A city waterlogging risk management method and system based on multi-model coupling

Abstract

The application provides a kind of urban waterlogging risk management method and system based on multi-model coupling, method contains constructing digital twin holographic base, through the deep coupling of intelligent model of waterlogging identification and physical mechanism driven rain flood simulation engine, constructs urban rain flood analysis model, generates high-precision waterlogging risk heat map and dynamic forecast threshold, real-time broadcast risk situation through voice interaction module;Relying on digital twin holographic base to build storm runoff pipe network three-dimensional coupling simulation system, update deduction result at minute level frequency, and real-time transport rainfall spatio-temporal distribution data to urban rain flood analysis model;Establish multi-objective collaborative decision engine, input water spread rate to multi-objective collaborative decision engine, generate pump station scheduling scheme, traffic control strategy and material delivery path.System contains model construction module, data analysis module and collaborative decision module.The application constructs a complete urban waterlogging risk management system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of waterlogging risk management, and particularly relates to a city waterlogging risk management method and system based on multi-model coupling. BACKGROUND

[0002] With the acceleration of urbanization and the frequent occurrence of extreme weather events, the problem of urban waterlogging is becoming increasingly serious. Traditional drainage systems are difficult to cope with high-intensity rainfall and complex surface runoff changes, leading to frequent waterlogging disasters, threatening urban safety and residents' lives and property. Existing waterlogging risk management technologies mostly rely on single models or static parameter systems. For example, mechanism models based on hydrology principles can simulate runoff processes, but they are difficult to reflect dynamic waterlogging evolution and multi-factor coupling effects. While hydrodynamic models have high accuracy, they are difficult to achieve rapid response and real-time decision support due to their high computational complexity and data requirements. In addition, existing solutions lack deep fusion of multi-source heterogeneous data (such as terrain, pipe network, weather, and real-time monitoring data), resulting in poor model adaptability and inability to dynamically optimize parameters to respond to sudden environmental changes (such as instantaneous heavy rain or pipe blockage).

[0003] Prior art one, application number: CN202410878481.X discloses a road water depth measurement method based on point cloud generation of road contour lines, including the following steps: using a laser scanner to obtain road point cloud data and processing the road point cloud data; using OTM software and based on the processed data, forming a relatively regular triangular mesh to construct a digital elevation model; by comprehensively using DEM and Toggle Contours, the contour lines are drawn, and the drawn contour lines are quality detected and post-processed; the contour map is projected and transformed; and the road water depth is calculated. Although the road water depth measurement method based on point cloud generation of road contour lines can effectively solve the city waterlogging risk management work in the city management system and effectively save installation and maintenance costs; but it only relies on road geometric data, and the measurement error is increased due to vegetation obstruction and sudden drainage blockage.

[0004] The prior art two, application number: CN202411018878.8 discloses an intelligent early warning and directional release method and system for chain disaster of external flood and internal waterlogging, obtains city monitoring data set of a target city by connecting city monitoring Internet of Things; performs internal waterlogging correlation identification based on the city monitoring data set to obtain a monitoring correlation identification result; constructs an external flood and internal waterlogging chain disaster prediction model; obtains a city internal waterlogging risk assessment report based on the monitoring correlation identification result and according to the external flood and internal waterlogging chain disaster prediction model; generates first early warning information based on the city internal waterlogging risk assessment report; determines early warning target groups based on the target city; and releases the first early warning information to the early warning target groups. Although the technical problem of low information transmission efficiency caused by the inaccurate traditional early warning release mode is solved, the technical effect of intelligent directional release of early warning information and improvement of information transmission efficiency and accuracy is achieved; however, the internal waterlogging correlation identification accuracy needs to be further improved using a single prediction model.

[0005] The prior art three, application number: CN202410760621.3 discloses a city rain flood monitoring control system based on the concept of sponge city, comprising a server, a rainfall monitoring module, a water level monitoring module, a rain flood combination analysis module, an intelligent control module and a control performance evaluation module; the rain flood combination analysis module is used to perform real-time processing and analysis on rainfall data and water level data, the intelligent control module is used to determine whether corresponding regulation and control are needed according to the analysis result of the rain flood combination analysis module, and if the corresponding regulation and control are needed, the corresponding city drainage facilities are controlled based on a control strategy. Although real-time monitoring, early warning and prediction and intelligent regulation and control of city rain flood are realized, the efficiency and precision of city rain flood management are significantly improved, the city internal waterlogging risk is reduced, the control performance of drainage facilities can be accurately fed back, the subsequent control performance is ensured to improve the rain flood management effect, and the intelligent degree is high; however, the drainage facility control is triggered only according to the water level / rainfall data, and the pre-regulation and control failure is relatively lagging.

[0006] The prior art one, the prior art two and the prior art three have the problems of single data dimension, update delay, static modeling and physically unreasonable pure data driving, and therefore, the present application provides a city internal waterlogging risk management method and system based on multi-model coupling. SUMMARY

[0007] To solve the above technical problems, the present application provides a city internal waterlogging risk management method based on multi-model coupling, comprising the following steps:

[0008] Based on urban geographic information, pipe network topology and meteorological and hydrological data, a digital twin holographic base is constructed, and a distributed big data framework is deployed to realize real-time access of meteorological radar, IoT sensor and social media waterlogging report data; through deep coupling of waterlogging identification intelligent model and physical mechanism driven rainstorm simulation engine, a city rainstorm analysis model is constructed to generate high-precision waterlogging risk heat map and dynamic prediction threshold, and the risk situation is reported in real time through voice interaction module;

[0009] Relying on the digital twin holographic base, a storm-runoff-pipe network three-dimensional coupling simulation system is constructed to update and deduce the results at a minute level, and the rainfall spatio-temporal distribution data is transmitted to the city rainstorm analysis model in real time; when the waterlogging diffusion rate output by the city rainstorm analysis model reaches the preset critical value, a red / orange / yellow three-level early warning signal is triggered;

[0010] A multi-objective collaborative decision engine is established, the waterlogging diffusion rate is input into the multi-objective collaborative decision engine to generate a pump station scheduling scheme, a traffic control strategy and a material delivery path, and the material delivery path is returned to the digital twin holographic base through the Internet of Things middle station.

[0011] Optionally, the process of constructing the city rainstorm analysis model comprises the following steps:

[0012] Obtain urban geographic information data including high-resolution topographic map, building distribution and road network; obtain pipe network topology data including drainage pipe network, rainwater box culvert and pump station location; obtain meteorological and hydrological data including historical rainfall, river water level and underground water level; integrate into a three-dimensional digital twin platform to construct a digital twin holographic base;

[0013] Based on real-time access of meteorological radar data, IoT sensor data and social media waterlogging report data, a waterlogging identification intelligent model is trained, the waterlogging area in the city is located through feature extraction and pattern recognition, and the waterlogging position is labeled in real time on the digital twin holographic base;

[0014] Based on the physical mechanism of meteorological and hydrological data, a rainstorm simulation engine is constructed, the formation and diffusion process of waterlogging is deduced through solving the coupling equation of rainfall-runoff-pipe network, and the key parameters of rainwater flow direction, flow and waterlogging depth are generated; the waterlogging identification intelligent model and the physical mechanism driven rainstorm simulation engine are deeply coupled to form a city rainstorm analysis model.

[0015] Optionally, the process of locating the waterlogging area in the city comprises the following steps:

[0016] The meteorological radar data, IoT sensor data and social media waterlogging report data are fused and processed, the features related to waterlogging are extracted, including preliminary prediction of possible waterlogging area, verification of specific location and specific spatial coordinate information of waterlogging occurrence;

[0017] The extracted features are subjected to pattern recognition and spatial correlation analysis, and meteorological radar data, IoT sensor data and social media waterlogging report data are integrated into a waterlogging recognition model; the waterlogging recognition model learns from historical waterlogging events to identify typical patterns of waterlogging occurrence, and dynamically updates the prediction results of waterlogging areas through continuous input of real-time meteorological radar data, IoT sensor data and social media waterlogging report data.

[0018] Optionally, the intensity and coverage of rainfall are captured through meteorological radar data; IoT sensor networks are deployed at key points in the city to monitor real-time ground water level, drainage pipe network flow and local humidity information; waterlogging reports uploaded by users on social media reflect waterlogging phenomena in the city.

