Background anomaly recognition method based on center logarithm ratio and fractal theory

Abstract

The invention relates to the technical field of exploration geochemistry, in particular to a background anomaly recognition method based on a center logarithm ratio and a fractal theory, which is mainly characterized by comprising the following steps: performing component data theory CLR conversion on chemical components, counting the number N of samples greater than or equal to a specific # imgabs0 # value, calculating the corresponding logarithm # imgabs1 #, and calculating the number N of samples greater than or equal to the specific # imgabs0 # value; carrying out scatter projection on the center logarithm ratio # imgabs2 # of all qualified sample to-be-researched elements m and the corresponding # imgabs3 #, and carrying out least square fitting on the projected scatter diagram; the method comprises the following steps: processing original concentration data through a logarithmic ratio of a component data center; background and abnormal threshold values are determined through a fractal theory, a center logarithm ratio # imgabs4 # obtained after conversion of a component data theory obeys Gaussian distribution, and a statistical analysis result can better reflect element distribution characteristics compared with original concentration.

Description

Technical Field

[0001] The invention relates to the technical field of exploration geochemistry, and in particular to a background anomaly recognition method based on central logarithmic ratio and fractal theory. Background Art

[0002] Identifying background and anomaly thresholds is an important part of geochemical exploration. Accurately identifying background and anomaly thresholds can effectively understand the distribution characteristics of elements and provide decision support for mineral exploration. Traditionally, the thresholds of element background and anomaly are based on the assumption that the original concentration data follows a Gaussian distribution, and are achieved through mathematical statistics, such as iterative elimination of deviations that are higher than the average value plus or minus several times (generally 1, 2, 2.5, 3 times). However, the original concentration data is component data, and the closed operation causes the element concentration to not be an independent random variable, which does not meet the prerequisite requirements of some mathematical statistics methods. Direct statistical analysis of the original concentration data may produce erroneous conclusions. Compositional data theory can solve this type of problem well, but the ALR transformation of P-dimensional original concentration data can only produce P-1-dimensional effective variables, and the ALR transformation does not retain distance, while the ILR transformation can only generate an effective variable consistent with the original information, which is difficult to interpret. CLR changes can produce effective variables with the same dimension as the original data and retain distance.

[0003] The distribution of elements in the earth does not conform to the Gaussian distribution. The material composition of the earth is affected by multiple geological processes in its evolutionary history, and the concentration of each element is the result of the superposition of multiple geological processes. Its distribution has nonlinear characteristics and in most cases obeys fractal statistics rather than Gaussian distribution. The background and anomaly thresholds identified by traditional statistical analysis methods may not necessarily conform to geological facts.

[0004] Therefore, a background anomaly recognition method based on central logarithmic ratio and fractal theory is proposed. Summary of the invention

[0005] The present invention aims to solve the above problems and provides a background anomaly recognition method based on central logarithmic ratio and fractal theory.

[0006] The present invention achieves the above-mentioned purpose through the following technical scheme, a background anomaly identification method based on central logarithmic ratio and fractal theory, comprising: performing theoretical CLR conversion of component data on chemical components, performing fractal processing on CLR conversion data, determining the background anomaly boundary through the least squares fitting function, and determining the background and anomaly thresholds through the data close to the third quartile (75%) on the QQ plot.

[0007] The specific steps of the present invention are as follows:

[0008] S1. Clean the data and remove samples with element concentration values ​​below the detection limit;

[0009] S2. Calculate the concentration of each component of qualified samples The corresponding central logarithmic ratio , where n is the number of components, m is the component to be studied, 1 ≤ i ≤ n;

[0010] S3, statistics greater than or equal to a specific The number of samples N of the value, calculate its corresponding logarithm ;

[0011] S4, the central logarithm ratio of all qualified samples to be studied component m and its corresponding Perform scatter plotting;

[0012] S5. Perform least squares fitting on the scatter plot obtained in S4. The number of fitted lines is determined by the specific shape of the scatter plot, and the fitting criterion is the coefficient of determination R of each line. 2 Greater than or equal to 0.75, that is:

[0013]

[0014] A and B are obtained by least squares fitting, and the slopes A of different straight lines are different;

[0015] S6. Adjacent fitted straight lines will form intersections, and record the intersections they represent. value;

[0016] S7. Drawing QQ graph, determine the intersection points The first value close to the 3rd quartile The original concentration C m The background and abnormal thresholds are the other intersection points, and the other intersection points are set as abnormal thresholds of different levels according to the size of the original concentration values ​​they represent. "Close to the third quantile" means that the distance between its quantile value and the third quantile is the shortest.

