A method for grading the accuracy of polluted land survey

Abstract

The present invention provides a method for grading the survey accuracy of polluted land, comprising: step 10, constructing a polluted site data attribute table; step 20, using the polluted site data attribute table, training to obtain a risk impact model; step 30, according to the risk impact model, obtaining the weight value of the influencing factor; step 40, obtaining a grading score table of the influencing factor; step 50, comparing the grading score table of the influencing factor, according to the factor data corresponding to the influencing factor of the target polluted site, assigning points to the influencing factor to obtain the score value of the influencing factor; according to the weight value and score value of the influencing factor, obtaining the risk index of the target polluted site; step 60, according to the risk index of the target polluted site, grading and dividing the survey accuracy of the target polluted site. The present invention provides a method for grading the survey accuracy of polluted land, which realizes the refined survey of different types of polluted land, reduces the survey cost, and improves the remediation efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of soil pollution control and treatment, and in particular, relates to a method for grading the accuracy of polluted land survey. Background Art

[0002] In recent years, my country has gradually formed a soil pollution investigation method and technical system, but it is difficult to meet the needs of detailed investigation of different types of sites in different regions in terms of investigation accuracy and point distribution methods. my country's current technical guidelines limit the sampling grid, without taking into account the specific production conditions and hydrogeological conditions of different types of plots. The traditional random sampling and grid sampling methods have weak point representativeness, resulting in the coexistence of many invalid points and missed points during the investigation process, making it difficult to accurately depict the degree and scope of site pollution. To a certain extent, it may lead to the omission of contaminated areas of the plots, and may also lead to an increase in contaminated areas, greatly reducing the efficiency of remediation of contaminated sites. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for grading the survey accuracy of polluted plots, so as to realize the refined survey of different types of polluted plots, reduce the survey cost and improve the restoration efficiency.

[0004] To solve the above technical problems, the embodiments of the present invention adopt the following technical solutions:

[0005] The embodiment of the present invention provides a method for grading the accuracy of polluted land survey, comprising the following steps:

[0006] Step 10, constructing a contaminated site data attribute table, wherein the contaminated site data attribute table includes influencing factors, factor data samples corresponding to the influencing factors, result indicators, and result data samples corresponding to the result indicators;

[0007] Step 20, using the contaminated site data attribute table, training at least two machine learning coupling models, and selecting the trained machine learning coupling model with the best preset evaluation index as the risk impact model;

[0008] Step 30, obtaining weight values ​​of influencing factors according to the risk impact model;

[0009] Step 40, obtaining a grading score table of the influencing factors according to the weight values ​​of the influencing factors;

[0010] Step 50, comparing the graded scoring table of influencing factors, assigning scores to the influencing factors according to the factor data corresponding to the influencing factors of the target contaminated site, and obtaining the score values ​​of the influencing factors; obtaining the risk index of the target contaminated site according to the weight values ​​and score values ​​of the influencing factors;

[0011] Step 60: classify the target contaminated site into different levels of survey accuracy according to the risk index of the target contaminated site.

[0012] As a further improvement of an embodiment of the present invention, the influencing factors include: industry category, years of production, land area, area of ​​key areas, toxicity of pollutants, mobility of pollutants, volatility of pollutants, presence of persistent pollutants, type of soil medium in the aeration zone, type of medium in the saturated zone soil layer, groundwater depth, rainfall in the plot, terrain slope, distribution of surrounding sensitive targets, emergency management, anti-seepage measures, distribution of groundwater pipelines and facilities, wastewater and exhaust gas monitoring, soil and groundwater monitoring, incidence rate of pollution accidents in the plot and surrounding environment, and abnormal conditions on site.

[0013] As a further improvement of the embodiment of the present invention, the result indicators include the type of exceeding standard factor, the sample exceeding standard rate and the maximum exceeding standard multiple.

[0014] As a further improvement of the embodiment of the present invention, in step 20, the preset evaluation indicators include determination coefficient, root mean square error and mean absolute error.

[0015] As a further improvement of the embodiment of the present invention, a trained gradient boosting decision tree model, an extreme gradient boosting model or a random forest model is selected as the risk impact model.

[0016] As a further improvement of the embodiment of the present invention, the step 40 specifically includes:

[0017] Each influencing factor is graded and divided respectively, and each grade is scored according to the weight value of the influencing factor to obtain a graded score table for each influencing factor.

