Methods and systems for risk diagnosis and control of toxic effects of soil and groundwater pollutants

Abstract

The soil and groundwater pollutant toxic effect risk diagnosis and control method and system of the present application relate to the field of soil pollution analysis technology. By collecting emission data and toxicological parameters, calculating partial correlation coefficients, and constructing a pollutant causal relationship diagram; dividing pollutants into samples, adding toxic effect nodes, and extracting sub-causal relationship diagrams, determining the pollutant parent node set and the toxic effect parent node set; calculating the first conditional probability of the pollutant node, calculating the second conditional probability and the third conditional probability of the toxic effect node, and obtaining the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node; constructing a Bayesian network, obtaining the joint probability distribution of the pollutant combination, obtaining the emission concentration to form an evidence vector, and calculating the posterior probability distribution; obtaining the probability of the toxic effect node being in different states, taking the state corresponding to the maximum probability as the severity of the toxic effect, and selecting the pollutant node with the highest pollution weight for priority treatment.

Description

Technical Field

[0001] The present application relates to the technical field of soil pollution analysis, and in particular to a method and system for diagnosing and controlling the risk of toxic effects of soil and groundwater pollutants. Background Art

[0002] With changes in chemical production and social lifestyles, many new chemicals are being produced and used in large quantities. Some of these substances are environmentally persistent, bioaccumulative, and toxic. These new pollutants enter the soil and groundwater environment through atmospheric deposition, sewage irrigation, and solid waste landfills. Traditional environmental monitoring and risk assessment methods are unable to fully assess their hazards. Soil and groundwater are important natural resources. Once contaminated, remediation is difficult, time-consuming, and costly. Therefore, it is urgent to strengthen risk warning and pollution prevention and control.

[0003] A Chinese patent application with publication number CN118761628A discloses a method for prioritizing the management and evaluation of water environmental risks of toxic and hazardous pollutants, comprising the following steps: evaluating the toxic effects, detection frequency, and exposure concentration of each pollutant in the target environmental area: the toxic effects include but are not limited to ecotoxicity, health toxicity, and physicochemical properties; obtaining the risk quotient of pollutants in the target environmental area; and dividing pollutants into three levels based on the toxic effects, detection frequency, and exposure concentration of each pollutant, and the risk quotient of pollutants in the target environmental area.

[0004] Currently, there is a lack of a quantitative, probability-based approach to assessing the toxic effects of pollutant combinations. This approach primarily relies on toxicological experimental data and expert judgment, making it difficult to accurately estimate the probability of a pollutant combination causing a toxic effect under different conditions, hindering the accuracy of toxic risk assessments. Summary of the Invention

[0005] This application aims to solve, at least to some extent, one of the technical problems in the related art. To this end, one purpose of this application is to propose a method and system for diagnosing and controlling the risk of toxic effects of soil and groundwater pollutants, thereby achieving risk control of soil and groundwater pollution.

[0006] One aspect of the present application provides a method for diagnosing and controlling the risk of toxic effects of soil and groundwater pollutants, including:

[0007] Step S100: collecting pollutant emission data and toxicological parameters of N pollution sources, calculating partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients;

[0008] Step S200: For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples based on the emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship graph, and sub-causal relationship graphs are extracted. The pollutant parent node set and the toxic effect parent node set are determined based on the sub-causal relationship graphs;

[0009] Step S300: Calculate the first conditional probability of the pollutant node based on the sample, calculate the second conditional probability and the third conditional probability of the toxic effect node, and obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S;

[0010] Step S400: Constructing a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtaining the joint probability distribution of the pollutant combination based on the Bayesian network, obtaining the current emission concentration of the pollutant combination to form an evidence vector, and calculating the posterior probability distribution based on the evidence vector;

[0011] Step S500: Obtain the probability of the toxic effect node being in different states based on the posterior probability distribution, and take the state corresponding to the maximum probability as the current severity of the toxic effect. When the state is greater than the state threshold, select the pollutant node with the highest pollution weight in the pollutant combination S for priority treatment.

[0012] The specific method of collecting pollutant emission data and toxicological parameters of N pollution sources, calculating the partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients is as follows:

[0013] Step S110: Set the statistical interval t1, collect the emission data of pollutants from N pollution sources within n1 statistical intervals, and construct the emission data matrix X for each statistical interval, where the emission data of the jth pollutant from the ath pollution source is x aj ;

[0014] Step S120: For each statistical interval, collecting toxicological parameters of the pollution source;

[0015] Step S130: For the statistical interval closest to the current time, subtract the mean of all elements in each column from each element in the emission data matrix of the statistical interval to obtain the centralization matrix X c , calculate the covariance matrix X of the centered matrix f , perform the inverse operation on the covariance matrix to obtain the precision matrix X p , calculate the partial correlation coefficient r of pollutant pair (i, j) based on the precision matrix ij ;

[0016] Step S140: setting a partial correlation coefficient threshold, determining that the pollutant pair whose partial correlation coefficient is greater than the partial correlation coefficient threshold has a significant partial correlation coefficient, and performing causal relationship judgment on the pollutant pair based on the toxicological parameters;

[0017] Step S150: Based on the directed edges and undirected edges between pollutants, a pollutant causal relationship graph is constructed. The nodes in the pollutant causal relationship graph represent pollutants, the edges represent causal relationships, and the weights of the edges are the partial correlation coefficients between pollutants.

[0018] The specific method for determining the causal relationship between the pollutants based on toxicological parameters is as follows:

[0019] Step S141: Obtain the first emission time of pollutant i and pollutant j. When the first emission time of pollutant i is earlier than that of pollutant j, and the first emission time difference between pollutant i and pollutant j is |T j -T i | is greater than the time threshold Δt, then pollutant i is determined to be the cause of pollutant j, and a directed edge i→j is added; where T i represents the first emission time of pollutant i, T j represents the first emission time of pollutant j;

[0020] Step S142: If pollutant j is a metabolite of pollutant i, then pollutant i is a cause of pollutant j, and a directed edge i→j is added;

[0021] Step S143: If pollutant i and pollutant j come from the same pollution source, there is a common cause relationship between pollutant i and pollutant j, and an undirected edge i-j is added.