[0019] Optionally, the process of forming a city rainstorm flood analysis model includes the following steps:

[0020] The real-time waterlogging location and range information output by the waterlogging recognition intelligent model is converted into input conditions for the rainstorm flood simulation engine to correct the initial state and boundary conditions of the simulation;

[0021] The rainwater flow direction, flow and waterlogging depth key parameters generated by the rainstorm flood simulation engine are fed back to the waterlogging recognition model to distinguish between noise signals and real waterlogging signals in the IoT sensor data;

[0022] The results of the waterlogging recognition model and the rainstorm flood simulation engine are compared in real time, and when the result deviation exceeds the preset threshold, a dynamic adjustment mechanism is started to prioritize correcting problems in the data sources; the output data of the waterlogging recognition model and the rainstorm flood simulation engine are fused into the digital twin holographic base to form a city rainstorm flood analysis model that includes spatial location and range labeling, time sequence evolution and mechanism-driven multi-dimensional reconstruction, as well as dynamic restoration and prediction of city rainstorm flood scenarios.

[0023] Optionally, the process of constructing a rainstorm-runoff-pipe network three-dimensional coupled simulation system includes the following steps:

[0024] Based on the precipitation process and the physical mechanism of surface runoff, a hydrological model under rainstorm events is constructed to simulate the distribution and aggregation of rainfall on the ground surface; the dynamic process of waterlogging formation and diffusion is described through surface runoff, and the development trend of waterlogging is simulated in combination with the terrain conditions; the dynamic response of the underground drainage system is simulated using pipe network hydraulics;

[0025] Surface runoff and pipe network hydraulics respond synchronously, and a bidirectional data exchange mechanism is established between surface runoff and pipe network hydraulics; the dynamic relationship between surface waterlogging and pipe network flow is adjusted through iteration until the simulation results converge; that is, rainstorm-runoff-pipe network three-dimensional coupling;

[0026] The storm-runoff-pipe network three-dimensional coupling simulation system carries out simulation deduction at a minute level; each round of deduction is based on the latest meteorological radar data and IoT sensor feedback, and the rainfall space-time distribution and the ground water accumulation condition are dynamically updated.

[0027] Optionally, the ground runoff outputs water accumulation depth and flow, as the inflow condition of pipe network water; the pipe network water outputs node pressure and pipe flow, reflecting the feedback of drainage capacity to ground water accumulation.

[0028] Optionally, dynamic response includes pipe flow change, node pressure distribution and pump station operation state.

[0029] Optionally, the process of simulation deduction at a minute level comprises the following steps:

[0030] Rainfall intensity, spatial distribution and evolution trend data are acquired in minutes, and the data are transmitted to the storm-runoff-pipe network three-dimensional coupling simulation system in real time; discrete data are converted into continuous spatial distribution data by using Kriging interpolation;

[0031] Water accumulation depth, pipe network flow and node pressure data are acquired in minutes, and the data are transmitted to the storm-runoff-pipe network three-dimensional coupling simulation system in real time; abnormal values are filled by using linear interpolation;

[0032] The time stamps of meteorological radar data and IoT sensor data are unified to the same time reference, and the meteorological radar data and the IoT sensor data are fused.

[0033] The application provides a city waterlogging risk management system based on multi-model coupling, comprising:

[0034] A model construction module is responsible for constructing a digital twin holographic base based on city geographic information, pipe network topology and meteorological and hydrological data, deploying a distributed big data framework to realize real-time access of meteorological radar, IoT sensors and social media water accumulation report data; through deep coupling of a water accumulation identification intelligent model and a physical mechanism driven rain flood simulation engine, a city rain flood analysis model is constructed, a high-precision waterlogging risk heat map and a dynamic prediction threshold are generated, and a voice interaction module is used to broadcast the risk situation in real time;

[0035] A data analysis module is responsible for constructing a storm-runoff-pipe network three-dimensional coupling simulation system based on the digital twin holographic base, updating and deducing results at a minute level, and delivering rainfall space-time distribution data to the city rain flood analysis model in real time; when the water accumulation diffusion rate output by the city rain flood analysis model reaches a preset critical value, a red / orange / yellow three-level early warning signal is triggered;

[0036] A collaborative decision-making module is responsible for establishing a multi-objective collaborative decision-making engine, inputting the waterlogging diffusion rate into the multi-objective collaborative decision-making engine, generating a pumping station scheduling scheme, a traffic control strategy and a material delivery path, and returning to the digital twin holographic base through the Internet of Things middle station.

[0037] The present application constructs a digital twin holographic base and a city rain flood analysis model, integrates city geographic information, pipe network topology and meteorological hydrological data to construct a high-precision digital twin holographic base, deploys a distributed big data framework to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data, ensures the comprehensiveness and timeliness of the data, deeply couples a waterlogging identification intelligent model and a physical mechanism driven rain flood simulation engine to generate a high-precision waterlogging risk heat map and a dynamic prediction threshold, establishes real-time monitoring and prediction capability of city waterlogging risk, and provides basic data support for risk early warning and decision-making.

[0038] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description and the accompanying drawings.

[0039] The technical solutions of the present application will be further described in detail below by means of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0040] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0041] Figure 1 The flow chart of the city waterlogging risk management method based on multi-model coupling in embodiment 1 of the present application;

[0042] Figure 2 Process diagram for constructing a city rain flood analysis model in the embodiment 2 of the present application;

[0043] Figure 3 Process diagram for positioning a waterlogging area in a city in the embodiment 3 of the present application;

[0044] Figure 4 Process diagram for forming a city rain flood analysis model in the embodiment 4 of the present application;

[0045] Figure 5 Process diagram for constructing a storm-runoff-pipe network three-dimensional coupling simulation system in the embodiment 5 of the present application;

[0046] Figure 6 Process diagram for storm-runoff-pipe network three-dimensional coupling in the embodiment 6 of the present application;

[0047] Figure 7 Process diagram for simulation deduction at a minute level in the embodiment 7 of the present application;

[0048] Figure 8 Process diagram for generating a pump station scheduling scheme, a traffic control strategy and a material delivery path in the embodiment 8 of the present application;

[0049] Figure 9 Block diagram of a city waterlogging risk management system based on multi-model coupling in the embodiment 9 of the present application. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present application will be described herein below with reference to the drawings; it should be understood that the preferred embodiments described herein are merely intended to describe and explain the present application, and are not intended to limit the present application.

[0051] The terms used in the embodiments of the present application are merely intended for the purpose of describing the specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms “a,” “an,” and “the” used in the embodiments of the present application are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0052] The following description refers to the accompanying drawings. Unless otherwise noted, same or similar components in different drawings have same or similar reference numerals. The implementations described in the following exemplary embodiments are not meant to represent all implementations consistent with the present disclosure. Instead they are simply examples of apparatuses and methods consistent with some aspects of the present disclosure. In the description of the present disclosure, it should be understood that the terms "first", "second", "third", etc. are used merely to distinguish like objects from one another, and are not necessarily used to describe a particular sequential or chronological order, unless otherwise noted. The terms "first", "second", "third", etc. are understood to be interchangeable depending on the context of the description.

[0053] Embodiment 1: As shown in the following table, the present embodiment provides a city waterlogging risk management method based on multi-model coupling, comprising the following steps: Figure 1

[0054] S100: Based on city geographic information, pipe network topology and meteorological and hydrological data, a digital twin holographic base is constructed, a distributed big data framework is deployed to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data; through deep coupling of waterlogging identification intelligent model and physical mechanism driven rain flood simulation engine, a city rain flood analysis model is constructed, a high-precision waterlogging risk heat map and dynamic prediction threshold are generated, and a voice interaction module is used to broadcast the risk situation in real time;

[0055] S200: Relying on the digital twin holographic base, a storm-runoff-pipe network three-dimensional coupling simulation system is constructed, the deduction result is updated at a minute level, and the rainfall spatio-temporal distribution data is transmitted to the city rain flood analysis model in real time; when the waterlogging diffusion rate output by the city rain flood analysis model reaches the preset critical value, a red / orange / yellow three-level early warning signal is triggered;

[0056] S300: A multi-objective collaborative decision engine is established, the waterlogging diffusion rate is input into the multi-objective collaborative decision engine, a pump station scheduling scheme, a traffic control strategy and a material delivery path are generated, and the material delivery path is returned to the digital twin holographic base through the Internet of Things middle platform.

