Underground water pollution tracing method and device and terminal equipment

Abstract

The invention is suitable for the technical field of data processing, and provides an underground water pollution tracing method and device and terminal equipment. The method comprises the following steps: performing water quality fluorescent fingerprint detection on a first underground water sample collected at a first monitoring point position in a target area to obtain a first water quality fluorescent fingerprint; comparing the first water quality fluorescent fingerprint with a water quality fluorescent fingerprint pre-stored in a database, and determining a first similarity between the first water quality fluorescent fingerprint and the pre-stored water quality fluorescent fingerprint; when a first similarity between a second water quality fluorescent fingerprint in the database and the first water quality fluorescent fingerprint is greater than or equal to a threshold value, collecting an underground water sample at a downstream point location of the first monitoring point location, and determining a second similarity between the water quality fluorescent fingerprint of the underground water sample and the second water quality fluorescent fingerprint; and when the second similarity is greater than a threshold value, determining the pollution source of the first monitoring point position according to the known pollution source of the second water quality fluorescence fingerprint. The spatial rationality and reliability of groundwater pollution tracing can be improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a method, apparatus, and terminal equipment for tracing the source of groundwater pollution. Background Technology

[0002] With rapid industrialization and urbanization, groundwater pollution has become an increasingly serious problem. Tracing the source of groundwater pollution is a crucial step in pollution control. Traditional groundwater pollution tracing methods primarily involve detecting the concentration of specific chemical components of pollutants in groundwater (such as ion ratios and characteristic organic matter), analyzing their spatial distribution patterns, and comparing them with potential pollution sources to infer the pollution source. The inventors of this application have discovered that traditional groundwater pollution tracing methods rely solely on static comparisons based on concentration similarity, lacking dynamic verification of the "spatial continuity of the pollution plume." When the chemical fingerprint of a target water sample is highly similar to that of a pollution source, this similarity may stem from coincidence (such as different types of pollution sources producing similar chemical characteristics) or background interference, potentially leading to misjudgment. Summary of the Invention

[0003] In view of this, the present application provides a groundwater pollution source tracing method, device and terminal equipment. After obtaining a high similarity result by comparing fluorescent fingerprints in the database, the upstream points of the monitoring point are sampled for verification. If the upstream points also show the same pollution fingerprint, it confirms the continuous spatial distribution of pollution characteristics, thereby improving the spatial rationality and reliability of groundwater pollution source tracing.

[0004] To achieve the above objectives, this application adopts the following technical solution:

[0005] Firstly, this application provides a method for tracing the source of groundwater pollution, including:

[0006] A water quality fluorescence fingerprint was obtained by performing water quality fluorescence fingerprint detection on the first groundwater sample at the first monitoring point in the target area. The first monitoring point can be any monitoring point in the target area.

[0007] The first water quality fluorescent fingerprint is compared with the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint; wherein, each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a known pollution source.

[0008] When the first similarity between the second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is greater than or equal to the threshold, a second groundwater sample is collected from a downstream monitoring point adjacent to the first monitoring point, and the second similarity between the water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint is determined.

[0009] When the second similarity is greater than or equal to the threshold, the groundwater pollution source of the first monitoring point is determined based on the known pollution source corresponding to the second water quality fluorescent fingerprint.

[0010] Based on the first aspect, in some embodiments, the method includes:

[0011] If the first similarity between the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is less than the threshold, then monitoring points are densely deployed upstream of the first monitoring point, and sampling is carried out in multiple time periods. The water quality fluorescent fingerprints of the sampled samples are compared one by one with the water quality fluorescent fingerprints in the water quality fluorescent fingerprint database. Based on the obtained similarity, the investigation scope and location are continuously narrowed down until the pollution source in the target area is determined.

[0012] Based on the first aspect, in some embodiments, the method further includes:

[0013] If the source of pollution in the target area cannot be determined based on similarity, then a specific pollutant is identified between the first groundwater sample and multiple suspected sources of pollution, wherein the multiple suspected sources of pollution are sources of pollution outside the water quality fluorescence fingerprint database; wherein, the specific pollutant is a pollutant whose concentration in the source of pollution is greater than or equal to a concentration threshold, or a pollutant whose concentration ratio of certain pollutants in the source of pollution is consistent.

[0014] Isotope analysis was performed on specific pollutants in the first groundwater sample and specific pollutants from multiple suspected pollution sources to determine the final pollution source from the multiple suspected pollution sources.

[0015] Based on the first aspect, in some embodiments, the step of comparing the first water quality fluorescent fingerprint with pre-stored water quality fluorescent fingerprints in a water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint includes:

[0016] The first water quality fluorescent fingerprint is converted into a first vector, and each water quality fluorescent fingerprint pre-stored in the water quality fluorescent fingerprint database is converted into a second vector. Each water quality fluorescent fingerprint pre-stored in the water quality fluorescent fingerprint database corresponds to a weight vector.

[0017] Calculate the weighted distance between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the first vector, the second vector, and the weight vector;

[0018] The first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint is determined based on the weighted distance.

[0019] Based on the first aspect, in some embodiments, the method for determining the weight vector is as follows:

[0020] Obtain intensity data of various fluorescent components of known pollution sources over time in a simulated aging environment;

[0021] Based on the intensity data, calculate the mean and standard deviation of each fluorescent component of the known pollution source, and calculate the coefficient of variation of each fluorescent component by dividing the standard deviation by the mean.

[0022] Based on the intensity data, calculate the first mean and first standard deviation of the first fluorescent component of a known pollution source, as well as the second mean and second standard deviation of the first fluorescent components of other known pollution sources. The first fluorescent component is any fluorescent component of the known pollution source.

[0023] The discrimination of the first fluorescent component was calculated based on the first mean, the first standard deviation, the second mean, and the second standard deviation;

[0024] The weights of each fluorescent component of a known pollution source are determined based on the coefficient of variation and the discrimination of each fluorescent component, thus obtaining the weight vector of each known pollution source.

