An artificial intelligence-based method for predicting groundwater pollution diffusion in a chemical industrial park

Abstract

The present application relates to the technical field of data processing, and more particularly to a chemical industry park groundwater pollution diffusion prediction method based on artificial intelligence. The content includes: obtaining the pollutant concentration information data in the chemical industry park groundwater and the geological exploration parameter data, and preprocessing to obtain the preprocessed data; based on the preprocessed data, introducing a self-shaping function embedding mechanism, constructing a pollution function, and generating a pollution potential field embedding tensor; introducing a geological disturbance function generator, constructing a spatial disturbance tensor, and combining the pollution potential field embedding tensor to generate a spatial correction tensor of pollution transmission; based on the pollution potential field embedding tensor and the spatial correction tensor of pollution transmission, modeling pollution propagation to realize groundwater pollution diffusion prediction. The technical problem of traditional groundwater pollution diffusion prediction method being inaccurate and incomplete in data processing and analysis, resulting in serious deviation of groundwater pollution diffusion prediction results from actual diffusion trajectory is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a method for predicting groundwater pollution diffusion in a chemical industrial park based on artificial intelligence. BACKGROUND

[0002] Due to the complexity of the process flow, the difficulty of emission management and other factors, chemical industrial parks have become high-risk areas of groundwater pollution. Once a pollution event occurs, pollutants will often enter the groundwater system through multiple paths such as leakage, surface infiltration, and landfill filtration, and diffuse in a unpredictable way in the heterogeneous porous medium, not only polluting water resources, but also possibly causing regional ecological damage and drinking water safety risks. Therefore, how to accurately predict the spatio-temporal evolution path of pollution diffusion in the early stage has become a key supporting technology for pollution prevention, emergency response and scientific management.

[0003] Traditional methods for predicting groundwater pollution diffusion are mainly based on traditional physical models and numerical simulation techniques, such as finite difference method (FDM), finite volume method (FVM) and finite element method (FEM) based on Advection-Dispersion equation. This kind of method emphasizes the simulation of the conservation process of pollutants migrating with hydrodynamic conditions, requires obtaining high-precision hydrogeological parameters, and accurately expressing the convection, diffusion and attenuation behavior of pollutants in the control equation. However, due to the complex geological conditions, drastic parameter changes and diverse pollution source distribution in chemical industrial parks, the actual collected data is often incomplete, discontinuous or even missing, making it difficult to fully model the actual diffusion behavior based on the precise numerical simulation of the control equation; in addition, the pollution diffusion process is often driven by external disturbances such as rainfall, underground pumping, artificial water injection, etc., and traditional static models cannot respond to the periodic disturbance effects and propagation path reconstruction problems in the dynamic process of pollution.

[0004] In summary, the traditional method for predicting groundwater pollution diffusion has the technical problems of inaccurate and incomplete data processing and analysis, resulting in a serious deviation of the predicted results from the actual diffusion trajectory. SUMMARY

[0005] The present application provides a method for predicting groundwater pollution diffusion in a chemical industrial park based on artificial intelligence to solve the technical problem of inaccurate and incomplete data processing and analysis of traditional methods for predicting groundwater pollution diffusion, resulting in a serious deviation of the predicted results from the actual diffusion trajectory.

[0006] The present application provides a method for predicting groundwater pollution diffusion in a chemical industrial park based on artificial intelligence to solve the technical problem of inaccurate and incomplete data processing and analysis of traditional methods for predicting groundwater pollution diffusion, resulting in a serious deviation of the predicted results from the actual diffusion trajectory.

[0007] A method for predicting groundwater pollution diffusion in a chemical industrial park based on artificial intelligence, comprising the following steps:

[0008] S1. Obtain the pollutant concentration information data in the groundwater of the chemical industry park and the geological exploration parameter data, and preprocess to obtain the preprocessed data; based on the preprocessed data, introduce a self-shaping function embedding mechanism, construct a pollution action function, and generate a pollution potential field embedding tensor;

[0009] S2. Introduce a geological disturbance function generator to construct a spatial disturbance tensor, and combine the pollution potential field embedding tensor to generate a spatial correction tensor for pollution transmission; based on the pollution potential field embedding tensor and the spatial correction tensor for pollution transmission, model the pollution propagation to obtain a pollution diffusion prediction tensor, and realize the prediction of groundwater pollution diffusion.

