Urban water body runoff control method and system

Abstract

The application provides a kind of urban water body runoff control method and system, it is related to urban water body runoff control technical field, comprising: obtaining historical rainfall data, real-time data, modeling parameter and urban drainage system parameter;Characteristic information is obtained by rainfall frequency analysis processing based on the preset bayesian inference mathematical model to historical rainfall data;According to characteristic information and rainwater pipe network layout map, regional division processing is carried out to obtain segmentation result;According to segmentation result, rainfall prediction model is constructed to obtain prediction change result;According to prediction result and rainwater pipe network layout map, simulation processing is carried out to obtain simulation result;Based on simulation result, initial control scheme is optimized to obtain urban water body runoff dynamic control scheme.The application comprehensively considers topography, building and soil permeability characteristics of multiple factors, adopts graph embedding algorithm and multi-objective optimization method, and comprehensively analyzes and optimizes urban drainage system, improves the overall efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of urban water body runoff control technology, and more specifically, to a method and system for controlling urban water body runoff. Background Technology

[0002] With the accelerating pace of urbanization, the design and management of urban drainage systems have become increasingly important. Urban runoff control plays a crucial role in preventing floods, reducing water pollution, and protecting urban infrastructure. However, current urban drainage system management technologies suffer from numerous problems. Existing technologies commonly design and evaluate drainage systems based on empirical formulas and simple statistical analyses. For example, conventional rainfall frequency analysis methods are used to predict rainfall intensity and duration, which are then simulated using drainage network models. However, these methods have significant drawbacks: firstly, they cannot effectively utilize real-time data to dynamically adjust the operation of the drainage system; secondly, traditional models cannot accurately reflect changes in complex urban terrain, building density, and soil permeability, leading to significant discrepancies between simulation results and actual conditions.

[0003] Based on the shortcomings of the existing technologies, there is an urgent need for a method and system for controlling urban water runoff. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for controlling urban water runoff, in order to improve the aforementioned problems. To achieve this objective, the technical solution adopted by this invention is as follows:

[0005] Firstly, this application provides a method for controlling urban water body runoff, including:

[0006] The system acquires historical rainfall data, real-time data, modeling parameters, and urban drainage system parameters. The real-time data includes real-time rainfall data and flow monitoring data. The modeling parameters include urban topographic maps, building models, and soil property distribution maps. The urban drainage system includes stormwater pipe network layout maps and initial control schemes.

[0007] Based on a pre-set Bayesian inference mathematical model, the historical rainfall data is processed by rainfall frequency analysis to obtain feature information, which includes urban rainfall characteristics and trends.

[0008] Based on the feature information and the rainwater pipe network layout map, the area is divided into at least one drainage zone based on the drainage characteristics to obtain the segmentation result.

[0009] A rainfall prediction model is constructed based on the segmentation results, and the real-time data is used as the input value of the rainfall prediction model to calculate the predicted change results, which include the changes in rainfall and flow rate within a preset time period in the future.

[0010] Based on the prediction results and the stormwater pipe network layout diagram, simulation processing is performed. The urban drainage system is modeled as a directed graph structure, and the node pressure, pipe flow and overflow risk under different scenarios are analyzed to obtain simulation results.

[0011] Based on the simulation results, the initial control scheme is optimized to obtain a dynamic control scheme for urban water body runoff.

[0012] Secondly, this application also provides a water body runoff control system for cities, including:

[0013] The acquisition module is used to acquire historical rainfall data, real-time data, modeling parameters, and urban drainage system parameters. The real-time data includes real-time rainfall data and flow monitoring data. The modeling parameters include urban topographic maps, building models, and soil property distribution maps. The urban drainage system includes stormwater pipe network layout maps and initial control schemes.

[0014] The analysis module performs rainfall frequency analysis on the historical rainfall data based on a preset Bayesian inference mathematical model to obtain feature information, which includes urban rainfall characteristics and trends.

[0015] The segmentation module is used to perform regional segmentation processing based on the feature information and the rainwater pipe network layout map, and to divide the city into at least one drainage zone based on drainage characteristics to obtain the segmentation result.

[0016] The prediction module is used to construct a rainfall prediction model based on the segmentation results, and to use the real-time data as the input value of the rainfall prediction model to calculate the predicted change results, which include the changes in rainfall and flow rate within a preset future time period.

[0017] The simulation module is used to perform simulation processing based on the prediction results and the stormwater pipe network layout diagram. It obtains simulation results by modeling the urban drainage system as a directed graph structure and analyzing the node pressure, pipe flow and overflow risk under different scenarios.

[0018] The optimization module optimizes the initial control scheme based on the simulation results to obtain a dynamic control scheme for urban water body runoff.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention comprehensively analyzes and optimizes urban drainage systems by taking into account various factors such as topography, buildings, and soil permeability characteristics, and employing graph embedding algorithms and multi-objective optimization methods, thereby improving the overall efficiency of the system.

[0021] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the urban water body runoff control method described in an embodiment of the present invention;

[0024] Figure 2 This is a flowchart illustrating the hierarchical clustering process. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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 some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. 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.

[0026] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0027] Example 1:

[0028] This embodiment provides a method for controlling urban water runoff.

[0029] See Figure 1 The figure shows that the method includes steps S100 to S600.

[0030] Step S100: Obtain historical rainfall data, real-time data, modeling parameters and urban drainage system parameters. Real-time data includes real-time rainfall data and flow monitoring data. Modeling parameters include urban topographic map, building model and soil property distribution map. Urban drainage system includes rainwater pipe network layout map and initial control scheme.

[0031] Understandably, historical rainfall data can reveal urban rainfall patterns under different seasons and climatic conditions, providing a foundation for subsequent rainfall frequency analysis. Real-time data, including real-time rainfall data and flow monitoring data, dynamically reflects current rainfall conditions and the operational status of the drainage system. This data is collected in real-time by automated monitoring devices such as rain gauges and flow meters and transmitted to the central control system via IoT technology to ensure timeliness and accuracy. Urban topographic maps and building models in the modeling parameters are constructed using remote sensing technology and Geographic Information Systems (GIS). This high-precision spatial data provides detailed descriptions of the city's topographic features and building layout, reflecting the impact of the urban surface on stormwater runoff. Soil property distribution maps are obtained through field sampling and laboratory analysis, focusing on soil permeability, saturation, and water content, which directly affect stormwater infiltration and runoff velocity. Stormwater pipe network layout maps and initial control schemes are derived from urban planning department and drainage system design drawings. The stormwater pipe network layout map details the routes, diameters, and materials of drainage pipes, while the initial control scheme includes operational strategies and scheduling rules for existing drainage facilities. This information is integrated into a comprehensive database, providing complete foundational data for subsequent model building and simulation.