[0025] Based on the first aspect, in some embodiments, calculating the discrimination of the first fluorescent component based on the first mean, the first standard deviation, the second mean, and the second standard deviation includes:

[0026] The discrimination of the first fluorescent component is calculated as the absolute value of the difference between the first mean and the second mean, and the sum of the first standard deviation and the second standard deviation.

[0027] The quotient of the absolute value and the sum is used to obtain the discrimination of the first fluorescent component.

[0028] Based on the first aspect, in some embodiments, determining the groundwater pollution source at the first monitoring point based on the known pollution source corresponding to the second water quality fluorescent fingerprint includes:

[0029] If two or more second water quality fluorescent fingerprints have a first similarity greater than a threshold with the first water quality fluorescent fingerprint, then the two or more second water quality fluorescent fingerprints and the first water quality fluorescent fingerprint are input into a positive definite matrix factorization model to obtain the contribution ratio of the known pollution sources corresponding to the two or more second water quality fluorescent fingerprints to the first groundwater sample, and the pollution source of the first groundwater sample is determined based on the contribution ratio.

[0030] Based on the first aspect, in some embodiments, the step of comparing the first water quality fluorescent fingerprint with pre-stored water quality fluorescent fingerprints in a water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint includes:

[0031] The intensity of each fluorescent component of the first water quality fluorescent fingerprint and the intensity of each fluorescent component of the pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database are determined. Each known pollution source in the water quality fluorescent fingerprint database corresponds to a fluorescent component. The fluorescent components of the first water quality fluorescent fingerprint are the same as the fluorescent components of the pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database.

[0032] The fluorescence component data matrix was processed by principal component analysis. The fluorescence component data matrix includes the fluorescence component intensities of each fluorescence component for each known pollution source and the fluorescence component intensities of each fluorescence component for the first water quality fluorescence fingerprint.

[0033] For the fluorescence component data matrix processed by principal component analysis, calculate the center location of each known pollution source and the center location of the first groundwater sample;

[0034] Based on the center location of each known pollution source and the center location of the first groundwater sample, calculate the distance between the first groundwater sample and each known pollution source;

[0035] Based on the distance between the first groundwater sample and each known pollution source, the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint is determined.

[0036] Secondly, embodiments of this application provide a groundwater pollution tracing device, comprising:

[0037] The fingerprint detection module is used to perform water quality fluorescent fingerprint detection on the first groundwater sample at the first monitoring point in the target area to obtain the first water quality fluorescent fingerprint. The first monitoring point can be any monitoring point in the target area.

[0038] The fingerprint comparison module is used to compare the first water quality fluorescent fingerprint with the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint; wherein, each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a known pollution source.

[0039] The verification module is used to collect a second groundwater sample from a downstream monitoring point adjacent to the first monitoring point when the first similarity between the second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is greater than or equal to a threshold, and to determine the second similarity between the water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint.

[0040] The pollution source determination module is used to determine the groundwater pollution source of the first monitoring point based on the known pollution source corresponding to the second water quality fluorescent fingerprint when the second similarity is greater than or equal to the threshold.

[0041] Thirdly, embodiments of this application provide a terminal device, including a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the steps of the groundwater pollution tracing method described in any of the first aspects above.

[0042] The beneficial effects of the embodiments of this application compared with the prior art include:

[0043] The aforementioned groundwater pollution tracing method involves performing water quality fluorescent fingerprinting on a first groundwater sample at a first monitoring point in the target area to obtain a first water quality fluorescent fingerprint; comparing the first water quality fluorescent fingerprint with pre-stored water quality fluorescent fingerprints in a water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint; when the first similarity between the second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is greater than or equal to a threshold, collecting a second groundwater sample from a downstream monitoring point adjacent to the first monitoring point to determine the second similarity between the water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint; when the second similarity is greater than or equal to the threshold, determining the groundwater pollution source at the first monitoring point based on the known pollution source corresponding to the second water quality fluorescent fingerprint.

[0044] In this embodiment of the application, after obtaining a high similarity result by comparing the fluorescent fingerprint of the first monitoring point with the water quality fluorescent fingerprint database, the upstream monitoring point of the first monitoring point is simultaneously sampled and verified. If the upstream monitoring point also shows the same pollution fingerprint, it confirms the continuous spatial distribution of pollution characteristics, which can reduce interference and misjudgment, thereby improving the spatial rationality and reliability of groundwater pollution source tracing. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating the groundwater pollution source tracing method provided in the embodiments of this application;

[0047] Figure 2 This is a structural block diagram of the groundwater pollution tracing device provided in the embodiments of this application;

[0048] Figure 3 This is a structural block diagram of the terminal device provided in the embodiments of this application. Detailed Implementation

[0049] The present application will be described more clearly below with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the function of the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application. These all fall within the protection scope of the present application.

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0051] See Figure 1 The groundwater pollution tracing method provided in this application embodiment may include the following steps:

[0052] Step 101: Perform water quality fluorescence fingerprint detection on the first groundwater sample at the first monitoring point in the target area to obtain the first water quality fluorescence fingerprint. The first monitoring point can be any monitoring point in the target area.

[0053] For example, geological and hydrogeological data (strata structure, aquifer characteristics, groundwater depth, flow direction, flow velocity, etc.) can be collected, and the history, location, and pollutant types of potential pollution sources in the target area can be investigated in order to arrange monitoring points.

[0054] For example, in areas where the pollution situation is unclear, a uniform grid can be used to arrange monitoring points; downstream of known or suspected point sources (such as leaks), monitoring points can be arranged in a fan shape along the groundwater flow; for multiple aquifers or pollutant density differentiation (such as light non-aqueous liquids (LNAPL) and heavy non-aqueous liquids (DNAPL), monitoring points need to be arranged at different depths. In addition, monitoring points can also be arranged upstream of the area (background value points), within the polluted area, at the edge of the pollution plume, and at the downstream boundary.

[0055] In this embodiment of the application, the water quality fluorescent fingerprint detection of groundwater samples can be performed using techniques well known in the art, which will not be described in detail here.

[0056] Step 102: Compare the first water quality fluorescent fingerprint with the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint. Each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a known pollution source.

