Small watershed-oriented mountain flood inundation risk dynamic assessment and early warning system

Abstract

This invention belongs to the field of flash flood early warning technology, specifically a dynamic assessment and early warning system for flash flood inundation risk in small watersheds. It includes a small watershed hydrological time-series sensing module, a flash flood causation coupling analysis module, a dynamic inundation range extrapolation module, a risk level stratified assessment module, and a precise early warning information delivery module. This invention involves real-time collection and standardized processing of multi-source hydrological, topographic, and environmental parameters in small watersheds, quantifying the coupling effect of flash flood causation, dynamically extrapolating inundation depth and range based on topographic data, integrating the distribution of disaster-bearing bodies to complete a five-level risk level assessment, and precisely delivering differentiated early warning information to medium-risk and above areas according to their risk levels. This significantly improves the dynamism of risk assessment and the accuracy and timeliness of early warnings, providing reliable support for flash flood prevention and mitigation in small watersheds.

Description

Technical Field

[0001] This invention relates to the field of flash flood early warning technology, specifically to a dynamic assessment and early warning system for flash flood inundation risk in small watersheds. Background Technology

[0002] Small watersheds are characterized by rapid runoff and sudden flash floods, making monitoring and control extremely difficult. Accurate inundation risk assessment and timely early warning are crucial for reducing casualties and property damage from flash floods in mountainous areas. Currently, flash flood early warning technologies in small watersheds are limited to basic monitoring and simple judgment, lacking a comprehensive, dynamic, and precise control system, and exhibiting several technical deficiencies. First, the analysis of flash flood causes is one-sided and fails to quantify the coupling effect. The projection of the inundation range relies on static models and historical data, resulting in large lag and bias. Second, the risk level classification does not take into account the distribution and vulnerability of disaster-bearing bodies. The early warning information is pushed out indiscriminately across the entire region without being differentiated and precise according to risk level. The early warning is seriously lacking in targeting and timeliness, making it difficult to meet the precise early warning needs for disaster prevention and mitigation.

[0003] Therefore, developing a dynamic assessment and early warning system for flash flood inundation risk in small watersheds, capable of coupled analysis of flash flood causes, dynamic projection of inundation range, hierarchical assessment of risk levels, and accurate delivery of early warning information, has become an urgent problem to be solved in the field of flash flood early warning. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic assessment and early warning system for flash flood risk in small watersheds, in order to solve the above-mentioned deficiencies.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a dynamic assessment and early warning system for flash flood inundation risk in small watersheds, including a small watershed hydrological time-series perception module, a flash flood induction coupling analysis module, an inundation range dynamic deduction module, a risk level hierarchical assessment module, and an early warning information accurate push module; The small watershed hydrological time series perception module is responsible for real-time acquisition, time series integration, and preliminary verification of multi-dimensional hydrological, topographic, and environmental parameters related to flash flood formation within the small watershed; the flash flood induction coupling analysis module is used for in-depth analysis of multi-source data, identifying the core inductions of flash flood formation, and quantitatively characterizing the coupling strength between various factors. The dynamic inundation range simulation module, relying on coupling degree and related parameters, combined with small watershed topographic data, dynamically simulates the inundation depth and range of flash floods; the risk level stratification assessment module, based on the simulation data and combined with the distribution data of disaster-bearing bodies in the small watershed, completes the quantitative assessment of flash flood inundation risk; the accurate early warning information push module, based on risk level data, combined with the population, facility distribution and contact information of the affected areas within the small watershed, accurately pushes early warning information.

[0006] Furthermore, the small watershed hydrological time series sensing module deploys sensing nodes in a grid pattern within the small watershed. Each node integrates a rainfall sensor, a water level sensor, a soil moisture sensor, a topographic elevation sensor, and a vegetation cover sensor, and each sensor collects raw data. After data acquisition, the raw data is time-series synchronized by uniformly calibrating the acquisition timestamps of different sensors to Beijing time. Then, abnormal data is verified by setting reasonable thresholds for each parameter, marking data that exceed the thresholds, and interpolating and correcting the data by combining the synchronous data of adjacent sensing nodes to remove invalid data. Finally, the verified valid data is time-series organized in the format of "time-node number-parameter type-value".

