A method for analyzing the test results of aquatic sediment samples

Abstract

This invention relates to the field of data processing technology, and more particularly to a method for analyzing the test results of aquatic sediment samples. This method constructs a three-dimensional clustering space by combining the elemental concentration of any element in each aquatic sediment sample with the sampling point location of each sample. When performing K-means clustering on the data points in the three-dimensional clustering space, the clustering distance between the data points and the centroid is corrected based on elemental concentration similarity to obtain a corrected distance. The corrected distance is then further corrected based on the elemental concentration differences between the data points and their neighboring data points, as well as the distribution characteristics of data points with similar elemental concentrations, to obtain a final corrected distance. K-means clustering is then performed on all data points based on all final corrected distances. The test results of the aquatic sediment samples are analyzed based on the clustering results, thus improving the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method for analyzing the test results of aquatic sediment samples. Background Technology

[0002] Analysis of laboratory results from stream sediment samples is typically a crucial step in geochemistry, hydrogeology, or environmental geology. By analyzing and processing the data, we can understand the content, enrichment dispersion, combination, and spatial distribution characteristics of the data. The aim is to identify the elemental distribution in the collected sediment samples, which is beneficial for discovering mineralized geological bodies, mineralization bodies, and ore bodies, thus guiding geological prospecting. Stream sediment samples are usually collected uniformly on 1:50,000 topographic maps. Current techniques typically employ the K-means clustering algorithm to cluster the elemental concentration data in the laboratory results of stream sediment samples. Through clustering, we can grasp the similarity structure between samples from different sampling points, reveal the distribution of geochemical types in the sediments, separate potential anomalous enrichment areas, and assist in the detection of mineralization areas.

[0003] However, traditional methods of K-means clustering, which rely solely on elemental concentration data, neglect the spatial continuity or proximity of samples. This can lead to spatially fragmented clustering, resulting in decreased accuracy in representing mineral distribution. Furthermore, stream sediment samples often exhibit additional enrichment in certain samples, leading to relatively high or anomalous concentrations of individual elements. For example, localized heavy metal pollution might cause a high concentration of a particular element, but this concentration does not reflect the true mineral content. K-means clustering can be skewed by such extreme anomalies, affecting the classification results.

[0004] Therefore, improving the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method for analyzing the test results of aquatic sediment samples to solve the problem of how to improve the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm.

[0006] This invention provides a method for analyzing the laboratory results of aquatic sediment samples, the method comprising the following steps:

[0007] At least two stream sediment samples are obtained. For any element, the element concentration of the element is extracted in each stream sediment sample. A three-dimensional clustering space is constructed by combining the location of the sampling point of each stream sediment sample in a two-dimensional topographic map and the element concentration of the element in each stream sediment sample.

[0008] When performing K-means clustering on data points in a three-dimensional clustering space, for any iteration process, the centroid of the iteration process is obtained. Based on the element concentration similarity between any data point and any centroid, the clustering distance between any data point and any centroid is first weighted and corrected to obtain the corrected distance.

[0009] Based on the element concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar element concentrations to any data point, a second weighted correction is applied to the correction distance to obtain the final corrected distance between any data point and any centroid.

[0010] The final corrected distance between each data point and each centroid is obtained in each iteration. Based on all the final corrected distances, K-means clustering is performed on all data points in the three-dimensional clustering space to obtain the clustering result of any element. Based on the clustering results of all elements, the test results of all aquatic sediment samples are analyzed.

[0011] Preferably, the construction of a three-dimensional clustering space by combining the location of the sampling point of each stream sediment sample on the two-dimensional topographic map and the elemental concentration of any element in each stream sediment sample includes:

[0012] Based on the coordinates of the sampling point of each stream sediment sample in the two-dimensional topographic map, and the element concentration of any element in each stream sediment sample, the element concentration of any element in each stream sediment sample is mapped to the same three-dimensional cluster space. The x-axis of the three-dimensional cluster space represents the horizontal coordinate of the corresponding sampling point of each stream sediment sample in the two-dimensional topographic map, the y-axis represents the vertical coordinate of the corresponding sampling point of each stream sediment sample in the two-dimensional topographic map, and the z-axis represents the element concentration of any element in each stream sediment sample.

[0013] Preferably, the step of performing a first weighted correction on the clustering distance between any data point and any centroid based on the elemental concentration similarity between any data point and any centroid to obtain the corrected distance includes:

[0014] Calculate the reciprocal of the absolute value of the difference between the element concentration of any data point and the element concentration of any centroid to obtain the first element concentration similarity index;

[0015] In the three-dimensional clustering space, the standard deviation of the element concentration of all data points in the neighborhood of any data point is calculated and denoted as the first standard deviation. The standard deviation of the element concentration of all data points in the neighborhood of any centroid is calculated and denoted as the second standard deviation. The reciprocal of the absolute value of the difference between the first standard deviation and the second standard deviation is calculated to obtain the second element concentration similarity index.