[0010] Preferably, the S1 specifically comprises:

[0011] The preprocessed data includes preprocessed pollutant concentration information data and preprocessed geological exploration parameter data; in the implementation process of the self-shaping function embedding mechanism, based on the soil permeability coefficient in the preprocessed pollutant concentration information data and the preprocessed geological exploration parameter data, and introducing a spatial decay modulation term, a pollution action function is constructed to quantify the diffusion behavior of pollutants in three-dimensional space.

[0012] Preferably, the S1 specifically comprises:

[0013] Introduce an attention weighting mechanism to weight the pollution action function to construct a pollution potential field embedding tensor.

[0014] Preferably, the S2 specifically comprises:

[0015] The geological disturbance function generator simulates nonlinear disturbance points at spatial positions by setting disturbance kernels.

[0016] Preferably, the S2 specifically comprises:

[0017] Based on the preprocessed geological exploration parameter data, obtain the disturbance intensity time weight, the disturbance angle, and the space-time oscillation frequency factor; based on the disturbance angle and the space-time oscillation frequency factor, construct the spatial decay function of the disturbance intensity and the space-time disturbance frequency modulation function, and combine the disturbance intensity time weight to construct the spatial disturbance tensor.

[0018] Preferably, the S2 specifically comprises:

[0019] Based on the spatial disturbance tensor and the pollution potential field embedding tensor, introduce an exponential function and a time modulation term to generate a spatial correction tensor for pollution transmission.

[0020] Preferably, the S2 specifically comprises:

[0021] A diffusion channel weight factor is introduced, combined with a diffusion frequency modulation factor, to perform a path direction-aware convolution operation on the pollution potential field embedding tensor and the spatial correction tensor of pollution transport, to obtain a pollution diffusion prediction tensor.

[0022] Preferably, S2 specifically includes:

[0023] The diffusion frequency modulation factor dynamically modulates the diffusion process by analyzing the time-frequency characteristics of the preprocessed pollution concentration information data and combining a cosine function.

[0024] The technical scheme of the present application has the following advantages:

[0025] 1. The pollution concentration information data in the groundwater and the geological exploration parameter data are fused through a self-shaping function embedding mechanism to construct a pollution action function, and a three-dimensional pollution potential field embedding tensor with geological response and source location is formed through function mapping; not only the heterogeneous data is converted into a unified format, but also the continuous expression is realized while the spatial heterogeneity is preserved, solving the problem that the traditional pollution monitoring data can only be statistically processed but cannot generate an evolution model, thereby improving the accuracy and expressiveness of the data.

[0026] 2. A geological disturbance function generator is designed, which fuses multiple factors such as geological fault influence, permeability discontinuity, and soil layer adsorption, and constructs a spatial correction tensor of pollution transport in combination with the spatial gradient of the pollution potential field embedding tensor, to realize dynamic processing of the disturbance response in different regions of the pollution propagation path, which is different from the simplified modeling method of the traditional groundwater pollution diffusion prediction method that uniformly smooths the geological structure, and can more truly reflect the diffusion path deflection, slowing down or aggregation behavior of the pollutants in the heterogeneous geological conditions.

[0027] 3. A diffusion channel weight factor is introduced, which performs pollution propagation modeling through path direction-aware convolution operation, introduces the concept of structure-coupled direction convolution, realizes the modeling of pollution diffusion in multiple dominant directions (such as along the strata bedding, hydraulic main flow, and slope direction), and forms a complete pollution migration prediction through tensor fusion. Compared with the traditional overall field modeling method, the combination causes of pollution flowing into a target area from different source points and different paths can be clearly restored, providing a solid data foundation for subsequent pollution tracing and regional risk classification. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 A flowchart of the prediction method of the groundwater pollution diffusion in the chemical industry park based on artificial intelligence. DETAILED DESCRIPTION

[0029] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined object of the application, the technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0031] The specific scheme of the method for predicting the diffusion of groundwater pollution in a chemical industrial park based on artificial intelligence provided by the present application will be specifically described below in combination with the drawings.