[0022] For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples by combining emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship graph, and sub-causal relationship graphs are extracted. The specific method for determining the pollutant parent node set and the toxic effect parent node set based on the sub-causal relationship graph is as follows:

[0023] Step S210: Determine the pollutant combination S to be studied in the nth pollution source, the pollutant combination includes pollutants S1, S2, ..., S k , k is the number of pollutants in the pollutant combination; the pollutants appearing in each statistical interval are determined based on the emission data, and all pollutants appearing in each statistical interval and the corresponding emission data and toxicological parameters are taken as a sample and sorted in chronological order to obtain n1 samples; wherein, the emission data includes emission concentration, and when the emission concentration of a pollutant is greater than 0, it indicates that the pollutant is present;

[0024] Step S220: Add a toxic effect node Toxicity to the pollutant causal relationship diagram. For each sample, when the toxicological parameter of the sample is greater than the toxicity threshold, it means that the sample has a synergistic toxic effect. k Directed edges are drawn to the toxic effect nodes, and a sub-causal relationship graph is formed by the pollutant nodes and the toxic effect nodes;

[0025] Step S230: Determine the toxic effect parent node set pa (Toxicity) and the pollutant parent node set pa (S) in each sample according to the child causal relationship graph. i ).

[0026] The specific method for calculating the first conditional probability of the pollutant node based on the sample, calculating the second conditional probability and the third conditional probability of the toxic effect node, and obtaining the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S is as follows:

[0027] Step S310: In each statistical interval, if the pollutant node S in the pollutant combination i With the given pollutant parent node set pa(S i ) appear at the same time, the pollutant node S i The state in the statistical interval is recorded as 1, otherwise it is recorded as 0. The frequency N (S) of the state being 1 in all statistical intervals is counted. i ,pa(S i )), count the pollutant parent node set pa(S in all statistical intervals i ) appears at a frequency of N(pa(S i )), divide the frequency of state 1 in all statistical intervals by the pollutant parent node set pa(S i ) appears, and the pollutant node S is obtained i In a given pollutant parent node set pa(S i ) under the first conditional probability P(S i |pa(S i ));

[0028] Step S320: From pollutant node S i The first conditional probability under different pollutant parent node sets is used to obtain the pollutant node S i Conditional probability table of ;

[0029] Step S330: Count the frequencies of the parent node sets of the toxic effect nodes satisfying the given toxic effect parent node set pa(Toxicity) in all statistical intervals, recorded as N(pa(Toxicity)), and count the frequencies of the occurrence of toxic effects in the statistical intervals satisfying the given toxic effect parent node set, recorded as N(Toxicity, pa(Toxicity)), and divide the frequencies of the occurrence of toxic effects by the frequencies satisfying the given toxic effect parent node set to obtain the second conditional probability P(Toxicity|pa(Toxicity)) of the toxic effect node under the given toxic effect parent node set;

[0030] Step S340: Set the state of the toxicity effect node to {State 1, State 2, ..., State y}, with a total of y states. Count the frequency of each state occurring in the statistical interval that satisfies the given set of toxicity effect parent nodes, recorded as N(state, pa(Toxicity));

[0031] Step S350: Divide the frequency of each state by the frequency of satisfying a given toxic effect parent node set to obtain the third conditional probability P(state|pa(Toxicity)) of each state of the toxic effect node under the given toxic effect parent node set;

[0032] Step S360: A complete conditional probability table of the toxic effect node is constructed by the second conditional probability of the toxic effect node under different toxic effect parent node sets and the third conditional probability of each state of the toxic effect node under different toxic effect parent node sets.

[0033] The Bayesian network is constructed based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node. The joint probability distribution of the pollutant combination is obtained according to the Bayesian network. The current emission concentration of the pollutant combination is obtained to form an evidence vector. The specific method for calculating the posterior probability distribution based on the evidence vector is as follows:

[0034] Step S410: Corresponding the pollutant nodes and toxic effect nodes in the sub-causal relationship graph to each node in the Bayesian network, adding corresponding directed edges in the Bayesian network according to the directed edges in the sub-causal relationship graph, and using the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node as parameters of the corresponding node in the Bayesian network;

[0035] Step S420: Combine the conditional probability tables of each pollutant node according to the Bayesian network to obtain the joint probability distribution P (Toxicity, S1, S2, ..., S k );

[0036] Step S430: Obtain the emission concentration of each pollutant in the current statistical interval, and form the emission concentration into an evidence vector E = (e1, e2, ..., e m ), substitute into the joint probability distribution, and calculate the probability P(E) of the evidence vector E; where e m It represents the emission concentration of the mth pollutant observed in the current statistical interval, where m is the number of pollutant types observed in the current statistical interval;

[0037] Step S440: Calculate the posterior probability distribution P(Toxicity|E) of the pollutant combination S under the given evidence vector E based on Bayesian network reasoning.

[0038] The probability of the toxic effect node being in different states is obtained according to the posterior probability distribution, and the state corresponding to the maximum probability is used as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment. The specific method is:

[0039] Step S510: Obtain the probability of the toxic effect node being in each state according to the posterior probability distribution;

[0040] Step S520: Find the state with the highest probability, and use the value of this state as the current severity of the toxic effect. When the severity of the toxic effect is greater than the state threshold, it indicates that the current toxic effect requires pollutant control.

[0041] Step S530: Obtain the sum of the weights of the edges of the shortest path between each pollutant node and the toxic effect node, and calculate the pollution weight of each pollutant node in the pollutant combination in combination with the current emission concentration of the pollutant;

[0042] Step S540: The pollutant node with the highest pollution weight is selected as the pollutant to be treated first.

