Water source pollution prediction method and device based on multi-source data fusion for water delivery line

Abstract

The application discloses a water source pollution prediction method and equipment based on multi-source data fusion of a water delivery line, and comprises the following steps: acquiring and preprocessing multi-source heterogeneous data to construct a time-space dynamic database of the water delivery line; a one-dimensional river network and a two-dimensional water area coupled water quality model are constructed and dynamically solved to simulate the migration and diffusion of pollutants; risk research and judgment of a sudden water pollution event are carried out based on the solving result; when the monitoring data triggers an early warning threshold, the concentration time-space distribution characteristics of each monitoring node are used for transmission path inversion, a suspected pollution outlet position is locked, and a diffusion track and an emergency scheduling scheme are output. Through multi-dimensional coupling simulation and intelligent tracing, the application realizes accurate early warning and scientific emergency scheduling of sudden pollution of the water delivery project, and improves the water source protection level.

Description

Technical Field

[0001] This invention relates to the field of information technology for water conservancy projects, and more specifically to a method and equipment for predicting water pollution in water transmission lines based on multi-source data fusion. Background Technology

[0002] Current large-scale water diversion projects rely heavily on offline manual monitoring at cross-sections or single hydrodynamic model simulations for water source protection, lacking dynamic and comprehensive assessments of multiple nodes and boundary conditions along the entire route. Traditional methods often analyze the main river channel and the key reservoir separately, making it difficult to simulate in real time the complex impact of the joint operation of dam groups on the spatiotemporal transport of pollutants. Furthermore, existing water environment early warning systems mostly display simple monitoring data, lacking intelligent source tracing analysis of sudden water pollution events and the ability to generate automated emergency dispatch support schemes driven by models. This results in delayed emergency response and insufficient decision support capabilities when facing complex pollution risks across river basins and administrative regions. Summary of the Invention

[0003] This invention provides a method and equipment for predicting water source pollution in water conveyance lines based on multi-source data fusion, which enables accurate early warning and scientific emergency dispatch of sudden pollution in water conveyance projects through multi-dimensional coupled simulation and intelligent dynamic source tracing. This method can at least solve one of the above-mentioned technical problems.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for predicting water source pollution in water conveyance lines based on multi-source data fusion includes the following steps: S1. Acquire multi-source heterogeneous data of the water transmission line. The multi-source heterogeneous data includes online monitoring data collected by monitoring equipment, engineering operation parameters, geospatial information and pollution source information. Based on the pre-processed and standardized multi-source heterogeneous data, construct a spatiotemporal dynamic database of the water transmission line. S2. Construct a multidimensional coupled water quality model for the water conveyance route. The multidimensional coupled water quality model for the water conveyance route includes a one-dimensional river network hydrodynamic water quality model for describing the transport of pollutants along the main stream and tributaries, and a two-dimensional hydrodynamic water quality model for describing the diffusion of pollutants in local waters of the hub and regulating lake. S3. Using the spatiotemporal dynamic database of water conveyance lines as boundary and initial conditions, dynamic coupling calculations are performed on the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model to simulate the migration and diffusion process of pollutants in the water conveyance lines. S4. Based on the dynamic coupling solution results, conduct risk assessment of sudden water pollution events. When the monitoring data triggers the early warning threshold, perform transmission path inversion based on the concentration spatiotemporal distribution characteristics of each monitoring node, lock the location of suspected sewage outlets, and output a pollution diffusion trajectory prediction map and emergency dispatch strategy auxiliary scheme.

[0005] Furthermore, in step S1, the process of preprocessing and standardizing the multi-source heterogeneous data and constructing the spatiotemporal dynamic database of the water transmission line includes: S1.1 Perform data cleaning on the online monitoring data, identify and remove abnormal data, and use interpolation algorithms to complete the missing time series data; S1.2 Construct a unified geospatial coordinate system and time reference, and perform spatiotemporal scale alignment and format conversion on the online monitoring data, the engineering operation parameters and the geospatial information to form standardized data items; S1.3 Based on the geospatial information, the water transmission line is divided into n node units, each node unit including a water source intake, a pumping station hub, a dam control section, and a sewage outlet, and a mapping index between each node unit and the associated water area is established. S1.4 Store the standardized data items in the spatiotemporal dynamic database of the water transmission line, and establish a dynamic indexing mechanism for the node unit.

[0006] Furthermore, in the S3 dynamic coupling solution process, the working process of the one-dimensional river network hydrodynamic and water quality model includes: S3.a.1 Based on the geospatial information, the main stream and tributaries of the water conveyance line are generalized into a node-river segment topology, wherein the nodes include diversion points, confluence points and water conservancy hub nodes; S3.a.2. Based on the engineering operation parameters, set boundary conditions at the nodes of the water conservancy hub. The boundary conditions include flow process, water level process and gate opening constraints at the nodes. S3.a.3. The Saint-Venant equations are used as the hydrodynamic control equations. The one-dimensional feature processing formula of hydraulic structures is introduced to solve the hydrodynamic parameters of the main and tributary river network and obtain the velocity field and water level field information of each river section. S3.a.4. Based on the velocity field and water level field information, the migration pattern of pollutants in each river section is described using a one-dimensional convection-diffusion equation. Pollution source terms are set according to the location of the sewage outlet, and a mixing model at the node is introduced to simulate the dynamic distribution of pollutant concentration at each confluence point.

[0007] Furthermore, in the S3 dynamic coupling solution process, the working process of the two-dimensional hydrodynamic water quality model includes: S3.b.1 Select the key points and regulating lakes in the water conveyance line as the target calculation area. Based on the geospatial information, construct an unstructured grid of the target calculation area and identify the boundaries of hydraulic structures and topographic elevation data. S3.b.2. Set the initial boundary conditions of the target calculation region. The boundary conditions include the inflow rate and water quality concentration of the associated river section, as well as the gate opening and closing status and pumping flow rate set based on the engineering operation parameters. S3.b.3. Using a two-dimensional shallow water equation set as the hydrodynamic control equation, the unstructured grid is numerically discretized, and the two-dimensional velocity field and water level field distribution within the target calculation area are solved by the finite volume method. S3.b.4. Based on the two-dimensional velocity field and water level field, a two-dimensional convection-diffusion equation is introduced, coupled with the source term function characterizing the diffusion of pollution sources, to dynamically simulate the two-dimensional concentration distribution and evolution trend of pollution plumes within the target calculation area. S3.b.5 Extract the average concentration of the target calculation area at each water conservancy hub section as the outlet concentration information, and calculate the outlet flow rate information based on the flow field vector integral; S3.b.6. Feed back the outlet concentration information and outlet flow information to the one-dimensional river network hydrodynamic water quality model to realize the dynamic boundary coupling solution of the one-dimensional model and the two-dimensional model.