[0007] Furthermore, after receiving the time-series dataset transmitted by the small watershed hydrological time-series sensing module, the flash flood cause coupling analysis module normalizes each parameter and maps the values ​​of each parameter to the [0,1] interval to obtain the normalized values ​​of precipitation Pn, topographic slope Sn, soil moisture content Wn, and vegetation coverage Vn. After normalization, a model for calculating the coupling degree of flash flood inducing factors is constructed to quantify the coupling effect of each inducing factor in order to obtain the coupling degree C of flash flood inducing factors. After calculating the coupling degree C, the data information including normalization parameters and the coupling degree C of flash flood inducing factors is transmitted to the dynamic simulation module of inundation range in real time.

[0008] Furthermore, the specific process by which the dynamic inundation range simulation module dynamically simulates the inundation depth and range of flash floods is as follows: Receive the mountain torrent induction coupling degree C and normalization parameters transmitted by the mountain torrent induction coupling analysis module, and simultaneously call the small watershed DEM topographic data, which includes the absolute elevation, topographic slope and valley orientation information of all areas in the small watershed. Determine the critical coupling degree Co for flash flood inundation. Co is the critical threshold for flash flood occurrence in a small watershed. Compare the flash flood induction coupling degree C with the critical coupling degree Co. When C≥Co, start the inundation range extrapolation; when C<Co, determine that there is no risk of flash flood inundation.

[0009] Furthermore, the specific initiation analysis process for the inundation range projection is as follows: After the simulation is started, the inundation depth H of each grid cell in the small watershed is first calculated. Then, the inundation depth calculation model is constructed by combining the coupling degree C, topographic elevation and historical inundation data. After calculating the real-time inundation depth H of each grid cell, the effective inundation area is selected. Combined with the spatial coordinates of the DEM topographic data, a real-time inundation range vector map is drawn, and the inundation depth and predicted inundation duration of each area are marked. After the simulation is completed, the inundation range vector map, the inundation depth data of each grid cell, and the predicted inundation duration data are transmitted to the risk level stratification assessment module in real time.

[0010] Furthermore, the effective flooding area selection method is as follows: Set a flooding depth threshold Hb, and compare the real-time flooding depth H of each grid cell with the flooding depth threshold Hb one by one. If the flooding depth H of the corresponding grid cell is less than Hb, it is judged as a non-flooded area; if the flooding depth H of the corresponding grid cell is greater than or equal to Hb, it is judged as an effective flooded area.

[0011] Furthermore, the risk level stratification assessment module assesses the risk of flash floods through the following specific assessment process: The system receives the inundation range vector map, real-time inundation depth H and inundation duration T of each grid unit from the dynamic inundation range simulation module. At the same time, it calls the built-in small watershed disaster-bearing body distribution database, which contains information on population distribution, housing distribution, farmland distribution and transportation facility distribution within the small watershed. Linear normalization is performed using the preset upper limit threshold of the number / area of ​​each type of disaster-bearing body to obtain the population normalization index U1, housing normalization index U2, farmland normalization index U3 and transportation facility normalization index U4. The normalization index is weighted and summed with the corresponding preset vulnerability coefficient to obtain the comprehensive vulnerability coefficient E of the disaster-bearing body. A risk level index R calculation model was constructed, and the comprehensive vulnerability coefficient E of the disaster-bearing body, the real-time inundation depth H, and the inundation duration T were integrated for analysis and calculation to obtain the flash flood inundation risk level index R of each grid unit; After calculating the flash flood inundation risk level index R for each grid unit, five risk levels are divided based on the flash flood inundation risk level index R, generating a flash flood inundation risk level distribution map for the small watershed. The risk level distribution map, along with the flash flood inundation risk level index R and risk level for each grid unit, are transmitted in real time to the early warning information precision push module.

[0012] Furthermore, the process of classifying risk levels based on the flash flood inundation risk level index R is as follows: The threshold values ​​for the risk level of flash floods, R1, R2, R3, and R4, are retrieved, and R1 > R2 > R3 > R4 > 0. If R ≥ R1, then the risk level of the corresponding grid cell is determined to be extremely high risk; If R2≤R<R1, then the risk level of the corresponding grid cell is determined to be high risk; If R3≤R<R2, then the risk level of the corresponding grid cell is determined to be medium risk; If R4≤R<R3, then the risk level of the corresponding grid cell is determined to be low risk; If R < R4, then the risk level of the corresponding grid cell is determined to be no risk.