[0016] Calculate the product between the first element concentration similarity index and the second element concentration similarity index to obtain the element concentration similarity between any data point and any centroid;

[0017] Based on the similarity of element concentrations, the corrected distance between any data point and any centroid is obtained.

[0018] Preferably, obtaining the corrected distance between any data point and any centroid based on the element concentration similarity includes:

[0019] The position weight is obtained by calculating the difference between constant 1 and the similarity of the element concentration, and the concentration weight is obtained by calculating the sum between constant 1 and the similarity of the element concentration.

[0020] In the three-dimensional clustering space, the square of the distance between any data point and any centroid on the x-axis is calculated to obtain the x-axis squared term; the square of the distance between any data point and any centroid on the y-axis is calculated to obtain the y-axis squared term; and the square of the distance between any data point and any centroid on the z-axis is calculated to obtain the z-axis squared term.

[0021] The position weights are used as the weights of the x-axis squared term and the y-axis squared term, respectively, and the concentration weights are used as the weights of the z-axis squared term. The x-axis squared term, the y-axis squared term, and the z-axis squared term are weighted and summed to obtain a weighted summation result. The arithmetic square root of the weighted summation result is used as the corrected distance between any data point and any centroid.

[0022] Preferably, the step of performing a second weighted correction on the correction distance based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to any data point, to obtain the final corrected distance between any data point and any centroid, includes:

[0023] The pollution level of any data point is determined based on the difference in element concentration between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar element concentrations to any data point.

[0024] If the degree of contamination is greater than or equal to the preset contamination threshold, then according to the degree of contamination of any data point, the correction distance between any data point and any centroid is subject to a second weighted correction to obtain the final correction distance between any data point and any centroid.

[0025] If the degree of contamination is less than a preset contamination threshold, then the corrected distance between any data point and any centroid is taken as the final corrected distance.

[0026] Preferably, the step of determining the contamination level of any data point based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to the data point, includes:

[0027] In the three-dimensional clustering space, similar data points with element concentrations similar to any data point are obtained. If the number of similar data points is 0, the contamination level of any data point is set to 1.

[0028] If the number of similar data points is not zero, calculate the Euclidean distance between each similar data point and any data point, and linearly normalize the average of all Euclidean distances to obtain the first distribution characteristic value of any data point.

[0029] In the three-dimensional clustering space, all data points in the neighborhood of any data point are recorded as neighborhood data points. The absolute value of the difference between the element concentration of each neighborhood data point and the element concentration of any data point is calculated. The average value of all the absolute values ​​of the difference is linearly normalized to obtain the second distribution characteristic value of any data point.

[0030] The contamination level of any data point is obtained by calculating the product between the first distribution characteristic value and the second distribution characteristic value.

[0031] Preferably, the step of performing a second weighted correction on the corrected distance between any data point and any centroid based on the degree of contamination of any data point to obtain the final corrected distance between any data point and any centroid includes:

[0032] Subtract the contamination level of any data point from the constant 1 to obtain the correction weight. Based on the correction weight, the correction distance between any data point and any centroid is weighted to obtain the final correction distance between any data point and any centroid.

[0033] Preferably, obtaining similar data points with element concentrations similar to any one of the data points in the three-dimensional clustering space includes:

[0034] In the three-dimensional clustering space, for any other data point besides the aforementioned data point, if the absolute value of the difference between the element concentration of the aforementioned data point and the element concentration of the other data point is less than or equal to a preset difference threshold, then the other data point is recorded as a similar data point with similar element concentration to the aforementioned data point.

[0035] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0036] This invention obtains at least two stream sediment samples. For any given element, the elemental concentration of that element is extracted from each stream sediment sample. Combining the location of the sampling point of each stream sediment sample on a two-dimensional topographic map and the elemental concentration of that element in each stream sediment sample, a three-dimensional clustering space is constructed. When performing K-means clustering on the data points in the three-dimensional clustering space, for any iteration, the centroid of that iteration is obtained. Based on the elemental concentration similarity between any data point and any centroid, a first weighted correction is applied to the clustering distance between any data point and any centroid. The corrected distance is obtained; based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to any data point, a second weighted correction is applied to the corrected distance to obtain the final corrected distance between any data point and any centroid; the final corrected distance between each data point and each centroid is obtained in each iteration; based on all the final corrected distances, K-means clustering is performed on all data points in the three-dimensional clustering space to obtain the clustering result of any element; based on the clustering results of all elements, the test results of all aquatic sediment samples are analyzed. This study constructs a three-dimensional clustering space by combining the location of sampling points of each stream sediment sample on a two-dimensional topographic map with the elemental concentration of any element in each sample. The K-means clustering algorithm is then used to perform three-dimensional clustering of the data points within this space. This approach integrates the characteristics of elemental concentration and geographic distribution, improving the accuracy of elemental clustering in ore bodies and enhancing the accuracy of ore body location through clustering results. This, in turn, improves the accuracy of analyzing the laboratory results of stream sediment samples. When performing K-means clustering on data points, the clustering distance between each data point and the centroid is corrected based on elemental concentration similarity and spatial distribution characteristics. This reduces the probability of spherical clusters appearing in high-dimensional clustering by the K-means clustering algorithm. At the same time, it reduces the interference of data points with abnormal elemental concentrations caused by heavy metal pollution on the clustering results of K-means clustering, thereby improving the accuracy of the K-means clustering algorithm in high-dimensional clustering and improving the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm. Attached Figure Description