[0043] One aspect of the present application provides a risk diagnosis and control system for the toxic effects of soil and groundwater pollutants, including:

[0044] The pollution data collection module is used to collect pollutant emission data and toxicological parameters of N pollution sources, calculate the partial correlation coefficients between pollutants, and construct a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients;

[0045] The parent node determination module is used to divide the pollutants into samples based on the emission data and toxicological parameters for the pollutant combination S to be studied in the nth pollution source, add toxic effect nodes to the pollutant causal relationship graph, extract the child causal relationship graph, and determine the pollutant parent node set and the toxic effect parent node set based on the child causal relationship graph;

[0046] A conditional probability table determination module is used to calculate the first conditional probability of the pollutant node based on the sample, calculate the second conditional probability and the third conditional probability of the toxic effect node, and obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S;

[0047] The posterior probability calculation module is used to construct a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtain the joint probability distribution of the pollutant combination based on the Bayesian network, obtain the current emission concentration of the pollutant combination to form an evidence vector, and calculate the posterior probability distribution based on the evidence vector;

[0048] The pollution weight calculation module is used to obtain the probability of the toxic effect node being in different states based on the posterior probability distribution, and take the state corresponding to the maximum probability as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment.

[0049] One aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for diagnosing and controlling the risk of toxic effects of soil and groundwater pollutants are implemented.

[0050] One aspect of the present application provides a readable storage medium storing a computer program suitable for loading by a processor to execute the steps in a method for diagnosing and controlling the risk of toxic effects of soil and groundwater pollutants.

[0051] The soil and groundwater pollutant toxicity risk diagnosis and control method and system proposed in this application have the following advantages over existing technologies:

[0052] This application constructs a pollutant causal relationship diagram to clarify the causal relationship between pollutants, which helps to understand the toxicity mechanism of complex pollutant combinations and provides a basis for formulating targeted pollution prevention and control strategies.

[0053] This application introduces toxic effect nodes to intuitively represent the impact of pollutant combinations on the environment and health, quantify toxicity risks, and provide a new approach for assessing the degree of harm of pollutant combinations. It also divides the pollutant combination of a pollution source into sub-causal relationship graphs, reducing the impact of irrelevant factors.

[0054] This application uses Bayesian networks for reasoning, which can dynamically evaluate the toxicity risk of pollutant combinations based on real-time monitored pollutant emission data, and provide support for the formulation of dynamically adjusted pollution prevention and control strategies.

[0055] The pollution weight indicators in this application provide a quantitative basis for determining priority pollutants for treatment. By prioritizing the control of pollutants with high pollution weights, the overall toxicity risk of the pollutant combination can be minimized, achieving optimal allocation of pollution prevention and control resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flow chart of the method for risk diagnosis and control of toxic effects of soil and groundwater pollutants provided for this application;

[0057] Figure 2 Sub-causal relationship diagram of synergistic toxic effects for the risk diagnosis and control methods of toxic effects of soil and groundwater pollutants provided in this application;

[0058] Figure 3 Flowchart of the calculation method of the posterior probability distribution provided in this application;

[0059] Figure 4 Functional module diagram of the soil and groundwater pollutant toxic effect risk diagnosis and control system provided for this application. DETAILED DESCRIPTION

[0060] To better understand the present application, various aspects of the present application will be described in more detail with reference to the accompanying drawings. It should be understood that these detailed descriptions are merely descriptions of exemplary embodiments of the present application and are not intended to limit the scope of the present application in any way. Throughout the specification, the same reference numerals refer to the same elements. The expression "and / or" includes any and all combinations of one or more of the associated listed items.

[0061] In the accompanying drawings, the size, dimensions, and shapes of the elements have been slightly adjusted for ease of illustration. The accompanying drawings are for illustration only and are not drawn strictly to scale. As used herein, the terms "substantially," "approximately," and similar terms are used to indicate approximate values, not degrees, and are intended to illustrate inherent deviations in measurements or calculations that would be recognized by a person of ordinary skill in the art. In addition, in this application, the order in which the steps are described does not necessarily represent the order in which these steps would occur in actual operation, unless otherwise specified or inferred from the context.

[0062] It should also be understood that expressions such as "including", "comprising", "having", "containing" and / or "comprising" are open rather than closed expressions in this specification, which indicate the presence of the stated features, elements and / or components, but do not exclude the presence of one or more other features, elements, components and / or combinations thereof. In addition, when expressions such as "at least one of..." appear after a list of listed features, they modify the entire list of features rather than just the individual elements in the list. In addition, when describing embodiments of the present application, "may" is used to mean "one or more embodiments of the present application". And, the term "exemplary" is intended to refer to an example or illustration.

[0063] Unless otherwise defined, all words used herein (including engineering terms and scientific and technological terms) have the same meaning as commonly understood by those skilled in the art to which this application belongs. It should also be understood that, unless otherwise specified in this application, words defined in commonly used dictionaries should be interpreted as having the same meaning as they do in the context of the relevant technology, and should not be interpreted in an idealized or overly formal sense.

[0064] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0065] Example 1

[0066] like Figure 1 As shown, the risk diagnosis and control methods for the toxic effects of soil and groundwater pollutants provided in this application include:

[0067] Step S100: collecting pollutant emission data and toxicological parameters of N pollution sources, calculating partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients;

[0068] The emission data refers to the concentration of each pollutant from each pollution source over a period of time. The emission data and toxicological parameters are data collected at each statistical interval over a historical period.

[0069] The pollutants include pollutants directly discharged from pollution sources and new pollutants generated by metabolism between pollutants.