[0032] Step S200: Based on the preset Bayesian inference mathematical model, perform rainfall frequency analysis on historical rainfall data to obtain feature information, including urban rainfall characteristics and trends;

[0033] Understandably, Bayesian inference provides a systematic approach that combines prior knowledge with actual data to obtain more accurate and reliable rainfall characteristic information. This characteristic information includes rainfall distribution (such as average rainfall, standard deviation, and extreme values), rainfall event frequency (such as the frequency and time interval distribution of rainfall events), rainfall duration (such as average rainfall duration and its fluctuations), rainfall intensity (such as the distribution of rainfall intensity and extreme rainfall intensity), seasonal variations (such as rainfall patterns in different seasons and monthly or quarterly rainfall), and long-term trends (such as the trend of rainfall changes over time and the impact of climate change). Rainfall characteristic information can comprehensively reveal the patterns and characteristics of urban rainfall, providing a scientific basis for urban drainage system design, extreme weather response, and flood control measures, thereby improving urban flood control capabilities and infrastructure safety.

[0034] Step S300: Perform regional division processing based on feature information and rainwater pipe network layout map, and divide the city into at least one drainage zone based on drainage characteristics to obtain the segmentation result;

[0035] It's important to note that the characteristic information includes the frequency, intensity, and duration of urban rainfall, reflecting the response of different areas to rainfall events. This information can be used to identify areas in the city prone to waterlogging and flooding. The stormwater drainage network layout map provides detailed information about the city's drainage system, including the path, diameter, and material of drainage pipes. Combining this information with other data allows for a detailed understanding of the drainage capacity of each area and the distribution of existing drainage facilities. Analyzing this data helps determine the drainage characteristics of each area; for example, some areas may have low drainage efficiency due to aging pipe networks or inadequate design, while other areas may be prone to waterlogging due to their low-lying terrain. When dividing areas, the city's topography, building distribution, and soil permeability must be considered. Topographic data helps determine the natural flow direction and convergence points of rainwater, building models reflect the obstruction of rainwater flow by buildings, and soil property distribution maps provide information on the permeability of soil in different areas. These factors combined determine the drainage characteristics of each area. By comprehensively analyzing this information, a reasonable urban zoning can be achieved. For example, high-density building areas and low-density residential areas can be divided into different drainage zones due to their different drainage needs and characteristics. Meanwhile, low-lying areas prone to waterlogging require special attention. Strengthening drainage facilities and management in these areas can effectively reduce the risk of urban flooding. This step, through zoning, enables precise management of the urban drainage system. The resulting drainage zones better reflect the specific drainage needs and capacities of each area, thus allowing for the development of more effective drainage management strategies.

[0036] Step S400: Construct a rainfall prediction model based on the segmentation results, and use real-time data as the input value of the rainfall prediction model to calculate the predicted change results. The predicted results include the changes in rainfall and flow rate within a preset time period in the future.

[0037] Understandably, this step, by building and running a rainfall prediction model, enables high-precision predictions of future rainfall and flow changes, providing a scientific basis for the dynamic management of urban drainage systems. The introduction of real-time data improves the model's response speed and accuracy, allowing cities to take preventative measures earlier and reduce the risk of urban flooding and waterlogging disasters. Simultaneously, the prediction results can help optimize drainage system operation strategies, improve overall drainage efficiency, and ensure the safety of urban residents and the stable operation of urban infrastructure.

[0038] Step S500: Based on the prediction results and the stormwater pipe network layout diagram, simulation processing is performed. The urban drainage system is modeled as a directed graph structure, and the node pressure, pipe flow and overflow risk under different scenarios are analyzed to obtain the simulation results.

[0039] Understandably, the first step is to construct a directed graph model of the urban drainage system based on rainfall forecasts and stormwater pipe network layout diagrams. Nodes in the directed graph structure represent key points in the drainage system, such as drainage wells, stormwater inlets, and pipe junctions, while edges represent the drainage pipes connecting these nodes. By assigning appropriate weights to nodes and edges—where the node weight represents its flow demand or pressure, and the edge weight represents the pipe's flow capacity and pressure bearing capacity—the characteristics of the actual drainage system can be more realistically reflected. Under simulated scenarios with different rainfall intensities and durations, the pressure changes at each node and the flow distribution in the pipes under these scenarios are calculated to identify potential overflow risk areas.

[0040] Step S600: Based on the simulation results, optimize the initial control scheme to obtain a dynamic control scheme for urban water body runoff.

[0041] Understandably, optimization first requires evaluating the initial control scheme and identifying its shortcomings. For example, some drainage pipes are prone to overload under high rainfall scenarios, or certain nodes may experience overflow risks under specific conditions. By identifying these problems, targeted improvement measures can be proposed. The key to optimization lies in using data from simulation results, employing multi-objective optimization algorithms and machine learning techniques to find the optimal combination of control measures. These measures include adjusting the flow distribution of drainage pipes, optimizing drainage paths, increasing the capacity of reservoirs, or deploying temporary drainage equipment in high-risk areas. The optimized dynamic control scheme needs to be able to respond in real time to changes in actual conditions. Therefore, it is also necessary to combine real-time monitoring data and use feedback control mechanisms to continuously adjust and optimize the control strategy. In this way, the urban drainage system can flexibly respond to different rainfall intensities and durations, ensuring the efficient operation of the drainage system.

[0042] Further, step S200 includes steps S210 to S240:

[0043] Step S210: Based on historical rainfall data, perform data processing, and obtain the rainfall dataset by filling in missing values ​​and removing outliers;

[0044] Step S220: Perform parameter estimation processing based on the rainfall dataset. Calculate the rainfall amount, duration, and interval for each rainfall event using the maximum likelihood estimation method to obtain a preliminary set of rainfall event feature parameters.

[0045] Specifically, we first define the basic elements of a rainfall event, where rainfall amount refers to the total precipitation measured within a specific time period, rainfall duration is the length of time from the start to the end of the event, and rainfall interval is the time interval between two rainfall events. Then, we make probability distribution assumptions for rainfall amount, duration, and interval. We assume that rainfall amount follows a gamma distribution with parameter θ1, duration follows a normal distribution with parameter θ2, and interval follows an exponential distribution with parameter θ3. Based on these assumptions, we construct the likelihood function L(θ), representing the joint probability of observing these rainfall events given parameter θ. For independent and identically distributed samples, the likelihood function is expressed as the product of the probability density functions of all observed data. The formula is as follows:

[0046]

[0047] Where L(θ1,θ2,θ3) represents the likelihood function; θ1, θ2, and θ3 represent the distribution parameters of rainfall amount, rainfall duration, and rainfall interval, respectively; i represents the sequence number of the rainfall event; z i This represents the total number of rainfall events; x i y represents the rainfall amount of the i-th rainfall event; i z represents the duration of the i-th rainfall event; i P(x) represents the interval between the i-th rainfall events; i |θ1), P(y i |θ2) and P(z) i |θ3) represents the probability density functions of rainfall, rainfall duration, and rainfall interval, respectively, given parameters θ1, θ2, and θ3.

