Ecological safety assessment method and system based on heavy metal pollution in sediments

Abstract

The present invention discloses an ecological safety assessment method and system based on heavy metal pollution in sediments, which belongs to environmental risk assessment technology, analyzes sediment samples to obtain the content value of target heavy metals at historical time points; divides the historical time points into a first time point and a second time point, and defines the corresponding content values ​​as a first content value and a second content value; establishes a prediction model, and the prediction model predicts the first prediction value at the second time point based on the first content value and additional information; compares the second content value with the first prediction value, obtains an error value, and generates multiple simulation values ​​based on the error value; the prediction model predicts the future second prediction value based on the second content value and additional information, synthesizes the second prediction value with different simulation values, generates multiple prediction results, and evaluates the river ecological environment based on the prediction results. The present invention predicts the development of the future ecological environment, so as to better help relevant personnel formulate protection measures for the river ecological environment.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental risk assessment, and in particular relates to an ecological safety assessment method and system based on heavy metal pollution in sediments. Background Art

[0002] Heavy metals discharged into rivers are transferred in water bodies, silt and sediments, and enter organisms through the food chain. Through bioaccumulation and biomagnification, they will seriously affect the survival safety of aquatic organisms. Therefore, it is necessary to evaluate the heavy metal pollution in rivers. The ecological risk of heavy metals in sediments generally adopts a qualitative judgment method compared with relevant standards.

[0003] For example, a Chinese patent document with publication number CN109308577A discloses a method for evaluating the environmental quality of heavy metals in sediments. The method first collects basic data on sediments, including regional geochemical background values ​​of heavy metals in sediments and physical and chemical properties of sediments, specifically the content of heavy metals in sediments, the content of aluminum oxide in sediments, the content of organic matter in sediments, and the median particle size of sediments; then, the content of heavy metals in sediments and the regional geochemical background values ​​of heavy metals in sediments are processed by triangular fuzzy numbers and α cutoffs; then, the enrichment index of heavy metals in sediments is calculated, correlation analysis is performed to determine the source of heavy metals in sediments, and the geoaccumulation index is calculated, and the pollution level of heavy metals in sediments is determined based on membership calculation; finally, the single factor pollution index is calculated, the potential ecological hazard index is calculated, and the ecological risk of heavy metals in sediments is evaluated based on membership calculation. For another example, a Chinese patent application with publication number CN112950044A discloses an ecological risk assessment method for heavy metals in sediments, which includes dividing a detection area, collecting plant samples in the detection area for detection and analysis, and calculating the concentration levels of heavy metal pollutants in plant samples in different sampling areas; obtaining the concentration levels of heavy metal pollutants in the sediments where the plants are located, obtaining a relationship model between the concentrations of heavy metal pollutants in plant samples and the concentrations of heavy metal pollutants in the sediments where the plants are located, and obtaining the concentration values ​​of heavy metal pollutants in river sediments in the sampling area based on the relationship model; and conducting an ecological risk assessment of the river based on the background values ​​of heavy metals and the concentration values ​​of heavy metal pollutants in river sediments in the sampling area.

[0004] The above-mentioned existing technologies only evaluate the current river ecological environment. With the development of artificial intelligence, it becomes possible to predict the pollution of the future ecological environment. By predicting the future pollution of the river, it can better help relevant personnel to formulate protection measures for the river ecological environment. Summary of the invention

[0005] To solve the above problems, the present invention provides an ecological safety assessment method and system based on heavy metal pollution in sediments to solve the problems in the prior art.

