A method and system for dynamic monitoring of urban flooding disasters based on big data

Abstract

This invention discloses a method and system for dynamic monitoring of urban flood disasters based on big data, belonging to the field of disaster management technology. The method includes: constructing a three-dimensional urban model; decoupling and evaluating the structural, functional, and load vulnerability to form a disaster monitoring architecture; supervising and training disaster monitoring components based on this architecture and embedding them into an urban information platform to achieve intelligent learning of disaster probability dependence and vulnerability entropy; combining multi-source sensor data for joint evaluation to determine the disaster monitoring status in real time; and implementing intelligent management and targeted monitoring guidance for urban flood disasters based on the monitoring results. This invention solves the technical problems of existing urban flood monitoring methods lacking comprehensive evaluation and real-time dynamic monitoring of the complex urban environment, leading to delayed disaster identification and insufficient monitoring accuracy. It achieves the technical effect of a big data-driven three-dimensional disaster monitoring architecture, enabling multi-dimensional dynamic evaluation of urban flood disasters and improving disaster identification accuracy and response efficiency.

Description

Technical Field

[0001] This invention relates to the field of disaster management technology, specifically to a method and system for dynamic monitoring of urban flood disasters based on big data. Background Technology

[0002] With the acceleration of urbanization and the frequent occurrence of extreme weather events, urban flooding has become a significant risk factor affecting urban safety and the lives and property of residents. Existing urban flood monitoring methods largely rely on single hydrological or meteorological data, lacking comprehensive assessment of the complex urban environment and the ability to monitor real-time dynamics. When faced with complex urban terrain, dense infrastructure, and variable weather conditions, these methods struggle to promptly detect the dynamic evolution of disasters, thus hindering precise prevention and real-time management. Summary of the Invention

[0003] This application provides a method and system for dynamic monitoring of urban flood disasters based on big data, which is used to solve the technical problems of existing urban flood monitoring methods lacking comprehensive assessment and real-time dynamic monitoring of complex urban environments, resulting in delayed disaster identification and insufficient monitoring accuracy.

[0004] The first aspect of this application provides a method for dynamic monitoring of urban flood disasters based on big data. The method includes: constructing a three-dimensional urban model; conducting a decoupled assessment of the structural vulnerability, functional vulnerability, and load vulnerability of the urban three-dimensional model based on the flood disaster dimension; constructing a disaster monitoring architecture; supervising and training disaster monitoring components according to the disaster monitoring architecture and embedding them in an urban information platform, wherein the training directions include disaster probability dependencies, flood vulnerability entropy quantification, and coupled assessment; performing joint assessment of the disaster monitoring components under data initialization by conducting multi-source sensing in the city and transmitting the data back to the urban information platform to determine the disaster monitoring status; and providing guidance for urban flood disaster management and targeted monitoring based on the disaster monitoring status.

[0005] The second aspect of this application provides a dynamic monitoring system for urban flood disasters based on big data. The system includes: a multi-dimensional decoupled assessment module for constructing a three-dimensional urban model and performing a decoupled assessment of the structural vulnerability, functional vulnerability, and load vulnerability of the urban three-dimensional model based on the flood disaster dimension, thereby constructing a disaster monitoring architecture; a monitoring component training module for supervising and training disaster monitoring components according to the disaster monitoring architecture and embedding them in an urban information platform, wherein the training directions include disaster probability dependencies, flood vulnerability entropy quantification, and coupling assessment; a joint assessment module for performing a joint assessment of the disaster monitoring components under data initialization by conducting multi-source sensing in the city and transmitting the data back to the urban information platform to determine the disaster monitoring status; and a disaster management module for providing urban flood disaster management and targeted monitoring guidance based on the disaster monitoring status.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application provides a big data-based dynamic monitoring method and system for urban flood disasters, which relates to the field of disaster management technology. By constructing a dynamic monitoring system for flood disasters based on big data and a three-dimensional urban model, it decouples and assesses the vulnerability of structure, function, and load, establishes a disaster supervision architecture, and trains embedded disaster supervision components to achieve real-time monitoring, entropy quantification analysis, and intelligent early warning under multi-source data fusion. This solves the technical problem that existing urban flood monitoring methods lack comprehensive assessment and real-time dynamic monitoring of complex urban environments, resulting in delayed disaster identification and insufficient monitoring accuracy. It realizes a big data-driven three-dimensional disaster supervision architecture, achieves multi-dimensional dynamic assessment of urban flood disasters, and improves the technical effect of disaster identification accuracy and response efficiency. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic diagram of the process for a dynamic monitoring method for urban flooding disasters based on big data, provided in an embodiment of this application;

[0010] Figure 2 A schematic diagram of the structure of a big data-based dynamic monitoring system for urban flood disasters provided in this application embodiment.

[0011] Figure labeling: Multidimensional decoupling assessment module 11, supervision component training module 12, joint assessment module 13, disaster management module 14. Detailed Implementation

[0012] This application provides a method and system for dynamic monitoring of urban flood disasters based on big data, which is used to solve the technical problems of existing urban flood monitoring methods lacking comprehensive assessment and real-time dynamic monitoring of complex urban environments, resulting in delayed disaster identification and insufficient monitoring accuracy.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, this application provides a method for dynamic monitoring of urban flood disasters based on big data, the method including:

[0016] P10: Construct a 3D model of the city, conduct a decoupled assessment of the structural vulnerability, functional vulnerability and load vulnerability based on the 3D model of the city for the dimension of flood disaster, and construct a disaster monitoring architecture.