[0008] Furthermore, in step S3, the simulated migration and diffusion process of pollutants in the water conveyance line includes: S3.c.1 Based on the coupling interface between the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model, a time synchronization calculation step size is set, and real-time dynamic boundary matching of water level and flow is performed at the coupling interface to perform hydrodynamic coupling calculation of one-dimensional river cross-section flow and two-dimensional regulation and storage area water body. S3.c.2. Based on the hydrodynamic coupling solution, the first-order kinetic degradation parameters of pollutants and longitudinal / lateral dispersion coefficients are introduced. Based on the flow field vector and concentration gradient, the convection-diffusion mass flux of pollutants at the water conveyance line and hub node is calculated. S3.c.3 For the gate and pumping station nodes in the node unit, dynamic source and sink terms are set in the multi-dimensional coupled water quality model of the water conveyance line according to the gate opening and scheduling rules. The source and sink terms characterize the hydraulic blockage, mixing and dilution and interception effect of pollutants at the hub nodes. S3.c.4 Record the spatiotemporal evolution matrix of pollutant concentration in each computing node unit in real time, and perform weighted interpolation and visualization projection on the spatiotemporal evolution matrix of pollutant concentration to generate a spatiotemporal evolution map of dynamic migration and diffusion of pollutants in the water conveyance line.

[0009] Furthermore, in step S4, the risk assessment process for sudden water pollution events based on the dynamic coupling solution results includes: S4.a.1 Extract the spatiotemporal evolution map of dynamic migration and diffusion, and obtain the spatiotemporal distribution data of pollutant concentration in each node unit at the current time and during the pre-simulation period; S4.a.2. Compare the pollutant concentration in the spatiotemporal distribution data of the pollutant concentration with the water quality target standard of the corresponding river section. If the monitoring data triggers the early warning threshold, identify the node unit threatened by pollution and generate an initial pollution risk assessment index. S4.a.3. Based on a pre-set set of sudden water pollution event scenarios, simulate the concentration change process of the pollution plume under different water conservancy hub scheduling conditions, and calculate the pollutant warning time and the scope of the impact of exceeding the standard for the node units threatened by pollution. S4.a.4. Use a fuzzy comprehensive evaluation model to classify the current pollution risk. The classification criteria include the peak concentration of pollutants, the warning time for reaching the downstream drinking water intake, and the vulnerability of the threatened key node units. S4.a.5. Based on the risk classification and combined with the operating conditions of the gate pump group in the water transmission line, automatically match and generate an emergency dispatch strategy auxiliary plan that includes gate opening adjustment, pump station start-up and shutdown, and control gate linkage blocking.

[0010] Furthermore, in step S4, the output process of the pollution diffusion trajectory prediction map includes: S4.b.1 Map the pollutant concentration data in the spatiotemporal evolution matrix of pollutant concentration to the geospatial information layer in chronological order, and generate a pollutant concentration contour map using the Gauss-Schrödinger interpolation algorithm; S4.b.2 Based on pollutant migration flux data, the particle tracking algorithm is used to calculate and plot the displacement trajectory of the pollutant plume center over time in real time. S4.b.3. Overlay the pollutant concentration contour map with the displacement trajectory line and output it in the form of a visual interface to indicate the impact range and pollution diffusion trend of the sudden water pollution event in real time.

[0011] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the above-described method for predicting water pollution in water transmission lines based on multi-source data fusion.

[0012] The beneficial effects of this invention are reflected in: 1. This invention effectively overcomes the limitations of traditional single-scale models in handling complex nodes of large-scale water diversion projects by dynamically coupling a one-dimensional river network model with a two-dimensional regulating lake model. Utilizing unstructured meshes and the finite volume method, it can precisely characterize the hydrodynamic features and pollutant diffusion behavior around hydraulic structures such as dams and pumping stations. This allows the simulation results to more accurately reflect the water quality evolution of water conveyance lines under multiple boundary conditions, providing reliable numerical support for water resource protection.

[0013] 2. The transmission path inversion and fuzzy comprehensive evaluation model based on historical cases introduced in this invention enable rapid quantitative assessment of the risk level of pollution incidents. After monitoring data triggers a threshold, the system can automatically complete pollution source tracing and generate an emergency plan covering both engineering scheduling and non-engineering measures. This intelligent process, from risk assessment to plan generation, shortens emergency response time and improves the level of dispatch and command in the face of sudden water environment risks.

[0014] 3. This invention, by constructing a spatiotemporal dynamic database and a dynamic indexing mechanism for key node units, achieves standardized integration and retrieval of multi-source heterogeneous monitoring data, solving the analysis challenges caused by fragmented data sources. The generated dynamic migration and diffusion spatiotemporal evolution map and the superimposed trajectory prediction map transform numerical simulation results into geospatial visualization information, facilitating real-time monitoring of pollutant leakage risks and impact ranges by management personnel, and optimizing the daily supervision and emergency management processes of water conveyance lines. Attached Figure Description

[0015] The accompanying drawings, which are provided to further illustrate this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application.

[0016] Figure 1 This is a schematic diagram of the overall process of the method according to an embodiment of the present invention.

[0017] Figure 2 This is a detailed flowchart illustrating the method of an embodiment of the present invention.

[0018] Figure 3 This is a flowchart of the multidimensional coupled water quality model of the water conveyance line according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It should be noted that the meaning of "and / or" throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that simultaneously satisfies A and B. Furthermore, "multiple" refers to two or more. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0021] See Figures 1-2 This invention provides a method for predicting water source pollution in water transmission lines based on multi-source data fusion, comprising the following steps: S1. Acquire multi-source heterogeneous data of the water transmission line. The multi-source heterogeneous data includes online monitoring data collected by monitoring equipment, engineering operation parameters, geospatial information and pollution source information. Based on the pre-processed and standardized multi-source heterogeneous data, construct a spatiotemporal dynamic database of the water transmission line. S2. Construct a multidimensional coupled water quality model for the water conveyance route. The multidimensional coupled water quality model for the water conveyance route includes a one-dimensional river network hydrodynamic water quality model for describing the transport of pollutants along the main stream and tributaries, and a two-dimensional hydrodynamic water quality model for describing the diffusion of pollutants in local waters of the hub and regulating lake. S3. Using the spatiotemporal dynamic database of water conveyance lines as boundary and initial conditions, dynamic coupling calculations are performed on the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model to simulate the migration and diffusion process of pollutants in the water conveyance lines. S4. Based on the dynamic coupling solution results, conduct risk assessment of sudden water pollution events. When the monitoring data triggers the early warning threshold, perform transmission path inversion based on the concentration spatiotemporal distribution characteristics of each monitoring node, lock the location of suspected sewage outlets, and output a pollution diffusion trajectory prediction map and emergency dispatch strategy auxiliary scheme.

