Reservoir downstream flooding development situation deduction method based on hydraulic model

Abstract

The invention discloses a reservoir downstream flooding development situation deduction method based on a hydraulic model, and belongs to the technical field of inference models. The problem that in the prior art, a traditional reservoir inundation development situation deduction method is low in prediction precision is solved. The method comprises the following steps of: acquiring reservoir downstream flooding related data, and performing data preprocessing on the reservoir downstream flooding related data to obtain a reservoir downstream flooding deduction multi-source data set; selecting and coupling a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model, and constructing a reservoir downstream flooding development situation deduction hydraulic model; setting a multi-scene inundation situation deduction scheme, and outputting a dynamic evolution animation of space-time sequence data and a multi-scene deduction result; defining inundation risk variable factors, dividing risk levels through risk values, completing reservoir downstream inundation risk level evaluation and constructing an inundation risk database; and reservoir downstream flooding development situation deduction is completed. The method effectively improves the efficiency of reservoir inundation prediction and risk management, and can be applied to flood prevention planning.

Description

TECHNICAL FIELD

[0001] The present application relates to a reservoir downstream inundation development trend deduction method, in particular to a reservoir downstream inundation development trend deduction method based on a hydraulic model, and belongs to the technical field of reasoning models. BACKGROUND

[0002] As a key project for regulating water resources and ensuring flood control safety, reservoirs are located in areas with dense population and concentrated economic activities, and the inundation risk is directly related to people's life and property safety and social stability. Climate change exacerbates the frequency of extreme flood events, and the risk of reservoir dam break and over-standard flood rises significantly. Therefore, accurate prediction and dynamic deduction of downstream inundation situation have become the core demand of water conservancy safety management.

[0003] The current inundation risk assessment has obvious limitations as follows: the traditional hydrological analysis relies on empirical formula or single model, and it is difficult to comprehensively consider topography, hydrology, engineering regulation and other factors; one-dimensional and two-dimensional models are independently applied, which has problems such as mismatch of time and space scale, simplification of boundary conditions, etc., resulting in insufficient prediction accuracy of inundation range, water depth and arrival time; the fusion ability of multi-source heterogeneous data is weak, which restricts the comprehensiveness of scenario simulation; the deduction results are mainly static reports, which lack dynamic visualization and multi-scenario comparison functions, and cannot meet the needs of emergency decision-making.

[0004] The existing application of hydraulic model has the following technical bottlenecks: one-dimensional model is difficult to simulate the inundation process of complex terrain, two-dimensional model has low calculation efficiency and imperfect coupling mechanism with river flood evolution; the scenario setting is limited to single return period flood, and the extreme scenario is not considered; the output results have low integration degree with risk assessment and flood avoidance planning, and the result transformation efficiency is low.

[0005] In summary, under the background of limited emergency resources and urgent decision-making time for reservoir operation and maintenance units, it is urgent to establish a technical method that integrates multi-source data, couples multi-scale models and supports multi-scenario dynamic deduction, that is, a reservoir downstream inundation development trend deduction method based on a hydraulic model is needed to improve the scientificity and accuracy of flood risk control. SUMMARY

[0006] In the following, a brief summary of the present application is given in order to provide a basic understanding of some aspects of the present application. It is understood that this summary is not a comprehensive summary of the present application. It is not intended to determine the key or important parts of the present application, nor to limit the scope of the present application. Its purpose is only to give some concepts in a simplified form as a prelude to the more detailed description discussed later.

[0007] In view of this, in order to solve the problem of low prediction accuracy of the traditional reservoir inundation development trend deduction method in the prior art, the present application provides a reservoir downstream inundation development trend deduction method based on a hydraulic model.