[0013] Furthermore, the early warning information precision push module filters risk levels, only pushing early warning information to areas with medium risk or above, while not pushing early warning information to low-risk or no-risk areas.

[0014] Furthermore, the precise early warning information push module determines the early warning level based on the risk level. Extremely high risk corresponds to a red warning, high risk to an orange warning, and medium risk to a yellow warning. Different warning levels correspond to different warning contents, as detailed below: Red Alert: Immediately evacuate affected personnel and prohibit all outdoor activities; Orange alert: Prepare for evacuation and closely monitor flood conditions; Yellow alert: Take precautions and avoid going to dangerous areas.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In this invention, by accurately quantifying the coupling effect of flash flood causes, dynamically inferring inundation depth and range, and combining the distribution of disaster-bearing bodies to complete a five-level risk stratification assessment, the accuracy of cause analysis, the dynamism of inference, and the accuracy of risk assessment are improved, providing core data and assessment support for flash flood early warning.

[0016] 2. In this invention, differentiated early warning information is pushed to medium-risk and above areas in a tiered manner, which accurately covers the affected population and ensures the success rate of the push, reduces invalid early warnings, and greatly improves the timeliness, pertinence and full coverage of early warnings, thereby enhancing the effectiveness of flood prevention and mitigation in small watersheds. Attached Figure Description

[0017] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings; Figure 1 This is a schematic diagram of the overall system structure of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Refer to Figure 1-2 As shown, the dynamic assessment and early warning system for flash flood inundation risk proposed in this invention includes a small watershed hydrological time series perception module, a flash flood induction coupling analysis module, an inundation range dynamic deduction module, a risk level hierarchical assessment module, and an early warning information accurate push module. The small watershed hydrological time series sensing module is used to collect, organize, and preliminarily verify multi-source hydrological, topographic, and environmental parameters related to flash flood formation within the small watershed in real time. This provides high-quality and timely basic data support for subsequent modules and helps avoid assessment errors caused by deviations in basic data.

[0020] Specifically, the small watershed hydrological time series sensing module deploys sensing nodes in a grid pattern within the small watershed. Each node integrates rainfall sensors, water level sensors, soil moisture sensors, topographic elevation sensors, and vegetation cover sensors. The deployment density of sensing nodes is dynamically adjusted according to the area and topographic complexity of the small watershed. For example, the deployment density of small watersheds with an area ≤ 50 km² is no less than 1 node / 5 km², while in areas with complex topography (slope ≥ 25°), the density is increased to 1 node / 3 km². Each sensor collects raw data at a frequency of 10 minutes / time. Among them, the rainfall sensor collects the precipitation during the period, the water level sensor collects the real-time water level of the river and ditch, the soil moisture sensor collects the volume water content of the 0-50cm soil layer, the topographic elevation sensor collects the absolute elevation of the node location, and the vegetation coverage sensor collects the vegetation coverage within 100m around the node. After data collection, the raw data is processed for time synchronization. The timestamps of different sensors are uniformly calibrated to Beijing time to eliminate data misalignment caused by time difference. Then, abnormal data is checked. By setting reasonable thresholds for each parameter (such as the maximum value of a single rainfall collection not exceeding 50 mm and soil moisture content not exceeding 120% of field capacity), data exceeding the thresholds are marked. At the same time, interpolation correction is performed by combining the synchronous data of adjacent sensing nodes to remove invalid data. Finally, the verified valid data is organized into a time series according to the format of "time-node number-parameter type-value" to form a standardized time series dataset, which is then transmitted in real time to the flash flood causation coupling analysis module to provide basic data for subsequent causation analysis.

[0021] The flash flood induction coupling analysis module is used to perform in-depth analysis of multi-source data transmitted by the small watershed hydrological time series sensing module, identify the core inductions of flash flood formation, quantify the coupling effect of each induction and output the coupling degree of the inductions. It overcomes the shortcomings of existing technologies in single-induction analysis and ignoring the interaction between inductions, realizes accurate quantitative analysis of flash flood inductions, and provides core parameter support for the inundation range projection.

