A method for predicting the influence of a discharge outlet on river water quality under rainfall conditions

Abstract

This invention discloses a method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions, belonging to the field of water environment management technology. By analyzing the underlying surface type and runoff coefficient, and combining this with pipeline network data to calculate pollutant inflow into the river, it achieves accurate quantification of pollution source load. Clustering algorithms are used to classify discharge outlets, improving the efficiency of subsequent modeling. A multi-model water quality prediction model is constructed to achieve dynamic and accurate prediction of water quality compliance. A multi-factor evaluation system scientifically quantifies the pollution contribution of each discharge outlet to the assessment section, generating a contribution ranking list, thus achieving accurate identification of key pollution outlets. Finally, through simulation prediction of engineering modification schemes, it realizes the scientific design and optimization of engineering measures guided by water quality compliance, providing scientific, economical, and feasible decision support for water environment management.

Description

Technical Field

[0001] This invention relates to the field of water environment management technology, specifically to a method for predicting the impact of discharge outlets on river water quality standards under rainfall conditions. Background Technology

[0002] With the acceleration of urbanization, rainwater runoff caused by rainfall will wash away and flush pollutants on the underlying surface, and carry pollutants from pipes into rivers through overflow outlets, leading to the deterioration of river water quality and seriously affecting the effectiveness of urban water environment management and river water quality compliance assessment.

[0003] To analyze the impact of rainfall on river water quality, related technical fields have developed analytical methods that couple watershed pipe networks with river channel models. For example, the existing patent CN110472887A discloses a method for analyzing the impact of rainfall on river water quality by coupling watershed pipe network-river channel model coupling. This method couples the generalized rainwater inlet parameters of the watershed pipe network model with the river channel model, inputs rainfall intensity and duration information, simulates rainwater runoff destination, and ultimately obtains the changes in river water level and pollutant concentration. However, the above-mentioned existing technologies still have significant shortcomings: 1. Its discharge outlet treatment is simplified to a generalized discharge outlet, without identifying or distinguishing discharge outlets with different pollution levels, and the boundary conditions are simple, so the prediction accuracy is limited.

[0004] 2. Its goal is limited to analyzing the impact of rainfall runoff on river water quality, and it can only output basic data, but cannot provide decision support for actual governance projects, thus being disconnected from engineering applications. Summary of the Invention

[0005] To address one or more shortcomings of the existing technology, this invention provides a method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions. This method enables the prediction of river water quality under different rainfall scales, accurate identification of key pollution discharge outlets, and scientific evaluation of engineering modification schemes.

[0006] To achieve the above objectives, the present invention adopts one or more of the following technical solutions: A method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions includes: S1. Analyze the satellite image of the target area to obtain the underlying surface type within the target area, and determine the pollutant runoff coefficient at each discharge outlet based on the underlying surface type; S2. Obtain pipeline information and outlet information within the target area, process the pipeline information and outlet information into standard data respectively, and calculate the amount of various runoff pollutants entering the river from each outlet based on the standard data and the pollutant runoff coefficient. S3. Based on the amount of each runoff pollutant entering the river, calculate the comprehensive pollution load of each discharge outlet. Based on the comprehensive pollution load, use a clustering algorithm to classify the discharge outlets in the target area to obtain the high-load discharge outlets. S4. Construct a river water quality prediction model. Based on the standard data, the amount of runoff pollutants entering the river at each discharge point, and the water conservancy scheduling data, predict the water quality compliance status of each river assessment section under different rainfall scales, and output a prediction table of water quality compliance status of each river assessment section based on the prediction results. The river water quality prediction model includes a non-point source pollution model and a river network hydrodynamic water quality model. S5. Construct a multi-factor pollution contribution assessment system. For river assessment sections whose water quality compliance prediction results are not met in the prediction table of each river assessment section, calculate the pollution contribution score of the high-load discharge outlets corresponding to the river assessment section based on the comprehensive pollution load, and generate a ranking list of discharge outlets' pollution contribution to the river based on the pollution contribution score. S6. Based on the ranking list of the contribution of the discharge outlets to river pollution, with the goal of achieving the standard of river water quality under rainfall conditions, design engineering transformation schemes for the river assessment sections where the predicted results do not meet the standards, and re-predict the water quality compliance status of the transformed river.

[0007] Preferably, the calculation of runoff pollutant inflow at each discharge point based on the standard data is performed using the following formula:

[0008] Where M is the amount of runoff pollutants entering the river, in t / h; X is the average concentration of runoff pollutants, in mg / L; q is the rainfall intensity, in mm / h; α is the runoff coefficient of the corresponding underlying surface; and A is the area of ​​the corresponding underlying surface, in m². 2 .

[0009] Preferably, step S3 specifically includes: Using the amount of each runoff pollutant entering the river as an indicator, a weighted scoring method is used to calculate the comprehensive pollution load of each discharge outlet. Using the service area of ​​each discharge outlet and the comprehensive pollution load as the core clustering factors, a hierarchical clustering agglomeration algorithm is used to initially cluster the discharge outlets in the target area. Then, the K-means method is used for iteration to obtain the discharge outlet classification results. The discharge outlet types include at least low-load small discharge outlets, medium-load discharge outlets, high-load large discharge outlets, and high-load small discharge outlets.

[0010] Preferably, the step of predicting the water quality compliance of each river assessment section under different rainfall scales based on the standard data, the amount of runoff pollutants entering the river at each discharge outlet, and water conservancy scheduling data specifically includes: The standard data, the amount of runoff pollutants entering the river at each discharge point, and the water conservancy scheduling data are input into the river water quality prediction model, and the flow change process curves of each discharge point and the pollutant concentration time history curves of each river assessment section are output. Based on the flow change process curve and the pollutant concentration time history curve, the average pollutant concentration of each river assessment section during each rainfall period is calculated. If the average pollutant concentration in all periods does not exceed the standard limit, the water quality of the river assessment section is predicted to meet the standard. If the average pollutant concentration in any period exceeds the standard limit, the water quality of the river assessment section is predicted to fail to meet the standard. Based on the prediction results, a prediction table of water quality compliance status of each river assessment section under different rainfall scales is output.

