Intelligent early warning method and system for heavy metal migration of phosphate tailing improved soil

Abstract

The invention provides an intelligent early warning method and system for phosphorus tailing improved soil heavy metal migration, relates to the technical field of computers, and accurately predicts the main migration direction of heavy metal pollutants by combining factors such as terrain gradient, hydrological flow direction and soil structure heterogeneity. The method comprises the following steps: acquiring soil information of a to-be-detected area and soil potential information of a detection point in the to-be-detected area; determining heavy metal concentration information of the detection point according to the soil potential information; the heavy metal concentration information is used for reflecting heavy metal concentration distribution conditions corresponding to different soil depths of the detection points; determining migration direction information of the heavy metal at the detection point; the migration direction information is used for reflecting the corresponding heavy metal migration direction distribution condition of the heavy metal of the detection point in the three-dimensional space; and predicting a pollution diffusion area based on the migration direction information, the organic decomposition rate and the underground water level fluctuation information, and performing visual display.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to an intelligent early warning method and system for heavy metal migration in soil improved by phosphorus tailings. Background Art

[0002] Soil pollutants can be roughly divided into two categories: inorganic pollutants and organic pollutants. Inorganic pollutants mainly include acids, alkalis, heavy metals, salts, compounds of radioactive elements cesium and strontium, and compounds containing arsenic, selenium, and fluorine. Organic pollutants mainly include organic pesticides, phenols, cyanides, petroleum, synthetic detergents, and harmful microorganisms brought by urban sewage, sludge, and manure. When the soil contains too many harmful substances and exceeds the self-purification capacity of the soil, it will cause changes in the composition, structure, and function of the soil, inhibit the activities of microorganisms, and constitute soil pollution.

[0003] Currently, there are still deficiencies in the existing soil pollution detection systems. The existing soil pollution early warning systems basically carry out early warning of soil pollution information through sample collection. The detection of soil pollutants is limited by the collection frequency and manpower, and it is impossible to accurately predict the soil pollution situation in a timely manner, resulting in inaccurate and untimely detection and early warning of soil pollution. Summary of the Invention

[0004] Based on the above technical problems, this application provides an intelligent early warning method and system for heavy metal migration in soil improved by phosphorus tailings. By using intelligent means to monitor the potential changes of heavy metal ions at different depths and regions in real time, combined with factors such as terrain slope, hydrological flow direction, and soil structure heterogeneity, accurately predict the main migration direction of heavy metal pollutants, and achieve accurate division and risk assessment of polluted areas.

[0005] In the first aspect, this application provides an intelligent early warning method for heavy metal migration in soil improved by phosphorus tailings. The method includes: obtaining soil information of the area to be detected and soil potential information of the detection points in the area to be detected; the soil information includes the organic decomposition rate of the soil and the fluctuation of the groundwater level; the soil potential information is used to reflect the potential changes of heavy metal ions at different soil depths of the detection points; determining the heavy metal concentration information of the detection points according to the soil potential information; the heavy metal concentration information is used to reflect the heavy metal concentration distribution corresponding to different soil depths of the detection points; determining the migration direction information of the heavy metals at the detection points; the migration direction information is used to reflect the heavy metal migration direction distribution corresponding to the heavy metals at the detection points in three-dimensional space; predicting the pollution diffusion area based on the migration direction information, organic decomposition rate, and groundwater level fluctuation information, and performing visual display.

[0006] In a possible implementation, based on the soil potential information, the heavy metal concentration information of the detection point is determined, including: obtaining the target soil potential and target heavy metal concentration corresponding to the target depth of the detection point; determining the heavy metal concentrations corresponding to multiple depths at the target depth according to the soil permeability of the detection point, to obtain multiple heavy metal concentrations; using the Kriging spatial interpolation algorithm to interpolate the multiple heavy metal concentrations, to obtain the heavy metal concentration distribution corresponding to different soil depths of the detection point, so as to determine the heavy metal concentration information of the detection point.

[0007] In a possible implementation, the migration direction information of the heavy metal at the detection point is determined, including: obtaining the terrain slope information and soil water retention time of the area to be detected, and according to the terrain slope information and soil water retention time, using hydrodynamics to simulate the migration direction of the heavy metal in the soil, to obtain the preliminary migration direction information of the heavy metal at the detection point; obtaining the soil particle size distribution information and microbial metabolic activity value of the area to be detected, and analyzing the influence of the soil particle size distribution information and microbial metabolic activity value on the migration direction through the random forest algorithm, to obtain the migration direction adjustment coefficient; in the case where the migration direction adjustment coefficient is greater than or equal to the preset threshold, adjusting the preliminary migration direction information based on the migration direction adjustment coefficient, to obtain the migration direction information of the heavy metal at the detection point.

[0008] In a possible implementation, analyzing the influence of the soil particle size distribution information and microbial metabolic activity value on the migration direction through the random forest algorithm, to obtain the migration direction adjustment coefficient, including: using the preliminary migration direction information as the label, and the soil particle size distribution information and microbial metabolic activity value as the input features, and using the random forest algorithm to construct a migration direction prediction model; determining the migration direction adjustment coefficient based on the preliminary migration direction information and the actual output result of the migration direction prediction model.

[0009] In a possible implementation, based on the migration direction information, organic decomposition rate, and groundwater level fluctuation information, the pollution diffusion area is predicted, including: inputting the migration direction information, organic decomposition rate, and groundwater level fluctuation information into the trained pollution area prediction model, to obtain the pollution diffusion area; after the pollution area prediction model reduces the dimensions of the input migration direction information, organic decomposition rate, and groundwater level fluctuation information through the batch normalization layer and the max pooling layer, using the long short-term memory network to extract the time series features, and then predicting the pollution diffusion trend in the future time period; the pollution diffusion area includes the core area, diffusion area, and edge area.

[0010] In a possible implementation manner, the method further includes: obtaining the surface runoff scouring coefficients of the core area, the diffusion area, and the edge area, and the ion saturation thresholds of multiple preset heavy metals; calculating the heavy metal pollution risk indexes of each area respectively based on the ion saturation thresholds and the surface runoff scouring coefficients, and clustering the calculation results to obtain a clustering result; the clustering result includes the high-risk proportion, the medium-risk proportion, and the low-risk proportion.

