An integrated system and method for urban flood control early warning and emergency command

Abstract

The present invention relates to the technical field of flood control early warning and emergency command, and particularly to an integrated system and method for urban flood control early warning and emergency command. The multi-source data acquisition module collects meteorological, hydrological, urban drainage pipe network, and video monitoring data in real time, and uses mobile terminals to report data. The dynamic early warning analysis module couples the meteorological time series prediction value based on the gated recurrent unit, the output of the drainage pipe network hydraulic model, and the terrain influence factor function, and calibrates the weight coefficient in real time through the KL divergence to generate a high-precision dynamic early warning map. The three-dimensional visualization command module generates a three-dimensional model of the dynamic inundation area, and uses the particle swarm optimization algorithm to add a new danger gradient term to realize the visualization of path planning. In the intelligent scheduling decision module, the improved ant colony optimization algorithm combines environmental risk factors to realize the dynamic coupling of path planning and real-time disaster changes. The present invention greatly improves the accuracy of urban flood control early warning and the efficiency of emergency command, and effectively responds to urban flood disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of flood control warning and emergency command, and specifically provides an integrated system and method for urban flood control warning and emergency command. Background Technique

[0002] With the global climate change, extreme weather events have become increasingly frequent, and the probability and intensity of urban flood disasters have increased significantly. The rapid development of urban construction has changed the original underlying surface conditions, with a substantial increase in impervious areas and an increase in the rainfall runoff coefficient, putting unprecedented pressure on the urban drainage system.

[0003] There are many deficiencies in existing urban flood control warning systems. On the one hand, the multi-source data collection lacks systematicness and real-time nature. Meteorological departments, hydrological monitoring stations, drainage management departments, etc. collect data independently, with inconsistent data formats and update frequencies, making it difficult to achieve efficient integration and sharing. For example, meteorological data may not be able to accurately and timely reflect the changes in the local microclimate of the city, and hydrological data has a lag in monitoring the water flow state inside the urban drainage network, resulting in deviations in the judgment of flood conditions.

[0004] On the other hand, the accuracy of the warning analysis model is poor. The coupling degree of traditional meteorological prediction models and urban waterlogging models is low, and it is impossible to comprehensively consider the interaction of multiple factors such as meteorological factors, topography, and the operation status of drainage networks. In some complex terrain areas, existing models are difficult to accurately simulate the formation and diffusion process of water accumulation, and the warning maps cannot accurately reflect the actual flood conditions, resulting in insufficient reliability of warning information and being unable to provide strong support for flood control decision-making.

[0005] In the emergency command link, the problems are also prominent. It is difficult to cooperate among various departments, and there is a lack of a unified visual command platform. Departments such as urban planning, transportation, and emergency rescue have poor information communication and low command and dispatch efficiency when dealing with flood conditions. For example, when allocating emergency resources, due to the lack of real-time and comprehensive understanding of traffic conditions, resource distribution, and the needs of affected areas, the resource allocation is unreasonable and the transportation route planning is poor, making it difficult to deliver resources to the affected locations within the best time, seriously affecting the timeliness and effectiveness of emergency rescue work. At the same time, the communication guarantee between the on-site and the command center is unstable, and communication interruptions often occur in extreme weather, affecting the timely feedback of on-site information and the transmission of command instructions. Therefore, an integrated system and method for urban flood control warning and emergency command are proposed to address the above problems. Summary of the Invention

[0006] The purpose of the present invention is to provide an integrated system and method for urban flood control warning and emergency command to solve the problems raised in the above background technique.

[0007] To achieve the above purpose, the present invention provides the following technical solutions:

[0008] An integrated urban flood prevention warning and emergency command system, comprising:

[0009] Multi-source data acquisition module: used to obtain meteorological data, hydrological data, urban drainage network data and video surveillance data in real time, and report data using mobile terminals;

[0010] Dynamic warning analysis module: generates a dynamic warning map by coupling a meteorological prediction model and an urban waterlogging model, and the coupled meteorological prediction model uses a machine learning algorithm to correct prediction parameters in real time;

[0011] Three-dimensional visualization command module: integrates the BIM urban model and GIS geographic information data to realize three-dimensional spatial visualization display of flood conditions and a multi-department collaborative operation interface;

[0012] Intelligent scheduling decision module: includes an emergency resource path planning unit based on an improved ant colony optimization algorithm and a resource matching decision unit based on dynamic weight multi-objective optimization;

[0013] Cross-platform communication module: supports dual-link guarantee of 5G-MEC edge computing and satellite communication, and realizes multimedia transmission of command instructions and augmented reality AR interaction of on-site terminals.

[0014] As a preferred solution, the dynamic warning analysis module adopts an improved spatio-temporal fusion prediction model, and its core algorithm is defined as:

[0015] ,

[0016] Where, is the predicted value at time; , , are dynamic weight coefficients, which are calibrated in real time through KL divergence; is the meteorological time series predicted value based on the gated recurrent unit, is the meteorological data; is the rate of change of the output of the drainage network hydraulic model with respect to time, is the pipeline water level height, is the surface runoff flow; is the terrain influence integral value, is the terrain influence factor function, and are the integration intervals.

[0017] As a preferred solution, the improved ant colony optimization algorithm in the intelligent scheduling decision module satisfies:

[0018] ,

[0019] Among them, is the pheromone concentration on the time path ; is the pheromone concentration on the time path ; is the pheromone evaporation coefficient; is the th ant's pheromone increment left on the path ; is the total number of ants; is the environmental risk adjustment factor; is the combined risk assessment value of road water depth and bridge load-bearing capacity at time

[0020] As a preferred solution, the resource matching decision-making unit adopts a dynamic weight multi-objective optimization model:

[0021] ,

[0022] Among them, , , are dynamically adjusted weight parameters, linked to the early warning level; is the emergency response time; is the emergency resource cost; is the emergency resource allocation efficiency;

[0023] Constraint conditions:

[0024] , among which, is the quantity of the th type of emergency resource; is the number of types of emergency resources; is the total demand for emergency resources; is the demand volatility, which is updated in real time according to public opinion data; is the arrival time of emergency resources; is the length of the emergency resource transportation path; is the emergency resource transportation speed at time is the path water accumulation influence coefficient; is the road section water depth.