[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes an ecological safety assessment method based on heavy metal pollution in sediments, comprising:

[0007] Determine a plurality of collection points and target heavy metals, collect samples at the collection points, and analyze sediment samples to obtain first data, wherein the first data includes content values ​​of the target heavy metal at a plurality of historical time points;

[0008] Dividing the historical time point into a first time point and a second time point, the content value corresponding to the first time point is defined as a first content value, and the content value at the second time point is defined as a second content value;

[0009] Acquiring additional information, the additional information including estuary confluence information within the target area, and aquaculture information, industrial activity information, and human activity information at the historical time point;

[0010] Establishing a prediction model, wherein the prediction model predicts a first predicted value of the target heavy metal at different collection points at the second time point based on the first content value and the additional information;

[0011] Comparing the second content value with the first predicted value, obtaining an error value at the second time point, and generating a plurality of simulation values ​​based on the error value;

[0012] The prediction model predicts a second predicted value of the target heavy metal at a third time point in the future based on the second content value and the additional information, synthesizes the second predicted value with different simulation values ​​to generate multiple prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results.

[0013] Further, calculating the second prediction value includes the following steps:

[0014] Calculating first differences between the same target heavy metal content values ​​at the same collection point in adjacent years, clustering all the first differences to obtain a clustering result, wherein the clustering result includes a plurality of categories;

[0015] Establish multiple prediction models based on the content value in each category and the additional information respectively, define the current time point as the fourth time point, and the previous time point of the fourth time point as the fifth time point, calculate the second difference between the fourth time point and the fifth time point, determine the first difference that is closest to the second difference value, and define it as a target difference, determine the prediction model to be used based on the target difference, and define it as a target model, and the target model calculates the second prediction value according to the content value at the fourth time point and the additional information.

[0016] Further, clustering the first differences comprises the following steps:

[0017] Clustering the first differences into a first number, extracting a cluster center from each of the categories, calculating a third difference between the cluster center and other first differences in the same category, adding all the third differences to obtain a first score for the category, and if there is a first score greater than a first threshold, re-clustering the content values ​​into a second number, the second number being greater than the first number, and calculating the first score for each of the categories when clustered into the second number;

[0018] If there are still cases where the first score is greater than the first threshold, then calculate the first average of all the first scores under the first number, and the second average of all the first scores under the second number, and continue to cluster the content values ​​into a third number, where the third number is greater than the second number, and repeat this step until there are no more cases where the first score is greater than the first threshold in the clustering result;

[0019] A coordinate system is established with the number of clusters as the horizontal axis and the average value as the vertical axis, a curve showing the change of the average value with the number of clusters is plotted in the coordinate system, the curvature of each coordinate point in the curve is calculated, the number of clusters corresponding to the point with the largest curvature is taken as the final number of clusters, and the category under the final number of clusters is taken as the clustering result.

[0020] Further, generating the simulation value comprises the following steps:

[0021] Generate multiple first time series, where the first time series are the error values ​​of the target heavy metal at different collection points at the same second time point, and calculate the mean and variance of each first time series, and the covariance between different first time series;

[0022] The distribution function of the error value is calculated based on the first time series, and based on the Monte Carlo method, a set of random number sequences is generated in combination with the mean, variance and covariance of the distribution function, and a target matrix is ​​generated based on the random number sequence, wherein the row vector of each row of the target matrix corresponds to the simulation value of different acquisition points at the second time point.

[0023] Furthermore, the estuary intersection information includes water inflow, which is the annual water inflow into the river inlet that is less than the first predetermined distance from the collection point. The breeding information includes the annual amount of aquatic products released within the second predetermined distance range of the collection point, and the value of the feed released. The industrial activity information includes the annual emissions of various target heavy metals discharged by the target factory, and the target factory is a factory that is less than the third predetermined distance from the collection point. The human activity information is the number of tourists within the fourth predetermined distance range of the collection point.

[0024] Further, obtaining the first data comprises the following steps:

[0025] Columnar sediment cores are collected at the collection point, and sampling is performed at intervals of 0.5 cm to obtain a plurality of sediment samples. The formation year of the sediment samples is determined based on isotope dating technology, and the content value of the target heavy metal in the sediment samples is obtained based on atomic absorption and atomic fluorescence methods.