[0018] As a further improvement of the embodiment of the present invention, in step 50, the risk index of the target contaminated site is calculated using formula (1) according to the weight value and score value of the influencing factor:

[0019] X = A × a + B × b + C × c + D × d + E × e + F × f + G × g + H × h + I × i + J × j + K × k + L × l + M × m + N × n + O × o + P × p + Q × q + R × r + S × s + T × t + W × w Formula (1)

[0020] In the formula, X represents the risk index of the target contaminated site; A represents the weight value of the industry category, and a represents the score value of the industry category; B represents the weight value of the production years, and b represents the score value of the production years; C represents the weight value of the land area, and c represents the score value of the land area; D represents the weight value of the key area, and d represents the score value of the key area; E represents the weight value of the pollutant toxicity, and e represents the score value of the pollutant toxicity; F represents the weight value of the pollutant mobility, and f represents the score value of the pollutant mobility; G represents the weight value of the pollutant volatility, and g represents the score value of the pollutant volatility; H represents the weight value of whether there are persistent pollutants, and h represents the score value of whether there are persistent pollutants; I represents the weight value of the medium type of the vadose zone soil, and i represents the score value of the medium type of the vadose zone soil; J represents the weight value of the medium type of the saturated zone soil layer, and j represents the score value of the medium type of the saturated zone soil layer; K represents the weight value of the groundwater depth , k represents the scoring value of groundwater depth; L represents the weight value of rainfall in the plot, l represents the scoring value of rainfall in the plot; M represents the weight value of terrain slope, m represents the scoring value of terrain slope; N represents the weight value of distribution of surrounding sensitive targets, n represents the scoring value of distribution of surrounding sensitive targets; O represents the weight value of emergency management, o represents the scoring value of emergency management; P represents the weight value of anti-seepage measures, p represents the scoring value of anti-seepage measures; Q represents the weight value of groundwater pipelines and facilities, q represents the scoring value of groundwater pipelines and facilities; R represents the weight value of wastewater and waste gas monitoring, r represents the scoring value of wastewater and waste gas monitoring; S represents the weight value of soil and groundwater monitoring, s represents the scoring value of soil and groundwater monitoring; T represents the weight value of the occurrence rate of pollution accidents in the plot and surrounding environment, t represents the scoring value of the occurrence rate of pollution accidents in the plot and surrounding environment; W represents the weight value of abnormal conditions on site, w represents the scoring value of abnormal conditions on site.

[0021] As a further improvement of the embodiment of the present invention, the step 60 specifically includes:

[0022] According to the graded scoring table of influencing factors, the maximum risk index of regional contaminated sites is obtained;

[0023] If the risk index of the target contaminated site is less than one-third of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is low, and the survey point density in the key area of ​​the target contaminated site shall not be less than 40m×40m;

[0024] If the risk index of the target contaminated site is greater than or equal to one-third of the maximum risk index of the regional contaminated site and less than two-thirds of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is medium, and the survey point density of the key area of ​​the target contaminated site shall not be less than 20m×20m;

[0025] If the risk index of the target contaminated site is greater than or equal to two-thirds of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is high, and the survey point density in the key areas of the target contaminated site shall not be less than 10m×10m.

[0026] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0027] (1) The present invention provides a method for grading the survey accuracy of polluted land, first constructing a polluted site data attribute table; then using the polluted site data attribute table, training at least two machine learning coupling models, selecting the trained machine learning coupling model with the best preset evaluation index as the risk impact model; then obtaining the weight value of the influencing factor according to the risk impact model; then obtaining the grading score table of the influencing factor according to the weight value of the influencing factor; then comparing the grading score table of the influencing factor, assigning points to the influencing factor of the target polluted site according to the factor data corresponding to the influencing factor of the target polluted site, and obtaining the score value of the influencing factor; obtaining the risk index of the target polluted site according to the weight value and score value of the influencing factor; finally, according to the risk index of the target polluted site, the survey accuracy of the target polluted site is graded and divided. The present invention adopts survey point distribution methods with different precisions for different types of polluted land, which makes up for the shortcomings of traditional grid sampling and random sampling, can realize the refinement of site survey distribution of different types of land, reduce survey costs, and improve remediation efficiency.