[0070] The specific method of collecting pollutant emission data and toxicological parameters of N pollution sources, calculating the partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients is as follows:

[0071] Step S110: Set the statistical interval t1, collect the emission data of pollutants from N pollution sources within n1 statistical intervals, and construct the emission data matrix X for each statistical interval, where the emission data of the jth pollutant from the ath pollution source is x aj ;

[0072] Step S120: For each statistical interval, collecting toxicological parameters of the pollution source;

[0073] Step S130: For the statistical interval closest to the current time, subtract the mean of all elements in each column from each element in the emission data matrix of the statistical interval to obtain a centering matrix, calculate the covariance matrix of the centering matrix, invert the covariance matrix to obtain a precision matrix, and calculate the partial correlation coefficient of the pollutant pair (i, j) based on the precision matrix;

[0074] The calculation formula of the centralization matrix is: Among them, μ represents the mean of all elements in each column, and T represents the transpose symbol;

[0075] The calculation formula of the covariance matrix is:

[0076] The calculation formula of the precision matrix is:

[0077] The calculation formula of the partial correlation coefficient of the pollutant pair (i, j) is: Among them, p ij is the element in row i and column j of the precision matrix, p ii is the element in row i and column i of the precision matrix, p jj is the element in row j and column j of the precision matrix;

[0078] The partial correlation coefficient is used to describe the correlation between pollutants after controlling the influence of other pollutants;

[0079] The statistical interval closest to the current time is used as the statistical interval for calculating the partial correlation coefficient because this statistical interval is closest to the real-time pollution situation and toxic effects.

[0080] Step S140: setting a partial correlation coefficient threshold, determining that the pollutant pair whose partial correlation coefficient is greater than the partial correlation coefficient threshold has a significant partial correlation coefficient, and performing causal relationship judgment on the pollutant pair based on the toxicological parameters;

[0081] The partial correlation coefficient threshold is set by those skilled in the art according to actual needs;

[0082] The specific method for determining the causal relationship between the pollutants based on toxicological parameters is as follows:

[0083] Step S141: Obtain the first emission time of pollutant i and pollutant j. When the first emission time of pollutant i is earlier than that of pollutant j, and the first emission time difference between pollutant i and pollutant j is |T j -T i | is greater than the time threshold Δt, then pollutant i is determined to be the cause of pollutant j, and a directed edge i→j is added; where T i represents the first emission time of pollutant i, T j represents the first emission time of pollutant j;

[0084] The time threshold Δt is set by those skilled in the art according to actual needs;

[0085] Step S142: If pollutant j is a metabolite of pollutant i, then pollutant i is a cause of pollutant j, and a directed edge i→j is added;

[0086] Step S143: If pollutant i and pollutant j come from the same pollution source, there is a common cause relationship between pollutant i and pollutant j, and an undirected edge i-j is added;

[0087] When there is a bidirectional causal relationship between pollutant i and pollutant j, a bidirectional edge When pollutants i and j satisfy both the causal relationship of step S141 and step S142 and the common cause relationship of step S143, the directed edge of the causal relationship is added first, and the undirected edge is added on the basis of the directed edge.

[0088] Step S150: Based on the directed edges and undirected edges between pollutants, a pollutant causal relationship graph is constructed. The nodes in the pollutant causal relationship graph represent pollutants, the edges represent causal relationships, and the weights of the edges are the partial correlation coefficients between pollutants.

[0089] The above steps collect emission data and toxicological parameters of pollution sources to provide a data basis for subsequent analysis, calculate the partial correlation coefficients between pollutants to quantify the correlation between pollutants, and construct a pollutant causal relationship diagram based on toxicological parameters and partial correlation coefficients to reveal the causal relationship between pollutants. It provides a framework for adding toxic effect nodes, extracting sub-causal relationship diagrams, and constructing Bayesian networks.

[0090] Step S200: For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples based on the emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship graph, and sub-causal relationship graphs are extracted. The pollutant parent node set and the toxic effect parent node set are determined based on the sub-causal relationship graphs;

[0091] The toxic effect node refers to the node of the synergistic toxic effect caused by the combination of pollutants;

[0092] For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples by combining emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship graph, and sub-causal relationship graphs are extracted. The specific method for determining the pollutant parent node set and the toxic effect parent node set based on the sub-causal relationship graph is as follows:

[0093] Step S210: Determine the pollutant combination S to be studied in the nth pollution source, the pollutant combination includes pollutants S1, S2, ..., S k , k is the number of pollutants in the pollutant combination; the pollutants appearing in each statistical interval are determined based on the emission data, and all pollutants appearing in each statistical interval and the corresponding emission data and toxicological parameters are taken as a sample and sorted in chronological order to obtain n1 samples; wherein, the emission data includes emission concentration, and when the emission concentration of a pollutant is greater than 0, it indicates that the pollutant is present;

[0094] Since the total number of statistical intervals is n1, the samples are divided according to the statistical intervals, so the total number of samples obtained is also n1;

[0095] Step S220: Add a toxic effect node Toxicity to the pollutant causal relationship diagram. For each sample, when the toxicological parameter of the sample is greater than the toxicity threshold, it means that the sample has a synergistic toxic effect. k Directed edges are drawn to the toxic effect nodes, and a sub-causal relationship graph is formed by the pollutant nodes and the toxic effect nodes;

[0096] Optionally, if there are other pollutants in the pollutant causal relationship graph that affect the toxic effect of the pollutant, then the other pollutant nodes are added to the sub-causal relationship graph;

[0097] Step S230: Determine the toxic effect parent node set pa (Toxicity) and the pollutant parent node set pa (S) in each sample according to the child causal relationship graph. i ).

[0098] The above steps divide the pollutants into different samples according to the pollutant combination to be studied, combined with emission data and toxicological parameters, add toxic effect nodes to the pollutant causal relationship diagram, connect the pollutant nodes with the toxic effect nodes, extract the sub-causal relationship diagram containing the pollutant nodes and the toxic effect nodes, and determine the pollutant parent node set and the toxic effect parent node set based on the sub-causal relationship diagram. Dividing the samples can analyze the impact of different pollutant combinations in a targeted manner; adding toxic effect nodes can associate pollutants with toxic effects; extracting the sub-causal relationship diagram can simplify the analysis scope; and determining the parent node set provides a basis for calculating conditional probability.