[0057] In this embodiment, the water quality fluorescence fingerprint has dozens or even hundreds of wavelengths of data. The fingerprint differences of different pollution sources may be scattered across many dimensions, but the most significant differences are often concentrated in a few key directions. In this step, principal component analysis can be used to reduce the dimensionality of the water quality fluorescence fingerprint data matrix, find the new coordinate axis that "best distinguishes different pollution sources," discard noise and redundant information, retain only the most important features, project all complex fingerprint data into a simplified principal component space with clear physical meaning, and then calculate the distance of the water quality fluorescence fingerprint in the principal component space to determine the first similarity.

[0058] For example, step 102 may include: determining the intensity of each fluorescent component of the first water quality fluorescent fingerprint and the intensity of each fluorescent component of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database, wherein each known pollution source in the water quality fluorescent fingerprint database corresponds to a fluorescent component, and the fluorescent components of the first water quality fluorescent fingerprint are the same as those of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database; processing the fluorescent component data matrix using principal component analysis, wherein the fluorescent component data matrix includes the intensity of each fluorescent component of each known pollution source and the intensity of each fluorescent component of the first water quality fluorescent fingerprint; calculating the center position of each known pollution source and the center position of the first groundwater sample on the fluorescent component data matrix processed by principal component analysis; calculating the distance between the first groundwater sample and each known pollution source based on the center position of each known pollution source and the center position of the first groundwater sample; and determining the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the distance between the first groundwater sample and each known pollution source.

[0059] For example, the above-described processing of the fluorescence component data matrix using principal component analysis may include: constructing a fluorescence component data matrix, wherein rows of the fluorescence component data matrix represent samples, columns represent fluorescence components, and each element of the fluorescence component data matrix represents the intensity of a certain fluorescence component in a certain sample; generating an n×n covariance matrix based on the fluorescence component data matrix, wherein the covariance matrix characterizes the relationship between the n fluorescence components; determining the eigenvectors of the principal components based on the covariance matrix and constructing a projection matrix; and performing dimensionality reduction projection on the fluorescence component data matrix based on the projection matrix.

[0060] For example, suppose there are 3 known pollution sources (domestic sewage S, chemical plant wastewater C, and machining wastewater M), with 2 samples from each source, for a total of 6 known samples. Each sample has 3 fluorescent components (F1, F2, F3). First, the intensities of the 6 samples from the 3 known pollution sources and the 3 fluorescent components from the groundwater sample are organized into an original fluorescence component data matrix, f1 S1 f2 represents the specific value of the F1 fluorescent component in sample S1. S1f3 represents the specific value of the F2 fluorescent component in sample S1. S1 The values ​​of the F3 fluorescent component in sample S1 are shown in Table 1. Similarly, the values ​​of other components are shown in Table 1.

[0061] Table 1 Original Fluorescent Component Data Matrix

[0062]

[0063] To eliminate the influence of differences in the absolute intensity magnitudes of different fluorescent components, the fluorescent component data matrix can be standardized first. The mean μ and standard deviation σ of each fluorescent component from all known pollution sources are calculated, and then the intensity of that fluorescent component is standardized based on these mean μ and standard deviation σ. For example, for the F1 fluorescent component, the mean μ1 and standard deviation σ1 of the F1 fluorescent component intensity are calculated for six samples from three known pollution sources. The F1 fluorescent component intensity values ​​for all samples, including groundwater samples, are then transformed as follows: F1' = (F1 - μ1) / σ1. The same calculation process is performed for the F2 and F3 fluorescent components to obtain the standardized fluorescent component data matrix.

[0064] For the standardized fluorescence component data matrix, its covariance matrix is ​​calculated. This covariance matrix describes the linear relationship between the n fluorescence components. Specifically, the covariance matrix Σ is an n×n matrix, and each element Σ[i,j] of the covariance matrix Σ represents the covariance between feature i and feature j. The calculation formula is as follows: Z(k,i) represents the element in the k-th row and i-th column of the standardized fluorescence component data matrix, m is the row number of the standardized fluorescence component data matrix, and i and j are integers from 1 to n.

[0065] After obtaining the covariance matrix Σ, we solve for the eigenvalues ​​and eigenvectors of the covariance matrix Σ by using det(Σ -λI) = 0, where I is the identity matrix, to obtain the eigenvalues ​​λ1 to λ2. n Based on the eigenvalues ​​λ1 to λ n Selecting some eigenvalues ​​as principal components, these eigenvalues ​​should satisfy the condition from λ1 to λ2. n The proportion is relatively large, for example, greater than 90%.

[0066] After obtaining the eigenvalues ​​and eigenvectors of the covariance matrix Σ, the eigenvectors are used to form a projection matrix W, where each column of the projection matrix W corresponds to an eigenvector. Then, the standardized fluorescence component data is projected onto the space formed by the principal components, i.e., the standardized fluorescence component data is multiplied by the projection matrix W, thus completing the dimensionality reduction of the fluorescence component data matrix.

[0067] Next, the position of each known pollution source in the principal component space is calculated based on the average value of the data corresponding to each principal component in the projected matrix. Then, the distance between the groundwater sample and each known pollution source is calculated based on the position of the groundwater sample in the projected matrix. The first similarity between the groundwater sample and each known pollution source is determined based on this distance. The larger the distance between the groundwater sample and each known pollution source, the smaller the first similarity; the smaller the distance between the groundwater sample and each known pollution source, the larger the first similarity.

[0068] Specifically, each principal component corresponds to a coordinate axis in the principal component space. In the projected matrix, each known pollution source has a value corresponding to each principal component. For any principal component, the average value of all values ​​for the same known pollution source under that principal component is calculated to obtain the coordinates of that known pollution source on the coordinate axis corresponding to that principal component. Similarly, the coordinates of the known pollution source on the coordinate axes corresponding to other principal components are calculated, thus obtaining the position of the known pollution source in the principal component space, which has coordinate values ​​corresponding one-to-one with each principal component. In addition, the projected matrix also contains the values ​​of the groundwater sample. Based on the above content, the position of the groundwater sample in the principal component space can be obtained.

[0069] After obtaining the locations of known pollution sources and groundwater samples in the principal component space, the distance between the groundwater samples and each known pollution source is calculated. Then, the distance can be converted to a similarity score using the formula: "Similarity = 1 / (1 + Distance)".

