Early warning method and system for dynamic personnel transfer in flood storage and detention area

Abstract

The invention relates to the technical field of flood control and disaster reduction, and particularly discloses a flood storage and detention area dynamic personnel transfer early warning method and system, and the method comprises the steps: collecting and fusing hydrological monitoring data, personnel dynamic distribution information and geographic information in real time, and constructing a unified dynamic data base; based on the dynamic data base, generating a gridded real-time risk level by fusing a flood routing model and real-time personnel density data; according to the real-time risk grade, dynamically triggering and issuing a grading early warning instruction; based on the real-time submerging prediction result and the dynamic road condition information, generating and dynamically updating an optimal transfer path for the threatened individual; pushing customized early warning information corresponding to the early warning instruction and the optimal transfer path to an individual terminal; according to the invention, based on hectometer-level gridding real-time risk assessment and a three-level early warning mechanism, a submerging prediction model and a special group positioning technology are fused, and high-risk point location coverage is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flood control and disaster reduction, and particularly relates to a dynamic personnel transfer early warning method and system for a flood storage and detention area. BACKGROUND

[0002] With the complication of the flood control engineering system and the frequent occurrence of extreme climate events, personnel transfer early warning for a flood storage and detention area has become a core link to protect people's lives and safety. In a dynamically changing disaster environment, real-time fusion analysis of hydrological monitoring data, population flow information and geographic spatial data is the key to realizing accurate early warning. However, due to the diversity of data types and the complexity of the scene, traditional technologies face many challenges in actual application, mainly in the following aspects: Multi-source data coordination failure: hydrological sensors (minute-level update), population statistics data (year / month-level update), and geographic information (static storage) are fragmented in terms of time and space scale, resulting in a high risk misjudgment rate and failing to meet the use of real scenes.

[0003] Dynamic response missing: relying on fixed water level thresholds, without fusion of storm runoff acceleration or night population aggregation variables, the early warning omission rate is extremely high, and accurate early warning cannot be achieved.

[0004] Decision execution disconnection: early warning instructions are disconnected from real-time path planning and evacuation resource scheduling, increasing the probability that part of the evacuees will encounter a secondary danger.

[0005] Therefore, how to provide a dynamic personnel transfer early warning system for a flood storage and detention area is a problem to be solved by those skilled in the art. SUMMARY

[0006] The purpose of the present application is to provide a dynamic personnel transfer early warning method and system for a flood storage and detention area to solve the problems raised in the background.

[0007] To achieve the above purpose, the present application provides the following technical solutions: A dynamic personnel transfer early warning method for a flood storage and detention area, the method comprising: real-time collection and fusion of hydrological monitoring data, personnel dynamic distribution information and geographic information to construct a unified dynamic data base; based on the dynamic data base, generating a grid-based real-time risk level by fusing a flood evolution model and real-time personnel density data; according to the real-time risk level, dynamically triggering and issuing graded early warning instructions; based on real-time inundation prediction results and dynamic road condition information, generating and dynamically updating an optimal transfer path for threatened individuals; pushing customized early warning information corresponding to the early warning instructions and the optimal transfer path to individual terminals.

[0008] As a further scheme of the present application, the step of collecting and fusing hydrological monitoring data, personnel dynamic distribution information and geographic information in real time to construct a unified dynamic data basement specifically comprises: Through the sensor network arranged at the nodes of the river channel, gate and flood storage area, high-frequency collection of water level, flow rate, rainfall and soil moisture data is performed, and the data is returned and processed through wireless communication; A refined dynamic population distribution model is constructed by fusing a hundred-meter grid population heat map generated based on operator base station signaling, a civil population database containing special group identification and a traffic hub flow state obtained through AI visual recognition. A high-precision digital elevation model, road network vector data, building contour and real-time capacity information of shelters are integrated, and multi-source spatio-temporal data superposition analysis is performed through a geographic information system engine.