[0026] Furthermore, the river ecological environment is evaluated based on the geoaccumulation index method.

[0027] Furthermore, the prediction model is a linear regression model.

[0028] Furthermore, the target heavy metals include Cr, Mn, Ni, Zn, As, Pb, Cu, and Co.

[0029] The present invention also provides an ecological safety assessment system based on heavy metal pollution in sediments, which is used to implement the above-mentioned ecological safety assessment method based on heavy metal pollution in sediments, and the system includes:

[0030] An analysis module, determining a plurality of collection points and target heavy metals, collecting samples at the collection points, and analyzing sediment samples to obtain first data, wherein the first data includes content values ​​of the target heavy metal at a plurality of historical time points;

[0031] A prediction module, dividing the historical time point into a first time point and a second time point, defining the content value corresponding to the first time point as a first content value, and defining the content value at the second time point as a second content value, acquiring additional information, the additional information including estuary intersection information in the target area, and aquaculture information, industrial activity information, and human activity information at the historical time point, establishing a prediction model, the prediction model predicting a first prediction value of the target heavy metal at different collection points at the second time point based on the first content value and the additional information;

[0032] An expansion module, comparing the second content value with the first predicted value, obtaining an error value at the second time point, and generating a plurality of simulation values ​​based on the error value;

[0033] An evaluation module, wherein the prediction model predicts a second predicted value of the target heavy metal at a third time point in the future based on the second content value and the additional information, synthesizes the second predicted value with different simulation values ​​to generate a plurality of prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results.

[0034] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0035] The present invention first collects columnar sediment cores at multiple collection points, analyzes them to obtain the contents of various heavy metals in historical years, thereby realizing the collection of basic data, and also incorporates estuary intersection information, aquaculture information, industrial activity information and human activity information, further improving the accuracy and rationality of the prediction; then, a prediction model is established based on the above basic data, and before the prediction is made, the historical time point is divided into a first time point and a second time point, and the data at the second time point is predicted based on the data at the first time point, so that the prediction errors of various heavy metals can be obtained, and then according to these errors, simulation values ​​with a distribution close to their numerical laws can be generated; finally, the prediction value output by the prediction model is added to different simulation values, so as to generate richer prediction results for relevant personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of the steps of an ecological safety assessment method based on heavy metal pollution in sediments of the present invention;

[0037] Figure 2 The schematic diagram is a structural diagram of an ecological safety assessment system based on heavy metal pollution in sediments according to the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0039] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.

[0040] like Figure 1 As shown, an ecological safety assessment method based on heavy metal pollution in sediments includes:

[0041] Step S1: determine multiple collection points and target heavy metals, collect samples at the collection points, and analyze sediment samples to obtain first data, where the first data includes content values ​​of the target heavy metals at multiple historical time points.

[0042] First, determine the river area to be evaluated, select multiple collection points in the river area, and determine the content values ​​of the target heavy metals to be obtained this time. The target heavy metals in this embodiment include Cr, Mn, Ni, Zn, As, Pb, Cu, Co, etc.; collect columnar sediment cores at the collection points, and then sample the columnar sediment cores at intervals of 0.5 cm to obtain multiple sediment samples. Each sediment sample is analyzed to obtain the year in which the sediment sample was formed and the content values ​​of various target heavy metals in the sediment sample. This embodiment generates historical time points in years, such as each year from 1970 to 2024 as a historical time point.

[0043] Step S2: Divide the historical time point into a first time point and a second time point, the content value corresponding to the first time point is defined as a first content value, and the content value at the second time point is defined as a second content value.

[0044] Step S3: Acquire additional information, including estuary confluence information within the target area, as well as aquaculture information, industrial activity information, and human activity information at historical time points.