[0011] Preferably, the non-point source pollution model includes:

[0012]

[0013] In the formula, B represents the cumulative amount of pollutants, expressed in kg / hm². 2 C1 represents the maximum cumulative amount of pollutants, expressed in kg / hm². 2 C2 is the cumulative rate constant; t is the cumulative time in minutes; W is the pollutant flushing rate in mg / L; S1 is the flushing coefficient; S2 is the flushing index; q1 is the runoff per unit area in mm / h; B is the pollutant accumulation in kg / hm². 2 .

[0014] Preferably, the river network hydrodynamic water quality model includes:

[0015]

[0016]

[0017] Where A is the river channel area; Q is the cross-sectional discharge. t represents the velocity of the lateral incoming water in the direction of the river channel; x represents the horizontal coordinate along the direction of water flow; and q represents the lateral inflow discharge of the river channel. y is the momentum correction factor; g is the gravitational acceleration; y is the water level height; denoted as frictional gradient; c represents the concentration of pollutants in the runoff. 1 represents the average flow velocity across the river cross section; E x K is the convective diffusion coefficient. c t represents the attenuation coefficient of runoff pollutants; t represents time.

[0018] Preferably, the construction of the multi-factor pollution contribution assessment system involves calculating the pollution contribution score of all discharge outlets corresponding to the river assessment sections whose predicted water quality compliance status in the river assessment section prediction table is not met. This specifically includes: Based on the comprehensive pollution load, a multi-factor pollution contribution assessment system including pollution load factor, spatial attenuation factor, hydraulic dilution factor and temporal synergy factor is constructed and weighted to obtain the pollution contribution score of each discharge outlet.

[0019] Preferably, the specific process of generating the ranking list of discharge outlets' pollution contribution to the river based on the pollution contribution score is as follows: All discharge outlets are sorted in descending order of their pollution contribution scores to generate a ranking list of their pollution contribution to the river. A predetermined proportion of discharge outlets are then selected and marked as critical pollution discharge outlets.

[0020] Preferably, the pipeline information includes the category, attributes, burial depth, cross-sectional dimensions, pipe material, ancillary structures and flow direction of each pipeline, and the outlet information includes the outlet type, attributes, discharge method, river entry method, material condition, upstream interception well location, interception ratio and service range of each outlet.

[0021] Preferably, the engineering renovation plan includes one or more of the following: sponge city construction project, rainwater and sewage diversion renovation project, and interceptor well renovation within the service area of ​​the outfall.

[0022] On the other hand, a prediction system for the impact of discharge outlets on river water quality compliance under rainfall conditions is provided, including: The analysis module is used to analyze satellite imagery of the target area to obtain the underlying surface type within the target area, and to determine the pollutant runoff coefficient at each discharge outlet based on the underlying surface type. The data processing module is used to acquire pipeline information and outlet information within the target area, process the pipeline information and outlet information into standard data respectively, and calculate the amount of various runoff pollutants entering the river from each outlet based on the standard data and the pollutant runoff coefficient. The clustering module is used to calculate the comprehensive pollution load of each discharge outlet based on the amount of each runoff pollutant entering the river, and to classify the discharge outlets in the target area based on the comprehensive pollution load using a clustering algorithm to obtain the high-load discharge outlets. The prediction module is used to construct a river water quality prediction model. Based on the standard data, the amount of runoff pollutants entering the river at each discharge point, and water conservancy scheduling data, it predicts the water quality compliance status of each river assessment section under different rainfall scales, and outputs a prediction table of water quality compliance status for each river assessment section based on the prediction results. The river water quality prediction model is constructed from a river network hydrodynamic water quality model, a surface runoff model, and a surface and pipeline confluence model. The assessment module is used to construct a multi-factor pollution contribution assessment system. For river assessment sections whose water quality compliance prediction results are not met in the prediction table of each river assessment section, the pollution contribution score of the high-load discharge outlets corresponding to the river assessment section is calculated based on the comprehensive pollution load, and a ranking list of discharge outlets' pollution contribution to the river is generated based on the pollution contribution score. The design module is used to design engineering modification schemes for river assessment sections that do not meet the predicted standards, based on the list of the contribution of the discharge outlets to river pollution, with the goal of achieving river water quality standards under rainfall conditions, and to re-predict the water quality compliance status of the modified river.

[0023] By adopting the above technical solution, the beneficial effects of the present invention are as follows: 1. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions of this invention achieves accurate quantification of pollution source load by analyzing the underlying surface type and runoff coefficient, and combining pipeline network data to calculate the amount of pollutants entering the river; it improves the efficiency of subsequent modeling by classifying discharge outlets through clustering algorithms; it achieves dynamic and accurate prediction of water quality compliance by constructing a water quality prediction model containing multiple models; it achieves accurate identification of key pollution discharge outlets by scientifically quantifying the pollution contribution of each discharge outlet to the assessment section through a multi-factor evaluation system and generating a contribution ranking list; and it achieves scientific design and optimization of engineering measures guided by water quality compliance through simulation prediction of engineering modification schemes.

[0024] 2. The prediction method of the present invention for the impact of discharge outlets on river water quality compliance under rainfall conditions supports the simulation and prediction of engineering transformation measures such as sponge city construction and rainwater and sewage separation. It can seek the optimal transformation scheme with water quality compliance as the guide and combined with engineering costs, providing scientific, economic and feasible decision support for water environment management. Attached Figure Description

[0025] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0026] Figure 1 This is a schematic diagram of the method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions, according to an embodiment of the present invention. Detailed Implementation

[0027] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0028] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0029] Example 1 In one typical embodiment of this application, a method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions is provided, such as... Figure 1 As shown, it includes the following steps: S1. Analyze the satellite imagery of the target area to obtain the underlying surface type within the target area, and determine the pollutant runoff coefficient at each discharge outlet based on the underlying surface type.

[0030] Specifically, the catchment area of ​​the river to be studied was selected as the target area for quantification, and high-resolution satellite imagery and land use status maps of the area were obtained. Since the intensity of non-point source pollution load under rainfall conditions is closely related to the underlying surface type, in this embodiment, based on the high-resolution satellite imagery of the study area, ArcGIS was used to interpret and identify the images, and the underlying surface was divided into four levels: A, B, C, and D, according to the pollution load intensity. Category A: Low-pollution underlying surfaces such as non-urban construction land, park green space, and woodland, with a runoff coefficient of 0.2; Category B: Low-pollution underlying surfaces such as upscale residential communities, public buildings, and science parks, with a runoff coefficient of 0.3; Category C: Moderately polluted underlying surfaces such as general commercial areas, general residential areas, well-managed factory / industrial areas, and municipal roads, with a runoff coefficient of 0.6; Category D: Highly polluted underlying surfaces such as farmers' markets, livestock and poultry breeding and slaughterhouses, garbage transfer stations, food streets, auto repair shops, urban villages, and village-run industrial zones, with a runoff coefficient of 0.7.