[0009] As a further scheme of the present application, the step of generating a grid-based real-time risk level by fusing a flood evolution model and real-time personnel density data based on the dynamic data basement specifically comprises: A flood evolution model based on two-dimensional hydrodynamic equations is used to simulate and output the spatio-temporal evolution of the flood inundation range, depth and flow rate of a hundred-meter grid in the future period; The population aggregation degree in a specified area is quantified in real time by fusing the population heat map, AI visual recognition data and special group information; A risk quantification index updated at a minute level is generated by coupling the flood inundation depth, personnel density and terrain evacuation difficulty coefficient in a specified grid through a weighted algorithm, and the risk quantification index is used to divide the risk level into four levels of low, medium, high and urgent.

[0010] As a further scheme of the present application, the risk value = α × D + β × P + γ × T; Where D is the standardized flood inundation depth, P is the standardized personnel density, and T is the terrain evacuation difficulty coefficient; α, β, γ are preset weight coefficients, and α + β + γ = 1.

[0011] As a further scheme of the present application, the values of the weight coefficients α, β and γ are obtained through machine learning training based on historical disaster data, and / or can be configured and adjusted according to the regional characteristics of different flood storage areas.

[0012] As a further scheme of the present application, the step of generating and dynamically updating an optimal transfer path for a threatened individual based on real-time inundation prediction results and dynamic road condition information specifically comprises: Based on the flood evolution model, the flood inundation situation in the next 2 to 6 hours is predicted; With a response speed of seconds, combined with real-time updated flooding prediction results, traffic state data and infrastructure state information, the optimal risk avoidance path is calculated for individuals; Path re-planning is triggered when path risk is monitored.

[0013] The application also provides a dynamic personnel transfer early warning system for flood storage areas, which is used to realize the dynamic personnel transfer early warning method for flood storage areas. The multi-source data acquisition module is used to acquire and fuse hydrological monitoring data, personnel dynamic distribution information and geographic information in real time, and construct a unified dynamic data base. The dynamic risk assessment module is used to generate a grid-based real-time risk level by fusing a flood evolution model and real-time personnel density data based on the dynamic data base. The hierarchical early warning decision module is used to dynamically trigger and issue hierarchical early warning instructions according to the real-time risk level. The intelligent path planning module is used to generate and dynamically update the optimal transfer path for threatened individuals based on real-time flooding prediction results and dynamic road condition information. The early warning information publishing module is used to push customized early warning information corresponding to the early warning instructions and the optimal transfer path to individual terminals.

[0014] Compared with the prior art, the application has the beneficial effects that: the application constructs a dynamic fusion engine of hydrological sensor network, mobile phone signaling population heat map and GIS geographic information, realizes minute-level spatiotemporal alignment of multi-source data, significantly improves the timeliness of risk determination, significantly reduces the misjudgment rate, supports rapid and efficient risk scanning, and completely solves the problem of decision lag caused by data island.

[0015] Based on the real-time risk assessment of a hundred-meter grid and the three-level early warning mechanism, the flooding prediction model and the special group positioning technology are fused to realize high-risk point coverage. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only some embodiments of the application.

[0017] Figure 1 The real-time risk level calculation and response process example diagram provided for the embodiments of the application.

[0018] Figure 2 The core technical architecture and flowchart of "second-level" response provided for the embodiments of the application.

[0019] Figure 3 The dynamic path generation and update example diagram provided for the embodiments of the application.

[0020] Figure 4 This is a structural block diagram of a dynamic personnel evacuation early warning system for flood storage and detention areas, provided as an embodiment of the present invention. Detailed Implementation

[0021] To make the technical problems to be solved, the technical solutions, and the beneficial effects 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 merely illustrative of the present invention and are not intended to limit the present invention.

[0022] In this embodiment of the invention, a method for dynamic early warning of population relocation in flood storage and detention areas is provided, the method comprising: Real-time collection and integration of hydrological monitoring data, dynamic distribution information of personnel, and geographic information to construct a unified dynamic data base; Based on the aforementioned dynamic data base, a gridded real-time risk level is generated by fusing flood evolution models with real-time population density data. Based on the real-time risk level, dynamically trigger and issue graded early warning instructions; Based on real-time flooding prediction results and dynamic road condition information, the optimal evacuation path is generated and dynamically updated for threatened individuals. Customized early warning information corresponding to the early warning command and the optimal transfer path is pushed to individual terminals.