[0045] For example, the years between 1980 and 2020 are taken as the first time point, and the years between 2020 and 2024 are taken as the second time point. The river confluence information includes which rivers are confluenced, the annual inflow of the confluenced rivers, which is defined as A in this embodiment, the breeding information includes the number of various aquatic products released each year, which is defined as B, and the feed value, which is defined as C, the industrial activity information includes the annual heavy metal emissions disclosed by the factories on both sides of the river, which is defined as D, and the human activity information includes the population tourism information, which is defined as E.

[0046] Step S4: establishing a prediction model, wherein the prediction model predicts a first predicted value of the target heavy metal at different collection points at a second time point based on the first content value and the additional information.

[0047] The prediction model of this embodiment is a linear regression model. In other embodiments, it can also be a BP neural network model, a Bayesian model, etc. After the prediction model is established, the prediction model is first tested. Specifically, the first content value of the target heavy metal at each collection point is input into the prediction model to obtain the first predicted value of each target heavy metal at the second time point.

[0048] Step S5: Compare the second content value with the first predicted value, obtain an error value at a second time point, and generate a plurality of simulation values ​​based on the error value.

[0049] Since the second content value is the actual measured value, by comparing the first predicted value and the second content value at the same second time point, the prediction error for each target heavy metal at each collection point can be obtained, that is, the error value. Then, multiple simulation values ​​are generated based on the error value, and each simulation value has a similar numerical distribution to all error values.

[0050] Step S6: The prediction model predicts the second predicted value of the target heavy metal at a third time point in the future based on the second content value and the additional information, synthesizes the second predicted value with different simulation values ​​to generate a variety of prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results.

[0051] For example, the third time point is 2030. First, the collection point to be predicted is determined, and additional information of the collection point and the second content values ​​of various target heavy metals at the surrounding collection points are obtained, which are input into the prediction model to obtain the second prediction values ​​of various target heavy metals at the collection point in 2030. Since this embodiment includes 8 target heavy metals, the second prediction value is a numerical sequence including 8 elements, which is defined as a prediction sequence here. The simulation value is also a numerical sequence including 8 elements. If three groups of model values ​​are generated, they are defined as model sequence 1, model sequence 2 and simulation sequence 3. The prediction sequence is added to the simulation sequences 1-3 respectively to obtain prediction results 1-3. Since the prediction results are also the contents of various target heavy metals, the river ecological environment in 2030 can be evaluated in combination with the existing technology.

[0052] The present invention first collects columnar sediment cores at multiple collection points, analyzes them to obtain the contents of various heavy metals in historical years, thereby realizing the collection of basic data, and also incorporates estuary intersection information, aquaculture information, industrial activity information and human activity information, further improving the accuracy and rationality of the prediction; then, a prediction model is established based on the above basic data, and before the prediction is made, the historical time point is divided into a first time point and a second time point, and the data at the second time point is predicted based on the data at the first time point, so that the prediction errors of various heavy metals can be obtained, and then according to these errors, simulation values ​​with a distribution close to their numerical laws can be generated; finally, the prediction value output by the prediction model is added to different simulation values, so as to generate richer prediction results for relevant personnel.

[0053] It is particularly noteworthy that the present invention can predict the development of the future ecological environment, thereby better helping relevant personnel to formulate protection measures for the river ecological environment.

[0054] In this embodiment, calculating the second prediction value includes the following steps:

[0055] The first difference of the same target heavy metal content value in adjacent years at the same collection point is calculated, and all the first differences are clustered to obtain a clustering result, which includes multiple categories.

[0056] Multiple prediction models are established based on the content values ​​and additional information in each category, the current time point is defined as the fourth time point, the previous time point of the fourth time point is defined as the fifth time point, the second difference between the fourth time point and the fifth time point is calculated, the first difference closest to the second difference is determined, defined as the target difference, the prediction model to be used is determined based on the target difference, defined as the target model, the target model calculates the second prediction value according to the content value and additional information at the fourth time point.