[0011] In a possible implementation manner, calculating the heavy metal pollution risk indexes of each area respectively based on the ion saturation thresholds and the surface runoff scouring coefficients includes: for any area, multiplying the surface runoff scouring coefficient of the area by the ion saturation thresholds of each preset heavy metal respectively to obtain multiple heavy metal pollution indexes; performing a weighted process on the multiple heavy metal pollution indexes to obtain the heavy metal pollution risk index of the area.

[0012] The technical solution provided by this application at least brings the following beneficial effects: (1) Considering that heavy metal ions have obvious potential characteristics, this application can successfully determine the heavy metal concentration distribution corresponding to different soil depths at the detection point by using the potential information of the soil. Considering the factor of geological movement, this application also needs to determine the heavy metal migration direction distribution corresponding to the heavy metals at the detection point in three-dimensional space. Further, combining the organic decomposition rate of the soil and the fluctuation of the groundwater level, accurately predict the main migration direction of heavy metal pollutants, and then successfully obtain the pollution diffusion area, and visually display the pollution diffusion area to the user to facilitate the user to intuitively understand the soil pollution situation.

[0013] (2) In order to save the measurement cost and improve the measurement efficiency, this application only needs to combine the target soil potential and the target heavy metal concentration corresponding to a certain target depth at the detection point with the soil permeability of the detection point to determine the heavy metal concentrations corresponding to multiple depths at the target depth, and obtain multiple heavy metal concentrations. Further, use the Kriging spatial interpolation algorithm to interpolate the multiple heavy metal concentrations to obtain the heavy metal concentration distribution corresponding to different soil depths at the detection point to determine the heavy metal concentration information at the detection point.

[0014] (3) This application first simulates the migration direction of heavy metals in the soil according to the terrain slope information and the soil water retention time of the area to be detected, and combines fluid mechanics to obtain the preliminary migration direction information. Further, this application also needs to verify the simulated preliminary migration direction information to determine the corresponding migration direction adjustment coefficient. If the adjustment coefficient is large, the preliminary migration direction information will be adjusted based on the migration direction adjustment coefficient to obtain the migration direction information of the heavy metals at the detection point to improve the accuracy of the final migration direction information.

[0015] (4) This application uses the random forest algorithm to construct a migration direction prediction model, which can conveniently compare the preliminary migration direction information with the actual output results of the migration direction prediction model, and then determine the accurate migration direction adjustment coefficient.

[0016] (5) This application integrates the migration direction and pollution distribution information, combines with the risk assessment model, and calculates the pollution risk index, which can achieve the accurate division of the polluted area.

[0017] In a second aspect, this application provides an intelligent early warning system for heavy metal migration in phosphorite tailing-improved soil. The intelligent early warning system includes an intelligent warning device, and the device includes an acquisition unit and a determination unit; the acquisition unit is used to acquire the soil information of the area to be detected and the soil potential information of the detection points in the area to be detected; the soil information includes the organic decomposition rate of the soil and the fluctuation of the groundwater level; the soil potential information is used to reflect the potential change of heavy metal ions at different soil depths of the detection point; the determination unit is used to determine the heavy metal concentration information of the detection point according to the soil potential information; the heavy metal concentration information is used to reflect the heavy metal concentration distribution corresponding to different soil depths of the detection point; the determination unit is also used to determine the migration direction information of the heavy metal at the detection point; the migration direction information is used to reflect the heavy metal migration direction distribution corresponding to the heavy metal at the detection point in three-dimensional space; the determination unit is also used to predict the pollution diffusion area based on the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information, and perform visual display.

[0018] In a possible implementation, the determination unit is specifically used to: acquire the target soil potential and the target heavy metal concentration corresponding to the target depth of the detection point; determine the heavy metal concentrations corresponding to multiple depths at the target depth according to the soil permeability of the detection point to obtain multiple heavy metal concentrations; use the Kriging spatial interpolation algorithm to interpolate the multiple heavy metal concentrations to obtain the heavy metal concentration distribution corresponding to different soil depths of the detection point, so as to determine the heavy metal concentration information of the detection point.

[0019] In a possible implementation, the determination unit is specifically used to: acquire the terrain slope information and the soil water retention time of the area to be detected, and use hydrodynamics to simulate the migration direction of heavy metals in the soil according to the terrain slope information and the soil water retention time to obtain the preliminary migration direction information of the heavy metal at the detection point; acquire the soil particle size distribution information and the microbial metabolic activity value of the area to be detected, and analyze the influence of the soil particle size distribution information and the microbial metabolic activity value on the migration direction through the random forest algorithm to obtain the migration direction adjustment coefficient; in the case where the migration direction adjustment coefficient is greater than or equal to the preset threshold, adjust the preliminary migration direction information based on the migration direction adjustment coefficient to obtain the migration direction information of the heavy metal at the detection point.

[0020] In a possible implementation manner, the determining unit is specifically configured to: use the preliminary migration direction information as a label, the soil particle size distribution information and the microbial metabolic activity value as input features, and construct a migration direction prediction model by using a random forest algorithm; determine a migration direction adjustment coefficient based on the preliminary migration direction information and the actual output result of the migration direction prediction model.

[0021] In a possible implementation manner, the determining unit is specifically configured to: input the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information into a trained pollution area prediction model to obtain a pollution diffusion area; after the pollution area prediction model reduces the dimension of the input migration direction information, organic decomposition rate, and groundwater level fluctuation information through a batch normalization layer and a max pooling layer, use a long short-term memory network to extract time series features, and then predict the pollution diffusion trend in a future time period; the pollution diffusion area includes a core area, a diffusion area, and an edge area.

[0022] In a possible implementation manner, the determining unit is further configured to: obtain the surface runoff scouring coefficients of the core area, the diffusion area, and the edge area, and the ion saturation thresholds of multiple preset heavy metals; calculate the heavy metal pollution risk indexes of each area respectively based on the ion saturation thresholds and the surface runoff scouring coefficients, and cluster the calculation results to obtain a clustering result; the clustering result includes the high-risk proportion, the medium-risk proportion, and the low-risk proportion.

[0023] In a possible implementation manner, the determining unit is specifically configured to: for any area, multiply the surface runoff scouring coefficient of the area by the ion saturation thresholds of each preset heavy metal to obtain multiple heavy metal pollution indicators; perform weighted processing on the multiple heavy metal pollution indicators to obtain the heavy metal pollution risk index of the area.