[0057] ​The working principle and beneficial effects of the above technical solution are: firstly, based on city geographic information, pipe network topology and meteorological hydrological data, a digital twin holographic base is constructed, a distributed big data framework is deployed to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data; through deep coupling of waterlogging identification intelligent model and physical mechanism driven rainstorm simulation engine, a city rainstorm analysis model is constructed, a high-precision waterlogging risk heat map and dynamic prediction threshold are generated, and a voice interaction module is used to report the risk situation in real time; secondly, a rainstorm-runoff-pipe network three-dimensional coupled simulation system is constructed based on the digital twin holographic base, the deduction result is updated at a minute level, and the rainfall spatio-temporal distribution data is real-time delivered to the city rainstorm analysis model; when the waterlogging diffusion rate output by the city rainstorm analysis model reaches the preset critical value, a red / orange / yellow three-level early warning signal is triggered; finally, a multi-objective collaborative decision engine is established, the waterlogging diffusion rate is input into the multi-objective collaborative decision engine, a pump station scheduling scheme, a traffic control strategy and a material delivery path are generated, and the data is returned to the digital twin holographic base through the Internet of Things middle station. Step S100 of the above scheme constructs the digital twin holographic base and the city rainstorm analysis model, a high-precision digital twin holographic base is constructed by integrating city geographic information, pipe network topology and meteorological hydrological data; a distributed big data framework is deployed to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data, ensuring the comprehensiveness and timeliness of the data; through deep coupling of waterlogging identification intelligent model and physical mechanism driven rainstorm simulation engine, a high-precision waterlogging risk heat map and dynamic prediction threshold are generated; real-time monitoring and prediction capability of city waterlogging risk is established, providing basic data support for risk warning and decision-making. Step S200 constructs a rainstorm-runoff-pipe network three-dimensional coupled simulation system and an early warning mechanism, relying on the digital twin holographic base, a rainstorm-runoff-pipe network three-dimensional coupled simulation system is constructed, which can update the deduction result at a minute level and real-time deliver rainfall spatio-temporal distribution data to the city rainstorm analysis model; when the waterlogging diffusion rate reaches the preset critical value, a red / orange / yellow three-level early warning signal is automatically triggered; fine simulation and dynamic early warning of waterlogging risk are realized, providing a scientific basis for timely response measures. Step S300 multi-objective collaborative decision engine and emergency response scheme generation, a multi-objective collaborative decision engine is established, the waterlogging diffusion rate is taken as an input parameter, an emergency response scheme such as pump station scheduling scheme, traffic control strategy and material delivery path is generated, and the data is returned to the digital twin holographic base through the Internet of Things middle station; rapid response and collaborative decision of waterlogging risk are realized, resource allocation and emergency management efficiency are optimized, and the loss caused by waterlogging is minimized.

[0058] In summary, the embodiment from the data integration and model construction, dynamic simulation and risk warning, collaborative decision-making and emergency response three levels, to build a complete urban waterlogging risk management system, to achieve the whole process of risk prediction to response measures closed-loop management.

[0059] Embodiment 2: as shown in embodiment 1, on the basis of the process of constructing the urban rain flood analysis model provided by the embodiment of the application, the following steps are included: Figure 2

[0060] S101: Obtain the city geographic information data including high-resolution topographic map, building distribution and road network; Obtain the pipe network topology data including drainage pipe network, rainwater box culvert and pump station location; Obtain the meteorological and hydrological data including historical rainfall, river water level and groundwater level; Integrated into the three-dimensional digital twin platform to build a digital twin holographic base;

[0061] S102: Based on the real-time access of meteorological radar data, IoT sensor data and social media waterlogging report data, train the waterlogging identification intelligent model, locate the waterlogging area in the city through feature extraction and pattern recognition, and mark the waterlogging position on the digital twin holographic base in real time;

[0062] S103: Based on the physical mechanism of meteorological and hydrological data, build a rain flood simulation engine, deduce the formation and diffusion process of waterlogging by solving the coupling equation of rainfall-runoff-pipe network, generate key parameters such as rainwater flow direction, flow and waterlogging depth; Deeply couple the waterlogging identification intelligent model and the physical mechanism driven rain flood simulation engine to form the urban rain flood analysis model.

[0063] ​The working principle and beneficial effects of the technical solution are: the embodiment first acquires city geographic information data including high-resolution topographic map, building distribution and road network; acquires pipe network topology data including drainage pipe network, rainwater box culvert and pump station position; acquires meteorological and hydrological data including historical rainfall, river water level and underground water level; integrated into a three-dimensional digital twin platform to build a digital twin holographic base; secondly, based on the real-time access of meteorological radar data, IoT sensor data and social media waterlogging report data, train the waterlogging identification intelligent model, locate the waterlogging area in the city through feature extraction and pattern recognition, and mark the waterlogging position on the digital twin holographic base in real time; finally, based on the physical mechanism of meteorological and hydrological data, build a rainstorm simulation engine, solve the coupling equation of rainfall-runoff-pipe network, deduce the formation and diffusion process of waterlogging, and generate key parameters such as rainwater flow direction, flow and waterlogging depth; the waterlogging identification intelligent model and the physical mechanism driven rainstorm simulation engine are deeply coupled to form a city rainstorm analysis model. The step S101 of the above scheme acquires city geographic information data, and the high-resolution topographic map provides accurate ground elevation information for the model, which helps to simulate the flow direction and accumulation of rainwater on the ground; the building distribution data provides the information of the shielding and intercepting effect of the city microtopography for the model, which affects the local flow path of rainwater; the road network data provides the main water collection channel of the city surface for the model, which helps to simulate the flow characteristics of rainwater in urbanized areas; together they build the basic spatial framework of the city surface, providing basic geographic support for rainstorm simulation. Acquiring pipe network topology data, drainage pipe network data provides the structure and capacity information of the underground rainwater discharge system for the model, which affects the flow direction and flow distribution of underground rainwater; rainwater box culvert data provides the capacity and position information of temporary water storage facilities, which helps to simulate the temporary accumulation and discharge process of rainwater; pump station position data provides the operation state information of drainage facilities for the model, which affects the drainage efficiency of local areas; together they build the topology structure of the city underground drainage system, providing underground drainage support for rainstorm simulation. Acquiring meteorological and hydrological data, historical rainfall data provides long-term rainfall characteristic information for the model, which helps to build a rainfall probability model and an extreme rainfall scenario; river water level data provides the influence information of external water bodies on the city drainage system, which helps to simulate river backflow and river overflow; underground water level data provides the potential pressure of underground water on the city drainage system, which affects the underground drainage capacity; together they build the basic characteristics of city rainfall and hydrology, providing environmental background support for subsequent rainstorm simulation. By integrating the above data into a three-dimensional digital twin platform, a digital twin holographic base is built to provide a unified spatial data framework; realize the all-round digital expression of city surface and underground drainage system, provide bottom support for high-precision simulation of subsequent models.Step S102 provides dynamic waterlogging monitoring information for the model based on real-time access to meteorological radar data, IoT sensor data, and social media waterlogging report data; through feature extraction and pattern recognition, the intelligent model can automatically identify waterlogging areas in the city, reducing the error and lag of manual identification; the identified waterlogging areas are labeled on the digital twin holographic base to realize the spatial expression of waterlogging information; and dynamic input data is provided for rain flood simulation to improve the real-time performance and accuracy of the model. Step S103 constructs a numerical simulation framework based on the physical mechanism of meteorological and hydrological data; by solving the coupled equations of rainfall-runoff-pipe network, the whole process from surface runoff to pipe network transmission to waterlogging formation is simulated; the generated key parameters (such as rainwater flow direction, flow rate, and waterlogging depth) provide quantitative support for analysis; the intelligent model is deeply coupled with the physical engine to integrate the results of the waterlogging identification intelligent model with the simulation process of the physical mechanism driven engine, thereby improving the overall simulation capability of the model; and by combining real-time data with physical laws, high-precision and high-efficiency rain flood analysis is realized in complex urban environments.

[0064] In summary, the embodiment acquires and integrates multi-source data to construct the spatial framework and data support of the model; trains the intelligent model to realize dynamic identification and labeling of waterlogging areas; and constructs a physical engine and couples the intelligent model to realize high-precision simulation of the rain flood process.