[0017] Specifically, the first step is to construct a three-dimensional model of the city. This model is based on digital twin city technology and integrates geographic information systems (GIS), remote sensing data, building information models, and real-time sensing data to achieve high-precision modeling of urban spatial morphology, infrastructure structure, and surface hydrological characteristics, providing refined spatial support for subsequent flood and waterlogging vulnerability analysis.

[0018] Building upon this foundation, a multi-dimensional vulnerability decoupling assessment of urban systems is conducted, focusing on flood disasters. Structural vulnerability refers to the resistance of urban physical structures to flood impacts, with assessment parameters including building material properties, drainage system bearing capacity, and terrain elevation distribution. Functional vulnerability refers to the resilience and recovery capabilities of key urban functional units, such as transportation, power, and communication, identified through topological analysis of the infrastructure network to pinpoint functional coupling and potential failure chains. Load vulnerability characterizes the sensitivity of urban population density, traffic flow, and economic activity intensity to flood events, calculated using socioeconomic big data and dynamic travel models. Decoupling analysis methods, such as principal component analysis or multi-objective hierarchical regression models, can be introduced to separate the correlations among the three types of vulnerability factors, eliminating statistical interference between different disaster impact dimensions and obtaining independent and quantifiable risk characteristic indicators. Finally, based on the aforementioned vulnerability indicator system, a disaster monitoring architecture for flood disaster monitoring and assessment is constructed. This architecture defines monitoring objects, monitoring indicators, and data flow relationships, providing a standardized structural foundation and technical interfaces for the subsequent training and embedded deployment of disaster monitoring components.

[0019] Furthermore, regarding the structural vulnerability decoupling assessment based on a three-dimensional urban model for the dimension of flood disaster, step P10 of this application embodiment also includes:

[0020] P11: Based on the city's three-dimensional model, taking drainage catchment areas and pipe networks as the first urban elements, a joint assessment of topographic convergence potential and drainage path dependence is performed on each drainage catchment area to determine the obstruction linear relationship, wherein the drainage catchment area and the obstruction linear relationship correspond one-to-one; P12: Using the first urban elements as the basis, the obstruction linear relationship is marked by basis location matching to generate a first structural vulnerability map.

[0021] Optionally, the process of decoupling the structural vulnerability assessment based on the three-dimensional urban model for the flood disaster dimension can be further refined to more accurately identify and quantify the structural vulnerability of cities in flood disasters.

[0022] Specifically, based on the constructed 3D urban model, the drainage catchment area and pipe network system are identified as the first urban element, which is the basic unit representing the city's drainage function and flood response characteristics. The drainage catchment area refers to the natural or artificial water collection area in the urban topography that collects rainwater or surface runoff, and its spatial boundary can be automatically defined through topographic elevation model (DEM) and flow direction analysis; the pipe network system includes basic drainage facilities such as rainwater pipes, inspection wells, pumping stations, and drainage outlets.

[0023] Based on this, a joint assessment of topographic convergence potential and drainage path dependence is performed on each drainage catchment area. Topographic convergence potential measures the intensity of the trend of topography in water flow accumulation and catchment, and is usually input by topographic factors such as slope aspect, catchment area, and minimum potential path. Drainage path dependence characterizes the degree of dependence of the drainage system on specific pipes or nodes, and can be calculated based on the pipe network topology and flow distribution model to identify critical paths under system overload or blockage conditions. By jointly assessing the above two indicators, a bottleneck linear relationship can be established. The bottleneck linear relationship refers to the critical bottleneck nodes that will lead to large-scale paralysis once blockage or overload occurs, as well as the degree of quantification of drainage and water potential under different water conditions. It is worth noting that there is a one-to-one correspondence between each drainage catchment area and the bottleneck linear relationship, which means that each drainage catchment area has its specific critical bottleneck nodes and corresponding quantitative indicators of water potential.

[0024] Next, using the first urban element as a base, the determined obstruction linear relationships are matched and marked with base location data. This process involves precisely matching the specific location information of the obstruction linear relationships with the corresponding locations in the city's 3D model and then marking them. In this way, a first structural vulnerability map can be generated. This map can visually display the key obstruction nodes and their locations in various drainage catchment areas of the city, providing a quantitative and visual basis for subsequent disaster monitoring and management.

[0025] Furthermore, regarding the functional vulnerability decoupling assessment based on a three-dimensional urban model for the dimension of flood disaster, step P10 of this application embodiment also includes:

[0026] P13: Based on the city's three-dimensional model, using the road network as the second urban element, conduct road network traffic assessments on key transportation hubs to determine traffic flow resilience relationships, wherein key transportation hubs correspond one-to-one with traffic flow resilience relationships; P14: Using the second urban element as a base, perform base location matching and marking on the traffic flow resilience relationships to generate a second functional vulnerability map.

[0027] It should be understood that the process of decoupling and assessing the functional vulnerability of cities based on three-dimensional urban models can be further refined to more accurately identify and quantify the functional vulnerability of cities in flood disasters.

