Method for monitoring soil pollutants in mine remediation area

Abstract

The invention relates to the technical field of pollutant monitoring, in particular to a monitoring method for soil pollutants in a mine remediation area. Comprising the following steps: acquiring a soil sample, detecting the soil sample by adopting a pollutant monitoring technology to obtain pollutant data, preprocessing the pollutant data, and intelligently processing the preprocessed pollutant data by utilizing a self-adaptive multi-dimensional nonlinear pollutant fusion algorithm to obtain comprehensive pollutant data; constructing a pollution prediction model, and performing prediction analysis on the comprehensive pollutant data by using the pollution prediction model to obtain a layered prediction result; and performing risk assessment on the pollutants based on the comprehensive pollutant data to obtain a comprehensive risk assessment value, further obtaining a comprehensive risk assessment result, and making a coping strategy based on the hierarchical prediction result and the comprehensive risk assessment result. The technical problems of insufficient pollution data processing precision and inaccurate risk assessment are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of pollutant monitoring, and particularly to a method for monitoring soil pollutants in a mine restoration area. Background Art

[0002] Mining activities have had a serious impact on the ecological environment globally, especially the problem of soil pollution is the most prominent. Soil pollution in mining areas mainly comes from harmful substances released during ore mining, ore dressing, smelting, tailings stacking, etc., including heavy metals (such as lead, cadmium, arsenic, mercury, etc.), acidic wastewater, organic pollutants (such as polycyclic aromatic hydrocarbons, petroleum hydrocarbons, etc.). These pollutants not only affect the physical and chemical properties of the soil, but also pollute groundwater through infiltration, and then endanger the ecological environment and human health. Therefore, how to effectively monitor the soil pollution status in the mine restoration area and formulate scientific and reasonable restoration strategies is one of the research focuses in the current environmental governance field.

[0003] Existing soil pollution monitoring methods mainly include artificial sampling and detection, remote sensing technology, on-line sensor detection, etc. Among them, the artificial sampling and detection method relies on laboratory analysis techniques, such as X-ray fluorescence spectroscopy (XRF), gas chromatography-mass spectrometry (GC-MS), inductively coupled plasma mass spectrometry (ICP-MS), etc. Although these methods have high detection accuracy, they have problems such as long detection cycles, inability to achieve real-time monitoring of pollutants, and limited spatial coverage. Although remote sensing technology can conduct a preliminary assessment of large-scale soil pollution, it is difficult to provide accurate pollutant concentration data and is greatly affected by factors such as surface vegetation and weather. Although on-line sensor detection technology can achieve real-time monitoring, the current detection sensitivity and stability of sensors still have certain limitations, especially in the case of coexistence of multiple pollutants in mining areas and complex terrain affecting signal propagation, it is difficult to guarantee the monitoring accuracy.

[0004] However, the above existing technologies also have technical problems of insufficient accuracy in processing pollution data and inaccurate risk assessment. Summary of the Invention

[0005] The present invention provides a method for monitoring soil pollutants in a mine restoration area to solve the technical problems of insufficient accuracy in processing pollution data and inaccurate risk assessment.

[0006] A method for monitoring soil pollutants in a mine restoration area of the present invention specifically includes the following technical solutions:

[0007] A method for monitoring soil pollutants in a mine restoration area includes the following steps:

[0008] S1. Obtain soil samples, detect the soil samples using pollutant monitoring techniques to obtain pollutant data, preprocess the pollutant data, and intelligently process the preprocessed pollutant data using an adaptive multi-dimensional non-linear pollutant fusion algorithm to obtain comprehensive pollutant data;

[0009] S2. Construct a pollution prediction model, use the pollution prediction model to perform predictive analysis on the comprehensive pollutant data to obtain a hierarchical prediction result; based on the comprehensive pollutant data, conduct a risk assessment on the pollutants to obtain a comprehensive risk assessment value, and then obtain a comprehensive risk assessment result, and formulate a response strategy based on the hierarchical prediction result and the comprehensive risk assessment result.

[0010] Preferably, the S1 specifically includes:

[0011] The adaptive multi-dimensional non-linear pollutant fusion algorithm extracts the primary pollutant characteristic data of the pollutants from the preprocessed pollutant data, and performs enhancement processing on the primary pollutant characteristic data to obtain enhanced characteristic data; and through non-linear mapping and data compression processing, optimizes the dimension of the enhanced characteristic data.