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 This is a flowchart of a method for analyzing the test results of aquatic sediment samples according to Embodiment 1 of the present invention. Detailed Implementation

[0039] Embodiments of this disclosure are described in detail below, with examples of these embodiments illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting it.

[0040] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.

[0041] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0042] See also Figure 1 This is a flowchart of a method for analyzing the test results of aquatic sediment samples according to Embodiment 1 of the present invention. Figure 1 As shown, the method may include:

[0043] Step S101: Obtain at least two aquatic sediment samples. For any element, extract the element concentration of the element in each aquatic sediment sample. Combine the location of the sampling point of each aquatic sediment sample on the two-dimensional topographic map and the element concentration of the element in each aquatic sediment sample to construct a three-dimensional clustering space.

[0044] On a 2D topographic map with a scale of 1:50000, according to The basic sampling unit evenly distributes sampling points to obtain at least two sampling points, and collects one stream sediment sample at each sampling point. Sampling locations are chosen in areas where the water flow slows down and where various particle sizes of the stream sediment easily converge, such as where the water flow changes from rapid to slow, where the water flow stagnates, on the inside of a river bend, or behind large rocks, ensuring that the particle size ratios in the stream sediment samples collected from each sampling point are in a naturally mixed state. Each stream sediment sample undergoes drying, sieving, mixing, and packaging before being sent to the laboratory for analysis. In the laboratory, methods such as tetraacid digestion, aqua regia digestion, and nitric acid-hydrofluoric acid method are used to extract the concentrations of different mineral elements (Au, Ag, Mo, Sn, Co, Cr, Cu, Mn, Ni, Pb, Ti, V, W, Zn, As, Sb, Bi, and Hg) from each stream sediment sample. The extraction of the concentrations of each mineral element from each stream sediment sample is a current technique and will not be elaborated upon here.

[0045] Traditionally, the K-means clustering algorithm is used to cluster the concentration of ore elements in stream sediment samples to grasp the similarity structure between stream sediment samples from different sampling points, reveal the geochemical type distribution of stream sediments, separate possible anomalous enrichment areas, and assist in the detection of mineralization areas.

[0046] However, traditional methods only perform K-means clustering based on the concentration of elements in each ore body, ignoring the continuity or proximity of samples in geographic space. This may result in spatially fragmented clustering, leading to a decrease in the accuracy of the clustering results in representing the distribution of ore bodies. Therefore, in this embodiment of the invention, K-means clustering is performed on the concentration of each ore body element in each stream sediment sample, taking into account the geographic distribution of each stream sediment sample. This allows the clustering results to integrate the concentration of each ore body element and the geographic spatial distribution characteristics of each stream sediment sample, thereby improving the accuracy of the representation of ore body distribution.

[0047] In this embodiment of the invention, taking element Au as an example, firstly, the concentration of element Au in each aquatic sediment sample is obtained; then, all concentrations corresponding to element Au are linearly normalized to obtain the elemental concentration of element Au in each aquatic sediment sample. Simultaneously, the longitude of each aquatic sediment sample on the two-dimensional topographic map is linearly normalized to obtain the abscissa of each aquatic sediment sample, and the latitude of each aquatic sediment sample on the two-dimensional topographic map is linearly normalized to obtain the ordinate of each aquatic sediment sample. Linear normalization is a prior art technique and will not be elaborated here. Finally, the elemental concentration of element Au in each aquatic sediment sample is mapped to the same three-dimensional space, denoted as the three-dimensional clustering space. In this three-dimensional clustering space, the x-axis represents the abscissa of the sampling point corresponding to each aquatic sediment sample, denoted as the abscissa axis; the y-axis represents the ordinate of the sampling point corresponding to each aquatic sediment sample, denoted as the ordinate axis; and the z-axis represents the elemental concentration of element Au in each aquatic sediment sample, denoted as the concentration axis.

[0048] Thus, by combining the position of each stream sediment sample on the two-dimensional topographic map with the elemental concentration of Au in each stream sediment sample, a three-dimensional clustering space is obtained, which is used to perform K-means clustering of the elemental concentration of Au in each stream sediment sample in three-dimensional space.