[0028] Specifically, based on the constructed 3D urban model, the road network is defined as the second urban element, characterizing the city's traffic flow and functional maintenance capabilities under flood conditions. On this basis, a road network traffic assessment is conducted on key transportation hubs to determine traffic flow resilience. Traffic flow resilience refers to the recovery and anti-interference capabilities of traffic flow at key transportation hubs under different flood conditions. This assessment process requires comprehensive consideration of multiple factors, including road submersion depth, traffic volume, road network connectivity, and the availability of alternative routes. For example, by combining simulated water depth data under flood scenarios with a dynamic traffic flow simulation model, a time-series analysis of the traffic capacity of each hub node is performed to obtain the variation patterns of traffic flow under different water intensities. By comparing the traffic flow recovery speed, detour capacity, and flow attenuation coefficient under normal traffic conditions and flood disturbance conditions, a traffic flow resilience relationship is established, i.e., a system of indicators characterizing the resilience and recovery of the transportation system under disaster impact. This relationship is used to quantify the ability of key transportation hubs to maintain their functions under flood disaster conditions and to realize a one-to-one mapping between key nodes and traffic flow resilience, providing a quantitative basis for the distribution of functional vulnerability.

[0029] Next, using the second urban element as a base, the determined traffic flow resilience relationships are matched and marked with base locations. Specifically, the aforementioned traffic flow resilience assessment results are spatially mapped in a 3D model, and geographical features such as road elevation, terrain slope, and drainage conditions are combined to identify road sections and nodes highly affected by flooding. Through multi-dimensional spatial data fusion and geographic registration, a second functional vulnerability map is generated. This map presents the functional weaknesses of the urban transportation system under flood disasters in the form of resilience level, traffic risk index, or probability of traffic disruption. This map not only helps identify the functional weaknesses of the city during flood disasters but also provides a scientific basis for developing targeted traffic diversion measures and emergency plans, thereby improving the city's ability to cope with flood disasters.

[0030] Furthermore, for the dimension of flood disaster, a decoupled assessment of load vulnerability based on a three-dimensional urban model is conducted, and a disaster monitoring architecture is constructed. Step P10 of this application embodiment also includes:

[0031] P15: Using the first and second urban elements as the basis, the drainage impact is assessed based on the spatiotemporal concentration of rainfall and the amount of areal rainfall. The drainage impact relationship is determined and the base location is matched and marked to generate the third load vulnerability map. P16: The disaster monitoring architecture is constructed by using the three-dimensional urban model as one layer, the third load vulnerability map as the second layer, and the first structural vulnerability map and the second functional vulnerability map in parallel as the third layer.

[0032] Optionally, the process of conducting load vulnerability decoupling assessment based on a three-dimensional urban model and constructing a disaster monitoring architecture for the dimension of flood disaster can be further refined to more comprehensively identify and quantify the load vulnerability of cities in flood disasters, and integrate the results of various vulnerability assessments to construct a disaster monitoring architecture.

[0033] Specifically, the study first assesses the drainage impact by evaluating the spatiotemporal concentration of rainfall and areal rainfall, using the first urban element (drainage catchment area and pipe network) and the second urban element (road network) as a basis. Specifically, high-resolution spatiotemporal distribution information of rainfall is obtained by combining radar-retrieved rainfall data with measured data from ground meteorological monitoring stations. The concentration of rainfall in the temporal dimension (i.e., the degree of concentration of rainfall per unit time) and the spatial concentration dimension (i.e., the concentration or dispersion characteristics of rainfall distribution) are calculated. Based on this, areal rainfall calculation models, such as the Thiessen polygon method or the grid-weighted average method, are used to obtain areal rainfall indices for different regions to reflect the spatial distribution of rainfall's impact on the overall drainage system load. Next, based on the combined results of rainfall concentration and areal rainfall, a drainage impact relationship is established, describing the probabilistic relationship model of drainage system overload or failure under specific rainfall conditions. This relationship is supported by a hydrodynamic model, quantifying the overload risk of different catchment areas and road nodes by simulating rainfall runoff processes and drainage system responses. Subsequently, the drainage impact relationship was matched and marked with base location data. High-risk drainage nodes and traffic disruption areas were then spatially projected onto the city's 3D model, ultimately generating a third load vulnerability map. This map comprehensively reflects the dynamic impact of rainfall's spatiotemporal characteristics on urban drainage and transportation systems, providing a quantitative basis for disaster response at the load level.

[0034] Next, a multi-dimensional integrated disaster monitoring architecture is constructed by overlaying the city's 3D model as the first layer of the overall spatial framework, the third load vulnerability map as the second layer, and the first structural vulnerability map and the second functional vulnerability map as the third layer. This architecture uses the city's 3D model as the base layer, the third load vulnerability map as the intermediate layer, and the first structural vulnerability map and the second functional vulnerability map as the top layer, forming a multi-layered and comprehensive disaster monitoring system. This architecture design allows the results of various vulnerability assessments to complement and corroborate each other, thereby reflecting the overall vulnerability of the city in flood disasters more comprehensively and accurately. Through this architecture, the monitoring, prediction, and response to urban flood disasters can operate collaboratively in a multi-dimensional data environment, providing systematic model support and spatial semantic foundation for the supervised training and real-time monitoring of subsequent disaster monitoring components.

[0035] P20: Based on the disaster monitoring architecture, supervise the training of disaster monitoring components and embed them in the urban information platform. The training directions include disaster probability dependency, flood vulnerability entropy quantification and coupling assessment.