[0028] (2) When constructing the polluted site data attribute table, the present invention selects the distribution of pollutants in the plot as the result indicator of the polluted site, and selects the basic data information, pollutant information, sensitive receptor conditions, environmental management level of the plot and hydrogeological conditions information that affect the result indicators of the polluted site as the influencing factors, which can cover all the influences of the site production activities on the site pollution results. Therefore, the machine learning coupling model trained by the polluted site data attribute table can more accurately reflect the influence of the site influencing factors in the region on the pollution results, so that the risk index of the target polluted site is more accurate.

[0029] (3) The present invention uses the contaminated site data attribute table to train at least two machine learning coupling models, and then selects the optimal trained machine learning coupling model as the risk impact model based on preset evaluation indicators. The obtained risk impact model can accurately reflect the impact of site influencing factors in the region on the pollution results, thereby obtaining a more accurate risk index for the target contaminated site.

[0030] (4) The present invention obtains the weight value of the influencing factor through the risk impact model, obtains the grading score table of the influencing factor according to the weight value of the influencing factor, and when conducting a target contaminated site survey, the influencing factor is scored according to the factor data corresponding to the influencing factor of the target contaminated site, and the score value of the influencing factor is obtained. The risk index of the target contaminated site is obtained according to the weight value and score value of the influencing factor. Compared with the traditional risk grading method of direct assignment, the risk index obtained by the method of the present invention according to the site regional characteristics, industry characteristics, etc. is more adaptable.

[0031] (5) The present invention classifies the target contaminated sites into low, medium or high levels according to their risk index, and then conducts surveys at different densities according to the levels, thereby avoiding the coexistence of many invalid points and missed points during the survey process. The present invention can accurately characterize the degree and scope of site contamination, improve the accuracy of the survey, and further improve the efficiency of remediation. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The present invention is a flowchart of a method for grading the accuracy of a polluted land survey according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The technical solution of the present invention is described in detail below.

[0034] The embodiment of the present invention provides a method for grading the accuracy of polluted land survey, such as Figure 1 As shown, the following steps are included:

[0035] Step 10: construct a polluted site data attribute table, which includes influencing factors, factor data samples corresponding to the influencing factors, result indicators, and result data samples corresponding to the result indicators.

[0036] Step 20: Use the contaminated site data attribute table to train at least two machine learning coupling models, and select the trained machine learning coupling model with the best preset evaluation index as the risk impact model.

[0037] Step 30: Obtain weight values ​​of influencing factors according to the risk impact model.

[0038] Step 40: Obtain a grading score table of the influencing factors according to the weight values ​​of the influencing factors.

[0039] Step 50, comparing the graded scoring table of influencing factors, assigning points to the influencing factors according to the factor data corresponding to the influencing factors of the target contaminated site, and obtaining the scoring values ​​of the influencing factors; obtaining the risk index of the target contaminated site according to the weight values ​​and scoring values ​​of the influencing factors.

[0040] Step 60: classify the target contaminated site into different levels of survey accuracy according to the risk index of the target contaminated site.

[0041] In step 10, sample data of existing polluted sites are collected, including basic information of the plot, hydrogeological conditions, and pollutant distribution information of the plot. Basic information of the plot includes the production history of the site, the industry category involved, the production years, the land area, and the area of ​​key areas; pollutant information includes pollutant toxicity, pollutant mobility, pollutant volatility, and whether there are persistent pollutants; sensitive receptors include the distribution of kindergartens / schools / hospitals / drinking water sources; the environmental management level of the plot includes emergency management, anti-seepage measures, distribution of groundwater pipelines and facilities, wastewater and exhaust gas monitoring, soil and groundwater monitoring, the incidence of pollution accidents in the plot and surrounding environment, and whether there are abnormal conditions on site; hydrogeological conditions include groundwater depth, soil medium in the aeration zone, soil medium in the saturated zone, terrain slope, rainfall in the plot, etc.; pollutant distribution information of the plot includes detection factor, detection rate, exceeding factor, exceeding rate, and exceeding multiple.