[0025] As a preferred solution, in the spatio-temporal fusion prediction model:

[0026] The LSTM network is improved by adopting a dual-channel attention mechanism, and its attention weight calculation is:

[0027] , where is the attention weight at time , , are trainable parameter matrices; is the hidden layer state at time is the input feature at the current time; is the bias term;

[0028] The SWMM model introduces a pipeline sedimentation correction factor:

[0029] , where is the actual flow; is the theoretical flow; is the pipeline aging coefficient; is the maintenance impact factor; is the time since the last maintenance.

[0030] As a preferred solution, the 3D visualization command module includes:

[0031] Generating a 3D model of the dynamic inundation area based on the improved Marching Cubes algorithm, and its isosurface extraction formula is:

[0032] , where is the isosurface function value at the spatial point ; is the octree node weight; is the improved Sigmoid function, which increases the response to terrain mutations; is the water level height at the grid point; is the water level threshold;

[0033] Path planning visualization uses the particle swarm optimization algorithm to generate a navigation beam:

[0034] ,

[0035] where is the -th generation and the -th particle's velocity in the -th dimension; is the inertia weight; is the -th generation and the -th particle's velocity in the -th dimension; , , are learning factors; , , is a random number within the range; is the th particle's individual optimal position in the th dimension; is the th generation's th particle's position in the th dimension; is the global optimal position's coordinate in the th dimension; is the danger gradient term, which can avoid high-risk areas in real time.

[0036] As a preferred solution, it further includes:

[0037] An evaluation and feedback module that calculates the comprehensive score of the disposal plan using the improved TOPSIS algorithm:

[0038] , where is the comprehensive score of the th disposal plan; is the weight of the th indicator, and a time decay function is introduced to dynamically adjust it; is the distance from the th disposal plan to the positive ideal solution on the th indicator; is the distance from the th disposal plan to the negative ideal solution on the th indicator; is the risk penalty factor; is the number of indicators.

[0039] An urban flood prevention warning and emergency command method, which uses an urban flood prevention warning and emergency command integrated system to conduct urban flood prevention warning and emergency command.

[0040] It can be seen from the technical solution provided by the present invention above that the beneficial effects of the urban flood prevention warning and emergency command integrated system and method provided by the present invention are:

[0041] High-precision warning: The dynamic warning analysis module integrates meteorological prediction and urban waterlogging model, and uses an innovative spatio-temporal fusion prediction model, which combines LSTM meteorological time series prediction, the output of the drainage pipe network hydraulic model, and the terrain influence factor function; through KL divergence, the dynamic weight coefficient is calibrated in real time, enabling the model to flexibly adjust the weights of various factors according to different scenarios, greatly improving the accuracy and timeliness of waterlogging warning, and providing solid and reliable data support for flood prevention decision-making;

[0042] Efficient resource allocation: In the intelligent scheduling decision module, the path planning unit of the improved ant colony optimization algorithm innovatively introduces an environmental risk adjustment factor that is positively correlated with the real-time rainfall intensity, and combines the joint risk assessment value of road waterlogging depth and bridge load-bearing capacity, enabling the emergency resource transportation path to avoid dangerous areas in real time, ensuring efficient and safe transportation; at the same time, when constructing the resource matching decision unit, a three-objective optimization system with time-varying weights is built, and the weight parameters are dynamically adjusted according to the warning level, comprehensively weighing the emergency response time, resource cost, and allocation efficiency to achieve the optimal allocation of emergency resources;

[0043] Intuitive and visual command: The three-dimensional visual command module integrates the BIM city model and GIS geographic information data. With the dynamic inundation area three-dimensional model generated by the improved Marching Cubes algorithm and the use of the improved Sigmoid function to enhance the response to terrain mutations, the inundation area is displayed more accurately and intuitively; for path planning visualization, the particle swarm optimization algorithm is used to generate navigation beams, and a new dangerous gradient term is added to avoid high-risk areas in real time, providing clear scheduling guidance for commanders and effectively improving the efficiency and accuracy of multi-department collaborative operations;

[0044] Reliable communication guarantee: The cross-platform communication module adopts a dual-link guarantee mechanism of 5G-MEC edge computing and satellite communication to ensure the stable and reliable multimedia transmission of command instructions in complex and harsh environments; the augmented reality interaction function of on-site terminals enables on-site staff to intuitively obtain command information and on-site flood situation data, significantly improving the timeliness and accuracy of emergency response;

[0045] Continuous optimization decision: The evaluation and feedback module uses the improved TOPSIS algorithm to calculate the comprehensive score of the disposal plan, dynamically adjusts the weights by introducing a time decay function, and at the same time considers the risk penalty factor to comprehensively and dynamically evaluate different disposal plans; this provides strong data support for the improvement and optimization of subsequent flood control work, promoting the continuous improvement of the overall level of urban flood control early warning and emergency command. Brief description of the drawings

[0046] Figure 1 It is a schematic diagram of the overall structure of an integrated system and method for urban flood control early warning and emergency command of the present invention. Detailed implementation manners

[0047] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0048] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the specification drawings and specific implementation manners.

[0049] As Figure 1 shown in the figure, the embodiment of the present invention provides an integrated system for urban flood control early warning and emergency command, which includes a multi-source data acquisition module, a dynamic early warning analysis module, a three-dimensional visualization command module, an intelligent scheduling decision-making module, and a cross-platform communication module.

