Urban river and lake water pollution assessment method and system based on space-time evolution characteristics

Abstract

This invention relates to the field of water environment monitoring and assessment technology, specifically a method and system for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics. The method includes acquiring water quality concentrations, flow velocity vectors, and regional rainfall-runoff data at monitoring nodes; identifying pollution types using the response time curves of water quality parameters and rainfall-runoff, and the concentration ratios between nodes; subsequently constructing a spatial transport matrix based on the velocity vector and spatial gradient distribution of concentration to quantify the mass load of pollutants migrating with the water flow per unit time; tracking the trajectory of the load evolution based on the matrix and calculating the self-purification degradation rate during the transport process; finally, establishing a spatiotemporal diffusion model by combining the load, degradation rate, and trajectory to simulate the coverage area and peak concentration of the pollution diffusion cloud within a preset time period, generating assessment results. This invention can achieve spatiotemporal dynamic tracking and accurate prediction of urban water pollution, providing a scientific basis for environmental governance decisions.

Description

Technical Field

[0001] This invention relates to the field of water environment monitoring and assessment technology, and in particular to a method and system for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics. Background Technology

[0002] With the acceleration of urbanization, urban river and lake water pollution has become increasingly severe, posing a key constraint on the sustainable development of the urban ecological environment. Current water pollution assessment methods primarily rely on static analysis of fixed-point monitoring data. This involves deploying several water quality monitoring stations to periodically collect conventional indicators such as chemical oxygen demand (COD), ammonia nitrogen, and total phosphorus, and then comparing these data with standard thresholds to determine the degree of pollution. However, these methods only provide information on the pollution status at discrete points in time and space, failing to effectively characterize the dynamic migration paths and diffusion evolution of pollutants in water bodies, making it difficult to trace pollution sources and predict pollution development trends.

[0003] Furthermore, traditional assessment methods often rely on single-indicator thresholds or empirical model matching for pollution type identification, lacking a comprehensive consideration of rainfall-runoff response characteristics, pollutant component fingerprint information, and hydrodynamic conditions. This results in insufficient accuracy in distinguishing different pollution types, such as stormwater and sewage overflows, direct discharge from production and domestic facilities, and construction slurry injection, easily leading to misjudgments or omissions. In terms of pollution diffusion prediction, existing models often simplify the water transport process, neglecting the dynamic effects of convection-diffusion coupling and pollutant self-degradation. This causes significant deviations between the predicted results and the actual spatiotemporal distribution of pollution clouds, failing to provide accurate spatial positioning and temporal early warning for pollution emergency prevention and control.

[0004] Therefore, the main problems with existing technologies are: insufficient spatiotemporal resolution in pollution assessment, making it difficult to accurately track pollutant migration paths; a single pollution type identification mechanism, lacking comprehensive diagnostic capabilities through multi-source data fusion; and simplified pollution diffusion prediction models that fail to effectively couple hydrodynamic transport and biochemical degradation processes, resulting in assessment results that are difficult to meet the actual needs of refined management of urban river and lake water environments in terms of accuracy and predictability. Summary of the Invention

[0005] To address the aforementioned shortcomings, the present invention aims to propose a method and system for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics. This system aims to construct a multidimensional monitoring system integrating water quality concentration parameters, water flow velocity vectors, and rainfall-runoff data. By combining response time curve analysis and pollutant component fingerprinting technology, it achieves accurate identification of pollution types. Furthermore, it characterizes the mass load flux of pollutants migrating with water flow based on a spatial transport matrix. Finally, by tracking the evolution trajectory of pollutants and calculating self-purification degradation rates, a spatiotemporal diffusion model is established to simulate and predict the coverage area and concentration peak of future pollution diffusion clouds. This solves the technical problems of insufficient spatiotemporal resolution in pollution assessment, low accuracy in pollution type identification, and poor accuracy in diffusion prediction in existing technologies.

[0006] To achieve this objective, the present invention adopts the following technical solution: A method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics includes the following steps: S1: Obtain hydrological monitoring data for the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall and runoff data for several monitoring nodes. S2: Using the response time curves of the water quality concentration parameters and the regional rainfall-runoff data, combined with the pollutant concentration ratios between different monitoring nodes, the pollution type of the urban river and lake area to be monitored is identified; S3: Based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, construct a spatial transport matrix of water pollutants and quantify the mass load of pollutants migrating with the water flow per unit time. S4: Based on the spatial transport matrix, track the evolution trajectory of the pollutant mass load in the water area to determine the pollution destination, and calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. S5: Combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, establish a spatiotemporal diffusion model for pollutants, simulate the coverage range and peak concentration of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

[0007] Preferably, the pollution type in the urban river and lake area to be monitored includes stormwater and sewage overflow pollution; Step S2 includes: By analyzing the flow growth slope of the regional rainfall-runoff data, the initial rainfall period is identified, and the instantaneous increase of the water quality concentration parameter during the initial rainfall period is extracted to construct a response time curve reflecting the trend of pollutant concentration with cumulative runoff. Using the water quality concentration parameters of the aforementioned monitoring nodes, the cross-concentration ratios between chemical oxygen demand, total nitrogen, and total phosphorus are calculated to generate the current water body's pollution component fingerprint information. The pollution component fingerprint information is then subjected to normalized correlation analysis with the preset combined sewer overflow characteristic spectrum. Calculate the peak phase lag time of the response time curve. If the peak phase lag time is within the preset pipe network confluence time interval, and the correlation coefficient between the pollution component fingerprint information and the characteristic spectrum of the combined sewer overflow meets the preset intensity threshold, then the pollution type of the urban river and lake area to be monitored is identified as stormwater and sewage overflow pollution.

[0008] Preferably, the pollution types in the urban river and lake areas to be monitored include direct discharge pollution from production and daily life. Step S2 includes: Real-time access to regional rainfall and runoff data; when the regional rainfall and runoff data are consistently below a preset dry season threshold, extract peak data that deviates from the conventional benchmark characteristic value in the water quality concentration parameter, and calculate the rising slope and duration of the peak data to construct a response time curve for external interference during the rainless period. The pollutant concentrations at each monitoring node during the occurrence of the peak data are collected, and a real-time pollution fingerprint characterizing the current chemical characteristics of the water body is generated by calculating the instantaneous concentration ratios between ammonia nitrogen, total phosphorus, and chemical oxygen demand. The vector space distance between the real-time pollution fingerprint and the conventional benchmark feature value of the water body area to be monitored is calculated to obtain the feature offset that reflects the degree of structural variation of pollutant components. The real-time pollution fingerprint is input into the pollution source strong fingerprint database, and the feature source type with the highest matching degree is obtained by normalized correlation algorithm. The pollution source strong fingerprint database includes the feature spectrum of domestic sewage and the feature spectrum of industrial wastewater. The intensity of external disturbance is quantified based on the rising slope of the response time curve, and the pollutant component variation weight is determined according to the characteristic offset. Multidimensional coupling calculation is performed using the intensity of external disturbance and the component variation weight to obtain the direct emission confidence level characterizing the probability of non-natural emissions. By comparing the direct discharge confidence level with the preset judgment threshold, if the direct discharge confidence level exceeds the preset judgment threshold and the retrieval result of the feature source type matches the feature of organic wastewater, then the current pollution type is determined to be direct discharge pollution from production and domestic use.

[0009] Preferably, the pollution type in the urban river and lake area to be monitored includes construction mud injection pollution; Step S2 includes: By comparing the changing trends of the water flow velocity vector and the water quality concentration parameter, if it is found that there is no time correlation between the increase in concentration and the fluctuation of flow velocity, a response time curve reflecting the distribution of pollutants is independent of the dynamic disturbance of the flow field is constructed. The concentration of characteristic pollutants during the abnormal fluctuations of the response time curve is collected, and combined with the instantaneous deviation of the pH value of the water body at the monitoring node, a mineralization characteristic fingerprint characterizing the discharge of building materials is generated. The mineralization fingerprint is normalized and correlated with the preset construction mud feature spectrum to obtain the matching coefficient reflecting the degree of inorganic component abnormality, and the dynamic decoupling weight is determined according to the leading or lagging characteristics of concentration increase in the response time curve. The matching coefficient and the dynamic decoupling weight are used to perform multidimensional coupling calculation to obtain the mud injection confidence level. If the mud injection confidence level exceeds the preset mud judgment threshold, the current pollution type is determined to be construction mud injection type pollution.

[0010] Preferably, constructing a spatial transport matrix of water pollutants based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter includes: Based on the river morphology and monitoring node distribution of the water body area to be monitored, a gridded water flow topology map is established, and the water flow velocity vector at the center of each grid is projected onto the topology axis to calculate the unit flow exchange coefficient between adjacent grids. The differences in water quality concentration parameters between adjacent monitoring nodes are extracted, and combined with the physical distance between nodes, the concentration gradient vector of pollutants in the three-dimensional spatial coordinate system is calculated to characterize the spatial diffusion trend of pollutants. The unit flow exchange coefficient is coupled with the concentration gradient vector, and a convection-diffusion discrete equation describing the migration of pollutants with water flow is established based on the law of conservation of mass. The convection-diffusion discrete equations are transformed into matrix form, where the elements of the matrix represent the mass load flux of pollutants migrating from a certain grid point to an adjacent grid point per unit time, forming a spatial transport matrix that reflects the real-time spatial dynamic distribution of pollutants.

[0011] Preferably, tracking the evolution trajectory of the pollutant mass load in the water body region based on the spatial transport matrix includes: Based on the mass load flux at each grid point in the spatial transport matrix, the net migration rate of pollutants between adjacent grids is calculated to obtain the spatial evolution intensity. Satisfying the relation: ; in, Indicates from grid point Point to adjacent grid points Spatial evolution intensity, Represents the grid points in the spatial transport matrix and The flow exchange coefficient between them Represents grid points Real-time concentration of pollutants at the location This indicates the preset turbulent diffusion tensor in directional components, Represents grid points and The spatial gradient vector of pollutants between them; By performing spatiotemporal integration of the spatial evolution intensity and the water flow velocity vector, the displacement vector of the pollutant agglomerate within a continuous time step is determined, thus obtaining the pollutant evolution trajectory coordinates. Satisfying the relation: ; in, Indicates in The coordinates of the pollutant evolution trajectory at any given moment. express The trajectory coordinates at any given moment This represents the effective migration velocity vector after being corrected for the spatial evolution intensity. The effective migration velocity vector is formed by superimposing the water flow velocity vector and the convection-diffusion component derived from the spatial transport matrix. Indicates the time variable of integration; Based on the hydrological environmental factors corresponding to the pollutant evolution trajectory coordinates, the biochemical reaction losses of pollutants along the migration path are verified, and trajectory correction coefficients are obtained. Satisfying the relation: ; in, Indicates the trajectory correction coefficient. This represents the evolution path space curve formed by the trajectory coordinates. Indicates temperature and dissolved oxygen Constrained pollutant self-cleaning degradation function, Elements representing the arc length along the path curve. This represents the magnitude of the effective migration velocity vector; By mapping and coupling the pollutant evolution trajectory coordinates with the trajectory correction coefficient, the entire pollution evolution trajectory, including mass decay characteristics, is output, yielding the final evolution trajectory tracking result. The following relation is satisfied: ; in, This indicates the final evolution trajectory tracking result. This represents the total mass load of pollutants released by the pollution source at the initial moment. This represents the cumulative trajectory correction coefficient as it evolves over time.