[0057] For example, the target heavy metal for clustering this time is Cr. The Cr content value of each collection point in each historical year is obtained. Then, for collection point 1, the first difference of Cr content between 1980 and 1981, and the first difference of Cr content between 1981 and 1982 are calculated. The subsequent calculation process is similar. After performing the above calculations on all collection points, a large number of first differences can be obtained. These first differences are clustered to obtain multiple categories. Through clustering, the values ​​with smaller changes in Cr content values ​​can be divided into one category, the values ​​with moderate changes can be divided into one category, and the values ​​with larger changes can be divided into one category. In this way, the data can be distinguished, and then a prediction model can be established in a targeted manner. For example, in 1980, -1990, the Cr content values ​​of some sampling points in adjacent years changed little. By fitting the additional information data during this period, for example, the following function can be obtained: Cr content value = constant 1 + (0.1*A+0.2*B+0.3*C+0.28*D+0.22*E) * number of years. Between 1990 and 2000, the Cr content values ​​of some sampling points changed greatly, and the following function can be obtained: Cr content value = constant 2 + (0.25*A+0.2*B+0.3*C+0.15*D+0.1*E) * number of years, where the number of years is the time interval with the current time. For example, if 2030 is predicted in 2024, the number of years is 6.

[0058] For example, if you want to predict collection point 1, the current year is 2024, which corresponds to the fourth time point, and the previous year is 2023, which corresponds to the fifth time point, then calculate the second difference of the Cr content value of collection point 1 at the fourth time point and the fifth time point, determine the first difference value closest to the second difference value, and then determine which prediction model is established based on the first difference value according to the matching first difference value, and set the corresponding model as the target model, and then substitute the Cr content value, additional information and year number of the year into the prediction model, so as to obtain the second prediction value. In other embodiments, you can also predict the additional information of each year in the future, and predict the first difference value that may appear in each year based on the additional information, so that a more accurate second prediction value can be obtained.

[0059] In this embodiment, clustering the first difference includes the following steps:

[0060] Cluster the first differences into a first number, extract the cluster center from each category, calculate the third difference between the cluster center and other first differences in the same category, add all the third differences to obtain the first score of the category, and if there is a first score greater than the first threshold, re-cluster the content values ​​into a second number, the second number is greater than the first number, and calculate the first score of each category when clustered into the second number.

[0061] Specifically, the K-means algorithm is used for clustering. First, a first number with a smaller value is set, and then the first difference is clustered into the first number of categories. After the clustering is completed, if there are 3 elements 1-3 in category 1 and the cluster center is 1, the third difference 1 between element 1 and element 2, and the third difference 2 between element 1 and element 3 are calculated, and the third difference 1 and the third difference 2 are added as the first score of category 1. If the first score is too large, greater than the first threshold, it indicates that there is a large gap between the elements in category 1, which means that the number of clusters is not set appropriately. At this time, the clustering data is increased for clustering again, that is, the second number is set, and the first difference is clustered again.

[0062] If there are still first scores greater than the first threshold, calculate the first average of all first scores under the first number, and the second average of all first scores under the second number, then continue to cluster the content values ​​into a third number, the third number is greater than the second number, and repeat this step until there are no first scores greater than the first threshold in the clustering results.

[0063] After clustering based on the second number, if there is still a first score greater than the first threshold, it indicates that the clustering is better or worse. However, in order to evaluate the clustering effect of the last time and this time, the first average and second average of the first scores of all categories are calculated. It can be seen that as the number of clusters increases, the average value will become smaller and smaller, and the second average value will be smaller than the first average value. After that, the first difference is clustered into the third number until there is no first score greater than the first threshold.

[0064] A coordinate system is established with the number of clusters as the horizontal axis and the average value as the vertical axis. A curve showing the change of the average value with the number of clusters is plotted in the coordinate system. The curvature of each coordinate point in the curve is calculated. The number of clusters corresponding to the point with the largest curvature is taken as the final number of clusters, and the category under the final number of clusters is taken as the clustering result.