[0050] In this embodiment, the multi-source data acquisition module is used to obtain meteorological data, hydrological data, urban drainage network data, and video surveillance data in real time, and report the data using a mobile terminal;

[0051] Furthermore, the multi-source data acquisition module in the integrated system for urban flood control early warning and emergency command undertakes the important task of collecting and integrating various key data, provides indispensable basic support for subsequent early warning analysis and command decision-making, and is an important link to ensure the efficient implementation of urban flood control work; specifically:

[0052] I. Data acquisition equipment and technologies:

[0053] Meteorological data acquisition equipment: Professional meteorological monitoring stations are used, equipped with equipment such as rain gauges, anemometers and wind vanes, and temperature and pressure monitors to accurately collect meteorological data; the rain gauge can monitor precipitation and precipitation intensity in real time, providing a key basis for predicting the risk of waterlogging; for example, before a heavy rain comes, it can timely capture precipitation changes and give an early warning of possible water accumulation; at the same time, satellite meteorological cloud map receiving equipment is used to obtain macroscopic meteorological information, combined with ground monitoring data, to improve the accuracy of meteorological prediction; the meteorological data acquisition frequency is high, usually updated in minutes or even seconds to ensure that meteorological changes can be reflected in a timely manner;

[0054] Hydrological data acquisition facilities: Hydrological monitoring equipment such as water level gauges and flow meters are set up in waters such as rivers, lakes, and reservoirs; the water level gauge can monitor water level changes in real time, and the flow meter can accurately measure water body flow; for example, in areas where rivers meet or are prone to floods, the water situation can be timely grasped through these devices; in addition, advanced technologies such as acoustic Doppler current profilers (ADCP) are used to obtain more comprehensive hydrological information, including flow velocity distribution, etc., providing detailed data for flood evolution simulation; these devices transmit data to the system in real time through wired or wireless transmission methods;

[0055] Urban Drainage Network Data Acquisition Technology: Use equipment such as pipeline robots and liquid level sensors to collect drainage network data; the pipeline robot can penetrate into the interior of the drainage pipeline, take pictures of the internal conditions of the pipeline, and detect problems such as damage and siltation; the liquid level sensor is installed at key nodes of the pipeline to monitor the water level height of the pipeline in real time; at the same time, combined with Geographic Information System (GIS) technology, digital management of information such as the layout, pipe diameter, and slope of the drainage network is carried out to build an accurate drainage network model, providing a basis for analyzing drainage capacity and waterlogging risk;

[0056] Video Surveillance Data Acquisition Equipment: Install high-definition cameras in key areas such as urban roads, bridges, and low-lying areas to form a video surveillance network; these cameras have night vision functions and automatic zoom capabilities, and can clearly capture pictures of the water accumulation depth, water flow speed, and road traffic conditions, etc.; through video analysis technology, it can automatically identify abnormal situations, such as serious water accumulation and vehicles being trapped, etc., and issue alarms in a timely manner; the video data is transmitted to the command center in real time through the network, providing intuitive on-site information for the staff;

[0057] II. Data Acquisition and Transmission Process:

[0058] Data Acquisition: Various data acquisition devices continuously collect corresponding data according to preset time intervals or trigger conditions; meteorological monitoring stations, hydrological monitoring equipment, drainage network sensors, and video surveillance cameras work simultaneously to ensure all-round and multi-dimensional data acquisition; during the data acquisition process, preliminary quality inspection is carried out on the collected data, and obviously incorrect or abnormal data is excluded to ensure the accuracy of the data;

[0059] Data Transmission: The collected data is transmitted to the data processing center through various communication methods; for data that is close in distance and has high real-time requirements, such as urban drainage network data and some video surveillance data, wired networks (such as optical fibers) are used for transmission to ensure the stability and high speed of data transmission; for meteorological data, hydrological data, and video surveillance data with a wide distribution, wireless communication technologies (such as 4G / 5G, Wi-Fi, etc.) are used for transmission; in addition, to prevent data loss caused by communication interruption, some devices also have local storage functions and automatically retransmit data after communication is restored;

[0060] Data Integration: The data transmitted to the data processing center is classified and integrated according to different types and sources; meteorological data, hydrological data, drainage network data, and video surveillance data are associated and matched to form a unified data set; through data cleaning and preprocessing, duplicate data is further removed and missing data is filled to provide a high-quality data basis for subsequent analysis and applications;

[0061] III. Importance in Flood Control Work:

[0062] Precise early warning support: The rich data collected by the multi-source data acquisition module provides strong support for the dynamic early warning analysis module. By coupling the meteorological prediction model with the urban waterlogging model and combining real-time meteorological, hydrological, and drainage pipe network data, it is possible to generate a more precise dynamic early warning map, predicting in advance the time, location, and severity of waterlogging, and gaining more preparation time for urban flood control work.

[0063] Scientific decision-making basis: It provides accurate data basis for the intelligent dispatching decision module. Based on the collected real-time data, the intelligent dispatching decision module can plan the emergency resource path through the improved ant colony optimization algorithm and make resource matching decisions using the dynamic weight multi-objective optimization model, realizing the reasonable allocation of emergency resources and improving the efficiency and effect of flood control emergency response.

[0064] Real-time monitoring and command: The three-dimensional visualization command module, with the data provided by the multi-source data acquisition module, integrates the BIM urban model and GIS geographical information data to realize the three-dimensional spatial visualization display of the flood situation. Staff can intuitively understand the urban flood situation, conduct multi-department collaborative command, issue accurate command instructions in a timely manner, and improve the scientificity and timeliness of flood control command.

[0065] The multi-source data acquisition module plays a key role in urban flood control early warning and emergency command work through diversified data acquisition devices and technologies and rigorous data acquisition and transmission processes, providing a solid data guarantee for ensuring the safe flood season of the city.

[0066] In this embodiment, the dynamic early warning analysis module generates a dynamic early warning map by coupling the meteorological prediction model with the urban waterlogging model, and the coupled meteorological prediction model uses machine learning algorithms to correct prediction parameters in real time.

