Environment diagnosis method and device based on numerical simulation and SVM efficient modeling

Abstract

The invention relates to an environment diagnosis method and device based on numerical simulation and SVM efficient modeling, and belongs to the technical field of soil-groundwater pollution prevention and control and risk management and control. An environment diagnosis model based on numerical simulation and SVM efficient modeling is established for a complex groundwater environment system under a typical condition framework, and reliable diagnosis and pre-judgment of soil and groundwater environment states, evolution processes and related influences in different application scenes can be simply and conveniently achieved on the premise that the given condition framework is met. The accuracy, efficiency and practicability of evaluation and pre-judgment of environment evolution, risk level and the like of the complex underground water system are improved, powerful support is provided for water and soil environment system protection and resource safety utilization, especially for precise prevention, control and repair of soil and underground water pollution and efficient risk management and control, and the method has important practical value.

Description

Technical Field

[0001] The present invention belongs to the technical field of soil-groundwater pollution prevention and control and risk management, and specifically relates to an environmental diagnosis method and device based on numerical simulation and SVM efficient modeling. Background Art

[0002] At present, in the monitoring and risk assessment of groundwater pollution, the application of numerical simulation systems to evaluate and quantify groundwater pollution status and environmental risks is a commonly used research method. However, accurate numerical simulation requires comprehensive consideration of many influencing factors and complex evolution processes, which is time-consuming and highly professional. In addition, there is a large uncertainty in the modeling and prediction of complex medium groundwater systems. Uncertainty assessment using methods such as Monte Carlo requires hundreds or thousands of numerical simulation tests, but the computational cost is huge or even unacceptable, and the efficiency is low. Therefore, there is an urgent need for practical, efficient, simple and accurate groundwater environment quantification and risk identification methods. Summary of the invention

[0003] To this end, the present invention provides an environmental diagnosis method and device based on numerical simulation and SVM efficient modeling to solve the problems of large uncertainty, high cost and low efficiency in environmental risk diagnosis in the prior art.

[0004] To achieve the above objectives, the present invention adopts the following technical solutions:

[0005] In a first aspect, the present invention provides an environmental diagnosis method based on numerical simulation and SVM efficient modeling, comprising:

[0006] Select the associated characteristic variables of the groundwater environment and construct a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under typical conditions; select and obtain a numerical simulation system based on mathematical control equations and numerical models;

[0007] Based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, n model input parameter variables that affect the spatiotemporal evolution of the associated characteristic variables are obtained, wherein the specific values ​​of the model input parameter variables are model input parameters; the variation interval of each model input parameter variable is selected and then the model input parameter variable interval set is obtained;

[0008] According to a given preset number of typical scenarios, different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system are obtained; based on the numerical simulation system, corresponding associated characteristic variable simulation values ​​of the different typical scenarios are obtained, and a statistical modeling data set is obtained by combining all the model input parameter combinations of the different typical scenarios and their corresponding associated characteristic variable simulation value sets; a multivariate regression statistical prediction model of the associated characteristic variables and the model input parameter variables is established and optimized according to the statistical modeling data set;

[0009] Based on n input parameter variables of the model, a preset number of scenario samples are obtained to obtain a first set of data-enhanced scenario parameter combinations, and a first set of data-enhanced scenario simulation values ​​corresponding to associated characteristic variables of each scenario are obtained through the numerical simulation system, and the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​are combined as a model verification data set; based on the model verification data set, the prediction error value of the associated characteristic variable and its distribution characteristics are obtained through the multivariate regression statistical prediction model; according to the prediction error value and its distribution characteristics, a distribution interval of the prediction error random variable corresponding to a given preset confidence is determined;

[0010] Based on the n model input parameter variables, a preset number of scenario samples are obtained to obtain a second set of data-enhanced scenario parameter combinations, and a second set of data-enhanced scenario prediction values ​​corresponding to the associated characteristic variables of each scenario are obtained through the multivariate regression statistical prediction model; the associated characteristic variable prediction confidence interval is determined according to the second set of data-enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to the given preset confidence; based on the n model input parameter variables, a preset number of scenario parameter combinations are obtained and the prediction confidence interval is then combined to obtain a statistical analysis enhanced modeling data set;

[0011] If a single environmental diagnostic threshold is given, then the first set of output label values, the second set of output label values ​​and the third set of output label values ​​of the SVM classification machine learning are obtained in sequence according to the statistical modeling data set, the statistical analysis enhanced modeling data set and the model verification data set;

[0012] The statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set correspond to a preset number of scenario parameter combinations, respectively, and are merged with the first group of output label values, the second group of output label values, and the third group of output label values ​​respectively corresponding to each other to obtain a single threshold SVM classification machine learning data set;

[0013] If multiple environmental diagnosis thresholds are given, different thresholds are selected according to the preset threshold value variation interval and the preset threshold value number, and based on the statistical modeling data set and the statistical analysis enhanced modeling data set, as well as based on the model verification data set, the environmental diagnosis threshold is added as a model input parameter variable for SVM classification machine learning modeling, thereby obtaining an SVM classification machine learning data set containing threshold parameter variables;

[0014] According to the single threshold SVM classification machine learning data set and the SVM classification machine learning data set containing threshold parameters, according to and applying the SVM classification model modeling principle and optimization method, the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model containing threshold parameters are obtained in turn;

[0015] For actual application scenarios that meet the typical condition framework, the specific values ​​of the model parameters are obtained according to the actual conditions and collected data of the specific groundwater environment system, and the output label value corresponding to the threshold is obtained through the single threshold SVM environmental diagnosis model, and the output label value is the environmental diagnosis prediction result, and / or based on the actual conditions and collected data of the specific groundwater environment system and the preset environmental diagnosis threshold, the environmental diagnosis prediction result is obtained through the SVM environmental diagnosis model containing threshold parameters.

[0016] Furthermore, the method of obtaining a statistical modeling data set by combining the model input parameters of all the different typical scenarios and their corresponding associated characteristic variable simulation value sets includes:

[0017] Selecting the change level of each model input parameter; obtaining different typical scenarios consisting of different model input parameter combinations of the groundwater environment system through an orthogonal test method or a full combination method of input parameters, which is deterministic scenario sampling;

[0018] Based on the numerical simulation system, corresponding simulation values ​​of the associated characteristic variables of the different typical scenarios are obtained; according to different model input parameter combinations of the different typical scenarios and their corresponding simulation values, statistical modeling data sets of the deterministic scenario sampling are obtained;

[0019] and obtaining a preset number of random scenario samples based on random values ​​of n model input parameter variables, that is, obtaining a certain number of different random scenario model input parameter combinations, obtaining the simulated value of each random scenario-related characteristic variable through the numerical simulation system, and combining the random scenario model parameter combinations and the simulated value sets of the random scenario-related characteristic variables to obtain the statistical modeling data set of the random scenario samples;

[0020] And the statistical modeling data set of the deterministic scenario sampling and the statistical modeling data set of the random scenario sampling are combined to obtain a statistical modeling data set integrated with the deterministic scenario sampling and the random scenario sampling. Further, the SVM classification machine learning data set includes: an SVM classification machine learning training set and an SVM classification machine learning test set;

[0021] And the statistical modeling data set and the statistical analysis enhanced modeling data set are respectively corresponding to a preset number of scenario parameter combinations, and are merged with the corresponding first group of output label values ​​and second group of output label values ​​to obtain a single threshold SVM classification machine learning training set; the model verification data set is corresponding to a preset number of scenario parameter combinations and the third group of output label values ​​to obtain a single threshold SVM classification machine learning test set;

[0022] And the method of obtaining a SVM classification machine learning data set containing threshold parameters also includes: if multiple environmental diagnosis thresholds are given, different thresholds are selected according to the preset threshold value variation range and the preset threshold value number, based on the statistical modeling data set and the statistical analysis enhanced modeling data set, and based on the model verification data set, the environmental diagnosis threshold is used as a new model input parameter to obtain a training set and a test set of SVM classification machine learning containing threshold parameters, respectively.

[0023] Furthermore, it also includes: given a preset value of modeling accuracy of the SVM environmental diagnostic model, including a preset value of training accuracy and a preset value of testing accuracy, if the obtained modeling accuracy of the SVM environmental diagnostic model is lower than the preset value, then increasing the preset number of different given typical scenarios and / or increasing the preset number of different scenario samples to perform SVM environmental diagnostic modeling optimization to obtain a corresponding accuracy SVM environmental diagnostic model; the corresponding accuracy SVM environmental diagnostic model includes a single threshold SVM environmental diagnostic model and a threshold parameter SVM environmental diagnostic model.