[0048] Next, the optimal parameter estimates are found by maximizing the logarithmic form of the likelihood function (log-likelihood function). Preferably, this process is implemented using numerical optimization algorithms, such as gradient descent or the Newton-Raphson method. The calculation formula is as follows:

[0049]

[0050] in, and These represent the optimal parameter estimates for rainfall, rainfall duration, and rainfall interval, respectively.

[0051] By using maximum likelihood estimation, the most probable parameter values ​​for rainfall amount, duration, and interval for each rainfall event can be estimated. Ultimately, a preliminary set of characteristic parameters for rainfall events is generated, containing the statistical characteristics of all rainfall events, providing foundational data for further Bayesian inference and rainfall frequency analysis.

[0052] Step S230: Perform Bayesian inference processing based on the preliminary rainfall event feature parameter set. By setting the gamma distribution as the prior distribution and updating the posterior distribution using historical rainfall data, the posterior distribution results of each rainfall event parameter are obtained.

[0053] Specifically, the basic idea of ​​Bayesian inference is to update our understanding of parameters by combining prior information and observational data. In this embodiment, the gamma distribution is chosen as the prior distribution for the rainfall parameter because it effectively describes the positively skewed distribution characteristics of rainfall. The probability density function of the gamma distribution is defined as:

[0054]

[0055] Where α and β are the shape and scale parameters of the gamma distribution; θ represents the parameters of the rainfall event; P(θ|α,β) is the probability density function of parameter θ, describing the prior knowledge of parameter θ before observation data is available; Γ(α) represents the gamma function; and e represents the natural constant.

[0056] Next, we will perform likelihood estimation using historical rainfall data. The likelihood function is defined as follows:

[0057]

[0058] Where D = {x1, x2, ..., x} n} represents the observed historical rainfall dataset; i represents the sequence number of the rainfall event; x i Let P(x) represent the i-th rainfall event; n represents the total number of rainfall events; i |θ) represents the observed rainfall event x under parameter θ. i The probability density function.

[0059] In Bayesian inference, it is necessary to calculate the posterior distribution P(x). i |θ), which is defined according to Bayes' theorem as:

[0060]

[0061] The posterior distribution can be calculated as follows:

[0062] P(θ|D)∝P(D|θ)P(θ);

[0063] Where P(θ|D) represents the likelihood function; P(θ) represents the prior distribution; and P(D) represents the marginal likelihood of the observed data.

[0064] Furthermore, using the gamma distribution as the prior distribution, the posterior distribution obtained after updating with historical rainfall data is still in the form of a gamma distribution, but its parameters α and β will be updated based on the observed data. The updated parameters α' and β' can be calculated using the following formulas:

[0065] α′=α+n;

[0066]

[0067] Where i represents the sequence number of the rainfall event; x i Let represent the i-th rainfall event; n represents the total number of rainfall events; α and β are the shape and scale parameters of the gamma distribution; α' and β' represent the updated shape and scale parameters.

[0068] Bayesian inference processing allows for more accurate estimation of rainfall event parameters, improving the precision of rainfall prediction models. Posterior distributions provide information on parameter uncertainties, enhancing the reliability and scientific rigor of predictions and providing a more solid data foundation for the design and management of urban drainage systems.

[0069] Step S240: Perform rainfall frequency analysis based on the posterior distribution results. By simulating rainfall events, the frequency distribution of rainfall amount, rainfall duration, and rainfall interval is statistically analyzed to obtain feature information.

[0070] Understandably, this step uses the Monte Carlo method to randomly sample from the posterior distribution to generate rainfall events. Through extensive repeated random sampling, possible rainfall scenarios are simulated. Specifically, a certain number of samples are drawn from each posterior distribution, and the generated sample data are combined to form a complete set of rainfall events. Then, statistical analysis is performed on the simulated rainfall event data to calculate the frequency distributions of rainfall amount, rainfall duration, and rainfall interval. Furthermore, by plotting histograms or cumulative distribution functions, the distribution of these parameters is displayed, thereby obtaining characteristic information about rainfall events, including the common range of rainfall amount, the typical length of rainfall duration, and the interval between rainfall events. This method allows for a detailed understanding of the characteristics and patterns of urban rainfall events, providing reliable data support for subsequent rainfall trend prediction and urban drainage system optimization. This not only enhances the understanding of urban rainfall patterns but also provides a scientific basis for flood control, disaster reduction, and drainage system optimization, improving the city's ability to cope with extreme weather events.

[0071] Example 2

[0072] The only difference between this embodiment and Embodiment 1 is that step S300 includes steps S310 to S330:

[0073] Step S310: Perform runoff simulation processing based on feature information and modeling parameters. By simulating rainfall events with different rainfall intensities and durations, analyze the runoff collection, diversion, and water accumulation areas to obtain urban drainage feature vectors.

[0074] Further, step S310 includes steps S311 to S314:

[0075] Step S311: Perform terrain modeling processing based on the city topographic map and building model, and generate static terrain modeling data containing terrain features and building distribution through a multi-scale terrain modeling algorithm;

[0076] Understandably, multi-scale terrain modeling (MSTM) algorithms can process terrain data at different scales to generate high-precision 3D terrain models. MSTM combines large-scale overall terrain data with small-scale local detail data to generate a comprehensive model that includes both macroscopic terrain features and microscopic details. This method not only improves the accuracy of terrain modeling but also effectively reduces computational resource consumption. Specifically, the urban topographic map is first preprocessed to extract terrain elevation data and important terrain features, such as rivers, hills, and low-lying areas. Next, building model data is integrated into the topographic map, reflecting the specific location, shape, and height of buildings. Then, the multi-scale terrain modeling algorithm fuses these terrain and building data to generate static terrain modeling data containing terrain features and building distribution. The static terrain modeling data generated in this step provides a detailed description of the city's terrain features and building distribution, accurately reflecting the city's surface morphology and structure.

[0077] Step S312: Based on the feature information and soil property distribution map, perform dynamic change process simulation processing. By conducting time series analysis on the changes in soil saturation during rainfall events with different rainfall intensities and durations, obtain dynamic infiltration data for different regions.