[0070] Referring to Table 2, let the projected matrix correspond to principal component 1 and principal component 2, f 1 S1P Let f2 be the coordinates of sample S1 and principal component 1 on the corresponding coordinate axes. S1P Let f1 be the coordinates of sample S1 and principal component 2 on the corresponding coordinate axes. S2P Let f2 be the coordinates of sample S2 and principal component 1 on the coordinate axes. S2P Let S2 be the coordinates of the corresponding axes of principal component 2. Then, the position of the domestic sewage pollution source in the principal component space is ((f 1)). S1P +f 1 S2P ) / 2, (f 2 S1P +f 2 S2P The location of the groundwater sample in the principal component space is (f 1) / 2). X1P f 2 X1P If the distance between the groundwater sample and the known pollution source, domestic sewage, is ((f 1) X1P -(f 1 S1P +f 1 S2P ) / 2) 2 +(f 2 X1P-(f 2 S1P +f 2 S2P ) / 2) 2 ) 1 / 2 .

[0071] Table 2 Projected Matrix

[0072]

[0073] In this embodiment, weights can be assigned to each fluorescent component of a known pollution source, and calculations can be performed based on these weights. In practical applications, fluorescence signals may be altered by environmental processes (biodegradation, photolysis, adsorption), thus affecting the accuracy of source tracing. For example, certain high-molecular-weight, structurally complex fluorescent components or specific peak patterns are relatively insensitive to environmental processes. Therefore, by using extensive simulation experiments and site data, the fluorescent components with the slowest rate of change under specific environmental conditions (e.g., underground anaerobic environments) can be identified. Assigning higher weights to these fluorescent components increases their influence in source tracing, thereby improving the accuracy of source tracing.

[0074] For example, step 102 may include: converting the first water quality fluorescent fingerprint into a first vector; converting each pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database into a second vector, wherein each pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a weight vector; calculating the weighted distance between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the first vector, the second vector, and the weight vector; and determining the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the weighted distance. For the aforementioned feature values ​​λ1 to λ2... n When the proportions of each feature value are relatively close, this method can be used to calculate the first similarity.

[0075] For example, in domestic sewage, the most stable fluorescent component is the humic acid-like peak, while the most variable fluorescent components are the tyrosine-like peak and the microbial product peak. In chemical plant wastewater, the most stable fluorescent component is the specific artificial dye peak, while the most variable fluorescent component is the solvent-related peak. In mechanical processing wastewater, the most stable fluorescent component is the polycyclic aromatic hydrocarbon peak, while the most variable fluorescent component is the surfactant peak. Based on the above, in domestic sewage samples, a higher weight is assigned to the humic acid-like fluorescent component, and lower weights are assigned to the tyrosine-like peak and the microbial product peak. In chemical plant wastewater samples, a higher weight is assigned to the specific artificial dye fluorescent component, and lower weights are assigned to the solvent-related fluorescent component. In mechanical processing wastewater samples, a higher weight is assigned to the polycyclic aromatic hydrocarbon fluorescent component, and lower weights are assigned to the surfactant fluorescent component.

[0076] For example, the method for determining the weight vector can be as follows: obtain the intensity data of each fluorescent component of a known pollution source changing over time in a simulated aging environment; calculate the mean and standard deviation of each fluorescent component of the known pollution source based on the intensity data, and calculate the coefficient of variation of each fluorescent component by dividing the standard deviation by the mean; calculate the first mean and first standard deviation of the first fluorescent component of a known pollution source, and the second mean and second standard deviation of the first fluorescent components of other known pollution sources based on the intensity data, wherein the first fluorescent component is any fluorescent component of the known pollution source; calculate the discrimination of the first fluorescent component of the known pollution source based on the first mean, first standard deviation, second mean, and second standard deviation; determine the weight of each fluorescent component of the known pollution source based on the coefficient of variation and discrimination of each fluorescent component, and obtain the weight vector of each known pollution source.

[0077] In this embodiment, the specific method for determining the discrimination degree can be as follows: calculate the discrimination degree of the first fluorescent component as equal to the absolute value of the difference between the first mean and the second mean, and the sum of the first standard deviation and the second standard deviation; calculate the quotient of the absolute value and the sum to obtain the discrimination degree of the first fluorescent component.

[0078] For example, for a known pollution source, the most likely environmental process at the site is simulated to age the known pollution source, and multiple time points (e.g., 0 days, 7 days, 14 days, 30 days, 60 days and 90 days) are set for sampling. Three-dimensional fluorescence spectroscopy analysis is performed on the samples at all time points, and a unified parallel factor analysis model is performed to obtain a data matrix, where each row is a sample at a time point and each column is the intensity of a fluorescent component.

[0079] For example, for a known pollution source, at day 0, the intensities of fluorescent components F1, F2, and F3 are F10, F20, and F30, respectively; at day 7, the intensities are F17, F27, and F37, respectively; and at day 14, the intensities are F1... 14 F2 14 and F3 14 At the 30-day time point, the intensities of the F1, F2, and F3 fluorescent components were respectively F1 30 F2 30 and F3 30 At the 60-day time point, the intensities of the F1, F2, and F3 fluorescent components were respectively F1 60 F2 60 and F3 60 At the 90-day time point, the intensities of the F1, F2, and F3 fluorescent components were respectively F190 F2 90 and F3 90 .

[0080] According to F10, F17, F1 14 F1 60 and F1 90 Calculate the mean and standard deviation of the F1 fluorescent component, and then calculate the coefficient of variation (CV) of the F1 fluorescent component by dividing the standard deviation by the mean. For the CVs of the F2 and F3 fluorescent components, please refer to the calculation process for the CV of the F1 fluorescent component; it will not be repeated here. After calculating the CV, take the reciprocal of the CV to determine the weight of the fluorescent component.