[0031] Based on the service area of ​​each discharge outlet, one or more complete underlying surface blocks are identified for each outlet, and the underlying surface type and area of ​​each block are recorded. Under a certain rainfall intensity, the pollutants entering the river from each discharge outlet are determined according to the underlying surface type of its service area.

[0032] In this embodiment, a certain drainage outlet corresponds to a Class D underlying surface, serving farmers' markets and urban villages, with areas of 3500m² respectively. 2 and 300000m 2 The average concentrations of pollutants in the runoff at the discharge outlet were measured to be: COD 88 mg / L, ammonia nitrogen 23.1 mg / L, and TP 2.51 mg / L.

[0033] S2. Obtain pipeline and outlet information within the target area and process it into standard data. Calculate the amount of runoff pollutants entering the river from each outlet based on the standard data.

[0034] Specifically, pipeline information includes the category, attributes, burial depth, cross-sectional dimensions, pipe material, ancillary structures, and flow direction of each pipeline within the target area. Outlet information includes the outlet type, attributes, discharge method, river discharge method, material condition, upstream interception well location, interception ratio, and service area of ​​each outlet within the target area. In this embodiment, ArcGIS is used to standardize the pipeline and outlet information to form standard data. Based on this standard data, the runoff pollutant discharge from a single outlet is calculated using equation (1): (1) In the formula, M represents the amount of runoff pollutants entering the river as indicated by a certain index (such as COD, ammonia nitrogen, TP), in t / h; X represents the average concentration of runoff pollutants, in mg / L; q represents the rainfall intensity, in mm / h; α represents the runoff coefficient of the corresponding underlying surface; and A represents the area of ​​the corresponding underlying surface, in m². 2 .

[0035] In this embodiment, the inflow rates of COD, ammonia nitrogen, and total phosphorus at the discharge outlet were 5.36 t / h, 1.42 t / h, and 0.15 t / h, respectively.

[0036] S3. Based on the amount of various runoff pollutants entering the river at each discharge point, the comprehensive pollution load of each discharge point is calculated. Then, the discharge points in the target area are classified using hierarchical clustering agglomeration algorithm and K-means method to obtain the high-load discharge points.

[0037] The specific process is as follows: S3.1. Based on the COD inflow R1, ammonia nitrogen inflow R2, and total phosphorus (TP) inflow R3 obtained in step S2, the comprehensive pollution load R at the discharge outlet is calculated using the weighted scoring method and equation (3): R = W1*R1 + W2*R2 + W3*R3 (3) In this embodiment, W1=0.4, W2=0.4, and W3=0.2.

[0038] S3.2. Data preprocessing is performed using the Z-score standardization method.

[0039] In this embodiment, the service area of ​​the discharge outlet and the comprehensive pollution load are used as two core clustering factors. The service area of ​​the discharge outlet can be obtained by analyzing satellite imagery in step S1. However, due to the significant differences in the dimensions and orders of magnitude between the service area of ​​the discharge outlet and the comprehensive pollution load, direct clustering would be distorted. Therefore, standardization processing is performed first, specifically as follows: Data on the service area and comprehensive pollution load of all discharge outlets were extracted, and the data for each indicator were standardized according to equation (2): (2) Where Z represents the standardized data, X represents the original data, μ represents the mean of all data, and σ represents the standard deviation of all data.

[0040] After standardization, all data have a mean of 0 and a standard deviation of 1, placing them on the same scale and allowing them to be directly used for clustering.

[0041] S3.3. For the standardized data, hierarchical clustering agglomerative algorithm and K-means method are used for iterative classification to complete the final classification.

[0042] First, based on engineering experience in urban river outfall management, a priori strategy is adopted to directly set the number of outfall classifications to 3 to 5 categories. In this embodiment, a 5-category classification method is preferred.

[0043] Secondly, a hierarchical clustering agglomeration algorithm is used for preliminary clustering: each discharge outlet is regarded as an initial independent category. Based on the two dimensions of standardized service area and comprehensive pollution load, the Euclidean distance between any two discharge outlets is calculated, and the two closest discharge outlets are merged into a new category. This process is repeated until all discharge outlets are gradually aggregated to generate a clustering dendrogram.

[0044] Next, based on the clustering dendrogram obtained from hierarchical clustering, the K-means method is used for further optimization to make outlets of the same category more similar and outlets of different categories more different: K=5 is determined according to the prior strategy, and the preliminary cluster centers obtained from hierarchical clustering are used as the initial cluster centers for K-means. Each outlet is assigned to the nearest cluster center, forming a temporary classification, and the center point of each category is recalculated. The process of assigning and updating center points is repeated until the center points no longer change, at which point the iteration terminates, and the final outlet classification result is output. The five categories of outlets are: low-load small outlets, medium-load outlets, high-load large outlets, high-load small outlets, and low-load large outlets. Here, high, medium, and low describe the degree of pollution, and size describes the service area.

[0045] Based on the discharge outlet classification results, the pollution levels and spatial distribution of different types of discharge outlets can be characterized, which improves the efficiency of subsequent modeling.

[0046] S4. Construct a river water quality prediction model. Input standard data, pollutant inflow from each discharge point, and water conservancy scheduling data into the river water quality prediction model. Output the flow change process curves of each discharge point and the pollutant concentration time history curves of each river assessment section to obtain the water quality compliance status of each river assessment section. The river water quality prediction model mainly includes a non-point source pollution model and a river network hydrodynamic water quality model.

[0047] Specifically, regarding non-point source pollution, the Horton infiltration model was first used to analyze the surface infiltration process, and then the exponential function equations of equations (4) and (5) were used to simulate the pollutant accumulation and scouring process, respectively, to obtain the pollutant accumulation amount B and the pollutant scouring amount W.