[0024] Furthermore, the associated characteristic variables are at least one; the associated characteristic variables include: groundwater pollutant concentration, saturated zone pollutant cross-section or interface flux, pollution plume migration and diffusion front position, pollution plume stability time, stable pollution plume distribution area, stable pollution plume front distance from the pollution source, stable pollution plume average concentration and pollution plume dissipation time;

[0025] As well as soil pollutant concentration, gas-phase pollutant concentration in the vadose zone, pollutant flux in the vadose zone section, pollutant interface flux between the vadose zone and saturated zone, pollutant flux at the soil-air interface, pollutant flux at the rock-air interface and the spatiotemporal distribution of NAPL pollutants.

[0026] Furthermore, the construction of the conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the typical condition framework includes:

[0027] Determine the characteristics of the aeration zone, wherein the characteristics of the aeration zone include medium type, lithological structure, medium heterogeneity and anisotropy;

[0028] Determine the characteristics of the saturated zone, wherein the characteristics of the saturated zone include medium type, water-bearing system structure, seepage dimension, flow pattern, medium heterogeneity and anisotropy characteristics;

[0029] Determine pollution source characteristics, including the phase state of the pollutants of interest, the type of pollutants of interest, the spatial distribution characteristics of the pollution sources, source intensity changes and dynamic characteristics;

[0030] Determining boundary condition characteristics, wherein the boundary condition characteristics include hydrodynamic field boundary condition characteristics and hydrochemical field boundary condition characteristics;

[0031] Initial condition characteristics are determined, wherein the initial condition characteristics include initial condition characteristics of the hydrodynamic field and initial condition characteristics of the hydrochemical field.

[0032] Furthermore, the associated characteristic variables and the model input parameter variables include:

[0033] The associated characteristic variables and the model input parameter variables are processed by any one or any combination of preset data processing methods, and the preset data processing methods include: no conversion, proportional scaling, logarithmic conversion, positive and negative consistency and normalization.

[0034] Furthermore, obtaining a preset number of scenario samples based on n model input parameters includes:

[0035] Select the number of random samples, determine or select the random distribution type and corresponding distribution parameters applicable to each model input parameter, complete the random number generation of the model input parameter and thus form a new parameter combination scenario.

[0036] Furthermore, the prediction error value of the associated characteristic variable and its distribution characteristics are obtained through the multivariate regression statistical prediction model; according to the prediction error value and its distribution characteristics, the distribution interval of the prediction error random variable corresponding to a given preset confidence is determined, including:

[0037] Based on the model validation set data, the multivariate regression statistical prediction model is used to obtain a prediction error value data set, wherein the prediction error value data set includes multiple prediction errors; the mean and variance of the prediction error random variable are estimated, and its probability density function is fitted and identified, and the sum of the cumulative probabilities on both sides is determined according to the preset confidence and the distribution interval of the prediction error random variable with the corresponding confidence is obtained.

[0038] Furthermore, the SVM classification model modeling optimization method in the SVM classification model modeling principle and optimization method includes:

[0039] Optimization of SVM classification model modeling optimization method and modeling parameters;

[0040] The modeling parameters preferably include

[0041] Optimization of kernel function types and their associated parameters;

[0042] The kernel function types preferably include linear kernel, polynomial kernel, radial basis function kernel and sigmoid kernel;

[0043] The optimization of the radial basis function kernel-related parameters includes the optimization of the regularization parameter C and the data mapping complexity parameter gamma;

[0044] The SVM modeling optimization method preferably includes a grid search method, a particle swarm optimization method, and a genetic particle swarm hybrid algorithm.

[0045] Further, the obtaining of the first group of output label values, the second group of output label values ​​and the third group of output label values ​​of the SVM classification machine learning includes:

[0046] If each of the associated characteristic variables corresponds to a single environmental diagnostic threshold, then based on the statistical modeling data set and the model verification data set, compare the relative size of each of the associated characteristic variable simulation values ​​and the corresponding single environmental diagnostic threshold, and obtain the first group of output label values ​​and the third group of output label values ​​of the SVM classification machine learning in turn; based on the statistical analysis enhanced modeling data set, if the predicted confidence interval of each of the associated characteristic variables does not include the corresponding single environmental diagnostic threshold, then based on the relative size of the upper and lower limits of the predicted confidence interval and the corresponding single environmental diagnostic threshold, obtain the second group of output label values ​​of the SVM classification machine learning;

[0047] In a second aspect, the present invention provides an environmental diagnosis device based on numerical simulation and SVM efficient modeling.

[0048] The device comprises:

[0049] The conceptual model building module is used to select the associated characteristic variables of the groundwater environment and build a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the framework of typical conditions; select and obtain a numerical simulation system based on mathematical control equations and numerical models;

[0050] A numerical simulation module is used to obtain n model input parameter variables that affect the spatiotemporal evolution of associated characteristic variables based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, wherein the specific values ​​of the model input parameter variables are model input parameters; select the variation interval of each model input parameter variable and then collect them to obtain a model input parameter variable interval set; and obtain the corresponding numerical simulation values ​​of the typical scenario and scenario sampling based on the numerical simulation system;

[0051] A statistical model building module is used to obtain different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system based on a given preset number of typical scenarios; obtain corresponding numerical simulation values ​​of the different typical scenarios based on the numerical simulation system, and obtain a statistical modeling data set by combining the model input parameter combinations of all the different typical scenarios and their corresponding associated characteristic variable simulation value sets; establish and optimize a multivariate regression statistical prediction model of the associated characteristic variables and the model input parameter variables based on the statistical modeling data set;

[0052] The statistical model prediction error analysis module is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a first set of data-enhanced scenario parameter combinations, obtain a first set of data-enhanced scenario simulation values ​​of associated characteristic variables corresponding to each scenario through the numerical simulation system, and use the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​as a model verification data set; based on the model verification data set, obtain the prediction error value of the associated characteristic variable and its distribution characteristics through the multivariate regression statistical prediction model; determine the distribution interval of the prediction error random variable corresponding to a given preset confidence according to the prediction error value and its distribution characteristics;

[0053] A machine learning enhanced data set generation module is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a second set of data enhanced scenario parameter combinations, obtain a second set of data enhanced scenario prediction values ​​for associated feature variables corresponding to each scenario through the multivariate regression statistical prediction model; determine the associated feature variable prediction confidence interval based on the second set of data enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to the given preset confidence; obtain a preset number of scenario parameter combinations based on the n model input parameter variables and the prediction confidence interval and then merge them to obtain a statistical analysis enhanced modeling data set;

[0054] It is also used for obtaining, if a single environmental diagnostic threshold is given, a first group of output label values, a second group of output label values, and a third group of output label values ​​of SVM classification machine learning in sequence according to the statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set; merging the statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set corresponding to a preset number of scenario parameter combinations, respectively, with the first group of output label values, the second group of output label values, and the third group of output label values ​​respectively corresponding to obtain a single threshold SVM classification machine learning data set;

[0055] It is also used for, if multiple environmental diagnosis thresholds are given, selecting different thresholds according to the preset threshold value variation interval and the preset threshold value number, adding the environmental diagnosis threshold as a model input parameter variable of SVM classification machine learning modeling based on the statistical modeling data set and the statistical analysis enhanced modeling data set, and based on the model verification data set, thereby obtaining the SVM classification machine learning data set containing the threshold parameter variable;

[0056] A universal SVM environmental diagnosis model construction module is used to obtain the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model containing threshold parameters in turn according to the single threshold SVM classification machine learning data set and the SVM classification machine learning data set containing threshold parameters, based on and applying the SVM classification model modeling principle and optimization method;

[0057] The application scenario environmental diagnosis module obtains the specific values ​​of the model parameters according to the actual conditions and collected data of the specific groundwater environmental system for actual application scenarios that meet the typical condition framework, and obtains the output label value of the corresponding threshold through the single threshold SVM environmental diagnosis model, and the output label value is the environmental diagnosis prediction result, and / or obtains the environmental diagnosis prediction result through the SVM environmental diagnosis model containing threshold parameters according to the actual conditions and collected data of the specific groundwater environmental system and the preset environmental diagnosis threshold.

[0058] The present invention adopts the above technical solution and has at least the following beneficial effects:

[0059] An environmental diagnosis method and device based on numerical simulation and SVM efficient modeling are provided. Based on the associated characteristic variables of the groundwater environment, an SVM-based environmental diagnosis model is established for complex groundwater environmental systems under a typical condition framework. The SVM-based environmental diagnosis model can simply realize reliable diagnosis and prediction of groundwater environmental status, evolution process and related impacts in different application scenarios under the premise of meeting a given condition framework, thereby improving the accuracy, efficiency and practicality of environmental diagnosis and risk assessment of complex groundwater systems, providing strong support for groundwater pollution risk diagnosis, prevention, control and remediation, and has important practical value.