[0099] Step S300: Calculate the first conditional probability of the pollutant node based on the sample, calculate the second conditional probability and the third conditional probability of the toxic effect node, and obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S;

[0100] The specific method for calculating the first conditional probability of the pollutant node based on the sample, calculating the second conditional probability and the third conditional probability of the toxic effect node, and obtaining the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S is as follows:

[0101] Step S310: In each statistical interval, if the pollutant node S in the pollutant combination i With the given pollutant parent node set pa(S i ) appear at the same time, the pollutant node S i The state in the statistical interval is recorded as 1, otherwise it is recorded as 0. The frequency N (S) of the state being 1 in all statistical intervals is counted. i ,pa(S i )), count the pollutant parent node set pa(S in all statistical intervals i ) appears at a frequency of N(pa(S i )), divide the frequency of state 1 in all statistical intervals by the pollutant parent node set pa(S i ) appears, and the pollutant node S is obtained i In a given pollutant parent node set pa(S i ) under the first conditional probability;

[0102] The pollutant node S i In a given pollutant parent node set pa(S i The calculation formula of the first conditional probability under ) is:

[0103] Step S320: From pollutant node S i The first conditional probability under different pollutant parent node sets is used to obtain the pollutant node S i Conditional probability table of ;

[0104] Step S330: Count the frequencies of the parent node set of the toxic effect node satisfying the given toxic effect parent node set pa(Toxicity) in all statistical intervals, recorded as N(pa(Toxicity)), and count the frequencies of the toxic effect occurring in the statistical intervals satisfying the given toxic effect parent node set, recorded as N(Toxicity, pa(Toxicity)), and divide the frequencies of the toxic effect occurring by the frequencies satisfying the given toxic effect parent node set to obtain the second conditional probability of the toxic effect node under the given toxic effect parent node set;

[0105] The calculation formula for the second conditional probability of the toxic effect node under a given toxic effect parent node set is:

[0106] Step S340: Set the state of the toxicity effect node to {State 1, State 2, ..., State y}, with a total of y states. Count the frequency of each state occurring in the statistical interval that satisfies the given set of toxicity effect parent nodes, recorded as N(state, pa(Toxicity));

[0107] Preferably, the status of the toxic effect node is divided into four states: 4, 3, 2, and 1, corresponding to severe, moderate, mild, and non-toxicity, respectively;

[0108] Step S350: Divide the frequency of each state by the frequency of satisfying a given set of toxic effect parent nodes to obtain a third conditional probability of each state of the toxic effect node under the given set of toxic effect parent nodes;

[0109] The calculation formula of the third conditional probability is:

[0110] Step S360: A complete conditional probability table of the toxic effect node is constructed by the second conditional probability of the toxic effect node under different toxic effect parent node sets and the third conditional probability of each state of the toxic effect node under different toxic effect parent node sets.

[0111] The above steps are used to calculate the conditional probability table, which quantifies the probability distribution of pollutant nodes and toxic effect nodes under different conditions and is an important parameter for constructing the Bayesian network.

[0112] For example, Figure 2As shown, this is a sub-causal relationship graph of the synergistic toxic effect provided by this application. The probability of the synergistic toxic effect of the pollutant combination A, B, and C is studied. The toxic effect node Toxicity is added to the pollutant causal relationship graph, and directed edges are drawn from the pollutant combinations A, B, and C to the toxic effect node. From the pollutant causal relationship graph, it is known that the metabolism between pollutants B and C will produce pollutant D. Therefore, pollutant D, the pollutant combinations A, B, and C, and the toxic effect node Toxicity constitute a sub-causal relationship graph;

[0113] According to the chain rule of Bayesian networks, we can obtain P(Toxicity,A,B,C,D)=P(Toxicity|D)×P(D|B,C)×P(B|A)×P(C|A)×P(A). Given the value of the parent node, the child node is independent of its non-descendant nodes.

[0114] Step S400: Constructing a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtaining the joint probability distribution of the pollutant combination based on the Bayesian network, obtaining the current emission concentration of the pollutant combination to form an evidence vector, and calculating the posterior probability distribution based on the evidence vector;

[0115] The Bayesian network is constructed based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node. The joint probability distribution of the pollutant combination is obtained according to the Bayesian network. The current emission concentration of the pollutant combination is obtained to form an evidence vector. The specific method for calculating the posterior probability distribution based on the evidence vector is as follows:

[0116] Step S410: Corresponding the pollutant nodes and toxic effect nodes in the sub-causal relationship graph to each node in the Bayesian network, adding corresponding directed edges in the Bayesian network according to the directed edges in the sub-causal relationship graph, and using the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node as parameters of the corresponding node in the Bayesian network;

[0117] The nodes in the Bayesian network are the pollutant nodes and toxic effect nodes in the sub-causal relationship graph;

[0118] Step S420: Combine the conditional probability tables of each pollutant node according to the Bayesian network to obtain the joint probability distribution P (Toxicity, S1, S2, ..., S k );

[0119] The conditional probability table of each pollutant node is combined according to the Bayesian network to obtain the calculation formula of the joint probability distribution of the pollutant combination S: P(Toxicity, S1, S2, ..., S k )=P(Toxicity|S1,S2,...,Sk )×∏ i P(S i |pa(S i )), where P(Toxicity|S1,S2,...,S k ) represents the conditional probability of the toxic effect node, P(S i |pa(S i )) is the pollutant node S i The first conditional probability, pa(S i ) represents the pollutant parent node set;

[0120] The conditional probability of the toxic effect node is when the given toxic effect parent node set pa (Toxicity) is (S1, S2, ..., S k ) when the second conditional probability;

[0121] The joint probability distribution can be decomposed into the product of the conditional probabilities of each node, obtained from the conditional probability tables of the pollutant nodes and the toxic effect nodes calculated in step S300;

[0122] The Bayesian network outputs the conditional probability value of each pollutant node decomposed in the joint probability distribution, which is substituted into the calculation formula of the joint probability distribution to calculate the value of the joint probability distribution of the pollutant combination S.