[0067] Furthermore, the dynamic early warning analysis module occupies a core early warning analysis position in the urban flood control early warning and emergency command integrated system. It generates accurate dynamic early warning maps by integrating multiple models and algorithms and deeply analyzing multi-source data, providing key decision-making basis for urban flood control work and effectively improving the city's ability to respond to flood disasters. Specifically:

[0068] I. Core algorithm and model integration:

[0069] Improved spatio-temporal fusion prediction model: The improved spatio-temporal fusion prediction model adopted by this module innovatively integrates multiple data prediction methods. The core algorithm formula is

[0070] ;

[0071] Among them, is the predicted value at time , , is the dynamic weight coefficient, which is calibrated in real time through KL divergence; is the meteorological time series prediction value based on the gated recurrent unit, is the meteorological data; is the rate of change of the output of the drainage pipe network hydraulic model with respect to time, is the water level height of the pipeline, is the surface runoff flow; is the terrain influence integral value, is the terrain influence factor function, and are the integration intervals;

[0072] Performing time series prediction on meteorological data based on the gated recurrent unit can effectively capture the time series characteristics in meteorological data and accurately predict the future meteorological change trend. For example, predicting key meteorological elements such as rainfall intensity and duration provides meteorological basic data for subsequent waterlogging risk assessment;

[0073] Coupling meteorology and urban waterlogging model: By coupling the meteorological prediction model and the urban waterlogging model, an accurate mapping from meteorological data to urban waterlogging risk is achieved; The meteorological prediction model provides meteorological information such as rainfall, and the urban waterlogging model (such as the SWMM model) based on these meteorological data, combined with urban drainage pipe network data, terrain data, etc., simulates processes such as the formation of surface runoff and the water flow movement in the drainage pipe network; is the rate of change of the output of the drainage pipe network hydraulic model with respect to time, reflecting the operating state of the drainage pipe network at different times, helping to analyze whether the drainage capacity of the pipe network can cope with the runoff generated by the current rainfall, thereby predicting the possible areas and times of waterlogging;

[0074] Dynamic weight coefficient calibration: In the formula, , , are the dynamic weight coefficients, which are calibrated in real time through KL divergence; This calibration mechanism can dynamically adjust the weights of each part of the data in the prediction result according to the importance changes of meteorological data, drainage pipe network data, and terrain data at different times; In the initial stage of the rainstorm, the meteorological data has a greater impact on predicting the waterlogging risk. At this time, the weight may be relatively high; As the rainfall continues, the importance of the drainage pipe network operating state data increases, the weight will be adjusted accordingly to ensure that the prediction model can adapt to the changes in the actual situation in real time and improve the prediction accuracy;

[0075] II. Algorithm optimization and improvement:

[0076] Improved LSTM Network with Dual-channel Attention Mechanism: In the spatio-temporal fusion prediction model, an LSTM network is improved using a dual-channel attention mechanism; the calculation method of the attention weight is , where is the attention weight at time ; ; are trainable parameter matrices; is the hidden layer state at time ; is the input feature at the current time;

[0077] SWMM Model Introduces Pipe Siltation Correction Factor: The SWMM model is optimized by introducing a pipe siltation correction factor; the actual flow calculation formula is , where is the actual flow; is the theoretical flow; is the pipe aging coefficient; is the maintenance impact factor; is the time since the last maintenance; Considering that the drainage capacity of the pipe will decrease due to siltation during long-term use, this correction factor can correct the theoretical flow according to the aging degree and maintenance situation of the pipe, making the model more in line with the actual drainage situation and improving the reliability of the waterlogging prediction;

[0078] III. Early Warning Atlas Generation and Application:

[0079] Dynamic Early Warning Atlas Generation: Based on the above algorithms and models, the dynamic early warning analysis module generates a dynamic early warning atlas; the atlas presents the waterlogging risk levels of different regions in the city at different future times in an intuitive visual way. Combining with GIS geographic information, it can accurately locate the risk areas; for example, different colors are used to represent different risk levels, red represents high-risk areas, yellow represents medium-risk areas, and green represents low-risk areas, enabling decision-makers to quickly understand the overall distribution and change trend of the urban waterlogging risk;

[0080] Provide support for emergency command: The generated early warning map provides an important basis for subsequent emergency command work; the intelligent dispatching decision-making module plans the transportation route of emergency resources according to the early warning map, combines the emergency resource reserve situation, and uses the improved ant colony optimization algorithm, and makes resource matching decisions through the dynamic weight multi-objective optimization model to achieve the efficient allocation of emergency resources; the 3D visualization command module, based on the early warning map, displays the flood situation and provides intuitive information support for multi-department collaborative command, enabling each department to carry out flood control and rescue work targeted according to the risk areas and levels.

[0081] The dynamic early warning analysis module realizes the accurate prediction and dynamic early warning of urban waterlogging risks through complex and sophisticated algorithm models, plays an indispensable key role in the urban flood control early warning and emergency command system, and provides strong support for ensuring urban safety and the safety of people's lives and property.

[0082] In this embodiment, the 3D visualization command module integrates the BIM city model and GIS geographic information data to realize the 3D spatial visualization display of the flood situation and the multi-department collaborative operation interface.

[0083] Furthermore, the 3D visualization command module is a key component of the urban flood control early warning and emergency command integrated system. It uses advanced technical means to present complex flood situation information in an intuitive and vivid 3D form, providing an efficient collaborative operation platform for flood control command, and greatly improving the scientificity and accuracy of command decision-making; specifically:

[0084] I. Data integration and fusion:

[0085] Integration of BIM and GIS data: This module integrates the BIM city model and GIS geographic information data; the BIM model details the 3D structural information of urban buildings, infrastructure, etc., accurate to the location and attributes of each floor of the building and each key facility; while the GIS geographic information data covers macro-geographic information such as terrain, water system distribution, and transportation network; integrating the two can construct a comprehensive and accurate urban 3D geographic information scene; for example, when viewing the flood situation in a certain area of the city, one can not only see the terrain undulation and road directions in the area, but also clearly understand the building structure and layout, providing a comprehensive data basis for analyzing the flood impact range and formulating countermeasures.