[0042] Select industry category, production years, land area, key area, pollutant toxicity, pollutant mobility, pollutant volatility, whether there are persistent pollutants, aeration zone soil medium type, saturated zone soil medium type, groundwater depth, land rainfall, terrain slope, distribution of surrounding sensitive targets, emergency management, anti-seepage measures, groundwater pipeline and facility distribution, wastewater and waste gas monitoring, soil and groundwater monitoring, land and surrounding environmental pollution accident rate and site abnormal conditions as influencing factors, detection factor, detection rate, exceeding factor, exceeding rate, exceeding multiple, and select exceeding factor type, sample exceeding rate and maximum exceeding multiple as result indicators. Standardize and normalize the sample data, and correspond the sample data to the influencing factors and result indicators one by one. Standardize and encode the sample data corresponding to each influencing factor and the sample data corresponding to each result indicator respectively, and define them through data type, width and non-empty constraints to obtain factor data samples corresponding to the influencing factors and result data samples corresponding to the result indicators, forming a polluted site data attribute table.

[0043] In step 20, two or more machine learning coupling models are trained using the contaminated site data attribute table. According to the preset evaluation indicators, the trained machine learning coupling model with the best preset evaluation indicators is selected as the risk impact model. The preset evaluation indicators include the determination coefficient (R 2 ), root mean square error (RMSE) and mean absolute error (MAE). 2 The closer it is to 1, the closer the RMSE is to 1, the smaller the MAE value is, the higher the simulation accuracy is, and the better the model effect is.

[0044] This embodiment selects a trained gradient boosting decision tree model, an extreme gradient boosting model or a random forest model as the risk impact model. The above three models have the following characteristics: (1) they can screen a large number of environmental features; (2) they can prevent overfitting; (3) they are insensitive to slight changes in data; and (4) even if the environment is complex and volatile, they can still obtain better results through strong optimization and supervision capabilities.

[0045] In step 30, the weight value of the influencing factor is obtained through the risk impact model. The weight value of the influencing factor is expressed as the contribution rate to the pollution risk. The larger the weight value, the higher the impact on the pollution risk.

[0046] Step 40 specifically includes:

[0047] Each influencing factor is graded and divided respectively, and each grade is scored according to the weight value of the influencing factor to obtain a graded scoring table for each influencing factor. Specifically, an influencing factor is divided into V grades, and the weight value U is divided into V equal parts. The first grade score is U, the second grade score is U / V×(V-1), the third grade score is U / V×(V-2), and so on. For example, taking the production years as an example, through the risk impact model, the weight value of the production years is 7. The production years are divided into 3 levels: the first level, the production years ≥ 30 years, the score is 7 points; the second level, 10 years ≤ production years < 30 years, the score is 4.6 points; the third level, the production years < 10 years, the score is 2.3 points.

[0048] In step 50, the influencing factors are scored according to the factor data corresponding to the influencing factors of the target contaminated site, and the scoring values ​​of the influencing factors are obtained by comparing the grading scoring table of the influencing factors. For example, the production years of the target contaminated site are 15 years, and the scoring value of the production years of the target site is 4.6 points according to the grading scoring table of the production years.

[0049] According to the weight and score of the influencing factors, the risk index of the target contaminated site is obtained using formula (1):

[0050] X = A × a + B × b + C × c + D × d + E × e + F × f + G × g + H × h + I × i + J × j + K × k + L × l + M × m + N × n + O × o + P × p + Q × q + R × r + S × s + T × t + W × w Formula (1)

[0051] In the formula, X represents the risk index of the target contaminated site; A represents the weight value of the industry category, and a represents the score value of the industry category; B represents the weight value of the production years, and b represents the score value of the production years; C represents the weight value of the land area, and c represents the score value of the land area; D represents the weight value of the key area, and d represents the score value of the key area; E represents the weight value of the pollutant toxicity, and e represents the score value of the pollutant toxicity; F represents the weight value of the pollutant mobility, and f represents the score value of the pollutant mobility; G represents the weight value of the pollutant volatility, and g represents the score value of the pollutant volatility; H represents the weight value of whether there are persistent pollutants, and h represents the score value of whether there are persistent pollutants; I represents the weight value of the medium type of the vadose zone soil, and i represents the score value of the medium type of the vadose zone soil; J represents the weight value of the medium type of the saturated zone soil layer, and j represents the score value of the medium type of the saturated zone soil layer; K represents the weight value of the groundwater depth , k represents the scoring value of groundwater depth; L represents the weight value of rainfall in the plot, l represents the scoring value of rainfall in the plot; M represents the weight value of terrain slope, m represents the scoring value of terrain slope; N represents the weight value of distribution of surrounding sensitive targets, n represents the scoring value of distribution of surrounding sensitive targets; O represents the weight value of emergency management, o represents the scoring value of emergency management; P represents the weight value of anti-seepage measures, p represents the scoring value of anti-seepage measures; Q represents the weight value of groundwater pipelines and facilities, q represents the scoring value of groundwater pipelines and facilities; R represents the weight value of wastewater and waste gas monitoring, r represents the scoring value of wastewater and waste gas monitoring; S represents the weight value of soil and groundwater monitoring, s represents the scoring value of soil and groundwater monitoring; T represents the weight value of the occurrence rate of pollution accidents in the plot and surrounding environment, t represents the scoring value of the occurrence rate of pollution accidents in the plot and surrounding environment; W represents the weight value of abnormal conditions on site, w represents the scoring value of abnormal conditions on site.

