Method and system for monitoring soil pollutants of power transmission and transformation project

Abstract

The invention relates to the technical field of soil monitoring, and discloses a power transmission and transformation project soil pollutant monitoring method and system, and the method comprises the steps: collecting hyperspectral image data in a power transmission and transformation project region; processing the hyperspectral image data by adopting a multi-stage spectrum transformation mechanism to obtain a characteristic wave band representing the pollutant content; inputting the characteristic wave band into a preset spectrum inversion model to generate a concentration spatial distribution diagram of the soil pollutants; and constructing a soil and groundwater diffusion model according to the concentration spatial distribution map, and generating a migration path identification map of soil pollutants and an underground pollution risk area identification map based on the soil and groundwater diffusion model. The method has the beneficial effect of improving pollution identification precision and migration prediction accuracy.

Description

Technical Field

[0001] The present application relates to the field of soil monitoring technology, and in particular to a method and system for monitoring soil pollutants in power transmission and transformation projects. Background Art

[0002] As the core vehicle of ecosystems, soil performs critical functions such as material transformation, energy flow, and water regulation. Its quality is directly linked to food security and public health. However, industrialization has led to the intrusion of heavy metals (such as cadmium and lead), persistent organic pollutants (POPs), and petroleum hydrocarbons into soil, causing a surge in pollution levels exceeding permitted standards. Power transmission and transformation projects, as crucial energy infrastructure, also pose a significant challenge to soil pollution within their areas. These projects involve the construction and operation of substations, switchyards, and transmission line towers. During these processes, soil may become contaminated with characteristic pollutants due to equipment leakage (such as transformer oil and sulfur hexafluoride decomposition products), material corrosion (such as heavy metal residues), and the accumulation of construction waste. This type of pollution is highly concealed, has complex diffusion pathways (influenced by groundwater flow and soil infiltration), and is closely linked to the distribution of engineering equipment. This places special demands on the timeliness, accuracy, and spatial coverage of monitoring technologies. Existing soil pollution monitoring methods for power transmission and transformation projects suffer from slow response, low accuracy, and difficulty in prediction, hindering timely and targeted pollution control. Summary of the Invention

[0003] The present application provides a method and system for monitoring soil pollutants in power transmission and transformation projects, which solves the technical problems of insufficient sensitivity and difficulty in prediction of related monitoring methods, and achieves the technical effect of improving the accuracy of pollution identification and migration prediction.

[0004] In order to achieve the above objectives, the main technical solutions adopted in this application include: In a first aspect, an embodiment of the present application provides a method for monitoring soil pollutants in a power transmission and transformation project, the method comprising: collecting hyperspectral image data within the power transmission and transformation project area; processing the hyperspectral image data using a multi-level spectral transformation mechanism to obtain characteristic bands characterizing the pollutant content; inputting the characteristic bands into a preset spectral inversion model to generate a spatial distribution map of soil pollutant concentrations; constructing a soil and groundwater diffusion model based on the spatial distribution map of concentrations, and generating a migration path identification map of soil pollutants and an underground pollution risk area identification map based on the soil and groundwater diffusion model.

[0005] The method for monitoring soil pollutants in power transmission and transformation projects proposed in the embodiment of the present application collects hyperspectral image data within the power transmission and transformation project area; uses a multi-level spectral transformation mechanism to process the hyperspectral image data to obtain characteristic bands that characterize the pollutant content; inputs the characteristic bands into a preset spectral inversion model to generate a spatial distribution map of soil pollutant concentrations; constructs a soil and groundwater diffusion model based on the concentration spatial distribution map, and based on the soil and groundwater diffusion model, generates a migration path identification map of soil pollutants and an underground pollution risk area identification map, thereby solving the technical problems of insufficient sensitivity and difficult prediction of related monitoring methods and achieving the technical effect of improving the accuracy of pollution identification and migration prediction.

[0006] Optionally, the method further includes: collecting the hyperspectral image data through a plurality of monitoring points preset within the power transmission and transformation project area; wherein each monitoring point is provided with a differentiated marker pollutant, and each marker pollutant has a sensitive characteristic in a specific band; the sensitive characteristic is a regular change pattern of the spectral reflectance curve caused by an increase in pollutant concentration.

[0007] Optionally, the method also includes: performing spatiotemporal alignment of multi-phase hyperspectral image data with ground in-situ data to generate a pollutant baseline concentration matrix; the pollutant baseline concentration matrix contains four-tuple data of position, time, spectrum and concentration; based on the pollutant baseline concentration matrix and the groundwater diffusion model, a pollutant diffusion model is constructed, and based on the pollutant diffusion model, time series spectral features are extracted; the time series spectral features are used to characterize the concentration changes and spectral dynamic changes of pollutants; according to the time series spectral features and a preset concentration change rate threshold, the pollutant concentration enhanced area is graded and calibrated.

[0008] Optionally, the method further includes: processing the hyperspectral image data through a recursive feature elimination method to obtain characteristic bands characterizing the pollutant content; based on the characteristic bands, judging the pollutant concentration level of each monitoring point through adaptive threshold mapping, and generating a pollutant distribution map, wherein the adaptive threshold mapping is used to associate the pollutant type and the characteristic band threshold, and the width of the characteristic band is used to characterize the pollutant concentration.

[0009] Optionally, the method further includes: constructing an adversarial hyperspectral sample that simulates the interference environment of a power transmission and transformation project; identifying abnormal disturbances in the adaptive threshold mapping process based on the adversarial hyperspectral sample; and correcting the pollutant concentration of the characteristic band based on the abnormal disturbance.