[0024] In a third aspect, the present application provides an electronic device, including: a processor and a memory; the memory stores instructions executable by the processor; when the processor is configured to execute the instructions, the electronic device implements the method of the first aspect as described above.

[0025] In a fourth aspect, the present application provides a computer program product, when the computer program product runs in an electronic device, enabling the electronic device to execute the relevant method of the first aspect to implement the method of the first aspect.

[0026] In a fifth aspect, the present application provides a computer-readable storage medium, the readable storage medium includes: software instructions; when the software instructions run in an electronic device, enabling the electronic device to implement the method of the first aspect as described above.

[0027] The beneficial effects of the second to fifth aspects described above can be referred to the first aspect, and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0029] Figure 1 It is a schematic structural diagram of an intelligent early warning system for soil pollution provided by an embodiment of the present application; Figure 2 It is a schematic composition diagram of an electronic device provided by an embodiment of the present application; Figure 3 It is a schematic flow diagram of an intelligent early warning method for heavy metal migration in phosphorus tailing-improved soil provided by an embodiment of the present application; Figure 4 It is a schematic composition diagram of an intelligent early warning device for soil pollution provided by an embodiment of the present application. Detailed implementation manners

[0030] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above accompanying drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0032] In addition, in the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B can represent A or B. The "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality of" means two or more than two.

[0033] Before explaining the embodiments of the present application in detail, some related terms and related technologies involved in the embodiments of the present application will be introduced.

[0034] The treatment and early warning of soil heavy metal pollution are key areas in environmental science and agricultural sustainable development. Its importance lies in its direct relation to the stability of the ecosystem, food safety, and human health. The migration and accumulation of heavy metal pollutants in the soil not only threaten the quality of cultivated land but also may spread through the hydrological cycle and food chain, triggering regional environmental crises. Therefore, studying the migration law of heavy metals in the soil and achieving intelligent monitoring and risk assessment have become scientific propositions that urgently need to be broken through.

[0035] Currently, the monitoring and treatment methods of soil heavy metal pollution mostly rely on traditional sampling analysis and static model prediction. However, when facing complex terrains and dynamic environments, these methods often have defects such as insufficient sampling representativeness, poor real-time performance, and low prediction accuracy. Especially for the dynamic changes in the migration of heavy metal ions in the deep soil, existing technologies are difficult to capture their spatio-temporal heterogeneity, resulting in distorted pollution range division and risk assessment results.

[0036] The core challenges in this field focus on how to accurately monitor the potential changes of heavy metal ions at different depths and regions, accurately obtain the soil heavy metal concentration, and effectively integrate key factors such as terrain slope, hydrological flow direction, and soil structure heterogeneity. The difficulty in monitoring the potential changes of heavy metal ions stems from the complexity of the soil medium and the obstacle in obtaining deep-layer data, while the dynamic influence of terrain slope and hydrological flow direction increases the uncertainty of migration direction prediction. In addition, the heterogeneity of soil structure makes the pollutant distribution show non-uniform characteristics, and the traditional homogenization assumption can no longer meet the needs of accurate analysis. These unsolved technical problems lead to large deviations in the prediction of heavy metal pollution migration paths and the scientific division of polluted areas.

[0037] In view of the above problems, the embodiments of the present application provide an intelligent early warning method for the migration of heavy metals in phosphorus tailing-improved soil. By using intelligent means to monitor the potential changes of heavy metal ions at different depths and regions in real time, combining factors such as terrain slope, hydrological flow direction, and soil structure heterogeneity, accurately predict the main migration direction of heavy metal pollutants, and achieve accurate division and risk assessment of polluted areas.

[0038] The following will detail the intelligent early warning method for the migration of heavy metals in phosphorus tailing-improved soil provided by the embodiments of the present application with reference to the accompanying drawings.

[0039] The intelligent early warning method for the migration of heavy metals in phosphorus tailing-improved soil provided by the embodiments of the present application can be applied to the intelligent early warning system for soil pollution. Figure 1 shows a schematic structural diagram of the intelligent early warning system for soil pollution. As Figure 1As shown in the figure, the intelligent soil pollution early warning system 10 includes an intelligent soil pollution early warning device 11 and a plurality of sensors 12. Among them, the sensors 12 are connected in a wired or wireless manner. The intelligent soil pollution early warning device 11 is connected to the plurality of sensors 12 in a wired or wireless manner. Specifically, the intelligent soil pollution early warning device 11 can be connected to the plurality of sensors separately, or can be connected to a total sensor. The embodiments of the present application do not limit this.

[0040] In the embodiments of the present application, the sensors 12 can be deployed in the soil or around the soil to complete the soil monitoring task. Each type of sensor can perform precise measurement for specific soil characteristics. For example, the sensors 12 can be divided into the following main types: soil moisture sensors, soil temperature sensors, soil conductivity sensors, soil pH sensors, etc.

[0041] The intelligent soil pollution early warning device 11 can be any electronic device with data processing functions. For example, the intelligent soil pollution early warning device 11 can be a server, a computer, or can also be a server cluster composed of multiple servers. In some embodiments, the server cluster can also be a distributed cluster. Optionally, the server can be a central server, and the server can also be implemented on a cloud platform. For example, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud, and a multi-cloud, etc., or any combination thereof. The embodiments of the present application do not limit this.

[0042] The execution subject of the intelligent early warning method for heavy metal migration in phosphorus tailing improved soil provided by the embodiments of the present application can be the above-mentioned intelligent soil pollution early warning device 11. As described above, the intelligent soil pollution early warning device 11 can be an electronic device with data processing functions such as a computer or a server. Optionally, the intelligent soil pollution early warning device 11 can also be a processor (such as a central processing unit (CPU)) in the aforementioned electronic device; or, the intelligent soil pollution early warning device 11 can also be an application program (APP) with model training functions installed in the aforementioned electronic device; or, the intelligent soil pollution early warning device 11 can also be a functional module with model training functions in the aforementioned electronic device. The embodiments of the present application do not limit this.

[0043] For simplicity of description, the following will uniformly introduce the intelligent soil pollution early warning device 11 as an example of an electronic device.