[0065] After the coordinate system is established, the horizontal coordinate of each coordinate point in the curve corresponds to the number of clusters each time. The point with the largest curvature is the inflection point of the curve. Before the inflection point, the average value drops faster, and after the inflection point, the average value drops slower, indicating that after the inflection point, increasing the number of clusters will also obtain lower benefits. Therefore, the number of clusters corresponding to the point with the largest curvature is taken as the optimal number.

[0066] The present embodiment generates a simulation value including the following steps:

[0067] Generate multiple first time series, the first time series are the error values ​​of target heavy metals at different collection points at the same second time point, calculate the mean value and variance of each first time series, and the covariance between different first time series.

[0068] For example, for the element Cr, in 2024, there are first predicted values ​​1-10 and second content values ​​1-10 at collection points 1-10. The first predicted value 1 is subtracted from the first content value 1 to obtain the error value 1. Similarly, the error values ​​2-9 can be obtained. The error values ​​1-10 are organized into the first time series 1. Similarly, for the element Mn, the first time series 2 is obtained. The mean value 1 and variance 1 of the first time series 1, the mean value 2 and variance 2 of the first time series are calculated, and then the covariance 1 of the first time series 1 and the first time series 2, and the covariance 2 of the first time series 2 and the first time series 1 are calculated.

[0069] The distribution function of the error value is calculated based on the first time series. Based on the Monte Carlo method, a set of random number sequences is generated by combining the mean, variance and covariance of the distribution function. The target matrix is ​​generated based on the random number sequence, and the row vector of each row of the target matrix corresponds to the simulation value of different acquisition points at the second time point.

[0070] Here, a distribution function is constructed for the data of the same collection point for many years. The distribution function represents the frequency of occurrence of elements and the corresponding probability. Constructing a distribution function based on existing data is a prior art and will not be repeated here. After constructing the distribution function, a group of random numbers can be generated. The group of random numbers includes a random number sequence with the same number of collection points. For example, collection points 1 and 2 have first time series 1 and 2. After calculating the distribution function of the element, the average value, variance and covariance of the value according to the first time series 1 and 2, two random number sequences can be generated according to these parameters, specifically including random number sequences 1 and 2. This embodiment is generated by the Monte Carlo method. In random number sequences 1 and 2, the probability of occurrence of each element frequency, or the frequency of occurrence is close to the distribution function. For example, in the distribution function, the frequency of occurrence of the Cr element is 5, and the probability of occurrence is 0.1. In the generated random number sequence, the frequency of occurrence of the Cr element is 5, and the probability of occurrence is 0.11. Then, the target matrix is ​​generated according to the random number sequence, so the constructed simulation value, its distribution probability, average value and variance, and the relationship between the sequences are close to the previous error value.

[0071] In this embodiment, the estuary intersection information includes the water inflow, which is the annual water inflow into the river inlet that is less than the first predetermined distance from the collection point. The breeding information includes the annual amount of aquatic products released within the second predetermined distance range of the collection point, as well as the feed value released. The industrial activity information includes the annual emissions of various target heavy metals emitted by the target factory, and the target factory is a factory that is less than the third predetermined distance from the collection point. The human activity information is the number of tourists within the fourth predetermined distance range of the collection point.

[0072] The first to fourth predetermined distances can be set according to actual conditions, and their values ​​can be the same or different.

[0073] In this embodiment, obtaining the first data includes the following steps:

[0074] Columnar sediment cores were collected at the collection point, and sampling was performed at intervals of 0.5 cm to obtain multiple sediment samples. The formation year of the sediment samples was determined based on isotope dating technology, and the content values ​​of target heavy metals in the sediment samples were obtained based on atomic absorption and atomic fluorescence methods.