[0010] Optionally, the characteristic band is input into a spectral inversion model, including: determining a dynamic sensitive band based on the pollutant absorption peak matched by the characteristic band, wherein the dynamic sensitive band is used to characterize a spectral band with significant sensitivity to changes in pollutant concentration or migration processes; processing the dynamic sensitive band through the spectral inversion model to generate a spatial distribution map of soil pollutant concentrations, wherein the spatial distribution map of soil pollutants includes initial distribution data of soil pollutants and real-time distribution data of soil pollutants diffusing with groundwater at any moment.

[0011] Optionally, the concentration spatial distribution map includes a geographic elevation display map based on the power transmission and transformation project area; the geographic elevation display map is constructed based on the fusion of dynamic environmental parameters and ground in-situ monitoring data; based on the geographic elevation display map and the migration risk value, a migration path prediction map of soil pollutants is generated, and the migration path prediction map is used to characterize the migration trend of pollutants; based on the migration path prediction map and the migration risk value, a migration path identification map of soil pollutants is generated.

[0012] Optionally, based on the concentration spatial distribution map, constructing a soil and groundwater diffusion model includes: loading the concentration spatial distribution map into a preset physical diffusion model to generate a pollutant diffusion matrix based on three-dimensional migration data; the pollutant diffusion matrix includes a soil invasive diffusion matrix and a groundwater high-speed dispersion diffusion matrix; based on the pollutant diffusion matrix, determining the spatiotemporal state characteristics of pollutants, the spatiotemporal state characteristics are the diffusion state data of pollutants in the time dimension and the space dimension; based on the spatiotemporal state characteristics, constructing a soil and groundwater diffusion model.

[0013] Optionally, based on the soil and groundwater diffusion model, a migration path identification map of soil pollutants and an underground pollution risk area identification map are generated, including: generating three-dimensional migration data based on the soil and groundwater diffusion model; determining the diffusion weights of different migration and diffusion coordinate points on the migration trajectory based on the three-dimensional migration data, and determining the emergency control risk value based on the diffusion weight; generating a migration path identification map based on the emergency control risk value; performing vulnerability calculation of the power transmission and transformation project area based on the migration path identification map to obtain vulnerability values ​​of different areas, and generating an underground pollution risk area identification map based on the vulnerability values ​​and the emergency control risk value.

[0014] In a second aspect, an embodiment of the present application provides a monitoring system for soil pollutants in a power transmission and transformation project, comprising: an acquisition module for acquiring hyperspectral image data within the power transmission and transformation project area; a transformation module for performing spectral transformation processing on the hyperspectral image data to obtain characteristic bands representing the pollutant content; a generation module for inputting the characteristic bands into a spectral inversion model to generate a spatial distribution map of soil pollutant concentrations; an identification module for constructing a soil and groundwater diffusion model based on the spatial distribution map of concentrations; and for generating a migration path identification map of soil pollutants and an underground pollution risk area identification map based on the soil and groundwater diffusion model.

[0015] In a third aspect, an embodiment of the present application provides a computer device comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the above-mentioned method for monitoring soil pollutants in power transmission and transformation projects by executing the computer instructions.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the above-mentioned method for monitoring soil pollutants in power transmission and transformation projects.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the above-mentioned method for monitoring soil pollutants in power transmission and transformation projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1 A flow chart of a method for monitoring soil pollutants in a power transmission and transformation project provided in an embodiment of the present application; Figure 2 A schematic diagram of a soil pollutant monitoring system for power transmission and transformation projects provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0021] As a core component of ecosystems, soil is the central medium for key ecological processes such as material transformation, energy flow, and water regulation. Its quality is directly linked to food security and public health. However, industrialization has led to the influx and accumulation of heavy metals (such as cadmium and lead), persistent organic pollutants (POPs), and petroleum hydrocarbons in the soil, causing severe pollution. This type of pollution is complex, differentiated, and covert. This is particularly true in power transmission and transformation areas, where unique pollution sources such as transformer oil leakage and the release of sulfur hexafluoride decomposition products further complicate remediation.

[0022] Traditional monitoring methods primarily rely on laboratory chemical analysis, such as atomic absorption spectroscopy (AAS) and gas chromatography-mass spectrometry (GC-MS). While highly accurate, these methods suffer from long sampling cycles (often requiring several weeks), high costs (over 1,000 yuan per sample), and limited spatial coverage, making them inadequate for large-scale, dynamic monitoring. In recent years, the integrated application of 3S technologies (remote sensing - RS, geographic information system (GIS), and global positioning system (GPS)) has significantly improved monitoring efficiency. For example, hyperspectral remote sensing can identify specific absorption bands of heavy metals in soil, enabling rapid delineation of contamination areas when combined with GIS spatial analysis. Biomonitoring techniques indirectly reflect the degree of pollution stress by detecting changes in soil microbial community structure or plant physiological responses, but their quantitative accuracy is significantly affected by environmental factors. Furthermore, the development of IoT sensing technology has led to the development of real-time soil parameter monitoring networks. Electrochemical sensors and fluorescent probes can provide online data on pH, conductivity, and specific pollutant concentrations. However, their application in complex soil matrices remains unsatisfactory due to limitations in sensor stability and interference tolerance. Overall, existing soil pollution monitoring technology faces three major challenges: Traditional laboratory methods are inefficient, requiring multiple manual steps from sampling, pre-processing, and analysis, making them difficult to meet the emergency response needs of sudden pollution incidents; on-site rapid detection equipment lacks sensitivity, limiting its ability to identify low-concentration pollutants (such as heavy metals at the μg / kg level), which can easily lead to missed detections and false positives; and multi-source data fusion capabilities are weak, with a lack of a unified scaling system for remote sensing inversion, ground sensing, and laboratory data, limiting the accuracy of pollution migration modeling. For example, in a chemical site restoration project, traditional grid sampling required the deployment of hundreds of points and took three months to complete the initial assessment. During this time, the pollution plume had already spread to downstream water sources, highlighting the passive situation brought about by these challenges.