[0060] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 is a flow chart of an environmental diagnosis method based on numerical simulation and SVM efficient modeling shown in an exemplary embodiment of the present invention;

[0063] Figure 2 It is a schematic diagram of the structure of a vadose zone-diving interface under a typical condition framework of layered heterogeneity shown in an exemplary embodiment of the present invention;

[0064] Figure 3 It is a schematic block diagram of an environmental diagnosis device based on numerical simulation and SVM efficient modeling shown in an exemplary embodiment of the present invention.

[0065] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0067] The application of numerical simulation system to evaluate and quantify groundwater pollution status and environmental risks is a common research method. However, in order to accurately simulate the numerical simulation, it is necessary to comprehensively consider many influencing factors and complex evolution processes, which is time-consuming and highly professional. In addition, the modeling and prediction of complex medium groundwater systems themselves have large uncertainties. The use of Monte Carlo and other methods for uncertainty assessment requires hundreds or thousands of numerical simulation tests, but the computational cost is huge or even unacceptable. Building an efficient machine learning prediction model based on numerical simulation provides another option for groundwater environmental process quantification and risk diagnosis. The premise of establishing a good machine learning model is to have data of a certain scale and quality. However, due to the complexity of the model and the influence of computing power, even supercomputers cannot obtain a large amount of data sets for training; therefore, combining numerical simulation with machine learning methods to more efficiently and practically accurately model, thereby reliably supporting groundwater environmental process quantification and risk diagnosis, has become an environmental problem that needs to be solved urgently.

[0068] The embodiments of the present invention provide an environmental diagnosis method and device based on numerical simulation and SVM efficient modeling, which can establish an SVM environmental problem diagnosis and prediction model with certain versatility for a complex groundwater environmental system under a certain type of conditions, and effectively provide rapid and accurate predictions of the groundwater environmental status, evolution process and related impacts in different application scenarios under a given condition framework, so as to improve the accuracy, efficiency and practicality of environmental diagnosis and risk assessment of complex groundwater systems.

[0069] The method and device of the present invention are described below through specific embodiments.

[0070] See also Figure 1 , Figure 1 is a flowchart of an environmental diagnosis method based on numerical simulation and SVM efficient modeling shown in an exemplary embodiment of the present invention, see Figure 1 , the method comprising:

[0071] Step S11, selecting the associated characteristic variables of the groundwater environment and constructing a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under a typical condition framework; selecting and obtaining a numerical simulation system based on mathematical control equations and numerical models;

[0072] Step S12: based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, n model input parameter variables that affect the spatiotemporal evolution of the associated characteristic variables are obtained, and the specific values ​​of the model input parameter variables are model input parameters; the variation interval of each model input parameter variable is selected and then the model input parameter variable interval set is obtained;

[0073] Step S13, according to a given preset number of typical scenarios, different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system are obtained; based on the numerical simulation system, corresponding associated characteristic variable simulation values ​​of different typical scenarios are obtained, and a statistical modeling data set is obtained by combining the model input parameter combinations of all different typical scenarios and their corresponding associated characteristic variable simulation value sets; a multivariate regression statistical prediction model of associated characteristic variables and model input parameter variables is established and optimized according to the statistical modeling data set;

[0074] Step S14: obtaining a preset number of scenario samples based on n model input parameter variables, obtaining a first set of data-enhanced scenario parameter combinations, obtaining a first set of data-enhanced scenario simulation values ​​of associated characteristic variables corresponding to each scenario through a numerical simulation system, and combining the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​as a model verification data set; obtaining the associated characteristic variable prediction error value and its distribution characteristics through a multivariate regression statistical prediction model based on the model verification data set; determining the prediction error random variable distribution interval corresponding to a given preset confidence level according to the prediction error value and its distribution characteristics;

[0075] Step S15: based on the n model input parameter variables, a preset number of scenario samples are obtained to obtain a second set of data-enhanced scenario parameter combinations, and a second set of data-enhanced scenario prediction values ​​corresponding to the associated characteristic variables of each scenario are obtained through a numerical simulation system; the prediction confidence interval of the associated characteristic variables is determined according to the second set of data-enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to a given preset confidence level; based on the n model input parameter variables, a preset number of scenario parameter combinations and prediction confidence intervals are obtained and then merged to obtain a statistical analysis-enhanced modeling data set;

[0076] Step S16: if a single environmental diagnostic threshold is given, then the first set of output label values, the second set of output label values ​​and the third set of output label values ​​of the SVM classification machine learning are obtained in turn according to the statistical modeling data set, the statistical analysis enhanced modeling data set and the model verification data set;

[0077] Step S17, merging the statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set corresponding to a preset number of scenario parameter combinations, and the corresponding first set of output label values, second set of output label values, and third set of output label values ​​to obtain a single threshold SVM classification machine learning data set;

[0078] Step S18: If multiple environmental diagnosis thresholds are given, different thresholds are selected according to the preset threshold value variation interval and the preset threshold value quantity, and based on the statistical modeling data set and the statistical analysis enhanced modeling data set, as well as based on the model verification data set, the environmental diagnosis threshold is added as a model input parameter variable for SVM classification machine learning modeling, thereby obtaining an SVM classification machine learning data set containing threshold parameter variables;

[0079] Step S19, based on the single threshold SVM classification machine learning data set and the SVM classification machine learning data set containing threshold parameters, according to and applying the SVM classification model modeling principle and optimization method, the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model containing threshold parameters are obtained in turn;

[0080] Step S20: For actual application scenarios that meet the typical condition framework, based on the actual conditions and collected data of the specific groundwater environmental system, obtain specific values ​​of model parameters, and obtain the output label value corresponding to the threshold through a single threshold SVM environmental diagnosis model, where the output label value is the environmental diagnosis prediction result, and / or based on the actual conditions and collected data of the specific groundwater environmental system and the preset environmental diagnosis threshold, obtain the environmental diagnosis prediction result through an SVM environmental diagnosis model containing threshold parameters.

[0081] It should be noted that the technical solution provided in this embodiment can be loaded into existing applications in the form of a small program or a plug-in in specific practice, or in the form of a separate application to implement the speed regulation function through an external interface. Applicable scenarios include but are not limited to: diagnosis of groundwater environmental systems.

[0082] Specifically, the preset number or preset quantity in each step is set according to the specific application scenario and specific diagnostic requirements; the preset threshold value variation range and the preset threshold value quantity are also set according to the specific application scenario and specific diagnostic requirements.

[0083] It can be understood that the method provided in this embodiment is based on the associated characteristic variables of the groundwater environment, and establishes an SVM-based environmental diagnosis model for complex groundwater environmental systems under a typical condition framework. The SVM-based environmental diagnosis model can simply realize reliable diagnosis and prediction of the groundwater environmental status, evolution process and related impacts in different application scenarios under the premise of meeting a given condition framework, thereby improving the accuracy, efficiency and practicality of environmental diagnosis and risk assessment of complex groundwater systems.

[0084] In specific practice, step S13 of "combining model input parameter combinations of all different typical scenarios and their corresponding associated characteristic variable simulation value sets to obtain a statistical modeling data set" includes: selecting the change level of each model input parameter; obtaining different typical scenarios consisting of different model input parameter combinations of the groundwater environment system through an orthogonal test method or a full combination method of input parameters, which is deterministic scenario sampling; obtaining corresponding simulation values ​​of characteristic variables associated with different typical scenarios based on a numerical simulation system; obtaining a statistical modeling data set of deterministic scenario sampling based on different model input parameter combinations of different typical scenarios and their corresponding simulation values; and obtaining a preset number of random scenario samplings based on random values ​​of n model input parameter variables, that is, obtaining a certain number of different random scenario model input parameter combinations, obtaining the simulation value of each random scenario associated characteristic variable through a numerical simulation system, and obtaining a statistical modeling data set of random scenario sampling by combining the random scenario model parameter combinations and the simulation values ​​of the random scenario associated characteristic variables; and combining the statistical modeling data set of deterministic scenario sampling with the statistical modeling data set of random scenario sampling to obtain a statistical modeling data set integrating deterministic scenario sampling and random scenario sampling.

[0085] In specific practice, the SVM classification machine learning data set includes an SVM classification machine learning training set and an SVM classification machine learning test set; the method also includes: merging the statistical modeling data set and the statistical analysis enhanced modeling data set corresponding to a preset number of scenario parameter combinations, and the corresponding first group of output label values ​​and second group of output label values ​​to obtain a single threshold SVM classification machine learning training set; merging the model verification data set corresponding to a preset number of scenario parameter combinations and the third group of output label values ​​to obtain a single threshold SVM classification machine learning test set; if multiple environmental diagnosis thresholds are given, different thresholds are selected according to the preset threshold value variation range and the preset threshold value number, and based on the statistical modeling data set and the statistical analysis enhanced modeling data set, as well as based on the model verification data set, the environmental diagnosis threshold is used as a new model input parameter variable, and a training set and a test set of SVM classification machine learning containing threshold parameter variables are obtained respectively.