[0075] Dating using cesium-137 and lead-210 isotopes can be combined with historical data, industrial production time in the river area, meteorological records and sedimentation rates to further improve the accuracy of the calculated age of each sediment sample.

[0076] This embodiment evaluates the river ecological environment based on the geoaccumulation index method.

[0077] like Figure 2As shown, the present invention also provides an ecological safety assessment system based on heavy metal pollution in sediments, which is used to implement the above-mentioned ecological safety assessment method based on heavy metal pollution in sediments, and the system includes:

[0078] An analysis module, which determines a plurality of collection points and target heavy metals. The collection points in this embodiment include P1, P2 and P3. The collection points are collected and the sediment samples are analyzed to obtain first data. The first data includes the content values ​​of the target heavy metals at a plurality of historical time points.

[0079] A prediction module divides the historical time point into a first time point and a second time point, defines the content value corresponding to the first time point as the first content value, and defines the content value at the second time point as the second content value, obtains additional information, and the additional information includes estuary intersection information in the target area, and aquaculture information, industrial activity information, and human activity information at the historical time point, and establishes a prediction model. The prediction model predicts the first prediction value of the target heavy metal at different collection points at the second time point based on the first content value and the additional information;

[0080] An expansion module compares the second content value with the first predicted value, obtains an error value at a second time point, and generates a plurality of simulation values ​​based on the error value;

[0081] Evaluation module, the prediction model predicts the second predicted value of the target heavy metal at a third time point in the future based on the second content value and additional information, synthesizes the second predicted value with different simulation values ​​to generate a variety of prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results.

[0082] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0083] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0085] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An ecological safety assessment method based on heavy metal pollution in sediments, characterized in that: include: Determine a plurality of collection points and target heavy metals, collect samples at the collection points, and analyze sediment samples to obtain first data, wherein the first data includes content values ​​of the target heavy metal at a plurality of historical time points; Dividing the historical time point into a first time point and a second time point, the content value corresponding to the first time point is defined as a first content value, and the content value at the second time point is defined as a second content value; Acquiring additional information, the additional information including estuary confluence information within the target area, and aquaculture information, industrial activity information, and human activity information at the historical time point; Establishing a prediction model, wherein the prediction model predicts a first predicted value of the target heavy metal at different collection points at the second time point based on the first content value and the additional information; Comparing the second content value with the first predicted value, obtaining an error value at the second time point, and generating a plurality of simulation values ​​based on the error value; The prediction model predicts a second predicted value of the target heavy metal at a third time point in the future based on the second content value and the additional information, adds the same second predicted value to different simulation values ​​to obtain multiple calculation results, uses the calculation results as prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results; Generating the simulation value comprises the following steps: Generate multiple first time series, where the first time series are the error values ​​of the target heavy metal at different collection points at the same second time point, and calculate the mean and variance of each first time series, and the covariance between different first time series; The distribution function of the error value is calculated based on the first time series, and based on the Monte Carlo method, a set of random number sequences is generated in combination with the mean, variance and covariance of the distribution function, and a target matrix is ​​generated based on the random number sequence, wherein the row vector of each row of the target matrix corresponds to the simulation value of different acquisition points at the second time point.

2. The method according to claim 1, characterized in that Calculating the second predicted value comprises the following steps: Calculating first differences between the same target heavy metal content values ​​at the same collection point in adjacent years, clustering all the first differences to obtain a clustering result, wherein the clustering result includes a plurality of categories; Establish multiple prediction models based on the content value in each category and the additional information respectively, define the current time point as the fourth time point, and the previous time point of the fourth time point as the fifth time point, calculate the second difference between the fourth time point and the fifth time point, determine the first difference that is closest to the second difference value, and define it as a target difference, determine the prediction model to be used based on the target difference, and define it as a target model, and the target model calculates the second prediction value according to the content value at the fourth time point and the additional information.