[0044] Figure 2 It is a schematic diagram of the composition of the electronic device provided by the embodiments of the present application. As Figure 2As shown, the electronic device may include: Processor 1, Memory 21, Communication Line 22, Communication Interface 23, and Input / Output Interface 24.

[0045] Among them, Processor 1, Memory 21, Communication Interface 23, and Input / Output Interface 24 may be connected through Communication Line 22.

[0046] Processor 1 is used to execute the instructions stored in Memory 21 to implement the fault analysis method provided in the following embodiments of this application. Processor 1 may be a CPU, a general-purpose processor, a network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller (MCU), a programmable logic device (PLD), or any combination thereof. Processor 1 may also be any other device with processing functions, such as a circuit, a device, or a software module. This application embodiment does not limit this. In one example, Processor 1 may include one or more CPUs, such as Figure 2 CPU0 and CPU1 in. As an alternative implementation, the electronic device may include multiple processors. For example, in addition to Processor 1, it may also include Processor 2 ( Figure 2 shown by the dashed line in).

[0047] Memory 21 is used to store instructions. For example, the instructions may be computer programs. Optionally, Memory 21 may be a read-only memory (ROM) or other types of static storage devices that can store static information and / or instructions, or it may be a random access memory (RAM) or other types of dynamic storage devices that can store information and / or instructions. It may also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices. This application embodiment does not limit this.

[0048] It should be noted that Memory 21 may exist independently of Processor 1 or may be integrated with Processor 1. Memory 21 may be located inside the electronic device or outside the electronic device. This application embodiment does not limit this.

[0049] A communication line 22 for transmitting information between components included in the electronic device.

[0050] A communication interface 23 for communicating with other devices (such as the above-mentioned image acquisition device 100) or other communication networks. The other communication network can be an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc. The communication interface 23 can be a module, a circuit, a transceiver, or any device capable of implementing communication.

[0051] An input / output interface 24 for implementing human-computer interaction between the user and the electronic device. For example, implementing action interaction or information interaction between the user and the electronic device.

[0052] Exemplarily, the input / output interface 24 can be a mouse, a keyboard, a display screen, or a touch display screen, etc. Action interaction or information interaction between the user and the electronic device can be achieved through a mouse, a keyboard, a display screen, or a touch display screen, etc.

[0053] It should be noted that Figure 2 the structure shown in Figure 2 does not constitute a limitation on the electronic device. In addition to the components shown, the electronic device may include more or fewer components than those shown in the figure, or a combination of certain components, or a different component arrangement.

[0054] The following introduces an intelligent early warning method for heavy metal migration in phosphorus tailing-improved soil provided by the embodiments of the present application.

[0055] Figure 3 is a schematic flowchart of the intelligent early warning method for heavy metal migration in phosphorus tailing-improved soil provided by the embodiments of the present application. Optionally, this method can be executed by an electronic device having the above-mentioned Figure 2 shown hardware structure. As shown in Figure 3 this method includes S301 to S304.

[0056] S301. Obtain the soil information of the area to be detected and the soil potential information of the detection points in the area to be detected.

[0057] Among them, the soil information includes the organic decomposition rate of the soil and the fluctuation of the groundwater level; the soil potential information is used to reflect the potential change of heavy metal ions at different soil depths of the detection points.

[0058] Specifically, soil organic decomposition rate refers to the rate at which organic matter in the soil is decomposed by microorganisms within a certain period of time, and is a key link in the soil organic matter cycle. In the embodiment of the present application, the electronic device can calculate the organic decomposition rate of the area to be detected by the reduction of soil organic matter per unit time or the ratio of the amount of mineralized decomposition carbon to the initial original carbon amount.

[0059] The groundwater level in most areas shows seasonal fluctuations throughout the year. For example, the water level in area A is relatively stable from January to February, continues to decline from March to June due to reduced precipitation and irrigation, rebounds from July to October due to increased precipitation replenishment, and remains basically stable from November to December. In the embodiment of the present application, a fluctuation coefficient can be used to measure the fluctuation of the groundwater level, where the fluctuation coefficient is between 0 and 1, and the closer the fluctuation coefficient is to 1, the more violent the fluctuation of the groundwater level, and the closer the fluctuation coefficient is to 0, the more stable the fluctuation of the groundwater level. In the embodiment of the present application, the electronic device can determine the fluctuation of the groundwater level in the area to be detected in combination with the current month. For example, if the current month is January, the electronic device can determine the fluctuation of the groundwater level in the area to be detected as 0.1.

[0060] Soil potential information is an important parameter in soil science, reflecting the redox state, ion activity and electrochemical properties of the soil, and has an important impact on soil fertility, pollutant migration and microbial activity. In the embodiment of the present application, electrochemical sensors can be deployed at a certain detection point in the area to be detected, and electronic equipment can use these electrochemical sensors to obtain the potential change data of heavy metal ions in different soil depths at the detection point, and then obtain the soil potential information of the detection point.

[0061] Specifically, in the deployment of the sensor network, electrochemical sensors are used to set up monitoring points at intervals of 5 cm at soil depths of 0 to 50 cm. The potential data of cadmium ions (Cd²⁺) are collected at each point, and the sampling frequency is 1 time / minute. After the analog signal is converted into a digital signal through an AD converter, the Kalman filter algorithm is used to reduce the noise of the original data. The filter parameters are set to process noise covariance Q=0.01 and observation noise covariance R=0.1. Combined with the soil particle size distribution data obtained by the laser diffraction method (clay particles account for 18%, silt particles 45%, and sand particles 37%), a three-dimensional soil model is established using finite element analysis, and the grid size is set to 0.5 mm. The pore water flow rate is calculated by Darcy's law, where the permeability coefficient k=3.2×10⁻ 5m / s, and the hydraulic gradient i = 0.15. After coupling the potential data with the flow velocity field, the electronic device uses the time series ARIMA model (p = 2, d = 1, q = 1) to predict the potential change trend in the next 2 hours. The model residuals are confirmed to have white noise characteristics through the Ljung-Box test (lag order 10, p-value 0.23). Finally, a non-linear mapping model is constructed based on the support vector regression (SVR) algorithm. The kernel function is selected as RBF (γ = 0.5, C = 1.0). The input features include the potential gradient, the particle size fractal dimension (2.71), and the damping coefficient (0.68 s⁻¹), and the output prediction error is controlled within ±5 mV.