[0086] In specific practice, it also includes: given a preset value of the modeling accuracy of the SVM environmental diagnostic model, including a preset value of the training accuracy and a preset value of the testing accuracy; if the modeling accuracy of the obtained SVM environmental diagnostic model is lower than the preset value, then increasing the preset number of different given typical scenarios and / or increasing the preset number of different scenario samples to optimize the modeling of the SVM environmental diagnostic model to obtain a corresponding accuracy SVM environmental diagnostic model; the corresponding accuracy SVM environmental diagnostic model includes a single threshold SVM environmental diagnostic model and a SVM environmental diagnostic model with threshold parameters.

[0087] In specific practice, the "associated characteristic variables" in step S11 include: groundwater pollutant concentration, saturated zone pollutant cross-section or interface flux, pollution plume migration and diffusion front position, pollution plume stabilization time, stable pollution plume distribution area, stable pollution plume front distance from the pollution source, stable pollution plume average concentration and pollution plume dissipation time; as well as soil pollutant concentration, vadose zone gas phase pollutant concentration, vadose zone cross-section pollutant flux, vadose zone-saturated zone pollutant interface flux, soil-air interface pollutant flux, rock-air interface pollutant flux and NAPL pollutant spatiotemporal distribution.

[0088] In specific practice, the associated characteristic variables and model input parameter variables are processed by any one or any combination of preset data processing methods. The preset data processing methods include: no conversion, proportional scaling, logarithmic conversion, positive and negative consistency, and normalization.

[0089] In specific practice, a preset number of scenario samples are obtained based on n model input parameters, including selecting the number of random sampling, determining or selecting the random distribution type and corresponding distribution parameters applicable to each model input parameter, completing the random number generation of the model input parameter, and then forming a new parameter combination scenario.

[0090] In specific practice, the prediction error value of the associated characteristic variable and its distribution characteristics are obtained through a multivariate regression statistical prediction model; based on the prediction error value and its distribution characteristics, the distribution interval of the prediction error random variable corresponding to a given preset confidence is determined, including: based on the model validation set data, using the multivariate regression statistical prediction model to obtain a prediction error value data set, the prediction error value data set includes multiple prediction errors; estimating the mean and variance of the prediction error random variable, fitting and identifying its probability density function, determining the sum of the cumulative probabilities on both sides according to the preset confidence and obtaining the corresponding confidence prediction error random variable distribution interval.

[0091] In specific practice, the SVM classification model modeling optimization method includes the optimization of SVM classification model modeling optimization method and modeling parameter optimization; modeling parameter optimization includes kernel function type optimization and its associated parameter optimization; kernel function type optimization includes linear kernel, polynomial kernel, radial basis function kernel and Sigmoid kernel selection; radial basis function kernel associated parameter optimization includes regularization parameter C and data mapping complexity parameter gamma selection; and SVM modeling optimization method optimization includes grid search method, particle swarm optimization method and genetic particle swarm hybrid algorithm selection.

[0092] In specific practice, in order to obtain the first set of output label values, the second set of output label values ​​and the third set of output label values ​​of the SVM classification machine learning, if each associated characteristic variable is given to correspond to a single environmental diagnostic threshold, then based on the statistical modeling data set and the model verification data set, the simulation value of each associated characteristic variable is compared with the corresponding single environmental diagnostic threshold, and the first set of output label values ​​and the third set of output label values ​​of the SVM classification machine learning are obtained in turn; based on the statistical analysis of the enhanced modeling data set, if the predicted confidence interval of each associated characteristic variable does not contain the corresponding single environmental diagnostic threshold, then the second set of output label values ​​of the SVM classification machine learning is obtained based on the relative size of the upper and lower limits of the predicted confidence interval and the corresponding single environmental diagnostic threshold;

[0093] See also Figure 2 , Figure 2 is a schematic diagram of the structure of a layered heterogeneous aeration zone-diving interface under typical conditions, see Figure 2 The aeration zone-phreatic surface interface under the framework of typical layered heterogeneous conditions includes: the pollution source 1 on the uppermost surface, and below the pollution source 1 are the upper aeration zone 2, the lower aeration zone 3, the phreatic surface 4 and the phreatic layer 5 in sequence; the pollution source 1 also contains accumulated water 6; it also includes a constant flow and constant concentration position 7 and a one-dimensional unsaturated flow direction 8.

[0094] In specific practice, step S11 of "constructing a conceptual model of the spatiotemporal evolution of the corresponding groundwater environmental system under a typical condition framework" includes: determining the characteristics of the aeration zone, including medium type, lithological structure, medium heterogeneity and anisotropy; determining the characteristics of the saturated zone, including medium type, water-bearing system structure, seepage dimension, flow state, medium heterogeneity and anisotropy; determining the characteristics of the pollution source, including the phase state of the pollutants of concern, the type of pollutants of concern, the spatial distribution characteristics of the pollution source, source intensity changes and dynamic characteristics; determining the characteristics of the boundary conditions, including the boundary conditions of the hydrodynamic field and the boundary conditions of the hydrochemical field; determining the characteristics of the initial conditions, including the initial conditions of the hydrodynamic field and the initial conditions of the hydrochemical field.

[0095] Specifically, groundwater environment-related characteristic variables are selected, and a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under a typical conditions framework is constructed. The mathematical control equations and numerical models suitable for characterizing the spatiotemporal evolution of the corresponding groundwater environment system are further obtained, and the numerical simulation system that can quantify the spatiotemporal evolution of the groundwater environment-related characteristic variables is screened.

[0096] In specific practice, the process of step S12 is to determine all model input parameter variables that affect the spatiotemporal evolution of the selected environmental associated characteristic variables based on the spatiotemporal evolution conceptual model and the numerical model and numerical simulation system used, and form a model input parameter variable set X = [X1, X2, ..., Xn], where n is the total number of model input parameter variables considered; considering the variable range of each model input parameter variable, select the variation interval of each model input parameter variable to be considered, and form a model input parameter variable interval set V = [V1, V2, ..., Vn];

[0097] In specific practice, in step S13, "different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system are obtained". Specifically, the change level of each model input parameter is selected, and the full combination of model input parameters or orthogonal experimental design method is applied to obtain the typical scenarios of different parameter combinations of the selected groundwater environment system, and obtain the simulated value set Y of the groundwater environment-related characteristic variables to form a statistical modeling data set for deterministic scenario sampling, wherein the number of orthogonal experimental design scenarios should satisfy the statistical modeling without overfitting.

[0098] In specific practice, the step S14 of "obtaining a preset number of scenario samples based on n model input parameters" includes: selecting the number of random samplings, determining or selecting the random distribution type and corresponding distribution parameters applicable to each model input parameter, completing the random number generation of the model input parameter and then forming a new parameter combination scenario; for the orthogonal experimental design situation, a new parameter combination scenario can also be formed by a random combination of the level values ​​of each model input parameter.

[0099] Specifically, the model input parameter combinations are changed to produce different scenario sampling sets, and the numerical simulation system is used to obtain the corresponding characteristic variable simulation value sets, which are integrated with the statistical modeling data set of the aforementioned deterministic scenario sampling to form a statistical modeling data set. Appropriate data processing methods are selected to establish and optimize a multivariate regression statistical prediction model that associates the characteristic variables Y with the model input parameter variables X.

[0100] In specific practice, it also includes further selecting an appropriate number of random sampling, generating random numbers of model input parameters again and forming corresponding new scenarios for more model input parameter combinations, applying a numerical simulation system to obtain simulation values ​​of corresponding groundwater environment-related characteristic variables, integrating and forming a model verification data set, and applying the established regression statistical prediction model to obtain prediction error values ​​and their distribution characteristics, thereby further determining the distribution interval (a, b) of the prediction error random variable of the groundwater environment-related characteristic variable statistical prediction model corresponding to a given confidence level.

[0101] In specific practice, step S14 "based on the model verification data set, obtain the prediction error value of the associated characteristic variable and its distribution characteristics through the multivariate regression statistical prediction model; determine the prediction error random variable distribution interval corresponding to the given preset confidence according to the prediction error value and its distribution characteristics" includes: based on the model verification set data, use the multivariate regression statistical prediction model to obtain the prediction error value data set, the prediction error value data set includes multiple prediction errors; estimate the mean and variance of the prediction error random variable, fit and identify its probability density function, and determine the sum of the cumulative probabilities on both sides as the prediction error random variable distribution interval.

[0102] Specifically, the appropriate number of random sampling is further selected, and the random number of model input parameters is generated again to form a corresponding new more parameter combination scenario realization, and the aforementioned multivariate regression statistical prediction model is used to obtain the predicted value Y of the groundwater environment associated characteristic variable corresponding to each scenario realization S Based on the distribution interval (a, b) of the prediction error random variable corresponding to the given confidence level of the statistical prediction model of the groundwater environment-related characteristic variables determined above, the upper and lower limits of the true value of the groundwater environment-related characteristic variables under the given confidence level, i.e., the prediction confidence interval (Y P1 ,Y P2 ), and then generate a statistical analysis enhanced modeling data set based on the random model input parameter combination scenario and characteristic variable prediction confidence interval; similarly, the model verification data set can be supplemented and updated as needed.