[0065] Embodiment 3: As shown in Figure 3 the process of positioning the waterlogging area in the city provided by the embodiment of the application based on embodiment 2 comprises the following steps:

[0066] S1021: The intensity and coverage of rainfall are captured through meteorological radar data; IoT sensor networks are deployed at key points in the city to monitor real-time information such as ground water level, drainage pipe network flow, and local humidity; and users spontaneously upload waterlogging reports on social media to reflect waterlogging phenomena in the city;

[0067] S1022: The meteorological radar data, IoT sensor data, and social media waterlogging report data are fused and processed to extract features related to waterlogging, including preliminary prediction of possible waterlogging areas, verification of specific locations where waterlogging occurs, and specific spatial coordinate information;

[0068] S1023: The extracted features are subjected to pattern recognition and spatial correlation analysis, and the meteorological radar data, IoT sensor data, and social media waterlogging report data are integrated into a waterlogging identification model; the waterlogging identification model identifies typical patterns of waterlogging occurrence through learning from historical waterlogging events, and dynamically updates the prediction results of waterlogging areas through continuous input of real-time meteorological radar data, IoT sensor data, and social media waterlogging report data.

[0069] The working principle and beneficial effects of the technical solution are: firstly, the intensity and coverage of rainfall are captured through meteorological radar data; IoT sensor networks are deployed at key points in the city to monitor real-time ground water level, drainage pipe network flow and local humidity information; users spontaneously upload waterlogging reports on social media to reflect waterlogging phenomena in the city; secondly, meteorological radar data, IoT sensor data and social media waterlogging report data are fused and processed to extract waterlogging-related features, including preliminary prediction of possible waterlogging areas, verification of specific locations and specific spatial coordinate information of waterlogging occurrence; finally, the extracted features are subjected to pattern recognition and spatial correlation analysis, and meteorological radar data, IoT sensor data and social media waterlogging report data are integrated into a waterlogging recognition model; the waterlogging recognition model learns from historical waterlogging events to identify typical patterns of waterlogging occurrence, and dynamically updates the prediction results of waterlogging areas through continuous input of real-time meteorological radar data, IoT sensor data and social media waterlogging report data. The step S1021 of the above scheme collects meteorological radar data, IoT sensor networks and social media waterlogging report data to build a three-dimensional multi-source data monitoring system; meteorological radar data provides macro rainfall information, providing a basis for preliminary assessment of waterlogging risk; IoT sensor networks provide micro dynamic data support by monitoring real-time ground water level, pipe network flow and other key parameters, enabling accurate capture of the occurrence and development of local waterlogging phenomena; social media waterlogging report data reflects waterlogging in areas not covered by sensors, ensuring the integrity of the monitoring range; through the coordinated action of multiple data sources, the city's waterlogging phenomenon is preliminarily monitored in all directions and at multiple scales, laying a data foundation for feature extraction and pattern recognition. Step S1022 fuses and processes multiple sources of data to extract waterlogging-related features; through analysis of meteorological radar data, the possible waterlogging area is preliminarily predicted to provide a target range for fine verification; combined with real-time monitoring data from IoT sensors, the specific location of waterlogging occurrence is accurately verified to ensure the reliability of the prediction results; social media waterlogging report data further supplements and improves the positioning of waterlogging areas through the extraction of spatial coordinate information; through feature extraction and fusion of multiple sources of data, the core information directly related to waterlogging is selected from massive data to provide high-quality input for pattern recognition and dynamic prediction. Step S1023 integrates the extracted features into a waterlogging recognition model through pattern recognition and spatial correlation analysis; through learning from historical waterlogging events, the model can identify typical patterns of waterlogging occurrence, improving the prediction ability of waterlogging phenomena; at the same time, continuous input of real-time data enables the model to dynamically update the prediction results of waterlogging areas, ensuring the timeliness and accuracy of the prediction; multiple sources of data are converted into waterlogging area prediction results with practical value, providing scientific basis and technical support for decision-making of urban rainwater and flood management.

[0070] Based on the physical mechanism of meteorological and hydrological data, a rain flood simulation engine is constructed, the formation and diffusion process of accumulated water is deduced by solving the coupled equation of rainfall-runoff-pipe network, and key parameters such as rainwater flow, flow and accumulated water depth are generated; the accumulated water recognition intelligent model is deeply coupled with the physical mechanism driven rain flood simulation engine to form a city rain flood analysis model.

[0071] Embodiment 4: as shown in the embodiment 2, on the basis of the embodiment 2, the process of forming the city rain flood analysis model provided by the embodiment of the application comprises the following steps: Figure 4

[0072] S1031: the real-time accumulated water position and range information output by the accumulated water recognition intelligent model is converted into input conditions of the rain flood simulation engine, which is used to correct the initial state and boundary conditions of the simulation;

[0073] S1032: the key parameters such as rainwater flow, flow and accumulated water depth generated by the rain flood simulation engine are fed back to the accumulated water recognition model to distinguish the noise signal from the real accumulated water signal in the IoT sensor data;

[0074] S1033: the result deviation of the output of the accumulated water recognition model and the rain flood simulation engine is compared in real time, when the result deviation exceeds the preset threshold, the dynamic adjustment mechanism is started, and the problems existing in the data source are preferentially corrected; the output data of the accumulated water recognition model and the rain flood simulation engine are fused into the digital twin holographic foundation to form the city rain flood analysis model including the annotation of spatial position and range, time sequence evolution and mechanism driven multi-dimensional reconstruction, and dynamic restoration and prediction of city rain flood scene.

[0075] ​The working principle and beneficial effects of the above technical solution are: firstly, the real-time waterlogging position and range information output by the waterlogging recognition intelligent model is converted into input conditions of the rain flood simulation engine, which is used to correct the initial state and boundary conditions of the simulation; secondly, the key parameters such as rainwater flow direction, flow and waterlogging depth generated by the rain flood simulation engine are fed back to the waterlogging recognition model to distinguish noise signals from real waterlogging signals in IoT sensor data; finally, the deviation of the output results of the waterlogging recognition model and the rain flood simulation engine is compared in real time, and when the deviation exceeds the preset threshold, the dynamic adjustment mechanism is started to correct the problems in the data source first; the output data of the waterlogging recognition model and the rain flood simulation engine are fused into the digital twin holographic base to form a city rain flood analysis model including spatial position and range labeling, time sequence evolution and mechanism-driven multi-dimensional reconstruction, and dynamic restoration and prediction of urban rain flood scenes. The step S1031 of the above scheme converts and corrects the initial conditions of the waterlogging recognition intelligent model, and converts the real-time waterlogging position and range information output by the waterlogging recognition intelligent model into input conditions of the rain flood simulation engine; the initial state and boundary conditions of the waterlogging recognition intelligent model are accurately corrected to ensure that the starting point of the simulation is highly consistent with the actual scene; the initial accuracy of the waterlogging recognition intelligent model is improved to provide reliable starting point support for rain flood simulation. The step S1032 of the rain flood simulation engine parameter feedback and noise signal filtering, the key parameters such as rainwater flow direction, flow and waterlogging depth generated by the rain flood simulation engine are fed back to the waterlogging recognition model; the noise signals in the IoT sensor data are effectively distinguished from the real waterlogging signals, the influence of data interference on the accuracy of the waterlogging recognition intelligent model is reduced, the noise resistance of the waterlogging recognition intelligent model is significantly improved, and the reliability of the output result is ensured. The step S1033 of result deviation comparison and dynamic adjustment mechanism, the output deviation of the waterlogging recognition model and the rain flood simulation engine is compared in real time, and when the deviation exceeds the preset threshold, the dynamic adjustment mechanism is started to correct the problems in the data source first; the accuracy and stability of the waterlogging recognition intelligent model are continuously optimized to ensure its adaptability in different scenarios; in addition, the output data of the waterlogging recognition model and the rain flood simulation engine are fused into the digital twin holographic base to finally form a city rain flood analysis model containing spatial position and range labeling, time sequence evolution and mechanism-driven multi-dimensional reconstruction, and dynamic restoration and prediction of urban rain flood scenes, completing the full-process closed loop of the model from data acquisition, processing to multi-dimensional analysis.

[0076] In summary, the three steps of the embodiment realize high-precision construction and continuous optimization of the city rain flood analysis model from the dimensions of initial condition correction, data quality optimization, and model dynamic adjustment and fusion.