[0023] An embodiment of the present application provides a method for monitoring soil pollutants in a power transmission and transformation project. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0024] Please refer to Figure 1 , Figure 1 The flowchart of the method for monitoring soil pollutants in power transmission and transformation projects provided in the embodiment of the present application is as follows: Figure 1 As shown, the process includes the following steps: Step S100: collecting hyperspectral image data within the power transmission and transformation project area.

[0025] Hyperspectral imagery data contains reflectance information on ground objects and can reveal the spectral characteristics of soil pollutants. During the acquisition process, aerial scanning is performed using a drone equipped with a hyperspectral imager. Continuous spectral imaging across multiple wavelengths is used to obtain a soil surface reflectance curve. This curve reveals characteristic absorption valleys of pollutants in the visible and near-infrared bands. Second-order derivative spectroscopy is used to amplify weak characteristic signals, revealing the unique spectral absorption characteristics of pollutants such as heavy metals and oil.

[0026] Step S300 : Processing the hyperspectral image data using a multi-level spectral transformation mechanism to obtain characteristic bands representing pollutant content.

[0027] Raw hyperspectral image data contains noise and redundancy. Spectral transformation involves applying a second-order derivative transform to the hyperspectral image data to highlight and enhance spectral curve variations that characterize pollutants. This multi-stage spectral transformation mechanism includes a first-stage spectral transformation mechanism to eliminate noise, a second-stage spectral transformation mechanism to enhance differences in pollutant characteristic bands, and a third-stage spectral transformation mechanism to determine the multi-scale spectral characteristics of pollutants. The first-stage processing utilizes a wavelet packet decomposition and reconstruction algorithm to separate instrument noise and environmental interference in the frequency domain. By establishing a noise signature library, the optimal denoising basis function is dynamically matched to preserve the effective spectral information of pollutants. The second-stage processing develops a spectral differential enhancement operator that uses fractional derivatives to amplify detailed differences in reflectance curves within characteristic bands. Adaptive selection of first- and second-order differentials is employed for different pollutant types (such as oils and heavy metals), overcoming the sensitivity limitations of traditional integer derivatives. The third-stage processing constructs a pyramidal analysis model to extract cross-scale spectral characteristics of pollutant particles through Gaussian scale-space transformation. Morphological opening and closing operations are combined to separate the spectral response patterns of pollutant clusters from those of discrete distributions.

[0028] A recursive feature elimination method is used to screen out band combinations that show a significant correlation with the target pollutant concentration, and the band index is used to construct a quantitative pollution degree. In the process of quantifying the pollution degree, the hyperspectral image data is first used as the initial feature set, and a preset regression model is trained to determine the mapping relationship between spectral reflectance and pollutant concentration. Based on the importance weights of different bands (weights are set according to the bands and corresponding pollutants), the bands with low importance rankings are eliminated to screen out characteristic bands. Characteristic bands are bands that show a significant correlation with the target pollutant concentration. For example, the characteristic band of Pb is 620-650nm, and the 620-650nm band is then screened as the characteristic band of Pb. The band index represents an index for quantifying the pollution degree, based on the physical correlation between optical reflectance characteristics and the chemical properties of pollutants, such as: heavy metals cause specific absorption peak shifts.

[0029] Step S500: input the characteristic waveband into a preset spectral inversion model to generate a spatial distribution map of soil pollutant concentrations.

[0030] Spectral inversion models, based on machine learning algorithms or physical models (such as radiation transfer models), determine the mapping relationship between spectral features and concentrations through training samples. Spatial concentration distribution maps include both initial soil contaminant distribution data and real-time distribution data of soil contaminants as they diffuse with groundwater at any given moment. Initial distribution data is a static distribution map generated by calculating pollutant concentrations at the moment of monitoring using the spectral inversion model. Real-time diffusion data is dynamically updated by combining groundwater flow direction (obtained through hydrological models) with time-series hyperspectral imagery.

[0031] Step S700: constructing a soil and groundwater diffusion model according to the concentration spatial distribution map, and generating a soil pollutant migration path identification map and an underground pollution risk area identification map based on the soil and groundwater diffusion model.

[0032] Based on the initial distribution data and real-time distribution data, a soil and groundwater diffusion model is constructed to generate a soil pollutant migration path identification map and an underground pollution risk area identification map. The diffusion model is based on the convection-diffusion equation. By inputting soil parameters, groundwater parameters and pollutant parameters, numerical simulation is performed to calculate the migration path of pollutants in soil and groundwater, and then combined with the risk assessment standards, the risk areas are divided, and the corresponding migration path identification map and underground pollution risk area identification map are generated. The embodiment of the present application uses a mobile hyperspectral inspection device to achieve rapid scanning of the radius range of power transmission and transformation equipment, which is particularly suitable for power transmission and transformation corridors in mountainous areas with complex terrain. Spectral transformation processing reduces the detection limit of heavy metals and improves the accuracy of identifying oil pollutants. By integrating soil permeability coefficient and groundwater velocity field data, the future diffusion range of pollution plumes can be predicted, which is more timely than the static assessment model. The location of pollution barrier trenches is accurately delineated through the migration path identification map, which reduces the amount of soil remediation work.

[0033] In some embodiments, the hyperspectral image data is collected by presetting a plurality of monitoring points within the power transmission and transformation project area; wherein each monitoring point is provided with a differentiated marker pollutant, and each marker pollutant has a sensitive characteristic in a specific band; the sensitive characteristic is a data pattern in which the spectral reflectance curve of the pollutant shows a regular change with increasing concentration within a preset time period.

[0034] When pollutant monitoring is carried out in any power transmission and transformation project area, monitoring points will first be set up according to the main power transmission and transformation equipment in the power transmission and transformation project area. The main power transmission and transformation equipment includes substations, switch stations, series compensation stations, and transmission line towers and other project equipment. The monitoring points are set up in a ring around the main power transmission and transformation equipment. The grid method and zoning method can also be used to arrange the points, with the coverage of sensitive areas around the main power transmission and transformation equipment as the standard.