[0008] The technical scheme is as follows: a reservoir downstream inundation development trend deduction method based on a hydraulic model, comprising the following steps:

[0009] S1. Collecting reservoir downstream inundation related data to form reservoir downstream inundation deduction multi-source data, and performing data preprocessing on the data through coordinate unification, terrain correction and data standardization to obtain a reservoir downstream inundation deduction multi-source data set;

[0010] S2. Selecting and coupling a one-dimensional river channel hydrodynamic model and a two-dimensional hydrodynamic model, setting coupling nodes and time steps, defining control equations of the models, performing parameter calibration through zoned roughness coefficient assignment and setting upstream and downstream boundary conditions, and constructing a reservoir downstream inundation development trend deduction hydraulic model;

[0011] S3. Based on the reservoir downstream inundation deduction multi-source data set and the reservoir downstream inundation development trend deduction hydraulic model, setting multi-scenario inundation trend deduction schemes, calculating inundation ranges, water depth distributions and arrival times under different scenarios, and outputting spatiotemporal sequence data and dynamic evolution animations of multi-scenario deduction results;

[0012] S4. Defining inundation risk variable factors, dividing risk levels through risk values, and completing reservoir downstream inundation risk level assessment and constructing an inundation risk database according to the multi-scenario deduction results of step S3;

[0013] S5. Based on the inundation risk database constructed in step S4, integrating dynamic inundation process three-dimensional visualization, risk level thematic map automatic rendering and flood diversion route planning functions, and completing reservoir downstream inundation development trend deduction.

[0014] Further, in the S1, the following steps are included:

[0015] S11. Collecting terrain data, hydrological data, social and economic data and real-time monitoring data, i.e., reservoir downstream inundation related data, to form reservoir downstream inundation deduction multi-source data;

[0016] In the S11, the terrain data includes digital elevation model, contour line data and administrative division vector map;

[0017] The hydrological data includes historical flood hydrograph, river cross-section data and reservoir dispatching rules;

[0018] The social and economic data includes population distribution, housing construction and infrastructure data;

[0019] The real-time monitoring data includes reservoir water level, river flow and rainfall station data;

[0020] S12. Extracting required information from the reservoir downstream inundation deduction multi-source data respectively to construct respective databases;

[0021] In step S12, elevation values, resolution, and projection information fields are extracted from the digital elevation model to construct a terrain attribute database in .mdb format;

[0022] Extract the administrative code and area fields from the administrative division vector map to generate administrative division attribute data in .shp format;

[0023] Extract flood discharge, water level, and occurrence time fields from historical flood data obtained based on historical flood hydrographs, and construct a hydrological time series database in .xlsx format;

[0024] The scheduling curves and flood discharge capacity parameters are extracted from the reservoir scheduling rules to form a scheduling parameter attribute table;

[0025] Extract regional population size and density fields from population distribution data to construct a population attribute database;

[0026] Extract building type, number of floors, and coordinate fields from building vector data to generate a socioeconomic element layer in .shp format;

[0027] The monitoring time, water level, and flow fields are extracted from sensor data of reservoir water level, river flow, and rain gauge stations. A real-time monitoring database in .csv format is constructed and associated with the monitoring station ID and spatial coordinates.

[0028] S13. Perform data preprocessing on the respective databases generated in step S12 to obtain a multi-source dataset for downstream inundation simulation of the reservoir.

[0029] In step S13, the terrain attribute database, hydrological time series database, socio-economic element layer and real-time monitoring database from step S12 are imported into the geographic information system platform software. At the same time, high-resolution remote sensing image data in GeoTIFF format is loaded and subjected to coordinate unification, terrain correction and data standardization.

[0030] Coordinate unification: The national 2000 geodetic coordinate system is used to transform the coordinates of all spatial data to ensure consistent spatial reference for topographic, hydrological, and socio-economic data;

[0031] Terrain correction: Using the 3D Analyst tool in the geographic information system platform software, outliers in the digital elevation model data are removed to generate a smooth terrain surface;

[0032] Data standardization: Hydrological data is converted to a uniform time step, socioeconomic data is aggregated according to a set grid unit, and real-time monitoring data format is converted to GeoJSON;

[0033] Finally, the integrated multi-source data was published using ArcGIS Server to provide map services, forming a standardized multi-source dataset for downstream inundation simulation of reservoirs.