[0012] Preferably, the S1 specifically includes:

[0013] In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, feature extraction is performed on the preprocessed pollutant data to obtain the primary pollutant characteristic data of each stratified soil sample, including: surface primary pollutant characteristic data, middle layer primary pollutant characteristic data, and deep layer primary pollutant characteristic data.

[0014] Preferably, the S1 specifically includes:

[0015] In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, taking the surface primary pollutant characteristic data as an example, the surface primary pollutant characteristic data is enhanced using a non-linear enhancement method to obtain enhanced surface characteristic data.

[0016] Preferably, the S1 specifically includes:

[0017] In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, a high-order polynomial transformation is used for non-linear mapping, and the enhanced surface characteristic data is smoothed using a Gaussian radial basis function to obtain compressed surface characteristic data.

[0018] Preferably, the S2 specifically includes:

[0019] Based on historical pollutant data, combined with remote sensing images, geographic information system data, and hydro-environmental data, as input data, perform timestamp synchronization on the input data to obtain synchronized input data. After filling in the missing data of the synchronized input data, perform outlier identification and data standardization to obtain standardized input data. Then, use the adaptive multi-dimensional non-linear pollutant fusion algorithm in step S1 to intelligently process the standardized input data to obtain fused data.

[0020] Preferably, S2 specifically includes:

[0021] Based on the fused data, construct a pollution prediction model, use the pollution prediction model to predict the fused data, and train and validate the pollution prediction model to obtain a trained and validated pollution prediction model; and use the trained and validated pollution prediction model to perform predictive analysis on the comprehensive pollutant data to obtain hierarchical prediction results, that is, prediction results for the surface layer, middle layer, and deep layer.

[0022] Preferably, S2 specifically includes:

[0023] Based on the comprehensive pollution data, taking the comprehensive risk assessment of the surface layer as an example, perform weighted processing on the repair difficulty coefficient of pollutants in the surface soil sample, the diffusion degree of pollutants in the surface soil sample, and the attenuation factor to conduct a comprehensive risk assessment of pollutants and obtain a comprehensive risk assessment value.

[0024] Preferably, S2 specifically includes:

[0025] Compare the comprehensive risk assessment value with the threshold to obtain a comprehensive risk assessment result, including low pollution, medium pollution, high pollution, and severe pollution; based on the prediction results of the surface layer, middle layer, and deep layer and the comprehensive risk assessment results of the surface layer, middle layer, and deep layer, formulate coping strategies.

[0026] The beneficial effects of the technical solution of the present invention are:

[0027] 1. By introducing an adaptive multi-dimensional non-linear pollutant fusion algorithm, efficient fusion and optimized calculation of preprocessed pollutant data are realized. This algorithm first extracts primary pollutant feature data from the preprocessed pollutant data and improves the expression ability of the primary pollutant feature data through non-linear enhancement methods to more accurately describe the behavior patterns of pollutants in the soil; through the combination of high-order polynomial transformation and Gaussian radial basis function, non-linear mapping and data compression processing are carried out, effectively reducing data redundancy, improving calculation efficiency, and retaining the core feature information of pollutants; the adaptive multi-dimensional non-linear pollutant fusion algorithm makes the comprehensive pollutant data more accurate, which helps to improve the calculation efficiency and prediction accuracy of the pollution prediction model.

[0028] 2. After constructing the pollution prediction model, conduct a risk assessment on the comprehensive pollution data, calculate the comprehensive risk assessment value by integrating multiple factors, and classify it into four levels: low pollution, medium pollution, high pollution, and severe pollution according to the severity of the pollution risk. Determine the repair importance of different pollutants by calculating the repair priority weights. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flowchart of a method for monitoring soil pollutants in a mine restoration area according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0030] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0032] The following specifically describes the specific solution of a method for monitoring soil pollutants in a mine restoration area provided by the present invention with reference to the accompanying drawings.