[0035] Among these, monitoring points contain marker pollutants, each with a different type and sensitive characteristics in specific wavelength bands. These sensitive characteristics are the pollutant spectral reflectance curves within the power transmission and transformation project area, which show a diffusion and migration trend within a preset monitoring period, and the enhanced pollutant concentration data for dynamic pollutant hotspots. Power transmission and transformation project areas may contain multiple pollutants. By setting marker pollutants at different monitoring points, targeted hyperspectral imagery data strongly correlated with these pollutants can be collected. For typical power transmission and transformation project pollutants (transformer oil, sulfur hexafluoride decomposition products, heavy metals, etc.), a database of spectral reflectance benchmark curves with project characteristics is established. By setting differentiated marker pollutants, a multi-dimensional monitoring indicator system is formed, thereby addressing the problem of misjudgment caused by confusion over pollutant characteristics.

[0036] This embodiment of the application uses a time-domain convolutional neural network (TCN) to analyze the morphological evolution of spectral curves over a continuous monitoring period, capturing the gradient changes in reflectivity within characteristic bands. A dynamic threshold adjustment mechanism is used to identify band shifts that characterize pollutant migration. A pollutant diffusion path network is constructed based on graph theory, mapping changes in the concentration of signature pollutants at each monitoring point to changes in the weights of network nodes. The spatial boundaries of areas of increased concentration are determined by calculating the strength of connections between nodes.

[0037] Differentiated markers are used to identify pollution sources. Tracking characteristic bands allows for precise identification of responsible sources in mixed pollution scenarios, for example, differentiating between combined pollution from transformer oil leaks and battery acid leaks. Dynamic sensitive feature recognition technology captures subtle spectral changes in the early stages of pollution spread, advancing the detection window for pollution events by multiple monitoring cycles compared to traditional fixed-threshold monitoring methods. Analyzing the spatial distribution characteristics of marker pollutants allows for intelligent planning of drone inspection routes, increasing the frequency of monitoring in key areas without increasing equipment load.

[0038] In some embodiments, multi-phase hyperspectral image data are temporally and spatially aligned with ground in-situ data to generate a pollutant baseline concentration matrix; the pollutant baseline concentration matrix contains four-tuple data of position, time, spectrum and concentration; based on the pollutant baseline concentration matrix and the groundwater diffusion model, a pollutant diffusion model is constructed, and based on the pollutant diffusion model, time series spectral features are extracted; the time series spectral features are used to characterize the concentration changes and spectral dynamic changes of pollutants; according to the time series spectral features and a preset concentration change rate threshold, the pollutant concentration enhanced area is graded and calibrated.

[0039] Specifically, multi-temporal hyperspectral images and ground-based in-situ data are collected and spatially aligned to generate a pollutant baseline concentration matrix. The pollutant baseline concentration matrix is ​​used to represent the four-tuple data of the initial position, time, spectrum, and concentration of pollutants in the power transmission and transformation project area. Multi-temporal hyperspectral images are data collected by drone / satellite hyperspectral sensors at different time points. Multi-temporal hyperspectral images cover the spectral reflectance information of the power transmission and transformation project area. On the same date of the corresponding temporal phase, the ground-based in-situ data obtains the actual pollutant concentration data at each point through soil sampling. Spatial and temporal alignment is to match the pixels of the hyperspectral image with the positions of the ground sampling points through GPS coordinates and synchronize the timestamps. The pollutant baseline concentration matrix integrates the aligned multi-temporal data into a three-dimensional matrix. Each element stores the correlation value of the hyperspectral reflectance and actual concentration at the corresponding position and time. Dynamic calibration is performed during this process. Dynamic calibration means that the matrix is ​​automatically updated as new temporal phase data is added.

[0040] Based on the pollutant baseline concentration matrix, a pollutant diffusion model for the power transmission and transformation project area is constructed to determine the time-series spectral characteristics. The pollutant diffusion model is composed of a groundwater diffusion model for the power transmission and transformation project area combined with the pollutant baseline concentration matrix. The time-series spectral characteristics include the concentration change rate of the pollutants and the spectral dynamic change characteristics. Using the baseline concentration matrix as input, combined with hydrogeological parameters, the convection-diffusion equation is used to simulate pollutant migration, thereby constructing the pollutant diffusion model. The concentration change rate is calculated by calculating the concentration increment at each location and time phase in the baseline matrix. The spectral dynamic change characteristics are determined by extracting the spectral reflectance changes at the same spatial location and different time points in the baseline matrix, and then calculating the quantitative value of the change trend represented by a specific band combination.

[0041] Based on time-series spectral characteristics, the area of ​​increased pollutant concentration corresponding to each marker pollutant within the power transmission and transformation project area is graded and calibrated under a preset concentration change rate threshold. The concentration change rate threshold is set based on the risk acceptance level of the power transmission and transformation project, such as specific values ​​for high, medium, and low level thresholds. Based on the spectral dynamic change separation model, independent component analysis (ICA) is used to decouple pollutant concentration changes from soil background interference (humidity and organic matter changes), extracting time-series characteristics that only reflect the pollution diffusion process. A concentration grading system based on migration trends dynamically divides core pollution areas, diffusion-affected areas, and potential risk areas according to the concentration gradient change rate along the diffusion direction of the pollution plume, forming a hierarchical control map with spatial topological relationships.