[0034] Furthermore, step S2 includes the following steps:

[0035] S21. Based on the coupling mechanism of wet and dry boundaries, a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model are selected for coupling. That is, the river flood process calculated by the one-dimensional model is used as the upstream inflow boundary of the two-dimensional model to realize the dynamic connection between the river and the inundation zone flow.

[0036] S22. Set up coupling nodes at key sections at the junction of the river channel and the inundation area, extract water level / discharge data from the one-dimensional model (i.e., the one-dimensional river hydrodynamic model) as the boundary input of the two-dimensional model (i.e., the two-dimensional hydrodynamic model), and simultaneously collect feedback on the inundation range calculated by the two-dimensional model to correct the downstream boundary conditions of the one-dimensional model.

[0037] In S22, an adaptive time step algorithm is adopted. The time step of the one-dimensional model is controlled within a set range according to the Courant number, and the time step of the two-dimensional model is dynamically adjusted based on the minimum grid size and water flow velocity to ensure the stability of coupled calculation. The time steps of the two models are synchronized by interpolation algorithm to achieve data exchange.

[0038] S23. Based on steps S21 and S22, mathematical expressions are given for the one-dimensional river hydrodynamic model and the two-dimensional hydrodynamic model.

[0039] In S23, the mathematical expression of the one-dimensional river hydrodynamic model is as follows:

[0040]

[0041]

[0042] Where B is the width of the water surface, Q is the flow rate, Z is the water level, q is the lateral flow rate, and t is the time. Where A is the distance along the direction of water flow, g is the gravitational acceleration, R is the hydraulic radius, and C is the Chezy coefficient.

[0043] The mathematical expression of a two-dimensional hydrodynamic model includes the continuity equation and the momentum equation;

[0044] The continuity equations of the two-dimensional hydrodynamic model are expressed as follows:

[0045]

[0046] The momentum equation for the two-dimensional hydrodynamic model is expressed as:

[0047]

[0048] Where H is the water depth and Z is the water level. For the source and sink terms in the continuity equation, M and N are the vertical average unit width flow rates in the x and y directions, respectively, u and v are the components of the vertical average velocity in the x and y directions, n is the Manning roughness coefficient, and g is the gravitational acceleration.

[0049] S24. Assign values ​​to the roughness coefficient n based on land use type, and use ArcGIS spatial analysis tools to convert the assigned values ​​of the roughness coefficient into grid data of a two-dimensional model;

[0050] S25. The reservoir discharge flow process line or water level-discharge relationship curve is used as the inlet boundary of the one-dimensional model, and the river outlet water level process line or zero gradient condition is set as the outlet boundary of the one-dimensional model. The downstream boundary of the two-dimensional model adopts sponge boundary treatment to allow floodwater to flow freely out of the computational domain. The measured water level and discharge data of historical flood events are selected, and the model is calibrated by adjusting the roughness coefficient and the river roughness correction coefficient to ensure that the Nash efficiency coefficient of the calculated value and the measured value is ≥0.85, so as to ensure that the accuracy of the two models meets the simulation requirements.

[0051] S26. Integrate the above parameters and model equations, realize data interaction between the one-dimensional model and the two-dimensional model through the model interface, construct a hydraulic model for the inundation development trend of the downstream reservoir, verify it using typical flood event data, output the simulation results of the inundation range and water depth distribution at different times, and compare it with the actual inundation range monitored by remote sensing, control the error within 10%, and complete the reliability verification of the hydraulic model for the inundation development trend of the downstream reservoir.

[0052] Furthermore, step S3 includes the following steps:

[0053] S31. Based on the preprocessed data from step S13 and the downstream inundation development trend of the reservoir, a hydraulic model is used to deduce three major scenarios: design flood, reservoir dam failure, and extreme precipitation.

[0054] In S31, the flood process line of a typical scenario is designed for flood simulation input, and the flood process and normal reservoir operation rules are determined by hydrological frequency analysis.

[0055] The reservoir dam failure simulation scenario sets the location of the breach in the middle of the dam or on the left bank, and the failure mode is divided into instantaneous total failure and gradual failure.