[0086] Multi-source data access and association: In addition to BIM and GIS data, the 3D visualization command module also accesses multi-source information such as meteorological data, hydrological data, urban drainage pipe network data, and video surveillance data, and realizes real-time association with the 3D scene; for example, the real-time water level data is associated with rivers and lakes in the 3D scene, and when the water level changes, the water body height in the 3D scene will also change accordingly; the rainfall area and intensity in the meteorological data are intuitively displayed on the 3D map, and combined with terrain and drainage pipe network information, the waterlogging area and the risk of waterlogging can be predicted more accurately; the video surveillance data can be played in real time at the corresponding points in the 3D scene, providing on-site real scene pictures for the commanders and enhancing the intuitive feeling of the flood situation;

[0087] II. 3D visualization display function:

[0088] Real-time display of flood situation: Using the integrated data, the 3D visualization command module realizes the 3D spatial visualization display of the flood situation; presenting information such as the flood inundation range, water level change trend, and road waterlogging depth in an intuitive way; for example, by rendering with different colors and transparencies, the waterlogging depth of different regions is displayed, with red representing areas with deeper waterlogging and higher danger levels, and blue representing areas with shallower waterlogging; using dynamic water flow effects to simulate the flow direction and speed of floods, enabling commanders to quickly grasp the development trend of the flood situation; at the same time, combined with the time axis function, historical flood situation data can be traced back to analyze the evolution process of the flood situation and provide a reference for predicting the future development of the flood situation;

[0089] Construction of 3D model for dynamic inundation area: Based on the improved Marching Cubes algorithm, a 3D model of the dynamic inundation area is generated, and its isosurface extraction formula is , where is the isosurface function value at the spatial point ; is the octree node weight; is the improved Sigmoid function, which increases the response to terrain mutations; is the water level height of the grid point; is the water level threshold; this algorithm calculates the isosurface function value at the spatial point by comparing the water level height of different grid points with the water level threshold through the octree node weight and the improved Sigmoid function , and then generates a 3D model of the dynamically changing inundation area; the improved Sigmoid function increases the response to terrain mutations, enabling the model to more accurately reflect the flood inundation situation under complex terrain conditions and providing a more accurate basis for command decisions;

[0090] III. Path Planning and Cooperative Operation:

[0091] Path Planning Visualization: Particle Swarm Optimization algorithm is adopted for path planning visualization to generate navigation beams. The formula is

[0092] ,

[0093] where is the velocity of the th generation of the th particle in the th dimension; is the inertia weight; is the velocity of the th generation of the th particle in the th dimension; , , are the learning factors; , , are random numbers within the interval; is the individual optimal position of the th particle in the th dimension; is the position of the th generation of the th particle in the th dimension; is the coordinate of the global optimal position in the th dimension; is the danger gradient term, which can avoid high-risk areas in real time. When planning the path, this algorithm considers the inertia weight , learning factors , , , as well as the individual optimal position of the particle, global optimal position and danger gradient term . Through the collaborative action of these parameters, it can generate the optimal path to avoid high-risk areas (such as severely waterlogged areas and damaged road sections) and display it in the three-dimensional scene in the form of navigation beams. For example, when allocating emergency resources, the best driving route for rescue vehicles can be planned according to this path planning function to improve the resource transportation efficiency;

[0094] Multi - department collaborative operation interface: The three - dimensional visualization command module provides a collaborative operation interface for multiple departments, integrating the business functions of multiple departments such as emergency management, fire protection, transportation, and water conservancy. Personnel from each department can share information and conduct collaborative operations in the same three - dimensional scene, such as marking key protection areas, issuing task instructions, and reporting work progress. For example, the emergency management department can mark the areas in need of emergency rescue in the three - dimensional scene and issue rescue tasks to the fire department; the transportation department can adjust the traffic control plan in real - time and display the controlled areas and diversion routes through the three - dimensional map; the water conservancy department can remotely control the operation of water conservancy facilities in the three - dimensional scene according to the water level changes, realizing the efficient collaborative operation of multiple departments and improving the overall efficiency of flood control emergency command.

[0095] Through data integration and fusion, visualization display, and collaborative operation functions, the three - dimensional visualization command module provides an all - around and visual decision - making support platform for urban flood control command, effectively improving the ability and level of urban flood control emergency command, and playing an important role in ensuring the safety of the city during the flood season.

[0096] In this embodiment, the intelligent scheduling and decision - making module includes an emergency resource path planning unit based on an improved ant colony optimization algorithm and a resource matching decision - making unit based on dynamic weight multi - objective optimization.

[0097] Furthermore, the intelligent scheduling and decision - making module is a core component of the urban flood control early warning and emergency command integrated system, playing a key role in flood control emergency response. It integrates advanced algorithms, comprehensively considers various factors, provides a scientific decision - making basis for emergency resource allocation and path planning, and ensures a rapid and efficient response in the face of flood conditions, minimizing losses to the greatest extent. Specifically:

[0098] I. Emergency resource path planning unit:

[0099] Principle of the improved ant colony optimization algorithm: This unit uses the improved ant colony optimization algorithm to plan the transportation path of emergency resources, and its core formula is

[0100] ,

[0101] where, is the pheromone concentration on path at time . The pheromone concentration will volatilize over time, and the pheromone evaporation coefficient , ; is the pheromone concentration on path at time ; is the pheromone increment left by the nd ant on path ; is the total number of ants; is the environmental risk adjustment factor; is the combined risk assessment value of road waterlogging depth and bridge load-bearing capacity at time Each ant will leave a pheromone increment on the path , and the total pheromone increment left by all ants will affect the pheromone concentration of the path; at the same time, the environmental risk adjustment factor and the combined risk assessment value of road waterlogging depth and bridge load-bearing capacity at time are introduced, which enables the algorithm to fully consider environmental risk factors when calculating the pheromone concentration of the path;

[0102] Algorithm advantages and application scenarios: Compared with traditional algorithms, the improved ant colony optimization algorithm can better adapt to the complex and changeable environment of urban flood control; for example, in the case of heavy rain causing road waterlogging and limited bridge load-bearing capacity, the algorithm will, according to the real-time assessment value, reduce the pheromone concentration of high-risk paths, guiding ants (representing the selection of emergency resource transportation paths) to avoid roads with serious waterlogging or insufficient bridge load-bearing capacity and choose safer and more efficient paths; this algorithm can ensure that emergency resources are quickly and safely transported to the disaster area in a harsh flood situation, improving the rescue efficiency; in practical applications, whether it is a fire truck going to an area with concurrent fires and floods or medical supplies being transported to the centralized resettlement point for affected people, the best path can be planned with the help of this algorithm;

[0103] II. Resource matching decision-making unit:

[0104] Construction of dynamic weight multi-objective optimization model: The resource matching decision-making unit uses a dynamic weight multi-objective optimization model to make resource matching decisions, and the objective function is