[0023] As a preferred embodiment of the present invention, the step of real-time acquisition and fusion of hydrological monitoring data, dynamic distribution information of personnel, and geographic information to construct a unified dynamic data base specifically includes: The sensor network deployed at nodes in river channels, sluice gates and flood storage areas collects high-frequency data on water level, flow velocity, rainfall and soil moisture, and transmits the data back via wireless communication and performs data cleaning. By integrating a 100-meter-level grid population heat map generated based on operator base station signaling, a civil affairs population database containing special group identifiers, and the flow status of people at transportation hubs obtained through AI visual recognition, a refined dynamic population distribution model is constructed. It integrates high-precision digital elevation models, road network vector data, building outlines, and real-time capacity information of shelters, and performs overlay analysis of multi-source spatiotemporal data through a geographic information system engine.

[0024] In this embodiment, the hydrological sensor network consists of a water level radar station, a flow velocity sensor, a rainfall sensor, and a soil moisture meter, covering river / sluice gate / flood storage area nodes. It collects multi-physical quantity data at a high frequency of 1-5 minutes, transmits it back to the edge nodes via LoRa / NB-IoT, and filters out abnormal values ​​of equipment failure and environmental interference based on an adaptive cleaning algorithm. The dynamic population distribution information is integrated by combining the population heat map of the operator's base station grid at a level of 100 meters, the permanent / temporary resident population database of the civil affairs department (identifying special groups such as the elderly, children, the sick and disabled), and the flow of people at transportation hubs identified by AI cameras to construct a refined dynamic population distribution model; The geographic information integration platform integrates facility attribute layers such as DEM elevation model, road network vector map, building outline and real-time capacity of shelters, and incorporates GPS dynamic data; it achieves multi-source layer overlay analysis through a unified spatiotemporal coordinate system using a GIS engine, supporting flood simulation and evacuation route planning decisions.

[0025] As a preferred embodiment of the present invention, the step of generating a gridded real-time risk level based on the dynamic data base by fusing the flood evolution model and real-time population density data specifically includes: Using a flood evolution model based on two-dimensional hydrodynamic equations, we simulate and output the spatiotemporal evolution of flood inundation range, depth, and flow velocity in a grid of 100 meters in the future period; By integrating the population heat map, the AI ​​visual recognition data, and the information on special groups, the population concentration in a specified area can be quantified in real time. By coupling the flood inundation depth, population density, and terrain evacuation difficulty coefficient within a specified grid using a weighted algorithm, a risk quantification index updated every minute is generated. This risk quantification index is used to classify four risk levels: low, medium, high, and emergency.

[0026] In this embodiment, the flood evolution model is constructed based on two-dimensional hydrodynamic equations. Driven by a high-resolution digital elevation model and real-time hydrological data, it dynamically simulates the spatiotemporal evolution of flood inundation range, depth, and flow velocity, and outputs a gridded risk distribution map at the hundred-meter level, providing accurate spatiotemporal early warning basis for personnel evacuation. By integrating mobile signaling grid heat maps at a scale of 100 meters, AI visual recognition of pedestrian flow at transportation hubs, and databases of special groups under civil affairs, population density data is used to construct a dynamic distribution model, which quantifies the degree of regional population concentration in real time, supporting grid-based assessment and graded early warning decision-making for flood risk.

[0027] The step of dynamically triggering and issuing graded early warning instructions based on the real-time risk level is as follows: Real-time risk levels, coupled with dynamic coupling of flood inundation depth, population density and terrain evacuation difficulty coefficient within the grid, generate minute-updated risk quantification indicators (low / medium / high / emergency levels) based on a weighted algorithm, driving the issuance of graded early warning instructions and precise evacuation resource scheduling; The three-level early warning system dynamically generates progressive early warning instructions based on real-time risk levels, driving precise individual evacuation and resource allocation.

[0028] The intelligent route planning module generates the optimal transfer route for each individual within seconds based on real-time flooding prediction and road condition changes, dynamically avoiding high-risk road sections. Achieving "second-level" path planning response (typically referring to completing calculation and push within 1-3 seconds) is crucial for this system's ability to cope with sudden flood emergencies. Its core technology lies in the combination of preprocessing, distributed computing, and intelligent algorithms, rather than the "calculation from scratch" mode of traditional navigation software.