[0022] Specifically, after receiving the time-series dataset transmitted by the small watershed hydrological time-series sensing module, the flash flood cause coupling analysis module first normalizes each parameter to eliminate the influence of the difference in the dimensions of different parameters. The normalization adopts the linear normalization method, mapping the values ​​of each parameter to the [0,1] interval to obtain the normalized values ​​of precipitation Pn, topographic slope Sn, soil moisture content Wn, and vegetation coverage Vn. After normalization, the weight coefficients of each inducing factor were determined using the analytic hierarchy process (AHP). Precipitation, topographic slope, soil moisture content, and vegetation cover were selected as primary evaluation indicators. A judgment matrix was constructed, and the weights of each indicator, ω1 (precipitation weight), ω2 (topographic slope weight), ω3 (soil moisture content weight), and ω4 (vegetation cover weight), were determined through consistency testing. Among them, ω1+ω2+ω3+ω4=1. Preferably, considering the characteristics of flash flood formation in small watersheds, the values ​​of ω1 are 0.4-0.5, ω2 is 0.2-0.3, ω3 is 0.15-0.25, and ω4 is 0.05-0.15. Subsequently, a model for calculating the coupling degree of flash flood inducing factors was constructed to quantify the coupling effect of each inducing factor in order to obtain the coupling degree C of flash flood inducing factors. The formula for calculating the coupling degree C of flash flood inducing factors is as follows: ; Wherein, C: the coupling degree of flash flood inducing factors, with a value range of [0,1]. The larger the C value, the stronger the coupling effect of each inducing factor, and the higher the probability of flash floods. In addition, (1-Vn) reflects the negative influence of vegetation cover (the lower the vegetation cover, the higher the risk of flash flood formation). After calculating the coupling degree C, the data information, including the normalized parameters and the coupling degree C of the flash flood induction, is transmitted in real time to the dynamic simulation module of the inundation range, providing core analytical parameters for the dynamic calculation and simulation of the inundation range.

[0023] The dynamic inundation range simulation module, based on the coupling degree and related parameters transmitted by the flash flood causation coupling analysis module and combined with small watershed topographic data, dynamically simulates the inundation depth and range of flash floods. It overcomes the shortcomings of static simulation in existing technologies that ignore real-time changes in causation factors. It can update the inundation simulation results in real time according to the dynamic changes in the causation coupling effect, providing data support for risk level assessment.

[0024] Specifically, the dynamic simulation module for inundation range first receives the coupling degree C and normalized parameters of flash flood induction from the flash flood induction coupling analysis module, and at the same time calls up the DEM topographic data of the small watershed (resolution not less than 30m). This data is imported into the system in advance and includes information such as the absolute elevation, topographic slope, and valley orientation of all areas within the small watershed. Subsequently, the critical coupling degree Co for flash flood inundation was determined. Co is the critical threshold for flash floods in a small watershed. It is obtained by calibration through historical flash flood data. For example, historical cases of flash floods in the small watershed in the past 30 years are selected, the coupling degree of the causes at the time of each case is extracted, and the minimum value is taken as Co. The coupling degree C of flash flood induction is compared with the critical coupling degree Co; When C < Co, it is determined that there is no risk of flash flooding, and the simulation module is in standby mode; When C ≥ Co, the flooding range estimation is initiated, and the specific initiation analysis process is as follows: After initiating the simulation, the inundation depth H of each grid cell within the small watershed is first calculated. Then, combining the coupling degree C, topographic elevation, and historical inundation data, an inundation depth calculation model is constructed, as shown in the following formula: ; Wherein, H: the real-time flooding depth of a certain grid cell, which is the core result of the simulation; C: Real-time flash flood induction coupling degree, transmitted by the flash flood induction coupling analysis module, is the real-time coupling degree value of the corresponding time period for this grid cell; Hh: The historical maximum inundation elevation of this grid unit is obtained from the local hydrological department's historical flash flood inundation database. It corresponds to the highest water level elevation during the historical flash floods in this grid unit to ensure that the data accurately matches the grid unit. He: The real-time topographic elevation of this grid cell is collected by the topographic elevation sensor of the small watershed hydrological time series sensing module, or extracted from the DEM topographic data built into the system, to ensure the real-time performance and accuracy of the elevation data. K: Topographic correction coefficient, determined according to the topographic type of the grid cell. For example, K=1.2 for valley areas (slope ≤ 5°), K=1.0 for gentle slope areas (5° < slope ≤ 25°), and K=0.8 for steep slope areas (slope > 25°). The K value is used to correct the influence of different topographic features on the inundation depth. In valley areas, water flows converge and the inundation depth is greater, while in steep slope areas, the water flow velocity is faster and the inundation depth is relatively smaller. Furthermore, after calculating the real-time flooding depth H of each grid cell, the effective flooding area can be selected, as shown below: Set a flooding depth threshold Hb, and compare the real-time flooding depth H of each grid cell with the flooding depth threshold Hb one by one. If the flooding depth H of the corresponding grid cell is less than Hb, it is judged as a non-flooded area; if the flooding depth H of the corresponding grid cell is greater than or equal to Hb, it is judged as an effective flooded area. By combining the spatial coordinates of DEM topographic data, a real-time inundation range vector map is drawn, and the inundation depth and predicted inundation duration of each area are marked (based on the trend of coupling change). Meanwhile, the updated mountain torrent induction coupling degree C is received in real time from the mountain torrent induction coupling analysis module, and the inundation depth and inundation range projection results are updated every 10 minutes to ensure the dynamic nature of the projection. After the projection is completed, the inundation range vector map, inundation depth data of each grid unit and inundation duration prediction data are transmitted in real time to the risk level stratification assessment module to provide data support for risk level assessment.