[0022] In this embodiment, step S1, which involves preprocessing and standardizing the multi-source heterogeneous data and constructing the spatiotemporal dynamic database of the water transmission line, includes: S1.1 Perform data cleaning on the online monitoring data, identify and remove abnormal data, and use interpolation algorithms to complete the missing time series data; S1.2 Construct a unified geospatial coordinate system and time reference, and perform spatiotemporal scale alignment and format conversion on the online monitoring data, the engineering operation parameters and the geospatial information to form standardized data items; S1.3 Based on the geospatial information, the water transmission line is divided into n node units, each node unit including a water source intake, a pumping station hub, a dam control section, and a sewage outlet, and a mapping index between each node unit and the associated water area is established. S1.4 Store the standardized data items in the spatiotemporal dynamic database of the water transmission line, and establish a dynamic indexing mechanism for the node unit.

[0023] The process of using interpolation algorithms to complete missing time-series data is as follows: Fritsch-Carlson piecewise monotonically cubic interpolation is employed to ensure that the completed values ​​always lie within the numerical range of adjacent observation points, avoiding non-physical overshoot near pollution concentration peaks; in online early warning mode, the completion operator only calls... Historical observation data prior to the specified time is used, and future values ​​are prohibited from being used to ensure the causal nature of the early warning. When the duration of continuous missing data exceeds three monitoring steps, the system terminates interpolation and marks it as "data quality missing". Through leave-one-out cross-validation experiments, the average absolute error of the completed values ​​is controlled within 5%, and the peak deviation is less than 8%, ensuring the accuracy of the completed data in the analysis of the evolution of sudden pollution events.

[0024] During the spatiotemporal scale alignment process, a coordinate transformation matrix is ​​used based on the geographic information standard of the Yangtze River-Huaihe River Water Diversion Project. Multi-source data were uniformly mapped to the CGCS2000 National Geodetic Coordinate System, with spatial alignment errors controlled within 5 meters. The division of the n node units followed the "hydraulic topology consistency principle," that is, based on the control gates, pump station spacing, and abrupt changes in river topography along the water diversion main line. Spatial segmentation is performed using a baseline scale, and a mapping index table between nodes and water areas is established to classify each node unit. Position coordinates With the associated two-dimensional hydrodynamic water quality model unstructured grid cells Dynamic matching is performed, and the matching process is conducted through... Spatial index tree pre-computes nodes With grid The inclusion relationship is established and stored as a mapping index table. The dynamic indexing mechanism for node units employs a timestamp-based B-Tree index structure, utilizing trigger mechanisms in the SQL database. Whenever a monitoring site uploads new data... At that time, the method of the present invention automatically according to The unique identifier ID updates the pollutant boundary conditions of the associated water area, enabling dynamic retrieval of data for specific time periods and key hub sections.

[0025] To address the challenge of spatial retrieval in unstructured grids, this method establishes an unstructured grid topology connectivity table, recording the topology connectivity of each grid. vertex coordinates and adjacent mesh index. Node element. With grid The spatial mapping relationship is pre-calculated using an R-Tree-based point-falling polygon algorithm, and the results are stored in a relational table in the database. When monitoring site data is uploaded, the corresponding node ID is directly queried from the relational table to obtain the relevant information. The initial source of pollutants.

[0026] This invention constructs a spatiotemporal dynamic database based on a unified geospatial coordinate system and time benchmark, achieving standardized integration of multi-source heterogeneous monitoring data. This effectively solves the analytical challenges caused by fragmented data sources and inconsistent spatiotemporal scales in water transfer projects. By using interpolation algorithms to complete missing data and establishing a dynamic indexing mechanism for node units, the continuity and integrity of model input data can be ensured, improving the retrieval efficiency of water quality parameters for water transfer lines.

[0027] See Figure 3 In this embodiment, during the S3 dynamic coupling solution process, the working process of the one-dimensional river network hydrodynamic and water quality model includes: S3.a.1 Based on the geospatial information, the main stream and tributaries of the water conveyance line are generalized into a node-river segment topology, wherein the nodes include diversion points, confluence points and water conservancy hub nodes; S3.a.2. Based on the engineering operation parameters, set boundary conditions at the nodes of the water conservancy hub. The boundary conditions include flow process, water level process and gate opening constraints at the nodes. S3.a.3. The Saint-Venant equations are used as the hydrodynamic control equations. The one-dimensional feature processing formula of hydraulic structures is introduced to solve the hydrodynamic parameters of the main and tributary river network and obtain the velocity field and water level field information of each river section. The one-dimensional feature processing formula for hydraulic structures, specifically for arc-shaped gates in engineering projects, employs a segmented flow calculation formula for free outflow and submerged outflow, as follows: when During (free outflow), the flow rate The calculation is as follows: ; when (During submerged outflow) Flow rate The calculation is as follows: ; in, The water level upstream of the sluice gate. The water level downstream of the sluice gate. The flow area of ​​the gate. For flow coefficient, This is the submergence correction coefficient, whose value is obtained by looking up a table or by linear interpolation based on the gate opening and water level difference.

[0028] S3.a.4. Based on the velocity field and water level field information, the migration pattern of pollutants in each river section is described using a one-dimensional convection-diffusion equation. Pollution source terms are set according to the location of the sewage outlet, and a mixing model at the node is introduced to simulate the dynamic distribution of pollutant concentration at each confluence point.

[0029] The mixing model at the nodes is used to simulate the complete mixing process of the main and tributary water bodies and pollutant concentrations at the confluence point. It is constructed based on the law of conservation of mass, and the specific calculation formula is as follows:

[0030] in, The concentration of the mixed node water ( ), and The inflow rates of the main stream nodes are respectively ( ) and background concentration ( ), and The inflow rates of the tributary nodes are respectively ( ) and pollutant concentration ( ).

[0031] This hybrid model receives flow and concentration boundary data from each branch of the one-dimensional model in real time, and performs hydrodynamic coupling calculations at each time step. Internal dynamic refresh This ensures the quality balance at each confluence node in the river network topology.