[0078] In simulating dynamic processes, it is necessary to consider the changes in soil infiltration and saturation under different rainfall conditions. Different types of soil exhibit different infiltration capacities and saturation characteristics during rainfall. For example, sandy soils infiltrate faster than clay soils. To accurately simulate this process, time series analysis methods are required, and the specific steps are as follows:

[0079] Step S3121: Based on the soil property distribution map and using existing soil sample data, construct a preliminary soil permeability model for each region;

[0080] Specifically, soil samples were first collected from different areas of the city and analyzed in the laboratory to obtain physical and chemical property data such as particle size distribution, water content, permeability coefficient, and saturation. These laboratory analysis results were then integrated with soil property distribution maps to ensure a comprehensive description of the soil properties in each area. Next, preliminary soil permeability models for each area were constructed using statistical and numerical modeling methods, preferably such as multiple regression analysis, geostatistical methods, or machine learning algorithms, including permeability coefficient and saturation characteristics. Finally, the accuracy of the models was verified and corrected by comparing them with actual observation data and known soil properties to ensure they accurately reflect the permeability characteristics of soils in different areas. By constructing preliminary soil permeability models, a detailed understanding of the permeability and saturation characteristics of soils in different urban areas can be obtained, providing necessary basic parameters for subsequent dynamic simulation of rainfall events and runoff analysis.

[0081] Step S3122: Simulate various rainfall events based on the data of different rainfall intensities and durations in the feature information;

[0082] It should be noted that this step utilizes the extracted rainfall characteristic information to determine the rainfall amount and rainfall time curve for each simulation event. Then, based on the preliminary soil infiltration model, initial conditions are set, including the initial soil moisture content and saturation. Next, a hydrological model or soil moisture balance model is used for numerical simulation to simulate the infiltration, retention, and runoff processes of water in the soil during rainfall, focusing on analyzing changes in soil saturation and infiltration rate during rainfall events.

[0083] Step S3123: For each simulated rainfall event, perform time series analysis to track changes in soil infiltration rate and saturation. Calculate soil saturation changes during and after rainfall using numerical simulation and infiltration equations.

[0084] Understandably, this step uses time series analysis to identify trends and periodic changes in the data, focusing on tracking soil changes at each stage of rainfall. Next, numerical simulation techniques are applied to calculate the soil infiltration process under different rainfall intensities and durations using the Richards equation, analyzing the changes in soil saturation from its initial state to the end of the rainfall. Through these analyses and calculations, a comprehensive understanding of the soil's infiltration characteristics and saturation variation patterns under different rainfall conditions can be achieved.

[0085] The one-dimensional form of the Richards equation is:

[0086]

[0087] Where θ represents volumetric water content; t represents time; z represents soil depth; K(θ) represents soil moisture permeability coefficient; and ψ represents soil matrix potential.

[0088] In practical applications, numerical solutions need to be obtained by combining the characteristic curves of specific soils. In this embodiment, the soil characteristic curve is assumed to be:

[0089]

[0090] Where ψ(θ) represents the soil matrix potential as a function of volumetric water content θ; K(θ) represents the soil water permeability coefficient; ψ s θ represents the matrix potential of soil under saturated conditions. s K represents the saturated water content of the soil. s The coefficient of permeability of soil under saturation is represented by λ; λ and η represent empirical parameters reflecting soil properties. Substituting these relationships into the Richards equation, numerical simulations can be performed to solve for the distribution and variation of soil moisture content under rainfall conditions.

[0091] Step S3124: Organize the results of time series analysis to obtain dynamic permeation data.

[0092] Understandably, this step transforms the processed data into datasets that reflect the changes in soil infiltration rate and saturation over time under different rainfall intensities and durations. Each dataset contains data at multiple time points, describing the soil response throughout the rainfall event.

[0093] Step S313: Construct a dynamic heterogeneous map model based on static terrain modeling data and dynamic infiltration data;

[0094] It should be noted that this step combines static topographic data with dynamic infiltration data to form a comprehensive model that can dynamically reflect water flow and soil infiltration characteristics during rainfall events. Nodes in the model represent important topographic features, building locations, and soil sample points, while edges represent the interactions between these nodes, including water flow paths and infiltration paths. The dynamic aspect is reflected in the model's ability to reflect hydrological changes during and after rainfall events in real time. Static topographic modeling data provides the city's topographic features and building distribution, serving as relatively fixed foundational data. Dynamic infiltration data, derived from time-series analysis, reflects changes in soil infiltration rate and saturation under different rainfall intensities and durations. This dynamic data allows the model to update soil moisture content and infiltration in real time, simulating dynamic hydrological changes during and after rainfall. Secondly, heterogeneity is reflected in the model's multi-layered and multi-element integration. Hydrological processes in urban environments are complex and variable, with different regions exhibiting varying topographic, building, and soil characteristics. The nodes in the model, representing important topographic features, building locations, and soil sample points, are inherently heterogeneous. For example, the water flow and infiltration characteristics differ significantly between high-density built-up areas and open green spaces, requiring the model to address these heterogeneous features separately. Edges represent the interactions between these nodes, including water flow paths and infiltration paths. Through this construction method, the dynamic heterogeneous graph model can not only reflect the hydrological response characteristics of different urban areas under various rainfall conditions in real time, but also accurately describe the interactions and combined effects between different areas.

[0095] Step S314: Based on the dynamic heterogeneous graph model, the node feature vectors are convolved and aggregated using the graph embedding algorithm to generate node embedding vectors, and the node embedding vectors are mapped to a high-dimensional feature space to obtain urban drainage feature vectors.

[0096] Specifically, firstly, the feature data of nodes in the dynamic heterogeneous graph model are integrated to form initial node feature vectors. Then, a graph embedding algorithm is used to convolve and aggregate these initial node feature vectors. Convolution captures the feature information of the node and its neighbors, while aggregation integrates the features of neighboring nodes to generate an embedding vector for each node. These embedding vectors not only contain the node's original feature information but also its relationship information with other nodes, allowing for a more comprehensive representation of node characteristics in a high-dimensional space. Next, the node embedding vectors are mapped to a high-dimensional feature space. This mapping process enables the node embedding vectors to be represented in a higher-dimensional space, capturing complex features and relationships. Finally, the high-dimensional embedding vectors of all nodes are integrated to obtain urban drainage feature vectors. These feature vectors comprehensively describe the performance of the urban drainage system under different rainfall conditions, including drainage capacity, flooding risk, and drainage efficiency.

[0097] Step S320: Perform clustering processing based on the urban drainage model and stormwater pipe network layout diagram to obtain clustering results;

[0098] Further, step S320 includes steps S321 to S324:

[0099] Step S321: Perform similarity calculation processing based on the city drainage feature vectors. Calculate the similarity between each city drainage feature vector using the cosine similarity method to obtain a similarity matrix.

[0100] Understandably, similarity calculations determine the cosine of the angle between two vectors using their dot product and norm. The similarity value ranges from -1 to 1, with values ​​closer to 1 indicating greater similarity. The resulting similarity matrix is ​​a symmetric matrix, where each element represents the similarity between two corresponding eigenvectors.