[0123] Step S430: Obtain the emission concentration of each pollutant in the current statistical interval, and form the emission concentration into an evidence vector E = (e1, e2, ..., e m ), substitute into the joint probability distribution and calculate the probability of the evidence vector E; where e m It represents the emission concentration of the mth pollutant observed in the current statistical interval, where m is the number of pollutant types observed in the current statistical interval;

[0124] The calculation formula of the probability of the evidence vector E is:

[0125] The evidence vector refers to the emission concentration of each pollutant in the current statistical interval;

[0126] Step S440: Calculating the posterior probability distribution P(Toxicity|E) of the pollutant combination S under the given evidence vector E based on Bayesian network reasoning;

[0127] The calculation formula of the posterior probability distribution is: Among them, P(Toxicity, E) represents the joint probability of the toxic effect node and the evidence vector occurring simultaneously;

[0128] The posterior probability distribution represents the state distribution of the toxic effect node under the condition that the evidence vector E is known.

[0129] Figure 3 Flowchart of the calculation method of the posterior probability distribution provided in this application.

[0130] The Bayesian network in the above steps can comprehensively consider the causal relationship and conditional probability between pollutants to calculate the joint probability distribution; the posterior probability distribution can update the state probability of the toxic effect node according to the current evidence vector, providing a basis for evaluating the severity of the toxic effect.

[0131] Step S500: Obtain the probability of the toxic effect node being in different states based on the posterior probability distribution, and take the state corresponding to the maximum probability as the current toxic effect severity. When the state is greater than the state threshold, select the pollutant node with the highest pollution weight in the pollutant combination S for priority treatment;

[0132] The probability of the toxic effect node being in different states is obtained according to the posterior probability distribution, and the state corresponding to the maximum probability is used as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment. The specific method is:

[0133] Step S510: Obtain the probability of the toxic effect node being in each state according to the posterior probability distribution;

[0134] Step S520: Find the state with the highest probability, and use the value of this state as the current severity of the toxic effect. When the severity of the toxic effect is greater than the state threshold, it indicates that the current toxic effect requires pollutant control.

[0135] The state threshold is set by those skilled in the art according to actual needs;

[0136] Step S530: Obtain the sum of the weights of the edges of the shortest path between each pollutant node and the toxic effect node, and calculate the pollution weight of each pollutant node in the pollutant combination in combination with the current emission concentration of the pollutant;

[0137] The calculation formula of the pollution weight of each pollutant node is: Among them, θ, β, γ are adjustment parameters, sumRi represents the pollutant node S i The sum of the weights of the shortest paths to the toxic effect nodes, V i Represents the pollutant node S i Current emission concentrations;

[0138] The adjustment parameters are set to appropriate adjustment parameter values based on the experience of field experts and data analysis results;

[0139] Step S540: The pollutant node with the highest pollution weight is selected as the pollutant to be treated first.

[0140] By comparing the posterior probabilities in the above steps, the severity of the current toxic effect can be determined; combined with the pollution weights, the pollutants with the greatest impact on the toxic effect can be determined, and targeted treatment plans can be formulated.

[0141] Example 2

[0142] like Figure 4 As shown in the figure, the soil and groundwater pollutant toxicity risk diagnosis and control system provided in this application includes:

[0143] The pollution data collection module is used to collect pollutant emission data and toxicological parameters of N pollution sources, calculate the partial correlation coefficients between pollutants, and construct a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients;

[0144] The parent node determination module is used to divide the pollutants into samples based on the emission data and toxicological parameters for the pollutant combination S to be studied in the nth pollution source, add toxic effect nodes to the pollutant causal relationship graph, extract the child causal relationship graph, and determine the pollutant parent node set and the toxic effect parent node set based on the child causal relationship graph;

[0145] A conditional probability table determination module is used to calculate the first conditional probability of the pollutant node based on the sample, calculate the second conditional probability and the third conditional probability of the toxic effect node, and obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S;

[0146] The posterior probability calculation module is used to construct a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtain the joint probability distribution of the pollutant combination based on the Bayesian network, obtain the current emission concentration of the pollutant combination to form an evidence vector, and calculate the posterior probability distribution based on the evidence vector;

[0147] The pollution weight calculation module is used to obtain the probability of the toxic effect node being in different states based on the posterior probability distribution, and take the state corresponding to the maximum probability as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment.

[0148] Example 3

[0149] According to another aspect of the present application, an electronic device is provided. The electronic device may include one or more processors and one or more memories. The memories may store computer-readable code that, when executed by the one or more processors, may implement the above-described soil and groundwater contaminant toxicity risk diagnosis and control method.

[0150] The method or system according to the embodiment of the present application can also be implemented with the help of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. The storage device in the electronic device, such as a ROM or a hard disk, can store the soil and groundwater pollutant toxic effect risk diagnosis and control method provided in this application. The soil and groundwater pollutant toxic effect risk diagnosis and control method may, for example, include: collecting pollutant emission data and toxicological parameters from N pollution sources, calculating the partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and the partial correlation coefficients; for the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples based on the emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship diagram, and the sub-causal relationship diagram is extracted, and the pollutant parent node set and the toxic effect parent node set are determined according to the sub-causal relationship diagram; the first conditional probability of the pollutant node is calculated based on the sample, and the second conditional probability and third conditional probability of the toxic effect node are calculated. Conditional probability, obtain the conditional probability table of each pollutant node in the pollutant combination S and the conditional probability table of the toxic effect node; construct a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtain the joint probability distribution of the pollutant combination according to the Bayesian network, obtain the current emission concentration of the pollutant combination to form an evidence vector, and calculate the posterior probability distribution based on the evidence vector; obtain the probability of the toxic effect node in different states according to the posterior probability distribution, and take the state corresponding to the maximum probability as the current severity of the toxic effect. When the state is greater than the state threshold, select the pollutant node with the highest pollution weight in the pollutant combination S for priority treatment. Furthermore, the electronic device may also include a user interface. Of course, when implementing different devices, one or more components in the electronic device can be omitted according to actual needs.