[0103] In specific practice, during step S16 and step S17, appropriate data processing methods are used to process the statistical modeling data set, the model verification data set, and the statistical analysis enhanced modeling data set. Given an environmental diagnosis threshold Vc, for the statistical modeling data set and the model verification data set, the output label value of the SVM machine learning can be directly obtained by comparing the threshold with the size of the environmental associated characteristic value; and for the statistical analysis enhanced modeling data set, the output label value of the SVM machine learning is identified by comparing the relative size of the threshold with the upper and lower limits of the prediction interval of the environmental associated characteristic variable; based on the statistical modeling data set and the statistical analysis enhanced modeling data set corresponding to the parameter variable combination scenarios and their corresponding SVM machine learning output label values ​​together constitute a single threshold SVM machine learning training set, and based on the parameter variable combination scenarios corresponding to the model verification data set and their corresponding SVM machine learning output label values ​​together constitute a single threshold SVM machine learning test set.

[0104] In specific practice, the "SVM classification model modeling optimization method" in step 19 includes the optimization of SVM classification model modeling optimization method and modeling parameter optimization; modeling parameter optimization includes kernel function type optimization and its associated parameter optimization; kernel function type optimization includes linear kernel, polynomial kernel, radial basis function kernel and Sigmoid kernel selection; radial basis function kernel associated parameter optimization includes regularization parameter C and data mapping complexity parameter gamma optimization; and SVM modeling optimization method optimization includes grid search method, particle swarm optimization method and genetic particle swarm hybrid algorithm selection.

[0105] Specifically, appropriate SVM machine learning training and testing accuracy are selected, and based on the SVM machine learning modeling optimization method, overfitting is avoided, so as to preferably obtain the SVM environmental diagnosis model for a single specific environmental diagnosis threshold Vc.

[0106] In specific practice, during step S20, for a specific groundwater environmental system, based on the survey results of the specific groundwater environmental system, confirm that the specific application scenario meets the requirements of the defined typical condition framework, and then further obtain the specific values ​​of the corresponding input parameters of the corresponding SVM environmental diagnosis model. For the specific environmental diagnosis threshold considered, the SVM environmental diagnosis model obtained by the single specific threshold SVM environmental diagnosis model determination method can obtain the corresponding accuracy prediction result, or for the specific environmental diagnosis threshold considered, the established general SVM environmental diagnosis model containing threshold parameters can be used to quickly and easily obtain the corresponding accuracy prediction result.

[0107] In a specific embodiment, polycyclic aromatic hydrocarbons (PAHs), as organic matter widely present in the environment, are highly valued by countries around the world due to their persistence, carcinogenicity and mutagenicity. The carcinogenic concentration of PAHs in groundwater published by the United States is in the range of 0.2-6.9 ng / L, but its corresponding concentration in surface water is 0.1-800 ng / L, most of which are in the range of 2-50 ng / L, and the solubility of PAHs in water is mostly above 1000 ng / L. The polluted surface organic polluted water body continues to infiltrate, and will undergo seepage, adsorption and interception, hydrodynamic dispersion and biodegradation in the vadose zone, and enter the shallow groundwater along the vadose zone soil with relatively weak anti-fouling performance, thereby causing the PAHs concentration in the groundwater to continue to increase. According to the analysis of the solubility and carcinogenic concentration range of PAHs, PAHs in groundwater are prone to exceed the carcinogenic standard. Therefore, it is necessary to pay attention to the migration and transformation of polycyclic aromatic hydrocarbon pollutants in the vadose zone, including quantitative research on the flux changes of polycyclic aromatic hydrocarbon pollutants at the vadose zone-diving interface under typical conditions. This will provide an important basis for the temporal and spatial distribution of relevant organic pollutants in the water and soil environment of the vadose zone under typical conditions and for the risk assessment of groundwater pollution. This embodiment uses limited numerical simulation data to establish an efficient and reliable vadose zone pollutant flux prediction model, demonstrating the important role of the present invention in environmental protection.

[0108] In view of the above problems, an environmental diagnosis method based on numerical simulation and SVM efficient modeling is proposed. The specific steps are as follows:

[0109] Step S31, select the stable flux of pollutants at the interface of the vadose zone and the saturated zone as the groundwater environment-related characteristic variable, and construct a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the framework of typical conditions, that is, a conceptual model for flux prediction of the vadose zone-phreatic interface under the framework of layered heterogeneous typical conditions, and further obtain mathematical control equations and numerical models suitable for characterizing the spatiotemporal evolution of the corresponding groundwater environment system, and screen the numerical simulation system that can quantify the spatiotemporal evolution of the groundwater environment-related characteristic variables.

[0110] Specifically, there are polycyclic aromatic hydrocarbon pollution problems in coking plant sites, petrochemical sites, power plant pollution sites, etc. in my country. When precipitation forms surface water in these sites, the water infiltrates steadily to form a stable seepage field. In this case, the concentration of organic matter in the infiltrating water is high, even close to the solubility, which can be regarded as a constant flow and constant concentration infiltration process. Due to construction excavation or filling, ground compaction or hardening, horizontal drainage and other engineering measures and the natural sedimentary characteristics of the strata, the stratigraphic structure of actual contaminated sites, landfills, etc. usually has horizontal layered heterogeneous characteristics. This study sets the simulation area as a vertical layered heterogeneous profile, and the entire research range is divided into upper and lower layers with different lithologies. The surface polycyclic aromatic hydrocarbon organic pollutant water enters the groundwater through the double-layer heterogeneous vadose zone soil, one-dimensional vertical flow, the upper boundary is a given flow and a given concentration condition, and the stable flux of pollutants at the vadose zone-groundwater interface is used as the associated characteristic variable.

[0111] Specifically, the selected organic pollutants are polycyclic aromatic hydrocarbons represented by benzo[a]pyrene. Such pollutants are semi-volatile pollutants with limited volatilization effect in a relatively closed underground environment, so the volatilization effect of pollutants is not considered in this study.

[0112] Specifically, CHEMFLO-2000 software is used. CHEMFLO-2000 is a numerical simulation system used to simulate water flow and solute transport in the unsaturated zone. The water flow process is described by the Richard equation, see formula (1); and the solute migration and transformation process is characterized by the convection diffusion and adsorption degradation equations, see formula (2). The two are coupled and solved by the finite difference method.

[0113]

[0114] Where: h is the matrix potential; t is the time; x is the distance from the surface downward; K(h) is the hydraulic conductivity coefficient under the matrix potential condition.

[0115] The partial differential governing equation for solute transport considering linear adsorption and first-order degradation has the general form:

[0116]

[0117] In the formula, t is time; θ is soil moisture content; R is the hysteresis coefficient of solute in soil, which is expressed as: ρ=ρ(x) is soil density, k is the distribution coefficient; C is the concentration of solute in water; x is the distance from the surface downward; D is the hydrodynamic diffusion coefficient (including molecular diffusion and mechanical dispersion); q=q(x,t) is the water flow rate; α=α(x) is the first-order degradation coefficient of microorganisms in water.

[0118] The Van Genuchten model, see formula (3), is used to describe the soil moisture characteristic curve, which represents the functional relationship between water content and matrix potential and other hydraulic parameters.

[0119]

[0120] In the formula: θ is soil moisture content; h is matrix potential; θs is saturated moisture content; θr is residual moisture content; α, m, n are VG model empirical coefficients, where m = 1-1 / n.

[0121] Step S32: Based on the established conceptual model, the numerical model and the numerical simulation system, determine the parameters that affect the spatiotemporal evolution of the selected environmental associated characteristic variables (see Table 1), and form a parameter set X = [X1, X2, ..., Xn], where n is the total number of parameters considered, n = 16; consider the variable range of each parameter, select the variation interval of each parameter to be considered (see Table 1), and form a parameter interval set V = [V1, V2, ..., Vn] 16 ].

[0122] Step S33, select the change level of each parameter (see Table 1 below for input parameter level and value table), apply the orthogonal experimental design method, obtain typical scenarios of different parameter combinations of the groundwater environment system, wherein the number of orthogonal experimental design scenarios meets the non-overfitting condition of statistical modeling;

[0123] Table 1 Model input parameter levels and values

[0124]

[0125] Note: F represents pollutant flux; p represents the percentage of upper layer thickness; T represents the thickness of the vadose zone; K upper and K lower denote the permeability coefficients of the upper and lower layers, respectively; and represent the saturated water content of the upper and lower layers respectively; and Respectively represent the saturated water content of the upper and lower layers; α upper and α lower Represent the distribution coefficients of the upper layer and the lower layer respectively; n upper and n lower represent the exponential parameters in the upper and lower layers of the Van Genuchten model, respectively; Q represents the pollutant flow; C represents the pollutant concentration; D represents the dispersion; k d Represents the degradation coefficient.