[0029] The core technical architecture and process for achieving "second-level" response are described in [link to documentation]. Figure 2 The core of this approach lies in overcoming the computational bottlenecks of traditional algorithms through a combination of preprocessing, distributed computing, and intelligent algorithms. Specifically, it is achieved through the following four technologies: preprocessing and caching mechanisms; real-time lightweight data injection; distributed edge computing architecture; and lightweight dynamic weighting algorithms.

[0030] As a preferred embodiment of the present invention, the risk value = α×D + β×P + γ×T; Where D is the standardized flood inundation depth, P is the standardized personnel density, and T is the terrain evacuation difficulty coefficient; α, β, and γ are preset weighting coefficients, and α+β+γ=1.

[0031] like Figure 1 As shown in this embodiment, the example scenario is set up as follows: Geographic location: A 500m×500m grid area (numbered G-07) within a flood storage and detention area.

[0032] Timeline: 9:00 to 9:30 a.m. on a certain day during the flood's evolution.

[0033] Key parameters: flood inundation depth (D), population density (P), terrain evacuation difficulty coefficient (T), and weighting coefficients (α=0.5, β=0.3, γ=0.2).

[0034] The risk level is updated minute by minute as the flood evolves (9:00 low risk → 9:05 high risk → 9:30 emergency), breaking through the traditional fixed threshold response mode.

[0035] 9:05 High risk cause: Although the water depth is only 0.8m (moderate threat), the population density (60 people / km²) is the dominant factor (weight β=0.3), triggering a level 2 warning.

[0036] 9:30 Emergency Cause: Despite the reduction in personnel, the water depth (1.5m) and bridge interruption (terrain coefficient T=0.8) significantly increase individual risk, requiring the highest level of intervention.

[0037] Level 2 alert: Dispatch buses to transfer the large group (62 people), matching the capacity of the shelters; Level 3 warning: For a small number of stranded people (10 people), switch to drones + speedboats for precise rescue to avoid wasting resources.

[0038] In a preferred embodiment of the present invention, the values ​​of the weighting coefficients α, β, and γ are obtained through machine learning training based on historical disaster data, and / or can be configured and adjusted according to the regional characteristics of different flood storage and detention areas.

[0039] In this embodiment, the weights are derived through training on historical disaster situations. For example, if a region has complex terrain (many hills), γ is increased to 0.3; if a region has a sparse population, β is decreased to 0.2.

[0040] As a preferred embodiment of the present invention, the step of generating and dynamically updating the optimal evacuation path for threatened individuals based on real-time flooding prediction results and dynamic road condition information specifically includes: Based on the aforementioned flood evolution model, the flood inundation situation is predicted for the next 2 to 6 hours; With a response time of seconds, combined with real-time updated flooding prediction results, traffic status data and infrastructure status information, the optimal evacuation path is calculated for individuals; When a path risk is detected, path replanning is triggered.

[0041] In this embodiment, inundation prediction is based on a two-dimensional hydrodynamic model, which integrates real-time hydrological data and high-precision DEM topography to dynamically simulate the flood evolution process (including changes in inundation range, depth and flow velocity) in the next 2-6 hours, and outputs a gridded spatiotemporal evolution map at the hundred-meter level, providing core decision-making basis for path planning and graded early warning. Based on real-time flooding predictions and dynamic road conditions, the system generates optimal evacuation routes for individuals, updating and avoiding high-risk road sections in real time.

[0042] The early warning information release module dynamically matches multi-level early warning strategies and accurately pushes customized evacuation instructions and real-time risk avoidance routes to individual terminals; like Figure 3 As shown, the example scenario is set up as follows: User: Ms. Li, located in grid G-07 (residential area) of the flood storage and detention area.

[0043] Destination: Designated shelter S3 (5 km in a straight line).

[0044] Timeline: Within 30 minutes of the flood's evolution.