[0032] Referring to the drawings Figure 1 , which shows a flow chart of a method for predicting the diffusion of groundwater pollution in a chemical industrial park based on artificial intelligence according to an embodiment of the present application. The method comprises the following steps:

[0033] S1. Obtain the pollutant concentration information data in the groundwater of the chemical industrial park and the geological exploration parameter data, and pre-process to obtain pre-processed data; based on the pre-processed data, introduce a self-shaping function embedding mechanism, construct a pollution function, and generate a pollution potential field embedding tensor;

[0034] According to the expert experience method, deploy monitoring points in the underground of the chemical industrial park, and through a data acquisition device (such as a hydrogeological sensor), obtain the pollutant concentration information data in the groundwater of each monitoring point at different times, and the geological exploration parameter data of the location of the monitoring point, such as soil permeability coefficient, hydraulic gradient, groundwater flow information (such as underground flow rate change, direction, runoff angle, etc.), underground structure disturbance spectrum and sewage activity record, etc. The pollutant concentration information data and the geological exploration parameter data are pre-processed to obtain pre-processed data. The pre-processed data includes pre-processed pollutant concentration information data and pre-processed geological exploration parameter data. The pre-processing process includes denoising, missing value filling, cleaning, standardization and normalization, etc. All of them adopt technical means well known to those skilled in the art, which will not be described here.

[0035] In order to map the time discontinuous and spatial non-uniform distribution of pollution monitoring data into a continuous spatio-temporal tensor structure, based on the pre-processed data, the self-shaping function embedding mechanism is introduced, based on the nonlinear water flow diffusion formula in seepage theory, the model of non-Euclidean disturbance field is constructed, the pollution function is constructed, and the diffusion behavior of pollutants in three-dimensional space of each monitoring point is quantified; The pollution monitoring data is the pre-processed data; The pollution function of the monitoring point at a certain time point is the pollution influence projection of the three-dimensional space, and the specific form is:

[0036] ,

[0037] Among them, is the output value of the pollution function of the first monitoring point at time to point ; is the pre-processed pollution concentration information data of the first monitoring point at time ; is the soil permeability coefficient in the pre-processed geological exploration parameter data of the first monitoring point, which is an important geological parameter representing the water migration ability, and the reference value range is ; is the pollution influence dynamic amplification coefficient of the first monitoring point at time , which is determined by the pollution growth rate and the change of groundwater flow rate in the pre-processed geological exploration parameter data, and the reference value range is ; is the disturbance main frequency of the geological disturbance of the first monitoring point in the direction (used to simulate the irregular structure of the stratum), which is obtained according to stratum analysis and hydrological experiment simulation, and the reference value range is ; is the horizontal coordinate of the underground space; is the phase frequency of the geological disturbance of the first monitoring point in the direction (used to control wave superposition), and the reference value range is , which is obtained based on the existing geological model and numerical fitting, and is a technical means familiar to those skilled in the art, which will not be repeated here; is the vertical coordinate of the underground space; is the depth impedance factor of the first monitoring point, which is used to control the adsorption and attenuation degree of pollution with depth, and is obtained by soil layering resistance experiment fitting, and the reference value range is Technical means well known to those skilled in the art will not be described here; is the underground depth coordinate; represents the net effective diffusion potential of the pollution source, is the exponential scaling result of the coupling of the intensity of the pollution source (preprocessed pollutant concentration information data) and its geological release capacity (soil permeability coefficient in the preprocessed geological exploration parameter data), and is the energy source term of the outward action of the pollution source; is a spatial decay modulation term, used to simulate the nonlinear decay law of the action intensity of the pollution source with the spatial position in three-dimensional space;

[0038] The purpose of the above pollution action function is to convert the preprocessed pollutant concentration information data of the monitoring point into an influence function of the spatially deformed pollution field. Since the influence of each monitoring point is different, an attention weighting mechanism is introduced to assign weights to the pollution action function of each monitoring point The attention weighting mechanism is a technical means well known to those skilled in the art and will not be described here;

[0039] Further, the pollution action function is weighted based on the weights to construct a pollution potential field embedding tensor:

[0040] ,

[0041] wherein, is the pollution potential field embedding tensor at time , represents a dynamic pollution influence field with spatial coordinate dependence, and is the basic input for path simulation and disturbance modeling; is the weight of the pollution action function of the th monitoring point at time ; is the total number of monitoring points.

[0042] S2. Introduce a geological disturbance function generator to construct a spatial disturbance tensor, and combine the pollution potential field embedding tensor to generate a spatial correction tensor for pollution transmission; based on the pollution potential field embedding tensor and the spatial correction tensor for pollution transmission, model the pollution propagation to obtain a pollution diffusion prediction tensor, and realize the prediction of groundwater pollution diffusion.

[0043] The migration of groundwater in the medium does not follow the Euclidean continuity law, but is coupled with nonlinear factors such as stratigraphic boundary, porosity change, and permeability. Therefore, in order to accurately simulate the nonlinear propagation path of pollution in the ground, a geological disturbance function generator is introduced to model the disturbance behavior of the underground structure during pollution diffusion; the geological disturbance function generator constructs a spatial disturbance tensor by setting a series of disturbance kernels, each kernel simulating a nonlinear disturbance point at a spatial position, and based on a multi-disturbance coupling mechanism, etc.