[0032] This invention generalizes water conveyance lines into node-river segment topologies and introduces one-dimensional feature processing formulas for hydraulic structures, enabling accurate simulation of hydrodynamic processes in complex networks of main streams and tributaries. This effectively overcomes the computational distortion problem at key dam nodes in traditional models. By coupling the Saint-Venant equations with the convection-diffusion equations, efficient calculations of pollutant migration patterns are achieved. Combined with a nodal hybrid model, the invention accurately characterizes the dynamic distribution of concentrations at confluence points, providing a reliable quantitative analysis tool for water quality evolution in large-scale water transfer projects and enhancing the scientific rigor and timeliness of water quality safety monitoring across the entire water source protection system.

[0033] See Figure 3In this embodiment, during the S3 dynamic coupling solution process, the working process of the two-dimensional hydrodynamic water quality model includes: S3.b.1 Select the key points and regulating lakes in the water conveyance line as the target calculation area. Based on the geospatial information, construct an unstructured grid of the target calculation area and identify the boundaries of hydraulic structures and topographic elevation data. S3.b.2. Set the initial boundary conditions for the target calculation region. The boundary conditions include the inflow rate and water quality concentration of the associated river section, as well as the gate opening and closing status and pumping flow rate set based on the engineering operation parameters. S3.b.3. Using a two-dimensional shallow water equation set as the hydrodynamic control equation, the unstructured grid is numerically discretized, and the two-dimensional velocity field and water level field distribution within the target calculation area are solved by the finite volume method. The two-dimensional velocity and water level fields within the target computational region are calculated using the finite volume method, employing a second-order accurate Godunov scheme, targeting unstructured triangular mesh elements with unique global IDs. Its discrete equation in the conservation form is:

[0034] in, For the first The average water depth and velocity vector of each grid cell For unit area, For the numerical flux across the cell boundary, The unit normal vector of the element boundary. Let be the length of the boundary side. It is a source term vector that includes terrain slope and subsurface friction.

[0035] S3.b.4. Based on the two-dimensional velocity field and water level field, a two-dimensional convection-diffusion equation is introduced, coupled with the source term function characterizing the diffusion of pollution sources, to dynamically simulate the two-dimensional concentration distribution and evolution trend of pollution plumes within the target calculation area. S3.b.5 Extract the average concentration of the target calculation area at each water conservancy hub section as the outlet concentration information, and calculate the outlet flow rate information based on the flow field vector integral; S3.b.6. Feed back the outlet concentration information and outlet flow information to the one-dimensional river network hydrodynamic water quality model to realize the dynamic boundary coupling solution of the one-dimensional model and the two-dimensional model.

[0036] The feedback mechanism between the outlet concentration information and the outlet flow rate information employs a boundary propagation algorithm based on cross-sectional flow rate weighted averaging. (Outlet concentration...) The acquisition follows the law of conservation of mass flux, and the calculation formula is as follows:

[0037] in, For the first Pollutant concentration at each grid edge node ( ), For the first The velocity component perpendicular to the outlet section at the edge of each grid ( ), For the first The flow area corresponding to each grid edge ( That is, the product of the grid edge length and the average water depth of the cell. This represents the total number of grid edges that make up the hub's exit section.

[0038] The above formula transforms the complex spatial distribution of the local flow field in the two-dimensional model into a cross-sectional representative value that can be called by the one-dimensional model, ensuring the conservation of mass flux at the coupling interface between the one-dimensional and two-dimensional models.

[0039] The process of calculating the outlet flow rate information based on the flow field vector integral is as follows: On the boundary of the two-dimensional grid cells, select the components that make up the outlet section of the water conservancy project. Calculate the instantaneous flow rate at the outlet section using the grid edges. The formula is as follows:

[0040] The above formula transforms the complex local flow field state of the two-dimensional model into the total cross-sectional flow data that the one-dimensional model can access, thereby realizing the closed-loop coupling of water flux and mass conservation between the one-dimensional river network model and the two-dimensional local hub region at the boundary.

[0041] This invention selects key water structures and regulating lakes as target areas and constructs unstructured grids to finely characterize the water flow characteristics around complex hydraulic structures, effectively compensating for the lack of detail in local water area simulations by one-dimensional models. By employing the finite volume method to solve the two-dimensional shallow water and convection-diffusion equations, accurate simulation of pollutants in complex two-dimensional flow fields is achieved. Furthermore, by feeding back the concentration and flow information of the key water structure cross-section to the one-dimensional river network model in real time, dynamic boundary coupling is realized. This not only improves the accuracy of early warning responses to localized sudden pollution events but also provides high-precision numerical technical support for the scientific water quality coordination and emergency dispatching of large-scale water diversion projects under complex boundary conditions.

[0042] See Figure 3 In this embodiment, step S3, simulating the migration and diffusion process of pollutants in the water conveyance line, includes: S3.c.1 Based on the coupling interface between the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model, a time synchronization calculation step size is set, and real-time dynamic boundary matching of water level and flow is performed at the coupling interface to perform hydrodynamic coupling calculation of one-dimensional river cross-section flow and two-dimensional regulation and storage area water body. S3.c.2. Based on the hydrodynamic coupling solution, the first-order kinetic degradation parameters of pollutants and longitudinal / lateral dispersion coefficients are introduced. Based on the flow field vector and concentration gradient, the convection-diffusion mass flux of pollutants at the water conveyance line and hub node is calculated. The first-order kinetic degradation parameters (unit: Calibration was performed separately for different pollutants (such as COD and ammonia nitrogen), and calculations were performed based on the improved Arrhenius degradation model:

[0043] in, for The basic degradation rate constant at time (unit: ), For real-time monitoring of water temperature ( ), The dimensionless temperature response coefficient has a range of values ​​of . The model is achieved through... As a benchmark, this effectively eliminates the problem of the non-physical dimensions of the exponential term. For COD and ammonia nitrogen, respectively, based on indoor static degradation experiments and in-situ river tracer experiments, their respective values ​​were fitted to obtain... and Value library to ensure the simulation accuracy of the model under different seasons and water temperature conditions.

[0044] The longitudinal and lateral dispersion coefficients and In this context, based on Elder's empirical formula... and Fischer's empirical formula Calculate, where, Because of the water depth, The shear velocity is calculated using the following formula: (in, It is the acceleration due to gravity. The roughness coefficient is Manning's roughness coefficient. (The vertical average velocity). This physical discrete parameter is dynamically corrected in real time as the hydrodynamic characteristics (velocity and water depth) of the river cross-section change, effectively characterizing the convection-diffusion transport law of pollutants in complex water conveyance trunk lines.