[0036] Furthermore, in supervising the training of disaster monitoring components, step P30 of this application embodiment also includes:

[0037] P21: Perform first-order training on the disaster monitoring architecture based on disaster probability dependency to determine the first monitor; P22: Perform second-order training on the disaster monitoring architecture based on flood vulnerability entropy quantification to determine the second monitor; P23: Deploy the first monitor and the second monitor in parallel, and perform third-order training with coupled evaluations of parallel output to build the disaster monitoring component.

[0038] It should be understood that the disaster monitoring component is trained according to the disaster monitoring architecture and embedded in the city information platform to ensure that the disaster monitoring component can comprehensively and accurately assess and monitor urban flood disasters.

[0039] Specifically, the disaster monitoring architecture is first trained using a first-order model based on the probabilistic dependencies of disasters. This training, based on probabilistic graphical models such as Bayesian networks or Markov random fields, quantifies the mutual information relationships among relevant elements of urban flooding disasters, establishing a causal dependency chain for disasters. Specifically, rainfall concentration, topographic convergence potential, and traffic flow resilience are used as the main analytical variables. Historical disaster data and real-time sensing data are used to calculate the conditional and joint probability relationships between different variables. For example, when rainfall concentration increases, the probability of water accumulation in areas with high topographic convergence potential increases, leading to a synchronous change in the probability of decreased traffic flow resilience. By calculating the mutual information among these variables, the dependence strength between various elements in the disaster propagation path can be quantified, and key nodes in disaster evolution can be identified. Based on the training results, a first monitor is constructed to monitor the dynamic changes in disaster causal relationships in real time, achieving early warning and trend identification at the probabilistic dependency level.

[0040] Next, the disaster monitoring architecture undergoes second-order training based on flood vulnerability entropy quantification. Entropy is used here as an indicator of the stability and predictability of the urban system. High entropy indicates a highly unstable and unpredictable state; a small anomaly, such as a clogged manhole cover, can trigger a chain reaction through the network, leading to a rapid escalation of the disaster. Conversely, low entropy indicates a stable system with strong resilience. By quantifying flood vulnerability entropy, a heat map of urban flood vulnerability entropy can be generated, visually demonstrating the vulnerability levels of different urban areas. Based on this analysis, a second monitor is constructed to continuously track the trend of system entropy changes, enabling the monitoring and quantification of the dynamic evolution of urban vulnerability.

[0041] Finally, the first and second monitors are deployed in parallel, and their outputs are used as input for third-order training of the coupled assessment. Coupled assessment refers to risk assessment that comprehensively considers two dimensions: the probability dependency of disasters and the quantification of flood vulnerability entropy. Specifically, through parallel computing and feature-weighted fusion, a comprehensive response function of the system under multi-dimensional perturbations is obtained, achieving a comprehensive assessment of disaster propagation paths, risk diffusion speed, and system resilience. Ultimately, a complete disaster monitoring component is built through third-order training. This component can monitor, assess, and issue early warnings for urban flood disasters in real time, providing accurate data support for the urban information platform.

[0042] P30: By performing multi-source sensing in the city and transmitting the data back to the city information platform, a joint assessment of the disaster monitoring components is conducted under data initialization to determine the disaster monitoring status.

[0043] Furthermore, in this embodiment, step P30 further includes data initialization of the disaster monitoring component:

[0044] P31: Collect multi-source datasets based on urban multi-source sensors, wherein the multi-source datasets include at least radar rainfall data, terrain data, GPS data, and communication information flow field; P32: Initialize the urban 3D model within the disaster monitoring component based on the multi-source datasets; P33: Analyze the disaster monitoring status based on the disaster monitoring component after data initialization.

[0045] Specifically, the first step is to collect multi-source datasets based on urban multi-source sensing. These datasets include at least radar rainfall data, topographic data, GPS data, and communication information flow fields. Radar rainfall data is used to monitor the spatiotemporal distribution of rainfall in real time; topographic data provides detailed information on urban terrain; GPS data is used to locate and track the positions of critical facilities and vehicles; and the communication information flow fields reflect the real-time status of the urban communication network. These multi-source datasets provide rich foundational data for disaster monitoring components, ensuring the comprehensiveness and accuracy of subsequent analyses.

[0046] Next, based on the collected multi-source datasets, the urban 3D model within the disaster monitoring component is initialized. Data initialization refers to integrating and loading the collected multi-source datasets into the urban 3D model to ensure that the model reflects the current real-world state of the city. Specifically, radar rainfall data is overlaid onto the meteorological layer of the urban 3D model to achieve dynamic spatial mapping of rainfall; topographic data and drainage system topology data are spatially registered to update surface runoff paths and pipe network connectivity parameters; and GPS and communication flow data are injected in real-time into the model's functional and load layers to dynamically correct traffic flow and population migration characteristics. Through data initialization, the 3D model is transformed from a static structural model to a dynamic operational model, enabling the disaster monitoring component to have a real-time response capability that reflects the current environmental state.

[0047] Finally, based on the disaster monitoring component after data initialization, the disaster monitoring status is analyzed. By calling the outputs of the aforementioned first and second monitors, the probability dependency relationship and flood vulnerability entropy distribution are jointly determined. Combined with the dynamic characteristics of real-time multi-source data, the disaster risk level, affected area range, and risk evolution trend are calculated. Multi-dimensional indicators of the disaster monitoring status are output, including water level warning index, traffic interruption probability, communication failure risk, and drainage system overload rate, providing accurate decision-making basis for real-time monitoring of urban flood disasters.