[0049] Step S102: When performing K-means clustering on data points in the three-dimensional clustering space, for any iteration process, obtain the centroid of the iteration process. Based on the element concentration similarity between any data point and any centroid, perform a first weighted correction on the clustering distance between any data point and any centroid to obtain the corrected distance.

[0050] Because the K-means clustering algorithm relies on the distance between data points and centroids, it is more likely to produce spherical clusters when clustering high-dimensional data. However, in reality, the elemental concentration of ore bodies in stream sediments is affected by geological genesis or environmental factors (such as water flow velocity and mineralization zones), and the resulting clusters are often non-spherical. Therefore, the clustering accuracy of the K-means clustering algorithm will decrease when clustering high-dimensional data.

[0051] Therefore, in this embodiment of the invention, when performing K-means clustering on data points in a three-dimensional clustering space, taking the i-th iteration as an example, the centroid of the i-th iteration is obtained. Based on the similarity of element concentration between the data points and the centroid, different weights are assigned to the distance between the data points and the centroid on each axis in the three-dimensional clustering space: the more similar the element concentration between the data points and the centroid, the more likely the data points and the sampling points corresponding to the centroid belong to the same ore body, and the more likely the data points should belong to the cluster where the centroid is located. Therefore, a larger weight should be assigned to the distance between the data points and the centroid on the concentration axis. Conversely, a smaller weight is assigned to the distance between a data point and its centroid on the concentration axis. On the other hand, the more similar the elemental concentration distribution in the neighborhood of a data point is to the elemental concentration distribution in the neighborhood of the centroid, the more likely the data point and the corresponding sampling point belong to the same ore body. In this case, a larger weight should be assigned to the distance between the data point and its centroid on the concentration axis, and vice versa. This allows data points that are relatively far apart on a two-dimensional topographic map to be more easily clustered based on elemental concentration, reducing the probability of spherical clusters. Furthermore, based on the weights of the distances between data points and their centroids on each axis in the three-dimensional clustering space, a corrected distance between the data point and its centroid is obtained to improve the clustering accuracy of the K-means clustering algorithm when processing high-dimensional data.

[0052] Taking the q-th data point and the c-th centroid in a 3D clustering space as an example, the steps to obtain the corrected distance between the q-th data point and the c-th centroid are as follows:

[0053] (1) Obtain the element concentration similarity between the qth data point and the cth centroid.

[0054] Specifically, the reciprocal of the absolute value of the difference between the element concentration of the q-th data point and the element concentration of the c-th centroid is calculated to obtain the first element concentration similarity index;

[0055] In the three-dimensional clustering space, the standard deviation of the element concentration of all data points in the 26 neighborhoods of the q-th data point is calculated and denoted as the first standard deviation. The standard deviation of the element concentration of all data points in the 26 neighborhoods of the c-th centroid is calculated and denoted as the second standard deviation. There is no restriction here. The implementer can set the neighborhood range according to the specific scenario. The reciprocal of the absolute value of the difference between the first standard deviation and the second standard deviation is calculated to obtain the second element concentration similarity index.

[0056] Calculate the product between the first element concentration similarity index and the second element concentration similarity index to obtain the element concentration similarity between the q-th data point and the c-th centroid.

[0057] In one embodiment, the formula for calculating the similarity of elemental concentration between the q-th data point and the c-th centroid is:

[0058]

[0059] in, This indicates the similarity in element concentration between the q-th data point and the c-th centroid. This represents the elemental concentration at the c-th centroid. This represents the element concentration at the q-th data point. Let represent the standard deviation of the element concentration of all data points within the 26-neighborhood of the c-th centroid. This represents the standard deviation of the element concentration of all data points within the 26-neighborhood of the q-th data point. Represents the absolute value symbol.

[0060] It should be noted that The smaller the value, the more similar the element concentrations are between the q-th data point and the c-th centroid, and thus... The larger the value, the more likely the sampling points corresponding to the q-th data point and the c-th centroid belong to the same ore body, and the more likely the q-th data point should belong to the cluster containing the c-th centroid. The greater the weight assigned to the distance between the q-th data point and the c-th centroid on the concentration axis, the higher the value. The standard deviation can characterize the uniformity of the data distribution. The smaller the value, the more similar the element concentration distribution in the neighborhood of the q-th data point is to the element concentration distribution in the neighborhood of the c-th centroid, and thus... The larger the value, the more likely the sampling points corresponding to the q-th data point and the c-th centroid belong to the same ore body, the more likely the q-th data point should belong to the cluster where the c-th centroid is located, and the greater the weight assigned to the distance between the q-th data point and the c-th centroid on the concentration axis.

[0061] (2) Based on the similarity of element concentration between the qth data point and the cth centroid, perform a first weighted correction on the clustering distance between the qth data point and the cth centroid to obtain the corrected distance between the qth data point and the cth centroid.