[0042] The embodiment of the present application realizes the dual functions of reverse deduction and forward prediction of the diffusion path of pollutants through the spatiotemporal fusion of multi-phase data, and can accurately restore the dynamic evolution process after the pollution incident occurs. The environmental background noise is effectively removed through the time series feature decoupling technology, so that the signal-to-noise ratio of weak pollution signals is improved to an engineering recognizable level, which is particularly suitable for continuous monitoring in scenarios where soil moisture changes drastically during the rainy season. Based on the dynamic calibration system of concentration gradients, differentiated emergency response strategies are automatically generated: physical barriers are implemented in the core area, chemical remediation is initiated in the diffusion area, and a three-level prevention and control network for monitoring and early warning is deployed in the risk area.

[0043] In some embodiments, the hyperspectral image data is processed by a recursive feature elimination method to obtain characteristic bands that characterize the pollutant content; based on the characteristic bands, the pollutant concentration level of each monitoring point is determined by adaptive threshold mapping to generate a pollutant distribution map, wherein the adaptive threshold mapping is used to associate the pollutant type and the characteristic band threshold, and the width of the characteristic band is used to characterize the pollutant concentration.

[0044] Hyperspectral image data is adaptively threshold mapped through a multi-level spectral transformation mechanism to generate a pollutant distribution map based on the concentration markers of the monitoring points; the adaptive threshold mapping is an association mapping between pollutant type and characteristic band threshold, and the width of the characteristic band is the pollutant concentration.

[0045] A hierarchical progressive processing mechanism effectively extracts characteristic spectra of pollutants that are often obscured by the environmental background, making it particularly suitable for early identification of trace leaks in aging equipment. Multi-scale feature analysis technology simultaneously identifies oil film contamination and heavy metal particulate contamination, resolving the problem of misidentification caused by spectral mixing. An adaptive threshold mapping mechanism automatically matches optimal processing parameters for different transmission and transformation project scenarios (such as converter stations and switch stations), reducing reliance on experts for manual parameter adjustment.

[0046] During the implementation process, pollutant concentrations are determined through spectral signatures to generate a pollutant distribution map. First, for hyperspectral image data, various types of representative hyperspectral image data samples are acquired from different monitoring points. Then, a multi-level spectral transformation mechanism is used to perform spectral measurements on these hyperspectral image data samples, obtaining characteristic band data as the independent variable and determining pollutant content as the dependent variable. The collected spectral data and pollutant element content data are cleaned to remove outliers and erroneous data, and standardized to make the data comparable. Spectral signatures are then used to establish an effective pollutant content prediction model and generate a pollutant distribution map.

[0047] In some embodiments, an adversarial hyperspectral sample is constructed to simulate the interference environment of a power transmission and transformation project; based on the adversarial hyperspectral sample, abnormal disturbances are identified during the adaptive threshold mapping process; and based on the abnormal disturbances, the pollutant concentration of the characteristic band is corrected.

[0048] Adversarial hyperspectral samples are used to simulate the interference environment in power transmission and transformation project areas. Based on a pre-set library of typical transmission and transformation interference spectral signatures (including electromagnetic pulse interference, equipment thermal radiation noise, and seasonal vegetation changes), a generative adversarial network constructs a physically interpretable adversarial hyperspectral sample set, forming a spectral interference dictionary covering all operating conditions. During adaptive threshold mapping, the adversarial hyperspectral samples are transferred to hyperspectral image data to determine whether anomalous disturbances exist. A feature transfer algorithm based on the residual attention mechanism maps the interference feature space of the adversarial hyperspectral samples to the actual monitoring data domain, and anomalous disturbance components are identified using a channel attention weight matrix. Based on these anomalous disturbances, pollutant concentrations attributable to characteristic bands at different monitoring points are determined. A dual-channel parallel processing architecture is constructed: the primary channel performs the concentration mapping process, while the auxiliary channel dynamically adjusts the characteristic band weight coefficients output by the primary channel by comparing the perturbation patterns of the adversarial hyperspectral samples.

[0049] The embodiments of the present application solve the problem of spectral distortion in the strong electromagnetic environment of power transmission and transformation projects, ensure the validity of pollution monitoring data in the domain during equipment operation, and solve the pain point of strong coupling between monitoring data and equipment start and stop status in traditional methods. Through the dynamic expansion mechanism of the adversarial hyperspectral sample library, the system has the adaptive ability to cope with special scenarios such as seasonal vegetation cover changes and metal dust interference during equipment maintenance. Through abnormal disturbance separation, it is possible to effectively distinguish between real pollution caused by equipment leakage and instantaneous interference caused by construction activities, providing technical evidence for the determination of accident responsibility.

[0050] In some embodiments, the characteristic band is input into a spectral inversion model, including: determining a dynamic sensitive band based on the pollutant absorption peak matched by the characteristic band, wherein the dynamic sensitive band is used to characterize a spectral band with significant sensitivity to changes in pollutant concentration or migration processes; processing the dynamic sensitive band through the spectral inversion model to generate a spatial distribution map of soil pollutant concentrations, wherein the spatial distribution map of soil pollutants includes initial distribution data of soil pollutants and real-time distribution data of soil pollutants diffusing with groundwater at any moment.

[0051] Match the characteristic bands with the pollutant absorption peaks that conform to the corresponding bands, and determine the dynamic sensitive bands; the dynamic sensitive bands are used to characterize the spectral bands whose sensitivity to pollutant concentration or migration process changes significantly with time or environmental conditions; by establishing a dynamic tracking algorithm for the pollutant spectral fingerprint, the absorption valley morphological characteristics of the target pollutant in the characteristic band are matched through a sliding window (such as absorption valley width, symmetry, and redshift), and the absorption peak deformation law affected by environmental factors is identified. By developing a band optimization algorithm based on migration trends, the sensitive band combination weights are dynamically adjusted according to the spectral response gradient changes in the pollutant diffusion direction to form a band set that adapts to different migration stages.