[0012] Preferably, the self-degradation rate of pollutants during transport is calculated based on the time history of concentration decay between adjacent monitoring nodes: Extract the concentration observation values ​​of adjacent monitoring nodes along the evolution trajectory at corresponding times, and calculate the total attenuation coefficient of pollutants during water transport. The following relation is satisfied: ; in, This represents the total attenuation coefficient of pollutants between adjacent monitoring nodes. This represents the observed pollutant concentration at the upstream monitoring node of the evolution trajectory. This indicates that the downstream monitoring nodes have undergone evolution time. Subsequent pollutant concentration observations This indicates the time interval between adjacent monitoring nodes as pollutant clumps migrate with the water flow; The spatial transport matrix is ​​used to quantify the contribution of concentration reduction due to physical diffusion between adjacent monitoring nodes, thus obtaining the physical dilution ratio. The following relation is satisfied: ; in, This indicates the percentage of concentration decrease caused by physical diffusion. This represents the longitudinal dispersion coefficient of the river channel obtained by mapping from the spatial transport matrix. This represents the second-order spatial gradient of concentration along the evolution trajectory. After removing the effect of physical dilution from the total attenuation coefficient, the net attenuation rate caused by biochemical degradation is extracted to obtain the self-cleaning degradation rate. The following relation is satisfied: ; in, This indicates the self-cleaning degradation rate of pollutants under the current water temperature environment. This represents the preset temperature correction factor. This indicates the water temperature parameters obtained in real time through the monitoring node. This indicates the preset biochemical degradation baseline reference temperature; Verify the nonlinear mapping relationship between the self-cleaning degradation rate and the dissolved oxygen concentration in the water, and output the corrected final self-cleaning degradation rate. The following relation is satisfied: ; in, This indicates the corrected final self-cleaning degradation rate. This indicates the real-time dissolved oxygen concentration in the water body obtained through the monitoring node. This represents the self-cleaning reaction half-saturation constant corresponding to the pollution type.

[0013] Preferably, by combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, a spatiotemporal diffusion model for pollutants is established to simulate the coverage area and peak concentration of the pollution diffusion cloud within a preset time period, including: Based on the migration path determined by the evolution trajectory, calculate the future... The spatial location of the center point of the pollution cloud at any given time is used to calculate the total amount of residual pollutants, combined with the final self-cleaning degradation rate. Satisfying the relation: ; in, Indicates the future The total mass of pollutants remaining in the pollution cloud at all times. This represents the total mass load of pollutants released by the pollution source at the initial moment. This indicates the corrected final self-cleaning degradation rate. This indicates the preset simulation deadline. Indicates the starting point of the pollution event; A two-dimensional Gaussian diffusion model was used to characterize the mass distribution of the pollution cloud on the water surface, and the predicted pollutant concentrations at each coordinate point were calculated. Satisfying the relation: ; in, Indicates in Time coordinates are Predicted pollutant concentration values ​​at the location This indicates that the evolutionary trajectory is determined by the evolutionary trajectory. The center coordinates of the cloud cluster at any given time and These represent the discrete coefficients along the water flow direction and perpendicular to the water flow direction, respectively, obtained by mapping from the spatial transport matrix; Based on the spatial gradient evolution of the predicted pollutant concentrations, a set of boundaries satisfying a preset concentration threshold is extracted to define the coverage area of ​​the pollution diffusion cloud, and the concentration maxima within the coverage area are searched to obtain the concentration peak. The following relation is satisfied: ; in, Indicates the predicted peak concentration. Indicates that by satisfying The coverage area of ​​the pollution diffusion cloud formed by the constrained coordinate points. This indicates the preset pollution determination boundary concentration threshold.

[0014] Preferably, the water pollution prediction and assessment results include: The evolution trajectory is overlaid with a geographic information layer to generate a dynamic distribution map that includes the extent of polluted water areas, overlapping locations of ecologically sensitive areas, and the degree of spatial damage. Statistical analysis of the spatiotemporal evolution data of pollutant mass load in water areas yields structured charts reflecting concentration evolution trends, migration of areas exceeding standards, and pollution reduction processes. Based on the evolution trajectory and the time of reaching the preset section, a digital early warning dashboard is obtained, including a risk countdown, pollution transit duration, and expected recovery period. By integrating the dynamic distribution map, the structured chart, and the digital early warning dashboard, the water pollution prediction and assessment results are obtained.

[0015] A city river and lake water pollution assessment system based on spatiotemporal evolution characteristics includes: The data acquisition module is used to acquire hydrological monitoring data of the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall runoff data of several monitoring nodes. The pollution type identification module is used to identify the pollution type of the urban river and lake area to be monitored by using the response time curve of the water quality concentration parameter and the regional rainfall runoff data, combined with the pollutant concentration ratio between different monitoring nodes. The spatial transport construction module is used to construct a spatial transport matrix of water pollutants based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, and to quantify the mass load of pollutants migrating with the water flow per unit time. The trajectory tracking and degradation calculation module is used to track the evolution trajectory of the pollutant mass load in the water area based on the spatial transport matrix to determine the pollution destination, and to calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. The prediction and assessment module is used to combine the pollutant mass load, the self-cleaning degradation rate and the evolution trajectory to establish a pollutant spatiotemporal diffusion model, simulate the coverage range and concentration peak of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

[0016] One of the above technical solutions has the following advantages or beneficial effects: This invention establishes a multi-dimensional information foundation covering pollutant chemical characteristics, hydrodynamic conditions, and meteorological driving factors by acquiring hydrological monitoring data including water quality concentration parameters, water flow velocity vectors, and regional rainfall-runoff data, thus overcoming the information limitations of traditional single-indicator water quality monitoring. Secondly, by utilizing the response time curves of water quality concentration parameters and rainfall-runoff data, combined with the pollutant concentration ratios between different monitoring nodes, pollution type identification is performed. This allows different pollution sources, such as stormwater overflow, direct discharge from production and domestic facilities, and construction slurry injection, to be effectively distinguished through their unique response characteristics and component fingerprints, significantly improving the accuracy of pollution source tracing. Finally, a spatial transport matrix of water pollutants is constructed based on the spatial gradient distribution of water flow velocity vectors and water quality concentration parameters, transforming the originally discrete point-like monitoring data... This process transforms pollutants into a continuous field quantity reflecting the spatial dynamic distribution of pollutants, enabling a quantitative characterization of the pollutant mass load migrating with the water flow per unit time. Based on the spatial transport matrix, the evolution trajectory of the pollutant mass load is tracked to determine the pollution destination. The self-purification degradation rate of pollutants is calculated based on the time history of concentration decay between adjacent monitoring nodes, thus organically coupling the hydrodynamic transport process with the biochemical degradation process, overcoming the shortcomings of the traditional model that treats the two separately. Finally, by integrating the pollutant mass load, self-purification degradation rate, and evolution trajectory, a spatiotemporal diffusion model of pollutants is established to simulate the coverage range and concentration peak of the pollution diffusion cloud within a preset time period, generating water pollution prediction and assessment results. This allows the assessment results to not only reflect the current pollution status but also predict the spatiotemporal evolution trend of pollution. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 This is a flowchart of the urban river and lake water pollution assessment method based on spatiotemporal evolution characteristics provided in this embodiment of the invention; Figure 2 This is a schematic diagram of the urban river and lake water pollution assessment system based on spatiotemporal evolution characteristics provided in an embodiment of the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0020] In this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0021] Urban river and lake water pollution assessment methods based on spatiotemporal evolution characteristics, such as Figure 1 As shown, a preferred embodiment of the present invention includes the following steps: S1: Obtain hydrological monitoring data for the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall and runoff data for several monitoring nodes. It should be noted that monitoring nodes refer to automatic water quality monitoring stations deployed within the river and lake areas of the city to be monitored. Each node can be equipped with a multi-parameter water quality analyzer, an acoustic Doppler current profiler, and rainfall data acquisition equipment for real-time collection of aquatic environmental information. Water quality concentration parameters refer to chemical index data characterizing the degree of water pollution obtained through water quality sensors, including but not limited to chemical oxygen demand (COD, reflecting the degree of organic pollution in the water), ammonia nitrogen (reflecting the degree of nitrogen-containing organic pollution and eutrophication), total phosphorus (reflecting the risk of eutrophication), and total nitrogen. These parameters are used to characterize the chemical pollution state of the water body. Water flow velocity vectors refer to the velocity vectors of water flow with direction and magnitude obtained through velocity measurement equipment, including longitudinal, lateral, and vertical velocity components along the river channel, used to describe the dynamic conditions of pollutant migration with the water flow. Regional rainfall-runoff data refer to the rainfall intensity, cumulative rainfall, and resulting surface runoff data of the monitoring area obtained through rain gauges and watershed hydrological models, used to characterize the driving effect of rainfall events on pollutant flushing into rivers. Hydrological monitoring data refers to the collection of the above three types of data, which together constitute the multidimensional information foundation for pollution assessment.

[0022] Understandably, by deploying multiple monitoring nodes at key sections and sensitive areas of urban rivers and lakes, a spatial gridded monitoring network is formed, overcoming the spatial limitations of traditional single-point monitoring. Real-time acquisition of water quality concentration parameters provides information on the chemical composition of pollutants; measuring water flow velocity vectors reveals the dynamic conditions of pollutant transport; and acquiring regional rainfall and runoff data identifies rainfall-driven pollution events. The simultaneous acquisition and fusion of these multi-source data provides comprehensive data support for subsequent pollution type identification, spatial transport modeling, and diffusion prediction. This allows the assessment process to shift from static section concentration comparisons to dynamic spatiotemporal evolution analysis, significantly improving the information completeness and spatiotemporal coverage of pollution assessments.