[0048] (4) In the formula: B represents the cumulative amount of pollutants, in kg / hm². 2 C1 represents the total amount of pollutants accumulated on the surface before rainfall, and is the source of erosion materials; C2 is the maximum cumulative amount of pollutants, expressed in kg / hm². 2 C2 is the cumulative rate constant; t is the cumulative time, in minutes. Specifically, the maximum cumulative amount of pollutants, C1, represents the asymptotic upper limit of surface pollutant accumulation during the dry season. It is usually obtained in two ways: one is by field measurement, which involves sampling and weighing surface sediments of different land use types (road surfaces, roofs, green spaces, etc.) in the target catchment area, measuring the maximum dry weight of sediments per unit area and the pollutant content, and converting it into a pollutant equivalent load; the other is by referencing the empirical values ​​in the literature of similar cities and similar underlying surfaces when the field measurement conditions are limited. The cumulative rate constant, C2, reflects the speed at which pollutants accumulate on the surface. It can be obtained by collecting surface sediments after different dry season days (e.g., 1 day, 3 days, 7 days, 14 days), measuring the cumulative amount at each time period, and then substituting the measured data into formula (4) for regression fitting; or by referring to the typical value range given in model manuals such as SWMM and calibrating it in combination with local rainfall characteristics. The cumulative time t is the duration of the rainless period since the end of the last rainfall. In this embodiment, it is extracted from the hourly rainfall records of the weather station and is the time interval accumulated from the end time of the last rainfall event to the current time.

[0049] (5) In the formula: W is the pollutant flushing amount, in mg / L; S1 is the flushing coefficient; S2 is the flushing index; q1 is the runoff per unit area, in mm / h.

[0050] Specifically, the runoff per unit area q1 reflects the intensity of surface runoff generated by rainfall. It can be obtained by first acquiring the hourly rainfall intensity from the rain gauge, and then substituting it into the Horton infiltration model to calculate the surface infiltration rate. The surface runoff rate is obtained by subtracting the infiltration rate from the rainfall intensity, i.e., q1=if (i is the rainfall intensity, and f is the infiltration rate calculated by the Horton model). The scour coefficient S1 reflects the ability of runoff to scour surface pollutants and is a dimensionless calibration coefficient. The scour index S2 reflects the degree of nonlinear influence of runoff on scour volume and is generally taken between 1 and 3. In this embodiment, by setting up a flow and water quality synchronous monitoring device at the outlet of the target catchment area, the runoff and pollutant concentration at different times during the rainfall process are measured. B and q1 are substituted into formula (5), and the measured concentration process line is fitted by the least squares method. S1 and S2 are obtained by joint calibration. In addition, in other embodiments, the recommended values ​​for different land use types in the SWMM model manual can also be referred to, and S1 and S2 can be obtained by combining local measured data.

[0051] In terms of river network hydrodynamic and water quality simulation, the river network hydrodynamic and water quality model includes the river network hydrodynamic control equations represented by equations (6) and (7), including the continuity equation and the momentum equation. It is used to simulate the changes in flow, water level and velocity with time and space at different locations of the river under different rainfall, different pump station scheduling and different diversion and drainage conditions. It can truly reflect the hydraulic characteristics of the river when the water level rises, the flow increases and the velocity accelerates during the rainy season.

[0052] (6) (7) In the formula: A is the water flow area of ​​the river channel, in m². 2 Q represents the cross-sectional flow rate, in cubic meters per second (m³). 3 / s; t represents the lateral inflow velocity along the river channel in m / s; t represents time in seconds; x represents the horizontal coordinate along the flow direction in meters; and q represents the lateral inflow discharge in the river channel in cubic meters per second. 3 / s; is the momentum correction factor, dimensionless; g is the gravitational acceleration; y is the water level height, in meters. is the friction gradient coefficient, which is dimensionless.

[0053] Specifically, the river channel flow area A is determined by the river channel cross-sectional topographic data and the current water level. By measuring the cross-sectional shape of the river channel (using a depth sounder or cross-sectional measurement), a water level-area relationship curve (hA curve) is established. Given any water level, the corresponding river channel flow area A can be found. Under rainfall conditions, the river channel flow area A usually changes in real time with the water level h. The initial cross-sectional flow rate Q is given by the upstream boundary conditions (such as the measured flow process line). The cross-sectional flow rate Q at subsequent times can be obtained by solving equations (6) and (7) simultaneously and progressing through each time period. Time t is the simulation time step, which can be set automatically or manually, and is usually 1 min to 15 min. The horizontal coordinate x along the water flow direction reflects the spatially discrete node spacing of the river channel. It is determined after dividing the river channel into several calculation cross-sections, and the spacing is generally tens to hundreds of meters. The lateral inflow q represents the lateral inflow per unit river length, including inflow from drainage outlets on both banks, tributary inflow, and rainfall runoff inflow. The rainfall runoff inflow is calculated using the aforementioned Horton infiltration model and non-point source scour model, determining the outflow process of each catchment sub-region, and then distributed to the corresponding river segment according to the catchment relationship. Tributary and drainage outlet inflows can be provided by measured flow rates or pump station scheduling data. The initial water level y is given by the measured water level or initial water surface line. Upstream and downstream boundary conditions require water level process lines or flow-water level relationships (water level boundaries are provided by tide gauge stations or sluice gate scheduling, and flow boundaries are provided by measured upstream inflow). The momentum correction factor α reflects the non-uniformity of the cross-sectional velocity distribution; α is generally taken as 1.05~1.15, and is often approximated as 1.0 in calculations. Friction gradient S f The frictional resistance of the riverbed to the water flow is reflected by Manning's formula S. f =n 2 Q|Q| / (A 2 R 4 / 3The formula is obtained, where n is the channel roughness, which is determined by referring to the hydraulic handbook and taking the value according to the riverbed material; Q is the cross-sectional discharge; A is the channel surface area; and R is the hydraulic radius. The longitudinal velocity u of the lateral inflow is the component of the lateral inflow velocity along the channel direction. When the lateral inflow is perpendicular to the main channel, u≈0; when the lateral inflow has a longitudinal component (such as when a tributary obliquely converges), the longitudinal component u=v is calculated based on the inflow velocity and the angle of intersection. 侧 •cosθ. In most cases, we simplify and take u=0, so this term has a relatively small impact.

[0054] The river network hydrodynamic water quality model also includes the river network water quality control equation represented by equation (8), which is used to simulate the process of pollutants entering the river migrating, spreading, diluting and degrading with the water flow. Finally, it outputs the concentration of pollutants such as COD, ammonia nitrogen or total phosphorus at any cross section and at any time in the river. Combined with the river network hydrodynamic control equation, the two-way coupling relationship between hydrodynamic driving water quality and water quality responding to hydrodynamic can be obtained, making the simulation results closer to the real river channel change law.