[0151] Example 4

[0152] According to one embodiment of the present application, a readable storage medium is provided. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, the soil and groundwater pollutant toxic effect risk diagnosis and control method according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0153] In addition, according to the implementation mode of the present application, the process described in the above reference flowchart can be implemented as a computer software program. For example, the present application provides a non-temporary machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be run by a processor to execute instructions corresponding to the method steps provided in the present application, such as: collecting pollutant emission data and toxicological parameters of N pollution sources, calculating the partial correlation coefficients between pollutants, and constructing a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients; for the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples based on the emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship diagram, and the sub-causal relationship diagram is extracted, and the pollutant parent node set and the toxic effect parent node set are determined according to the sub-causal relationship diagram; the pollutant is calculated according to the sample The first conditional probability of the pollutant node is calculated, and the second and third conditional probabilities of the toxic effect node are calculated to obtain a conditional probability table for each pollutant node in the pollutant combination S and a conditional probability table for the toxic effect node; a Bayesian network is constructed based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, and a joint probability distribution of the pollutant combination is obtained based on the Bayesian network. The current emission concentration of the pollutant combination is obtained to form an evidence vector, and a posterior probability distribution is calculated based on the evidence vector; the probability of the toxic effect node being in different states is obtained based on the posterior probability distribution, and the state corresponding to the maximum probability is used as the current toxic effect severity. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment. When the computer program is executed by the central processing unit (CPU), it performs the above functions defined in the method of this application.

[0154] The methods, apparatus, and devices of the present application may be implemented in many ways. For example, the methods, apparatus, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the methods according to the present application. Therefore, the present application also covers recording media that store programs for executing the methods according to the present application.

[0155] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0156] The above-described specific embodiments further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is merely a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for diagnosing and controlling the toxic effects of soil and groundwater pollutants, characterized in that: include: Collect pollutant emission data and toxicological parameters from N pollution sources, calculate the partial correlation coefficients between pollutants, and construct a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients. The specific method is as follows: set the statistical interval t1, collect the emission data of pollutants from N pollution sources within n1 statistical intervals, and for each statistical interval, construct the emission data matrix X, where the emission data of the jth pollutant from the ath pollution source is ; For each statistical interval, toxicological parameters of the pollution sources were collected; For the statistical interval closest to the current time, subtract the mean of all elements in each column from each element in the emission data matrix of the statistical interval to obtain the central matrix , calculate the covariance matrix of the centered matrix , perform the inverse operation on the covariance matrix to obtain the precision matrix , calculate the partial correlation coefficient of pollutant pair (i, j) based on the precision matrix ; A partial correlation coefficient threshold is set. Pollutant pairs with partial correlation coefficients greater than the threshold are judged to have significant partial correlation coefficients, and the causal relationship of the pollutant pairs is determined based on toxicological parameters. Based on the directed and undirected edges between pollutants, a pollutant causal relationship graph is constructed. The nodes in the pollutant causal relationship graph represent pollutants, the edges represent causal relationships, and the edge weights are the partial correlation coefficients between pollutants. The specific method for determining the causal relationship between the pollutants based on toxicological parameters is as follows: Get the first emission time of pollutant i and pollutant j. When the first emission time of pollutant i is earlier than that of pollutant j, and the first emission time difference between pollutant i and pollutant j is Greater than the time threshold , then it is determined that pollutant i is the cause of pollutant j, and a directed edge i→j is added; where, represents the first emission time of pollutant i, represents the first emission time of pollutant j; If pollutant j is a metabolite of pollutant i, then pollutant i is the cause of pollutant j, and a directed edge i→j is added; If pollutant i and pollutant j come from the same pollution source, there is a common cause relationship between pollutant i and pollutant j, and an undirected edge i-j is added; For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples based on the emission data and toxicological parameters. The toxic effect nodes are added to the pollutant causal relationship graph, and the sub-causal relationship graph is extracted. The pollutant parent node set and the toxic effect parent node set are determined based on the sub-causal relationship graph; According to the sample, the first conditional probability of the pollutant node is calculated, and the second conditional probability and the third conditional probability of the toxic effect node are calculated to obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S. The specific method is: in each statistical interval, if the pollutant node in the pollutant combination With the given pollutant parent node set If they appear at the same time, the pollutant node The state within the statistical interval is recorded as 1, otherwise it is recorded as 0, and the frequency of the state being 1 in all statistical intervals is counted. , count the pollutant parent node sets in all statistical intervals Frequency of occurrence , divide the frequency of state 1 in all statistical intervals by the set of pollutant parent nodes The frequency of occurrence, get the pollutant node In a given pollutant parent node set The first conditional probability under ; By pollutant node The first conditional probability under different pollutant parent node sets obtains the pollutant node Conditional probability table of ; Count all statistical intervals, the parent node set of the toxic effect node satisfies the given toxic effect parent node set The frequency of , in the statistical interval that satisfies the given set of toxic effect parent nodes, the frequency of the occurrence of statistical toxic effects is recorded as , divide the frequency of toxic effect occurrence by the frequency of satisfying a given set of toxic effect parent nodes to obtain the second conditional probability of the toxic effect node under the given set of toxic effect parent nodes ; Set the state of the toxic effect node to {State 1, State 2, ..., State y}, with a total of y states. Count the frequency of each state occurring in the statistical interval that satisfies the given set of toxic effect parent nodes, and record it as ; Divide the frequency of each state by the frequency of satisfying the given toxic effect parent node set to obtain the third conditional probability of each state of the toxic effect node under the given toxic effect parent node set ; A complete conditional probability table of the toxic effect node is formed by the second conditional probability of the toxic effect node under different toxic effect parent node sets and the third conditional probability of each state of the toxic effect node under different toxic effect parent node sets; A Bayesian network is constructed based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node. The joint probability distribution of the pollutant combination is obtained based on the Bayesian network. The current emission concentration of the pollutant combination is obtained to form an evidence vector. The posterior probability distribution is calculated based on the evidence vector. According to the posterior probability distribution, the probability of the toxic effect node being in different states is obtained, and the state corresponding to the maximum probability is taken as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment.