[0101] Step S322: Perform hierarchical clustering based on the similarity matrix and the rainwater pipe network layout map. By combining the geographical proximity and pipe network connectivity in the pipe network, a preliminary clustering result is obtained.

[0102] It is important to note that the focus of this step is to fully utilize the geographical proximity and pipe network connectivity of the urban drainage system to ensure that the clustering results reflect both the similarity of drainage characteristics and the actual drainage network structure. Specifically, firstly, a similarity matrix and a stormwater pipe network layout map are used as input data. The former details the similarity between urban drainage feature vectors, while the latter shows the geographical location and connectivity of the drainage system. Next, weights are defined for the proximity and connectivity of each region. The geographical proximity weight reflects the degree of geographical proximity between regions, while the pipe network connectivity weight reflects the actual connection of drainage pipes. By introducing these weights, the hierarchical clustering algorithm is improved, ensuring that the actual geographical and pipe network structure is given priority during the clustering process. That is, in each cluster merging, not only the cosine similarity of the region feature vectors is considered, but also the geographical proximity weight and pipe network connectivity weight are introduced, and the merging criterion is modified to weighted similarity. Then, the hierarchical clustering algorithm gradually merges the clusters with the highest similarity to generate a hierarchical tree. The merging process is as follows: Figure 2 As shown in the diagram, nodes A to E represent various regions within the urban drainage system. The final clusters 1 and 2 demonstrate the division of the city into two main drainage zones. During the merging process, regions with high geographical proximity and pipe network connectivity are prioritized for merging, resulting in clusters that better reflect the actual drainage system structure. Finally, a hierarchical tree structure is used to determine the preliminary clustering results for each region.

[0103] Step S323: The preliminary clustering results are refined and optimized using a density-based spatial constraint clustering algorithm. By combining the flow direction in the pipe network and the density of water accumulation points in the area, boundary points and noise points are processed to obtain optimized clustering results.

[0104] It should be noted that in this embodiment, boundary points refer to points located at the boundaries of different clusters, with indistinct features and easily influenced by the surrounding area. These points exhibit uncertain drainage behavior during actual rainfall events. Noise points, on the other hand, are spatially scattered and isolated points caused by abnormal rainfall or special terrain conditions. Specifically, boundary points and noise points in the preliminary clustering results are first identified. Boundary points are often located at turning points in the drainage system or at the boundaries of different terrain features, while noise points are small-scale water accumulation points or areas with poor drainage. Next, a density-based spatial constraint clustering algorithm is used to calculate the density of points to identify core points, boundary points, and noise points. In this process, core points are high-density points with many neighboring points, while boundary points and noise points are low-density points and isolated points, respectively. The advantage of the density-based spatial constraint clustering algorithm is that it can handle clusters of arbitrary shapes and effectively identify noise points. During the refinement and optimization process, the flow direction reflects the actual path of water flow in the urban drainage system. For example, if an upstream boundary point shows a tendency to move downstream during rainfall, it should be reclassified into the downstream cluster instead of remaining in its original cluster. This adjustment more accurately reflects the actual drainage situation. High-density waterlogging areas are typically locations with high drainage pressure; these areas exhibit significant waterlogging characteristics during rainfall events and require separate clustering for better management. For instance, a city square experiencing persistent waterlogging during multiple rainfall events should be identified as a high-density waterlogging point and treated separately during clustering to allow for targeted drainage measures. This step improves the accuracy and effectiveness of clustering results by reallocating boundary points and noise points, ensuring each point is appropriately assigned to a cluster.

[0105] Step S324: Perform feature fusion processing based on the optimized clustering results. The drainage feature vectors in each cluster are fused using a weighted average method to obtain the final clustering result.

[0106] The formula for the weighted average method is as follows:

[0107]

[0108] Among them, v f represents the drained feature vector after fusion; i represents the index of the drained feature vector; n represents the total number of drained feature vectors in the cluster; v i w represents the i-th drainage feature vector in each cluster; iThis represents the weight of the i-th drainage feature vector.

[0109] Step S330: Divide the city into at least one drainage zone based on the clustering results to obtain the segmentation results.

[0110] Further, step S330 includes steps S331 to S333:

[0111] Step S331: Perform region identification processing based on the optimized clustering results to obtain region identification results;

[0112] Understandably, this step involves optimizing clustering results to identify various drainage zones within the city. These zones are divided based on similarity, density, and actual pipe network connections to ensure relatively consistent drainage characteristics for each zone. During the identification process, the city's Geographic Information System (GIS) data is used to map the clustering results onto the actual city map, determining the geographical boundaries of each cluster.

[0113] Step S332: Determine the drainage outlet and key nodes of each drainage zone based on the area identification results to obtain preliminary drainage zone results;

[0114] It should be noted that, based on the identification, the drainage outlets and key nodes of each drainage zone are further determined. Drainage outlets refer to the main outlets through which rainwater is discharged from the drainage system, while key nodes include important connection points, flow regulation points, and nodes prone to water accumulation within the drainage system. Identifying these nodes and outlets helps optimize the management and control of the entire drainage system.

[0115] Step S333: Evaluate and adjust the boundaries of the drainage zones based on the preliminary drainage zoning results to obtain the final zoning results.

[0116] Furthermore, based on actual geographical conditions and the operation of the drainage system, the preliminary drainage zoning results are evaluated and adjusted as necessary. Evaluation criteria include drainage efficiency, flooding risk, and ease of maintenance. If necessary, zoning boundaries are adjusted to avoid classifying areas with significantly different drainage characteristics into the same zoning, thereby improving the overall coordination of the drainage system within each zoning. After completing the evaluation and adjustments, the final zoning result, i.e., the complete drainage zoning map, is generated. This map shows in detail the boundaries, drainage outlets, and key nodes of each drainage zoning zone.

[0117] Through the specific steps described above, the optimized clustering results are transformed into actual drainage zones, ensuring that each zone has reasonable boundaries and clearly defined drainage nodes. This not only helps improve the efficiency and reliability of urban drainage systems but also provides a scientific basis for urban zoning management in response to extreme rainfall events.

[0118] Example 3

[0119] The only difference between this embodiment and embodiment 1 or 2 is that step S400 includes steps S410 to S430:

[0120] Step S410: Perform rainfall pattern recognition processing based on the segmentation results, and obtain rainfall pattern feature data by identifying and extracting typical rainfall patterns for each partition;

[0121] Step S420: Perform time series modeling based on rainfall pattern characteristics, establish a prediction model through a long short-term memory network, and combine historical rainfall data for model training and optimization to obtain a rainfall prediction model;

[0122] Step S430: Based on the rainfall prediction model, perform prediction processing on the real-time data to obtain the predicted change results.