[0105] , where , , are dynamically adjusted weight parameters, which are linked to the warning level; is the emergency response time; is the cost of emergency resources; is the efficiency of emergency resource allocation. When the warning level is high and the disaster situation is urgent, the weight of the emergency response time will increase, emphasizing the need to put resources into use as soon as possible; while when the warning level is relatively low and resource allocation pays more attention to cost-effectiveness, the weight of the cost of emergency resources will increase accordingly; the efficiency of emergency resource allocation is also considered in the model, As its weight, it ensures the efficient progress of the resource allocation process;

[0106] Constraint setting and significance: The model sets a series of constraints, including , to ensure that the quantity of emergency resources can meet the actual needs, while considering the demand volatility , which is updated in real time according to public opinion data to cope with possible changes in resource demand; , comprehensively considering the length of the emergency resource transportation path 、 the transportation speed at a certain moment and the influence coefficient of path waterlogging 、the depth of waterlogging on the road section , to ensure that resources can reach the disaster area within the specified time; , to ensure the rationality and effectiveness of the weight parameters; These constraints provide comprehensive restrictions and guidance for resource matching decisions from aspects such as resource quantity, arrival time, and weight balance, making the decision results more in line with the actual flood control needs (where is the quantity of the th kind of emergency resource; is the number of types of emergency resources; is the total demand for emergency resources; is the demand volatility, which is updated in real time according to public opinion data; is the arrival time of emergency resources; is the length of the emergency resource transportation path; is the transportation speed of emergency resources at a certain moment; is the influence coefficient of path waterlogging; is the depth of waterlogging on the road section);

[0107] III. Module collaboration and overall efficiency:

[0108] Data interaction with other modules: The intelligent scheduling decision-making module closely cooperates with other modules in the system; The multi-source data acquisition module provides it with data such as meteorology, hydrology, drainage pipe networks, and on-site real-time videos, which are used to evaluate the severity of flood conditions, road traffic conditions, etc., and provide basic information for path planning and resource matching; The warning maps and prediction data generated by the dynamic warning analysis module help the intelligent scheduling decision-making module determine the emergency response priorities and resource demand scales in different regions; The 3D visualization command module then displays the results of the intelligent scheduling decision-making module in an intuitive 3D form, facilitating decision-making and command coordination by command personnel, and realizing information sharing and collaborative work among modules;

[0109] Improving the efficiency of flood control emergency command: Through the collaborative work of the emergency resource path planning unit and the resource matching decision-making unit, the intelligent dispatching decision module can quickly plan the optimal emergency resource transportation path according to the real-time flood situation, reasonably allocate various emergency resources, and achieve the efficient utilization of resources; in the actual flood control work, this module can greatly improve the emergency response speed, ensure that the rescue force arrives at the affected area in time, improve the scientificity and accuracy of flood control emergency command, and play an important role in protecting the safety of the city and people's lives and property.

[0110] In this embodiment, the cross-platform communication module supports dual-link guarantee of 5G-MEC edge computing and satellite communication, realizing the multimedia transmission of command instructions and the augmented reality (AR) interaction of on-site terminals.

[0111] Furthermore, as a key hub of the urban flood control early warning and emergency command integration system, the cross-platform communication module plays an indispensable role in ensuring the efficient and stable transmission of information; it integrates advanced communication technologies, constructs a multi-link guarantee system, realizes the multimedia transmission of command instructions and the augmented reality (AR) interaction of on-site terminals, and effectively promotes the coordination and efficient development of flood control emergency command work; specifically:

[0112] I. Communication link guarantee:

[0113] 5G-MEC edge computing link: The cross-platform communication module relies on the high speed and low latency characteristics of the 5G network to ensure the rapid transmission of data; the large bandwidth capacity of the 5G network can meet the rapid upload and download requirements of a large amount of real-time data (such as high-definition video surveillance data, high-resolution GIS map data, etc.), enabling the command center to obtain detailed information about the on-site flood situation in real time, such as the depth of road waterlogging and the drainage status of drainage outlets, providing an accurate basis for decision-making; at the same time, combined with MEC (Mobile Edge Computing) technology, the computing and storage capabilities are sunk to the network edge, reducing data transmission latency and improving the system response speed; for example, the data collected by on-site terminal devices can be preliminarily processed and analyzed on the local edge server, and then key information is generated and transmitted to the command center, reducing the burden on the core network and further improving communication efficiency; this enables command instructions to be quickly conveyed to on-site terminals, realizing real-time command of flood control work.

[0114] Satellite communication link: To cope with complex flood control environments, such as when 5G network coverage is insufficient or the ground communication network is damaged due to disasters, the satellite communication link serves as a backup guarantee; satellite communication has the advantages of wide coverage and being unrestricted by geographical conditions, enabling the command center to maintain contact with remote disaster-stricken areas, mountainous areas, etc. where it is difficult to communicate through the ground network; it can transmit various information such as voice, data, and images, ensuring the transmission of key command instructions and important disaster situation information; when some base stations are unable to be used due to flood disasters in certain areas, on-site rescue personnel can report the disaster situation and request support to the command center through satellite communication equipment, and the command center can also issue rescue tasks and resource allocation instructions through satellite communication, maintaining the continuity of flood control command work;

[0115] II. Command instruction transmission function:

[0116] Multimedia transmission: The cross-platform communication module supports the multimedia transmission of command instructions, capable of integrating and transmitting information in various forms such as text, voice, images, and videos; commanders can quickly convey decision-making and deployment through voice instructions, improving communication efficiency and avoiding delays caused by text input; at the same time, combined with image and video materials, such as maps marked with key rescue areas, on-site disaster situation videos, etc., enabling on-site personnel to more intuitively understand the task requirements and accurately execute rescue operations; when commanding the rescue of trapped people, the command center can send map images containing the positions of trapped people and on-site rescue guidance videos to help rescue personnel quickly locate and implement rescues, enhancing the accuracy and efficiency of rescue operations;

[0117] Accurate transmission and reception of instructions: This module adopts advanced communication protocols and signal processing technologies to ensure the accuracy and integrity of command instructions during transmission; through mechanisms such as data verification and error correction coding, the transmitted data is monitored and repaired in real time to avoid the loss or incorrect transmission of instructions caused by signal interference or transmission errors; at the same time, on-site terminal devices have intelligent reception and decoding functions, capable of quickly and accurately parsing the received multimedia instructions and presenting them to on-site staff in an intuitive manner, such as displaying clear text prompts and playing voice instructions on the terminal device, ensuring that command instructions can be correctly understood and executed by on-site personnel;