[0045] S3.c.3 For the gate and pumping station nodes in the node unit, dynamic source and sink terms are set in the multi-dimensional coupled water quality model of the water conveyance line according to the gate opening and scheduling rules. The source and sink terms characterize the hydraulic blockage, mixing and dilution and interception effect of pollutants at the hub nodes. The dynamic source and sink items for dam and pumping station nodes The mathematical expression is:

[0046] in, Source-sink intensity per unit volume at the node ( ), The water volume of the control area of ​​the hub node ( ), The current water quality concentration at the node ( ), For the first Concentration at each outlet ( ), For the first The flow rate of each outlet (inflow is positive, outflow is negative). The comprehensive attenuation coefficient of hydraulic retention ( ), The dimensionless characteristic coefficients characterizing the gate blocking effect, This is the gate opening adjustment function. The total flow rate through the gate or pumping station. upstream mainstream concentration ( ).

[0047] S3.c.4 Record the spatiotemporal evolution matrix of pollutant concentration in each computing node unit in real time, and perform weighted interpolation and visualization projection on the spatiotemporal evolution matrix of pollutant concentration to generate a spatiotemporal evolution map of dynamic migration and diffusion of pollutants in the water conveyance line.

[0048] This invention, by introducing first-order kinetic degradation parameters and a dynamic source-sink mathematical model, can quantitatively characterize the hydraulic impediment, mixing, dilution, and interception effects of dams and pumping stations on pollutants, solving the problem of inaccurate pollutant migration simulation under complex boundary conditions. By introducing convection-diffusion mass flux calculation based on flow field vectors and concentration gradients, it achieves dynamic tracking of water quality throughout the entire process from macroscopic main channels to local hubs. The spatiotemporal evolution matrix and dynamic evolution map of pollutant concentration constructed by this method not only improve the calculation accuracy and mass balance completeness of pollutant migration and diffusion but also provide a quantitative analysis method for early warning of water quality safety risks in large-scale water transfer projects under complex scheduling conditions.

[0049] In this embodiment, step S4, the process of assessing the risk of a sudden water pollution event based on the dynamic coupling solution results, includes: S4.a.1 Extract the spatiotemporal evolution map of dynamic migration and diffusion, and obtain the spatiotemporal distribution data of pollutant concentration in each node unit at the current time and during the pre-simulation period; S4.a.2. Compare the pollutant concentration in the spatiotemporal distribution data of the pollutant concentration with the water quality target standard of the corresponding river section. If the monitoring data triggers the early warning threshold, identify the node unit threatened by pollution and generate an initial pollution risk assessment index. S4.a.3 Based on a pre-set set of sudden water pollution event scenarios, simulate the concentration change process of pollution plumes under different water conservancy hub scheduling conditions, and calculate the pollutant warning time and the scope of the impact of exceeding the standard for the node units threatened by pollution. S4.a.4. Use a fuzzy comprehensive evaluation model to classify the current pollution risk. The classification criteria include the peak concentration of pollutants, the warning time for reaching the downstream drinking water intake, and the vulnerability of the threatened key node units. The process of risk classification for sudden water pollution based on the fuzzy comprehensive evaluation model is as follows: establishing a risk assessment index set. ,in, The number of times the pollutant concentration exceeds the standard. For the warning time, A vulnerability index for sensitive targets; the weight vector of each index is calculated using the analytic hierarchy process (AHP). (satisfy A judgment matrix was constructed using expert scoring, and the consistency ratio was ensured through a consistency check. Based on environmental quality standards, construct fuzzy membership matrices for each indicator. After normalizing the dimensions of the evaluation factors, the fuzzy comprehensive evaluation vector is calculated. ,in," "This is a fuzzy operator (using a product-summation model); the final risk score." Mapped to a level range, when It was determined to be of extremely high risk. This assessment process pre-sets different judgment matrices for different water body functional zones (such as drinking water source protection areas). This ensures the relevance and scientific rigor of the evaluation system under different pollution scenarios.

[0050] When using a fuzzy comprehensive evaluation model to classify current pollution risks, the weight vector is determined through the analytic hierarchy process (AHP). The specific process is as follows: First, a judgment matrix is ​​constructed based on three evaluation indicators: peak pollutant concentration, warning time to downstream drinking water intake, and vulnerability level of threatened key node units. In constructing the judgment matrix, the peak pollutant concentration is set as the first dominant factor, the warning time as the second dominant factor, and the vulnerability level as the third dominant factor. A scaling value is assigned based on the relative importance of each factor, and the largest eigenvalue and corresponding eigenvector are calculated. Finally, the eigenvector is normalized to obtain the weight vector, and a consistency test is performed. If the consistency test result is less than 0.1, the weight allocation is considered reasonable.

[0051] S4.a.5. Based on the risk classification and combined with the operating conditions of the gate pump group in the water transmission line, automatically match and generate an emergency dispatch strategy auxiliary plan that includes gate opening adjustment, pump station start-up and shutdown, and control gate linkage blocking.

[0052] The process of automatically matching and generating emergency dispatch strategies involves constructing a logical rule mapping table based on an expert experience base: if the risk level is... or Then the scheduling policy generation function is called. ,in, For emergency dispatch strategy vectors, This represents the current operating state vector of the gate and the pumping station. The scheduling response gain matrix is ​​obtained from forward simulation training based on the hydrodynamic model. This is the set of physical constraint thresholds for gate opening adjustment.

[0053] The gate opening adjustment value is calculated using the above function. Adjustment amount of pump station flow :

[0054]

[0055] in, and These represent the maximum opening and the current opening of the gate, respectively. and These are the minimum flow rate and the current operating flow rate of the pumping station, respectively. For water quality to meet the standards, and All of these are scheduling feedback coefficients obtained by fitting simulated operating conditions. The water volume of the control area of ​​the hub node ( ), The pollutant warning time for node units threatened by pollution (i.e., the remaining time for the pollution plume to reach sensitive targets such as downstream water intakes). This represents the predicted peak pollutant concentration at the contaminated node. This strategy ensures that the hydraulic balance of the water conveyance line is maintained while reducing the peak concentration at the contaminated node.