[0081] Furthermore, for a known pollution source i, according to F10, F17, and F1 14 F1 60 and F1 90 Calculate the first mean and first standard deviation of the F1 fluorescent component; for other known pollution sources besides known pollution source i, calculate the F10, F17, and F1 values ​​of the other known pollution sources. 14 F1 60 and F1 90 Calculate the second mean and second standard deviation of the F1 fluorescent component. The discriminative power of the F1 fluorescent component for known pollution source i is equal to the absolute value of the difference between the first and second means, divided by the sum of the first and second standard deviations. A higher discriminative power of the F1 fluorescent component for known pollution source i indicates that the F1 fluorescent component can better distinguish known pollution source i from other known pollution sources; conversely, a lower discriminative power indicates that the F1 fluorescent component cannot effectively distinguish known pollution source i from other known pollution sources.

[0082] After obtaining the coefficient of variation and discrimination of each fluorescent component, the product of the reciprocal of the coefficient of variation and the discrimination can be used to determine the weight of each fluorescent component of the known pollution source, thus obtaining the weight vector of each known pollution source. Specifically, for a certain fluorescent component, if its coefficient of variation is small and its discrimination is large, the product of the reciprocal of the coefficient of variation and the discrimination is indeed large, and the weight corresponding to the intensity of that fluorescent component is large; otherwise, the weight corresponding to the intensity of that fluorescent component is small.

[0083] Step 103: When the first similarity between the second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is greater than or equal to the threshold, collect a second groundwater sample from a downstream monitoring point adjacent to the first monitoring point, and determine the second similarity between the water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint.

[0084] If the first similarity calculated in step 102 is greater than or equal to a threshold, it indicates that the pollution source of the first groundwater sample may be a known pollution source corresponding to the second water quality fluorescent fingerprint. To more accurately determine whether the pollution source of the first groundwater sample is a known pollution source corresponding to the second water quality fluorescent fingerprint, it is possible to detect whether the pollution source characteristics are spatially continuous. That is, a second groundwater sample is collected from a downstream monitoring point adjacent to the first monitoring point, the second similarity between the second groundwater sample's water quality fluorescent fingerprint and the second water quality fluorescent fingerprint is determined, and the spatial distribution of the pollution source characteristics at the first monitoring point is detected based on the second similarity. If the pollution source characteristics are spatially continuous, it conforms to the basic physicochemical laws of pollutant migration and diffusion in the groundwater environment, which can greatly enhance the reliability, credibility, and persuasiveness of the source tracing conclusion.

[0085] Step 104: When the second similarity is greater than or equal to the threshold, determine the groundwater pollution source of the first monitoring point based on the known pollution source corresponding to the second water quality fluorescent fingerprint.

[0086] Specifically, if both the first and second similarities are greater than or equal to the threshold, it indicates that the pollution source characteristics of the first monitoring point are spatially continuous. In this case, the groundwater pollution source of the first monitoring point can be determined based on the known pollution sources corresponding to the second water quality fluorescent fingerprint. Furthermore, if only one second water quality fluorescent fingerprint has a first similarity greater than the threshold with the first water quality fluorescent fingerprint, the known pollution source corresponding to that second water quality fluorescent fingerprint can be identified as the groundwater pollution source of the first monitoring point. If two or more second water quality fluorescent fingerprints have a first similarity greater than the threshold with the first water quality fluorescent fingerprint, further analysis is required.

[0087] Specifically, if there are two or more second water quality fluorescent fingerprints with a first similarity greater than a threshold with the first water quality fluorescent fingerprint, then the two or more second water quality fluorescent fingerprints and the first water quality fluorescent fingerprint are input into a positive definite matrix factorization model to obtain the contribution ratio of the known pollution sources corresponding to the two or more second water quality fluorescent fingerprints to the first groundwater sample, and the pollution source of the first groundwater sample is determined based on the contribution ratio.

[0088] The Positive Matrix Factorization (PMF) model decomposes the fluorescence fingerprint matrix of a groundwater sample into a pollution source feature matrix and a pollution source contribution matrix. Under the constraint that each matrix element is non-negative, it determines the contribution ratio of each known pollution source to the groundwater sample. Specifically, the PMF model takes the matrix X composed of the fluorescence fingerprint of the groundwater sample as input and decomposes it using the relation X = G×F + E to obtain the contribution ratio of each known pollution source. G is the pollution source contribution matrix, F is the pollution source feature matrix, and E is the residual matrix. All elements in G and F are non-negative.

[0089] To improve the stability and reliability of the pollution source contribution ratio results, the positive definite matrix factorization model was repeatedly calculated. Each calculation used different initial values ​​or resampled the input water quality fluorescence fingerprint data. Based on the contribution ratio results of each known pollution source to the groundwater sample obtained from the multiple calculations, the mean contribution ratio of each known pollution source was calculated, and the 95% confidence interval corresponding to the mean contribution rate was determined based on the statistical distribution of the multiple calculation results.

[0090] For example, the mean and standard deviation of the contribution ratio of multiple calculation results can be calculated, and the confidence interval can be calculated using the t-distribution. Based on the confidence level (95%) and degrees of freedom (number of calculations - 1), the corresponding t-critical value can be obtained by looking up the t-distribution table. The lower limit of the confidence interval = sample mean - t-critical value × standard error, and the upper limit of the confidence interval = sample mean + t-critical value × standard error. The standard error = standard deviation of contribution ratio / sqrt(N).

[0091] If the lower limit of the 95% confidence interval corresponding to the average contribution rate of a known pollution source is greater than zero, it indicates that the known pollution source makes a significant contribution to the pollution source of the first groundwater sample. If the lower limit of the 95% confidence interval corresponding to the average contribution rate of a known pollution source is zero, it indicates that the known pollution source may not contribute to the pollution source of the first groundwater sample. Based on this, this embodiment of the application identifies known pollution sources whose lower limit of the 95% confidence interval corresponding to the average contribution rate is greater than zero as the pollution source of the first groundwater sample. If the lower limit of the 95% confidence interval of two or more known pollution sources is greater than zero, it indicates that the pollution source of the first groundwater sample is a mixed pollution composed of the two or more known pollution sources.

[0092] In some embodiments, the above method may further include: if the first similarity between the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is less than a threshold, then densely deploy monitoring points upstream of the first monitoring point, sample at multiple time periods, compare the water quality fluorescent fingerprint of the sampled sample with the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database one by one, and continuously narrow the investigation range and location based on the obtained similarity until the pollution source in the target area is determined.