2. The method for diagnosing and controlling the toxic effects of soil and groundwater pollutants according to claim 1, wherein: For the pollutant combination S to be studied in the nth pollution source, the pollutants are divided into samples by combining emission data and toxicological parameters, toxic effect nodes are added to the pollutant causal relationship graph, and sub-causal relationship graphs are extracted. The specific method for determining the pollutant parent node set and the toxic effect parent node set based on the sub-causal relationship graph is as follows: Determine the pollutant combination S to be studied in the nth pollution source, the pollutant combination includes pollutants 、 ,..., , k is the number of pollutants in the pollutant combination; the pollutants appearing in each statistical interval are determined based on the emission data, and all pollutants appearing in each statistical interval and the corresponding emission data and toxicological parameters are taken as a sample and sorted in chronological order to obtain n1 samples; wherein, the emission data includes emission concentration, and when the emission concentration of a pollutant is greater than 0, it indicates that the pollutant is present; Adding a Toxic Effects Node to a Pollutant Causal Diagram For each sample, when the toxicological parameter of the sample is greater than the toxicity threshold, it means that the sample has a synergistic toxic effect. 、 ,..., Directed edges are drawn to the toxic effect nodes, and a sub-causal relationship graph is formed by the pollutant nodes and the toxic effect nodes; Determine the set of toxic effect parent nodes in each sample based on the child causal relationship graph and pollutant parent node set .

3. The method for diagnosing and controlling the toxic effects of soil and groundwater pollutants according to claim 2, characterized in that: The Bayesian network is constructed based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node. The joint probability distribution of the pollutant combination is obtained according to the Bayesian network. The current emission concentration of the pollutant combination is obtained to form an evidence vector. The specific method for calculating the posterior probability distribution based on the evidence vector is as follows: The pollutant nodes and toxic effect nodes in the sub-causal relationship graph correspond to each node in the Bayesian network. According to the directed edges of the sub-causal relationship graph, corresponding directed edges are added to the Bayesian network. The conditional probability table of each pollutant node and the conditional probability table of the toxic effect node are used as parameters of the corresponding node in the Bayesian network. Combine the conditional probability tables of each pollutant node according to the Bayesian network to obtain the joint probability distribution of the pollutant combination S ; Obtain the emission concentration of each pollutant in the current statistical interval and form the emission concentration into an evidence vector , substitute into the joint probability distribution and calculate the probability of the evidence vector E ;in, It represents the emission concentration of the mth pollutant observed in the current statistical interval, where m is the number of pollutant types observed in the current statistical interval; Calculate the posterior probability distribution of pollutant combination S under a given evidence vector E based on Bayesian network reasoning .

4. The method for diagnosing and controlling the toxic effects of soil and groundwater pollutants according to claim 3, wherein: The probability of the toxic effect node being in different states is obtained according to the posterior probability distribution, and the state corresponding to the maximum probability is used as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment. The specific method is: The probability of the toxic effect node being in each state is obtained according to the posterior probability distribution; Find the state with the highest probability and use the value of this state as the current severity of the toxic effect. When the severity of the toxic effect is greater than the state threshold, it means that the current toxic effect requires pollutant control. Obtain the sum of the weights of the edges of the shortest path between each pollutant node and the toxic effect node, and calculate the pollution weight of each pollutant node in the pollutant combination based on the current emission concentration of the pollutant; The pollutant nodes with the highest pollution weight will be selected as the pollutants to be treated first.

5. A soil and groundwater pollutant toxicity effect risk diagnosis and control system, which is used to implement the soil and groundwater pollutant toxicity effect risk diagnosis and control method according to any one of claims 1 to 4, characterized in that: include: The pollution data collection module is used to collect pollutant emission data and toxicological parameters of N pollution sources, calculate the partial correlation coefficients between pollutants, and construct a pollutant causal relationship diagram based on the toxicological parameters and partial correlation coefficients; The parent node determination module is used to divide the pollutants into samples based on the emission data and toxicological parameters for the pollutant combination S to be studied in the nth pollution source, add toxic effect nodes to the pollutant causal relationship graph, extract the child causal relationship graph, and determine the pollutant parent node set and the toxic effect parent node set based on the child causal relationship graph; A conditional probability table determination module is used to calculate the first conditional probability of the pollutant node based on the sample, calculate the second conditional probability and the third conditional probability of the toxic effect node, and obtain the conditional probability table of each pollutant node and the conditional probability table of the toxic effect node in the pollutant combination S; The posterior probability calculation module is used to construct a Bayesian network based on the conditional probability table of the pollutant node and the conditional probability table of the toxic effect node, obtain the joint probability distribution of the pollutant combination based on the Bayesian network, obtain the current emission concentration of the pollutant combination to form an evidence vector, and calculate the posterior probability distribution based on the evidence vector; The pollution weight calculation module is used to obtain the probability of the toxic effect node being in different states based on the posterior probability distribution, and take the state corresponding to the maximum probability as the current severity of the toxic effect. When the state is greater than the state threshold, the pollutant node with the highest pollution weight in the pollutant combination S is selected for priority treatment.

6. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for diagnosing and controlling the toxic effects of soil and groundwater pollutants as described in any one of claims 1 to 4 are implemented.

7. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which is suitable for loading by a processor to execute the steps in the soil and groundwater pollutant toxic effect risk diagnosis and control method as described in any one of claims 1 to 4.

Images