[0118] III. On-site terminal interaction function:

[0119] Augmented Reality (AR) Interaction: The cross-platform communication module enables the AR interaction function of on-site terminals, combining virtual information with the real-world scenario through AR technology. On-site workers wear AR devices (such as smart glasses) and can view virtual annotation information in the real environment, such as danger area warnings, rescue route planning, equipment operation guides, etc. When repairing drainage equipment, the AR device can display the schematic diagram of the equipment structure, fault troubleshooting steps, and maintenance guidance information in the real scenario, helping maintenance personnel quickly locate the fault point and perform repairs, improving the repair efficiency. At the same time, workers can also have real-time interaction with the command center through the AR device, such as transmitting the on-site perspective to the command center in real-time, enabling commanders to more intuitively understand the on-site situation and make more accurate decisions.

[0120] Improving On-site Work Efficiency and Safety: The AR interaction function greatly improves the efficiency and safety of on-site work. It reduces the time for workers to consult paper materials or operate complex terminal devices, enabling workers to focus more on actual work tasks. Through real-time risk prompts and operation guides, it effectively avoids safety accidents caused by human negligence or unfamiliarity with complex operations. When conducting rescues in dangerous waterlogged areas, the AR device can display information such as water depth and potential hazards in real-time, providing safety guarantees for rescue personnel and improving the efficiency and success rate of rescue operations.

[0121] The cross-platform communication module breaks down the barriers of information transmission and interaction through dual-link guarantee, multimedia command instruction transmission, and the AR interaction function of on-site terminals, realizing efficient communication and collaboration between the flood control command center and the on-site, providing strong communication support for urban flood control early warning and emergency command work, and effectively enhancing the flood control emergency response ability.

[0122] In this embodiment, the evaluation and feedback module calculates the comprehensive score of the disposal plan using the improved TOPSIS algorithm.

[0123] Furthermore, the evaluation and feedback module plays an important role in the urban flood control early warning and emergency command integration system. It quantitatively evaluates the flood control disposal plan through scientific algorithms and feeds back the evaluation results to the system, promoting the continuous optimization of the system and enhancing the overall efficiency of flood control work. Specifically:

[0124] Scheme Evaluation Algorithm: The evaluation and feedback module calculates the comprehensive score of the disposal plan using the improved TOPSIS algorithm. The formula is , where is the comprehensive score of the th disposal plan. The comprehensive score is the core indicator for measuring the quality of the plan. is the The weights of the indicators are dynamically adjusted by introducing a time decay function. In the initial stage of flood control, the weights of some indicators related to rapid response are relatively high. As the flood control work progresses, the weights of indicators such as resource utilization efficiency may change to reflect the key points of flood control work at different stages. For the th disposal plan to the positive ideal solution on the th indicator; For the th disposal plan to the negative ideal solution on the th indicator. These two distances are used to measure the gap between the plan and the best and worst cases on each indicator. is the risk penalty factor. For plans with high risks, the penalty points are increased to ensure that the evaluation results can truly reflect the risk level of the plans. is the number of indicators, covering multiple aspects such as emergency response time, resource allocation efficiency, and disaster loss control, to comprehensively evaluate the comprehensive effect of the plan.

[0125] Dynamic adjustment and feedback optimization of indicators: The evaluation indicators are dynamically adjusted according to the actual flood control situation and system operation feedback. By analyzing historical flood control data and comparing the implementation effects of different plans, new indicators that have a significant impact on flood control results are identified, or the weights of existing indicators are reallocated. If it is found in a certain flood control that the rescue is delayed in a certain area due to poor information transmission, the weight of information transmission efficiency in the evaluation indicators will be increased subsequently. The evaluation feedback module feeds back the plan evaluation results to all links of the system, provides reference for the intelligent dispatching decision-making module to help it optimize resource allocation and path planning strategies, assists the dynamic early warning analysis module to improve the early warning model and enhance the early warning accuracy, and can also enable the three-dimensional visualization command module to display more targeted information, improve the scientific nature of command decisions, and promote the continuous optimization of the entire flood control system.

[0126] Improving the overall efficiency of flood control work: Through the evaluation and feedback of disposal plans, this module effectively improves the overall efficiency of flood control work. In the actual flood control process, evaluating different emergency rescue plans and selecting the plan with a high comprehensive score for implementation can significantly improve the rescue efficiency and reduce casualties and property losses. After each flood control work is completed, the evaluation feedback module conducts a review and evaluation of the entire process, summarizes experience and lessons, provides valuable reference for subsequent flood control work, continuously improves the urban flood control early warning and emergency command system, and better responds to future flood disasters.

[0127] A method for urban flood control early warning and emergency command realizes real-time monitoring, accurate early warning, efficient command, and scientific evaluation of urban flood conditions through the collaborative work of multiple modules to ensure urban safety. The following are its detailed operation steps:

[0128] Step S1, Data Collection and Transmission: The multi-source data collection module uses professional equipment and mobile terminals to collect meteorological, hydrological, drainage network, and video surveillance data in real time; the meteorological monitoring station collects data such as precipitation and wind speed, the hydrological monitoring equipment monitors water level and flow rate, etc., the drainage network sensors obtain pipeline water level and drainage capacity information, and the video surveillance cameras capture on-site pictures; these data are stably transmitted to the data processing center through wired or wireless communication methods to provide comprehensive and accurate data support for subsequent analysis;

[0129] Step S2, Dynamic Early Warning Analysis: The dynamic early warning analysis module receives the collected data and processes it using an improved spatio-temporal fusion prediction model; by coupling the meteorological prediction model and the urban waterlogging model, combined with terrain data, calculate the waterlogging risk values at different times and regions, and generate a dynamic early warning map; for example, use the LSTM network to predict meteorological data, combine the SWMM model to simulate the water flow in the drainage network, and correct the prediction results based on the terrain influence factors; if the waterlogging risk in a certain area is predicted to be high, the system will promptly issue an early warning message to notify relevant departments and personnel;