[0062] Specifically, the difference between constant 1 and the similarity of the element concentration is calculated to obtain the position weight, and the sum between constant 1 and the similarity of the element concentration is calculated to obtain the concentration weight;

[0063] In the three-dimensional clustering space, the square of the distance between the q-th data point and the c-th centroid on the x-axis (i.e., the horizontal axis) is calculated to obtain the x-axis squared term. The square of the distance between any data point and any centroid on the y-axis (i.e., the vertical axis) is calculated to obtain the y-axis squared term. The square of the distance between any data point and any centroid on the z-axis (i.e., the concentration axis) is calculated to obtain the z-axis squared term.

[0064] The position weights are used as the weights of the x-axis squared term and the y-axis squared term, respectively, and the concentration weights are used as the weights of the z-axis squared term. The x-axis squared term, the y-axis squared term, and the z-axis squared term are weighted and summed to obtain a weighted summation result. The arithmetic square root of the weighted summation result is used as the corrected distance between the q-th data point and the c-th centroid.

[0065] In one embodiment, the formula for calculating the corrected distance between the q-th data point and the c-th centroid is:

[0066]

[0067] in, This represents the corrected distance between the q-th data point and the c-th centroid. This indicates the similarity in element concentration between the q-th data point and the c-th centroid. This represents the coordinates of the c-th centroid in the three-dimensional clustering space on the x-axis. This represents the coordinates of the q-th data point on the x-axis in the 3D clustering space. This represents the coordinate of the c-th centroid in the three-dimensional clustering space on the y-axis. This represents the y-coordinate of the q-th data point in the 3D clustering space. This represents the coordinates of the c-th centroid in the three-dimensional clustering space on the z-axis. This represents the coordinates of the q-th data point on the z-axis in the three-dimensional clustering space.

[0068] It should be noted that The larger the value, the more likely the sampling points corresponding to the q-th data point and the c-th centroid belong to the same ore body, and the more likely the q-th data point should belong to the cluster containing the c-th centroid. Therefore, the greater the weight assigned to the distance between the q-th data point and the c-th centroid on the concentration axis, the higher the probability that they belong to the same ore body. The larger the value, the greater the weight should be; to maintain weight balance, the weight of the distance between the q-th data point and the c-th centroid on the x-axis and y-axis should be reduced, i.e. .

[0069] Therefore, based on the element concentration similarity between the q-th data point and the c-th centroid, the clustering distance between the q-th data point and the c-th centroid is first weighted and corrected to obtain the corrected distance between the q-th data point and the c-th centroid. This makes it easier for data points that are relatively far apart on the two-dimensional topographic map to be clustered together based on element concentration, reducing the probability of spherical clusters and improving the clustering accuracy of K-means clustering when clustering high-dimensional data.

[0070] Step S103: Based on the element concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar element concentrations to any data point, a second weighted correction is applied to the correction distance to obtain the final corrected distance between any data point and any centroid.

[0071] Furthermore, considering that some sampled aquatic sediments may have additional enrichment, resulting in relatively high Au concentrations, and that some samples may be contaminated with heavy metals, this could also lead to relatively high Au concentrations. Since the Au concentration under heavy metal contamination is not representative of the actual ore body element data and represents anomalies, the K-means clustering algorithm can be skewed by these anomalies, affecting the clustering results. Therefore, in this embodiment of the invention, it is necessary to determine whether q data points are data points with abnormal element concentrations caused by heavy metal contamination, and to further correct the distance between the q-th data point and the c-th centroid. This reduces the possibility of the K-means clustering algorithm being skewed by anomalies, thereby improving the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm.

[0072] Although the elemental concentrations of ore-bearing elements in sediment samples from different river systems may vary, the differences in elemental concentrations between adjacent sampling points are relatively small. In contrast, heavy metal pollution is anthropogenic, resulting in significant differences in elemental concentrations between adjacent sampling points. Furthermore, heavy metal pollution is anthropogenic and smaller in scale than ore bodies. Additionally, heavy metal pollution from other industrial sites may exist on two-dimensional topographic maps, exhibiting a more dispersed distribution. Ore-bearing elements, on the other hand, are naturally formed, following a gradual extension process, resulting in specific elemental concentrations within specific regions. This means that sampling points in river sediments corresponding to data points with similar elemental concentrations tend to be more concentrated. Therefore, in this embodiment of the invention, the pollution level of the q-th data point is determined based on the elemental concentration differences between the q-th data point and its neighboring data points, as well as the spatial distribution characteristics of data points with similar elemental concentrations. This information is used to determine whether the q-th data point is a data point with abnormal elemental concentrations caused by heavy metal pollution. Specifically:

[0073] In the three-dimensional clustering space, for any other data point other than the q-th data point, if the absolute value of the difference between the element concentration of the q-th data point and the element concentration of the other data point is less than or equal to 0.1 (no restriction is imposed here, and the implementer can set it according to the specific scenario), then the other data point is recorded as a similar data point with similar element concentration to the q-th data point.