[0033] Refer to the attached Figure 1 , which shows a flowchart of a method for monitoring soil pollutants in a mine restoration area provided by an embodiment of the present invention. The method includes the following steps:

[0034] S1. Obtain soil samples, detect the soil samples using pollutant monitoring techniques to obtain pollutant data, preprocess the pollutant data, and perform intelligent processing on the preprocessed pollutant data using an adaptive multi-dimensional non-linear pollutant fusion algorithm to obtain comprehensive pollutant data;

[0035] In the mine restoration area, professional personnel determine the sampling depth according to the expert experience method for stratified sampling to obtain soil samples. The stratification includes the surface layer, middle layer, and deep layer. For example, the surface layer (0 - 20 cm) mainly samples the soil surface layer, representing the recent deposition of pollutants and environmental pollution; the middle layer (20 - 50 cm) samples the middle layer soil, which is greatly affected by mining activities and geological conditions and may have long-term pollutant accumulation; the deep layer (50 - 100 cm) of soil reflects the penetration and distribution of pollutants under the action of groundwater flow, especially having important reference significance for water-soluble pollutants.

[0036] Existing pollutant monitoring technologies such as X-ray fluorescence spectroscopy (XRF) analysis, gas chromatography-mass spectrometry (GC-MS) technology, pH value and conductivity measurement are used to detect soil samples to obtain pollutant data, including pollutant data such as heavy metals, acidity and alkalinity, and organic pollutants.

[0037] Furthermore, preprocessing such as denoising, cleaning, standardization, and dimensionality processing is performed on the pollutant data to obtain preprocessed pollutant data. The preprocessing methods used are all well-known technical means to those skilled in the art and will not be elaborated here.

[0038] The preprocessed pollutant data is intelligently processed using an adaptive multi-dimensional non-linear pollutant fusion algorithm to obtain comprehensive pollutant data; the adaptive multi-dimensional non-linear pollutant fusion algorithm extracts primary pollutant feature data of pollutants from the preprocessed pollutant data, and performs enhancement processing on the primary pollutant feature data to obtain enhanced feature data, so as to obtain a more accurate pollutant description in subsequent steps; then, through non-linear mapping and data compression processing, the dimension of the enhanced feature data is optimized to reduce the computational complexity and ensure the efficiency of subsequent calculations. The specific implementation process is as follows:

[0039] First, existing feature engineering technologies are used to extract features from the preprocessed pollutant data to obtain primary pollutant feature data of each stratified soil sample, including: surface primary pollutant feature data, middle layer primary pollutant feature data, and deep layer primary pollutant feature data;

[0040] Furthermore, the primary pollutant feature data is enhanced using existing non-linear enhancement methods to obtain enhanced feature data; taking the primary pollutant feature data at the r position on the surface layer as an example, the primary pollutant feature data on the surface layer is non-linearly enhanced to obtain enhanced surface layer feature data E j 。

[0041] Next, the enhanced surface layer feature data is compressed from a high-dimensional space to a lower-dimensional space through non-linear mapping to reduce data redundancy, reduce the computational complexity in subsequent analysis, and at the same time retain key non-linear features. A combination of high-order polynomial transformation and Gaussian radial basis function is used to map the pollutant concentration data of the soil sample to a low-dimensional space to achieve compression processing of the enhanced feature data.

[0042] First, high-order polynomial transformation is used for non-linear mapping, and then the enhanced surface layer feature data is smoothed using an existing Gaussian radial basis function to obtain compressed surface layer feature data. It is achieved through the following formula:

[0043]

[0044] Among them, C j is the compressed characteristic data of the j-th pollutant in the surface soil sample, that is, the comprehensive pollutant data of the surface layer; w ij is the weighting factor of the characteristic data after non-linear mapping, indicating the influence degree in the compressed characteristics; f j (E j ) is the non-linear transformation function of the enhanced characteristic data of the j-th pollutant in the surface soil sample, which is used to perform high-order polynomial transformation or other non-linear mapping on the enhanced surface characteristic data, capture the complex behavior of the pollutant in the surface soil sample, and is determined according to the expert experience method combined with the specific scenario; θ j is the polynomial order in the non-linear mapping, which is used to control the complexity of the characteristic transformation of the j-th pollutant and is determined by the experimental method; is the Gaussian radial basis function; E ij is the i-th characteristic data in the enhanced characteristic data of the j-th pollutant in the surface soil sample; μ j is the mean value of the enhanced characteristic data of the j-th pollutant in the surface soil sample; σ j is the standard deviation of the enhanced characteristic data of the j-th pollutant in the surface soil sample; m is the total number of characteristic data in the enhanced characteristic data of the j-th pollutant in the surface soil sample.

[0045] After the above processing, comprehensive pollution data is obtained, including comprehensive pollution data of the surface layer, comprehensive pollution data of the middle layer, and comprehensive pollution data of the deep layer.