[0123] Specifically, the process begins with rainfall pattern recognition based on the city segmentation results. For each sub-district, historical rainfall data is collected and analyzed, combined with the sub-district's topographic features and drainage system characteristics to identify and extract typical rainfall patterns, generating rainfall pattern feature data. In-depth analysis of these features allows for the extraction of pattern data reflecting the unique rainfall characteristics of each sub-district. Next, time series modeling is performed based on the rainfall pattern features. A Long Short-Term Memory (LSTM) network is used, combined with historical rainfall data, for model training and optimization. LTM networks excel at processing time series data and can capture data dependencies over long time spans. By incorporating features related to urban drainage, such as drainage system pressure, historical waterlogging data, and soil permeability, the model can more accurately predict future rainfall trends. For example, the model can learn from historical rainfall patterns, combined with current soil saturation and drainage system pressure, to predict the intensity and duration of future rainfall events. During the optimization training process, particular attention is paid to the model's adaptability to different sub-districts, ensuring it reflects the unique rainfall and drainage characteristics of each sub-district. Finally, based on the optimized rainfall prediction model, real-time data is used for prediction processing. By inputting real-time rainfall data and drainage system flow monitoring data into a long short-term memory network model, the model can predict rainfall and flow changes over a preset time period in real time. By combining real-time monitoring data, the model can not only predict the overall trend of future rainfall events but also provide accurate flow change predictions, helping to identify potential flooding risk areas.

[0124] Example 4

[0125] The only difference between this embodiment and embodiments 1, 2, or 3 is that step S500 includes steps S510 to S540:

[0126] Step S510: Based on the predicted change results, perform directed graph structure construction processing. Use the topology sorting algorithm to transform the key nodes and connection relationships in the urban drainage system into a directed graph structure to obtain a preliminary directed graph model.

[0127] It should be noted that, firstly, based on the predicted changes, key nodes in the urban drainage system are identified. These nodes include major drainage outlets, flow regulation nodes, and nodes in areas prone to flooding. The identification of key nodes relies on flow changes and flooding risk assessments in the predicted changes, identifying the points with the greatest impact on the drainage system under different rainfall scenarios. Next, a topological sorting algorithm is used to process these key nodes and their connections, constructing a directed graph structure. The topological sorting algorithm effectively handles directional relationship networks, ensuring that the logical order and flow paths between nodes are maintained during the construction process. Specifically, nodes are arranged in topological order according to the direction of water flow from upstream to downstream, ensuring the correctness and consistency of the water flow path. Topological sorting forms an acyclic directed graph structure by sorting the nodes of the entire drainage system according to their dependencies. Furthermore, during the construction of the directed graph, special attention is paid to the connections between nodes, which reflect the water flow paths and mutual influences between different nodes in the drainage system. By combining the predicted changes, the connection strength and direction between nodes are dynamically adjusted to ensure that the directed graph accurately reflects the actual operation of the drainage system under different rainfall conditions. The preliminary directed graph model not only shows the overall structure of the urban drainage system, but also reveals the dependencies and flow paths between key nodes, providing a clear framework for further detailed analysis and system optimization.

[0128] Step S520: Based on the preliminary directed graph model and rainwater pipe network layout diagram, assign node weights and edge weights. Using a weight calculation method based on flow and pressure, set the node weight to the node's flow demand and set the edge weight to the pipe's flow capacity and pressure bearing capacity to obtain a weighted directed graph model.

[0129] Understandably, in the node weight assignment process, the weight of each node is primarily based on its flow demand. Flow demand reflects the amount of water a node needs to handle under different rainfall scenarios. By analyzing and predicting changes, the maximum flow demand of each node can be determined. For edge weight assignment, the main factors are the pipeline's flow capacity and pressure bearing capacity. The pipeline's flow capacity refers to the maximum flow rate it can handle when operating at full load, while pressure bearing capacity reflects the pipeline's stability and safety under different water pressure conditions. By combining these physical characteristics, a reasonable weight can be assigned to each edge (i.e., the pipeline). The weighted directed graph model not only demonstrates the structure of the urban drainage system but also reflects the workload and capacity of each node and pipeline under different rainfall scenarios.

[0130] Step S530: Perform scenario analysis processing based on the weighted directed graph model. By simulating rainfall events with different rainfall intensities and durations, and combining the rainfall and flow changes in the predicted change results, use the dynamic shortest path algorithm to calculate the pressure and flow of each node under different scenarios to obtain the scenario simulation results.

[0131] Specifically, firstly, rainfall scenario parameters of different intensities and durations are defined to ensure the comprehensiveness of the scenario analysis. Then, different rainfall events are simulated based on the rainfall scenario parameters, and rainfall and flow data from the predicted changes are combined to generate rainfall data for each scenario. The dynamic shortest path algorithm is used because it can effectively handle path optimization problems in complex networks, and is particularly suitable for urban drainage systems under real-time changing rainfall and flow conditions. This algorithm can calculate the pressure and flow distribution of each node under different scenarios, find the optimal path from the source node to the target node, and identify potential bottlenecks and high-risk areas. In this embodiment, a weighted directed graph model is used, enabling the algorithm to dynamically adjust path selection, considering the pipeline's flow capacity and pressure bearing capacity, providing more accurate simulation results. Calculation formula:

[0132] d(s,t)=min∑ (u,v)∈E d(u,v);

[0133]

[0134] Where s represents the source node; t represents the target node; u and v represent any two nodes; E represents the edge set; d(s,t) represents the shortest path distance from source node s to target node t; w(u,v) represents the weight of edge (u,v), representing the pressure carrying capacity; d(u,v) represents the traffic demand from node u to node v; d(s,v) represents the shortest path distance from source node s to node v; P v F represents the pressure at node v. v This represents the flow of node v.

[0135] Step S540: Based on the scenario simulation results, perform overflow risk analysis and processing. Use a multi-objective optimization algorithm to integrate node pressure, pipeline flow and regional water accumulation, and identify high-risk areas and potential overflow points to obtain the final simulation results.

[0136] It should be noted that multi-objective optimization algorithms can consider multiple objectives simultaneously. In this embodiment, these objectives include node pressure, pipeline flow rate, and regional water accumulation. Node pressure represents the system's load level under different scenarios, pipeline flow rate reflects the water flow capacity through each pipeline, and regional water accumulation reveals the potential risk of flooding. By comprehensively considering these objectives, the optimization algorithm can balance the relationship between different factors and find the optimal system operating state.

[0137] Example 5

[0138] The only difference between this embodiment and embodiments 1, 2, 3, or 4 is that step S600 includes steps S610 to S630:

[0139] Step S610: Evaluate the effectiveness of the initial control scheme based on the simulation results to obtain the evaluation results;

[0140] Understandably, this step involves examining the performance of various measures in the plan under different rainfall scenarios to identify its effectiveness and shortcomings. The focus is on the response effectiveness at key points and in high-risk areas, assessing the performance of the drainage system in actual operation. The evaluation results include an overall effectiveness analysis of the initial control plan, specific performance under different rainfall scenarios, identified major problems, and improvement recommendations.