[0126] Step S34: for groundwater environment-related characteristic variables, based on numerical simulation, obtain simulated values ​​Y of groundwater environment-related characteristic variables in typical scenarios of orthogonal experimental design to form a statistical modeling data set for deterministic scenario sampling.

[0127] Step S35, add 20 different parameter combination scenario samplings, apply the numerical simulation system to obtain the corresponding characteristic variable simulation values, integrate them with the statistical modeling data set of the aforementioned deterministic scenario sampling to form a statistical modeling data set, take the natural logarithm after the characteristic variable is divided by 100, select appropriate data processing methods (no conversion, proportional scaling, logarithmic conversion, etc.) for other model input parameter variables, and establish and optimize the multivariate regression statistical prediction model of Y and X as follows:

[0128]

[0129] Where: F represents pollutant flux; p represents the proportion of upper layer thickness; T represents the thickness of the vadose zone; K upper and K lower denote the permeability coefficients of the upper and lower layers, respectively; and represent the saturated water content of the upper and lower layers respectively; and Respectively represent the saturated water content of the upper and lower layers; α upper and α lower Represent the distribution coefficients of the upper layer and the lower layer respectively; n upper and n lower represent the exponential parameters in the upper and lower layers of the Van Genuchten model, respectively; Q represents the pollutant flow; C represents the pollutant concentration; D represents the dispersion; k d Represents the degradation coefficient.

[0130] Step S36, further select the number of random sampling 35, generate random numbers of parameters again and form corresponding new more parameter combination scenarios, apply the numerical simulation system to obtain the corresponding groundwater environment associated characteristic variable simulation values, integrate and form a model verification data set, apply the established regression statistical prediction model to obtain the prediction error value and its distribution characteristics, and further determine the prediction error random variable distribution interval (-0.41, 0.40) of the groundwater environment associated characteristic variable statistical prediction model corresponding to a given confidence level.

[0131] Step S37, further select the random sampling quantity 1000, generate the parameter random number again and form the corresponding new more parameter combination scenario realization, and apply the aforementioned multivariate regression statistical prediction model to obtain the corresponding groundwater environment associated characteristic variable prediction value Y SBased on the distribution interval of the prediction error random variable corresponding to the given confidence level of the statistical prediction model of the groundwater environment-related characteristic variables determined above (-0.41, 0.40), the upper and lower limits of the true value of the groundwater environment-related characteristic variables under the given confidence level, namely the prediction confidence interval (Y P1 ,Y P2 ),in:

[0132] Y P1 =Y S / (1+0.4)

[0133] Y P2 =Y S / (1-0.41)

[0134] This random parameter combination scenario and the characteristic variable prediction interval together constitute the statistical analysis enhanced modeling data set.

[0135] Step S38: set the environmental diagnosis threshold Vc=-14.2 (equivalent to the pollutant stable flux threshold of 6.8×10 -5 ).

[0136] Step S39: For the statistical modeling data set and the model verification data set, the output label value can be directly obtained by comparing the threshold value with the size of the environment-related characteristic value; and for the statistical analysis enhanced modeling data set, the output label value is determined by comparing the relative size of the threshold value with the upper and lower limits of the prediction interval of the environment-related characteristic variable; the parameter combination scenarios corresponding to the statistical modeling data set and the statistical analysis enhanced modeling data set and their corresponding output label values ​​together constitute the SVM machine learning training set, and the parameter combination scenarios corresponding to the model verification data set and their corresponding output label values ​​together constitute the SVM machine learning test set.

[0137] Step S40, set the SVM machine learning modeling accuracy to 1.0, preset the training accuracy and test accuracy to 1.0, select the radial basis function (RBF) kernel, consider optimizing related parameters including the regularization parameter C and the data mapping complexity parameter gamma, use the grid search method for modeling optimization, and avoid overfitting.

[0138] Step S41, further obtaining a preferred SVM environmental diagnosis model for a single specific threshold Vc, whose corresponding accuracy (including training accuracy and test accuracy) is less than 0.9.

[0139] Step S42: further select the random sampling quantity of 10,000, generate random numbers of parameters again and form corresponding new more parameter combination scenarios, and apply the aforementioned multivariate regression statistical prediction model to obtain the corresponding groundwater environment associated characteristic variable prediction value Y SBased on the distribution interval of the prediction error random variable corresponding to the given confidence level of the statistical prediction model of the groundwater environment-related characteristic variables determined above (-0.41, 0.40), the upper and lower limits of the true value of the groundwater environment-related characteristic variables under the given confidence level, namely the prediction confidence interval (Y P1 ,Y P2 ),in:

[0140] Y P1 =Y S / (1+0.4)

[0141] Y P2 =Y S / (1-0.41)

[0142] This random parameter combination scenario and the characteristic variable prediction interval together constitute the statistical analysis enhanced modeling data set.

[0143] Step S43, repeating steps S38 to S39, achieving the preset training accuracy and test accuracy of 1.0, and obtaining the optimal SVM environmental diagnosis model that meets the modeling accuracy requirements for a single specific threshold Vc.

[0144] Step S44: Select 12 typical threshold values ​​V according to the threshold variation interval. C-k (k=1,2,……12)=(-8,-9,…,-19), for each V C-k Repeat step S39 to set each V C-i The corresponding statistical analysis enhanced modeling data set adds a new input parameter V C-k The sets are then combined to form an SVM machine learning training set containing threshold parameters, and the corresponding SVM machine learning test set is obtained at the same time.

[0145] Step 45, repeat step S40 to perform machine learning to further obtain the optimal SVM environmental diagnosis model containing threshold parameters.

[0146] Step S46, based on the results of a specific groundwater environmental system survey in a study area, confirm that the specific application scenario meets the requirements of the defined typical condition framework, and then obtain the specific parameter values ​​of the corresponding input parameters of the corresponding SVM environmental diagnosis model, give the groundwater pollutant stable flux risk threshold as the environmental diagnosis threshold (-14.2), and obtain accurate prediction results by using the SVM environmental diagnosis model obtained by the above-mentioned single threshold universal SVM environmental diagnosis model determination method, or directly apply the established universal SVM environmental diagnosis model containing threshold parameters to quickly and easily obtain accurate prediction results. The environmental diagnosis results are all groundwater pollution in which the stable flux of pollutants can cause the groundwater quality to be inferior to the third-class water quality standard.

[0147] See also Figure 3 , Figure 3 is a schematic block diagram of an environmental diagnosis device based on numerical simulation and SVM efficient modeling according to an exemplary embodiment of the present invention, see Figure 3 The environmental diagnosis device 100 based on numerical simulation and SVM efficient modeling includes:

[0148] The conceptual model building module 101 is used to select the associated characteristic variables of the groundwater environment and build a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under a typical condition framework; select and obtain a numerical simulation system based on mathematical control equations and numerical models;

[0149] The numerical simulation module 102 is used to obtain n model input parameter variables that affect the spatiotemporal evolution of the associated characteristic variables based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, wherein the specific values ​​of the model input parameter variables are model input parameters; select the variation interval of each model input parameter variable and then collect them to obtain a model input parameter variable interval set; obtain the corresponding numerical simulation values ​​of the typical scenario and scenario sampling based on the numerical simulation system;

[0150] The statistical model building module 103 is used to obtain different typical scenarios composed of different combinations of model input parameters of the groundwater environment system according to a given preset number of typical scenarios; obtain corresponding numerical simulation values ​​of different typical scenarios based on the numerical simulation system, and obtain a statistical modeling data set by combining the model input parameter combinations of all different typical scenarios and their corresponding associated characteristic variable simulation value sets; establish and optimize a multivariate regression statistical prediction model of associated characteristic variables and model input parameter variables according to the statistical modeling data set;

[0151] The statistical model prediction error analysis module 104 is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a first set of data-enhanced scenario parameter combinations, obtain a first set of data-enhanced scenario simulation values ​​of associated characteristic variables corresponding to each scenario through a numerical simulation system, and use the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​as a model verification data set; based on the model verification data set, obtain the prediction error value of the associated characteristic variable and its distribution characteristics through a multivariate regression statistical prediction model; determine the distribution interval of the prediction error random variable corresponding to a given preset confidence according to the prediction error value and its distribution characteristics;

[0152] The machine learning enhanced data set generation module 105 is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a second set of data enhanced scenario parameter combinations, obtain a second set of data enhanced scenario prediction values ​​for associated feature variables corresponding to each scenario through a multivariate regression statistical prediction model; determine the associated feature variable prediction confidence interval based on the second set of data enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to a given preset confidence; obtain a preset number of scenario parameter combinations and prediction confidence intervals based on n model input parameter variables and then merge them to obtain a statistical analysis enhanced modeling data set;