[0025] The risk level stratification assessment module assesses the risk of flash floods based on the simulation data transmitted by the dynamic simulation module of the inundation range and the distribution data of disaster-bearing bodies in small watersheds. It overcomes the shortcomings of existing technologies that have a single risk level classification and ignore the differences of disaster-bearing bodies. Based on the degree of inundation and the vulnerability of disaster-bearing bodies, it classifies different risk levels, providing a basis for accurate early warning.

[0026] Specifically, the risk level stratification assessment module first receives the inundation range vector map, the real-time inundation depth H and inundation duration T of each grid unit transmitted by the inundation range dynamic simulation module. At the same time, it calls the built-in small watershed disaster-bearing body distribution database, which contains information on population distribution, housing distribution, farmland distribution and transportation facility distribution within the small watershed. Linear normalization is performed using preset upper limit thresholds for the number / area of ​​each type of disaster-bearing body to eliminate the dimensional differences between population, housing, farmland, and transportation facilities. This yields the population normalization index U1 (calculated by the ratio of the population in the corresponding grid unit to the corresponding upper limit threshold; the calculation methods for U2 and U4 are the same), housing normalization index U2, farmland normalization index U3 (calculated by the ratio of the farmland area in the corresponding grid unit to the corresponding upper limit threshold for farmland area), and transportation facility (including roads and bridges) normalization index U4. The comprehensive vulnerability coefficient E of the disaster-bearing body is obtained by weighted summing the normalized index and the corresponding preset vulnerability coefficient, as shown in the following formula: ; Among them, D1, D2, D3, and D4 are preset vulnerability weight values; E: Comprehensive vulnerability coefficient of the disaster-bearing body of the grid element, with a value of [0,1]; Subsequently, a risk level index R calculation model was constructed, integrating the comprehensive vulnerability coefficient E of the disaster-bearing body, the real-time inundation depth H, and the inundation duration T (predicted by the dynamic inundation range extrapolation module based on the trend of coupling degree changes; a continuous increase in coupling degree extends the prediction time, while a decrease in coupling degree shortens the prediction time) for analysis and calculation, to obtain the flash flood inundation risk level index R for each grid unit, as shown in the following formula: ; Wherein, α: submersion depth weighting coefficient, preferably, α=4.0; β: Weighting coefficient for flooding duration, preferably β=2.0; γ: Vulnerability weighting coefficient of the disaster-bearing body, preferably γ=4.0; After calculating the flash flood inundation risk level index R for each grid unit, the flash flood inundation risk level thresholds R1, R2, R3, and R4 are retrieved, where R1 > R2 > R3 > R4 > 0. Based on the flash flood inundation risk level index R, five risk levels are divided, as shown below: If R ≥ R1, then the risk level of the corresponding grid cell is determined to be extremely high risk; If R2≤R<R1, then the risk level of the corresponding grid cell is determined to be high risk; If R3≤R<R2, then the risk level of the corresponding grid cell is determined to be medium risk; If R4≤R<R3, then the risk level of the corresponding grid cell is determined to be low risk; If R < R4, then the risk level of the corresponding grid cell is determined to be no risk.