[0062] Among them, AD conversion, that is, analog-to-digital conversion, is the process of converting analog signals into digital signals. The core device is the analog-to-digital converter (ADC). Potential data refers to the potential values obtained through electrochemical measurement or analysis, which are usually used to characterize the potential differences in electrochemical reactions or electrolyte solutions. In the embodiments of the present application, it can be used to reflect the potential differences in the soil.

[0063] S302. Determine the heavy metal concentration information of the detection point according to the soil potential information.

[0064] Among them, the heavy metal concentration information is used to reflect the heavy metal concentration distribution corresponding to different soil depths at the detection point.

[0065] As a possible implementation, the electronic device can first obtain the target soil potential and the target heavy metal concentration corresponding to the target depth of the detection point, and determine the heavy metal concentrations corresponding to multiple depths at the target depth according to the soil permeability of the detection point to obtain multiple heavy metal concentrations. Further, the electronic device uses the Kriging spatial interpolation algorithm to interpolate the multiple heavy metal concentrations to obtain the heavy metal concentration distribution corresponding to different soil depths at the detection point, so as to determine the heavy metal concentration information of the detection point.

[0066] Specifically, in soil heavy metal pollution monitoring, based on the potential change data, the Kriging spatial interpolation algorithm is used to calculate the heavy metal ion concentration distribution at each depth. During the interpolation process, the semi-variogram model is set as the spherical model, the nugget value is 0.05, the sill value is 0.8, and the range is 15 m. Combining with the source distance attenuation model, the concentration attenuation rates at 10 m, 20 m, and 30 m away from the pollution source are calculated to be 0.12, 0.25, and 0.38 respectively. Through the analysis of the sediment penetration rate difference, the penetration rates at different depths (0 - 10 cm, 10 - 20 cm, 20 - 30 cm) are calculated to be 1.2×10⁻ 6 m / s, 8.5×10⁻ 7 m / s, and 5.3×10⁻ 7m / s, and there is a significant correlation between the infiltration rate difference and the concentration distribution. A spatial feature map of the concentration distribution was constructed using Geographic Information System (GIS) with a spatial resolution of 0.1 m. The kernel density estimation algorithm was used to analyze the pollution hotspots, with a bandwidth set to 5 m and a Gaussian function selected as the kernel function. The main features of the concentration distribution were extracted through Principal Component Analysis (PCA). The cumulative contribution rate of the first three principal components reached 85%. The correlation coefficient between the first principal component and the distance from the pollution source was -0.76, and the correlation coefficient between the second principal component and the infiltration rate difference was 0.68. A concentration prediction model was constructed based on the random forest algorithm. The input features included the rate of change of potential, the distance from the pollution source, and the infiltration rate difference. The model parameters were set as follows: the number of trees was 100, the maximum depth was 10, the feature selection ratio was 0.8, and the prediction error was controlled within ±0.05 mg / kg.

[0067] S303. Determine the migration direction information of heavy metals at the detection points.

[0068] Among them, the migration direction information is used to reflect the distribution of the heavy metal migration directions corresponding to the heavy metals at the detection points in three-dimensional space.

[0069] As a possible implementation, the electronic device can obtain the terrain slope information and the soil water retention time of the area to be detected. According to the terrain slope information and the soil water retention time, the hydrodynamic simulation is used to simulate the migration direction of heavy metals in the soil, and the preliminary migration direction information of the heavy metals at the detection points is obtained. The electronic device can obtain the soil particle size distribution information and the microbial metabolic activity value of the area to be detected, and analyze the influence of the soil particle size distribution information and the microbial metabolic activity value on the migration direction through the random forest algorithm to obtain the migration direction adjustment coefficient. When the migration direction adjustment coefficient is greater than or equal to the preset threshold, the electronic device can adjust the preliminary migration direction information based on the migration direction adjustment coefficient to obtain the migration direction information of the heavy metals at the detection points.

[0070] Specifically, topographic slope data is extracted based on the Digital Elevation Model (DEM). The average slope of the study area is calculated to be 8.5° using the slope algorithm. Combining with the alluvial factor model, the alluvial factor weight is set to 0.75, and the alluvial factor value is calculated to be 0.63. Soil moisture data is obtained through the soil moisture sensor network. The time series analysis method is used to calculate the soil moisture residence time. The time window is set to 30 days, and the average residence time is obtained as 12.5 hours. For the migration direction of heavy metal ions, the Navier-Stokes equation is used for hydrodynamic simulation. The soil porosity is set to 0.45, and the simulation step size is 0.01 m. The simulation results show that the main migration direction of heavy metal ions in the soil is 15° south by southeast. Combining with the root adsorption competition factor model, the root density is set to 0.3 g / cm³, and the adsorption competition coefficient is set to 0.85. The interception rate of root adsorption on heavy metal ions is calculated to be 0.42. Through the organic matter decomposition rate model, the initial organic matter content is set to 2.5%, and the decomposition rate constant is set to 0.02 / day. The influence coefficient of organic matter decomposition on the migration of heavy metal ions is calculated to be 0.58. Based on the above data, the path analysis algorithm is used to determine the preliminary range of the heavy metal ion migration path. The path width is set to 0.5 m, and the path length is set to 25 m. The spatial distribution of the migration path is significantly correlated with the topographic slope and the soil moisture residence time.

[0071] In some embodiments, in order to obtain the migration direction adjustment coefficient, the electronic device may use the preliminary migration direction information as a label, the soil particle size distribution information and the microbial metabolic activity value as input features, and use the random forest algorithm to construct a migration direction prediction model. Further, the electronic device may determine the migration direction adjustment coefficient based on the preliminary migration direction information and the actual output result of the migration direction prediction model.