[0093] Specifically, when the first similarity score between the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is less than the threshold, it indicates that the pollution source at the first monitoring point is not on the known list of pollution sources and is very likely an unknown, undocumented pollution source. Furthermore, pollution always spreads from upstream to downstream; therefore, the upstream boundary where the pollutant appears is the location closest to the pollution source. Moreover, the water quality fluorescent fingerprint of the pollution source exhibits continuity during migration. From the pollution source's discharge point to the downstream contaminated point, the water quality fluorescent fingerprint gradually changes due to mixing and degradation, but near the source, the water quality fluorescent fingerprint characteristics are most obvious and closest to the pollution source itself. Based on this, monitoring points can be densely deployed upstream of the first monitoring point, and surprise sampling can be conducted at multiple time periods. The similarity score between the upstream and first monitoring point water quality fluorescent fingerprints can be compared, and the investigation scope and location can be continuously narrowed based on the obtained similarity scores, thereby gradually identifying the pollution source at the first monitoring point.

[0094] Optionally, the above method may further include: if the pollution source of the target area cannot be determined based on similarity, then a specific pollutant is identified between the first groundwater sample and multiple suspected pollution sources, wherein the multiple suspected pollution sources are pollution sources outside the water quality fluorescence fingerprint database; wherein the specific pollutant is a pollutant whose concentration in the pollution source is greater than or equal to a concentration threshold, or a pollutant whose concentration ratio of certain pollutants in the pollution source is consistent; isotopic analysis is performed on the specific pollutant of the first groundwater sample and the specific pollutants of the multiple suspected pollution sources to determine the final pollution source from the multiple suspected pollution sources.

[0095] Specifically, if the similarity between the groundwater sample from the first monitoring point and the groundwater sample from the upstream point and the known pollution source is both less than the threshold, then the pollution source at the first monitoring point cannot be determined by the similarity between the water quality fluorescence fingerprints. In this case, isotope analysis can be used to determine the pollution source. First, multiple suspected pollution sources other than the known pollution sources are identified, and specific pollutants are identified between the first groundwater sample and the multiple suspected pollution sources. Specific pollutants can be pollutants whose concentration in the pollution source is greater than or equal to the concentration threshold, or pollutants whose concentration proportions are consistent in the pollution source. Then, isotope analysis is performed on the specific pollutants in the first groundwater sample and the specific pollutants in the multiple suspected pollution sources.

[0096] For example, if the concentration of pollutant A in the first groundwater sample is greater than the concentration threshold, and the concentration in one or more suspected pollution sources is also greater than the concentration threshold, then pollutant A can be considered as a specific pollutant between the first groundwater sample and the suspected pollution sources.

[0097] For example, if the concentration ratio of pollutant A to pollutant B in the first groundwater sample is m, and the concentration ratio of pollutant A to pollutant B in one or more suspected pollution sources is also m or close to m, then the specific pollutants among the first groundwater sample and multiple suspected pollution sources are pollutant A, pollutant B, and the concentration ratio m between pollutant A and pollutant B.

[0098] Different isotopic masses of the same element can undergo slight separation (fractionation) during physical, chemical, and biological processes, resulting in unique isotopic ratios for substances from different sources, forming a stable "isotopic fingerprint." Accurately determining the stable isotopic ratios of specific pollutants in a first groundwater sample and each suspected pollution source is crucial. If the isotopic ratio of the first groundwater sample is consistent with the isotopic ratio of a suspected pollution source within the error range, and this isotopic ratio is significantly different from (much greater or much smaller than) the isotopic ratios of other suspected pollution sources, then that suspected pollution source is confirmed as the final pollution source.

[0099] The above-mentioned groundwater pollution tracing method obtains a high similarity result by comparing the fluorescent fingerprint of the first monitoring point with the water quality fluorescent fingerprint database. Then, it samples and verifies the upstream monitoring point. If the upstream monitoring point also shows the same pollution fingerprint, it confirms the continuous spatial distribution of pollution characteristics, which can reduce interference and misjudgment, thereby improving the spatial rationality and reliability of groundwater pollution tracing.

[0100] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0101] Corresponding to the groundwater pollution source tracing method described in the above embodiments, Figure 2 A structural block diagram of the groundwater pollution tracing device provided in the embodiments of this application is shown. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0102] See Figure 2 This application provides a groundwater pollution tracing device, which may include a fingerprint detection module 201, a fingerprint comparison module 202, a verification module 203, and a pollution source determination module 204.

[0103] The fingerprint detection module 201 is used to perform water quality fluorescent fingerprint detection on the first groundwater sample at the first monitoring point in the target area to obtain the first water quality fluorescent fingerprint. The first monitoring point can be any monitoring point in the target area.

[0104] The fingerprint comparison module 202 is used to compare the first water quality fluorescent fingerprint with the water quality fluorescent fingerprints pre-stored in the water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint; wherein, each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a known pollution source.

[0105] The verification module 203 is used to collect a second groundwater sample from a downstream monitoring point adjacent to the first monitoring point when the first similarity between the second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is greater than a threshold, and to determine the second similarity between the water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint.

[0106] The pollution source determination module 204 is used to determine the groundwater pollution source of the first monitoring point based on the known pollution source corresponding to the second water quality fluorescent fingerprint when the second similarity is greater than the threshold.

[0107] Optionally, the above-mentioned device further includes an upstream pollution determination module, which is used to: if the first similarity between the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is less than a threshold, then densely deploy monitoring points upstream of the first monitoring point, sample at multiple time periods, compare the water quality fluorescent fingerprint of the sampled sample with the water quality fluorescent fingerprint in the water quality fluorescent fingerprint database one by one, and continuously narrow the investigation range and location based on the obtained similarity until the pollution source in the target area is determined.

