be understood that such terms can be interchanged under appropriate circumstances, so that the embodiments of the present invention described herein can be implemented in an order other than that illustrated or described.

[0175] Although the embodiments of the present invention are disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims

1. A prospecting prediction method based on the fuzzy feature analysis of geological big data, characterized in that, Comprising: Establishing a geoscience big data spatial database for the target prospecting area; Dividing the target prospecting area to generate multiple grid cells; Selecting multiple model cells from the multiple grid cells, where the model cells are grid cells with a higher exploration degree and where one or more ore occurrences have been discovered; Selecting initial prediction variables from the geoscience big data spatial databases corresponding to the respective model cells, where the initial prediction variables are at least two of multiple different types of data materials; Respectively determining the fuzzy membership function of any one of the initial prediction variables in each model cell; Obtaining the fuzzy membership of the initial prediction variables of the model cell according to the fuzzy membership function; calculating the fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership of the initial prediction variables of the model cell; Determining the weight coefficient of each initial prediction variable according to the fuzzy matching coefficient; Selecting target prediction variables according to the weight coefficients of all the initial prediction variables; Obtaining the metallogenic favorability of all the grid cells in the target prospecting area according to the fuzzy membership function and weight coefficients of the target prediction variables.

2. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 1, wherein Before dividing the target prospecting area, it further includes: establishing a geoscience big data spatial database for the target prospecting area, and the steps are: Obtaining multiple different types of data materials of the target prospecting area, where the data scales or precisions of the respective data materials are the same, and include at least two of geology, geophysics, geochemistry, remote sensing, and ore occurrences; Based on the same coordinate system, performing spatial coordinate transformation processing and data format conversion processing on various types of data materials to generate the geoscience big data spatial database.

3. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 1, wherein Calculating the area of the grid cell according to the following formula: where d is the area of the grid cell, D is the preset data scale or precision, and β is the influencing factor.

4. The ore prospecting prediction method based on geoscience big data fuzzy feature analysis according to claim 3, characterized in that When the initial prediction variable is a discrete variable, the following formula is used to determine the fuzzy membership function of the initial prediction variable: where, μ A (x ij ) is the fuzzy membership degree that the j-th predictive variable of the i-th model unit belongs to the fuzzy set A, x ij is the value of the j-th predictive variable of the i-th model unit, m is the number of initial predictive variables, and n is the number of model units; When the initial prediction variable is a continuous variable, the following formula is used to determine the fuzzy membership function of the initial prediction variable: where a and b are constants, and a < b, and a and b are determined based on the cumulative distribution function curve of each initial prediction variable.

5. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 4, wherein, Calculating the fuzzy matching coefficient according to the following formula: where r jk is the fuzzy matching coefficient between the k-th initial prediction variable of the j-th initial prediction variable, and μ A (x ij ) is the fuzzy membership degree of the j-th initial prediction variable of the i-th model unit, and μ A (x ik ) is the fuzzy membership degree of the k-th initial prediction variable of the i-th model unit, m is the number of initial prediction variables, and n is the number of model units.

6. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 5, wherein Determining the weight coefficient of each initial prediction variable according to the following formula: Among them, a j is the weight coefficient of the j-th initial prediction variable.

7. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 1, wherein The step of selecting target prediction variables according to the weight coefficients of all the initial prediction variables includes: Based on the weight coefficients of all the initial prediction variables, drawing a weight coefficient change curve in descending order, and according to the weight coefficient drop turning point on the weight coefficient change curve, taking all the initial prediction variables before the drop turning point as the selected target prediction variables.

8. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 6, characterized in that The calculation formula of the metallogenic favorability is: Among them, f is the metallogenic favorability of the target ore prospecting area, μ A (x1), μ A (x2),......, μ A (x p ) are the fuzzy membership degrees of each target prediction variable, a1, a2,......, a p are the weight coefficients of each target prediction variable, and p is the number of target prediction variables.

9. The prospecting prediction method based on the fuzzy feature analysis of geological big data according to claim 1, wherein, It further includes: Dividing the types of each ore prospecting area in the target prospecting area according to the metallogenic favorability of the target prospecting area.

10. A prospecting prediction device based on the fuzzy feature analysis of geological big data, characterized in that, Comprising: A construction unit for establishing a geoscience big data spatial database for the target prospecting area, where the geoscience big data spatial database includes various different types of data materials; A dividing unit, configured to divide the target prospecting area to generate a plurality of grid units; A first selection unit, configured to select a plurality of model units from the plurality of grid units, where the model units are grid units with a relatively high exploration degree and where one or more ore occurrences have been discovered; and to select initial prediction variables from the geo-big data spatial databases corresponding to the respective model units, where the initial prediction variables are at least two of a plurality of different types of data materials; A processing unit, configured to respectively determine the fuzzy membership function of any one of the initial prediction variables in each model unit; and to obtain the fuzzy membership degree of the initial prediction variables of the model unit according to the fuzzy membership function, and to calculate the fuzzy matching coefficient between any initial prediction variable and any other initial prediction variable according to the fuzzy membership degree of the initial prediction variables of the model unit, and to determine the weight coefficient of each initial prediction variable according to the fuzzy matching coefficient; A second selection unit, configured to select target prediction variables according to the weight coefficients of all the initial prediction variables; A determination unit, configured to calculate the metallogenic favorability of all the grid units in the target prospecting area according to the fuzzy membership function and the weight coefficient of the target prediction variable.

Images