[0044] ,

[0045] wherein, is a spatial perturbation tensor, representing the geological perturbation intensity in three-dimensional space and time ; is a perturbation kernel index, used for accumulation of multiple perturbation centers in the underground structure; is the total number of perturbation kernels, representing the number of perturbation centers, which is set according to specific scenarios and is not limited here; is a perturbation intensity time weight, representing the perturbation intensity of the th perturbation kernel at time , which is estimated based on the pre-processed geological exploration parameter data, such as groundwater flow velocity, pressure head change rate, etc., and the reference value range is , and the methods used are well known to those skilled in the art, and are not described here; is a perturbation angle, representing the deviation angle of the perturbation direction relative to the main flow direction, which is calculated based on the hydraulic gradient direction and the runoff angle in the pre-processed geological exploration parameter data, and the reference value range is , and the calculation method is well known to those skilled in the art, and is not described here; is a transverse perturbation kernel function, which is of the form , representing the influence radius from the th perturbation kernel center to any point; is a longitudinal perturbation function, defined as , used to simulate the influence decay of the perturbation in the geological depth direction, is the vertical central depth coordinate of the th perturbation kernel in the underground space; is a space-time oscillation frequency factor, used to control the amplitude of spatial fluctuations of the perturbation, which is estimated based on the underground structure perturbation spectrum in the pre-processed geological exploration parameter data, and the reference value range is , which is well known to those skilled in the art, and is not described here; is a spatial decay function of the perturbation intensity, used to simulate the influence decay of the perturbation kernel on the spatial position; is a space-time perturbation frequency modulation function, used to simulate the propagation oscillation phenomenon produced by the pollution path in the geological periodic structure, reflecting actual perturbations such as seepage layering alternation, reflected wave, underground pumping period, etc.; is the tangent value of the perturbation angle, which amplifies the nonlinear perturbation behavior through the angle, and is used to simulate the degree of deflection of the pollution path;

[0046] Based on the control theory model of the perturbation response tensor field and the reversible perturbation physical field construction method of the diffusion-anti-diffusion system, the spatial perturbation tensor is coupled with the aforementioned pollution potential field embedding tensor to generate a spatial correction tensor for pollution transport:

[0047] ,

[0048] wherein, is the spatial correction tensor for pollution transport, representing the perturbation intensity coefficient of the underground pollution path at position and time ; is the spatial gradient of the pollution potential field embedding tensor at time , representing the local change rate of pollution potential in three-dimensional space, which is an important reference index for the main direction and pollution diffusion capacity of the pollution diffusion path. The spatial partial derivative of the pollution potential field embedding tensor is obtained, which is a well-known technical means for those skilled in the art and will not be described here; is the perturbation amplification coefficient, used to control the amplification effect of time, simulate the enhancement of perturbation in certain special period (such as pollution discharge peak period, intensified infiltration during heavy rain, etc.), and determined according to specific scene, the reference value range is ; is a time modulation term, which intervenes in the perturbation response by introducing a time periodicity factor to simulate the physical phenomenon of perturbation intensity fluctuating with time; is an exponential amplification term of local perturbation intensity, used to simulate the nonlinear amplification effect of perturbation on pollution propagation capacity, i.e. the stronger the perturbation, the more significant the influence on diffusion path deflection or acceleration, which is similar to the nonlinear growth of perturbation flux with energy in physical field; is the response adjustment factor, representing the resistance factor under the adjustment of perturbation, used to simulate the response difficulty of diffusion path after being disturbed;

[0049] Based on the pollution potential field embedding tensor and the spatial correction tensor for pollution transport, pollution propagation modeling is carried out; the pollution potential field embedding tensor and the spatial correction tensor for pollution transport are subjected to path direction perception convolution operation to obtain a pollution diffusion prediction tensor :