[0056] This invention achieves rapid quantitative assessment of the risk level of sudden water pollution events by constructing a fuzzy comprehensive evaluation model and a mapping table of expert experience and logical rules, solving the problem of the lag in manual judgment. Based on the scheduling response gain matrix obtained from the forward simulation training of the hydrodynamic model, the system automatically calculates the gate opening adjustment value and the pump station flow adjustment amount, which can not only accurately reduce the peak concentration of polluted nodes, but also effectively ensure the hydraulic balance of the water conveyance trunk line. This scheme realizes an automated closed loop from risk warning to emergency response strategy generation, improving the scientific decision-making ability and emergency response efficiency of large-scale water transfer projects in the face of sudden environmental risks.

[0057] In this embodiment, the output process of the pollution diffusion trajectory prediction map in step S4 includes: S4.b.1 Map the pollutant concentration data in the spatiotemporal evolution matrix of pollutant concentration to the geospatial information layer in chronological order, and generate a pollutant concentration contour map using the Gauss-Schrödinger interpolation algorithm; The semi-variogram model of the Gaussian model is selected as the Gaussian model. In the formula, For spatial distance, Value of a nugget. For sill values, For variable range; The above parameters were obtained by fitting an empirical semivariogram to historical monitoring section water quality data to ensure consistency across spatial scales. The concentration field transitions smoothly within the range, effectively suppressing interpolation oscillations in numerical simulations.

[0058] S4.b.2 Based on pollutant migration flux data, the particle tracking algorithm is used to calculate and plot the displacement trajectory of the pollutant plume center over time in real time. The process of using particle tracking algorithms to calculate the displacement trajectory of the center of a contamination plume over time in real time is as follows: the contamination plume is considered as a set of uniformly distributed Lagrange particles, and its center position is... In a two-dimensional flow field The migration in the middle follows a first-order ordinary differential equation:

[0059] in, The flow field vector is obtained by solving the two-dimensional shallow water equation. To characterize the diffusion tensor of turbulent diffusion effects, This represents the local pollutant concentration gradient.

[0060] The above equations are solved by time integration using the fourth-order Runge-Kutta method, with time steps... Perform iterative updates of particle positions:

[0061] in, This is the slope term of the fourth-order integral. Through this physical tracking process, not only is the advection transport effect of the flow field on the pollutant plume characterized, but the asymmetric diffusion deviation caused by the concentration gradient is also corrected by the diffusion tensor, thereby achieving high-precision trajectory depiction at complex hub boundaries.

[0062] S4.b.3. Overlay the pollutant concentration contour map with the displacement trajectory line and output it in the form of a visual interface to indicate the impact range and pollution diffusion trend of the sudden water pollution event in real time.

[0063] The process of locating suspected sewage outlets by inverting transmission paths based on the spatiotemporal distribution characteristics of concentrations at each monitoring node specifically includes: Establish each monitoring node Concentration time series By calculating the cross-correlation function between upstream nodes and early warning nodes Find the time delay that maximizes the correlation coefficient. ,in, This is the time offset. For the upstream Each monitoring node The concentration of pollutants at any given time; Based on the flow field vector calculated from the hydrodynamic model Time lag Converted to source distance ; will calculate distance With the database pre-stored Match the known sewage outlet location information. If the conditions are met... (in, The model grid space error threshold. If the actual water flow path distance from the known sewage outlet to the current warning node is known, then that node is identified as a suspected pollution source. This location process is achieved through a deterministic mapping between physical time delay and flow field trajectory, which can effectively locate pollution sources without complex inversion operators, ensuring the feasibility of this method in engineering scenarios.

[0064] This invention achieves high-precision visualization of pollutant concentration distribution and migration trajectories by organically combining the Gauss-Schrichn interpolation algorithm and the particle tracking algorithm. The use of semi-variogram fitting effectively suppresses interpolation oscillations, ensuring a smooth transition of the concentration field at complex boundaries. The physical tracking process based on the fourth-order Runge-Kutta method characterizes the combined effects of advection and turbulent diffusion in the flow field, overcoming asymmetric diffusion bias. This method transforms abstract numerical simulation results into intuitive geospatial evolution maps, not only improving the scientific rigor of predicting the impact range of sudden water pollution events but also providing managers with a visualized decision-making basis for real-time pollution containment and emergency response strategies.

[0065] To further verify the feasibility and superiority of this method, the present invention provides the following implementation case to demonstrate the emergency dispatch decision-making process of this method in the actual operation of a large-scale water diversion project.

[0066] During the routine monitoring period following the flood season of a certain year, the spatiotemporal dynamic database of the water conveyance line collects water quality data, engineering operation parameters, and surrounding meteorological and geographical information in real time through online monitoring nodes deployed along the Yangtze River-Huaihe River diversion main line and related hub areas. At a certain monitoring moment, the system received an abnormal alarm from the online monitoring data of a sewage outlet node located approximately 15 kilometers upstream of a certain pumping station hub. The COD and ammonia nitrogen concentrations at this point showed a sudden, stepwise increase, rapidly exceeding the preset safety baseline. The system immediately and automatically triggered the sudden water pollution event risk assessment process described in this invention, performing transmission path inversion based on the spatiotemporal distribution characteristics of concentrations at each monitoring node.

[0067] The system first calculates the cross-correlation function between upstream monitoring nodes and early warning nodes to identify the time lag characteristics of concentration fluctuations. Combining this with flow field vector data calculated from the hydrodynamic model, the time lag is converted into a source-tracing distance. By comparing and mapping this source-tracing distance with pre-stored information on known sewage outlet locations in the database, the system quickly pinpoints the suspected sewage outlet location. Subsequently, based on a pre-set set of sudden water pollution event scenarios, the system simulates the migration process of the pollution plume under the current water conservancy hub's operational conditions. The simulation results show that, due to the dry season water replenishment phase in this river section, the flow velocity is relatively slow. However, influenced by the water level difference at the control gates along the river, the pollution plume's downstream path is complex. It is estimated that it will reach a major drinking water intake approximately 20 kilometers downstream in 4.2 hours, and the impact of the exceedance will cover three key downstream dam nodes.

[0068] During the risk assessment phase, a fuzzy comprehensive evaluation model is used to quantify and classify the current pollution risk. Evaluation indicators include peak pollutant concentration, warning time to the water intake, and the vulnerability level of the regional node units. Based on the weight set determined by the analytic hierarchy process (AHP), a risk membership vector for the event is derived, ultimately classifying the emergency as "extremely high risk." Once the risk classification is confirmed, the emergency dispatch strategy auxiliary scheme generation function is immediately activated. Based on an expert experience database and the dispatch response gain matrix obtained from forward simulation training of the hydrodynamic model, the optimal emergency engineering dispatch measures are automatically calculated.