3. The method according to claim 2, characterized in that Clustering the first differences comprises the following steps: Clustering the first differences into a first number, extracting a cluster center from each of the categories, calculating a third difference between the cluster center and other first differences in the same category, adding all the third differences to obtain a first score for the category, and if there is a first score greater than a first threshold, re-clustering the content values ​​into a second number, the second number being greater than the first number, and calculating the first score for each of the categories when clustered into the second number; If there are still cases where the first score is greater than the first threshold, then calculate the first average of all the first scores under the first number, and the second average of all the first scores under the second number, and continue to cluster the content values ​​into a third number, where the third number is greater than the second number, and repeat this step until there are no more cases where the first score is greater than the first threshold in the clustering result; A coordinate system is established with the number of clusters as the horizontal axis and the average value as the vertical axis, a curve showing the change of the average value with the number of clusters is plotted in the coordinate system, the curvature of each coordinate point in the curve is calculated, the number of clusters corresponding to the point with the largest curvature is taken as the final number of clusters, and the category under the final number of clusters is taken as the clustering result.

4. The method according to claim 1, characterized in that: The estuary intersection information includes water inflow, which is the annual water inflow into the river entrance that is less than the first predetermined distance from the collection point. The breeding information includes the annual amount of aquatic products released within the second predetermined distance range of the collection point, and the value of the feed released. The industrial activity information includes the annual emissions of various target heavy metals discharged by the target factory, and the target factory is a factory that is less than the third predetermined distance from the collection point. The human activity information is the number of tourists within the fourth predetermined distance range of the collection point.

5. The method according to claim 1, characterized in that: Obtaining the first data comprises the following steps: Columnar sediment cores are collected at the collection point, and sampling is performed at intervals of 0.5 cm to obtain a plurality of sediment samples. The formation year of the sediment samples is determined based on isotope dating technology, and the content value of the target heavy metal in the sediment samples is obtained based on atomic absorption and atomic fluorescence methods.

6. The method according to claim 1, characterized in that The river ecological environment is evaluated based on the geoaccumulation index method.

7. The method according to claim 1, characterized in that The prediction model is a linear regression model.

8. The method according to claim 1, characterized in that: The target heavy metals include Cr, Mn, Ni, Zn, As, Pb, Cu, and Co.

9. An ecological safety assessment system based on heavy metal pollution in sediments, used to implement the method according to any one of claims 1 to 8, characterized in that: include: An analysis module, determining a plurality of collection points and target heavy metals, collecting samples at the collection points, and analyzing sediment samples to obtain first data, wherein the first data includes content values ​​of the target heavy metal at a plurality of historical time points; A prediction module, dividing the historical time point into a first time point and a second time point, defining the content value corresponding to the first time point as a first content value, and defining the content value at the second time point as a second content value, acquiring additional information, the additional information including estuary intersection information in the target area, and aquaculture information, industrial activity information, and human activity information at the historical time point, establishing a prediction model, the prediction model predicting a first prediction value of the target heavy metal at different collection points at the second time point based on the first content value and the additional information; An extension module compares the second content value and the first predicted value, obtains the error value at the second time point, generates multiple simulation values ​​based on the error value, generates multiple first time series when generating the simulation value, and the first time series are the error values ​​of the target heavy metal at different collection points at the same second time point, calculates the mean and variance of each first time series, and the covariance between different first time series, calculates the distribution function of the error value based on the first time series, and generates a set of random number sequences based on the Monte Carlo method in combination with the mean, variance and covariance of the distribution function, generates a target matrix based on the random number sequence, and the row vector of each row of the target matrix corresponds to the simulation value of different collection points at the second time point; An evaluation module, wherein the prediction model predicts a second predicted value of the target heavy metal at a third time point in the future based on the second content value and the additional information, adds the same second predicted value to different simulation values ​​to obtain multiple calculation results, uses the calculation results as prediction results, and evaluates the river ecological environment of the collection point based on the second predicted value and the prediction results.

Images