[0045] System inputs: real-time flooding prediction model (water depth / flow velocity in the next 30 minutes), real-time traffic conditions (GPS floating car data from the transportation department), and infrastructure status (bridge load-bearing sensors and camera water accumulation identification).

[0046] Initial path planning (9:00): Flooding prediction: The water depth in the area where bridge B1 is located is expected to be >1.0m in 20 minutes (critical value). Real-time traffic conditions: Main road H12 is clear (floating car average speed 40km / h). Bridge status: B1 load-bearing capacity is normal (sensor data: load < 30 tons).

[0047] Path generation P1: G-07 → Bridge B1 → Main Road H12 → Shelter S3 (Total length 5.2km, estimated time 10 minutes).

[0048] Ms. Li's mobile app displayed a blue navigation route: "Please evacuate along this route. It is estimated that you will arrive in 10 minutes."

[0049] First risk trigger: Bridge B1 load cell alarm (structural damage caused by foundation scour, load capacity drops sharply to <10 tons).

[0050] System response: Remove path P1 in real time (mark B1 as "high-risk section"); start alternative route calculation: select detour auxiliary road L9 (no bridge, but distance increased to 6.1km).

[0051] Update route P2: G-07 → Auxiliary road L9 → Shelter S3 (estimated time 14 minutes).

[0052] The app pops up a message: "Warning! Bridge B1 is dangerous. Your route has been replanned."

[0053] Second risk trigger: Flooding forecast update: The low-lying section of auxiliary road L9 is expected to accumulate more than 0.8m of water in 5 minutes (small vehicles will not be able to pass); AI-powered camera detection: There is already water accumulation at the L9 entrance (actual water depth 0.3m and the water level is rising).

[0054] System response: Based on comprehensive assessment, L9 is not passable; high-priority route initiated: elevated G7 (high altitude, no risk of water accumulation).

[0055] Update route P3: G-07 → Elevated G7 → Shelter S3 (Total length 6.5km, estimated travel time 12 minutes).

[0056] Emergency command push: APP voice broadcast + SMS: "Emergency! L9 road is flooded, please detour via elevated G7 immediately!"

[0057] Ms. Li eventually arrived at shelter S3 via route P3; Post-incident data review: Bridge B1 collapsed at 9:25; auxiliary road L9 was flooded to 1.2m at 9:20 (completely interrupted). The system successfully avoided two high-risk points, ensuring user safety.

[0058] like Figure 4 As shown, the present invention also provides a dynamic personnel evacuation early warning system for flood storage and detention areas, used to implement the aforementioned dynamic personnel evacuation early warning method for flood storage and detention areas, the system comprising: The multi-source data acquisition module is used to collect and integrate hydrological monitoring data, dynamic distribution information of personnel, and geographic information in real time to build a unified dynamic data base. The dynamic risk assessment module is used to generate a gridded real-time risk level based on the dynamic data base by integrating the flood evolution model and real-time population density data; The tiered early warning decision module is used to dynamically trigger and issue tiered early warning instructions based on the real-time risk level. The intelligent route planning module is used to generate and dynamically update the optimal evacuation route for threatened individuals based on real-time flooding prediction results and dynamic road condition information. The early warning information dissemination module is used to push customized early warning information corresponding to the early warning command and the optimal transfer path to individual terminals.

[0059] This invention constructs a dynamic fusion engine for hydrological sensor networks, mobile phone signal population heat maps, and GIS geographic information, achieving minute-level spatiotemporal alignment of multi-source data. This significantly improves the timeliness of risk assessment, reduces the false judgment rate, supports rapid and efficient risk scanning, and completely solves the problem of decision-making lag caused by data silos.

[0060] Based on a 100-meter-level grid-based real-time risk assessment and a three-level early warning mechanism, and by integrating inundation prediction models and special population positioning technologies, high-risk locations are covered.

[0061] By combining flood prediction with real-time road condition-driven second-level path planning, the success rate of hazard avoidance is improved and the rate of secondary encounters is reduced.