[0077] Embodiment 5: as Figure 5As shown, on the basis of Embodiment 1, the process of constructing the storm-runoff-pipe network three-dimensional coupling simulation system provided in the embodiment of the application comprises the following steps:

[0078] S201: Based on the physical mechanism of precipitation process and surface runoff, a hydrological model under a storm event is constructed to simulate the distribution and aggregation process of rainfall on the ground surface; the dynamic process of water accumulation and diffusion is described through surface runoff, and the development trend of the accumulated water is simulated in combination with the terrain condition; the dynamic response of the underground drainage system is simulated by using pipe network hydraulics, including the change of pipe flow, the distribution of node pressure and the operation state of the pump station;

[0079] S202: The dynamic responses of the surface runoff and the pipe network hydraulics are synchronized, a bidirectional data exchange mechanism is established between the surface runoff and the pipe network hydraulics, the accumulated water depth and flow of the surface runoff are output as the inflow conditions of the pipe network hydraulics; the node pressure and pipe flow of the pipe network hydraulics are output to reflect the feedback of the drainage capacity to the surface accumulated water; the dynamic relationship between the surface accumulated water and the pipe flow is adjusted through iteration until the simulation result converges; that is, the storm-runoff-pipe network three-dimensional coupling;

[0080] S203: The storm-runoff-pipe network three-dimensional coupling simulation system is simulated and deduced at a minute level frequency; each round of deduction is based on the latest meteorological radar data and IoT sensor feedback to dynamically update the spatio-temporal distribution of rainfall and the surface accumulated water condition.

[0081] The working principle and beneficial effects of the above technical solution are: firstly, based on the physical mechanism of precipitation process and surface runoff, a hydrological model under storm event is constructed to simulate the distribution and aggregation process of rainfall on the ground; the dynamic process of water accumulation and diffusion is described by surface runoff, and the development trend of water accumulation is simulated combined with the terrain conditions; the dynamic response of underground drainage system is simulated by using pipe network hydraulics, including pipe flow change, node pressure distribution and pump station operation state; secondly, the dynamic response of surface runoff and pipe network hydraulics is synchronized, a two-way data exchange mechanism is established between surface runoff and pipe network hydraulics, the surface runoff outputs water depth and flow as the inflow condition of pipe network hydraulics; the pipe network hydraulics outputs node pressure and pipe flow to reflect the feedback of drainage capacity to surface water; through iterative adjustment of the dynamic relationship between surface water and pipe flow, the simulation result converges; three-dimensional coupling of storm-runoff-pipe network; finally, the storm-runoff-pipe network three-dimensional coupling simulation system carries out simulation deduction at a minute level frequency; each round of deduction is based on the latest meteorological radar data and IoT sensor feedback to dynamically update the spatio-temporal distribution of rainfall and surface water conditions. The step S201 of constructing hydrological model, surface runoff model and pipe network hydraulic model in the above scheme is based on the physical mechanism of precipitation process and surface runoff to simulate the spatio-temporal distribution and aggregation process of rainfall on the ground; provides input data to lay the foundation for the dynamic response of surface runoff and pipe network model; accurately describes the initial influence of rainfall on surface runoff and pipe network system to provide reliable driving conditions for simulation. The surface runoff model simulates the formation and diffusion process of surface water, predicts the development trend of water accumulation combined with terrain conditions (such as slope and elevation), considers factors such as rainfall distribution, soil infiltration and surface roughness, and outputs water depth and flow; intuitively reflects the instantaneous hydrological response of the ground under storm event to provide inflow conditions for the pipe network model. The pipe network hydraulic model simulates the dynamic response of underground drainage system, including pipe flow change, node pressure distribution and pump station operation state, calculates water flow transmission and drainage capacity based on pipe network topology and pipe parameters; quantifies the drainage performance of underground pipe network system to provide data support for the feedback of surface water; provides basic model support for storm-runoff-pipe network coupling system to ensure the consistency of each sub-model in physical mechanism. The dynamic response and data interaction of surface runoff and pipe network hydraulics in step S202 ensures the dynamic response of surface runoff model and pipe network model is synchronized through uniform time step; avoids data deviation caused by inconsistent time steps to improve simulation accuracy; the surface runoff model outputs water depth and flow as the inflow condition of pipe network model; the pipe network hydraulic model outputs node pressure and pipe flow to reflect the feedback of drainage capacity to surface water; realizes the dynamic interaction of surface and underground systems to make the simulation result more close to the actual situation; through iterative calculation, the dynamic relationship between surface water and pipe flow is adjusted until the simulation result converges; optimizes the model cooperativity to ensure the stability and accuracy of the coupling system; realizes the deep integration of surface runoff and pipe network hydraulics to provide accurate expression for the overall impact of storm event.Step S203 simulates deduction and dynamic updating, simulates at a minute-level time step, captures the instantaneous dynamics of the rainstorm event, improves the spatiotemporal resolution of the simulation, provides real-time support for urban flood control decision-making; updates the rainfall spatiotemporal distribution using meteorological radar data, and combines with real-time monitoring of the surface water accumulation condition by the feedback of the IoT sensor; enables the simulation system to have dynamic updating capability, enhances the response speed and prediction accuracy to the sudden rainstorm event; improves the real-time and practicality of the system, and provides technical support for urban flood control planning and emergency management.

[0082] Embodiment 6: as shown in the embodiment 5, on the basis of the embodiment 5, the process of rainstorm- runoff-pipe network three-dimensional coupling provided by the embodiment of the application comprises the following steps: Figure 6

[0083] S2021: the change of the surface water depth with time is determined by the net rainfall and the pipe network inlet flow, the flow distribution of the surface runoff reflects the aggregation and diffusion process of the rainfall on the ground, and the pipe network inlet flow represents the rate of the water accumulation into the underground pipe network, and the dynamic process of the formation, diffusion and recession of the surface water accumulation is simulated;

[0084] S2022: the flow dynamics of the underground pipe network reflects the change of the flow in the pipe with time, the propagation of the water flow in the pipe is affected by the gravity, the pipe cross-sectional area and the water head height gradient; at the same time, the pipe network inlet flow as the interaction point of the surface runoff and the pipe network system depicts the hydraulic response of the underground pipe network, including the change of the pipe flow, the distribution of the node pressure and the dynamic adjustment of the pump station operation state;

[0085] S2023: the dynamic interaction of the surface water accumulation and the pipe network flow is realized through a set of iterative equations, in each iteration, the surface water depth is updated according to the net rainfall and the pipe network inlet flow, and the pipe network inlet flow is connected with the surface water and the pipe flow through the coupling function; the pipe network flow is adjusted according to the gravity driving and the surface runoff input; through multiple iterations, the dynamic relationship between the surface water accumulation and the pipe network flow gradually approaches the stable state until the simulation result converges.

[0086] Wherein, the dynamic change of the surface water accumulation is represented by the following equation:

[0087]

[0088] In the formula, H surf (x, y, t) represents the surface water depth (unit: m), and represents the water accumulation at the position (x, y) at time t; Q surf (x, y, t) represents the surface runoff flow (unit: m 3 / s), and represents the runoff rate at the position (x, y) at time t; R net (x, y, t) represents the net rainfall (unit: m​3 Q (x, y, t) = P (x, y, t) - I (x, y, t) inlet (x, y, t) represents the pipe flow (unit: m 3 / s) at location (x, y) at time t; this equation describes the dynamic changes of the surface water, including the rainfall input, runoff diffusion, and feedback of the pipe drainage.

[0089] The flow dynamics of the pipe network are represented by the following equation:

[0090]

[0091] where Q pipe (x, y, z, t) represents the pipe flow (unit: m 3 / s) at location (x, y, z) at time t; c wave represents the wave speed of the water flow (unit: m / s), representing the propagation speed of the water flow in the pipe; g represents the acceleration of gravity (unit: m / s 2 ), a constant approximately equal to 9.81; A pipe (x, y, z) represents the pipe cross-sectional area (unit: m 2 ), representing the pipe area at location (x, y, z); H pipe (x, y, z, t) represents the pipe water head height (unit: m), representing the water head at location (x, y, z) at time t; Q surf (x, y, t) represents the surface runoff flow (unit: m 3 / s) at location (x, y) at time t; this equation describes the dynamic changes of the flow in the pipe network, including the water flow propagation, gravity driving, and input of the surface runoff.