[0130] Step S3, Visualized Command and Decision-making: The three-dimensional visualization command module integrates the BIM city model and GIS geographic information data to display the flood situation; the command personnel view information such as flooded areas and water level changes in the three-dimensional scene for multi-department collaborative command; the intelligent scheduling decision-making module uses the improved ant colony optimization algorithm to plan the emergency resource transportation path according to the early warning information and resource reserve situation, and makes resource matching decisions through the dynamic weight multi-objective optimization model; such as determining the deployment plan of fire trucks, rescue supplies, etc., and issuing command instructions;

[0131] Step S4, Cross-platform Communication and Execution: The cross-platform communication module transmits the command instructions to the on-site terminal in multimedia form through the 5G-MEC edge computing and satellite communication dual links; the on-site staff receive the instructions with the help of the augmented reality AR interaction function of the terminal, view the task information and on-site markings, and perform emergency rescue tasks such as drainage and rescuing trapped people, and feedback the on-site situation to the command center in real time;

[0132] Step S5, Evaluation Feedback and Optimization: The evaluation feedback module uses the improved TOPSIS algorithm to comprehensively evaluate the disposal plan and calculate the score; according to the evaluation results, analyze the advantages and disadvantages of the plan, and provide the feedback information to other modules; for example, the intelligent scheduling decision-making module optimizes the resource allocation strategy according to the feedback, and the dynamic early warning analysis module improves the prediction model to continuously enhance the flood control ability of the system.

[0133] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An integrated system for urban flood prevention warning and emergency command, characterized in that: include: Multi-source data acquisition module: used to obtain meteorological data, hydrological data, urban drainage network data and video surveillance data in real time, and report the data using mobile terminals; Dynamic warning analysis module: Generates dynamic warning maps by coupling the meteorological forecast model with the urban waterlogging model. The coupled meteorological forecast model uses machine learning algorithms to correct forecast parameters in real time. 3D visualization command module: Integrates BIM city model and GIS geographic information data to realize 3D spatial visualization of flood situation and multi-department collaborative operation interface; Intelligent scheduling decision module: including emergency resource path planning unit based on improved ant colony optimization algorithm and resource matching decision unit based on dynamic weight multi-objective optimization; Cross-platform communication module: supports 5G-MEC edge computing and satellite communication dual-link guarantee, realizes multimedia transmission of command instructions and augmented reality AR interaction of on-site terminals; The dynamic warning analysis module adopts an improved spatiotemporal fusion prediction model, and its core algorithm is defined as: ,in, for The predicted value at the moment; , , is the dynamic weight coefficient, which is calibrated in real time through KL divergence; is the meteorological time series prediction value based on the gated recurrent unit, For meteorological data; Output the rate of change of the hydraulic model of the drainage network with respect to time, is the water level in the pipe, is the surface runoff flow; is the integral value of terrain influence, is the terrain influence factor function, and is the integration interval.

2. The integrated system of urban flood prevention warning and emergency command according to claim 1 is characterized in that: The improved ant colony optimization algorithm in the intelligent scheduling decision module satisfies: ,in, for Time Path pheromone concentration on; for Time Path pheromone concentration on; is the pheromone volatility coefficient; For the Ants on the path The increase of pheromones left on the is the total number of ants; is the environmental risk adjustment factor; for Joint risk assessment value of road water depth and bridge load-bearing capacity at all times.

3. The integrated system of urban flood prevention early warning and emergency command according to claim 2 is characterized in that: The resource matching decision unit adopts a dynamic weight multi-objective optimization model: ,in, , , It is a dynamically adjusted weight parameter linked to the warning level; For emergency response time; Cost of emergency resources; To improve the efficiency of emergency resource deployment; Constraints: ,in, For the The number of emergency resources; is the number of types of emergency resources; is the total amount of emergency resource demand; Demand volatility is updated in real time based on public opinion data; The arrival time of emergency resources; The length of the transportation route for emergency resources; for The speed of emergency resource transportation at all times; is the path waterlogging influence coefficient; The depth of water accumulation on the road section.

4. The integrated system of urban flood prevention warning and emergency command according to claim 3 is characterized in that: In the spatiotemporal fusion prediction model: The dual-channel attention mechanism is used to improve the LSTM network, and its attention weight is calculated as: ,in, for The attention weight of the moment; , , is the trainable parameter matrix; for Hidden layer state at all times; Enter features for the current moment; is the bias term; The SWMM model incorporates a pipe siltation correction factor: ,in, is the actual flow rate; is the theoretical flow rate; is the pipeline aging coefficient; To maintain the impact factor; The time since the last maintenance.

5. The integrated system of urban flood prevention early warning and emergency command according to claim 4 is characterized in that: The three-dimensional visualization command module includes: The three-dimensional model of the dynamic flooded area is generated based on the improved Marching Cubes algorithm, and the isosurface extraction formula is: ,in, For space point The isosurface function value at ; is the octree node weight; It is an improved Sigmoid function with increased terrain mutation response; is the water level height at the grid point; is the water level threshold; Path planning visualization uses particle swarm optimization algorithm to generate navigation beams: ,in, For the Daidi The particle in The speed of the dimension; is the inertia weight; For the Daidi The particle in The speed of the dimension; , , is the learning factor; , , for Random numbers in the interval; For the The particle in The optimal position of an individual in the dimension; For the Daidi The particle in The location of the dimension; The global optimal position is Dimensional coordinates; It is a hazard gradient item, which can avoid high-risk areas in real time.

6. The integrated system of urban flood prevention early warning and emergency command according to claim 5 is characterized in that: Also includes: The evaluation feedback module uses the improved TOPSIS algorithm to calculate the comprehensive score of the disposal plan: ,in, For the The overall score of the treatment options; For the The weight of each indicator is adjusted dynamically by introducing a time decay function; For the The ideal solution is the The distance on the indicator; For the The solution to the negative ideal solution is The distance on the indicator; is the risk penalty factor; is the number of indicators.

7. A method for urban flood prevention early warning and emergency command, characterized in that: The urban flood warning and emergency command method utilizes the urban flood warning and emergency command integrated system described in any one of claims 1-6 to perform urban flood warning and emergency command.

Images