[0052] Step 60 specifically includes:

[0053] According to the graded scoring table of influencing factors, the maximum risk index of regional contaminated sites is obtained.

[0054] Specifically, according to the weight value of the influencing factor and the highest score value of the influencing factor (i.e. the weight value of the influencing factor), the maximum risk index of the regional contaminated site is calculated using formula (2): X max =A×A+B×B+C×C+D×D+E×E+F×F+G×G+H×H+I×I+J×J+K×K+L×L+M×M+N×N+O×O+P×P+Q×Q+R×R+S×S+T×T+W×W Formula (2)

[0055] If the risk index X of the target contaminated site is less than X max / 3, the pollution risk of the target contaminated site is low, and the density of survey points in key areas of the target contaminated site shall not be less than 40m×40m, that is, at least one survey point shall be set up within 40m×40m.

[0056] If the risk index X of the target contaminated site is greater than or equal to X max / 3 and less than X max ×2 / 3, the pollution risk of the target contaminated site is medium, and the density of survey points in key areas of the target contaminated site shall not be less than 20m×20m, that is, at least one survey point shall be set up within 20m×20m.

[0057] If the risk index X of the target contaminated site is greater than or equal to X max ×2 / 3, the pollution risk of the target contaminated site is high, and the density of survey points in key areas of the target contaminated site shall not be less than 10m×10m, that is, at least one survey point shall be set up within 10m×10m.

[0058] Among them, key areas include but are not limited to production workshops, storage tank areas, raw and auxiliary material warehouses, sewage treatment areas, waste gas treatment areas, and hazardous waste warehouses.

[0059] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above specific embodiments, and the descriptions in the above specific embodiments and the specification are only for further illustrating the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed for protection.