[0055] (8) In the formula: c is the concentration of a certain pollutant; 1 represents the average flow velocity across the river cross section; E x K is the convective diffusion coefficient. c The attenuation coefficient of a certain pollutant, in units of d. -1 .

[0056] Specifically, the pollutant concentration c is the target variable for solving equation (8); the initial pollutant concentration c is given by the measured water quality concentration at each cross section (sampled and measured before the simulation starts), serving as the initial condition; subsequent times are obtained by solving the equation step by step. The average flow velocity u1 at the cross section is obtained by solving Q and A using formulas (6) and (7), and then calculated as u1=Q / A. u1 is not an independent input to this equation, but a real-time output of the hydrodynamic model. Each step of the calculation reads the current flow velocity at the cross section from the hydrodynamic model, i.e., the coupling relationship of "hydrodynamic driving water quality"—faster flow means faster pollutant transport; slower flow means pollutant retention. Longitudinal convection diffusion coefficient E x The ability of pollutants to diffuse and mix along the water flow direction, including the combined effects of molecular diffusion and turbulent diffusion, is often estimated using empirical formulas, such as Elder's formula E. x =5.93·d·u*, where d is the water depth and u* is the frictional velocity, u* = (gRS) f ) 1 / 2 Pollutant attenuation coefficient K c The Kc value reflects the rate at which pollutants are naturally reduced in water bodies through biodegradation, chemical oxidation, and sedimentation. Referencing literature on similar rivers, the Kc value for COD is generally taken as 0.05–0.35 dt. -1 Ammonia nitrogen should be taken as 0.05~0.20 d.-1 Total phosphorus was taken as 0.01~0.10 d. -1 .

[0057] In this embodiment, a river in a certain city is used as the research object, and a rainfall event lasting 6 hours with a total rainfall of 45 mm is selected for simulation. First, surface sediment samples were collected and weighed from the Class D underlying surface (farmers' market, urban village) in the target catchment area. The maximum cumulative pollutant C1 was measured to be COD 52.6 kg / hm², ammonia nitrogen 11.8 kg / hm², and TP 1.35 kg / hm². Surface sediment samples were collected after different rainless periods (1 day, 3 days, 7 days, 14 days) and regression was fitted to obtain the cumulative rate constant C2 as 0.0005 min. -1 The duration of the rainless period before the rain was extracted from the hourly rainfall records of the meteorological station and was t = 5 days (7200 min). Substituting the above values ​​into formula (4), the cumulative amount of pollutants B before the rain was calculated to be COD 51.2 kg / hm², ammonia nitrogen 11.5 kg / hm², and TP 1.31 kg / hm². Next, the hourly rainfall intensity was obtained from the rain gauge and substituted into the Horton infiltration model to calculate the peak rainfall period q1, which reached 28.3 mm / h. At the outlet of the target catchment area, flow and water quality synchronous monitoring equipment was set up to measure the runoff and pollutant concentration at different times during the rainfall process. The cumulative amount of pollutants B and the runoff per unit area q1 were substituted into formula (5), and the measured concentration process line was fitted by the least squares method. The scour coefficient S1 was determined to be COD 0.0053, ammonia nitrogen 0.0056, and TP 0.0070, and the scour index S2 was 1.7 for all of them. Substituting the above values ​​into formula (5), the pollutant flushing amount W during the peak rainfall period is calculated to be COD 80.0 mg / L, ammonia nitrogen 19.0 mg / L, and TP 2.7 mg / L.

[0058] In the simulation of river network hydrodynamics, a water level-area relationship curve (hA curve) was established by measuring the cross-sectional morphology of the river channel to determine the flow area A of each calculation section at different water levels. The initial water level y was given by the measured water surface line before the start of the simulation. The upstream boundary condition was taken as the measured flow process line (peak flow Q = 32.5 m³ / s), and the downstream boundary was taken as the tidal station water level process line. The Manning roughness n was taken as 0.035 according to the hydraulic calculation manual based on the riverbed material. The momentum correction coefficient α was taken as 1.08. The longitudinal velocity u of the lateral inflow was simplified to 0. The formulas (6) and (7) were solved simultaneously, and the calculation was carried out step by step to obtain the change process of the flow rate Q and water level y of each section with time and space. Among them, the test section A had the maximum flow rate of 38.7 m³ / s and the highest water level of 2.86 m 2 hours after the peak rainfall. The average velocity u1 = Q / A of the section reached a maximum of 0.62 m / s. In the river network water quality simulation, the initial concentration c is determined by actual water quality sampling at each cross-section before simulation, and the upstream boundary concentration is provided by W output from the non-point source scour model; the friction gradient S is calculated by back-calculation using the Manning formula. f =0.00016, and then the frictional velocity u is calculated. * =√(gRSf)=0.059m / s, longitudinal diffusion coefficient E x Using Elder's formula for estimation, and taking a water depth of d = 2.86m, E was calculated. x =5.93×2.86×0.059=1.00m² / s; the attenuation coefficient Kc is taken as COD 0.15d from the reference. -1 ammonia nitrogen 0.10d -1 TP0.05d -1 Substituting the above values ​​into formula (8) and solving it synchronously with the hydrodynamic equation, the time history curves of pollutant concentrations at each river assessment section were obtained. The final water quality compliance status of each river assessment section was as follows: At assessment section A, the COD concentration exceeded the Class IV standard limit for surface water (30 mg / L) within 3-8 hours after rainfall, with a peak concentration of 48.6 mg / L and an exceedance duration of 5 hours; ammonia nitrogen exceeded the Class IV standard limit (1.5 mg / L) within 2-6 hours after rainfall, with a peak concentration of 3.2 mg / L and an exceedance duration of 4 hours; TP exceeded the Class IV standard limit (0.3 mg / L) within 2-5 hours after rainfall, with a peak concentration of 0.52 mg / L and an exceedance duration of 3 hours. At assessment sections B and C, COD and ammonia nitrogen both met the standards during rainfall, while TP did not exceed the standard.