[0072] Specifically, when extracting the attenuation rate of the pollution source distance from the soil sample, an exponential decay model is used to fit the curve of the pollution concentration changing with distance. The initial pollution concentration is set to 1200 mg / kg, and the attenuation coefficient is 0.15 / m. It is calculated that the concentration at a distance of 10 meters from the pollution source decays to 245 mg / kg. The limit parameter of the soil buffer capacity is fitted through the acid-base titration experimental data. The Langmuir adsorption model is adopted, with the maximum adsorption capacity set to 2.8 mmol / kg and the equilibrium constant set to 0.45 L / mmol. It is calculated that the buffer capacity under the current pH = 6.2 condition is 1.7 mmol / kg. Combining the preliminary range data of the migration path, a migration direction prediction model is constructed using the random forest algorithm. The input features include the soil particle size distribution (clay content 18%, silt content 35%, sand content 47%) and the microbial metabolic activity value (using dehydrogenase activity as an index, the measured value is 32 μgTPF / g·h). The number of decision trees is set to 100, and the maximum depth is set to 8. Through feature importance analysis, the contribution weight of the particle size distribution to the migration direction is 0.62, and the contribution weight of the microbial activity is 0.38. Based on the model output results, a multiple linear regression is used to calculate the migration direction adjustment coefficient. The regression coefficients are -0.23 for the particle size distribution and 0.17 for the microbial activity. Finally, the adjustment coefficient is determined to be -0.09, indicating that the migration direction of heavy metal ions will deflect approximately 5° to the northwest under the current soil conditions.

[0073] When the migration direction adjustment coefficient exceeds the preset threshold of 0.1, a three-dimensional grid division technology is used to perform refined modeling on the soil area. The grid resolution is 1 m × 1 m × 0.5 m, and the coverage depth range is 0 to 5 m. The spatial interpolation of the heavy metal concentration in the soil is carried out through the Kriging interpolation algorithm. The semi-variogram model is set to the spherical model, the nugget value is 0.05, the sill value is 0.8, and the range is 15 m. The interpolation results show that the regional concentration distribution presents an obvious gradient change. Combining the toxicity attenuation rate model, a first-order kinetic equation is used to fit the heavy metal toxicity attenuation process. The initial toxicity value is set to 1000 TU, and the attenuation constant is 0.12 / day. It is calculated that the toxicity value decays to 32 TU after 30 days. At the same time, the Monte Carlo simulation method is used to perform a probability analysis on the regional exposure frequency. The number of simulations is set to 10000 times. The exposure frequency distribution results show that the high-risk area accounts for 15%, the medium-risk area accounts for 45%, and the low-risk area accounts for 40%. Based on the above data, a migration direction distribution map is generated through weighted overlay analysis. The weights are set to 0.6 for the toxicity attenuation rate and 0.4 for the exposure frequency. Finally, it is determined that the main migration direction of heavy metals is deflected to the southeast direction, with a deflection angle of 8°.

[0074] S304. Predict the pollution diffusion area based on the migration direction information, organic decomposition rate, and groundwater level fluctuation information, and perform visual display.

[0075] As a possible implementation, the electronic device can input the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information into the trained pollution area prediction model to obtain the pollution diffusion area. After the pollution area prediction model reduces the dimension of the input migration direction information, organic decomposition rate, and groundwater level fluctuation information through the batch normalization layer and the max pooling layer, it uses the long short-term memory network to extract the time series features, and then predicts the pollution diffusion trend in the future time period. The pollution diffusion area includes the core area, the diffusion area, and the edge area.

[0076] Specifically, a one-dimensional convolutional neural network is used to process the potential monitoring data. The number of input layer channels is set to 3, corresponding to the potential change, groundwater level fluctuation, and organic matter decomposition rate respectively. The convolutional kernel size is 5, the stride is 1, the ReLU activation function is selected, and the output feature map dimension is 64. After reducing the dimension through the batch normalization layer and the max pooling layer, the long short-term memory network (LSTM) is used to extract the time series features. The number of hidden layer units is set to 128, the time step is 24 hours, the monitoring data of the past 7 days is input, and the pollution diffusion trend in the next 3 days is predicted. The Pearson correlation coefficient is used to calculate the correlation between the potential change and the water level fluctuation. When the correlation coefficient exceeds 0.7, it is determined as a strongly correlated area. Combining with the organic matter decomposition rate threshold of 0.05 g / (kg·d), the high pollution risk area is identified. The random forest algorithm is used to classify the pollution range. The number of trees is set to 100, and the maximum depth is 10. The feature importance analysis shows that the contribution degree of the water level fluctuation is 45%, the potential change is 35%, and the organic matter decomposition is 20%. Based on the classification results, the Gaussian process regression is used to fit the pollution boundary. The radial basis function is selected as the kernel function, the length scale is set to 2.5 meters, and the noise level is 0.01. It is predicted that the proportion of the pollution core area is 12%, the diffusion area is 28%, and the edge area is 60%. The spatial autocorrelation analysis is used to verify the pollution distribution pattern. The calculated Moran's index value is 0.65, indicating that the pollution shows significant aggregation characteristics, and the hot spots are mainly distributed in the northeast of the study area.

[0077] In one design, the electronic device can also integrate the migration direction distribution map and the main distribution range of the pollution area through the risk assessment model, and calculate the heavy metal pollution risk index of each area in combination with the ion accumulation saturation threshold and the surface runoff scouring coefficient to determine the accurate division result of the pollution area including the ion exchange rate of the sediment layer.

[0078] As a possible implementation, the electronic device can obtain the surface runoff scouring coefficients of the core area, diffusion area, and edge area, as well as the ion saturation thresholds of multiple preset heavy metals. Further, the electronic device can calculate the heavy metal pollution risk indices of each area based on the ion saturation thresholds and surface runoff scouring coefficients, and cluster the calculation results to obtain a clustering result; the clustering result includes the high-risk proportion, medium-risk proportion, and low-risk proportion.

[0079] For example, for any area, the electronic device can multiply the surface runoff scouring coefficient of the area by the ion saturation thresholds of each preset heavy metal to obtain multiple heavy metal pollution indicators. Further, the electronic device can perform a weighted process on the multiple heavy metal pollution indicators to obtain the heavy metal pollution risk index of the area.