[0046] S2. Build a pollution prediction model, use the pollution prediction model to conduct prediction analysis on the comprehensive pollutant data, and obtain a hierarchical prediction result; based on the comprehensive pollutant data, conduct a risk assessment on the pollutant to obtain a comprehensive risk assessment value, and then obtain a comprehensive risk assessment result, and formulate a response strategy based on the hierarchical prediction result and the comprehensive risk assessment result.

[0047] Based on the historical pollutant data obtained from the existing database, combined with remote sensing images, geographic information system data, and hydro-environmental data, as input data, timestamp synchronization is performed on the input data to obtain the synchronized input data, ensuring that data from different sources can be processed within the same time frame. To ensure data integrity, existing time series interpolation algorithms are used to fill in the missing data in the synchronized input data. Then, an outlier identification method based on local density deviation detection is adopted. The method based on local density deviation detection eliminates abnormal soil samples by calculating the degree of deviation of the data in the multi-dimensional feature space. Finally, data standardization is carried out to obtain the standardized input data, and the standardized input data is intelligently processed using the adaptive multi-dimensional non-linear pollutant fusion algorithm in step S1 to obtain the fused data for subsequent construction of a pollution prediction model.

[0048] Further, based on the fused data, enter the pollution diffusion modeling stage. An adaptive pollution propagation perception network is used to construct a pollution prediction model. The adaptive pollution propagation perception network first constructs a three-dimensional pollutant distribution map, maps the fused data into the geographical space, and combines variables such as wind direction, precipitation, and groundwater flow velocity to simulate the pollution diffusion path. The simulation of the pollution diffusion path uses the graph neural network (GNN) method. During the pollutant diffusion process, dynamic connection weights between different pollution points are established, and by continuously updating the diffusion relationship, the prediction of the pollutant diffusion trend is realized. In addition, since the pollutant diffusion process may be affected by terrain barriers, a terrain influence factor correction strategy is adopted during the simulation of the pollution diffusion path. The migration rate of pollutants in different landforms is calculated through terrain data to ensure that the pollution prediction model can adapt to complex geographical environments.

[0049] After simulating the pollution diffusion path, a long short-term memory network is used to construct a pollution prediction model to predict the fused data, and existing training and validation methods are used for training and validation to obtain the trained and validated pollution prediction model. The trained and validated pollution prediction model is used to predict and analyze the comprehensive pollutant data to obtain hierarchical prediction results, namely the prediction results for the surface layer, middle layer, and deep layer, including low pollution, medium pollution, high pollution, and severe pollution.

[0050] At the same time, based on the comprehensive pollution data, a comprehensive risk assessment of the pollutants is carried out. Finally, the comprehensive risk assessment value is calculated through the following formula and corresponding remediation strategies are generated; taking the comprehensive risk assessment of the surface layer as an example:

[0051]

[0052] Among them, R final is the comprehensive risk assessment value of the pollutants in the surface soil samples; N is the total number of pollutants in the surface soil samples; is the remediation difficulty coefficient of the j-th pollutant in the surface soil sample, determined by the expert experience method; η j is the attenuation factor, which represents the influence range or influence attenuation rate of the j-th pollutant in the surface soil sample in space, and is used to control the spatial spread of pollutants in the remediation area, determined by the physical properties of pollutant diffusion; is the diffusivity of the j-th pollutant in the surface soil sample, which describes the diffusion ability or spread range of the pollutant in the surface soil sample, and is evaluated according to the physical and chemical properties of the pollutant (such as solubility, hydrophilicity, etc.) and environmental factors (such as soil moisture, temperature, etc.); ∈ j is the stability of the j-th pollutant in the surface soil sample, which represents the stability of the pollutant in the environment, and is evaluated according to the chemical stability of the pollutant and its interaction with the soil; is the remediation priority weight of the j-th pollutant in the surface soil sample, which represents the importance of the pollutant in the remediation process;

[0053]