[0062] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dynamic early warning of population relocation in flood storage and detention areas, characterized in that, The method includes: Real-time collection and integration of hydrological monitoring data, dynamic distribution information of personnel, and geographic information to construct a unified dynamic data base; Based on the aforementioned dynamic data base, a gridded real-time risk level is generated by fusing flood evolution models with real-time population density data. Based on the real-time risk level, dynamically trigger and issue graded early warning instructions; Based on real-time flooding prediction results and dynamic road condition information, the optimal evacuation path is generated and dynamically updated for threatened individuals. Customized early warning information corresponding to the early warning command and the optimal transfer path is pushed to individual terminals.

2. The method for dynamic personnel relocation early warning in flood storage and detention areas according to claim 1, characterized in that, The specific steps for real-time acquisition and fusion of hydrological monitoring data, dynamic population distribution information, and geographic information to construct a unified dynamic data base include: The sensor network deployed at nodes in river channels, sluice gates and flood storage areas collects high-frequency data on water level, flow velocity, rainfall and soil moisture, and transmits the data back via wireless communication and performs data cleaning. By integrating a 100-meter-level grid population heat map generated based on operator base station signaling, a civil affairs population database containing special group identifiers, and the flow status of people at transportation hubs obtained through AI visual recognition, a refined dynamic population distribution model is constructed. It integrates high-precision digital elevation models, road network vector data, building outlines, and real-time capacity information of shelters, and performs overlay analysis of multi-source spatiotemporal data through a geographic information system engine.

3. The method for dynamic personnel relocation early warning in flood storage and detention areas according to claim 1, characterized in that, The step of generating a gridded real-time risk level based on the dynamic data base by fusing the flood evolution model with real-time population density data specifically includes: Using a flood evolution model based on two-dimensional hydrodynamic equations, we simulate and output the spatiotemporal evolution of flood inundation range, depth, and flow velocity in a grid of 100 meters in the future period; By integrating the population heat map, the AI ​​visual recognition data, and the information on special groups, the population concentration in a specified area can be quantified in real time. By coupling the flood inundation depth, population density, and terrain evacuation difficulty coefficient within a specified grid using a weighted algorithm, a risk quantification index updated every minute is generated. This risk quantification index is used to classify four risk levels: low, medium, high, and emergency.

4. The method for dynamic personnel relocation early warning in flood storage and detention areas according to claim 3, characterized in that, Risk value = α×D + β×P + γ×T; Where D is the standardized flood inundation depth, P is the standardized personnel density, and T is the terrain evacuation difficulty coefficient; α, β, and γ are preset weighting coefficients, and α+β+γ=1.

5. The method for dynamic personnel relocation early warning in flood storage and detention areas according to claim 4, characterized in that, The values ​​of the weighting coefficients α, β, and γ are obtained through machine learning training based on historical disaster data, and / or can be configured and adjusted according to the regional characteristics of different flood storage and detention areas.

6. The method for dynamic personnel relocation early warning in flood storage and detention areas according to claim 1, characterized in that, The step of generating and dynamically updating the optimal evacuation path for threatened individuals based on real-time flooding prediction results and dynamic road condition information specifically includes: Based on the aforementioned flood evolution model, the flood inundation situation is predicted for the next 2 to 6 hours; With a response time of seconds, combined with real-time updated flooding prediction results, traffic status data and infrastructure status information, the optimal evacuation path is calculated for individuals; When a path risk is detected, path replanning is triggered.

7. A dynamic personnel evacuation early warning system for flood storage and detention areas, used to implement the dynamic personnel evacuation early warning method for flood storage and detention areas as described in any one of claims 1-6, characterized in that, The system includes: The multi-source data acquisition module is used to collect and integrate hydrological monitoring data, dynamic distribution information of personnel, and geographic information in real time to build a unified dynamic data base. The dynamic risk assessment module is used to generate a gridded real-time risk level based on the dynamic data base by integrating the flood evolution model and real-time population density data; The tiered early warning decision module is used to dynamically trigger and issue tiered early warning instructions based on the real-time risk level. The intelligent route planning module is used to generate and dynamically update the optimal evacuation route for threatened individuals based on real-time flooding prediction results and dynamic road condition information. The early warning information dissemination module is used to push customized early warning information corresponding to the early warning command and the optimal transfer path to individual terminals.

Images