[0050] ,

[0051] wherein, represents the evolution and distribution of pollutants in the entire three-dimensional space of the chemical industry park at time , which describes the dynamic trajectory of pollutants from the pollution source to the entire region in the form of tensor, i.e. the pollution diffusion prediction tensor; is the index of diffusion path / channel; is the total number of diffusion paths / channels, which is set according to specific scenarios based on existing propagation mechanisms and is not limited here; is the diffusion channel weight factor, indicating the pollutant The propagation proportion or energy weight on the diffusion path is obtained by regressing and fitting the historical pollution monitoring data from the existing database. The reference value range is , which are well known to those skilled in the art and will not be described in detail here; It is The pollution potential field on the channel is embedded in the tensor, indicating that at the current moment , pollution in the The potential energy impact formed by the superposition of multiple monitoring points on the channel; It is a directional three-dimensional path convolution operator, a tensor propagation operator with a spatially adjustable kernel, directional filtering and weight field mapping, which is equivalent to the finite volume method in geophysical simulation or the discrete propagator in FEM numerical diffusion modeling. In tensor calculations, represents the path direction aware convolution operation, which is used to propagate the potential contamination field to the surrounding grid nodes through the kernel weight in a certain direction; It is Spatially corrected tensor of contamination transport on the channel; It is a diffusion frequency modulation factor used to simulate the periodic fluctuations of pollution diffusion on a specific path, such as diurnal changes, tidal influences, and sewage discharge cycles. The diffusion frequency modulation factor uses Fourier transform to analyze the time-frequency characteristics of the pre-processed pollutant concentration information data and extracts the main frequency as the diffusion frequency adjustment parameter. , and finally the diffusion process is dynamically modulated by the cosine function; the diffusion frequency adjustment parameter The reference value range is ; The Fourier transform is a technical means well known to those skilled in the art and will not be described in detail here.

[0052] In summary, an artificial intelligence-based prediction method for groundwater pollution diffusion in chemical parks was completed.

[0053] The order in which the embodiments of the invention are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0055] The above examples are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that the technical solutions recorded in the foregoing examples can be modified, or some technical features can be replaced by equivalent features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for predicting the diffusion of groundwater pollution in a chemical industrial park based on artificial intelligence, characterized in that, The method comprises the following steps: S1. Obtain the pollutant concentration information data in the groundwater of the chemical industrial park and the geological exploration parameter data, and preprocess to obtain preprocessed data; Based on the preprocessed data, a self-shaping function embedding mechanism is introduced, a pollution action function is constructed through a spatial decay modulation term, and a pollution potential field embedding tensor is generated; S2. Introduce a geological disturbance function generator, obtain the disturbance intensity time weight, disturbance angle and space-time oscillation frequency factor based on the preprocessed geological exploration parameter data, construct the spatial decay function of disturbance intensity and the space-time disturbance frequency modulation function based on the disturbance angle and the space-time oscillation frequency factor, and construct the spatial disturbance tensor combined with the disturbance intensity time weight, and generate the spatial correction tensor of pollution transmission combined with the pollution potential field embedding tensor; based on the pollution potential field embedding tensor and the spatial correction tensor of pollution transmission, pollution propagation modeling is performed to obtain a pollution diffusion prediction tensor, realizing the prediction of groundwater pollution diffusion.

2. The method according to claim 1, wherein, The S1 specifically comprises: The preprocessed data comprises preprocessed pollutant concentration information data and preprocessed geological exploration parameter data; in the implementation process of the self-shaping function embedding mechanism, based on the soil permeability coefficient in the preprocessed pollutant concentration information data and the preprocessed geological exploration parameter data, a spatial decay modulation term is introduced to construct a pollution action function, and the diffusion behavior of pollutants in three-dimensional space is quantified.

3. The method according to claim 2, wherein, The S1 specifically comprises: An attention weighting mechanism is introduced to weight the pollution action function, and a pollution potential field embedding tensor is constructed.

4. The method according to claim 3, wherein, The S2 specifically comprises: The geological disturbance function generator simulates nonlinear disturbance points at spatial positions by setting disturbance kernels.

5. The method for predicting groundwater pollution spread in a chemical park based on artificial intelligence according to claim 1, characterized in that: The S2 specifically comprises: Based on the spatial disturbance tensor and the pollution potential field embedding tensor, an exponential amplification term of local disturbance intensity and a response adjustment factor are introduced to generate a spatial correction tensor of pollution transmission.

6. The method according to claim 5, wherein, The S2 specifically comprises: A diffusion channel weight factor is introduced to perform path direction perception convolution operation on the pollution potential field embedding tensor and the spatial correction tensor of pollution transmission combined with the diffusion frequency modulation factor, to obtain a pollution diffusion prediction tensor.

7. The method according to claim 6, wherein, The S2 specifically comprises: The diffusion frequency modulation factor dynamically modulates the diffusion process by analyzing the time-frequency characteristics of the preprocessed pollutant concentration information data combined with a cosine function.

Images