[0141] Step S620: Identify optimization requirements based on the evaluation results;

[0142] It should be noted that optimization requirements include increasing drainage capacity, adjusting pipeline layout, and optimizing pressure distribution at nodes. Specific optimization directions and objectives are determined for weaknesses and potential problems exposed under different scenarios.

[0143] Step S630: Develop and implement optimization plans based on optimization requirements, and make dynamic adjustments through real-time monitoring and feedback mechanisms.

[0144] Understandably, dynamic adjustments include opening or closing the gates of the storage tank, adjusting the flow rate of the drainage network, and optimizing the overflow outlet settings to adapt to real-time changes in rainfall and flow conditions.

[0145] Example 6:

[0146] This embodiment provides a water body runoff control system for cities, including:

[0147] The acquisition module is used to acquire historical rainfall data, real-time data, modeling parameters, and urban drainage system parameters. Real-time data includes real-time rainfall data and flow monitoring data. Modeling parameters include urban topographic maps, building models, and soil property distribution maps. Urban drainage system parameters include stormwater pipe network layout maps and initial control schemes.

[0148] The analysis module performs frequency analysis on historical rainfall data based on a pre-set Bayesian inference mathematical model to obtain feature information, including urban rainfall characteristics and trends.

[0149] The segmentation module is used to perform regional segmentation processing based on feature information and stormwater pipe network layout map, and to divide the city into at least one drainage zone based on drainage characteristics to obtain the segmentation result.

[0150] The prediction module is used to construct a rainfall prediction model based on the segmentation results, and to calculate the predicted changes by taking real-time data as the input value of the rainfall prediction model. The predicted changes include rainfall and flow changes within a preset time period in the future.

[0151] The simulation module is used to perform simulation processing based on the prediction results and the stormwater pipe network layout diagram. It obtains simulation results by modeling the urban drainage system as a directed graph structure and analyzing the node pressure, pipe flow and overflow risk under different scenarios.

[0152] The optimization module optimizes the initial control scheme based on the simulation results to obtain a dynamic control scheme for urban water body runoff.

[0153] In the embodiments disclosed in this application, the analysis module includes:

[0154] The first processing unit is used to organize and process historical rainfall data, and obtain the rainfall dataset by filling in missing values ​​and removing outliers.

[0155] The first computing unit is used to perform parameter estimation processing based on the rainfall dataset. It calculates the rainfall amount, rainfall duration, and rainfall interval for each rainfall event using the maximum likelihood estimation method, thereby obtaining a preliminary set of rainfall event feature parameters.

[0156] The second processing unit is used to perform Bayesian inference processing based on the preliminary rainfall event feature parameter set. By setting the gamma distribution as the prior distribution and updating the posterior distribution using historical rainfall data, the posterior distribution result of each rainfall event parameter is obtained.

[0157] The third processing unit is used to perform rainfall frequency analysis based on the posterior distribution results. It generates rainfall events by simulating them and obtains feature information by statistically analyzing the frequency distribution of rainfall amount, rainfall duration, and rainfall interval.

[0158] In the embodiments disclosed in this application, the partitioning module includes:

[0159] The first simulation unit is used to perform runoff simulation processing based on feature information and modeling parameters. By simulating rainfall events with different rainfall intensities and durations, it analyzes the runoff collection, diversion, and water accumulation areas to obtain urban drainage feature vectors.

[0160] The first clustering unit is used to perform clustering processing based on the urban drainage model and stormwater pipe network layout map to obtain clustering results;

[0161] The first partitioning unit is used to divide the city into at least one drainage zone based on the clustering results to obtain the segmentation result.

[0162] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling urban water body runoff, characterized in that, include: The system acquires historical rainfall data, real-time data, modeling parameters, and urban drainage system parameters. The real-time data includes real-time rainfall data and flow monitoring data. The modeling parameters include urban topographic maps, building models, and soil property distribution maps. The urban drainage system includes stormwater pipe network layout maps and initial control schemes. Based on a pre-set Bayesian inference mathematical model, the historical rainfall data is processed by rainfall frequency analysis to obtain feature information, which includes urban rainfall characteristics and trends. Based on the feature information and the rainwater pipe network layout map, the area is divided into at least one drainage zone based on the drainage characteristics to obtain the segmentation result. A rainfall prediction model is constructed based on the segmentation results, and the real-time data is used as the input value of the rainfall prediction model to calculate the predicted change results, which include the changes in rainfall and flow rate within a preset future time period. The simulation is performed based on the predicted changes and the stormwater pipe network layout diagram. The urban drainage system is modeled as a directed graph structure, and the node pressure, pipe flow and overflow risk under different scenarios are analyzed to obtain the simulation results. Based on the simulation results, the initial control scheme is optimized to obtain a dynamic control scheme for urban water body runoff. The region division process includes: Based on the feature information and the modeling parameters, runoff simulation is performed. By simulating rainfall events with different rainfall intensities and durations, the runoff collection, diversion, and waterlogging areas are analyzed to obtain urban drainage feature vectors. Clustering results were obtained by performing clustering processing based on the urban drainage model and the stormwater pipe network layout diagram. Based on the clustering results, the city is divided into at least one drainage zone to obtain the segmentation results; The runoff simulation process includes: Based on the urban topographic map and the building model, terrain modeling processing is performed, and static terrain modeling data containing terrain features and building distribution is generated through a multi-scale terrain modeling algorithm; Based on the aforementioned feature information and the aforementioned soil property distribution map, dynamic change process simulation processing is performed. By conducting time series analysis on the changes in soil saturation during rainfall events of different rainfall intensities and durations, dynamic infiltration data for different regions are obtained. A dynamic heterogeneous map model is constructed based on the static terrain modeling data and the dynamic infiltration data. Based on the dynamic heterogeneous graph model, the node feature vectors are convolved and aggregated using a graph embedding algorithm to generate node embedding vectors, and the node embedding vectors are mapped to a high-dimensional feature space to obtain urban drainage feature vectors.

2. The urban water body runoff control method according to claim 1, characterized in that, Based on a pre-defined Bayesian inference mathematical model, the historical rainfall data is processed by rainfall frequency analysis to obtain feature information, including: The historical rainfall data is processed by filling in missing values ​​and removing outliers to obtain a rainfall dataset. Based on the rainfall dataset, parameter estimation is performed. The rainfall amount, duration, and interval of each rainfall event are calculated using the maximum likelihood estimation method to obtain a preliminary set of rainfall event feature parameters. Bayesian inference is performed based on the preliminary set of rainfall event feature parameters. By setting the gamma distribution as the prior distribution and updating the posterior distribution using the historical rainfall data, the posterior distribution result of each rainfall event parameter is obtained. Based on the posterior distribution results, rainfall frequency analysis is performed. Rainfall events are generated by simulation, and feature information is obtained by statistically analyzing the frequency distribution of rainfall amount, rainfall duration, and rainfall interval.