[0052] Based on the dynamic sensitive bands, a hybrid-driven inversion model is constructed to generate a migration risk value based on pollutant concentration. The migration risk value is determined by fusing spectral features with simulated diffusion features. The hybrid-driven spectral inversion model is implemented by fusing data-driven (spectral features) with physical-driven (diffusion model). The data-driven approach inverts current pollutant concentrations based on the input reflectance data of the dynamic sensitive bands. The physical-driven approach is determined by inputting hydrogeological parameters and historical concentration data and simulating future pollution migration trends using the advection-dispersion equation (ADE).

[0053] By constructing a neural network architecture constrained by physical equations, the fluid dynamics diffusion equation is embedded in the network's hidden layer. A spatiotemporal coupled expression of pollution risk is achieved through tensor fusion of spectral and migration features. The current concentration from spectral inversion is integrated with features such as migration rate and direction from the diffusion model to determine the migration risk value.

[0054] This embodiment of the application integrates spectral inversion based on physical mechanisms, overcoming the overfitting limitations of traditional data-driven models and achieving millimeter-level spatial resolution of pollution migration paths under complex geological conditions. A sensitive band evolution mechanism enables the model to adapt to different diffusion phases (infiltration, stabilization, and acceleration), accurately capturing the trajectory of the advancing pollution plume. The spatiotemporal coupled expression of risk values ​​generates actionable prevention and control heat maps, guiding the timing of containment project implementation and the allocation of remediation resources.

[0055] In some embodiments, the concentration spatial distribution map includes a geographic elevation display map based on the power transmission and transformation project area; the geographic elevation display map is constructed based on the fusion of dynamic environmental parameters and ground in-situ monitoring data; based on the geographic elevation display map and the migration risk value, a migration path prediction map of soil pollutants is generated, and the migration path prediction map is used to characterize the migration trend of pollutants; based on the migration path prediction map and the migration risk value, a migration path identification map of soil pollutants is generated.

[0056] The spatial concentration distribution map includes a geographic elevation map based on the transmission and transformation project area. This map, based on dynamic environmental parameters and ground-based in-situ monitoring data, forms a pollutant migration path prediction map. This migration path prediction map then generates spatial dynamic evolution parameters for pollutant concentrations based on concentration predictions, based on migration risk values. A deep coupling algorithm, based on terrain elevation data and pollutant parameters, automatically corrects the spatial distribution functions of gravitational potential energy and seepage pressure in the diffusion model through slope and aspect analysis of the digital elevation model (DEM), addressing the error accumulation problem caused by simplified topographic factors in traditional two-dimensional models. A particle system is used to simulate the spatiotemporal evolution of pollution diffusion, mapping migration risk values ​​to particle emission rates, motion trajectories, and attenuation coefficients. Using fluid dynamics-constrained particle behavior rules (such as Brownian motion superimposed on convection vectors), the movement of the pollutant diffusion front can be visualized. An adaptive interpolation model is constructed that combines dynamic environmental parameters (wind speed and humidity) with ground-based monitoring data. Using the Kriging spatial interpolation algorithm, high-resolution environmental field data is generated, providing physically driven boundary conditions for migration path prediction.

[0057] The embodiments of this application achieve simultaneous simulation of multi-media diffusion processes involving underground infiltration, surface runoff, and atmospheric deposition through deep coupling of elevation data and fluid parameters, accurately identifying hidden migration paths in complex terrains such as mountain substations. Dynamic evolution parameters support timeline sliding control of the pollution diffusion process, simulating the inhibitory effects of different treatment options (such as excavation and isolation, chemical neutralization) on the movement of pollution fronts, and providing a digital testing ground for emergency response plan selection. The three-dimensional visualization simulation system breaks down the disciplinary barriers between environmental monitoring and civil engineering, and the output results can directly guide the design of engineering details such as the construction angle of the cutoff wall and the location of the repair agent injection point.

[0058] In some embodiments, constructing a soil and groundwater diffusion model based on the concentration spatial distribution map includes: loading the concentration spatial distribution map into a preset physical diffusion model to generate a pollutant diffusion matrix based on three-dimensional migration data; the pollutant diffusion matrix includes a soil invasive diffusion matrix and a groundwater high-speed dispersion diffusion matrix; determining the spatiotemporal state characteristics of pollutants based on the pollutant diffusion matrix, the spatiotemporal state characteristics being the diffusion state data of pollutants in the time dimension and the space dimension; and constructing a soil and groundwater diffusion model based on the spatiotemporal state characteristics.

[0059] A physical diffusion model is pre-built and the spatial concentration distribution map is loaded into the model to generate a pollutant diffusion matrix based on three-dimensional migration data. This matrix includes an intrusive diffusion matrix based on soil and a high-velocity dispersed diffusion matrix based on groundwater. A decoupling algorithm for the soil-groundwater system is developed, establishing independent governing equations for soil capillary infiltration (a modified Darcy's law model) and groundwater diffusion migration (a heterogeneous convection-diffusion equation). An interfacial flux coupler is used to achieve a dynamic balance of pollutant transport across the medium, overcoming the simplification error bottleneck of traditional single-medium models. The finite volume method is used to spatially discretize the physical model, mapping the concentration distribution map to a three-dimensional grid initial condition. A migration matrix is ​​generated through time-stepping, including the longitudinal penetration depth, lateral diffusion radius, and time evolution dimensions, achieving a fully dimensional parameterized representation of pollution diffusion. Based on the pollutant diffusion matrix, the spatiotemporal state characteristics of the dynamic diffusion of pollutants are determined. Based on these spatiotemporal state characteristics, the soil expansion model and groundwater diffusion model for the pollutants are then determined. Finally, by constructing the eigenvalue decomposition algorithm of the migration matrix, the diffusion rate tensor in the main infiltration direction, the pollution plume morphological parameters (aspect ratio / fractal dimension) and the time-varying stability coefficient are extracted to form an interpretable diffusion dynamics fingerprint.