[0017] Furthermore, the original concentration data are transformed into central logarithmic ratio (CLR) according to the compositional data theory, and the concentration and sample number are identified as abnormal according to the fractal theory. The denominator in the CLR transformation step can be the content of all elements, or the content of the element to be studied and the constraint constant 1 minus the content of the element to be studied m. The identified background and anomalies have nonlinear distribution properties.

[0018] Compared with the existing technical methods, the present invention can achieve the following beneficial effects:

[0019] The present invention uses the central logarithmic ratio of the component data theoretical processing The data obeys Gaussian distribution, and its statistical results can better reflect the distribution characteristics of elements than the original concentration; the central logarithmic ratio processed by composition data theory The dimension of the original concentration data is consistent with that of the original concentration data. The information dimension of the result is better than that of the additive logarithmic ratio and the interpretability is better than that of the equidistant logarithmic ratio. The fractal study of the logarithm of the sample number can accurately identify the background formed by non-metallogenic processes and the anomalies formed by metallogenic processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] FIG1 is a flow chart of the method of the present invention;

[0021] FIG. 2 is an embodiment of the present invention. Relationship diagram;

[0022] Figure 3 shows QQ plot of the data. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] A background anomaly recognition method based on central logarithmic ratio and fractal theory comprises the following steps:

[0025] Step 1: Clean the data and remove samples with element concentration values ​​below the detection limit to ensure that all sample data are qualified;

[0026] Step 2: Calculate the central logarithm ratio of all qualified samples corresponding to the proposed research component m ;

[0027] Step 3: Count the values ​​greater than or equal to a specific value The number of samples N of the value, calculate its corresponding logarithm ;

[0028] Step 4: Calculate the central logarithm ratio of all qualified samples of the proposed research component m and its corresponding Perform scatter plotting;

[0029] Step 5: Perform least squares fitting on the scatter plot obtained in step 4. The number of fitted lines is determined by the specific shape of the scatter plot, and the fitting standard is the coefficient of determination R of each fitted line. 2 Greater than or equal to 0.75, that is:

[0030]

[0031] In the formula, A is the fractal index and B is the constant term, both of which are obtained by least squares fitting, and the slope A of different straight lines is different;

[0032] Step 6: Adjacent fitted lines will form intersections, and record their representations. value;

[0033] Step 7: Drawing QQ graph, determine the intersection points The first value close to the 3rd quartile The original concentration C m is the background and abnormal threshold, and the other intersection points are set as abnormal thresholds of different levels according to the concentration values ​​they represent.

[0034] In addition, it should be understood that although the present specification is described in terms of implementation modes, not every implementation mode contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Ordinary technicians in this field should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by ordinary technicians in this field.

Claims

1. A background anomaly recognition method based on central logarithmic ratio and fractal theory, characterized by: The chemical composition data were transformed into the theoretical central logarithm ratio (CLR) of the composition data. The CLR transformed data and sample quantity were fractal processed. The background anomaly boundary was determined by the least squares fitting function. The background and anomaly thresholds were determined by the data close to the third quartile (75%) on the QQ plot.

2. The background anomaly identification method according to claim 1, characterized in that: The specific steps include: S1. Clean the data and remove samples with element concentration values ​​below the detection limit; S2. Calculate the concentration of each component of qualified samples The corresponding central logarithmic ratio , where n is the number of components, m is the component to be studied, 1 ≤ i ≤ n; S3, statistics greater than or equal to a specific The number of samples N of the value, calculate its corresponding logarithm ; S4, the central logarithm ratio of all qualified samples to be studied component m and its corresponding Perform scatter plotting; S5. Perform least squares fitting on the scatter plot obtained in S4. The number of fitted lines is determined by the specific shape of the scatter plot, and the fitting criterion is the coefficient of determination R of each line. 2 Greater than or equal to 0.75, that is: , where A and B are obtained by least squares fitting, and the slopes A of different straight lines are different; S6. Adjacent fitted straight lines will form intersections, and record the intersections they represent. value; S7. Drawing QQ graph, determine the intersection points The first value close to the 3rd quantile The original concentration C m The background and abnormal thresholds are set, and the other intersection points are set as abnormal thresholds of different levels according to the size of the original concentration values ​​they represent. "Close to the third quantile" means that the distance between its quantile value and the third quantile is the shortest.

3. The background anomaly recognition method based on central logarithmic ratio and fractal theory according to any one of claims 1 or 2, characterized in that: The original concentration data is converted into central logarithmic ratio (CLR) based on the compositional data theory. The concentration and sample number are identified as abnormal based on the fractal theory. The identified background and abnormality have nonlinear distribution properties.

4. The background anomaly recognition method based on central logarithmic ratio and fractal theory according to claim 2 is characterized in that: In step S5, A and B are obtained by least square fitting, and the slopes A of different straight lines are different.

Images