[0054] Among them, is the remediation difficulty adjustment coefficient of the j-th pollutant in the surface soil sample, which is used to reflect the remediation difficulty of the pollutant and is determined by the expert experiment method; is the Laplace operator of the j-th pollutant in the surface soil sample, that is, the second derivative of the characteristic distribution of pollutant concentration data in the remediation area Ω, which reflects the spatial change rate of pollutant concentration and is used to describe the diffusion characteristics of pollutants, calculated by numerical simulation or actual measurement; r is the spatial position, which represents the spatial coordinates of the change in pollutant concentration; C j (r) is the compression characteristic data of the j-th pollutant at position r in the surface soil sample; is the remediation importance coefficient of the j-th pollutant in the surface soil sample, which reflects the remediation importance of the pollutant or the degree to be considered in the remediation strategy, determined by the expert experience method; ρ j is the diffusion coefficient of the j-th pollutant in the surface soil sample, which is a parameter used to describe the diffusion rate of pollutants in the soil and is obtained by the experimental method; is in the surface soil sample for the th pollutant's remediation difficulty adjustment coefficient, used to reflect the remediation difficulty of the pollutant, determined by the expert experiment method; is in the surface soil sample for the th pollutant's Laplace operator; is at position r in the surface soil sample for the th pollutant's compression characteristic data;

[0055] Compare the comprehensive risk assessment value with the threshold preset according to the expert experience method to obtain the comprehensive risk assessment results, including low pollution, medium pollution, high pollution, and severe pollution.

[0056] After the above treatment, the prediction results and comprehensive risk assessment results of the surface layer, middle layer, and deep layer are obtained. Based on the prediction results and comprehensive risk assessment results of the surface layer, middle layer, and deep layer, coping strategies are formulated, including that the surface layer pollution mainly comes from atmospheric deposition, rain leaching, industrial emissions, and mining activities. Since the pollution of the surface soil is relatively easy to spread, green low-impact treatment methods should be preferentially adopted and strengthened in combination with physical remediation. In the case of low pollution, phytoremediation can be used, such as planting heavy metal hyperaccumulator plants (Indian mustard, sunflower) to absorb pollutants; applying microbial agents, such as iron-reducing bacteria, to degrade organic pollutants; adding organic matter, such as humic acid or biochar, to improve soil microbial activity. In the case of medium pollution, adsorbent treatment can be used, such as spreading modified bentonite or zeolite to fix heavy metals and reduce bioavailability; using Fenton reagent for catalytic oxidation to degrade organic pollutants; using shallow soil tillage technology to accelerate the natural degradation of pollutants. In the case of high pollution, deep phytoremediation can be used, selecting plants with developed roots, such as poplar trees, to improve the pollutant absorption efficiency; using phosphates or sulfiding agents for heavy metal passivation treatment. In the case of severe pollution, surface soil replacement can be carried out, removing the polluted soil and replenishing clean soil; using a water-permeable film to cover to prevent pollution from spreading.

[0057] Middle layer pollution is usually caused by the slow penetration of pollutants, and common pollution types include heavy metal accumulation and persistent organic pollutants. The remediation method needs to consider the stability of pollutants, and chemical and physical stabilization means should be preferentially adopted. In the case of low pollution, ecological stabilization methods can be used, such as deep-rooted plants (willows) for long-term absorption and fixation; applying compost, humic acid, etc. to improve the microbial degradation ability. In the case of medium pollution, chemical stabilization can be used, such as using iron oxides or modified clay to fix heavy metals; using microbial agents, such as Pseudomonas, to degrade organic pollutants. In the case of high pollution, chemical elution technology can be used, using chelating agents (such as EDTA) to extract heavy metals and recycle them; using electrokinetic remediation technology, applying an electric field in the polluted soil to promote the migration of pollutants to the recovery electrode. In the case of severe pollution, deep soil replacement methods can be taken, removing the polluted soil and filling clean soil; constructing an underground isolation layer, such as a bentonite wall, to prevent pollution from migrating to the deep layer.

[0058] Deep contamination is usually difficult to remove, and the treatment mainly relies on physical barriers and long-term monitoring. When necessary, forced remediation methods are adopted. In the case of low contamination, long-term monitoring measures are taken to regularly detect the spread of underground contamination to ensure that the pollutants do not migrate further. In the case of medium contamination, physical sealing technology can be used to construct an underground isolation barrier in the contaminated area, such as a bentonite wall, to reduce the spread of pollutants to deeper layers. In the case of high contamination, extraction remediation methods can be adopted, such as installing a groundwater extraction system to prevent pollutants from entering the groundwater; biological leaching technology can be used, and microorganisms are used to promote the dissolution of pollutants and remove them through an underground percolation system. In the case of severe contamination, underground barrier technology can be adopted, using high-density materials, such as HDPE membranes, to form a physical isolation in the contaminated area; for severe organic contamination, deep heat treatment methods can be adopted, and organic matter is volatilized through underground heating technology and then collected and treated.