[0060] The embodiments of the present application independently construct soil and groundwater diffusion models and couple them with the interface to accurately distinguish between vertical migration dominated by capillary seepage and horizontal diffusion driven by hydraulic gradient, thereby solving the prediction distortion problem of fuzzy processing of cross-media processes in traditional methods.

[0061] The spatiotemporal feature extraction technology used in this application can identify abnormal diffusion signals from hidden channels such as bedrock fissures and abandoned pipeline corridors, providing early warning of the risk of contamination bypass in heterogeneous media. The differential outputs of the dual-medium model guide the development of a coordinated remediation solution using soil vapor extraction and a groundwater permeable reaction wall, avoiding the back-infiltration of contaminants between media caused by a single remediation technology.

[0062] In some embodiments, based on the soil and groundwater diffusion model, a migration path identification map of soil pollutants and an underground pollution risk area identification map are generated, including: generating three-dimensional migration data based on the soil and groundwater diffusion model; determining the diffusion weights of different migration diffusion coordinate points on the migration trajectory based on the three-dimensional migration data, and determining the emergency control risk value based on the diffusion weight; generating a migration path identification map based on the emergency control risk value; performing vulnerability calculation of the power transmission and transformation project area based on the migration path identification map to obtain vulnerability values ​​of different areas, and generating an underground pollution risk area identification map based on the vulnerability values ​​and the emergency control risk value.

[0063] Among them, according to the three-dimensional migration data, the diffusion weights of different migration diffusion coordinate points on the migration trajectory are determined, and the emergency control risk value based on the diffusion weight is determined. A three-dimensional trajectory energy field model based on migration potential is developed, and a multi-factor coupling weight calculation function is constructed through the diffusion direction of pollutants, the medium adsorption coefficient and the terrain potential gradient to quantify the potential contribution of each migration path to the ecosystem. According to the emergency control risk value, a migration path identification map of the corresponding migration path is generated; a fuzzy membership function of the risk value-control level is established, and the continuous risk value is discretized into a three-level control system of emergency disposal area, monitoring and early warning area, and normal observation area to achieve a visual spatial mapping of risk decision-making. According to the migration path identification map, the vulnerability of the power transmission and transformation project area is calculated, and a migration path identification map is constructed according to the vulnerability values ​​of different areas. This application designs a vulnerability evaluation index system including soil ecological sensitivity, groundwater use function, and surface vegetation type, generates a spatial vulnerability gradient map through the hierarchical analysis method, and superimposes it with the risk identification map to form a prevention and control priority matrix.

[0064] The embodiment of the present application automatically identifies the path of least resistance for pollution diffusion through energy field analysis of migration trajectories, guides the layout of engineering facilities such as cut-off walls and monitoring wells along the optimal control axis, and improves interception efficiency. The two-dimensional assessment system simultaneously considers the sensitivity of pollution sources and receptors, avoids excessive governance of high-risk and low-vulnerability areas in traditional methods, and achieves precise allocation of restoration resources. The three-level management and control system forms a red-orange-yellow early warning mechanism, supports the automatic triggering of differentiated emergency processes at different risk levels, and significantly shortens the emergency response decision chain.

[0065] Please refer to Figure 2 , Figure 2 A schematic diagram of a monitoring system for soil pollutants in a power transmission and transformation project provided in an embodiment of the present application is shown in FIG. Figure 2As shown, the monitoring system for soil pollutants in power transmission and transformation projects includes: an acquisition module for collecting hyperspectral image data within the power transmission and transformation project area; a transformation module for performing spectral transformation processing on the hyperspectral image data to obtain characteristic bands that characterize the pollutant content; a generation module for inputting the characteristic bands into a spectral inversion model to generate a concentration spatial distribution map of soil pollutants; an identification module for constructing a soil and groundwater diffusion model based on the concentration spatial distribution map; and a module for generating a migration path identification map of soil pollutants and an underground pollution risk area identification map based on the soil and groundwater diffusion model.

[0066] The functional modules in the embodiments of the present application form a loosely coupled architecture through standardized data interfaces. The acquisition module integrates hyperspectral imaging and BeiDou positioning technologies, the transformation module embeds an adaptive learning algorithm, the generation module is equipped with a hybrid-driven inversion engine, and the identification module runs a multi-media diffusion model, forming a fully automatic processing chain from data acquisition to decision output. A microservice-based computing resource allocation algorithm is developed to dynamically allocate GPU computing power to spectral transformation (high real-time requirements) or diffusion modeling (computationally intensive) based on task priority to achieve optimal utilization of hardware resources. A unified spatiotemporal database architecture is constructed, compatible with multi-source data streams such as drone aerial images, ground sensor networks, and meteorological monitoring stations, enabling near-real-time data preprocessing and feature extraction through edge computing nodes.

[0067] This application transforms laboratory-grade monitoring technology into engineering equipment deployable in harsh field environments through deep hardware-algorithm-model integration, resolving the traditional dilemma of achieving both portability and computational efficiency. This collaborative chain, from autonomous drone inspections to real-time edge processing and cloud-based intelligent decision-making, enables pollution event discovery, assessment, and early warning. The standardized data interface design is compatible with mainstream GIS platforms and power operation and maintenance systems, and output results can be directly imported into power transmission and transformation asset management systems to trigger maintenance work orders.

[0068] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0069] The soil pollutant monitoring system for power transmission and transformation projects in this embodiment is presented in the form of functional units, where the units refer to ASIC (Application Specific Integrated Circuit) circuits, processors and memories that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0070] See also Figure 3 , Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application is shown in FIG. Figure 3As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of a GUI on an external input / output device (such as, a display device coupled to an interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0071] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0072] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0073] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0074] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0075] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0076] The embodiments of the present application also provide a computer-readable storage medium. The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0077] An embodiment of the present application provides a computer program product, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform a method according to any embodiment of the present application.