[0108] Optionally, the pollution source determination module 204 is further configured to: if the pollution source of the target area cannot be determined based on similarity, then determine a specific pollutant between the first groundwater sample and multiple suspected pollution sources, wherein the multiple suspected pollution sources are pollution sources outside the water quality fluorescence fingerprint database; wherein, the specific pollutant is a pollutant whose concentration in the pollution source is greater than or equal to a concentration threshold, or a pollutant whose concentration ratio of certain pollutants in the pollution source is consistent; perform isotope analysis on the specific pollutant of the first groundwater sample and the specific pollutant of the multiple suspected pollution sources, and determine the final pollution source from the multiple suspected pollution sources.

[0109] Optionally, the fingerprint comparison module 202 is specifically used to: convert the first water quality fluorescent fingerprint into a first vector; convert each water quality fluorescent fingerprint pre-stored in the water quality fluorescent fingerprint database into a second vector, wherein each water quality fluorescent fingerprint pre-stored in the water quality fluorescent fingerprint database corresponds to a weight vector; calculate the weighted distance between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the first vector, the second vector, and the weight vector; and determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the weighted distance.

[0110] Optionally, the weight vector can be determined as follows: obtain the intensity data of each fluorescent component of a known pollution source changing over time in a simulated aging environment; calculate the mean and standard deviation of each fluorescent component of the known pollution source based on the intensity data, and calculate the coefficient of variation of each fluorescent component by dividing the standard deviation by the mean; calculate the first mean and first standard deviation of the first fluorescent component of a known pollution source, and the second mean and second standard deviation of the first fluorescent components of other known pollution sources based on the intensity data, where the first fluorescent component is any fluorescent component of the known pollution source; calculate the discrimination of the first fluorescent component based on the first mean, first standard deviation, second mean, and second standard deviation; determine the weight of each fluorescent component of the known pollution source based on the coefficient of variation and discrimination of each fluorescent component, and obtain the weight vector of each known pollution source.

[0111] Optionally, the step of calculating the discrimination of the first fluorescent component based on the first mean, the first standard deviation, the second mean, and the second standard deviation includes: calculating that the discrimination of the first fluorescent component is equal to the absolute value of the difference between the first mean and the second mean, and the sum of the first standard deviation and the second standard deviation; and calculating the quotient of the absolute value and the sum to obtain the discrimination of the first fluorescent component.

[0112] Optionally, the pollution source determination module 204 is specifically used to: if there are two or more second water quality fluorescent fingerprints with a first similarity greater than a threshold with the first water quality fluorescent fingerprint, then input the two or more second water quality fluorescent fingerprints and the first water quality fluorescent fingerprint into a positive definite matrix factorization model to obtain the contribution ratio of the known pollution sources corresponding to the two or more second water quality fluorescent fingerprints to the first groundwater sample, and determine the pollution source of the first groundwater sample based on the contribution ratio.

[0113] Optionally, the fingerprint comparison module 202 is specifically used for: determining the intensity of each fluorescent component of the first water quality fluorescent fingerprint and the intensity of each fluorescent component of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database, wherein each known pollution source in the water quality fluorescent fingerprint database corresponds to a fluorescent component, and the fluorescent components of the first water quality fluorescent fingerprint are the same as those of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database; processing the fluorescent component data matrix using principal component analysis, wherein the fluorescent component data matrix includes the intensity of each fluorescent component of each known pollution source and the intensity of each fluorescent component of the first water quality fluorescent fingerprint; calculating the center position of each known pollution source and the center position of the first groundwater sample on the fluorescent component data matrix processed by principal component analysis; calculating the distance between the first groundwater sample and each known pollution source based on the center position of each known pollution source and the center position of the first groundwater sample; and determining the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint based on the distance between the first groundwater sample and each known pollution source.

[0114] Figure 3 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. Figure 3 As shown, the terminal device 300 in this embodiment includes a processor 310 and a memory 320. The memory 320 stores a computer program that can run on the processor 310, such as a groundwater pollution tracing program. When the processor 310 executes the computer program, it implements the steps in the above-described groundwater pollution tracing method embodiment, for example... Figure 1 As shown in 101 to 104. Alternatively, when the processor 310 executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 2 The functions of the fingerprint detection module 201 to the pollution source determination module 204 are shown.

[0115] The terminal device 300 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The terminal device may include, but is not limited to, a processor 310 and a memory 320. Those skilled in the art will understand that... Figure 3 This is merely an example of terminal device 300 and does not constitute a limitation on terminal device 300. It may include more or fewer components than shown, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0116] The processor 310 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0117] The memory 320 can be an internal storage unit of the terminal device 300, such as a hard disk or memory of the terminal device 300. The memory 320 can also be an external storage device of the terminal device 300, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device 300. Furthermore, the memory 320 can include both internal and external storage units of the terminal device 300. The memory 320 is used to store the computer program and other programs and data required by the terminal device. The memory 320 can also be used to temporarily store data that has been output or will be output.

[0118] The above-described 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 of groundwater pollution source tracing, characterized by, The method comprises: fluorescent fingerprint detection is performed on a first groundwater sample collected at a first monitoring point in a target area to obtain a first water quality fluorescent fingerprint, the first monitoring point being any monitoring point in the target area; the first water quality fluorescent fingerprint is compared with pre-stored water quality fluorescent fingerprints in a water quality fluorescent fingerprint database to determine a first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint, wherein each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponds to a known pollution source; when a second water quality fluorescent fingerprint in the water quality fluorescent fingerprint database has a first similarity with the first water quality fluorescent fingerprint greater than or equal to a threshold value, a second groundwater sample is collected at a downstream monitoring point adjacent to the first monitoring point, and a second similarity between a water quality fluorescent fingerprint of the second groundwater sample and the second water quality fluorescent fingerprint is determined; when the second similarity is greater than or equal to the threshold value, a groundwater pollution source of the first monitoring point is determined according to a known pollution source corresponding to the second water quality fluorescent fingerprint.

2. The groundwater pollution source apportionment method of claim 1, wherein, The method comprises: if the first similarity between each water quality fluorescent fingerprint in the water quality fluorescent fingerprint database and the first water quality fluorescent fingerprint is less than the threshold value, monitoring points are arranged in the upstream of the first monitoring point, sampling is performed at multiple time periods, the water quality fluorescent fingerprints of the sampling samples are compared with the water quality fluorescent fingerprints in the water quality fluorescent fingerprint database one by one, and the range and direction of investigation are continuously narrowed according to the obtained similarities until the pollution source of the target area is determined.