[0059] In summary, a method for monitoring soil pollutants in a mine restoration area has been completed.

[0060] The sequence of the invention embodiments is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] Each embodiment in this specification is described in a progressive manner. For the same or similar parts between the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

[0062] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A monitoring method for soil pollutants in a mine restoration area, characterized in that It includes the following steps: S1. Obtain soil samples, detect the soil samples using pollutant monitoring techniques to obtain pollutant data, preprocess the pollutant data, and intelligently process the preprocessed pollutant data using an adaptive multi-dimensional non-linear pollutant fusion algorithm to obtain comprehensive pollutant data; S2. Construct a pollution prediction model, use the pollution prediction model to perform prediction analysis on the comprehensive pollutant data to obtain hierarchical prediction results; Based on the comprehensive pollutant data, conduct a risk assessment on the pollutants to obtain a comprehensive risk assessment value, and then obtain a comprehensive risk assessment result. Formulate coping strategies based on the hierarchical prediction results and the comprehensive risk assessment results.

2. The monitoring method for soil pollutants in a mine restoration area according to claim 1, characterized in that The S1 specifically includes: The adaptive multi-dimensional non-linear pollutant fusion algorithm extracts primary pollutant feature data of the pollutants from the preprocessed pollutant data, and performs enhancement processing on the primary pollutant feature data to obtain enhanced feature data; and optimizes the dimension of the enhanced feature data through non-linear mapping and data compression processing.

3. The monitoring method for soil pollutants in a mine restoration area according to claim 2, characterized in that, The S1 specifically includes: In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, feature extraction is performed on the preprocessed pollutant data to obtain primary pollutant feature data of each stratified soil sample, including: surface primary pollutant feature data, middle layer primary pollutant feature data, and deep layer primary pollutant feature data.

4. The monitoring method for soil pollutants in a mine restoration area according to claim 3, characterized in that, The S1 specifically includes: In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, taking the surface primary pollutant feature data as an example, the surface primary pollutant feature data is enhanced using a non-linear enhancement method to obtain enhanced surface feature data.

5. The monitoring method of soil pollutants in a mine restoration area according to claim 4, characterized in that, The S1 specifically includes: In the implementation process of the adaptive multi-dimensional non-linear pollutant fusion algorithm, high-order polynomial transformation is used for non-linear mapping, and the enhanced surface feature data is smoothed using a Gaussian radial basis function to obtain compressed surface feature data.

6. The monitoring method for soil pollutants in a mine restoration area according to claim 1, characterized in that, The S2 specifically includes: Based on historical pollutant data, combined with remote sensing images, geographic information system data, and hydrological environment data, as input data, perform timestamp synchronization on the input data to obtain synchronized input data. Fill in the missing data of the synchronized input data, then perform outlier identification and data standardization to obtain standardized input data. Intelligently process the standardized input data using the adaptive multi-dimensional non-linear pollutant fusion algorithm in step S1 to obtain fused data.

7. A method for monitoring soil pollutants in a mine restoration area according to claim 6, characterized in that, The S2 specifically includes: Based on the fused data, construct a pollution prediction model, use the pollution prediction model to predict the fused data, and train and validate the pollution prediction model to obtain a trained and validated pollution prediction model; and use the trained and validated pollution prediction model to perform prediction analysis on the comprehensive pollutant data to obtain hierarchical prediction results, that is, prediction results for the surface layer, middle layer, and deep layer.

8. The monitoring method for soil pollutants in a mine restoration area according to claim 7, wherein The S2 specifically includes: Based on the comprehensive pollution data, taking the surface comprehensive risk assessment as an example, through weighted processing of the remediation difficulty coefficient of pollutants in the surface soil samples, the diffusion degree and attenuation factor of pollutants in the surface soil samples, the comprehensive risk assessment of pollutants is carried out to obtain the comprehensive risk assessment value.

9. The monitoring method for soil pollutants in a mine restoration area according to claim 8, wherein The S2 specifically includes: Compare the comprehensive risk assessment value with the threshold to obtain the comprehensive risk assessment results, including low pollution, medium pollution, high pollution, and severe pollution; formulate coping strategies based on the prediction results of the surface layer, middle layer, and deep layer and the comprehensive risk assessment results of the surface layer, middle layer, and deep layer.

Images