[0078] The systems or modules described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0079] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0080] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0081] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems, and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0082] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device that implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0083] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0084] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of other identical elements in the process, method, commodity, or apparatus comprising the element.

[0085] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences from the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0086] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

[0087] Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. A method for monitoring soil pollutants in power transmission and transformation projects, characterized in that: The method comprises: Collect hyperspectral image data within the power transmission and transformation project area; The hyperspectral image data is processed using a multi-level spectral transformation mechanism to obtain characteristic bands representing pollutant content; Inputting the characteristic waveband into a preset spectral inversion model to generate a concentration spatial distribution map of soil pollutants; A soil and groundwater diffusion model is constructed according to the concentration spatial distribution map, and based on the soil and groundwater diffusion model, a soil pollutant migration path identification map and an underground pollution risk area identification map are generated.

2. The method according to claim 1, characterized in that The method further includes: collecting the hyperspectral image data through a plurality of monitoring points preset within the power transmission and transformation project area; wherein each monitoring point is provided with a differentiated marker pollutant, and each marker pollutant has a sensitive characteristic in a specific band; the sensitive characteristic is a regular change pattern of the spectral reflectance curve caused by an increase in pollutant concentration.

3. The method according to claim 1, characterized in that The method further comprises: Performing spatiotemporal alignment of the multi-temporal hyperspectral image data with the ground in-situ data to generate a pollutant baseline concentration matrix; the pollutant baseline concentration matrix includes four-tuple data of position, time, spectrum, and concentration; Based on the pollutant baseline concentration matrix and the groundwater diffusion model, a pollutant diffusion model is constructed, and based on the pollutant diffusion model, time series spectral features are extracted; the time series spectral features are used to characterize concentration changes and spectral dynamic changes of pollutants; According to the time series spectral characteristics and the preset concentration change rate threshold, the pollutant concentration enhanced areas are graded and calibrated.

4. The method according to claim 1, wherein The method further comprises: Processing the hyperspectral image data through a recursive feature elimination method to obtain characteristic bands representing pollutant content; Based on the characteristic band, the pollutant concentration level of each monitoring point is determined by adaptive threshold mapping to generate a pollutant distribution map, wherein the adaptive threshold mapping is used to associate the pollutant type and the characteristic band threshold, and the width of the characteristic band is used to characterize the pollutant concentration.

5. The method according to claim 4, characterized in that The method further comprises: Construct adversarial hyperspectral samples that simulate the interference environment of power transmission and transformation projects; Based on the adversarial hyperspectral sample, identifying abnormal disturbances in the adaptive threshold mapping process; The pollutant concentration in the characteristic band is corrected based on the abnormal disturbance.

6. The method according to claim 1, characterized in that Inputting the characteristic band into a spectral inversion model includes: Determining a dynamic sensitive band based on the pollutant absorption peak matched to the characteristic band, wherein the dynamic sensitive band is used to characterize a spectral band with significant sensitivity to changes in pollutant concentration or migration processes; The dynamic sensitive band is processed by the spectral inversion model to generate a concentration spatial distribution map of soil pollutants, which includes initial distribution data of soil pollutants and real-time distribution data of soil pollutants in a diffusion state with groundwater at any moment.

7. The method according to claim 1, characterized in that The concentration spatial distribution map includes a geographic elevation display map based on the power transmission and transformation project area; the geographic elevation display map is constructed based on the fusion of dynamic environmental parameters and ground in-situ monitoring data; based on the geographic elevation display map and the migration risk value, a migration path prediction map of soil pollutants is generated, and the migration path prediction map is used to characterize the migration trend of pollutants; A migration path identification map of soil pollutants is generated based on the migration path prediction map and the migration risk value.

8. The method according to claim 1, characterized in that According to the concentration spatial distribution map, constructing a soil and groundwater diffusion model includes: Loading the concentration spatial distribution map into a preset physical diffusion model to generate a pollutant diffusion matrix based on the three-dimensional migration data; the pollutant diffusion matrix includes a soil intrusive diffusion matrix and a groundwater high-speed dispersion diffusion matrix; Determining the spatiotemporal state characteristics of pollutants based on the pollutant diffusion matrix, wherein the spatiotemporal state characteristics are diffusion state data of pollutants in time and space dimensions; Based on the spatiotemporal state characteristics, a soil and groundwater diffusion model is constructed.

9. The method according to claim 1, characterized in that Based on the soil and groundwater diffusion model, a soil pollutant migration path identification map and an underground pollution risk area identification map are generated, including: generating three-dimensional migration data based on the soil and groundwater dispersion model; Determining diffusion weights of different migration diffusion coordinate points on the migration trajectory according to the three-dimensional migration data, and determining an emergency management and control risk value based on the diffusion weights; Generate a migration path identification map according to the emergency management and control risk value; The vulnerability of the power transmission and transformation project area is calculated according to the migration path identification map to obtain vulnerability values ​​of different areas, and an underground pollution risk area identification map is generated based on the vulnerability values ​​and the emergency control risk values.

10. A system for monitoring soil pollutants in power transmission and transformation projects, capable of implementing the method for monitoring soil pollutants in power transmission and transformation projects as claimed in any one of claims 1 to 9, characterized in that: The system comprises: The acquisition module is used to collect hyperspectral image data within the power transmission and transformation project area; A transformation module, configured to perform spectral transformation processing on the hyperspectral image data to obtain characteristic bands representing pollutant content; A generation module, configured to input the characteristic bands into a spectral inversion model to generate a concentration spatial distribution map of soil pollutants; An identification module is used to construct a soil and groundwater diffusion model based on the concentration spatial distribution map; and is used to generate a soil pollutant migration path identification map and an underground pollution risk area identification map based on the soil and groundwater diffusion model.

Images