3. The groundwater pollution source apportionment method of claim 2, wherein, The method further comprises: if the pollution source of the target area cannot be determined according to the similarities, specific pollutants between the first groundwater sample and a plurality of suspected pollution sources outside the water quality fluorescent fingerprint database are determined, wherein the specific pollutants are pollutants with a concentration greater than or equal to a concentration threshold value in the pollution sources, or are pollutants with a consistent concentration ratio of certain pollutants in the pollution sources; isotope analysis is performed on the specific pollutants of the first groundwater sample and the specific pollutants of the plurality of suspected pollution sources to determine a final pollution source from the plurality of suspected pollution sources.

4. The groundwater pollution source apportionment method of claim 1, wherein, The method of comparing the first water quality fluorescent fingerprint with the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database to determine the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint comprises: the first water quality fluorescent fingerprint is converted into a first vector, and each pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database is converted into a second vector, each pre-stored water quality fluorescent fingerprint in the water quality fluorescent fingerprint database corresponding to a weight vector; a weighted distance between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint is calculated according to the first vector, the second vector and the weight vector; the first similarity between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint is determined according to the weighted distance.

5. The groundwater pollution source apportionment method of claim 4, wherein, The method of determining the weight vector comprises: intensity data of each fluorescent component of the known pollution source in a simulated aging environment over time is obtained; the mean and standard deviation of each fluorescent component of the known pollution source are calculated according to the intensity data, and the coefficient of variation of each fluorescent component is obtained by calculating the quotient of the standard deviation and the mean of each fluorescent component; According to the intensity data, a first mean and a first standard deviation of a first fluorescent component of a certain known pollution source are calculated, and a second mean and a second standard deviation of the first fluorescent component of other known pollution sources are calculated, the first fluorescent component being any fluorescent component of the certain known pollution source; A discrimination degree of the first fluorescent component is calculated according to the first mean, the first standard deviation, the second mean and the second standard deviation; Weights of each fluorescent component of the known pollution sources are determined according to the coefficient of variation and the discrimination degree of each fluorescent component, and a weight vector of each known pollution source is obtained.

6. The groundwater pollution source apportionment method of claim 5, wherein, The calculation of the discrimination degree of the first fluorescent component according to the first mean, the first standard deviation, the second mean and the second standard deviation comprises: The discrimination degree of the first fluorescent component is calculated as the absolute value of the difference between the first mean and the second mean, and the sum of the first standard deviation and the second standard deviation; The discrimination degree of the first fluorescent component is obtained by calculating the quotient of the absolute value and the sum.

7. The groundwater pollution source apportionment method according to any one of claims 1 to 6, characterized in that, The determination of the underground water pollution source of the first monitoring point according to the known pollution sources corresponding to the second water quality fluorescent fingerprints comprises: If the first similarity degrees between the two or more second water quality fluorescent fingerprints and the first water quality fluorescent fingerprint are greater than a threshold value, the two or more second water quality fluorescent fingerprints and the first water quality fluorescent fingerprint are input into a positive definite matrix factorization model to obtain the contribution proportions of the two or more second water quality fluorescent fingerprints to the first underground water sample, and the pollution source of the first underground water sample is determined according to the contribution proportions.

8. The groundwater pollution source apportionment method according to any one of claims 1 to 6, characterized in that, The comparison of the first water quality fluorescent fingerprint with the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database to determine the first similarity degrees between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint comprises: The intensities of each fluorescent component of the first water quality fluorescent fingerprint and the intensities of each fluorescent component of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database are determined, each known pollution source in the water quality fluorescent fingerprint database corresponding to one fluorescent component, the fluorescent components of the first water quality fluorescent fingerprint being the same as the fluorescent components of the pre-stored water quality fluorescent fingerprints in the water quality fluorescent fingerprint database; The fluorescent component data matrix is processed by principal component analysis, the fluorescent component data matrix comprising the intensities of each fluorescent component of each known pollution source and the intensities of each fluorescent component of the first water quality fluorescent fingerprint; The center positions of each known pollution source and the center position of the first underground water sample are calculated based on the fluorescent component data matrix processed by the principal component analysis; The distances between the first underground water sample and each known pollution source are calculated according to the center positions of each known pollution source and the center position of the first underground water sample; The first similarity degrees between the first water quality fluorescent fingerprint and each pre-stored water quality fluorescent fingerprint are determined according to the distances between the first underground water sample and each known pollution source.

9. An apparatus for groundwater pollution source tracing, characterized by, The fingerprint detection module is configured to perform water quality fluorescent fingerprint detection on a first underground water sample at a first monitoring point in a target region to obtain a first water quality fluorescent fingerprint, the first monitoring point being any monitoring point in the target region. ​ The fingerprint comparison module is configured to compare the first water quality fluorescence fingerprint with pre-stored water quality fluorescence fingerprints in a water quality fluorescence fingerprint database to determine a first similarity between the first water quality fluorescence fingerprint and each pre-stored water quality fluorescence fingerprint, wherein each water quality fluorescence fingerprint in the water quality fluorescence fingerprint database corresponds to a known pollution source. The verification module is configured to, when a second water quality fluorescence fingerprint in the water quality fluorescence fingerprint database has a first similarity with the first water quality fluorescence fingerprint greater than or equal to a threshold value, collect a second groundwater sample from a downstream monitoring point adjacent to the first monitoring point, and determine a second similarity between a water quality fluorescence fingerprint of the second groundwater sample and the second water quality fluorescence fingerprint. The pollution source determination module is configured to, when the second similarity is greater than or equal to the threshold value, determine a groundwater pollution source of the first monitoring point according to a known pollution source corresponding to the second water quality fluorescence fingerprint.

10. A terminal device comprising a memory and a processor, said memory having stored a computer program executable on said processor, characterized in that, The processor executes the computer program to implement the steps of the groundwater pollution source tracing method according to any one of claims 1 to 8.

Images