[0059] In this embodiment, the non-point source pollution model and the river network hydrodynamic and water quality model are coupled using an input-output method. Rainfall is first simulated and calculated by the non-point source pollution model to obtain the pollutant flow rate and concentration entering the river through the outlet after confluence. This is then used as a boundary condition and coupled into the aforementioned river network hydrodynamic and water quality model. The model incorporates the effects of upstream inflow, downstream tidal level (in the case of a tidal river network), and the shape of the river channel itself to calculate and output the curves showing the pollutant concentration changes over time at each cross-section of the river. To calibrate and verify the accuracy of the river water quality prediction model, the calculation results from the river network hydrodynamic and water quality model are imported into the non-point source pollution model for verification.

[0060] In this embodiment, the river network hydrodynamic and water quality model obtains the two-way coupling relationship between hydrodynamic driving water quality and water quality responding to hydrodynamics by simultaneously solving the river network hydrodynamic control equation and the river network water quality control equation, making the simulation results closer to the real river channel change patterns.

[0061] Among them, water conservancy scheduling data can be obtained from local water conservancy departments, including: the number of pump stations started and the operating flow under different rainfall scales, the water diversion and drainage flow of the river outside the river basin, and the normal water level and the control water level of the river during the rainy season. This can truly restore the actual hydraulic conditions of the river during the rainy season and avoid the disconnect between the river water quality model and reality.

[0062] In addition, to ensure the reliability of the prediction results, after the river water quality prediction model is constructed, parameter calibration is performed, specifically as follows: Historical hydrological monitoring data and historical water quality monitoring data of river assessment sections were retrieved. Data such as flow rate, water level, and concentrations of various pollutants at the assessment sections were compared with the model's concurrent calculations. Key model parameters, including pollutant degradation coefficients, convective diffusion coefficients, and lateral inflow velocity along the river channel, were adaptively adjusted. The model was considered validated and ready for formal prediction when the error between the calculated and actual concentrations did not exceed 15%.

[0063] After inputting standard data, pollutant inflow rates at each discharge point, and water conservancy scheduling data, the river water quality prediction model is used to perform time-period and cross-section-by-crossing calculations. The output includes flow change curves for each discharge point and pollutant concentration time-history curves for each river assessment section. Based on these curves, the highest and average pollutant concentrations at each assessment section during rainfall are analyzed and calculated. If the average pollutant concentration at all time points does not exceed the standard limit, the water quality at that assessment section is predicted to meet the standard. If the average pollutant concentration at any time point exceeds the standard limit, the water quality at that assessment section is determined to be substandard. Finally, a statistical table of water quality compliance at each river assessment section under different rainfall intensities is obtained.

[0064] S5. Construct a multi-factor pollution contribution assessment system, which includes pollution load factor, spatial attenuation factor, hydraulic dilution factor, and temporal synergy factor. Based on the multi-factor pollution contribution assessment system, rank all discharge outlets by pollution contribution and generate a ranking list of key discharge outlets' pollution contribution to the river.

[0065] Specifically, in real-world aquatic environments, even if a discharge outlet has a large total discharge volume, its contribution to exceeding the standards at the assessment section may not be significant if it is far from the assessment section or if the river has a large volume of water and strong dilution capacity at the time of discharge. Therefore, this embodiment introduces pollution load factor, spatial attenuation factor, hydraulic dilution factor, and temporal synergy factor to comprehensively evaluate the pollution contribution of each discharge outlet to the assessment section from four dimensions: pollution amount, spatial distance, hydraulic dilution, and temporal synchronicity.

[0066] Among them, the pollution load factor F1 is defined as the total amount of pollutants entering the river from the discharge outlet within a certain period of time. It directly reflects the scale of pollution discharge from the discharge outlet and is a basic indicator for assessing the contribution. Its data comes directly from the comprehensive pollution load of various runoff pollutants at each discharge outlet within a set period of time calculated in step S3.

[0067] The spatial attenuation factor F2 is inversely proportional to the product of the river channel distance from the discharge outlet to the assessment section and the velocity attenuation coefficient. After being discharged from the outlet, pollutants need to migrate with the water flow to the assessment section, during which physical diffusion and degradation occur. Generally speaking, the farther the discharge outlet is from the assessment section, the more significant the attenuation of pollutants during migration. At the same time, the water flow velocity also affects the mass transfer efficiency of pollutants; the faster the flow velocity, the less the pollutant attenuates, and the higher its contribution to pollution at the river channel assessment section. This embodiment uses the flow field data output by the model to quantify the spatial distance and hydraulic connection between the discharge outlet and the assessment section, thereby calculating this factor.

[0068] The hydraulic dilution factor F3 is defined as the ratio of the river flow rate at the discharge outlet to the discharge flow rate at the outlet. This factor reflects the river's immediate dilution capacity for pollutants. When the river flow rate is much greater than the discharge flow rate, the pollutants are rapidly diluted, and the peak concentration decreases. Conversely, if the river flow rate is small, a high-concentration pollution plume is more likely to form. The data for river flow rate and discharge flow rate are obtained from the flow rate change curves of each discharge outlet and the pollutant concentration time history curves of each river assessment section output from the river water quality prediction model in step S4.

[0069] The temporal synergy factor is used to analyze the temporal matching degree between the peak pollution level at the discharge outlet and the peak river flow, as well as the period of exceedance at the assessment section, i.e., whether the peak pollution level at the discharge outlet and the period of exceedance at the river assessment section occur simultaneously. During rainfall, the peak discharge level at the discharge outlet and the peak river flow often occur asynchronously. If the peak discharge level at the discharge outlet occurs during a period of low river flow and weak self-purification capacity, its threat to water quality compliance will increase significantly. In this embodiment, the temporal synergy factor F4 is defined as the ratio of the overlap time between the duration of the peak pollution level at the discharge outlet and the period of exceedance at the river assessment section to the total duration of exceedance at the section, thereby quantifying the temporal synergy effect.