[0092] The coupling of the surface water and the pipe network flow is achieved by the following iterative equation:

[0093]

[0094] where n represents the iteration number, representing the current iteration step; Δt represents the time step (unit: s), representing the time interval of each iteration; f coupling represents the coupling function, representing the mathematical relationship between the surface water and the pipe network flow; represents the surface water depth at the n-th iteration (unit: m); represents the pipe network inlet flow at the n-th iteration (unit: m 3 / s); represents the pipe flow at the n-th iteration (unit: m 3 / s); This represents the surface runoff flow rate at the (n+1)th iteration (unit: m³). 3 / s). This iterative equation realizes bidirectional data interaction between surface water accumulation and pipe network flow until the simulation results converge. The above set of equations constitutes the core mathematical model of the three-dimensional coupling of stormwater-runoff-pipe network. Through complex physical mechanisms and iterative calculations, it realizes the dynamic interaction between the surface and underground systems, ensuring the accuracy and stability of the simulation results. It achieves real-time bidirectional coupling between surface water accumulation (two-dimensional spatial diffusion) and underground pipe network flow (three-dimensional pipe network) through iterative equations, breaking through the one-way input mode of separation or simplification of surface and pipe network in traditional models. It realizes bidirectional data interaction between surface and pipe network through coupling functions and step-by-step iteration, solving the numerical instability problem caused by time step mismatch in traditional models. It introduces water flow wave velocity and gravitational acceleration into the pipe network flow equation, combined with the pipe cross-sectional area, to fully characterize the dynamic characteristics of non-steady flow (such as pressure wave propagation and head gradient change) in the pipe.

[0095] The working principle and beneficial effects of the technical solution are: firstly, the change of the surface water depth with time is determined by the net precipitation and the pipe network inlet flow, and the flow distribution of the surface runoff reflects the aggregation and diffusion process of the rainfall on the ground surface, and the pipe network inlet flow represents the rate of the accumulated water entering the underground pipe network, simulating the dynamic process of the formation, diffusion and recession of the surface accumulated water; secondly, the flow dynamics of the underground pipe network reflects the change of the flow in the pipeline with time, and the propagation of the water flow in the pipeline is affected by gravity, the cross-sectional area of the pipeline and the water head height gradient; at the same time, the pipe network inlet flow as the interaction point of the surface runoff and the pipe network system describes the hydraulic response of the underground pipe network, including the change of the pipeline flow, the distribution of the node pressure and the dynamic adjustment of the pump station operation state; finally, the dynamic interaction of the surface accumulated water and the pipe network flow is realized through a set of iterative equations, in each iteration, the surface accumulated water depth is updated according to the net precipitation and the pipe network inlet flow, and the pipe network inlet flow is coupled with the surface accumulated water and the pipeline flow through a coupling function; the pipe network flow is adjusted according to the gravity driving and the surface runoff input; through multiple iterations, the dynamic relationship between the surface accumulated water and the pipe network flow gradually approaches a stable state until the simulation result converges. The step S2021 of the simulation of the dynamic process of the surface accumulated water is determined by the net precipitation (rainfall minus infiltration) and the pipe network inlet flow, and reflects the influence of the rainfall input and the drainage capacity on the surface accumulated water; the net precipitation is the initial driving condition of the surface accumulated water, and the pipe network inlet flow reflects the feedback effect of the underground system on the surface accumulated water; the dynamic change process of the surface accumulated water is quantified to provide input data for the subsequent coupling with the underground pipe network. The flow distribution of the surface runoff reflects the aggregation and diffusion process of the rainfall on the ground surface, which is affected by factors such as topography, soil properties and ground roughness, and by simulating the runoff distribution, the spatio-temporal evolution of the accumulated water area and depth is described; the pipe network inlet flow provides a spatial distribution input condition, supporting the interaction between the surface and the underground system. The pipe network inlet flow represents the rate of the accumulated water entering the underground pipe network, which is the direct interaction point of the surface runoff and the pipe network system, connecting the surface accumulated water and the underground pipe network, and providing key connection data for the coupling; the formation, diffusion and recession process of the surface accumulated water under the rainstorm event is simulated to provide boundary conditions for the dynamic response of the pipe network system.Step S2022 simulating the dynamic process of pipe network flow, the dynamic change of pipe flow, reflecting the transmission process of water flow in the pipe, affected by gravity, pipe cross-sectional area and water head height gradient, the space-time distribution and change trend of pipe flow are described by solving the hydraulic equation; quantifying the hydraulic response of underground pipe network, providing calculation basis for feedback of surface water accumulation; the role of pipe network inlet flow as the interaction point of surface runoff and pipe network system directly affects the hydraulic state of pipe network, realizes the dynamic data exchange of surface and underground system, ensures the integrity of coupling; node pressure and pump station operation state reflect the pressure distribution and pump station operation efficiency of underground drainage system, affect the drainage capacity of pipe network, evaluate the performance of underground system, provide technical basis for decision making; simulate the dynamic response of underground pipe network, including real-time change of flow, pressure and pump station state, support the coupling of surface and underground system. Step S2023 dynamic coupling of surface water accumulation and pipe network flow, construction of iterative equation, through iterative calculation to realize the dynamic adjustment of surface water depth and pipe network flow, in each iteration, the surface water depth is updated according to the net precipitation and pipe network inlet flow, to ensure that the dynamic interaction process of surface water accumulation and pipe network flow can be accurately simulated. The role of coupling function, the mathematical relationship between surface water accumulation and pipe network flow is established, the two-way data exchange is realized, the synergy of surface and underground system is optimized, and the accuracy of simulation is improved. Convergence condition, through multiple iterations, adjust the dynamic relationship between surface water and pipe network flow, until the simulation result converges; ensure the stability and reliability of simulation result, provide high quality data support for decision making; realize the deep integration of surface water and pipe network flow, ensure that the overall influence of rainstorm event on urban hydrological system can be accurately expressed.

[0096] In the embodiment 5, the process of simulation and deduction at the frequency of minutes provided by the present application comprises the following steps: Figure 7 In the embodiment 5, the process of simulation and deduction at the frequency of minutes provided by the present application comprises the following steps:

[0097] S2031: obtain rainfall intensity, spatial distribution and evolution trend data in minutes, and transmit the data to the rainstorm-runoff-pipe network three-dimensional coupling simulation system in real time; use Kriging interpolation to convert discrete data into continuous spatial distribution data;

[0098] S2032: obtain water depth, pipe flow and node pressure data in minutes, and transmit the data to the rainstorm-runoff-pipe network three-dimensional coupling simulation system in real time; use linear interpolation to fill in the abnormal values;

[0099] S2033: unify the time stamps of meteorological radar data and IoT sensor data to the same time reference, and fuse the meteorological radar data and IoT sensor data.

[0100] The working principle and beneficial effects of the above technical solution are as follows: Firstly, this embodiment acquires rainfall intensity, spatial distribution, and evolution trend data on a minute-by-minute basis, and transmits the data in real-time to the three-dimensional coupled simulation system of rainstorm-runoff-pipeline network. Kriging interpolation is used to convert discrete data into continuous spatial distribution data. Secondly, water accumulation depth, pipeline flow rate, and node pressure data are acquired on a minute-by-minute basis and transmitted in real-time to the rainstorm-runoff-pipeline network three-dimensional coupled simulation system. Outliers are filled using linear interpolation. Finally, the timestamps of meteorological radar data and IoT sensor data are unified to the same time base, and the meteorological radar data and IoT sensor data are fused. Step S2031 of the above solution, rainfall data acquisition and spatial interpolation, ensures the real-time nature of rainfall information through high-frequency data acquisition, providing dynamically updated input conditions for the simulation system; spatial interpolation improves the spatial continuity of the data and enhances the accuracy of rainfall distribution simulation. Step S2032 Sensor Data Acquisition and Outlier Processing: Through outlier detection and correction, errors caused by sensor malfunctions or environmental interference are eliminated, enhancing data reliability. This provides high-precision local state data for the pipeline hydraulic model and surface runoff model, supporting refined simulation of water accumulation and pipeline dynamics. Step S2033 Data Fusion and Spatiotemporal Alignment: Through spatiotemporal alignment and fusion, information from different data sources is integrated, providing more comprehensive and consistent input data. Combining the advantages of radar and sensor data compensates for the shortcomings of single data sources, improving the accuracy of simulation results. It provides consistent boundary conditions for the dynamic coupling of surface and underground systems, ensuring the convergence and reliability of the simulation process.

[0101] In summary, this embodiment provides high-precision, dynamic rainfall input to the simulation system through rainfall data acquisition and spatial interpolation; ensures data reliability and refined simulation requirements through sensor data acquisition and outlier handling; and integrates multi-source data through data fusion and spatiotemporal alignment to support dynamic coupling and high-precision simulation of the system. This ensures the system's real-time performance, accuracy, and robustness.