[0069] The emergency dispatch plan specifically includes: firstly, by issuing instructions, the control gate downstream of the pollution plume is immediately closed to achieve a coordinated blockage and prevent the pollution plume from flowing downstream; simultaneously, the pumping flow rate of the pumping station is adjusted to the minimum operating load to avoid forced transport of pollutants via the pumping station's power; for upstream diversion nodes, the opening of relevant control gates is automatically adjusted to increase the tributary confluence ratio and thus increase the mixing and dilution efficiency of the main stream, thereby reducing local concentration peaks. The gate adjustment and pumping station flow rate adjustment in this process are both controlled by a dispatch function. Real-time calculations are performed to ensure that while reducing the peak concentration, the water level balance of the entire hydraulic system is maintained, and to avoid slope instability caused by a sudden drop in the water level in front of the gate due to sudden changes in scheduling.

[0070] During emergency response, the pollution diffusion trajectory prediction map is updated in real time based on pollutant migration flux data. Using a particle tracking algorithm, the pollution plume is treated as a set of Lagrange particles, and time integration is performed in the two-dimensional flow field using the fourth-order Runge-Kutta method to plot the displacement trajectory of the pollution plume center over time. Simultaneously, the system uses the Gauss-Schrichten interpolation algorithm to generate a contour map of the current concentration field, which is then overlaid with the trajectory line and output to the decision support platform in a visual interface. This allows dispatching and management personnel to intuitively view the dynamic process of the pollution plume being hydraulically impeded, mixed, diluted, and intercepted at complex hub boundaries.

[0071] Ultimately, with the scientific intervention of this emergency dispatch strategy, the pollution plume was effectively intercepted in the intermediate reservoir area before entering the core protection zone of the drinking water intake. The pollutant concentration was diluted and reduced to below the environmental quality standard limit. Monitoring data shows that, through the automatic assessment and emergency dispatch assistance of the method in this embodiment, the response time was shortened by approximately 65% ​​compared to traditional manual emergency dispatch, effectively ensuring the safety of water quality during the Yangtze River-Huaihe River Water Diversion Project. This embodiment not only verified the simulation accuracy of the multidimensional coupled water quality model under complex water transfer boundaries, but also demonstrated the significant value of the intelligent decision-making closed loop from risk assessment and source tracing analysis to strategy generation in practical engineering applications. Through the precise presentation of the spatiotemporal evolution map, managers were not only able to accurately handle this emergency, but also accumulated effective dispatch plans for different pollution scenarios in the region through historical data review, providing important data and experience support for the future construction of a long-term mechanism for water quality protection along the entire line.

[0072] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the above-described method for predicting water pollution in water transmission lines based on multi-source data fusion.

[0073] This invention also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the above-described method for predicting water pollution in water transmission lines based on multi-source data fusion.

[0074] This invention also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the steps of the above-described method for predicting water pollution in water transmission lines based on multi-source data fusion.

[0075] It is understood that the systems, devices and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples and beneficial effects of the relevant contents can be referred to the corresponding parts of the above-mentioned method for predicting water pollution of water sources in water transmission lines based on multi-source data fusion.

[0076] It should be noted that those skilled in the art will understand that all or part of the steps implemented in the embodiments of the present invention can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented in hardware, it can be implemented entirely or partially by purchasing standard parts or modifications. When implemented in software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0077] In summary, the water pollution prediction method for water conveyance lines based on multi-source data fusion provided by this invention integrates heterogeneous data from multiple sources and constructs a spatiotemporal dynamic database. This achieves standardized management and efficient fusion of environmental monitoring information along the entire water conveyance line, solving the technical problems of data fragmentation and incomplete spatial coverage in traditional monitoring methods. The one-dimensional and two-dimensional dynamically coupled water quality model constructed by this method can accurately simulate the pollutant migration process in tributaries and key lakes under complex boundary conditions, significantly improving the prediction accuracy of water quality evolution patterns in large-scale, long-distance water transfer projects. It realizes a closed-loop chain from automatic pollution risk assessment to intelligent matching of emergency strategies, enhancing the rapid response capability and scientific scheduling level for sudden water pollution events.

[0078] It should be understood that the examples and embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art can make various modifications or changes based on them. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A method for predicting water source pollution in water conveyance lines based on multi-source data fusion, characterized in that, Includes the following steps: S1. Acquire multi-source heterogeneous data of the water transmission line. The multi-source heterogeneous data includes online monitoring data collected by monitoring equipment, engineering operation parameters, geospatial information and pollution source information. Based on the pre-processed and standardized multi-source heterogeneous data, construct a spatiotemporal dynamic database of the water transmission line. S2. Construct a multidimensional coupled water quality model for the water conveyance route. The multidimensional coupled water quality model for the water conveyance route includes a one-dimensional river network hydrodynamic water quality model for describing the transport of pollutants along the main stream and tributaries, and a two-dimensional hydrodynamic water quality model for describing the diffusion of pollutants in local waters of the hub and regulating lake. S3. Using the spatiotemporal dynamic database of water conveyance lines as boundary and initial conditions, dynamic coupling calculations are performed on the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model to simulate the migration and diffusion process of pollutants in the water conveyance lines. S4. Based on the dynamic coupling solution results, conduct risk assessment of sudden water pollution events. When the monitoring data triggers the early warning threshold, perform transmission path inversion based on the concentration spatiotemporal distribution characteristics of each monitoring node, lock the location of suspected sewage outlets, and output a pollution diffusion trajectory prediction map and emergency dispatch strategy auxiliary scheme.

2. The method for predicting water source pollution in water transmission lines based on multi-source data fusion as described in claim 1, characterized in that, In step S1, the process of preprocessing and standardizing the multi-source heterogeneous data and constructing the spatiotemporal dynamic database of the water transmission line includes: S1.1 Perform data cleaning on the online monitoring data, identify and remove abnormal data, and use interpolation algorithms to complete the missing time series data; S1.2 Construct a unified geospatial coordinate system and time reference, and perform spatiotemporal scale alignment and format conversion on the online monitoring data, the engineering operation parameters and the geospatial information to form standardized data items; S1.3 Based on the geospatial information, the water transmission line is divided into n node units, each node unit including a water source intake, a pumping station hub, a dam control section, and a sewage outlet, and a mapping index between each node unit and the associated water area is established. S1.4 Store the standardized data items in the spatiotemporal dynamic database of the water transmission line, and establish a dynamic indexing mechanism for the node unit.