[0023] S2: Using the response time curves of the water quality concentration parameters and the regional rainfall-runoff data, combined with the pollutant concentration ratios between different monitoring nodes, the pollution type of the urban river and lake area to be monitored is identified; It should be noted that the response time curve is a curve plotted with time on the horizontal axis and water quality concentration parameters on the vertical axis. It characterizes the dynamic response relationship of pollutant concentration over time (especially with rainfall and runoff processes), reflecting the occurrence, development, and dissipation of pollution events. The pollutant concentration ratio between different monitoring nodes refers to the ratio between the concentrations of various pollutants monitored at multiple spatial locations, such as the ratio of chemical oxygen demand (COD) to total phosphorus, or the ratio of ammonia nitrogen to total nitrogen. These ratios constitute the fingerprint information of pollutant components, used to distinguish the chemical characteristics of different pollution sources. Pollution type refers to the classification of water pollution based on its causes and source characteristics. It mainly includes stormwater overflow pollution (caused by rainwater runoff from combined sewer systems), direct discharge pollution (caused by illegal discharge of domestic sewage or industrial wastewater during the dry season), and construction slurry injection pollution (caused by the illegal discharge of construction slurry from construction sites). The purpose of identifying pollution types is to determine the causes of pollution and provide a basis for subsequent targeted control measures.

[0024] Understandably, by constructing response-time curves between water quality concentration parameters and rainfall runoff data, the temporal correlation between pollutant concentration changes and rainfall processes can be analyzed: if the concentration peak lags behind the rainfall peak and conforms to the characteristics of pipe network runoff time, it indicates stormwater overflow pollution; if the concentration anomaly occurs during a rainless period, it indicates direct discharge pollution from production and domestic sources. By calculating the proportion of pollutant concentrations between different monitoring nodes, unique chemical fingerprint information for various pollution events can be extracted: stormwater overflow typically shows a simultaneous increase in chemical oxygen demand (COD), total nitrogen (TOD), and total phosphorus (TP), with relatively stable proportions; direct discharge of domestic sewage is characterized by high ammonia nitrogen, high COD, and characteristic detergent components; direct discharge of industrial wastewater is characterized by abnormal proportions of industry-specific pollutants (such as heavy metals and organic toxins); and construction slurry is characterized by high turbidity, high suspended solids, and alkaline pH. Combining the temporal characteristics of the response-time curve with the spatial distribution characteristics of pollutant component fingerprints allows for comprehensive identification of multiple pollution types, overcoming the limitations of traditional single-indicator threshold determination methods and significantly improving the accuracy and reliability of pollution source tracing.

[0025] S3: Based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, construct a spatial transport matrix of water pollutants and quantify the mass load of pollutants migrating with the water flow per unit time. It should be noted that the spatial gradient distribution refers to the rate of change of water quality concentration parameters with position in a three-dimensional spatial coordinate system. It is obtained by calculating the ratio of concentration difference to distance between adjacent monitoring nodes and is used to characterize the spatial diffusion trend and direction of pollutants. The spatial transport matrix is ​​a mathematical matrix that discretizes the convection-diffusion process into mass exchange relationships between grids. The rows and columns of the matrix correspond to spatially discrete grid points, and the matrix elements represent the mass load flux of pollutants migrating from a grid point to an adjacent grid point per unit time, used to quantify the spatiotemporal dynamic process of pollutant migration with water flow. The mass load refers to the total mass of pollutants passing through a certain cross-section per unit time, obtained by integrating the product of water flow rate and pollutant concentration, and is used to characterize the intensity of pollutant transport. The purpose of constructing the spatial transport matrix is ​​to transform the continuous hydrodynamic-water quality coupling process into a computable discrete mathematical model, achieving accurate characterization of pollutant migration paths.

[0026] Understandably, the water flow velocity vector determines the direction and rate of pollutant migration via convection, while the spatial gradient distribution of water concentration parameters determines the trend and intensity of pollutant diffusion from high-concentration areas to low-concentration areas. By discretizing the monitored water body into grid cells, the water flow velocity vector is projected onto the topological axis at the center of each grid to obtain the convective transport component. The concentration gradient between adjacent grids is calculated to obtain the diffusion transport component. Then, based on the law of conservation of mass, a convective-diffusion discrete equation describing the migration of pollutants with the water flow is established, ultimately transformed into a matrix form. The spatial transport matrix reflects the migration flux of pollutants between points in space in real time, expanding the originally discrete point monitoring data into a continuous field distribution, thereby enabling precise quantification of the mass load of pollutants migrating with the water flow per unit time.

[0027] S4: Based on the spatial transport matrix, track the evolution trajectory of the pollutant mass load in the water area to determine the pollution destination, and calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. It should be noted that the evolution trajectory refers to the spatial path curve formed by pollutant clumps migrating with water flow, consisting of a series of coordinate points that change over time, used to determine the destination of pollutants and the time it takes for them to reach various sensitive areas. Pollution destination refers to the final migration direction and distribution location of pollutants in the water body, including downstream transport, retention in local areas, or diffusion to tributaries. The time history refers to the record of the concentration change of pollutants over time during their migration between adjacent monitoring nodes, used to analyze concentration decay patterns. Self-purification degradation rate refers to the rate at which the concentration of pollutants decreases in the water body through physical, chemical, and biological processes (such as sedimentation, oxidation, and microbial decomposition), reflecting the water body's natural purification capacity. The purpose of tracking the evolution trajectory and calculating the self-purification degradation rate is to achieve dynamic tracking of pollutant migration paths and quantitative assessment of natural decay capacity.

[0028] Understandably, based on the mass load flux of each grid point in the spatial transport matrix, the net migration rate (spatial evolution intensity) of pollutants between adjacent grids can be calculated. Then, by combining this with the water flow velocity vector for spatiotemporal integration, the displacement vector of pollutant clumps within a continuous time step can be determined, yielding the pollutant evolution trajectory coordinates and thus accurately characterizing the pollutant migration path. Simultaneously, by extracting the concentration observations of adjacent monitoring nodes along the evolution trajectory at corresponding times, the total attenuation coefficient of pollutants during water transport can be calculated. The physical dilution effect (the contribution of concentration reduction due to physical diffusion quantified through the spatial transport matrix) is then removed from the total attenuation coefficient, and the net attenuation rate caused by biochemical degradation is extracted to obtain the self-cleaning degradation rate. Combining the tracking of the evolution trajectory with the calculation of the self-cleaning degradation rate allows for the simultaneous characterization of the pollutant migration and attenuation processes, overcoming the limitations of traditional models that treat transport and degradation separately.

[0029] S5: Combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, establish a spatiotemporal diffusion model for pollutants, simulate the coverage range and peak concentration of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

[0030] It should be noted that the pollutant spatiotemporal diffusion model is a mathematical model that integrates time and spatial dimensions to describe the diffusion process of pollutants in water bodies over time, comprehensively considering physicochemical processes such as convective transport, turbulent diffusion, and biochemical degradation. A pollution diffusion cloud refers to a three-dimensional water body region with a specific concentration distribution formed by the migration and diffusion of pollutants, its boundary defined by a preset concentration threshold. The coverage area refers to the spatial region occupied by the projection of the pollution diffusion cloud onto a horizontal plane, characterizing the spatial scale of the pollution impact. The concentration peak refers to the maximum concentration of pollutants within the pollution diffusion cloud and its location, characterizing the intensity center of the pollution impact. The water pollution prediction and assessment results refer to a comprehensive judgment on the future development trend of pollution obtained through model simulation, including information such as the spatiotemporal distribution of the pollution diffusion cloud, the time of arrival at sensitive areas, and the risk level, used to support pollution emergency prevention and control decisions.

[0031] Understandably, by combining the pollutant mass load quantified in step S3 (characterizing the intensity of pollution input), the self-cleaning degradation rate calculated in step S4 (characterizing the natural deceleration rate of pollution), and the determined evolution trajectory (characterizing the pollution migration path), a complete spatiotemporal diffusion model of pollutants can be established. Based on the migration path determined by the evolution trajectory, the spatial location of the pollution cloud center point at future moments can be predicted; combined with the self-cleaning degradation rate, the total amount of pollutants remaining during the migration process can be calculated; using a two-dimensional or three-dimensional diffusion model, the mass distribution of the pollution cloud in space can be characterized; by extracting the boundary set that meets the concentration threshold, the coverage area of ​​the pollution diffusion cloud can be defined; and by searching for the concentration maxima, the concentration peak can be determined. The integrated simulation of the above processes can generate predictive assessment results about the pollution development trend within a preset time period, extending the assessment from judging the current state to predicting future evolution trends.

[0032] Preferably, the pollution type in the urban river and lake area to be monitored includes stormwater and sewage overflow pollution; Step S2 includes: By analyzing the flow growth slope of the regional rainfall-runoff data, the initial rainfall period is identified, and the instantaneous increase of the water quality concentration parameter during the initial rainfall period is extracted to construct a response time curve reflecting the trend of pollutant concentration with cumulative runoff. Using the water quality concentration parameters of the aforementioned monitoring nodes, the cross-concentration ratios between chemical oxygen demand, total nitrogen, and total phosphorus are calculated to generate the current water body's pollution component fingerprint information. The pollution component fingerprint information is then subjected to normalized correlation analysis with the preset combined sewer overflow characteristic spectrum. Calculate the peak phase lag time of the response time curve. If the peak phase lag time is within the preset pipe network confluence time interval, and the correlation coefficient between the pollution component fingerprint information and the characteristic spectrum of the combined sewer overflow meets the preset intensity threshold, then the pollution type of the urban river and lake area to be monitored is identified as stormwater and sewage overflow pollution.

[0033] It should be noted that the flow growth slope refers to the rate of change of runoff over time in regional rainfall-runoff data. It is obtained by calculating the increment of runoff per unit time and is used to identify the initial stage and intensity variation characteristics of rainfall events. The initial rainfall period refers to the stage from the start of rainfall until the runoff reaches its peak. During this period, deposited pollutants in combined sewer systems are flushed into rivers by rainwater, making it a key identification window for stormwater overflow pollution. Instantaneous increase refers to the rapid rise in water quality concentration parameters over a short period, reflecting the intensity of pollutants carried into rivers by runoff. Cumulative runoff refers to the total runoff from the start of rainfall to a certain point in time, used to characterize the cumulative effect of rainfall flushing. Cross-concentration ratio refers to the ratio between the concentrations of three pollutants: chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP), such as the ratio of COD to TN, and the ratio of TN to TP. These ratios constitute the core content of the pollution component fingerprint information. Pollution component fingerprint information is a feature vector composed of the concentration ratios of multiple pollutants, used to characterize the chemical composition characteristics of a specific pollution source. The characteristic spectrum of combined sewer overflow refers to the standardized chemical composition pattern of pollutants contained in sewage overflowing from a combined sewer system during rainfall. It typically manifests as a specific ratio and concentration range of chemical oxygen demand (COD), total nitrogen (TNO), and total phosphorus (TP). Normalized correlation analysis (NCR) is a statistical method that calculates the similarity between the fingerprint information of the pollutant components to be analyzed and the standard characteristic spectrum after standardization. Cosine similarity or Pearson correlation coefficient are commonly used as measures. Peak phase lag time refers to the time delay between the peak water quality concentration and the peak rainfall runoff, reflecting the transport time of pollutants from the sewer network to the river channel. The sewer network confluence time interval refers to the reasonable time range within which pollutants are transported from the discharge point to the monitoring section, estimated based on sewer network length, slope, and rainfall intensity. The preset intensity threshold is the critical correlation coefficient value for determining the degree of pollution type matching; exceeding this threshold indicates a successful match.