[0080] Specifically, in the risk assessment model, the migration direction distribution map and the main distribution range of the polluted area are integrated through spatial overlay analysis. The Kriging interpolation method is used to perform spatial interpolation on the heavy metal pollution concentration. The interpolation parameters are set as follows: the semi-variogram model is the spherical model, the nugget value is 0.1, the sill value is 1.5, and the range is 500 meters. Combining with the ion cumulative saturation threshold, the saturation threshold of cadmium is set to 0.5 mg / kg, lead is 100 mg / kg, and zinc is 300 mg / kg. The heavy metal pollution risk indices of each area are calculated through a logistic regression model, and the regression coefficients are 0.8 for cadmium, 0.6 for lead, and 0.4 for zinc. The surface runoff scouring coefficient is calculated using the SCS curve number method. The soil type is set as type C, the land use type is farmland, the curve number is 75, the rainfall is 50 mm, and the calculated scouring coefficient is 0.3, dimensionless. The ion exchange rate of the sediment layer is measured by ion exchange chromatography. The exchange capacity is set to 10 cmol / kg, and the exchange rate is 0.02 cmol / (kg·d). Combining with the pollution risk index, the K-means clustering algorithm is used to accurately divide the polluted area. The number of clusters is 3, the number of iterations is 100, and the convergence threshold is 0.01. The final division result is that the high-risk area accounts for 15%, the medium-risk area accounts for 35%, and the low-risk area accounts for 50%. The division result is verified through spatial autocorrelation analysis. The calculated Moran's index value is 0.72, indicating that the pollution risk shows significant spatial aggregation characteristics, and the high-risk areas are mainly distributed in the southwestern part of the study area.

[0081] The technical solutions provided by the embodiments of the present application at least bring the following beneficial effects: Considering that heavy metal ions have obvious potential characteristics, the present application can successfully determine the heavy metal concentration distribution corresponding to different soil depths at the detection point by using the potential information of the soil. Considering the factors of geological movement, the present application also needs to determine the heavy metal migration direction distribution corresponding to the heavy metals at the detection point in three-dimensional space. Further, by combining the organic decomposition rate of the soil and the fluctuation of the groundwater level, the main migration direction of heavy metal pollutants can be accurately predicted, and then the pollution diffusion area can be successfully obtained and visually displayed to the user to facilitate the user's intuitive understanding of the soil pollution situation.

[0082] The above mainly introduces the solutions provided by the embodiments of the present application from the perspective of methods. To achieve the above functions, it includes the corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should easily realize that, combining the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in the way of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0083] In an exemplary embodiment, the embodiments of the present application further provide a soil pollution intelligent early warning device. Figure 4 It is a schematic diagram of the composition of the soil pollution intelligent early warning device provided by the embodiments of the present application. As Figure 4 shown, the soil pollution intelligent early warning device includes: an acquisition unit 401 and a determination unit 402.

[0084] The acquisition unit 401 is used to acquire the soil information of the area to be detected and the soil potential information of the detection points in the area to be detected; the soil information includes the organic decomposition rate of the soil and the fluctuation of the groundwater level; the soil potential information is used to reflect the potential change of heavy metal ions at different soil depths of the detection point; the determination unit 402 is used to determine the heavy metal concentration information of the detection point according to the soil potential information; the heavy metal concentration information is used to reflect the heavy metal concentration distribution corresponding to different soil depths of the detection point; the determination unit 402 is further used to determine the migration direction information of the heavy metals at the detection point; the migration direction information is used to reflect the heavy metal migration direction distribution corresponding to the heavy metals at the detection point in three-dimensional space; the determination unit 402 is further used to predict the pollution diffusion area based on the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information, and perform visual display.

[0085] In a possible implementation manner, the determining unit 402 is specifically configured to: obtain the target soil potential and the target heavy metal concentration corresponding to the target depth of the detection point; determine the heavy metal concentrations corresponding to multiple depths at the target depth according to the soil permeability of the detection point, so as to obtain multiple heavy metal concentrations; use the Kriging spatial interpolation algorithm to interpolate the multiple heavy metal concentrations to obtain the heavy metal concentration distribution corresponding to different soil depths of the detection point, so as to determine the heavy metal concentration information of the detection point.

[0086] In a possible implementation manner, the determining unit 402 is specifically configured to: obtain the terrain slope information and the soil water retention time of the area to be detected, and according to the terrain slope information and the soil water retention time, use hydrodynamics to simulate the migration direction of heavy metals in the soil to obtain the preliminary migration direction information of the heavy metals at the detection point; obtain the soil particle size distribution information and the microbial metabolic activity value of the area to be detected, and analyze the influence of the soil particle size distribution information and the microbial metabolic activity value on the migration direction through the random forest algorithm to obtain the migration direction adjustment coefficient; in the case where the migration direction adjustment coefficient is greater than or equal to a preset threshold, adjust the preliminary migration direction information based on the migration direction adjustment coefficient to obtain the migration direction information of the heavy metals at the detection point.

[0087] In a possible implementation manner, the determining unit 402 is specifically configured to: use the preliminary migration direction information as the label and the soil particle size distribution information and the microbial metabolic activity value as the input features, and use the random forest algorithm to construct a migration direction prediction model; determine the migration direction adjustment coefficient based on the preliminary migration direction information and the actual output result of the migration direction prediction model.

[0088] In a possible implementation manner, the determining unit 402 is specifically configured to: input the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information into the trained pollution area prediction model to obtain the pollution diffusion area; after the pollution area prediction model reduces the dimension of the input migration direction information, organic decomposition rate, and groundwater level fluctuation information through the batch normalization layer and the max pooling layer, use the long short-term memory network to extract time series features, and then predict the pollution diffusion trend in the future time period; the pollution diffusion area includes the core area, the diffusion area, and the edge area.

[0089] In a possible implementation manner, the determining unit 402 is further configured to: obtain the surface runoff scouring coefficients of the core area, the diffusion area, and the edge area, and the ion saturation thresholds of multiple preset heavy metals; calculate the heavy metal pollution risk indexes of each area respectively based on the ion saturation thresholds and the surface runoff scouring coefficients, and cluster the calculation results to obtain the clustering result; the clustering result includes the high-risk proportion, the medium-risk proportion, and the low-risk proportion.

[0090] In a possible implementation manner, the determining unit 402 is specifically configured to: for any region, multiply the surface runoff scouring coefficient of the region by the ion saturation thresholds of each preset heavy metal to obtain a plurality of heavy metal pollution indexes; perform a weighting process on the plurality of heavy metal pollution indexes to obtain the heavy metal pollution risk index of the region.

[0091] It should be noted that Figure 4 The division of the modules in [reference] is illustrative. It is only a logical function division. In actual implementation, there may be other division methods. For example, two or more functions may also be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software function units.

[0092] In an exemplary embodiment, the embodiment of the present application further provides a computer-readable storage medium, including software instructions, which when running on an electronic device, cause the electronic device to execute any one of the methods provided in the above embodiments.