[0153] It is also used for obtaining the first group of output label values, the second group of output label values ​​and the third group of output label values ​​of the SVM classification machine learning in sequence according to the statistical modeling data set, the statistical analysis enhanced modeling data set and the model verification data set if a single environmental diagnosis threshold is given; the statistical modeling data set, the statistical analysis enhanced modeling data set and the model verification data set correspond to a preset number of scenario parameter combinations, respectively, and the first group of output label values, the second group of output label values ​​and the third group of output label values ​​corresponding to each other are merged to obtain a single threshold SVM classification machine learning data set;

[0154] It is also used for selecting different thresholds according to the preset threshold value variation interval and the preset threshold value quantity if multiple environmental diagnosis thresholds are given, adding the environmental diagnosis threshold as a model input parameter variable of SVM classification machine learning modeling based on the statistical modeling data set and the statistical analysis enhanced modeling data set, and based on the model verification data set, thereby obtaining the SVM classification machine learning data set containing the threshold parameter variable;

[0155] The universal SVM environmental diagnosis model construction module 106 is used to obtain the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model with threshold parameter variable according to the single threshold SVM classification machine learning data set and the SVM classification machine learning data set with threshold parameter variable according to and applying the SVM classification model modeling principle and optimization method;

[0156] The application scenario environmental diagnosis module 107 obtains specific values ​​of model parameters according to the actual conditions and collected data of the specific groundwater environmental system for actual application scenarios that meet the typical condition framework, and obtains the output label value corresponding to the threshold through a single threshold SVM environmental diagnosis model, and the output label value is the environmental diagnosis prediction result, and / or obtains the environmental diagnosis prediction result through an SVM environmental diagnosis model with threshold parameters according to the actual conditions and collected data of the specific groundwater environmental system and the preset environmental diagnosis threshold.

[0157] It should be noted that the device provided in this embodiment is applicable to scenarios including, but not limited to, diagnosis of groundwater environment systems.

[0158] It can be understood that the device provided in this embodiment establishes an SVM-based environmental diagnosis model for complex groundwater environmental systems under a typical condition framework based on the associated characteristic variables of the groundwater environment. The SVM-based environmental diagnosis model can simply realize reliable diagnosis and prediction of the groundwater environmental status, evolution process and related impacts in different application scenarios under the premise of meeting a given condition framework, thereby improving the accuracy, efficiency and practicality of environmental diagnosis and risk assessment of complex groundwater systems.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0160] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0161] It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0162] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0163] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above 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.

[0164] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An environmental diagnosis method based on numerical simulation and SVM efficient modeling, characterized in that: include: Select the associated characteristic variables of the groundwater environment and construct a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the framework of typical conditions; Select and obtain the numerical simulation system based on the mathematical control equations and numerical models; Based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, n model input parameter variables that affect the spatiotemporal evolution of the associated characteristic variables are obtained, wherein the specific values ​​of the model input parameter variables are model input parameters; the variation interval of each model input parameter variable is selected and then the model input parameter variable interval set is obtained; According to a given preset number of typical scenarios, different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system are obtained; based on the numerical simulation system, corresponding associated characteristic variable simulation values ​​of the different typical scenarios are obtained, and a statistical modeling data set is obtained by combining all the model input parameter combinations of the different typical scenarios and their corresponding associated characteristic variable simulation value sets; a multivariate regression statistical prediction model of the associated characteristic variables and the model input parameter variables is established and optimized according to the statistical modeling data set; Based on n input parameter variables of the model, a preset number of scenario samples are obtained to obtain a first set of data-enhanced scenario parameter combinations, and a first set of data-enhanced scenario simulation values ​​corresponding to associated characteristic variables of each scenario are obtained through the numerical simulation system, and the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​are combined as a model verification data set; based on the model verification data set, the prediction error value of the associated characteristic variable and its distribution characteristics are obtained through the multivariate regression statistical prediction model; according to the prediction error value and its distribution characteristics, a distribution interval of the prediction error random variable corresponding to a given preset confidence is determined; Based on n input parameter variables of the model, a preset number of scenario samples are obtained to obtain a second set of data-enhanced scenario parameter combinations, and a second set of data-enhanced scenario prediction values ​​corresponding to associated characteristic variables of each scenario are obtained through the multivariate regression statistical prediction model; and the prediction confidence interval of the associated characteristic variables is determined according to the second set of data-enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to the given preset confidence level; Based on the n model input parameter variables, a preset number of scenario parameter combinations and the prediction confidence interval are obtained and then combined to obtain a statistical analysis enhanced modeling data set; If a single environmental diagnostic threshold is given, then the first set of output label values, the second set of output label values ​​and the third set of output label values ​​of the SVM classification machine learning are obtained in sequence according to the statistical modeling data set, the statistical analysis enhanced modeling data set and the model verification data set; The statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set correspond to a preset number of scenario parameter combinations, respectively, and are merged with the first group of output label values, the second group of output label values, and the third group of output label values ​​respectively corresponding to each other to obtain a single threshold SVM classification machine learning data set; If multiple environmental diagnosis thresholds are given, different thresholds are selected according to the preset threshold value variation interval and the preset threshold value number, and based on the statistical modeling data set and the statistical analysis enhanced modeling data set, as well as based on the model verification data set, the environmental diagnosis threshold is added as a model input parameter variable for SVM classification machine learning modeling, thereby obtaining an SVM classification machine learning data set containing threshold parameter variables; According to the single threshold SVM classification machine learning data set and the SVM classification machine learning data set containing threshold parameters, according to and applying the SVM classification model modeling principle and optimization method, the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model containing threshold parameters are obtained in turn; For actual application scenarios that meet the typical condition framework, the specific values ​​of the model parameters are obtained according to the actual conditions and collected data of the specific groundwater environment system, and the output label value corresponding to the threshold is obtained through the single threshold SVM environmental diagnosis model, and the output label value is the environmental diagnosis prediction result, and / or based on the actual conditions and collected data of the specific groundwater environment system and the preset environmental diagnosis threshold, the environmental diagnosis prediction result is obtained through the SVM environmental diagnosis model containing threshold parameters.

2. The diagnostic method according to claim 1, characterized in that The step of obtaining a statistical modeling data set by combining the model input parameters of all the different typical scenarios and the corresponding associated characteristic variable simulation value sets comprises: Selecting the change level of each model input parameter; obtaining different typical scenarios consisting of different model input parameter combinations of the groundwater environment system through an orthogonal test method or a full combination method of input parameters, which is deterministic scenario sampling; Based on the numerical simulation system, corresponding simulation values ​​of the associated characteristic variables of the different typical scenarios are obtained; according to different model input parameter combinations of the different typical scenarios and their corresponding simulation values, statistical modeling data sets of the deterministic scenario sampling are obtained; and obtaining a preset number of random scenario samples based on random values ​​of n model input parameter variables, that is, obtaining a certain number of different random scenario model input parameter combinations, obtaining the simulated value of each random scenario-related characteristic variable through the numerical simulation system, and combining the random scenario model parameter combinations and the simulated value sets of the random scenario-related characteristic variables to obtain the statistical modeling data set of the random scenario samples; And the statistical modeling data set of the deterministic scenario sampling and the statistical modeling data set of the random scenario sampling are combined to obtain a statistical modeling data set integrating the deterministic scenario sampling and the random scenario sampling.

3. The diagnostic method according to claim 1, characterized in that: The SVM classification machine learning data set includes an SVM classification machine learning training set and an SVM classification machine learning test set; the method further includes: The statistical modeling data set and the statistical analysis enhanced modeling data set correspond to a preset number of scenario parameter combinations, respectively, and are merged with the corresponding first group of output label values ​​and the second group of output label values ​​to obtain a single threshold SVM classification machine learning training set; the model verification data set corresponds to a preset number of scenario parameter combinations and the third group of output label values ​​to obtain a single threshold SVM classification machine learning test set; If multiple environmental diagnostic thresholds are given, different thresholds are selected according to the preset threshold value variation range and the preset threshold value number. Based on the statistical modeling data set and the statistical analysis enhanced modeling data set, as well as based on the model verification data set, the environmental diagnostic threshold is used as a new model input parameter variable to obtain a training set and a test set of SVM classification machine learning containing threshold parameters, respectively.

4. The diagnostic method according to claim 1, characterized in that: Also includes: Given a preset value of the modeling accuracy of the SVM environmental diagnostic model, including a preset value of training accuracy and a preset value of testing accuracy, if the obtained modeling accuracy of the SVM environmental diagnostic model is lower than the preset value, then the preset number of different given typical scenarios and / or the preset number of different scenario samples are increased to perform SVM environmental diagnostic modeling optimization to obtain a corresponding accuracy SVM environmental diagnostic model; the corresponding accuracy SVM environmental diagnostic model includes a single threshold SVM environmental diagnostic model and a threshold parameter SVM environmental diagnostic model.