[0034] Understandably, analyzing the flow growth slope of regional rainfall-runoff data to identify the initial rainfall period allows for the precise capture of the critical time window for combined sewer overflow sediments to be washed into rivers. By extracting the instantaneous increase in water quality concentration parameters during this period and constructing response-time curves, the dynamic response relationship of pollutants with cumulative runoff can be quantified, revealing the typical "flushing-peak-regression" process characteristics of stormwater overflows. Calculating the cross-concentration ratios of chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP) to generate pollutant fingerprint information, and performing normalized correlation analysis with the characteristic spectrum of combined sewer overflows, allows for verification of pollution sources from a chemical composition perspective. Calculating the peak phase lag duration of the response-time curve and comparing it with the confluence time interval of the sewer network allows for verification of pollution transport paths from a temporal response perspective. When both temporal and chemical characteristics simultaneously meet the judgment criteria, stormwater overflow pollution can be accurately identified, enabling precise source tracing of rainfall-driven pollution events, avoiding misjudgments caused by single-indicator determinations, and significantly improving the accuracy and reliability of pollution type identification in complex urban water environments.

[0035] For example, in the monitoring of a combined sewer system area, the system detected a rapid increase in runoff after the start of rainfall, with the flow rate increasing at a slope of [missing information]. The system identifies the initial rainfall period. Water quality concentration parameters during this period are extracted: Chemical Oxygen Demand (COD) rapidly increases from 25 mg / L to 78 mg / L, Total Nitrogen (TN) from 3.2 mg / L to 9.5 mg / L, and Total Phosphorus (TP) from 0.4 mg / L to 1.8 mg / L. Response time curves show an exponential increase in concentration with cumulative runoff. The cross-concentration ratios of COD, TN, and TP are calculated: COD / TN = 8.2, TN / TP = 5.3, generating a pollutant fingerprint. This fingerprint is then subjected to normalized correlation analysis with the pre-defined combined sewer overflow characteristic spectrum (COD / TN standard ratio of 7-9, TN / TP standard ratio of 4-6), yielding a cosine similarity of 0.94, exceeding the pre-defined intensity threshold of 0.85. Simultaneously, the peak phase lag of the response time curve is calculated to be 42 minutes, falling within the pre-defined confluence time range of 30-60 minutes for this watershed. Based on a comprehensive assessment of temporal and chemical characteristics, the system accurately identified the pollution type as stormwater overflow pollution.

[0036] Preferably, the pollution types in the urban river and lake areas to be monitored include direct discharge pollution from production and daily life. Step S2 includes: Real-time access to regional rainfall and runoff data; when the regional rainfall and runoff data are consistently below a preset dry season threshold, extract peak data that deviates from the conventional benchmark characteristic value in the water quality concentration parameter, and calculate the rising slope and duration of the peak data to construct a response time curve for external interference during the rainless period. The pollutant concentrations at each monitoring node during the occurrence of the peak data are collected, and a real-time pollution fingerprint characterizing the current chemical characteristics of the water body is generated by calculating the instantaneous concentration ratios between ammonia nitrogen, total phosphorus, and chemical oxygen demand. The vector space distance between the real-time pollution fingerprint and the conventional benchmark feature value of the water body area to be monitored is calculated to obtain the feature offset that reflects the degree of structural variation of pollutant components. The real-time pollution fingerprint is input into the pollution source strong fingerprint database, and the feature source type with the highest matching degree is obtained by normalized correlation algorithm. The pollution source strong fingerprint database includes the feature spectrum of domestic sewage and the feature spectrum of industrial wastewater. The intensity of external disturbance is quantified based on the rising slope of the response time curve, and the pollutant component variation weight is determined according to the characteristic offset. Multidimensional coupling calculation is performed using the intensity of external disturbance and the component variation weight to obtain the direct emission confidence level characterizing the probability of non-natural emissions. By comparing the direct discharge confidence level with the preset judgment threshold, if the direct discharge confidence level exceeds the preset judgment threshold and the retrieval result of the feature source type matches the feature of organic wastewater, then the current pollution type is determined to be direct discharge pollution from production and domestic use.

[0037] It should be noted that the preset dry season threshold refers to the critical value for runoff in the absence of rainfall. It is typically set according to the characteristics of the watershed, corresponding to the runoff level of light rainfall. Runoff below this threshold is considered to be in a drought period, and pollution events occurring under these conditions exclude rainfall-induced runoff. The conventional baseline characteristic value refers to the statistical characteristics of background water quality parameters in the monitored water body area under stable hydrodynamic conditions and without pollution events. This includes the mean, variance, and normal fluctuation range of each pollutant concentration, used to identify abnormal deviations. Peak data refers to local maxima in water quality concentration parameters that are significantly higher than the conventional baseline characteristic value, reflecting a sudden input of exogenous pollutants. The rise slope refers to the rate of concentration change during the rise of the peak data, reflecting the intensity and suddenness of the pollution input. Duration refers to the duration from the initial rise of the peak data to its return to the background level, reflecting the persistence of the pollution event. The real-time pollution fingerprint is a feature vector composed of the instantaneous concentration ratios of ammonia nitrogen, total phosphorus, and chemical oxygen demand, used to characterize the current chemical composition of the water body. Vector space distance refers to the geometric distance between the real-time pollution fingerprint vector and the conventional baseline feature value vector in a high-dimensional feature space. It is commonly measured by Euclidean or Mahalanobis distance; a larger distance indicates a higher degree of component structural variation. Feature offset is the quantified result of the vector space distance, reflecting the degree of deviation of pollutant components from the background state. The pollution source strength fingerprint database is a database containing chemical feature spectra of various known pollution sources, including domestic sewage feature spectra (high ammonia nitrogen, high chemical oxygen demand, specific detergent component proportions) and industrial wastewater feature spectra (industry-specific pollutant proportions). The normalized correlation algorithm is an algorithm that calculates the similarity between the fingerprint to be retrieved and each feature spectrum in the database after standardization, used to determine the most matching pollution source type. External disturbance intensity refers to the pollution input intensity index quantified based on the rising slope of the response time curve. Component variation weight refers to the weight of the degree of pollutant composition anomaly determined based on feature offset. Direct discharge confidence is a comprehensive index characterizing the probability that a pollution event is caused by non-natural direct discharge, obtained through multi-dimensional coupling calculation.

[0038] Understandably, by monitoring regional rainfall-runoff data consistently below the preset dry season threshold, interference from rainfall factors is first eliminated, focusing on abnormal pollution events during the rainless period. By extracting peak data deviating from conventional baseline feature values ​​and calculating their upward slope and duration, the sudden input characteristics of exogenous pollutants can be identified, and a response time curve for exogenous interference during the rainless period can be constructed. By calculating the instantaneous concentration ratios of ammonia nitrogen, total phosphorus, and chemical oxygen demand to generate a real-time pollution fingerprint, and calculating the feature offset by vector space distance with conventional baseline feature values, the degree of anomaly of pollutant components can be quantified from the perspective of chemical composition. By inputting the real-time pollution fingerprint into a pollution source strong fingerprint database for normalized correlation retrieval, the most likely pollution source type can be determined. By multidimensionally coupling the intensity of external disturbances with the weight of component variation, the confidence level of direct discharge is obtained. The judgment is made by combining time-series characteristics and chemical characteristics. When the confidence level of direct discharge exceeds the preset judgment threshold and the characteristic source type matches the characteristics of organic wastewater, the pollution of direct discharge from production and life can be accurately identified. This enables the precise capture of illegal sewage discharge events during the dry season, avoids misjudgment caused by natural fluctuations, and significantly improves the pertinence and effectiveness of pollution supervision during the rainless period.

[0039] For example, in the monitoring of a river downstream of an industrial zone, real-time monitoring showed that the regional rainfall-runoff data remained below the preset dry season threshold of 2 mm / h for 48 consecutive hours, indicating the start of a drought period. At this time, abnormal peaks appeared in the water quality concentration parameters at the monitoring nodes: Chemical Oxygen Demand (COD) rapidly increased from the background value of 22 mg / L to 65 mg / L, with an upward slope of 8.6 mg / (L·h), lasting approximately 6 hours; Ammonia Nitrogen (ANOD) increased from 0.8 mg / L to 4.5 mg / L; and Total Phosphorus (TP) increased from 0.3 mg / L to 1.2 mg / L. The response time curve of external disturbances during the drought period showed a typical "steep rise-plateau-slow decline" pattern. The instantaneous concentration ratios of ANOD, T, and COD were calculated: ANOD / COD = 0.069, T Phosphorus / COD = 0.018, generating a real-time pollution fingerprint. The fingerprint was compared with conventional baseline feature values ​​(ammonia nitrogen / COD background ratio approximately 0.036, total phosphorus / COD background ratio approximately 0.014) using vector space distance calculation. The Euclidean distance was 0.089, indicating a significant feature offset. The real-time pollution fingerprint was input into the pollution source intensity fingerprint database for retrieval. The normalized correlation coefficient with the feature spectrum of domestic sewage (ammonia nitrogen / COD standard ratio 0.06-0.08, total phosphorus / COD standard ratio 0.015-0.025) was 0.91, and the correlation coefficient with the feature spectrum of industrial wastewater was 0.34, indicating a match with domestic sewage features. Based on the rising slope... The quantified external disturbance intensity is 0.78, the component variation weight determined based on the feature offset is 0.82, and the confidence level of direct discharge calculated by multidimensional coupling is 0.80, exceeding the preset judgment threshold of 0.70. Based on these findings, the pollution type is determined to be direct discharge pollution from production and daily life.

[0040] Preferably, the pollution type in the urban river and lake area to be monitored includes construction mud injection pollution; Step S2 includes: By comparing the changing trends of the water flow velocity vector and the water quality concentration parameter, if it is found that there is no time correlation between the increase in concentration and the fluctuation of flow velocity, a response time curve reflecting the distribution of pollutants is independent of the dynamic disturbance of the flow field is constructed. The concentration of characteristic pollutants during the abnormal fluctuations of the response time curve is collected, and combined with the instantaneous deviation of the pH value of the water body at the monitoring node, a mineralization characteristic fingerprint characterizing the discharge of building materials is generated. The mineralization fingerprint is normalized and correlated with the preset construction mud feature spectrum to obtain the matching coefficient reflecting the degree of inorganic component abnormality, and the dynamic decoupling weight is determined according to the leading or lagging characteristics of concentration increase in the response time curve. The matching coefficient and the dynamic decoupling weight are used to perform multidimensional coupling calculation to obtain the mud injection confidence level. If the mud injection confidence level exceeds the preset mud judgment threshold, the current pollution type is determined to be construction mud injection type pollution.