[0048] Furthermore, in the joint evaluation under data initialization, step P30 of this application embodiment also includes:

[0049] P34: Based on the first monitor, perform anisotropic projection mapping of the second-layer and third-layer maps on the city's 3D model after data initialization; P35: Perform analysis based on the third load vulnerability map of the second layer to determine the first-step analysis result; P36: Transfer the first-step analysis result to the third layer and perform independent analysis based on the first structural vulnerability map and the second functional vulnerability map to determine the second-step analysis result; P37: Based on the first-step analysis result and the second-step analysis result, perform real-time disaster probability dependency analysis and output one-dimensional monitoring results.

[0050] Optionally, the joint assessment process can be further refined. First, based on the output of the first monitor, anisotropic projection mapping of the second-layer and third-layer maps is performed on the initialized urban 3D model. The second-layer map corresponds to the third load vulnerability map, and the third-layer map includes the first structural vulnerability map and the second functional vulnerability map. Anisotropic projection mapping refers to projecting the map data in different directions to capture vulnerability features in different directions within the urban 3D model. For example, under heavy rainfall scenarios, hydrodynamic effects exhibit significant directional influence in low-lying areas, while the distribution characteristics of structural damage and functional degradation show localized clustering in space. Through anisotropic mapping algorithms, such as weighted Gaussian kernel projection or spatial direction matrix transformation, spatially coupled projections between the rainfall field, drainage impact field, and traffic disturbance field can be achieved, providing a refined input dataset for subsequent analysis steps.

[0051] Next, an analysis based on a two-layer third load vulnerability map is performed to determine the first-step analysis results. This step primarily focuses on the immediate impact of rainfall on the urban drainage system. By analyzing the spatiotemporal concentration of rainfall and the impact of areal rainfall on the drainage system, the current load vulnerability state is determined. For example, through hydrodynamic modeling and time series analysis, the evolution trend of localized waterlogging caused by short-duration heavy rainfall and the system's critical state are identified, thus forming a spatial distribution result reflecting the rainfall impact, i.e., the first-step analysis result. This result describes the immediate response capability of the urban drainage system under the current rainfall scenario and is a key prior input condition for subsequent structural and functional vulnerability analyses.

[0052] Then, the results of the first-step analysis are transferred to the third layer, that is, input into the analysis modules of the first structural vulnerability map and the second functional vulnerability map, and independent analysis is performed to determine the results of the second-step analysis. This step uses the results of the first-step analysis as prior conditions to perform independent analyses of structural vulnerability and functional vulnerability respectively. By analyzing the structural vulnerability map and the functional vulnerability map, the actual situation of each is determined. Specifically, the impact of rainfall is used as a prior constraint to simulate the response of the structural layer and the functional layer respectively: in the structural layer, the distribution of local structural damage is determined by analyzing the bearing state and failure probability of drainage facilities, underground spaces and surface buildings; in the functional layer, the functional degradation areas are identified by assessing the traffic flow interruption rate, communication network connectivity and service accessibility. Further, probabilistic mutual information analysis is performed, that is, analyzing the degree of information sharing between different vulnerability maps to assess their interdependence and form the results of the second-step analysis.

[0053] Finally, based on the results of the first and second step analyses, a real-time disaster probability dependency analysis is performed, outputting one-dimensional monitoring results. This step comprehensively considers the immediate impact of rainfall on the drainage system (first step analysis results) and the actual situation of structural and functional vulnerability (second step analysis results), using a disaster probability dependency model to assess the probability of urban flooding in real time. The one-dimensional monitoring results will intuitively display the city's flood risk under current conditions, providing real-time disaster monitoring data for the city information platform.

[0054] Furthermore, step P30 in this embodiment of the application also includes:

[0055] P38: Based on the lateral interaction channel, the results of the first-step analysis and the second-step analysis are shared with the second monitor; P39: The second monitor performs real-time quantification of flood vulnerability entropy and outputs two-dimensional monitoring results; P310: The one-dimensional monitoring results and the two-dimensional monitoring results are coupled and evaluated by spatiotemporal mapping, and the disaster monitoring status is output.

[0056] In one possible embodiment of this application, the joint assessment process under data initialization can be further extended to ensure that the real-time status of urban flooding disasters can be comprehensively and accurately assessed and monitored, and the final disaster monitoring status can be output.

[0057] Specifically, the first step involves sharing the results of the first and second steps of analysis with the second monitor via a lateral interaction channel. This lateral interaction channel is a data-sharing mechanism established between different monitors, ensuring that each monitor can acquire and utilize the analysis results of other monitors in real time. This step ensures that the analysis results from the first monitor, including the immediate impact of rainfall on the drainage system and the actual status of structural and functional vulnerabilities, can be obtained by the second monitor, providing data support for subsequent flood vulnerability entropy quantification.

[0058] Next, the second monitor performs real-time quantification of flood vulnerability entropy, outputting two-dimensional monitoring results. This step, based on the functionality of the second monitor, utilizes flood vulnerability entropy quantification technology to conduct real-time assessment of the overall vulnerability status of the city. Specifically, by sampling in real-time the fluctuations in drainage system flow, the time-varying characteristics of traffic flow, and the connectivity of communication networks, the conditional entropy and joint entropy values ​​of each spatial node are calculated to quantify the system's complexity and stability. High entropy values ​​indicate that the urban system is in a highly unstable and unpredictable state, while low entropy values ​​indicate that the system is stable and has strong shock resistance. Through differential analysis of entropy values ​​at multiple time points, the temporal gradient field of flood vulnerability entropy is obtained, thus outputting two-dimensional monitoring results presented in the form of spatial thermal distribution. These results can reflect the dynamic disaster sensitivity and risk evolution trends of different areas of the city under the current environment.