[0093] In an exemplary embodiment, the embodiment of the present application further provides a computer program product including computer-executable instructions, which when running on an electronic device, cause the electronic device to execute any one of the methods provided in the above embodiments.

[0094] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer-executable instructions. When the computer-executable instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer-executable instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer-executable instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more integrated media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a solid state disk (SSD), etc.

[0095] Although the present application has been described in connection with various embodiments, those skilled in the art will recognize other variations of the disclosed embodiments upon review of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the articles "a" or "an" do not exclude a plurality. A single processor or other unit may implement several functions recited in the claims. Certain measures are recited in mutually different dependent claims, but this does not indicate that these measures cannot be combined to good effect.

[0096] Although the present application has been described in connection with specific features and their embodiments, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the application. Accordingly, the specification and drawings are merely exemplary illustrations of the application defined by the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the application. It is obvious that those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

[0097] The above is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An intelligent early warning method for improving the migration of heavy metals in soil by phosphorus tailings, characterized in that, The method includes: Obtaining soil information of the area to be detected and soil potential information of detection points within the area to be detected; the soil information includes the organic decomposition rate of the soil and the fluctuation of the groundwater level; the soil potential information is used to reflect the potential change of heavy metal ions at different soil depths of the detection points; Determining the heavy metal concentration information of the detection points according to the soil potential information; the heavy metal concentration information is used to reflect the heavy metal concentration distribution corresponding to different soil depths of the detection points; Determining the migration direction information of the heavy metals at the detection points; the migration direction information is used to reflect the heavy metal migration direction distribution corresponding to the heavy metals at the detection points in three-dimensional space; Predicting the pollution diffusion area based on the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information, and performing visual display.

2. The method according to claim 1, wherein The determining the heavy metal concentration information of the detection points according to the soil potential information includes: Obtaining the target soil potential and the target heavy metal concentration corresponding to the target depth of the detection point; Determining the heavy metal concentrations corresponding to multiple depths at the target depth according to the soil permeability of the detection point, to obtain multiple heavy metal concentrations; Interpolating the multiple heavy metal concentrations by using the Kriging spatial interpolation algorithm to obtain the heavy metal concentration distribution corresponding to different soil depths of the detection point, so as to determine the heavy metal concentration information of the detection point.

3. The method according to claim 1, characterized in that, The determining the migration direction information of the heavy metals at the detection points includes: Obtaining the terrain slope information and the soil water retention time of the area to be detected, and using hydrodynamics to simulate the migration direction of heavy metals in the soil according to the terrain slope information and the soil water retention time, to obtain the preliminary migration direction information of the heavy metals at the detection point; Obtaining the soil particle size distribution information and the microbial metabolic activity value of the area to be detected, and analyzing the influence of the soil particle size distribution information and the microbial metabolic activity value on the migration direction through a random forest algorithm, to obtain a migration direction adjustment coefficient; When the migration direction adjustment coefficient is greater than or equal to a preset threshold, adjusting the preliminary migration direction information based on the migration direction adjustment coefficient to obtain the migration direction information of the heavy metals at the detection point.

4. The method according to claim 3, wherein The analyzing the influence of the soil particle size distribution information and the microbial metabolic activity value on the migration direction through a random forest algorithm to obtain a migration direction adjustment coefficient includes: Taking the preliminary migration direction information as a label, and the soil particle size distribution information and the microbial metabolic activity value as input features, and constructing a migration direction prediction model by using a random forest algorithm; Determining the migration direction adjustment coefficient based on the preliminary migration direction information and the actual output result of the migration direction prediction model.

5. The method according to claim 1, wherein The predicting the pollution diffusion area based on the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information includes: Input the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information into the trained pollution area prediction model to obtain the pollution diffusion area; after the pollution area prediction model reduces the dimension of the input migration direction information, the organic decomposition rate, and the groundwater level fluctuation information through a batch normalization layer and a max pooling layer, it uses a long short-term memory network to extract time series features, and then predicts the pollution diffusion trend in the future time period; the pollution diffusion area includes a core area, a diffusion area, and an edge area.

6. The method according to claim 5, characterized in that, The method further includes: Obtain the surface runoff scouring coefficient of the core area, the diffusion area, and the edge area, and the ion saturation threshold of multiple preset heavy metals; Based on the ion saturation threshold and the surface runoff scouring coefficient, calculate the heavy metal pollution risk index of each area respectively, and cluster the calculation results to obtain a clustering result; the clustering result includes the high-risk proportion, the medium-risk proportion, and the low-risk proportion.

7. The method according to claim 6, wherein The calculating the heavy metal pollution risk index of each area respectively based on the ion saturation threshold and the surface runoff scouring coefficient includes: For any area, multiply the surface runoff scouring coefficient of the area by the ion saturation threshold of each preset heavy metal to obtain multiple heavy metal pollution indicators; Perform a weighted process on the multiple heavy metal pollution indicators to obtain the heavy metal pollution risk index of the area.

8. An intelligent early warning system for improving the migration of heavy metals in soil by phosphorus tailings, characterized in that, The intelligent early warning system includes an intelligent early warning device, and the device includes an acquisition unit and a determination unit; The acquisition unit is used to acquire the soil information of the area to be detected and the soil potential information of the detection points in the area to be detected; the soil information includes the organic decomposition rate of the soil and the groundwater level fluctuation condition; the soil potential information is used to reflect the potential change condition of heavy metal ions at different soil depths of the detection points; The determination unit is used to determine the heavy metal concentration information of the detection points according to the soil potential information; The heavy metal concentration information is used to reflect the heavy metal concentration distribution condition corresponding to different soil depths of the detection points; The determination unit is further used to determine the migration direction information of the heavy metals at the detection points; the migration direction information is used to reflect the heavy metal migration direction distribution condition corresponding to the heavy metals at the detection points in three-dimensional space; The determination unit is further used to predict the pollution diffusion area based on the migration direction information, the organic decomposition rate, and the groundwater level fluctuation information, and perform visual display.

9. An electronic device, characterized in that, It includes: A processor and a memory; The memory stores instructions executable by the processor; When the processor is configured to execute the instructions, the electronic device implements the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The readable storage medium includes: software instructions; When the software instructions run in an electronic device, the electronic device implements the method according to any one of claims 1-7.

Images