[0041] It should be noted that time correlation refers to the degree of association between the changing trends of two time-series variables (such as water flow velocity and pollutant concentration). It is usually measured through cross-correlation analysis or time-delay correlation analysis. The presence of time correlation indicates that concentration changes are driven by hydrodynamic forces, while the absence of time correlation suggests that concentration changes are caused by non-hydrodynamic factors. Dynamic decoupling refers to the phenomenon of pollutant concentration changes being decoupled from fluctuations in water flow velocity, reflecting that pollutant input is independent of natural hydrodynamic processes, suggesting possible anthropogenic, targeted discharges. Characteristic pollutant concentration refers to the concentration of a marker pollutant that can characterize a specific pollution source. For construction mud, this typically includes high concentrations of suspended solids, high turbidity, and particulate matter with a specific particle size distribution. Instantaneous pH deviation refers to the sudden change in the acidity or alkalinity of the water body relative to the background state. Construction mud often exhibits an alkaline deviation due to the leaching of building materials such as cement and lime. Mineralization fingerprint refers to a chemical characteristic vector characterizing the discharge of inorganic building materials, composed of characteristic pollutant concentrations and pH values. Construction mud characteristic profile refers to the standardized chemical composition pattern of construction site mud wastewater, typically characterized by high suspended solids, high turbidity, alkaline pH, and specific calcium and magnesium ion ratios. The matching coefficient is the similarity value obtained from normalized correlation analysis between the mineralization characteristic fingerprint and the construction mud characteristic profile. Leading or lagging characteristics refer to the temporal position of concentration increase relative to flow velocity change in the response time curve; leading indicates that the concentration change precedes the flow velocity change, while lagging indicates that it follows the flow velocity change; both indicate dynamic decoupling. The dynamic decoupling weight is a weighting coefficient quantified based on the leading or lagging characteristics, characterizing the degree to which the concentration change is independent of the hydrodynamic driving force. The mud injection confidence score is a comprehensive indicator characterizing the probability that a pollution event is caused by construction mud injection, calculated through multi-dimensional coupling of the matching coefficient and the dynamic decoupling weight.

[0042] Understandably, by comparing the changing trends of water flow velocity vectors and water quality concentration parameters, identifying the dynamic decoupling phenomenon where there is no time correlation between concentration increases and flow velocity fluctuations, it is possible to preliminarily determine the existence of anthropogenic emissions independent of natural hydrodynamics. By constructing a response time curve reflecting the distribution of pollutants independent of flow field dynamic disturbances, the temporal characteristics of concentration anomalies can be further analyzed. By collecting characteristic pollutant concentrations (high suspended solids, high turbidity) during abnormal fluctuations and combining them with the instantaneous deviation of pH (alkaline characteristics), a mineralization fingerprint characterizing the discharge of building materials is generated, identifying the inorganic component characteristics of construction mud from a chemical composition perspective. By performing normalized correlation analysis between the mineralization fingerprint and a preset construction mud characteristic spectrum to obtain matching coefficients, mud characteristics can be quantified from a chemical similarity perspective. By determining the dynamic decoupling weight based on the leading or lagging characteristics of concentration increases in the response time curve, anthropogenic emission characteristics can be quantified from a temporal independence perspective. By performing multidimensional coupling calculations of the matching coefficient and the dynamic decoupling weight, the confidence level of mud injection is obtained. When the confidence level exceeds the preset mud judgment threshold, construction mud injection pollution can be accurately identified, enabling precise capture of illegal mud discharge during the construction period, avoiding confusion with natural sediment erosion, and significantly improving the pertinence and effectiveness of urban water environment supervision in construction.

[0043] For example, in monitoring a construction site near a river, the system compared the trends of flow velocity vector and suspended solids concentration at the monitoring nodes: the flow velocity fluctuated between 0.3 and 0.8 m / s during the monitoring period, exhibiting normal tidal periodicity; however, the suspended solids concentration suddenly surged from 45 mg / L to 380 mg / L at 14:30, and then rapidly decreased after about 2 hours. There was no temporal correlation between the concentration change and the flow velocity fluctuation, with a cross-correlation coefficient of only 0.12, indicating a dynamic decoupling phenomenon. A response time curve independent of the flow field dynamic disturbance was constructed. During the abnormal period, the concentrations of characteristic pollutants were collected: suspended solids 380 mg / L, turbidity 280 NTU, and the pH value was simultaneously monitored to rise instantaneously from the background value of 7.2 to 8.6, a deviation of 1.4 units. A mineralization characteristic fingerprint (a feature vector composed of suspended solids concentration, turbidity, and pH deviation) was generated. The fingerprint was compared with preset construction mud characteristic spectra (suspended solids standard range 300-500 mg / L, turbidity standard range 200-400 NTU, pH deviation standard range 1.0-2.0) using normalized correlation analysis. The calculated cosine similarity was 0.93, and the matching coefficient was 0.93. Analysis of the response time curve showed that the concentration increase began at 14:30, while the flow rate was in a decreasing phase (flow rate decreased from 0.6 m / s to 0.4 m / s from 14:00 to 15:00). The concentration increase significantly lagged behind the peak flow rate and was in the opposite direction to the flow rate change. The dynamic decoupling weight was determined to be 0.85, and the confidence level of the mud injection calculated by multidimensional coupling was 0.89, exceeding the preset mud judgment threshold of 0.75. The system determined that this pollution type was construction mud injection pollution.

[0044] Preferably, constructing a spatial transport matrix of water pollutants based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter includes: Based on the river morphology and monitoring node distribution of the water body area to be monitored, a gridded water flow topology map is established, and the water flow velocity vector at the center of each grid is projected onto the topology axis to calculate the unit flow exchange coefficient between adjacent grids. The differences in water quality concentration parameters between adjacent monitoring nodes are extracted, and combined with the physical distance between nodes, the concentration gradient vector of pollutants in the three-dimensional spatial coordinate system is calculated to characterize the spatial diffusion trend of pollutants. The unit flow exchange coefficient is coupled with the concentration gradient vector, and a convection-diffusion discrete equation describing the migration of pollutants with water flow is established based on the law of conservation of mass. The convection-diffusion discrete equations are transformed into matrix form, where the elements of the matrix represent the mass load flux of pollutants migrating from a certain grid point to an adjacent grid point per unit time, forming a spatial transport matrix that reflects the real-time spatial dynamic distribution of pollutants.

[0045] It should be noted that a gridded flow topology map refers to a computational domain mapping that discretizes a continuous river channel space into several interconnected grid cells. The grid shape is selected according to the river morphology, such as rectangle, curved quadrilateral, or triangle. The topological connectivity reflects the adjacency relationship between the flow direction and the grid cells. The topological axis refers to the coordinate axis direction along the main flow direction of the river, usually the longitudinal axis of a curvilinear coordinate system. The unit flow exchange coefficient refers to the volumetric flow rate of water passing through the interface of adjacent grid cells per unit time. It is calculated by multiplying the projected component of the velocity vector at the interface along the topological axis with the flow area of ​​the interface, and characterizes the intensity of water exchange between grid cells. The physical distance refers to the actual spatial distance between adjacent monitoring nodes or grid centers. Considering the tortuous shape of the river channel, the distance along the path is used instead of the straight-line distance. The concentration gradient vector is the vector of the rate of change of pollutant concentration in various spatial directions. It is calculated by the ratio of the concentration difference between adjacent nodes to the physical distance. The direction points to the direction of the fastest increase in concentration, and the magnitude reflects the drasticness of the concentration change. The convection-diffusion discrete equations refer to the numerical equations obtained by discretizing the continuous convection-diffusion partial differential equations into a system of algebraic equations in space. The convection term describes the transport of pollutants with the water flow, while the diffusion term describes the dispersion of pollutants from high-concentration areas to low-concentration areas. The mass loading flux refers to the total mass of pollutants migrating through an interface per unit time, determined by the flow exchange coefficient and the concentrations (convection term) and concentration gradients (diffusion term) on both sides of the interface. The spatial transport matrix is ​​the coefficient matrix obtained by rearranging the convection-diffusion discrete equations into matrix algebraic form. The dimension of the matrix equals the total number of discrete grids. Off-diagonal elements represent the mass exchange intensity between adjacent grids, while diagonal elements represent the mass conservation relationship within the grid. The matrix as a whole reflects the dynamic transport relationship of pollutants between points in space in real time.

[0046] Understandably, by establishing a gridded flow topology map based on river morphology and monitoring node distribution, complex geometric boundary conditions can be transformed into a regular computational grid. By projecting the flow velocity vector onto the topological axis and calculating the unit flow exchange coefficient, the intensity of water exchange between grids can be quantified, providing a foundation for convective transport calculations. By extracting the differences in water quality concentration parameters between adjacent monitoring nodes and combining them with physical distance to calculate the concentration gradient vector, the spatial diffusion trend of pollutants can be characterized, providing a foundation for diffusion transport calculations. By coupling the unit flow exchange coefficient with the concentration gradient vector and establishing a convection-diffusion discrete equation based on the law of mass conservation, the physical process of pollutant migration with the water flow can be fully described. By transforming the discrete equation into matrix form, where matrix elements represent the mass load flux per unit time, a spatial transport matrix that reflects the real-time spatial dynamic distribution of pollutants can be formed.