[0074] If the number of similar data points is not zero, calculate the Euclidean distance between each similar data point and the qth data point, and linearly normalize the average of all Euclidean distances to obtain the first distribution characteristic value of the qth data point.

[0075] In the three-dimensional clustering space, all data points within the 26 neighborhoods of the q-th data point are denoted as neighborhood data points. There is no restriction here. The implementer can set the neighborhood range according to the specific scenario. The absolute value of the difference between the element concentration of each neighborhood data point and the element concentration of the q-th data point is calculated. The average value of all the absolute values ​​of the difference is linearly normalized to obtain the second distribution characteristic value of the q-th data point.

[0076] The contamination level of the q-th data point is obtained by multiplying the first distribution characteristic value and the second distribution characteristic value.

[0077] In one implementation, the formula for calculating the contamination level of the q-th data point is:

[0078]

[0079] in, This indicates the degree of contamination of the q-th data point. This represents the element concentration at the q-th data point. This represents the element concentration of the j-th neighboring data point of the q-th data point. This represents the number of all neighboring data points of the q-th data point. Let represent the Euclidean distance between the q-th data point and the k-th similar data point of the q-th data point in the 3D clustering space. This represents the number of all similar data points for the q-th data point, and norm() represents the linear normalization function.

[0080] It should be noted that The larger the value, the greater the difference in element concentration between the q-th data point and its neighboring data points, and thus... The larger the value, the greater the likelihood that the q-th data point is a data point with abnormal element concentration caused by heavy metal pollution; The larger the value, the more discrete the distribution of the q-th data point is with its similar data points, and the more it matches the spatial distribution characteristics of data points with abnormal element concentrations caused by heavy metal pollution. The larger the value, the greater the likelihood that the q-th data point is a data point with abnormal element concentration caused by heavy metal pollution.

[0081] Specifically, if the number of similar data points for the q-th data point is 0, it means that the q-th data point is the heavy metal pollution point with the most independent element concentration, rather than a vein point with a large coverage area of ​​the mineralization region. Therefore, the pollution level of the q-th data point is set to 1.

[0082] because The value range is [0,1]. Data points with abnormal element concentrations caused by heavy metal pollution and normal data points will be divided into the two ends of the value range [0,1]. Therefore, the middle boundary of 0.5 is used as the threshold, that is, the preset pollution threshold is set to 0.5. There is no restriction here, and the implementer can set it according to the specific scenario. If the pollution degree of the q-th data point is less than 0.5, the q-th data point is determined to be a normal data point. At this time, the correction distance between the q-th data point and the c-th centroid is used as the final correction distance. If the pollution degree of the q-th data point is greater than or equal to 0.5, the q-th data point is determined to be a data point with abnormal element concentrations caused by heavy metal pollution. At this time, according to the pollution degree of the q-th data point, the correction distance between the q-th data point and the c-th centroid needs to be corrected by a second weighting to obtain the final correction distance between the q-th data point and the c-th centroid. This reduces the possibility that the K-means clustering algorithm is skewed by abnormal extreme values ​​and improves the accuracy of the analysis of the test results of aquatic sediment samples using the K-means clustering algorithm. Specifically:

[0083] Subtract the contamination level of the q-th data point from the constant 1 to obtain the correction weight. Based on the correction weight, the correction distance between the q-th data point and the c-th centroid is weighted to obtain the final correction distance between the q-th data point and the c-th centroid.

[0084] In one embodiment, the formula for calculating the final corrected distance between the q-th data point and the c-th centroid is:

[0085]

[0086] in, This represents the final corrected distance between the q-th data point and the c-th centroid. This represents the corrected distance between the q-th data point and the c-th centroid. This indicates the degree of contamination of the q-th data point.

[0087] It should be noted that The larger the value, the greater the likelihood that the q-th data point is a data point with anomalies in elemental concentration caused by heavy metal pollution. Therefore, a smaller weight should be assigned to the corrected distance between the q-th data point and the c-th centroid. The smaller, then The smaller the value, the smaller the final corrected distance between the q-th data point and the c-th centroid, which means the smaller the clustering distance between the q-th data point and the c-th centroid used for K-means clustering.

[0088] Step S104: Obtain the final corrected distance between each data point and each centroid during each iteration. Based on all the final corrected distances, perform K-means clustering on all data points in the three-dimensional clustering space to obtain the clustering result of any element. Based on the clustering results of all elements, analyze the test results of all aquatic sediment samples.