[0070] Based on the four factors mentioned above, this embodiment uses a weighted comprehensive evaluation method to rank the pollution contribution of all discharge outlets. In practice, each factor is first normalized to eliminate the influence of dimensions, and then the weight coefficients of each factor are set according to the actual water environment characteristics. For example, for rivers with poor flow, the spatial attenuation factor has a weight as high as 0.4, while the pollution load factor, hydraulic dilution factor, and temporal synergy factor all have weights of 0.2. Finally, the pollution contribution score of each discharge outlet is calculated, and all discharge outlets are automatically ranked from highest to lowest according to their pollution contribution scores, generating a ranking list of key pollution discharge outlets' pollution contribution to the river. The top 20% of discharge outlets are marked as key pollution discharge outlets. This list includes at least the discharge outlet number, location coordinates, pollution contribution score, ranking, main influencing factors, and whether it is the main source of the section exceeding the standard. This visually demonstrates which discharge outlets are the key factors causing the assessment section to exceed the standard, providing a precise targeting basis for the subsequent step S6 to design targeted engineering modification schemes, avoiding blind treatment and significantly improving treatment efficiency.

[0071] In this embodiment, based on the calculation results of the river water quality prediction model in S4, a multi-factor pollution contribution assessment is performed on each discharge outlet. Taking the assessment section A as an example, 4 hours after the peak rainfall (the period of most severe exceedance) is selected as the assessment time. The pollution load factor is obtained by normalizing the comprehensive pollution load R of each discharge outlet in S2; the spatial decay factor is calculated according to the river distance from each discharge outlet to the assessment section A using an exponential decay model; the hydraulic dilution factor is determined according to the ratio of the inflow of each discharge outlet to the river section flow; and the temporal synergy factor is determined according to the degree of overlap between the time when the flow from each discharge outlet reaches section A and the period of exceedance. Multiplying the four factors, a multi-factor pollution contribution assessment system is obtained. The pollution contribution of all discharge outlets is ranked, and the following list of key discharge outlets' pollution contribution to the assessment section A is obtained, as shown in Table 1.

[0072] Table 1

[0073] The above ranking results indicate that the pollution contribution of outlet DP-03 to the assessment section A is much higher than that of other outlets. It is the primary pollution source causing the water quality of section A to exceed the standard during rainfall and should be given priority in the scope of interception and source control.

[0074] S6. For rivers where the prediction results in step S4 do not meet the standards, based on the key pollution outlets in S5, design an engineering renovation plan with the goal of achieving the water quality standard under rainfall conditions, and then re-predict the water quality standard of the renovated rivers.

[0075] Specifically, the engineering renovation plan includes one or more of the following: sponge city construction within the service area of ​​the outfall, rainwater and sewage separation renovation project, and intercepting well renovation. The implementation logic of these measures in the river water quality prediction model is different, but in essence, they all affect the output results by changing the input data or model parameters.

[0076] The main function of sponge city construction measures is to regulate, intercept, and purify rainwater runoff. In model simulation, this is mainly reflected in the adjustment of the underlying surface type and its runoff coefficient. For example, if a large area of ​​urban villages (Class D underlying surface, runoff coefficient 0.7) exists within the service area of ​​a key outfall, the plan is to renovate them through demolition and reconstruction or by adding sponge facilities such as rain gardens and permeable concrete. During the simulation, the system adjusts the underlying surface type parameter of this area from Class D to Class B or Class C, and the corresponding runoff coefficient decreases from 0.7 to 0.4~0.5. This parameter change can directly reduce the total surface runoff and pollutant scouring, thereby reducing the calculated value of pollutant weight M entering the river at the source. The system then re-runs steps S2 to S4 to evaluate the contribution of the renovation plan to achieving water quality standards.

[0077] The core objective of stormwater and sewage separation projects is to intercept sewage that would otherwise flow mixed into rivers and redirect it to sewage treatment plants. In model parameter adjustments, this is reflected in corrections to the service area of ​​discharge outlets or the concentration of pollutants discharged. For areas where stormwater and sewage separation has been completed, the model either excludes pollutants generated in that area from the total amount entering the river at that outlet, or significantly reduces the concentration parameter entering the river based on interception efficiency. In this way, the model can simulate the reduction rate of pollution load entering the river after the separation project, thereby predicting the degree of improvement in river water quality.

[0078] Interception well modifications primarily adjust rainwater overflow by changing the interception ratio. The interception ratio is a key parameter affecting overflow pollution in combined sewer systems. In the model, by modifying the interception ratio parameter of the upstream interception well, the ratio of water intercepted to wastewater treatment plants to water directly overflowing into rivers under different rainfall intensities can be dynamically simulated. For example, increasing the interception ratio from 1 to 3 times will result in the model calculating that more mixed wastewater is intercepted, significantly changing the overflow flow rate process curve at the outlet, with a lower peak value and a smaller total volume. Based on this, the system recalculates the amount of pollutants entering the river and predicts water quality compliance.

[0079] After adjusting the parameters of any one or more of the aforementioned engineering measures, the system executes a re-prediction step to ensure river water quality meets standards. The system re-inputs the adjusted data into the river water quality prediction model, runs the simulation calculation in step S4 again, and outputs a new time-history curve of pollutant concentration at the assessment section. If the prediction results show that the water quality meets standards, it proves that the engineering modification scheme is economically and technically feasible; if it still does not meet standards, the scheme can be further adjusted, such as increasing the area of ​​sponge facilities or increasing the interception ratio, until the simulation results meet the water quality compliance requirements. Through the iterative process of scheme design, parameter adjustment, and re-simulation verification, the system ensures that the finally selected engineering modification scheme has a scientific basis and achieves the goal of meeting river water quality standards at the most reasonable engineering cost.

[0080] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions, characterized in that, include: S1. Analyze the satellite image of the target area to obtain the underlying surface type within the target area, and determine the pollutant runoff coefficient at each discharge outlet based on the underlying surface type; S2. Obtain pipeline information and outlet information within the target area, process the pipeline information and outlet information into standard data respectively, and calculate the amount of various runoff pollutants entering the river from each outlet based on the standard data and the pollutant runoff coefficient. S3. Based on the amount of each runoff pollutant entering the river, calculate the comprehensive pollution load of each discharge outlet. Based on the comprehensive pollution load, use a clustering algorithm to classify the discharge outlets in the target area to obtain the high-load discharge outlets. S4. Construct a river water quality prediction model. Based on the standard data, the amount of various runoff pollutants entering the river from each discharge outlet, and water conservancy scheduling data, predict the water quality compliance status of each river assessment section under different rainfall scales, and output a prediction table of water quality compliance status of each river assessment section based on the prediction results. The river water quality prediction model includes a non-point source pollution model and a river network hydrodynamic water quality model. S5. Construct a multi-factor pollution contribution assessment system. For river assessment sections whose water quality compliance prediction results are not met in the prediction table of each river assessment section, calculate the pollution contribution score of the high-load discharge outlets corresponding to the river assessment section based on the comprehensive pollution load, and generate a ranking list of discharge outlets' pollution contribution to the river based on the pollution contribution score. S6. Based on the ranking list of the contribution of the discharge outlets to river pollution, with the goal of achieving the standard of river water quality under rainfall conditions, design engineering transformation schemes for the river assessment sections where the predicted results do not meet the standards, and re-predict the water quality compliance status of the transformed river.

2. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The calculation formula for the amount of runoff pollutants entering the river at each discharge point based on the standard data is as follows: Where M is the amount of runoff pollutants entering the river, in t / h; X is the average concentration of runoff pollutants, in mg / L; q is the rainfall intensity, in mm / h; α is the runoff coefficient of the corresponding underlying surface; and A is the area of ​​the corresponding underlying surface, in m². 2 .

3. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The specific process of step S3 includes: Using the amount of each runoff pollutant entering the river as an indicator, a weighted scoring method is used to calculate the comprehensive pollution load of each discharge outlet. Using the service area of ​​each discharge outlet and the comprehensive pollution load as the core clustering factors, a hierarchical clustering agglomeration algorithm is used to initially cluster the discharge outlets in the target area. Then, the K-means method is used for iteration to obtain the discharge outlet classification results. The discharge outlet types include at least low-load small discharge outlets, medium-load discharge outlets, high-load large discharge outlets, and high-load small discharge outlets.

4. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The prediction of water quality compliance at various river assessment sections under different rainfall scales, based on the standard data, the amount of runoff pollutants entering the river at each discharge point, and water conservancy scheduling data, specifically includes: The standard data, the amount of runoff pollutants entering the river at each discharge point, and the water conservancy scheduling data are input into the river water quality prediction model, and the flow change process curves of each discharge point and the pollutant concentration time history curves of each river assessment section are output. Based on the flow rate change curve and the pollutant concentration time history curve, the average pollutant concentration at each river assessment section during each rainfall period was calculated. If the average concentration of pollutants in all time periods does not exceed the standard limit, the water quality of the river section under assessment is predicted to meet the standard. If the average concentration of pollutants exceeds the standard limit at any time period, the water quality of the river section under assessment is predicted to be substandard. Based on the prediction results, a prediction table of water quality compliance status for each river assessment section under different rainfall scales is generated.

5. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The non-point source pollution model includes: In the formula, B represents the cumulative amount of pollutants, expressed in kg / hm². 2 C1 represents the maximum cumulative amount of pollutants, expressed in kg / hm². 2 C2 is the cumulative rate constant; t is the cumulative time in minutes; W is the pollutant flushing rate in mg / L; S1 is the flushing coefficient; S2 is the flushing index; q1 is the runoff per unit area in mm / h; B is the pollutant accumulation in kg / hm². 2 .

6. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The river network hydrodynamic and water quality model includes: Where A is the river channel area; Q is the cross-sectional discharge. t represents the velocity of the lateral incoming water in the direction of the river channel; x represents the horizontal coordinate along the direction of water flow; and q represents the lateral inflow discharge of the river channel. y is the momentum correction factor; g is the gravitational acceleration; y is the water level height; denoted as frictional gradient; c represents the concentration of pollutants in the runoff. 1 represents the average flow velocity across the river cross section; E x K is the convective diffusion coefficient. c t represents the attenuation coefficient of runoff pollutants; t represents time.

7. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The constructed multi-factor pollution contribution assessment system, for river assessment sections whose predicted water quality compliance status in the river assessment section prediction table is not met, calculates the pollution contribution score of all discharge outlets corresponding to that river assessment section, specifically including: Based on the comprehensive pollution load, a multi-factor pollution contribution assessment system including pollution load factor, spatial attenuation factor, hydraulic dilution factor and temporal synergy factor is constructed and weighted to obtain the pollution contribution score of each discharge outlet.

8. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The specific process for generating a ranking list of discharge outlets' pollution contribution to the river channel based on the pollution contribution score is as follows: All discharge outlets are sorted in descending order of their pollution contribution scores to generate a ranking list of their pollution contribution to the river. A predetermined proportion of discharge outlets are then selected and marked as critical pollution discharge outlets.

9. The method for predicting the impact of discharge outlets on river water quality compliance under rainfall conditions as described in claim 1, characterized in that, The pipeline information includes the category, attributes, burial depth, cross-sectional dimensions, pipe material, ancillary structures and flow direction of each pipeline. The outlet information includes the outlet type, attributes, discharge method, river entry method, material condition, upstream interception well location, interception ratio and service area of ​​each outlet.

10. A prediction system for the impact of discharge outlets on river water quality compliance under rainfall conditions, characterized in that, include: The analysis module is used to analyze the underlying surface type within the target area and determine the runoff pollutants at each discharge outlet based on the underlying surface type. The data processing module is used to acquire pipeline information and outlet information within the target area and process them into standard data, and calculate the amount of runoff pollutants entering the river at each outlet based on the standard data. The clustering module is used to classify the discharge outlets in the target area according to the amount of runoff pollutants entering the river at each discharge outlet using a clustering algorithm, and to perform a weighted calculation on the clustered discharge outlets to obtain the comprehensive pollution load, wherein the weighting factors include the runoff pollutant concentration and the discharge flow rate. The prediction module is used to construct a river water quality prediction model. Based on the standard data, the amount of runoff pollutants entering the river at each discharge point, and water conservancy scheduling data, it predicts the water quality compliance status of each river assessment section under different rainfall scales and outputs a prediction table of water quality compliance status for each river assessment section. The river water quality prediction model includes a river network hydrodynamic water quality model, a surface runoff model, and a surface and pipeline confluence model. The assessment module is used to construct a multi-factor pollution contribution assessment system. For river assessment sections whose water quality compliance prediction results are not met in the prediction table of each river assessment section, the module calculates the pollution contribution score of all discharge outlets corresponding to the river assessment section and generates a ranking list of discharge outlets' pollution contribution to the river based on the pollution contribution scores. The design module is used to design engineering modification schemes for river assessment sections that do not meet the predicted standards, based on the ranking list of the contribution of the discharge outlets to river pollution, with the goal of achieving river water quality standards under rainfall conditions, and to re-predict the water quality compliance status of the modified river.

Images