[0102] Example 8: As Figure 8 As shown, based on Example 1, the process of generating pump station scheduling schemes, traffic control strategies, and material delivery routes provided in this embodiment of the invention includes the following steps:

[0103] S301: Using the water diffusion rate output from the urban stormwater analysis model as input, combined with water depth, spatial distribution and evolution trend, the severity of the current urban flooding is assessed; based on the digital twin holographic base, the real-time operation status of pumping stations, traffic network flow and material reserve information are obtained to assess the dispatchability of pumping stations, the feasibility of traffic networks and the availability of materials.

[0104] S302: Based on the assessment results of urban flooding status, pumping station operation status, traffic network flow, and material reserve information, construct a multi-objective optimization problem with objectives including flood control benefits, traffic efficiency, and resource utilization; solve the multi-objective optimization problem to generate the optimal decision-making scheme.

[0105] S303: The generated pump station scheduling plan, traffic control strategy, and material delivery route will be distributed to relevant execution units through the Internet of Things platform, including: pump station control system, traffic management system, and emergency material dispatch system.

[0106] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first uses the water diffusion rate output from the urban stormwater analysis model as input, and combines the water depth, spatial distribution and evolution trend to assess the severity of the current urban flooding situation; based on the digital twin holographic base, it acquires the pump station operation status, traffic network flow and material reserve information in real time to assess the pump station dispatchability, traffic network feasibility and material availability; secondly, based on the assessment results of the urban flooding situation and the pump station operation status, traffic network flow and material reserve information, it constructs a multi-objective optimization problem, with objectives including: flood control benefits, traffic efficiency and resource utilization rate; solves the multi-objective optimization problem to generate the optimal decision scheme; finally, the generated pump station dispatch scheme, traffic control strategy and material delivery path are distributed to relevant execution units through the Internet of Things platform, including: pump station control system, traffic management system and emergency material dispatch system. The above solution achieves intelligent management of pump station scheduling, traffic control, and material delivery through data input and status assessment, multi-objective optimization and decision generation, and decision execution and feedback. It combines real-time assessment of urban flooding status, infrastructure status monitoring, multi-objective optimization solutions, and IoT execution feedback to ensure the efficiency, accuracy, and real-time nature of emergency management for urban flooding risks, providing strong technical support for safe urban operation.

[0107] Example 9: As Figure 9 As shown, based on Examples 1-8, the urban flooding risk management system based on multi-model coupling provided in this embodiment of the invention includes:

[0108] The model building module is responsible for constructing a digital twin holographic foundation based on urban geographic information, pipeline topology, and meteorological and hydrological data. It deploys a distributed big data framework to enable real-time access to water accumulation report data from meteorological radar, IoT sensors, and social media. Through the deep coupling of the water accumulation identification intelligent model and the physical mechanism-driven stormwater simulation engine, it constructs an urban stormwater analysis model, generates a high-precision urban flood risk heat map and dynamic forecast thresholds, and broadcasts the risk situation in real time through the voice interaction module.

[0109] a data analysis module responsible for building a storm-runoff-pipe network three-dimensional coupled simulation system relying on the digital twin holographic base, updating and deducing results at a minute level frequency, and real-time delivering rainfall spatio-temporal distribution data to the urban rain flood analysis model; when the waterlogging diffusion rate output by the urban rain flood analysis model reaches a preset critical value, triggering a red / orange / yellow three-level early warning signal;

[0110] a collaborative decision-making module responsible for establishing a multi-objective collaborative decision-making engine, inputting the waterlogging diffusion rate into the multi-objective collaborative decision-making engine, generating a pump station dispatching scheme, a traffic control strategy, and a material delivery path, and returning to the digital twin holographic base through the Internet of Things middle station.

[0111] The working principle and beneficial effects of the above technical solution are: the model construction module of the embodiment constructs a digital twin holographic base based on city geographic information, pipe network topology and meteorological and hydrological data, and realizes real-time access of meteorological radar, IoT sensors and social media waterlogging report data by deploying a distributed big data framework; a city rainstorm analysis model is constructed through deep coupling of a waterlogging identification intelligent model and a physical mechanism driven rainstorm simulation engine, a high-precision waterlogging risk heat map and a dynamic prediction threshold are generated, and a risk situation is broadcast in real time through a voice interaction module; the data analysis module constructs a storm-runoff-pipe network three-dimensional coupled simulation system relying on the digital twin holographic base, updates and deduces results at a minute level, and real-time delivers rainfall spatio-temporal distribution data to the city rainstorm analysis model; when the waterlogging diffusion rate output by the city rainstorm analysis model reaches a preset critical value, a red / orange / yellow three-level early warning signal is triggered; the collaborative decision-making module establishes a multi-objective collaborative decision-making engine, inputs the waterlogging diffusion rate into the multi-objective collaborative decision-making engine, generates a pump station scheduling scheme, a traffic control strategy and a material delivery path, and returns to the digital twin holographic base through the Internet of Things middle station. The model construction module of the above scheme realizes data integration and real-time access, deploys a distributed big data framework, realizes efficient access of meteorological radar, IoT sensors and social media waterlogging report data, and ensures the real-time and consistency of multi-source heterogeneous data; model construction and intelligent analysis, through deep coupling of a waterlogging identification intelligent model and a physical mechanism driven rainstorm simulation engine, a city rainstorm analysis model is constructed, not only improving the calculation accuracy of the model, but also realizing dynamic simulation of waterlogging in complex urban environment; risk visualization and early warning broadcast, a high-precision waterlogging risk heat map and a dynamic prediction threshold are generated, and a risk situation is broadcast in real time through a voice interaction module, providing an intuitive decision-making basis for city managers. The data analysis module realizes real-time deduction and dynamic update, updates and deduces results at a minute level, realizes dynamic simulation of rainfall spatio-temporal distribution and waterlogging formation process; waterlogging propagation analysis, real-time delivery of rainfall spatio-temporal distribution data to the city rainstorm analysis model, accurate prediction of waterlogging propagation range and trend combined with changes in waterlogging diffusion rate; early warning mechanism triggering, when the waterlogging diffusion rate reaches the preset critical value, a red, orange, yellow three-level early warning signal is triggered, helping city managers quickly identify the risk level and take appropriate measures. The collaborative decision-making module realizes multi-objective optimization and strategy generation, inputs the waterlogging diffusion rate into the multi-objective collaborative decision-making engine, considers multiple targets such as pump station scheduling, traffic control and material delivery path, and generates an optimal emergency scheduling scheme; closed-loop management and dynamic adjustment, the generated scheduling scheme is returned to the digital twin holographic base through the Internet of Things middle station, forming a closed-loop management mechanism, not only realizing dynamic allocation of emergency resources, but also ensuring the real-time and operability of the decision; collaborative response and efficiency improvement, through the optimization calculation of the multi-objective collaborative decision-making engine, the efficiency and accuracy of the city waterlogging emergency response are significantly improved, and the impact of waterlogging on city operation is minimized.

[0112] In summary, the data analysis module of the embodiment realizes real-time deduction and early warning of waterlogging risk, and the collaborative decision-making module generates efficient emergency strategies through multi-objective optimization; the three modules work together to form a complete urban waterlogging risk management system, providing urban managers with full-process technical support from data access to risk early warning to decision response.

[0113] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the equivalent technology of the present application, the present application also intends to include these modifications and variations.