[0047] Preferably, tracking the evolution trajectory of the pollutant mass load in the water body region based on the spatial transport matrix includes: Based on the mass load flux at each grid point in the spatial transport matrix, the net migration rate of pollutants between adjacent grids is calculated to obtain the spatial evolution intensity. Satisfying the relation: ; in, Indicates from grid point Point to adjacent grid points Spatial evolution intensity, Represents the grid points in the spatial transport matrix and The flow exchange coefficient between them Represents grid points Real-time concentration of pollutants at the location This indicates the preset turbulent diffusion tensor in directional components, Represents grid points and The spatial gradient vector of pollutants between them; By performing spatiotemporal integration of the spatial evolution intensity and the water flow velocity vector, the displacement vector of the pollutant agglomerate within a continuous time step is determined, thus obtaining the pollutant evolution trajectory coordinates. Satisfying the relation: ; in, Indicates in The coordinates of the pollutant evolution trajectory at any given moment. express The trajectory coordinates at any given moment This represents the effective migration velocity vector after being corrected for the spatial evolution intensity. The effective migration velocity vector is formed by superimposing the water flow velocity vector and the convection-diffusion component derived from the spatial transport matrix. Indicates the time variable of integration; Based on the hydrological environmental factors corresponding to the pollutant evolution trajectory coordinates, the biochemical reaction losses of pollutants along the migration path are verified, and trajectory correction coefficients are obtained. Satisfying the relation: ; in, Indicates the trajectory correction coefficient. This represents the evolution path space curve formed by the trajectory coordinates. Indicates temperature and dissolved oxygen Constrained pollutant self-cleaning degradation function, Elements representing the arc length along the path curve. This represents the magnitude of the effective migration velocity vector; By mapping and coupling the pollutant evolution trajectory coordinates with the trajectory correction coefficient, the entire pollution evolution trajectory, including mass decay characteristics, is output, yielding the final evolution trajectory tracking result. The following relation is satisfied: ; in, This indicates the final evolution trajectory tracking result. This represents the total mass load of pollutants released by the pollution source at the initial moment. This represents the cumulative trajectory correction coefficient as it evolves over time.

[0048] It should be noted that spatial evolution intensity refers to the net mass flux of pollutants migrating through adjacent grid interfaces per unit time. It is calculated from the difference between the convective transport term (the product of the flux exchange coefficient and the concentration) and the diffusion transport term (the product of the diffusion coefficient and the concentration gradient). A positive value indicates that the net migration direction is... Negative values ​​indicate the opposite direction. The turbulent diffusion tensor is a second-order tensor characterizing the anisotropic diffusion properties of pollutants under turbulent conditions, and its components... reflect directional concentration gradient in The diffusion flux generated by the direction is usually estimated using empirical formulas based on flow velocity, water depth, and river characteristics. The effective migration velocity vector refers to the macroscopic velocity of pollutant clumps after considering both convection and turbulent diffusion. It is obtained by superimposing and correcting the convection-diffusion component derived from the flow velocity vector and the spatial transport matrix, reflecting the composite result of the actual migration velocity and direction of pollutants. Spatiotemporal integration refers to the integration of the effective migration velocity vector over a time interval, obtaining the total displacement by accumulating the displacement vectors at each time step. Biochemical reaction loss refers to the mass loss of pollutants during migration due to reactions such as biodegradation and chemical oxidation. The trajectory correction coefficient is a decay factor characterizing the proportion of remaining mass of pollutants after migration along the evolution trajectory. It is calculated by the integral exponential decay of the self-cleaning degradation function along the path, ranging from 0 to 1. Values ​​closer to 0 indicate more severe loss, while values ​​closer to 1 indicate less severe loss. The evolution path space curve is a continuous curve formed by connecting a series of pollutant evolution trajectory coordinates; it is the path domain for integration calculations. The arc length element refers to the infinitesimal length segment along the space curve, used for line integral calculations. The cumulative trajectory correction coefficient refers to the cumulative product of trajectory correction coefficients from the start of pollution to the current time, reflecting the degree of quality degradation throughout the entire process.

[0049] It is understandable that by calculating the net migration rate (spatial evolution intensity) of pollutants between adjacent grids based on the mass loading flux of each grid point in the spatial transport matrix, the migration direction and intensity of pollutants between grids can be quantified, where the convection term... Reflects the transport and diffusion of pollutants carried by water flow. Reflecting dispersion driven by concentration gradients, the difference between the two yields the net migration flux. By integrating the spatial evolution intensity with the water flow velocity vector in time and space, the displacement vector of pollutant clumps over continuous time steps can be determined, where the effective migration velocity vector is... By integrating the coupling effects of convection and diffusion, the coordinates of the pollutant evolution trajectory are obtained through integral accumulation. This allows for the tracking of the spatial movement of pollution clouds from a Lagrange perspective. This is achieved by analyzing the hydrological environmental factors (temperature) corresponding to the coordinates of the pollutant evolution trajectory. Dissolved oxygen Verify biochemical reaction losses and calculate trajectory correction coefficients. It can quantify the natural decay of pollutants along their migration paths, where the self-cleaning degradation function... Reflecting the influence of environmental conditions on the degradation rate, along the path curve The integral takes into account the cumulative loss effect throughout the migration process. By mapping and coupling the pollutant evolution trajectory coordinates with trajectory correction coefficients, the output is a complete trajectory of the pollution evolution process, including mass decay characteristics. It can simultaneously obtain the spatial location and remaining mass information of pollution clouds, enabling a complete characterization of the migration path and decay process of pollutants.

[0050] Preferably, the self-degradation rate of pollutants during transport is calculated based on the time history of concentration decay between adjacent monitoring nodes: Extract the concentration observation values ​​of adjacent monitoring nodes along the evolution trajectory at corresponding times, and calculate the total attenuation coefficient of pollutants during water transport. The following relation is satisfied: ; in, This represents the total attenuation coefficient of pollutants between adjacent monitoring nodes. This represents the observed pollutant concentration at the upstream monitoring node of the evolution trajectory. This indicates that the downstream monitoring nodes have undergone evolution time. Subsequent pollutant concentration observations This indicates the time interval between adjacent monitoring nodes as pollutant clumps migrate with the water flow; The spatial transport matrix is ​​used to quantify the contribution of concentration reduction due to physical diffusion between adjacent monitoring nodes, thus obtaining the physical dilution ratio. The following relation is satisfied: ; in, This indicates the percentage of concentration decrease caused by physical diffusion. This represents the longitudinal dispersion coefficient of the river channel obtained by mapping from the spatial transport matrix. This represents the second-order spatial gradient of concentration along the evolution trajectory. After removing the effect of physical dilution from the total attenuation coefficient, the net attenuation rate caused by biochemical degradation is extracted to obtain the self-cleaning degradation rate. The following relation is satisfied: ; in, This indicates the self-cleaning degradation rate of pollutants under the current water temperature environment. This represents the preset temperature correction factor. This indicates the water temperature parameters obtained in real time through the monitoring node. This indicates the preset biochemical degradation baseline reference temperature; Verify the nonlinear mapping relationship between the self-cleaning degradation rate and the dissolved oxygen concentration in the water, and output the corrected final self-cleaning degradation rate. The following relation is satisfied: ; in, This indicates the corrected final self-cleaning degradation rate. This indicates the real-time dissolved oxygen concentration in the water body obtained through the monitoring node. This represents the self-cleaning reaction half-saturation constant corresponding to the pollution type.

[0051] It should be noted that the total attenuation coefficient refers to the total rate constant of concentration reduction during the migration of pollutants between adjacent monitoring nodes. It is calculated as the ratio of logarithmic concentration difference to transport time, comprehensively reflecting the combined effects of physical dilution and biochemical degradation. Transport duration refers to the time interval required for pollutant clumps to migrate from upstream to downstream monitoring nodes with the water flow, estimated by the node spacing and average flow velocity or determined through tracer experiments. The physical dilution ratio refers to the proportion of concentration reduction caused by physical diffusion (longitudinal dispersion, lateral mixing, etc.) to the total concentration reduction, calculated by the relative relationship between the product of the longitudinal dispersion coefficient and the second-order spatial gradient of concentration and the total attenuation. The longitudinal dispersion coefficient is a parameter characterizing the dispersion and mixing of pollutants in the longitudinal direction of the river channel due to uneven flow velocity profiles and turbulent diffusion, estimated by mapping the spatial transport matrix based on flow velocity distribution characteristics. The second-order spatial gradient of concentration refers to the rate of change of the spatial variation rate of pollutant concentration along the water flow direction, reflecting the curvature characteristics of the concentration distribution. The temperature correction factor is a parameter characterizing the influence of water temperature on the rate of biochemical reactions. It is usually determined based on the Arrhenius equation or empirical values ​​(e.g., 1.047-1.08), reflecting the factor by which the degradation rate increases by 1°C. The biochemical degradation baseline reference temperature is the standard reference temperature for temperature correction, typically 20°C or a laboratory-measured temperature. The nonlinear mapping relationship refers to the saturated functional relationship between the self-cleaning degradation rate and the dissolved oxygen concentration, characterized by a rapid increase in degradation rate with increasing dissolved oxygen at low dissolved oxygen levels, followed by a gradual plateau at high dissolved oxygen levels. The self-cleaning reaction half-saturation constant is the dissolved oxygen concentration at which the degradation rate reaches half of its maximum rate, reflecting the dependence of pollutant degradation on dissolved oxygen.

[0052] Understandably, by extracting concentration observations from adjacent monitoring nodes along the evolution trajectory and calculating the total attenuation coefficient, the rate of concentration reduction of pollutants during transport can be comprehensively quantified. By quantifying the physical dilution ratio using the spatial transport matrix, the contribution of concentration reduction caused by physical processes such as longitudinal dispersion can be separated from the total attenuation. The longitudinal dispersion coefficient is obtained by mapping from the spatial transport matrix, and the second-order concentration gradient reflects the non-uniformity of concentration distribution. The ratio of their product to the total attenuation yields the physical dilution ratio. By removing the influence of physical dilution from the total attenuation coefficient and introducing temperature correction, the net attenuation rate caused by biochemical degradation, i.e., the self-cleaning degradation rate, can be obtained. The temperature correction considers the impact of the difference between the actual water temperature and the reference temperature on microbial activity. By verifying the nonlinear mapping relationship between the self-cleaning degradation rate and dissolved oxygen concentration (using the Michaelis-Menten equation), the degradation rate under dissolved oxygen-limited conditions can be further corrected. The half-saturation constant reflects the aerobic characteristics of a specific pollution type. In summary, the quantitative separation of physical dilution and biochemical degradation, as well as the comprehensive correction of multiple environmental factors, have been achieved, making the estimation of the self-cleaning degradation rate more accurate and reliable, and significantly improving the characterization accuracy of natural purification processes in pollution prediction.

[0053] Preferably, by combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, a spatiotemporal diffusion model for pollutants is established to simulate the coverage area and peak concentration of the pollution diffusion cloud within a preset time period, including: Based on the migration path determined by the evolution trajectory, calculate the future... The spatial location of the center point of the pollution cloud at any given time is used to calculate the total amount of residual pollutants, combined with the final self-cleaning degradation rate. Satisfying the relation: ; in, Indicates the future The total mass of pollutants remaining in the pollution cloud at all times. This represents the total mass load of pollutants released by the pollution source at the initial moment. This indicates the corrected final self-cleaning degradation rate. This indicates the preset simulation deadline. Indicates the starting point of the pollution event; A two-dimensional Gaussian diffusion model was used to characterize the mass distribution of the pollution cloud on the water surface, and the predicted pollutant concentrations at each coordinate point were calculated. Satisfying the relation: ; in, Indicates in Time coordinates are Predicted pollutant concentration values ​​at the location This indicates that the evolutionary trajectory is determined by the evolutionary trajectory. The center coordinates of the cloud cluster at any given time and These represent the discrete coefficients along the water flow direction and perpendicular to the water flow direction, respectively, obtained by mapping from the spatial transport matrix; Based on the spatial gradient evolution of the predicted pollutant concentrations, a set of boundaries satisfying a preset concentration threshold is extracted to define the coverage area of ​​the pollution diffusion cloud, and the concentration maxima within the coverage area are searched to obtain the concentration peak. The following relation is satisfied: ; in, Indicates the predicted peak concentration. Indicates that by satisfying The coverage area of ​​the pollution diffusion cloud formed by the constrained coordinate points. This indicates the preset pollution determination boundary concentration threshold.