Claims

1. A method for grading the accuracy of polluted land surveys. It is characterized in that The following steps are involved: Step 10, constructing a contaminated site data attribute table, wherein the contaminated site data attribute table includes influencing factors, factor data samples corresponding to the influencing factors, result indicators, and result data samples corresponding to the result indicators; Step 20, using the contaminated site data attribute table, training at least two machine learning coupling models, and selecting the trained machine learning coupling model with the best preset evaluation index as the risk impact model; Step 30, obtaining weight values ​​of influencing factors according to the risk impact model; Step 40, obtaining a grading score table of the influencing factors according to the weight values ​​of the influencing factors; Step 50, comparing the graded scoring table of influencing factors, assigning points to the influencing factors according to the factor data corresponding to the target contaminated site and the influencing factors, and obtaining the scoring values ​​of the influencing factors; obtaining the risk index of the target contaminated site according to the weight values ​​and scoring values ​​of the influencing factors; Step 60, classifying the target contaminated site into survey accuracy grades according to the risk index of the target contaminated site; The step 40 specifically includes: Each influencing factor is graded and divided respectively, and scores are assigned to each grade according to the weight value of the influencing factor to obtain a graded score table for each influencing factor; In step 50, the risk index of the target contaminated site is calculated using formula (1) according to the weight values ​​and score values ​​of the influencing factors: X = A × a + B × b + C × c + D × d + E × e + F × f + G × g + H × h + I × i + J × j + K × k + L × l + M × m + N × n + O × o + P × p + Q × q + R × r + S × s + T × t + W × w Formula (1) In the formula, X represents the risk index of the target contaminated site; A represents the weight value of the industry category, and a represents the score value of the industry category; B represents the weight value of the production years, and b represents the score value of the production years; C represents the weight value of the land area, and c represents the score value of the land area; D represents the weight value of the key area, and d represents the score value of the key area; E represents the weight value of the pollutant toxicity, and e represents the score value of the pollutant toxicity; F represents the weight value of the pollutant mobility, and f represents the score value of the pollutant mobility; G represents the weight value of the pollutant volatility, and g represents the score value of the pollutant volatility; H represents the weight value of whether there are persistent pollutants, and h represents the score value of whether there are persistent pollutants; I represents the weight value of the medium type of the vadose zone soil, and i represents the score value of the medium type of the vadose zone soil; J represents the weight value of the medium type of the saturated zone soil layer, and j represents the score value of the medium type of the saturated zone soil layer; K represents the weight value of the groundwater depth , k represents the scoring value of groundwater depth; L represents the weight value of rainfall in the plot, l represents the scoring value of rainfall in the plot; M represents the weight value of terrain slope, m represents the scoring value of terrain slope; N represents the weight value of the distribution of surrounding sensitive targets, n represents the scoring value of the distribution of surrounding sensitive targets; O represents the weight value of emergency management, o represents the scoring value of emergency management; P represents the weight value of anti-seepage measures, p represents the scoring value of anti-seepage measures; Q represents the weight value of groundwater pipelines and facilities, q represents the scoring value of groundwater pipelines and facilities; R represents the weight value of wastewater and waste gas monitoring, r represents the scoring value of wastewater and waste gas monitoring; S represents the weight value of soil and groundwater monitoring, s represents the scoring value of soil and groundwater monitoring; T represents the weight value of the incidence rate of pollution accidents in the plot and surrounding environment, t represents the scoring value of the incidence rate of pollution accidents in the plot and surrounding environment; W represents the weight value of abnormal conditions on site, w represents the scoring value of abnormal conditions on site; Step 60 specifically includes: According to the graded scoring table of influencing factors, the maximum risk index of regional contaminated sites is obtained; Specifically, according to the weight value of the influencing factor and the highest score value of the influencing factor (i.e. the weight value of the influencing factor), the maximum risk index of the regional contaminated site is calculated using formula (2): X max =A×A+B×B+C×C+D×D+E×E+F×F+G×G+H×H+I×I+J×J+K×K+L×L+M×M+N×N+O×O+P×P+Q×Q+R×R+S×S+T×T+W×W Formula (2) If the risk index of the target contaminated site is less than one-third of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is low, and the survey point density in the key area of ​​the target contaminated site shall not be less than 40m×40m; If the risk index of the target contaminated site is greater than or equal to one-third of the maximum risk index of the regional contaminated site and less than two-thirds of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is medium, and the survey point density of the key area of ​​the target contaminated site shall not be less than 20m×20m; If the risk index of the target contaminated site is greater than or equal to two-thirds of the maximum risk index of the regional contaminated site, the pollution risk of the target contaminated site is high, and the survey point density in the key areas of the target contaminated site shall not be less than 10m×10m.

2. The method for grading the accuracy of polluted land survey according to claim 1, It is characterized in that The influencing factors include: industry category, years of production, land area, area of ​​key areas, toxicity of pollutants, mobility of pollutants, volatility of pollutants, presence of persistent pollutants, soil medium type in the aeration zone, medium type in the saturated zone soil layer, groundwater depth, rainfall in the plot, terrain slope, distribution of surrounding sensitive targets, emergency management, anti-seepage measures, distribution of groundwater pipelines and facilities, wastewater and exhaust gas monitoring, soil and groundwater monitoring, incidence rate of pollution accidents in the plot and surrounding environment, and abnormal conditions on site.

3. The method for grading the accuracy of polluted land survey according to claim 1, It is characterized in that The result indicators include the type of exceeding standard factors, the sample exceeding standard rate and the maximum exceeding standard multiple.

4. The method for grading the accuracy of polluted land survey according to claim 1, It is characterized in that In step 20, the preset evaluation indicators include determination coefficient, root mean square error and mean absolute error.

5. The method for grading the accuracy of polluted land survey according to claim 1, It is characterized in that In step 20, a trained gradient boosting decision tree model, an extreme gradient boosting model or a random forest model is selected as the risk impact model.

Images