3. The urban water body runoff control method according to claim 1, characterized in that, Clustering results were obtained by performing clustering processing based on the urban drainage model and the stormwater pipe network layout diagram, including: The similarity calculation is performed on the urban drainage feature vectors. The similarity between each urban drainage feature vector is calculated using the cosine similarity method to obtain a similarity matrix. Hierarchical clustering is performed based on the similarity matrix and the rainwater pipe network layout map. By combining the geographical proximity and pipe network connectivity in the pipe network, preliminary clustering results are obtained. The preliminary clustering results are refined and optimized using a density-based spatial constraint clustering algorithm. By combining the flow direction in the pipe network and the density of regional water accumulation points, boundary points and noise points are processed to obtain optimized clustering results. Based on the optimized clustering results, feature fusion processing is performed, and the drainage feature vectors in each cluster are fused using a weighted average method to obtain the final clustering result.

4. The urban water body runoff control method according to claim 1, characterized in that, A rainfall prediction model is constructed based on the segmentation results, and the real-time data is used as the input value of the rainfall prediction model to calculate the predicted change results, including: Based on the segmentation results, rainfall pattern recognition processing is performed, and rainfall pattern feature data is obtained by identifying and extracting typical rainfall patterns for each partition; Based on the characteristics of the rainfall pattern, time series modeling is performed, a prediction model is established through a long short-term memory network, and the model is trained and optimized by combining the historical rainfall data to obtain a rainfall prediction model. The real-time data is processed based on the rainfall prediction model to obtain the predicted changes.

5. The urban water body runoff control method according to claim 1, characterized in that, Based on the predicted changes and the stormwater pipe network layout diagram, simulation processing is performed. The urban drainage system is modeled as a directed graph structure, and simulation results are obtained by analyzing node pressure, pipe flow, and overflow risk under different scenarios, including: Based on the predicted changes, a directed graph structure is constructed. The key nodes and connections in the urban drainage system are transformed into a directed graph structure using a topology sorting algorithm, resulting in a preliminary directed graph model. Based on the preliminary directed graph model and the rainwater pipe network layout diagram, node weights and edge weights are assigned. Using a weight calculation method based on flow and pressure, the weight of a node is set to the flow demand of the node, and the weight of an edge is set to the flow capacity and pressure bearing capacity of the pipe, thus obtaining a weighted directed graph model. Scenario analysis is performed based on the weighted directed graph model. By simulating rainfall events with different intensities and durations, and combining the rainfall and flow changes in the predicted change results, the dynamic shortest path algorithm is used to calculate the pressure and flow of each node under different scenarios, and the scenario simulation results are obtained. Based on the simulation results, overflow risk analysis is performed. A multi-objective optimization algorithm is used to integrate node pressure, pipeline flow, and regional water accumulation, and high-risk areas and potential overflow points are identified to obtain the final simulation results.

6. A water body runoff control system for cities, characterized in that, include: The acquisition module is used to acquire historical rainfall data, real-time data, modeling parameters, and urban drainage system parameters. The real-time data includes real-time rainfall data and flow monitoring data. The modeling parameters include urban topographic maps, building models, and soil property distribution maps. The urban drainage system includes stormwater pipe network layout maps and initial control schemes. The analysis module performs rainfall frequency analysis on the historical rainfall data based on a preset Bayesian inference mathematical model to obtain feature information, which includes urban rainfall characteristics and trends. The segmentation module is used to perform regional segmentation processing based on the feature information and the rainwater pipe network layout map, and to divide the city into at least one drainage zone based on drainage characteristics to obtain the segmentation result. The prediction module is used to construct a rainfall prediction model based on the segmentation results, and to use the real-time data as the input value of the rainfall prediction model to calculate the predicted change results, which include the changes in rainfall and flow rate within a preset future time period. The simulation module is used to perform simulation processing based on the predicted change results and the stormwater pipe network layout diagram. It obtains simulation results by modeling the urban drainage system as a directed graph structure and analyzing the node pressure, pipe flow and overflow risk under different scenarios. The optimization module optimizes the initial control scheme based on the simulation results to obtain a dynamic control scheme for urban water body runoff. The partitioning module includes: The first simulation unit is used to perform runoff simulation processing based on the feature information and the modeling parameters. By simulating rainfall events with different rainfall intensities and durations, it analyzes the runoff collection, diversion, and water accumulation areas to obtain urban drainage feature vectors. The first clustering unit is used to perform clustering processing based on the urban drainage model and the stormwater pipe network layout diagram to obtain clustering results; The first partitioning unit is used to divide the city into at least one drainage zone based on the clustering results to obtain a segmentation result; The runoff simulation process includes: Based on the urban topographic map and the building model, terrain modeling processing is performed, and static terrain modeling data containing terrain features and building distribution is generated through a multi-scale terrain modeling algorithm; Based on the aforementioned feature information and the aforementioned soil property distribution map, dynamic change process simulation processing is performed. By conducting time series analysis on the changes in soil saturation during rainfall events of different rainfall intensities and durations, dynamic infiltration data for different regions are obtained. A dynamic heterogeneous map model is constructed based on the static terrain modeling data and the dynamic infiltration data. Based on the dynamic heterogeneous graph model, the node feature vectors are convolved and aggregated using a graph embedding algorithm to generate node embedding vectors, and the node embedding vectors are mapped to a high-dimensional feature space to obtain urban drainage feature vectors.

7. The urban water body runoff control system according to claim 6, characterized in that, The analysis module includes: The first processing unit is used to perform data processing based on the historical rainfall data, and obtain the rainfall dataset by filling in missing values ​​and removing outliers. The first calculation unit is used to perform parameter estimation processing based on the rainfall dataset, and calculate the rainfall amount, rainfall duration and rainfall interval for each rainfall event using the maximum likelihood estimation method to obtain a preliminary set of rainfall event feature parameters; The second processing unit is used to perform Bayesian inference processing based on the preliminary rainfall event feature parameter set. By setting the gamma distribution as the prior distribution and updating the posterior distribution using the historical rainfall data, the posterior distribution result of each rainfall event parameter is obtained. The third processing unit is used to perform rainfall frequency analysis processing based on the posterior distribution results. It obtains feature information by simulating rainfall events and statistically analyzing the frequency distribution of rainfall amount, rainfall duration, and rainfall interval.

Images