Claims

1. A multi-model coupling-based urban waterlogging risk management method, characterized in that, Comprising the following steps: Based on urban geographic information, network topology and meteorological and hydrological data, a digital twin holographic base is constructed, and a distributed big data framework is deployed to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data; through the deep coupling of the waterlogging identification intelligent model and the physical mechanism driven rainstorm simulation engine, a city rainstorm analysis model is constructed to generate high-precision waterlogging risk heat maps and dynamic prediction thresholds, and the risk situation is broadcast in real time through the voice interaction module; Relying on the digital twin holographic base, a rainstorm-runoff-pipe network three-dimensional coupled simulation system is constructed to update and deduce the results at a minute level, and the rainfall spatio-temporal distribution data is transmitted to the city rainstorm analysis model in real time; when the waterlogging diffusion rate output by the city rainstorm analysis model reaches the preset critical value, a red / orange / yellow three-level early warning signal is triggered; The process of constructing the rainstorm-runoff-pipe network three-dimensional coupled simulation system comprises the following steps: based on the physical mechanism of the precipitation process and the surface runoff model, a hydrological model under rainstorm events is constructed to simulate the distribution and aggregation of rainfall on the ground surface; The dynamic process of waterlogging formation and diffusion is depicted through the surface runoff model, and the development trend of waterlogging is simulated in combination with the terrain conditions; the dynamic response of the underground drainage system is simulated by using the pipe network hydraulic model; the surface runoff model and the pipe network hydraulic model are synchronized in dynamic response, and a bidirectional data exchange mechanism is established between the surface runoff model and the pipe network hydraulic model; the dynamic relationship between surface waterlogging and pipe network flow is adjusted through iteration until the simulation result converges; that is, rainstorm-runoff-pipe network three-dimensional coupling; the rainstorm-runoff-pipe network three-dimensional coupled simulation system performs simulation deduction at a minute level; each round of deduction is based on the latest meteorological radar data and IoT sensor feedback to dynamically update the rainfall spatio-temporal distribution and surface waterlogging conditions; The process of forming the city rainstorm analysis model comprises that the real-time waterlogging position and range information output by the waterlogging identification intelligent model is converted into input conditions of the rainstorm simulation engine for correcting the initial state and boundary conditions of the simulation; the rainwater flow direction, flow and waterlogging depth key parameters generated by the rainstorm simulation engine are fed back to the waterlogging identification model to distinguish noise signals from real waterlogging signals in the IoT sensor data; the deviation of the results output by the waterlogging identification model and the rainstorm simulation engine is compared in real time, and when the result deviation exceeds the preset threshold, a dynamic adjustment mechanism is started to preferentially correct problems in the data source; the output data of the waterlogging identification model and the rainstorm simulation engine are fused into the digital twin holographic base to form a city rainstorm analysis model including spatial position and range labeling, time sequence evolution and mechanism driven multi-dimensional reconstruction, as well as dynamic restoration and prediction of city rainstorm scenarios; A multi-objective collaborative decision engine is established, the waterlogging diffusion rate is input into the multi-objective collaborative decision engine to generate a pump station dispatching scheme, a traffic control strategy and a material delivery path, which are fed back to the digital twin holographic base through the Internet of Things middle station.

2. The multi-model coupling based urban waterlogging risk management method of claim 1, wherein, The process of constructing the city rainstorm analysis model comprises the following steps: Acquire urban geographic information data including high-resolution topographic map, building distribution and road network; acquire pipe network topology data including drainage pipe network, rainwater box culvert and pump station location; acquire meteorological and hydrological data including historical rainfall, river water level and groundwater level; integrate into the three-dimensional digital twin platform to build a digital twin holographic base; Based on real-time access of meteorological radar data, IoT sensor data and social media waterlogging report data, train waterlogging identification intelligent model, locate waterlogging area in the city through feature extraction and pattern recognition, and mark waterlogging location on digital twin holographic base in real time; Based on the physical mechanism of meteorological and hydrological data, build rain and flood simulation engine, deduce the formation and diffusion process of waterlogging by solving the coupled equations of rainfall-runoff-pipe network, generate rainwater flow direction, flow and waterlogging depth key parameters; Deeply couple waterlogging identification intelligent model and physical mechanism driven rain and flood simulation engine to form urban rain and flood analysis model.

3. The multi-model coupling based urban waterlogging risk management method of claim 2, wherein, The process of locating waterlogging area in the city includes the following steps: Fuse meteorological radar data, IoT sensor data and social media waterlogging report data, extract waterlogging related features, including preliminary prediction of possible waterlogging area, verification of specific location and specific spatial coordinate information of waterlogging occurrence; Perform pattern recognition and spatial correlation analysis on the extracted features, integrate meteorological radar data, IoT sensor data and social media waterlogging report data into waterlogging identification model; waterlogging identification model learns from historical waterlogging events, identifies typical patterns of waterlogging occurrence, and dynamically updates prediction results of waterlogging area through continuous input of real-time meteorological radar data, IoT sensor data and social media waterlogging report data.

4. The multi-model coupling based urban waterlogging risk management method of claim 3, wherein, Capture rainfall intensity and coverage range through meteorological radar data; deploy IoT sensor network at key points in the city to monitor real-time ground water level, drainage pipe network flow and local humidity information; waterlogging reports uploaded by users on social media reflect waterlogging phenomena in the city.

5. The multi-model coupling based urban waterlogging risk management method of claim 1, wherein, Surface runoff model outputs waterlogging depth and flow as input conditions for pipe network hydrology; pipe network hydrology model outputs node pressure and pipe flow to reflect feedback of drainage capacity on surface waterlogging.

6. The multi-model coupling based urban waterlogging risk management method of claim 1, wherein, Dynamic response includes pipe flow change, node pressure distribution and pump station operating state.

7. The multi-model coupling based urban waterlogging risk management method of claim 1, wherein, The process of simulation and deduction at minute level frequency includes the following steps: Acquire rainfall intensity, spatial distribution and evolution trend data in minutes, transmit data to rain-runoff-pipe three-dimensional coupling simulation system in real time; use Kriging interpolation to convert discrete data into continuous spatial distribution data; Acquire waterlogging depth, pipe network flow and node pressure data in minutes, transmit data to rain-runoff-pipe three-dimensional coupling simulation system in real time; use linear interpolation to fill in abnormal values; Unify time stamps of meteorological radar data and IoT sensor data to the same time reference, fuse meteorological radar data and IoT sensor data.

8. A multi-model coupling based urban waterlogging risk management system, characterized in that, Include: The model construction module is responsible for constructing a digital twin holographic base based on city geographic information, pipe network topology and meteorological and hydrological data, deploying a distributed big data framework to realize real-time access of meteorological radar, IoT sensors and social media waterlogging report data; through deep coupling of the waterlogging identification intelligent model and the physical mechanism driven rainstorm simulation engine, a city rainstorm analysis model is constructed to generate high-precision waterlogging risk heat maps and dynamic prediction thresholds, and the risk situation is reported in real time through a voice interaction module; The data analysis module is responsible for relying on the digital twin holographic base to construct a storm-runoff-pipe network three-dimensional coupled simulation system, updating and deducing results at a minute level, and real-time transmission of rainfall spatio-temporal distribution data to the city rainstorm analysis model; when the waterlogging diffusion rate output by the city rainstorm analysis model reaches the preset critical value, a red / orange / yellow three-level early warning signal is triggered; The process of constructing the storm-runoff-pipe network three-dimensional coupled simulation system includes the following steps: based on the physical mechanism of the precipitation process and the surface runoff model, a hydrological model under storm events is constructed to simulate the distribution and aggregation process of rainfall on the ground surface; The dynamic process of waterlogging formation and diffusion is described by the surface runoff model, and the development trend of waterlogging is simulated in combination with the terrain conditions; the dynamic response of the underground drainage system is simulated by using the pipe network hydraulic model; the surface runoff model and the pipe network hydraulic model are synchronized in dynamic response, and a bidirectional data exchange mechanism is established between the surface runoff model and the pipe network hydraulic model; the dynamic relationship between surface waterlogging and pipe network flow is adjusted through iteration until the simulation result converges; that is, the storm-runoff-pipe network three-dimensional coupling; the storm-runoff-pipe network three-dimensional coupled simulation system performs simulation and deduction at a minute level; each round of deduction is based on the latest meteorological radar data and IoT sensor feedback to dynamically update the rainfall spatio-temporal distribution and surface waterlogging conditions; the process of forming the city rainstorm analysis model includes that the real-time waterlogging position and range information output by the waterlogging identification intelligent model is converted into the input conditions of the rainstorm simulation engine for correcting the initial state and boundary conditions of the simulation; the rainwater flow direction, flow and waterlogging depth key parameters generated by the rainstorm simulation engine are fed back to the waterlogging identification model to distinguish the noise signals from the real waterlogging signals in the IoT sensor data; the result deviation of the output of the waterlogging identification model and the rainstorm simulation engine is compared in real time, and when the result deviation exceeds the preset threshold, the dynamic adjustment mechanism is started to preferentially correct the problems in the data source; the output data of the waterlogging identification model and the rainstorm simulation engine are fused into the digital twin holographic base to form the city rainstorm analysis model including spatial position and range labeling, time sequence evolution and mechanism driven multi-dimensional reconstruction, and dynamic restoration and prediction of city rainstorm scenarios; The collaborative decision-making module is responsible for establishing a multi-objective collaborative decision-making engine, inputting the waterlogging diffusion rate into the multi-objective collaborative decision-making engine to generate a pump station scheduling scheme, a traffic control strategy and a material delivery path, and returning to the digital twin holographic base through the Internet of Things middle station.

Images