5. The diagnostic method according to claim 1, characterized in that: The associated characteristic variables are at least one; the associated characteristic variables include: groundwater pollutant concentration, saturated zone pollutant cross-section or interface flux, pollution plume migration and diffusion front position, pollution plume stability time, stable pollution plume distribution area, stable pollution plume front distance from the pollution source, stable pollution plume average concentration and pollution plume dissipation time; As well as soil pollutant concentration, gas-phase pollutant concentration in the vadose zone, pollutant flux in the vadose zone section, pollutant interface flux between the vadose zone and saturated zone, pollutant flux at the soil-air interface, pollutant flux at the rock-air interface and the spatiotemporal distribution of NAPL pollutants.

6. The diagnostic method according to claim 1, characterized in that: The construction of the conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the typical condition framework includes: Determine the characteristics of the aeration zone, wherein the characteristics of the aeration zone include medium type, lithological structure, medium heterogeneity and anisotropy; Determine the characteristics of the saturated zone, wherein the characteristics of the saturated zone include medium type, water-bearing system structure, seepage dimension, flow pattern, medium heterogeneity and anisotropy characteristics; Determine pollution source characteristics, including the phase state of the pollutants of interest, the type of pollutants of interest, the spatial distribution characteristics of the pollution sources, source intensity changes and dynamic characteristics; Determining boundary condition characteristics, wherein the boundary condition characteristics include hydrodynamic field boundary condition characteristics and hydrochemical field boundary condition characteristics; Initial condition characteristics are determined, wherein the initial condition characteristics include initial condition characteristics of the hydrodynamic field and initial condition characteristics of the hydrochemical field.

7. The diagnostic method according to claim 1, characterized in that: Also includes: The associated characteristic variables and the model input parameter variables are processed by any one or any combination of preset data processing methods, and the preset data processing methods include: no conversion, proportional scaling, logarithmic conversion, positive and negative consistency and normalization.

8. The diagnostic method according to claim 1, characterized in that: The step of obtaining a preset number of scenario samples based on n model input parameters includes: Select the number of random samples, determine or select the random distribution type and corresponding distribution parameters applicable to each model input parameter, complete the random number generation of the model input parameter and thus form a new parameter combination scenario.

9. The diagnostic method according to claim 1, characterized in that: The method of obtaining the prediction error value of the associated characteristic variable and its distribution characteristics through the multivariate regression statistical prediction model; and determining the distribution interval of the prediction error random variable corresponding to a given preset confidence according to the prediction error value and its distribution characteristics, comprises: Based on the model validation set data, the multivariate regression statistical prediction model is used to obtain a prediction error value data set, wherein the prediction error value data set includes multiple prediction errors; the mean and variance of the prediction error random variable are estimated, and its probability density function is fitted and identified, and the sum of the cumulative probabilities on both sides is determined according to the preset confidence and the distribution interval of the prediction error random variable with the corresponding confidence is obtained.

10. The diagnostic method according to claim 1, characterized in that: The SVM classification model modeling optimization method in the SVM classification model modeling principle and optimization method includes: Optimization of SVM classification model modeling optimization method and modeling parameters; The modeling parameters preferably include Optimization of kernel function types and their associated parameters; The kernel function types preferably include linear kernel, polynomial kernel, radial basis function kernel and sigmoid kernel; The optimization of the radial basis function kernel-related parameters includes the optimization of the regularization parameter C and the data mapping complexity parameter gamma; The SVM modeling optimization method preferably includes a grid search method, a particle swarm optimization method, and a genetic particle swarm hybrid algorithm.

11. The diagnostic method according to claim 1, characterized in that: The obtaining of the first group of output label values, the second group of output label values ​​and the third group of output label values ​​of the SVM classification machine learning comprises: If each of the associated characteristic variables corresponds to a single environmental diagnostic threshold, then based on the statistical modeling data set and the model verification data set, compare the relative size of each of the associated characteristic variable simulation values ​​and the corresponding single environmental diagnostic threshold, and obtain the first group of output label values ​​and the third group of output label values ​​of the SVM classification machine learning in turn; based on the statistical analysis enhanced modeling data set, if the predicted confidence interval of each of the associated characteristic variables does not include the corresponding single environmental diagnostic threshold, then based on the relative size of the upper and lower limits of the predicted confidence interval and the corresponding single environmental diagnostic threshold, obtain the second group of output label values ​​of the SVM classification machine learning.

12. An environmental diagnosis device based on numerical simulation and SVM efficient modeling, characterized in that: The device comprises: The conceptual model building module is used to select the associated characteristic variables of the groundwater environment and build a conceptual model of the spatiotemporal evolution of the corresponding groundwater environment system under the framework of typical conditions; select and obtain a numerical simulation system based on mathematical control equations and numerical models; A numerical simulation module is used to obtain n model input parameter variables that affect the spatiotemporal evolution of associated characteristic variables based on the spatiotemporal evolution conceptual model, the numerical model and the numerical simulation system, wherein the specific values ​​of the model input parameter variables are model input parameters; select the variation interval of each model input parameter variable and then collect them to obtain a model input parameter variable interval set; and obtain the corresponding numerical simulation values ​​of the typical scenario and scenario sampling based on the numerical simulation system; A statistical model building module is used to obtain different typical scenarios consisting of different combinations of model input parameters of the groundwater environment system based on a given preset number of typical scenarios; obtain corresponding associated characteristic variable simulation values ​​of the different typical scenarios based on the numerical simulation system, and obtain a statistical modeling data set by combining all the model input parameter combinations of the different typical scenarios and their corresponding associated characteristic variable simulation value sets; establish and optimize a multivariate regression statistical prediction model of the associated characteristic variables and the model input parameter variables based on the statistical modeling data set; The statistical model prediction error analysis module is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a first set of data-enhanced scenario parameter combinations, obtain a first set of data-enhanced scenario simulation values ​​of associated characteristic variables corresponding to each scenario through the numerical simulation system, and use the first set of data-enhanced scenario parameter combinations and the first set of data-enhanced scenario simulation values ​​as a model verification data set; based on the model verification data set, obtain the prediction error value of the associated characteristic variable and its distribution characteristics through the multivariate regression statistical prediction model; determine the distribution interval of the prediction error random variable corresponding to a given preset confidence according to the prediction error value and its distribution characteristics; A machine learning enhanced data set generation module is used to obtain a preset number of scenario samples based on n model input parameter variables, obtain a second set of data enhanced scenario parameter combinations, obtain a second set of data enhanced scenario prediction values ​​for associated feature variables corresponding to each scenario through the multivariate regression statistical prediction model; determine the associated feature variable prediction confidence interval based on the second set of data enhanced scenario prediction values ​​and the prediction error random variable distribution interval corresponding to the given preset confidence; obtain a preset number of scenario parameter combinations based on the n model input parameter variables and the prediction confidence interval and then merge them to obtain a statistical analysis enhanced modeling data set; It is also used for obtaining, if a single environmental diagnostic threshold is given, a first group of output label values, a second group of output label values, and a third group of output label values ​​of SVM classification machine learning in sequence according to the statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set; merging the statistical modeling data set, the statistical analysis enhanced modeling data set, and the model verification data set corresponding to a preset number of scenario parameter combinations, respectively, with the first group of output label values, the second group of output label values, and the third group of output label values ​​respectively corresponding to obtain a single threshold SVM classification machine learning data set; It is also used for, if multiple environmental diagnosis thresholds are given, selecting different thresholds according to the preset threshold value variation interval and the preset threshold value number, adding the environmental diagnosis threshold as a model input parameter variable of SVM classification machine learning modeling based on the statistical modeling data set and the statistical analysis enhanced modeling data set, and based on the model verification data set, thereby obtaining the SVM classification machine learning data set containing the threshold parameter variable; A universal SVM environmental diagnosis model construction module is used to obtain the corresponding single threshold SVM environmental diagnosis model and the SVM environmental diagnosis model containing threshold parameters in turn according to the single threshold SVM classification machine learning data set and the SVM classification machine learning data set containing threshold parameters, based on and applying the SVM classification model modeling principle and optimization method; The application scenario environmental diagnosis module obtains the specific values ​​of the model parameters according to the actual conditions and collected data of the specific groundwater environmental system for actual application scenarios that meet the typical condition framework, and obtains the output label value of the corresponding threshold through the single threshold SVM environmental diagnosis model, and the output label value is the environmental diagnosis prediction result, and / or obtains the environmental diagnosis prediction result through the SVM environmental diagnosis model containing threshold parameters according to the actual conditions and collected data of the specific groundwater environmental system and the preset environmental diagnosis threshold.

Images