[0059] Finally, a spatiotemporal mapping coupling assessment is performed on the one-dimensional and two-dimensional monitoring results to output the disaster monitoring status. This process constructs a spatiotemporal fusion model to conduct multidimensional correlation analysis between one-dimensional monitoring results (reflecting the risk evolution trajectory under probability dependence) and two-dimensional monitoring results (reflecting the risk diffusion pattern under entropy spatial distribution), achieving global awareness of disaster risk and identification of local responses. Specifically, a spatiotemporal covariance matrix and mutual information mapping algorithm are used to calculate the risk coupling strength between different time slices and spatial units, thereby forming a disaster monitoring status matrix. Furthermore, based on this matrix, comprehensive disaster status indicators are output, including risk level, affected area, vulnerability distribution, and trend vector, realizing a dynamic closed loop from the probability layer to the entropy layer, and from local analysis to global assessment, which can provide comprehensive and accurate decision support for urban flood disaster management.

[0060] P40: Based on the disaster monitoring status, provide guidance on urban flood disaster management and targeted monitoring.

[0061] Furthermore, step P40 in this embodiment of the application also includes:

[0062] P41: Identify the disaster monitoring status, locate risk control points and determine risk control plans, and generate disaster management instructions; P42: Generate targeted monitoring instructions based on the risk control points and risk control trends; P43: Decouple the disaster management instructions and targeted monitoring instructions into minimum execution units and decentralize them to multiple threads to perform dynamic monitoring of urban flood disasters.

[0063] Specifically, the disaster monitoring status is first identified and analyzed. Based on a comprehensive judgment of risk level, vulnerability distribution, and trend vector in the monitoring status matrix, risk control points in the city are automatically located and corresponding disaster management instructions are generated. Risk control points refer to nodes in the disaster propagation chain that have high risk propagation potential or key control functions, including high-flooding-risk areas, critical nodes in drainage systems, traffic bottlenecks, and areas with high communication density. Furthermore, based on these risk control points, spatiotemporal clustering algorithms and multidimensional risk weight assessments are used to determine their spatial location and risk level, and corresponding risk control plans are generated according to different levels, including emergency response measures, resource allocation plans, and personnel evacuation schemes. Then, disaster management instructions are generated based on the risk control plans, which will guide subsequent disaster management actions.

[0064] Next, based on the identified risk control points and their risk evolution trends, targeted monitoring instructions are generated. This means that based on the identification of risk control points and the formulation of risk control plans, targeted monitoring instructions are further generated. Targeted monitoring instructions refer to monitoring tasks specific to particular risk control points. These tasks aim to track changes in the status of risk control points in real time, assess risk trends, and adjust disaster management measures promptly. For example, when the flood level rise rate corresponding to a risk control point exceeds a threshold, the system automatically schedules water level sensors, video surveillance, and drainage flow meters in the corresponding area to perform high-frequency sampling. If the risk control point is located at a transportation hub or underground passage, the instructions include priority tasks for traffic flow monitoring and water depth detection. Through targeted monitoring, the development dynamics of disasters can be grasped more accurately, providing real-time data support for disaster management.

[0065] Finally, disaster management instructions and targeted monitoring instructions are decoupled and decentralized to the smallest execution units, enabling dynamic monitoring of urban flooding disasters. This step involves breaking down disaster management instructions and targeted monitoring instructions into the smallest execution units. These units are the basic units for executing tasks and can operate independently to complete specific tasks, such as single-node drainage scheduling, local traffic control, or single-area video verification. Decoupling by the smallest execution units improves the flexibility and efficiency of task execution. Simultaneously, decentralizing tasks to multiple execution units for parallel processing ensures rapid response and execution. In this way, the city can achieve multi-departmental and multi-system coordinated response during disasters, ensuring the real-time nature and efficiency of disaster monitoring, management, and control.

[0066] In summary, the embodiments of this application have at least the following technical effects:

[0067] This application achieves spatially refined identification of urban flood risk by constructing a 3D urban model and decoupling the assessment of structural, functional, and load vulnerability; it establishes an intelligent analysis mechanism based on disaster probability dependence and flood vulnerability entropy quantification through supervised training of disaster monitoring components, enabling dynamic prediction of disaster evolution; it completes real-time data initialization and joint evaluation of the model by combining multi-source urban sensor data, improving monitoring accuracy and response speed; and it generates risk control and targeted monitoring instructions through disaster status identification, which are executed through multi-threaded scheduling, realizing an automated closed loop from disaster monitoring, assessment to management, significantly enhancing the intelligence and emergency response capabilities of urban flood control and drainage.

[0068] It has achieved a three-dimensional disaster monitoring architecture based on big data, enabling multi-dimensional dynamic assessment of urban flood disasters and improving the technical effect of disaster identification accuracy and response efficiency.