[0089] According to step S103, when performing K-means clustering on data points in the three-dimensional clustering space, the final corrected distance between each data point and each centroid is obtained in each iteration. Based on all the final corrected distances, K-means clustering is performed on all data points in the three-dimensional clustering space to obtain at least one cluster of element Au. The concentration of element Au in the aquatic sediment samples corresponding to data points belonging to the same cluster is similar, and the geographical distribution of the sampling points is clustered.

[0090] Similarly, by obtaining at least one cluster for each other ore body element, since the concentration of ore body elements in the stream sediment samples corresponding to data points belonging to the same cluster is similar, and the geographical distribution of the sampling points is clustered, the geochemical type distribution of stream sediments can be revealed based on the location of the sampling points in each cluster corresponding to each ore body element. Possible anomalous enrichment areas can be separated to assist in the detection of mineralization areas. For example, the more members in the same cluster, the more likely there is an ore body in the area of ​​the corresponding sampling points in that cluster, making the search for ore body distribution more accurate.

[0091] It is worth noting that the focus of this invention is to improve the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm by correcting the clustering distance between data points and centroids. Analyzing the test results of aquatic sediment samples based on clustering results is existing technology and will not be elaborated upon here.

[0092] In summary, this invention obtains at least two aquatic sediment samples. For any given element, the elemental concentration of that element is extracted from each aquatic sediment sample. Combining the location of the sampling point of each aquatic sediment sample on a two-dimensional topographic map and the elemental concentration of that element in each aquatic sediment sample, a three-dimensional clustering space is constructed. When performing K-means clustering on the data points in the three-dimensional clustering space, for any iteration, the centroid of that iteration is obtained. Based on the elemental concentration similarity between any data point and any centroid, a first weighted adjustment is applied to the clustering distance between any data point and any centroid. The corrected distance is obtained by: First, based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to the data point, a second weighted correction is applied to the corrected distance to obtain the final corrected distance between any data point and any centroid. Then, the final corrected distance between each data point and each centroid is obtained during each iteration. Based on all the final corrected distances, K-means clustering is performed on all data points in the three-dimensional clustering space to obtain the clustering result for any element. Finally, based on the clustering results of all elements, the analysis results of all aquatic sediment samples are analyzed. This study constructs a three-dimensional clustering space by combining the location of sampling points of each stream sediment sample on a two-dimensional topographic map with the elemental concentration of any element in each sample. The K-means clustering algorithm is then used to perform three-dimensional clustering of the data points within this space. This approach integrates the characteristics of elemental concentration and geographic distribution, improving the accuracy of elemental clustering in ore bodies and enhancing the accuracy of ore body location through clustering results. This, in turn, improves the accuracy of analyzing the laboratory results of stream sediment samples. When performing K-means clustering on data points, the clustering distance between each data point and the centroid is corrected based on elemental concentration similarity and spatial distribution characteristics. This reduces the probability of spherical clusters appearing in high-dimensional clustering by the K-means clustering algorithm. At the same time, it reduces the interference of data points with abnormal elemental concentrations caused by heavy metal pollution on the clustering results of K-means clustering, thereby improving the accuracy of the K-means clustering algorithm in high-dimensional clustering and improving the accuracy of analyzing the test results of aquatic sediment samples using the K-means clustering algorithm.

[0093] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for analyzing the test results of aquatic sediment samples, characterized in that, The method for analyzing the test results of a type of aquatic sediment sample includes: At least two stream sediment samples are obtained. For any element, the element concentration of the element is extracted in each stream sediment sample. A three-dimensional clustering space is constructed by combining the location of the sampling point of each stream sediment sample in a two-dimensional topographic map and the element concentration of the element in each stream sediment sample. When performing K-means clustering on data points in a three-dimensional clustering space, for any iteration process, the centroid of the iteration process is obtained. Based on the element concentration similarity between any data point and any centroid, the clustering distance between any data point and any centroid is first weighted and corrected to obtain the corrected distance. Based on the element concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar element concentrations to any data point, a second weighted correction is applied to the correction distance to obtain the final corrected distance between any data point and any centroid. The final corrected distance between each data point and each centroid is obtained in each iteration. Based on all the final corrected distances, K-means clustering is performed on all data points in the three-dimensional clustering space to obtain the clustering result of any element. Based on the clustering results of all elements, the test results of all aquatic sediment samples are analyzed.

2. The method for analyzing the test results of aquatic sediment samples according to claim 1, characterized in that, The construction of a three-dimensional clustering space by combining the location of the sampling point of each stream sediment sample on the two-dimensional topographic map and the elemental concentration of any element in each stream sediment sample includes: Based on the coordinates of the sampling point of each stream sediment sample in the two-dimensional topographic map, and the element concentration of any element in each stream sediment sample, the element concentration of any element in each stream sediment sample is mapped to the same three-dimensional cluster space. The x-axis of the three-dimensional cluster space represents the horizontal coordinate of the corresponding sampling point of each stream sediment sample in the two-dimensional topographic map, the y-axis represents the vertical coordinate of the corresponding sampling point of each stream sediment sample in the two-dimensional topographic map, and the z-axis represents the element concentration of any element in each stream sediment sample.