3. The method for predicting water source pollution in water conveyance lines based on multi-source data fusion as described in claim 1, characterized in that, In the S3 dynamic coupling solution process, the working process of the one-dimensional river network hydrodynamic and water quality model includes: S3.a.1 Based on the geospatial information, the main stream and tributaries of the water conveyance line are generalized into a node-river segment topology, wherein the nodes include diversion points, confluence points and water conservancy hub nodes; S3.a.

2. Based on the engineering operation parameters, set boundary conditions at the nodes of the water conservancy hub. The boundary conditions include flow process, water level process and gate opening constraints at the nodes. S3.a.

3. The Saint-Venant equations are used as the hydrodynamic control equations. The one-dimensional feature processing formula of hydraulic structures is introduced to solve the hydrodynamic parameters of the main and tributary river network and obtain the velocity field and water level field information of each river section. S3.a.

4. Based on the velocity field and water level field information, the migration pattern of pollutants in each river section is described using a one-dimensional convection-diffusion equation. Pollution source terms are set according to the location of the sewage outlet, and a mixing model at the node is introduced to simulate the dynamic distribution of pollutant concentration at each confluence point.

4. The method for predicting water source pollution in water conveyance lines based on multi-source data fusion as described in claim 3, characterized in that, In the S3 dynamic coupling solution process, the working process of the two-dimensional hydrodynamic water quality model includes: S3.b.1 Select the key points and regulating lakes in the water conveyance line as the target calculation area. Based on the geospatial information, construct an unstructured grid of the target calculation area and identify the boundaries of hydraulic structures and topographic elevation data. S3.b.

2. Set the initial boundary conditions of the target calculation region. The boundary conditions include the inflow rate and water quality concentration of the associated river section, as well as the gate opening and closing status and pumping flow rate set based on the engineering operation parameters. S3.b.

3. Using a two-dimensional shallow water equation set as the hydrodynamic control equation, the unstructured grid is numerically discretized, and the two-dimensional velocity field and water level field distribution within the target calculation area are solved by the finite volume method. S3.b.

4. Based on the two-dimensional velocity field and water level field, a two-dimensional convection-diffusion equation is introduced, coupled with the source term function characterizing the diffusion of pollution sources, to dynamically simulate the two-dimensional concentration distribution and evolution trend of pollution plumes within the target calculation area. S3.b.5 Extract the average concentration of the target calculation area at each water conservancy hub section as the outlet concentration information, and calculate the outlet flow rate information based on the flow field vector integral; S3.b.

6. Feed back the outlet concentration information and outlet flow information to the one-dimensional river network hydrodynamic water quality model to realize the dynamic boundary coupling solution of the one-dimensional model and the two-dimensional model.

5. The method for predicting water source pollution in water conveyance lines based on multi-source data fusion as described in claim 1, characterized in that, In step S3, the simulated migration and diffusion process of pollutants in the water conveyance line includes: S3.c.1 Based on the coupling interface between the one-dimensional river network hydrodynamic and water quality model and the two-dimensional hydrodynamic and water quality model, a time synchronization calculation step size is set, and real-time dynamic boundary matching of water level and flow is performed at the coupling interface to perform hydrodynamic coupling calculation of one-dimensional river cross-section flow and two-dimensional regulation and storage area water body. S3.c.

2. Based on the hydrodynamic coupling solution, the first-order kinetic degradation parameters of pollutants and longitudinal / lateral dispersion coefficients are introduced. Based on the flow field vector and concentration gradient, the convection-diffusion mass flux of pollutants at the water conveyance line and hub node is calculated. S3.c.3 For the gate and pumping station nodes in the node unit, dynamic source and sink terms are set in the multi-dimensional coupled water quality model of the water conveyance line according to the gate opening and scheduling rules. The source and sink terms characterize the hydraulic blockage, mixing and dilution and interception effect of pollutants at the hub nodes. S3.c.4 Record the spatiotemporal evolution matrix of pollutant concentration in each computing node unit in real time, and perform weighted interpolation and visualization projection on the spatiotemporal evolution matrix of pollutant concentration to generate a spatiotemporal evolution map of dynamic migration and diffusion of pollutants in the water conveyance line.

6. The method for predicting water source pollution in water transmission lines based on multi-source data fusion as described in claim 1, characterized in that, In step S4, the risk assessment process for sudden water pollution events based on the dynamic coupling solution results includes: S4.a.1 Extract the spatiotemporal evolution map of dynamic migration and diffusion, and obtain the spatiotemporal distribution data of pollutant concentration in each node unit at the current time and during the pre-simulation period; S4.a.

2. Compare the pollutant concentration in the spatiotemporal distribution data of the pollutant concentration with the water quality target standard of the corresponding river section. If the monitoring data triggers the early warning threshold, identify the node unit threatened by pollution and generate an initial pollution risk assessment index. S4.a.3 Based on a pre-set set of sudden water pollution event scenarios, simulate the concentration change process of pollution plumes under different water conservancy hub scheduling conditions, and calculate the pollutant warning time and the scope of the impact of exceeding the standard for the node units threatened by pollution. S4.a.

4. Use a fuzzy comprehensive evaluation model to classify the current pollution risk. The classification criteria include the peak concentration of pollutants, the warning time for reaching the downstream drinking water intake, and the vulnerability of the threatened key node units. S4.a.

5. Based on the risk classification and combined with the operating conditions of the gate pump group in the water transmission line, automatically match and generate an emergency dispatch strategy auxiliary plan that includes gate opening adjustment, pump station start-up and shutdown, and control gate linkage blocking.

7. The method for predicting water source pollution in water conveyance lines based on multi-source data fusion as described in claim 6, characterized in that, In step S4, the output process of the pollution diffusion trajectory prediction map includes: S4.b.1 Map the pollutant concentration data in the spatiotemporal evolution matrix of pollutant concentration to the geospatial information layer in chronological order, and generate a pollutant concentration contour map using the Gauss-Schrödinger interpolation algorithm; S4.b.2 Based on pollutant migration flux data, the particle tracking algorithm is used to calculate and plot the displacement trajectory of the pollutant plume center over time in real time. S4.b.

3. Overlay the pollutant concentration contour map with the displacement trajectory line and output it in the form of a visual interface to indicate the impact range and pollution diffusion trend of the sudden water pollution event in real time.

8. A computer device, characterized in that, The system includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the water pollution prediction method for water transmission lines based on multi-source data fusion as described in any one of claims 1-7.

Images