[0054] It should be noted that the simulation cutoff time refers to the end point of the pollution diffusion prediction, set according to emergency decision-making needs, usually several hours to several days after the pollution occurs. The pollution cloud center point refers to the geometric center or concentration-weighted center of the pollutant mass distribution, migrating along the evolution trajectory, representing the core location of the pollution impact. The residual total amount refers to the total mass of pollutants remaining at future times after self-purification degradation, calculated from the initial mass load according to the exponential decay law, reflecting the pollution load after natural purification. The two-dimensional Gaussian diffusion model is a simplified diffusion model that assumes that pollutants are distributed in a Gaussian (normal) manner on a horizontal plane, where the concentration decreases exponentially from the center to the surrounding areas, suitable for relatively uniform and open water bodies. The dispersion coefficient along the water flow direction refers to the parameter characterizing the dispersion of pollutants in the mainstream direction due to velocity shear and turbulent diffusion, obtained by mapping the spatial transport matrix according to the longitudinal velocity distribution characteristics. The dispersion coefficient perpendicular to the water flow direction refers to the parameter characterizing the dispersion of pollutants in the lateral direction due to lateral circulation and turbulent diffusion, obtained by mapping the spatial transport matrix according to the river width, water depth, and lateral velocity distribution. Predicted concentration values ​​refer to estimated pollutant concentrations at a specific location at a future time, calculated using a diffusion model. Pollution threshold concentrations define the critical concentration range for pollution impact, typically set based on water quality standards (such as Class III water quality limits) or ecological risk thresholds. Coverage area refers to the spatial region comprised of all coordinate points where predicted concentrations are greater than or equal to the concentration threshold, representing the geographical extent of pollution impact. Concentration maxima are the locations within the coverage area where predicted concentrations reach their maximum; the concentration peak at these points represents the intensity center of the pollution impact.

[0055] Understandably, by calculating the spatial location of the pollution cloud's center point at future moments based on the migration path determined by the evolution trajectory, it is possible to predict where the core area of ​​pollution impact will reach over time. Furthermore, by combining the final self-purification degradation rate to calculate the total amount of residual pollutants, the remaining pollution load after natural purification can be quantified, avoiding over-prediction caused by the simple assumption of no attenuation. A two-dimensional Gaussian diffusion model is used to characterize the mass distribution of the pollution diffusion cloud on the water surface, where the dispersion coefficient along the water flow direction... Coefficient of variation with respect to the vertical water flow direction Obtained by spatial transport matrix mapping, it reflects the influence of river hydrodynamic characteristics on diffusion patterns (usually...). (i.e., longitudinal diffusion is stronger than lateral diffusion), which allows for the calculation of pollutant concentration predictions at each coordinate point, thus obtaining the complete spatial distribution of concentrations at future times. By extracting the boundary set that meets the preset concentration threshold based on the spatial gradient evolution of the concentration predictions, the coverage area of ​​the pollution diffusion cloud is defined, and the concentration maxima within the coverage area are searched to obtain the concentration peak. This allows for the simultaneous acquisition of information on the spatial scale and intensity center of the pollution impact.

[0056] Preferably, the water pollution prediction and assessment results include: The evolution trajectory is overlaid with a geographic information layer to generate a dynamic distribution map that includes the extent of polluted water areas, overlapping locations of ecologically sensitive areas, and the degree of spatial damage. Statistical analysis of the spatiotemporal evolution data of pollutant mass load in water areas yields structured charts reflecting concentration evolution trends, migration of areas exceeding standards, and pollution reduction processes. Based on the evolution trajectory and the time of reaching the preset section, a digital early warning dashboard is obtained, including a risk countdown, pollution transit duration, and expected recovery period. By integrating the dynamic distribution map, the structured chart, and the digital early warning dashboard, the water pollution prediction and assessment results are obtained.

[0057] It should be noted that a geographic information layer refers to digital map data containing spatial information such as river vector data, water boundaries, land use types, and ecological protection zone boundaries, typically stored and managed in GIS (Geographic Information System) format. A dynamic distribution map refers to a thematic map reflecting the spatial distribution of pollution that can evolve and update over time; it is obtained by overlaying the evolution trajectory and predicted concentration field onto a geographic information layer. Overlapping locations of ecologically sensitive areas refer to the spatial intersections between the pollution diffusion range and ecologically sensitive areas such as drinking water source protection areas, fish spawning grounds, and wetland parks; these are key areas of focus for environmental risk management. Spatial damage level refers to the severity of pollution impact assessed comprehensively based on the number of times pollutant concentration exceeds water quality standards or the duration of such exceedances. Spatiotemporal evolution data refers to the sequence of changes in parameters such as pollutant mass load, concentration, and degradation rate, encompassing both temporal and spatial dimensions. Structured charts refer to data visualization results organized in the form of tables, graphs, heat maps, etc., that are easy to understand and analyze. Pre-set cross-sections refer to key monitoring locations set according to management needs, such as water intakes, provincial boundary cross-sections, and assessment cross-sections. Risk countdown refers to the estimated remaining time before a pollution cloud reaches a pre-defined cross-section, used to assess the urgency of emergency decisions. Pollution transit duration refers to the time it takes for a pollution cloud to pass through a pre-defined cross-section and return to background levels, reflecting the duration of the pollution event. Expected recovery period refers to the predicted time when pollution will recede to acceptable levels based on the self-degradation rate, used to assess the recovery prospects. Digital early warning dashboards are integrated, real-time updated visual monitoring interfaces that provide managers with readily available decision support information.

[0058] Understandably, by overlaying the evolution trajectory with geographic information layers to generate a dynamic distribution map, the spatial distribution of pollution and its relationship with ecologically sensitive areas can be visually displayed, enabling managers to quickly identify risk areas. By statistically analyzing the spatiotemporal evolution data of pollutant mass loads to generate structured charts, the concentration evolution trend, the migration of areas exceeding standards, and the pollution reduction process can be quantitatively revealed, supporting trend analysis and effect evaluation. By generating digital early warning dashboards based on the evolution trajectory and the time of arrival at preset cross-sections, key time information such as risk countdown, pollution transit duration, and expected recovery period can be provided, supporting emergency dispatch decisions. By integrating dynamic distribution maps, structured charts, and digital early warning dashboards, multi-dimensional and multi-level water pollution prediction and assessment results can be formed, meeting the information needs of managers at different levels, achieving information flow from technical personnel to command and decision-makers, and significantly improving the efficiency of pollution assessment results transformation and application, as well as the scientific nature of emergency prevention and control command.

[0059] Urban river and lake water pollution assessment system based on spatiotemporal evolution characteristics, such as Figure 2 As shown, it includes: The data acquisition module is used to acquire hydrological monitoring data of the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall runoff data of several monitoring nodes. The pollution type identification module is used to identify the pollution type of the urban river and lake area to be monitored by using the response time curve of the water quality concentration parameter and the regional rainfall runoff data, combined with the pollutant concentration ratio between different monitoring nodes. The spatial transport construction module is used to construct a spatial transport matrix of water pollutants based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, and to quantify the mass load of pollutants migrating with the water flow per unit time. The trajectory tracking and degradation calculation module is used to track the evolution trajectory of the pollutant mass load in the water area based on the spatial transport matrix to determine the pollution destination, and to calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. The prediction and assessment module is used to combine the pollutant mass load, the self-cleaning degradation rate and the evolution trajectory to establish a pollutant spatiotemporal diffusion model, simulate the coverage range and concentration peak of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

[0060] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0061] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics, characterized in that, Includes the following steps: S1: Obtain hydrological monitoring data for the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall and runoff data for several monitoring nodes. S2: Using the response time curves of the water quality concentration parameters and the regional rainfall-runoff data, combined with the pollutant concentration ratios between different monitoring nodes, the pollution type of the urban river and lake area to be monitored is identified; S3: Based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, construct a spatial transport matrix of water pollutants and quantify the mass load of pollutants migrating with the water flow per unit time. S4: Based on the spatial transport matrix, track the evolution trajectory of the pollutant mass load in the water area to determine the pollution destination, and calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. S5: Combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, establish a spatiotemporal diffusion model for pollutants, simulate the coverage range and peak concentration of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

2. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, The pollution types in the urban river and lake areas to be monitored include stormwater and sewage overflow pollution. Step S2 includes: By analyzing the flow growth slope of the regional rainfall-runoff data, the initial rainfall period is identified, and the instantaneous increase of the water quality concentration parameter during the initial rainfall period is extracted to construct a response time curve reflecting the trend of pollutant concentration with cumulative runoff. Using the water quality concentration parameters of the aforementioned monitoring nodes, the cross-concentration ratios between chemical oxygen demand, total nitrogen, and total phosphorus are calculated to generate the current water body's pollution component fingerprint information. The pollution component fingerprint information is then subjected to normalized correlation analysis with the preset combined sewer overflow characteristic spectrum. Calculate the peak phase lag time of the response time curve. If the peak phase lag time is within the preset pipe network confluence time interval, and the correlation coefficient between the pollution component fingerprint information and the characteristic spectrum of the combined sewer overflow meets the preset intensity threshold, then the pollution type of the urban river and lake area to be monitored is identified as stormwater and sewage overflow pollution.

3. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, The pollution types in the urban river and lake areas to be monitored include direct discharge pollution from production and daily life. Step S2 includes: Real-time access to regional rainfall and runoff data; when the regional rainfall and runoff data are consistently below a preset dry season threshold, extract peak data that deviates from the conventional benchmark characteristic value in the water quality concentration parameter, and calculate the rising slope and duration of the peak data to construct a response time curve for external interference during the rainless period. The pollutant concentrations at each monitoring node during the occurrence of the peak data are collected, and a real-time pollution fingerprint characterizing the current chemical characteristics of the water body is generated by calculating the instantaneous concentration ratios between ammonia nitrogen, total phosphorus, and chemical oxygen demand. The vector space distance between the real-time pollution fingerprint and the conventional benchmark feature value of the water body area to be monitored is calculated to obtain the feature offset that reflects the degree of structural variation of pollutant components. The real-time pollution fingerprint is input into the pollution source strong fingerprint database, and the feature source type with the highest matching degree is obtained by normalized correlation algorithm. The pollution source strong fingerprint database includes the feature spectrum of domestic sewage and the feature spectrum of industrial wastewater. The intensity of external disturbance is quantified based on the rising slope of the response time curve, and the pollutant component variation weight is determined according to the characteristic offset. Multidimensional coupling calculation is performed using the intensity of external disturbance and the component variation weight to obtain the direct emission confidence level characterizing the probability of non-natural emissions. By comparing the direct discharge confidence level with the preset judgment threshold, if the direct discharge confidence level exceeds the preset judgment threshold and the retrieval result of the feature source type matches the feature of organic wastewater, then the current pollution type is determined to be direct discharge pollution from production and domestic use.

4. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, The pollution types in the urban river and lake areas to be monitored include construction mud injection pollution; Step S2 includes: By comparing the changing trends of the water flow velocity vector and the water quality concentration parameter, if it is found that there is no time correlation between the increase in concentration and the fluctuation of flow velocity, a response time curve reflecting the distribution of pollutants is independent of the dynamic disturbance of the flow field is constructed. The concentration of characteristic pollutants during the abnormal fluctuations of the response time curve is collected, and combined with the instantaneous deviation of the pH value of the water body at the monitoring node, a mineralization characteristic fingerprint characterizing the discharge of building materials is generated. The mineralization feature fingerprint is subjected to normalized correlation analysis with the preset construction mud feature spectrum to obtain the matching coefficient reflecting the degree of abnormality of inorganic components, and the dynamic decoupling weight is determined according to the leading or lagging characteristics of concentration increase in the response time curve. The matching coefficient and the dynamic decoupling weight are used to perform multidimensional coupling calculation to obtain the mud injection confidence level. If the mud injection confidence level exceeds the preset mud judgment threshold, the current pollution type is determined to be construction mud injection type pollution.

5. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, Based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, the spatial transport matrix of water pollutants is constructed as follows: Based on the river morphology and monitoring node distribution of the water body area to be monitored, a gridded water flow topology map is established, and the water flow velocity vector at the center of each grid is projected onto the topology axis to calculate the unit flow exchange coefficient between adjacent grids. The differences in water quality concentration parameters between adjacent monitoring nodes are extracted, and combined with the physical distance between nodes, the concentration gradient vector of pollutants in the three-dimensional spatial coordinate system is calculated to characterize the spatial diffusion trend of pollutants. The unit flow exchange coefficient is coupled with the concentration gradient vector, and a convection-diffusion discrete equation describing the migration of pollutants with water flow is established based on the law of conservation of mass. The convection-diffusion discrete equations are transformed into matrix form, where the elements of the matrix represent the mass load flux of pollutants migrating from a certain grid point to an adjacent grid point per unit time, forming a spatial transport matrix that reflects the real-time spatial dynamic distribution of pollutants.

6. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 5, characterized in that, Tracking the evolution trajectory of the pollutant mass load in the water body based on the spatial transport matrix includes: Based on the mass load flux at each grid point in the spatial transport matrix, the net migration rate of pollutants between adjacent grids is calculated to obtain the spatial evolution intensity. Satisfying the relation: ； in, Indicates from grid point Point to adjacent grid points Spatial evolution intensity, Represents the grid points in the spatial transport matrix and The flow exchange coefficient between them Represents grid points Real-time concentration of pollutants at the location This indicates the preset turbulent diffusion tensor in directional components, Represents grid points and The spatial gradient vector of pollutants between them; By performing spatiotemporal integration of the spatial evolution intensity and the water flow velocity vector, the displacement vector of the pollutant agglomerate within a continuous time step is determined, thus obtaining the pollutant evolution trajectory coordinates. Satisfying the relation: ； in, Indicates in The coordinates of the pollutant evolution trajectory at any given moment. express The trajectory coordinates at any given moment This represents the effective migration velocity vector after being corrected for the spatial evolution intensity. The effective migration velocity vector is formed by superimposing the water flow velocity vector and the convection-diffusion component derived from the spatial transport matrix. Indicates the time variable of integration; Based on the hydrological environmental factors corresponding to the pollutant evolution trajectory coordinates, the biochemical reaction losses of pollutants along the migration path are verified, and trajectory correction coefficients are obtained. Satisfying the relation: ； in, Indicates the trajectory correction coefficient. This represents the evolution path space curve formed by the trajectory coordinates. Indicates temperature and dissolved oxygen Constrained pollutant self-cleaning degradation function, Elements representing the arc length along the path curve. This represents the magnitude of the effective migration velocity vector; By mapping and coupling the pollutant evolution trajectory coordinates with the trajectory correction coefficient, the entire pollution evolution trajectory, including mass decay characteristics, is output, yielding the final evolution trajectory tracking result. The following relation is satisfied: ； in, This indicates the final evolution trajectory tracking result. This represents the total mass load of pollutants released by the pollution source at the initial moment. This represents the cumulative trajectory correction coefficient as it evolves over time.

7. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, Based on the time history of concentration decay between adjacent monitoring nodes, the self-degradation rate of pollutants during transport is calculated: Extract the concentration observation values ​​of adjacent monitoring nodes along the evolution trajectory at corresponding times, and calculate the total attenuation coefficient of pollutants during water transport. The following relation is satisfied: ； in, This represents the total attenuation coefficient of pollutants between adjacent monitoring nodes. This represents the observed pollutant concentration at the upstream monitoring node of the evolution trajectory. This indicates that the downstream monitoring nodes have undergone evolution time. Subsequent pollutant concentration observations This indicates the time interval between adjacent monitoring nodes as pollutant clumps migrate with the water flow; The spatial transport matrix is ​​used to quantify the contribution of concentration reduction due to physical diffusion between adjacent monitoring nodes, thus obtaining the physical dilution ratio. The following relation is satisfied: ； in, This indicates the percentage of concentration decrease caused by physical diffusion. This represents the longitudinal dispersion coefficient of the river channel obtained by mapping from the spatial transport matrix. This represents the second-order spatial gradient of concentration along the evolution trajectory. After removing the effect of physical dilution from the total attenuation coefficient, the net attenuation rate caused by biochemical degradation is extracted to obtain the self-cleaning degradation rate. The following relation is satisfied: ； in, This indicates the self-cleaning degradation rate of pollutants under the current water temperature environment. This represents the preset temperature correction factor. This indicates the water temperature parameters obtained in real time through the monitoring node. This indicates the preset biochemical degradation baseline reference temperature; Verify the nonlinear mapping relationship between the self-cleaning degradation rate and the dissolved oxygen concentration in the water, and output the corrected final self-cleaning degradation rate. The following relation is satisfied: ； in, This indicates the corrected final self-cleaning degradation rate. This indicates the real-time dissolved oxygen concentration in the water body obtained through the monitoring node. This represents the self-cleaning reaction half-saturation constant corresponding to the pollution type.

8. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 7, characterized in that, Combining the pollutant mass load, the self-cleaning degradation rate, and the evolution trajectory, a spatiotemporal diffusion model for pollutants is established to simulate the coverage area and peak concentration of the pollution diffusion cloud within a preset time period, including: Based on the migration path determined by the evolution trajectory, calculate the future... The spatial location of the center point of the pollution cloud at any given time is used to calculate the total amount of residual pollutants, combined with the final self-cleaning degradation rate. Satisfying the relation: ； in, Indicates the future The total mass of pollutants remaining in the pollution cloud at all times. This represents the total mass load of pollutants released by the pollution source at the initial moment. This indicates the corrected final self-cleaning degradation rate. This indicates the preset simulation deadline. Indicates the starting time of the pollution; A two-dimensional Gaussian diffusion model was used to characterize the mass distribution of the pollution cloud on the water surface, and the predicted pollutant concentrations at each coordinate point were calculated. Satisfying the relation: ； in, Indicates in Time coordinates are Predicted pollutant concentration values ​​at the location This indicates that the evolutionary trajectory is determined by the evolutionary trajectory. The center coordinates of the cloud cluster at any given time and These represent the discrete coefficients along the water flow direction and perpendicular to the water flow direction, respectively, obtained by mapping from the spatial transport matrix; Based on the spatial gradient evolution of the predicted pollutant concentrations, a set of boundaries satisfying a preset concentration threshold is extracted to define the coverage area of ​​the pollution diffusion cloud, and the concentration maxima within the coverage area are searched to obtain the concentration peak. The following relation is satisfied: ； in, Indicates the predicted peak concentration. Indicates that by satisfying The coverage area of ​​the pollution diffusion cloud formed by the constrained coordinate points. This indicates the preset pollution determination boundary concentration threshold.

9. The method for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics according to claim 1, characterized in that, The results of the water pollution prediction and assessment include: The evolution trajectory is overlaid with a geographic information layer to generate a dynamic distribution map that includes the extent of polluted water areas, overlapping locations of ecologically sensitive areas, and the degree of spatial damage. Statistical analysis of the spatiotemporal evolution data of pollutant mass load in water areas yields structured charts reflecting concentration evolution trends, migration of areas exceeding standards, and pollution reduction processes. Based on the evolution trajectory and the time of reaching the preset section, a digital early warning dashboard is obtained, including a risk countdown, pollution transit duration, and expected recovery period. By integrating the dynamic distribution map, the structured chart, and the digital early warning dashboard, the water pollution prediction and assessment results are obtained.

10. A system for assessing urban river and lake water pollution based on spatiotemporal evolution characteristics, characterized in that: include: The data acquisition module is used to acquire hydrological monitoring data of the water body area to be monitored. The hydrological monitoring data includes water quality concentration parameters, water flow velocity vectors, and regional rainfall runoff data of several monitoring nodes. The pollution type identification module is used to identify the pollution type of the urban river and lake area to be monitored by using the response time curve of the water quality concentration parameter and the regional rainfall runoff data, combined with the pollutant concentration ratio between different monitoring nodes. The spatial transport construction module is used to construct a spatial transport matrix of water pollutants based on the spatial gradient distribution of the water flow velocity vector and the water quality concentration parameter, and to quantify the mass load of pollutants migrating with the water flow per unit time. The trajectory tracking and degradation calculation module is used to track the evolution trajectory of the pollutant mass load in the water area based on the spatial transport matrix to determine the pollution destination, and to calculate the self-cleaning degradation rate of the pollutant during the transport process based on the time history of concentration decay between adjacent monitoring nodes. The prediction and assessment module is used to combine the pollutant mass load, the self-cleaning degradation rate and the evolution trajectory to establish a pollutant spatiotemporal diffusion model, simulate the coverage range and concentration peak of the pollution diffusion cloud within a preset time period, and generate water pollution prediction and assessment results.

Images