[0069] Example 2, based on the same inventive concept as the big data-based dynamic monitoring method for urban flood disasters in the aforementioned examples, such as... Figure 2 As shown, this application provides a dynamic monitoring system for urban flood disasters based on big data. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0070] The multidimensional decoupling assessment module 11 is used to construct a three-dimensional urban model and to conduct a decoupling assessment of the structural vulnerability, functional vulnerability and load vulnerability based on the three-dimensional urban model for the dimension of flood disaster, thereby constructing a disaster supervision architecture.

[0071] The disaster supervision component training module 12 is used to supervise the training of disaster supervision components and embed them in the urban information platform according to the disaster supervision architecture. The training directions include disaster probability dependency, flood vulnerability entropy quantification and coupling assessment.

[0072] The joint assessment module 13 is used to conduct a joint assessment of the disaster monitoring components under data initialization by performing multi-source sensing in the city and transmitting the data back to the city information platform, so as to determine the disaster monitoring status.

[0073] The disaster management module 14 is used to manage urban flood disasters and provide targeted monitoring guidance based on the disaster monitoring status.

[0074] Furthermore, the multidimensional decoupling evaluation module 11 is also used to perform the following steps:

[0075] Based on the city's three-dimensional model, taking drainage catchment areas and pipe networks as the first urban elements, a joint assessment of topographic convergence potential and drainage path dependence is performed on each drainage catchment area to determine the obstruction linear relationship, wherein there is a one-to-one correspondence between drainage catchment areas and obstruction linear relationships; using the first urban elements as the basis, the obstruction linear relationship is marked by basis location matching to generate a first structural vulnerability map.

[0076] Furthermore, the multidimensional decoupling evaluation module 11 is also used to perform the following steps:

[0077] Based on the city's three-dimensional model, using the road network as the second urban element, a road network traffic assessment is conducted on key transportation hubs to determine traffic flow resilience relationships, wherein key transportation hubs correspond one-to-one with traffic flow resilience relationships; using the second urban element as a base, the base location matching and marking of the traffic flow resilience relationships are performed to generate a second functional vulnerability map.

[0078] Furthermore, the multidimensional decoupling evaluation module 11 is also used to perform the following steps:

[0079] Using the first and second urban elements as a base, the spatial and temporal concentration of rainfall and areal rainfall are assessed to evaluate drainage impact, determine drainage impact relationships, and perform base location matching and marking to generate a third load vulnerability map. A disaster monitoring framework is constructed with the urban 3D model as one layer, the third load vulnerability map as the second layer, and the first structural vulnerability map and the second functional vulnerability map in parallel as the third layer.

[0080] Furthermore, the supervision component training module 12 is also used to perform the following steps:

[0081] The disaster monitoring architecture is trained in a first-order manner based on the probability dependency of disasters to determine the first monitor; the disaster monitoring architecture is trained in a second-order manner based on the quantification of flood vulnerability entropy to determine the second monitor; the first monitor and the second monitor are deployed in parallel, and a third-order training is performed with the coupled evaluation of the parallel output to build the disaster monitoring component.

[0082] Furthermore, the joint evaluation module 13 is also used to perform the following steps:

[0083] Based on urban multi-source sensing, a multi-source dataset is collected, which includes at least radar rainfall data, terrain data, GPS data, and communication information flow field. Based on the multi-source dataset, the urban 3D model in the disaster monitoring component is initialized. Based on the initialized disaster monitoring component, the disaster monitoring status is analyzed.

[0084] Furthermore, the joint evaluation module 13 is also used to perform the following steps:

[0085] Based on the first monitor, anisotropic projection mapping of the second-layer and third-layer maps is performed on the city's three-dimensional model after data initialization; analysis based on the third load vulnerability map of the second layer is performed to determine the first-step analysis result; the first-step analysis result is transferred to the third layer, and independent analysis based on the first structural vulnerability map and the second functional vulnerability map is performed to determine the second-step analysis result; based on the first-step analysis result and the second-step analysis result, real-time disaster probability dependency analysis is performed, and one-dimensional monitoring results are output.

[0086] Furthermore, the joint evaluation module 13 is also used to perform the following steps:

[0087] According to the lateral interaction channel, the first-step analysis results and the second-step analysis results are shared to the second monitor; the second monitor performs real-time quantification of flood vulnerability entropy and outputs two-dimensional monitoring results; the one-dimensional monitoring results and the two-dimensional monitoring results are coupled and evaluated by spatiotemporal mapping, and the disaster monitoring status is output.

[0088] Furthermore, the disaster management module 14 is also used to perform the following steps:

[0089] The system identifies the disaster monitoring status, locates risk control points and determines risk control plans, and generates disaster management instructions. Based on the risk control points, it generates targeted monitoring instructions according to risk control trends. The system decouples the disaster management instructions and the targeted monitoring instructions into minimum execution units and decentralizes them through multiple threads to perform dynamic monitoring of urban flood disasters.