3. The method for analyzing the test results of aquatic sediment samples according to claim 1, characterized in that, The step of performing a first weighted correction on the clustering distance between any data point and any centroid based on the element concentration similarity between any data point and any centroid to obtain the corrected distance includes: Calculate the reciprocal of the absolute value of the difference between the element concentration of any data point and the element concentration of any centroid to obtain the first element concentration similarity index; In the three-dimensional clustering space, the standard deviation of the element concentration of all data points in the neighborhood of any data point is calculated and denoted as the first standard deviation. The standard deviation of the element concentration of all data points in the neighborhood of any centroid is calculated and denoted as the second standard deviation. The reciprocal of the absolute value of the difference between the first standard deviation and the second standard deviation is calculated to obtain the second element concentration similarity index. Calculate the product between the first element concentration similarity index and the second element concentration similarity index to obtain the element concentration similarity between any data point and any centroid; Based on the similarity of element concentrations, the corrected distance between any data point and any centroid is obtained.

4. The method for analyzing the test results of aquatic sediment samples according to claim 3, characterized in that, The step of obtaining the corrected distance between any data point and any centroid based on the element concentration similarity includes: The position weight is obtained by calculating the difference between constant 1 and the similarity of the element concentration, and the concentration weight is obtained by calculating the sum between constant 1 and the similarity of the element concentration. In the three-dimensional clustering space, the square of the distance between any data point and any centroid on the x-axis is calculated to obtain the x-axis squared term; the square of the distance between any data point and any centroid on the y-axis is calculated to obtain the y-axis squared term; and the square of the distance between any data point and any centroid on the z-axis is calculated to obtain the z-axis squared term. The position weights are used as the weights of the x-axis squared term and the y-axis squared term, respectively, and the concentration weights are used as the weights of the z-axis squared term. The x-axis squared term, the y-axis squared term, and the z-axis squared term are weighted and summed to obtain a weighted summation result. The arithmetic square root of the weighted summation result is used as the corrected distance between any data point and any centroid.

5. The method for analyzing the test results of aquatic sediment samples according to claim 1, characterized in that, The step of performing a second weighted correction on the corrected distance based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to any data point, to obtain the final corrected distance between any data point and any centroid, includes: The pollution level of any data point is determined based on the difference in element concentration between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar element concentrations to any data point. If the degree of contamination is greater than or equal to the preset contamination threshold, then according to the degree of contamination of any data point, the correction distance between any data point and any centroid is subject to a second weighted correction to obtain the final correction distance between any data point and any centroid. If the degree of contamination is less than the preset contamination threshold, then the corrected distance between any data point and any centroid is taken as the final corrected distance.

6. The method for analyzing the test results of aquatic sediment samples according to claim 5, characterized in that, The step of determining the contamination level of any data point based on the elemental concentration difference between any data point and its neighboring data points, and the spatial distribution characteristics of data points with similar elemental concentrations to the data point, includes: In the three-dimensional clustering space, similar data points with element concentrations similar to any data point are obtained. If the number of similar data points is 0, the contamination level of any data point is set to 1. If the number of similar data points is not zero, calculate the Euclidean distance between each similar data point and any data point, and linearly normalize the average of all Euclidean distances to obtain the first distribution characteristic value of any data point. In the three-dimensional clustering space, all data points in the neighborhood of any data point are recorded as neighborhood data points. The absolute value of the difference between the element concentration of each neighborhood data point and the element concentration of any data point is calculated. The average value of all the absolute values ​​of the difference is linearly normalized to obtain the second distribution characteristic value of any data point. The contamination level of any data point is obtained by calculating the product between the first distribution characteristic value and the second distribution characteristic value.

7. The method for analyzing the test results of aquatic sediment samples according to claim 5, characterized in that, The step of performing a second weighted correction on the corrected distance between any data point and any centroid based on the degree of contamination of any data point, to obtain the final corrected distance between any data point and any centroid, includes: Subtract the contamination level of any data point from the constant 1 to obtain the correction weight. Based on the correction weight, the correction distance between any data point and any centroid is weighted to obtain the final correction distance between any data point and any centroid.

8. The method for analyzing the test results of aquatic sediment samples according to claim 6, characterized in that, The step of obtaining similar data points with element concentrations similar to any one of the data points in the three-dimensional clustering space includes: In the three-dimensional clustering space, for any other data point besides the aforementioned data point, if the absolute value of the difference between the element concentration of the aforementioned data point and the element concentration of the other data point is less than or equal to a preset difference threshold, then the other data point is recorded as a similar data point with similar element concentration to the aforementioned data point.

Images