[0027] Finally, based on the risk level information of each grid unit, the risk level of each grid unit is overlaid with the inundation range and the distribution data of the disaster-bearing bodies to generate a small watershed flash flood inundation risk level distribution map, marking the core disaster-bearing bodies and risk hazard points in each risk level area.

[0028] Example 2: Refer to Figure 1-2 As shown, the difference between this embodiment and Embodiment 1 is that the risk level stratification assessment module is connected to the early warning information precision push module. The risk level stratification assessment module transmits the risk level distribution map and the flash flood inundation risk level index R and risk level of each grid unit to the early warning information precision push module in real time. Based on the risk level data, the early warning information precision push module combines the population, facility distribution and contact information of the affected area in the small watershed to accurately push early warning information, avoiding redundancy or omission of early warning information and improving the timeliness and pertinence of early warning.

[0029] Specifically, after receiving the risk level distribution map, risk index and risk level data of each grid unit transmitted by the risk level hierarchical assessment module, the early warning information precise push module first filters the risk level and only pushes early warning information to areas with medium risk and above, while not pushing early warning information to low-risk and no-risk areas, thereby reducing invalid pushes.

[0030] Subsequently, the warning level is determined based on the risk level, with the following correspondence: extremely high risk corresponds to a red warning, high risk corresponds to an orange warning, and medium risk corresponds to a yellow warning. Different warning levels correspond to different warning contents, as detailed below: Red Alert: Immediately evacuate affected personnel and prohibit all outdoor activities; Orange alert: Prepare for evacuation and closely monitor flood conditions; Yellow alert: Take precautions and avoid going to dangerous areas.

[0031] Furthermore, by combining the system's built-in disaster-bearing body distribution database, the system extracts the contact information (mobile phone numbers, village / community notification channels) of the population in medium-risk and above areas, as well as the contact information of enterprises and units, and organizes them by grid unit to ensure that every affected person and unit can be accurately covered.

[0032] Simultaneously, early warning information is automatically generated, including the warning level, affected area, flood risk, preventative measures, and evacuation routes (the optimal evacuation route is automatically planned based on the small watershed topography and transportation infrastructure distribution). Warning information is available in both text and voice formats, tailored to different groups (voice warnings are prioritized for the elderly). After the message is sent, the system provides real-time feedback on the sending status (sent, not sent, failed to send). For those whose messages failed to be sent, a secondary notification is sent through village / community channels to ensure full coverage of the warning information.

[0033] The working principle of this invention is as follows: When in use, by collecting and standardizing multi-source hydrological, topographic, and environmental parameters of small watersheds, the coupling effect of flash flood causes is quantified. Combined with topographic data, the inundation depth and range are dynamically inferred. The distribution of disaster-bearing bodies is integrated to classify five levels of risk. Differentiated early warnings are only pushed to areas with medium risk and above. This solves the problems of one-sided cause analysis, static inundation inference, and insufficient targeting of early warnings in small watershed flash flood warnings. It realizes dynamic updates of inundation range, accurate risk assessment, and efficient and non-redundant early warning push, which greatly enhances the response and handling capabilities of flash flood prevention and mitigation in small watersheds.

[0034] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, enabling those skilled in the art to better understand and utilize it. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A dynamic assessment and early warning system for flash flood risk in small watersheds, characterized in that, It includes a small watershed hydrological time-series sensing module, a flash flood cause coupling analysis module, an inundation range dynamic simulation module, a risk level stratified assessment module, and an early warning information precise push module; The small watershed hydrological time series perception module is responsible for real-time acquisition, time series integration, and preliminary verification of multi-dimensional hydrological, topographic, and environmental parameters related to flash flood formation within the small watershed; the flash flood induction coupling analysis module is used for in-depth analysis of multi-source data, identifying the core inductions of flash flood formation, and quantitatively characterizing the coupling strength between various factors. The dynamic inundation range simulation module, relying on coupling degree and related parameters, combined with small watershed topographic data, dynamically simulates the inundation depth and range of flash floods; the risk level stratification assessment module, based on the simulation data and combined with the distribution data of disaster-bearing bodies in the small watershed, completes the quantitative assessment of flash flood inundation risk; the accurate early warning information push module, based on risk level data, combined with the population, facility distribution and contact information of the affected areas within the small watershed, accurately pushes early warning information.

2. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 1, characterized in that, The small watershed hydrological time series sensing module deploys sensing nodes in a grid pattern within the small watershed, and each sensor collects raw data. After the data collection is completed, the raw data undergoes time series synchronization processing and abnormal data verification. Finally, the verified valid data is organized into a time series in the format of "time-node number-parameter type-value" to form a standardized time series dataset, which is then transmitted in real time to the flash flood causation coupling analysis module.

3. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 2, characterized in that, After receiving the time-series dataset, the flash flood causation coupling analysis module normalizes each parameter. After the normalization is completed, it constructs a flash flood causation coupling degree calculation model and quantifies the coupling effect of each causation to obtain the flash flood causation coupling degree C.

4. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 1, characterized in that, The specific process by which the dynamic inundation range simulation module dynamically simulates the inundation depth and range of flash floods is as follows: The system receives the coupling degree C of flash flood induction and the normalized parameters, and simultaneously calls the DEM topographic data of the small watershed to determine the critical coupling degree Co of flash flood inundation. The coupling degree C of flash flood induction is compared with the critical coupling degree Co. When C≥Co, the inundation range is extrapolated; when C<Co, it is determined that there is no risk of flash flood inundation.

5. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 4, characterized in that, The specific initiation analysis process for the flood range projection is as follows: After the simulation is started, the inundation depth H of each grid cell in the small watershed is calculated. After the real-time inundation depth H of each grid cell is calculated, the effective inundation area is selected. Combined with the spatial coordinates of the DEM topographic data, a real-time inundation range vector map is drawn, and the inundation depth and predicted inundation duration of each area are marked.

6. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 5, characterized in that, The selection method for effective flooding areas is as follows: Set a flooding depth threshold Hb. If the flooding depth H of the corresponding grid cell is less than Hb, it is determined to be a non-flooded area; if the flooding depth H of the corresponding grid cell is greater than or equal to Hb, it is determined to be an effective flooded area.

7. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 1, characterized in that, The risk level stratification assessment module assesses the risk of flash floods as follows: A risk level index R calculation model was constructed, and the comprehensive vulnerability coefficient E of the disaster-bearing body, the real-time inundation depth H, and the inundation duration T were integrated for analysis and calculation to obtain the flash flood inundation risk level index R of each grid unit; Based on the flash flood inundation risk level index R, five risk levels are divided, and a flash flood inundation risk level distribution map of a small watershed is generated. The risk level distribution map, as well as the flash flood inundation risk level index R and risk level of each grid unit, are transmitted in real time to the early warning information precision push module.

8. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 7, characterized in that, The specific process for classifying risk levels based on the flash flood inundation risk level index R is as follows: The threshold values ​​for the risk level of flash floods, R1, R2, R3, and R4, are retrieved, and R1 > R2 > R3 > R4 > 0. If R ≥ R1, the risk level of the corresponding grid cell is determined to be extremely high; if R2 ≤ R < R1, the risk level of the corresponding grid cell is determined to be high; if R3 ≤ R < R2, the risk level of the corresponding grid cell is determined to be medium; if R4 ≤ R < R3, the risk level of the corresponding grid cell is determined to be low; if R < R4, the risk level of the corresponding grid cell is determined to be no risk.

9. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 7, characterized in that, The early warning information precision push module filters risk levels and only pushes early warning information to areas with medium risk or above, while not pushing early warning information to low-risk or no-risk areas.

10. The dynamic assessment and early warning system for flash flood risk in small watersheds according to claim 9, characterized in that, The precise early warning information push module determines the early warning level based on the risk level. Extremely high risk corresponds to a red warning, high risk to an orange warning, and medium risk to a yellow warning. Different warning levels correspond to different warning contents, as detailed below: Red Alert: Immediately evacuate affected personnel and prohibit all outdoor activities; Orange Alert: Prepare for evacuation and closely monitor flood conditions; Yellow Alert: Strengthen precautions and avoid going to dangerous areas.

Images