[0090] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0091] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0092] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for dynamic monitoring of urban flood disasters based on big data, characterized in that, The method includes: A three-dimensional urban model is constructed, which is based on digital twin city technology. A decoupled assessment of the structural vulnerability, functional vulnerability and load vulnerability of the urban three-dimensional model is carried out for the dimension of flood disaster, and a disaster monitoring architecture is constructed. According to the disaster monitoring architecture, the disaster monitoring component is trained and embedded in the urban information platform. The training directions include disaster probability dependency, flood vulnerability entropy quantification and coupling assessment. By performing multi-source sensing in the city and transmitting the data back to the city information platform, a joint assessment of the disaster monitoring components is conducted under data initialization to determine the disaster monitoring status. Based on the disaster monitoring status, provide guidance for urban flood disaster management and targeted monitoring; A decoupled structural vulnerability assessment based on a 3D urban model was conducted to address flood disasters, including: Based on the aforementioned three-dimensional urban model, taking drainage catchment areas and pipe networks as the first urban elements, a joint assessment of topographic convergence potential and drainage path dependence is conducted on each drainage catchment area to determine the obstruction linear relationship, wherein there is a one-to-one correspondence between drainage catchment areas and obstruction linear relationships. Using the first urban element as a base, the base position matching and marking are performed on the obstruction linear relationship to generate a first structural vulnerability map; Functional vulnerability decoupling assessment based on a 3D urban model was conducted to address flood disasters, including: Based on the aforementioned three-dimensional urban model, with the road network as the second urban element, a road network traffic assessment is conducted on key transportation hubs to determine traffic flow resilience relationships. Here, key transportation hubs correspond one-to-one with traffic flow resilience relationships. Using the second urban element as a base, the traffic flow resilience relationship is matched and marked with the base location to generate a second functional vulnerability map. A decoupled assessment of load vulnerability based on a 3D urban model is conducted to address flood disasters, and a disaster monitoring architecture is constructed, including: Based on the first and second city elements, the drainage impact is assessed by the spatiotemporal concentration of rainfall and the areal rainfall, the drainage impact relationship is determined and the base location is matched and marked to generate a third load vulnerability map. A disaster monitoring architecture is constructed by using a three-dimensional urban model as one layer, a third load vulnerability map as the second layer, and the first structural vulnerability map and the second functional vulnerability map in parallel as the third layer. The training component for disaster monitoring includes: The disaster monitoring architecture is trained in a first-order manner based on the disaster probability dependency relationship to determine the first monitor; This training uses a probabilistic graphical model to quantify the mutual information relationships among factors related to urban flooding disasters and establish a causal dependency chain for disasters. The disaster monitoring architecture is subjected to second-order training based on flood vulnerability entropy quantization to determine the second monitor; Entropy is used as an indicator to measure the stability and predictability of urban systems. By quantifying flood vulnerability entropy, a heat map of urban flood vulnerability entropy is generated. The first and second monitors are deployed in parallel, and the coupled evaluations with parallel outputs are used for third-order training to build the disaster monitoring component. 2.The big data-based urban flood disaster dynamic monitoring method according to claim 1, wherein, Data initialization of the disaster monitoring component includes: Based on urban multi-source sensing, a multi-source dataset is collected, wherein the multi-source dataset includes at least radar rainfall data, terrain data, GPS data, and communication information flow field. Based on the multi-source dataset, the city 3D model within the disaster monitoring component is initialized. Based on the disaster monitoring component after data initialization, analyze the disaster monitoring status. 3.The big data-based urban flood disaster dynamic monitoring method according to claim 2, characterized in that, Perform joint evaluation under data initialization, including: Based on the first monitor, anisotropic projection mapping of the second-layer and third-layer maps is performed on the city's three-dimensional model after data initialization; Perform analysis based on the second-layer third load vulnerability map to determine the results of the first-step analysis; The results of the first step analysis are transferred to the third layer to perform independent analysis based on the first structural vulnerability map and the second functional vulnerability map to determine the results of the second step analysis. Based on the results of the first and second steps of analysis, a real-time disaster probability dependency analysis is performed, and a one-dimensional monitoring result is output. 4.The big data-based urban flood disaster dynamic monitoring method according to claim 3, characterized in that, According to the lateral interaction channel, the results of the first step analysis and the results of the second step analysis are shared with the second monitor; The second monitor performs real-time quantification of flood vulnerability entropy and outputs two-dimensional monitoring results; The one-dimensional monitoring results and the two-dimensional monitoring results are coupled and evaluated through spatiotemporal mapping, and the disaster monitoring status is output. 5.The big data-based urban flood disaster dynamic monitoring method according to claim 1, wherein, Provide guidance on urban flood disaster management and targeted monitoring, including: Identify the disaster monitoring status, locate risk control points and determine risk control plans, and generate disaster management instructions; Based on the risk control points, targeted monitoring instructions are generated according to the risk control trends; The disaster management instructions and the targeted monitoring instructions are decoupled into minimum execution units and decentralized to multiple threads to perform dynamic monitoring of urban flood disasters.

6. The city flood disaster dynamic monitoring system based on big data adopts the city flood disaster dynamic monitoring method based on big data according to any one of claims 1-5, characterized in that, The system includes: The multidimensional decoupled assessment module is used to construct a three-dimensional urban model and conduct a decoupled assessment of the structural vulnerability, functional vulnerability, and load vulnerability based on the three-dimensional urban model for flood disasters, thereby constructing a disaster monitoring architecture. The disaster supervision component training module is used to supervise and train the disaster supervision component according to the disaster supervision architecture and embed it in the urban information platform. The training directions include disaster probability dependency, flood vulnerability entropy quantification and coupling assessment. The joint assessment module is used to conduct a joint assessment of the disaster monitoring components under data initialization by performing multi-source sensing in the city and transmitting the data back to the city information platform, so as to determine the disaster monitoring status. The disaster management module is used to manage urban flood disasters and provide targeted monitoring guidance based on the disaster monitoring status.

Images