[0056] Extreme precipitation scenarios are coupled with meteorological data, and uniform or peak-shaped rainfall patterns are input to design rainfall patterns.

[0057] S32. Use the hydraulic model to calculate parameters for three major scenarios of downstream inundation development of the reservoir;

[0058] In step S32, the inundation range is determined by the dry and wet grids of the two-dimensional model, and grid cells with water depths greater than a set value are extracted. Vector boundaries are generated through GIS spatial analysis. The water depth distribution is processed by inverse distance weighted interpolation to process the water depth values ​​of the grid cells and generate a plane contour map. The arrival time is recorded as the moment when the water depth of each grid first exceeds the set value. Combined with the water flow velocity field data for correction, a spatial distribution map of the inundation arrival time is generated. During the calculation process, the influence of flow velocity changes on the arrival time is corrected synchronously to ensure time accuracy.

[0059] S33. Based on the scenario parameters in step S32, output standardized spatiotemporal sequence data from the hydraulic model derived from the downstream inundation development trend of the reservoir.

[0060] In step S33, the water level-flow process line of key sections is extracted from the time series and stored in .csv format, which includes scenario ID, time, section ID, water level, and flow field;

[0061] The spatial sequence outputs the inundation range in .shp format, the water depth in .tif format, and the arrival time raster data in .tif format according to the set time step, and correlates the scenario parameters with the inundation development trend of the downstream reservoir and the metadata of the hydraulic model.

[0062] Based on time series and spatial series, a spatiotemporal database containing standardized spatiotemporal sequence data is constructed using PostgreSQL+PostGIS database. Scenario information table and spatiotemporal data table are designed to realize unified storage and query of multi-scenario data.

[0063] S34. Based on standardized spatiotemporal sequence data, create dynamic animations of the inundation situation projection results output by the hydraulic model for the downstream inundation development trend projection of the reservoir.

[0064] In S34, the preprocessing stage aligns multi-step data according to the time axis, uses a hierarchical color scheme to map water depth, represents arrival time with grayscale gradient, generates single-frame images using Python software or ArcGIS Pro, synthesizes MP4 videos, and overlays administrative divisions, important facility labels, and time axis controls.

[0065] S35. Select measured data from historical flood events to verify the inundation situation projection results, ensuring that the mean absolute error is greater than or equal to the set value and the Nash efficiency coefficient is greater than or equal to the set value. By comparing and analyzing the risk differences of different scenarios through the dynamic curve of scenario-inundated area and the heat map of maximum water depth-population density, we can help identify high-risk areas and provide data support for emergency decision-making.

[0066] Furthermore, step S4 includes the following steps:

[0067] S41. Based on the flooding situation simulation results and the characteristics of the disaster-bearing body, two major categories of risk variables, namely disaster-causing factors and disaster-bearing body vulnerability factors, are selected and standardized.

[0068] In S41, the disaster-causing factors include flooding depth, flooding flow velocity, and flooding duration;

[0069] Vulnerability factors of disaster-bearing bodies include population density, economic density, and infrastructure exposure.

[0070] The flooding risk variable factor is standardized to the [0,1] interval by min-max, that is, standardized value = (actual value - minimum value) / (maximum value - minimum value).

[0071] S42. A two-dimensional risk matrix framework of probability-consequence is adopted. The risk matrix is ​​constructed by combining the characteristics of the inundation scenario with the quantified risk value R of the inundation risk variable factors. The probability L is determined based on the return period of the flood scenario, and the consequence C is calculated by coupling the disaster-causing factors and the vulnerability factors of the disaster-bearing body.

[0072] In S42, consequence C is represented as:

[0073]

[0074] in, The water depth in the flooded area, For the flow rate of the flood, For the duration of flooding, For population density, For economic exposure, Land use type;

[0075] The risk value R is represented as:

[0076] R = L × C

[0077] S43. Based on the risk value R and the actual disaster bearing capacity of the region, the flooding risk is divided into 5 levels. The level division results are generated by GIS spatial overlay to generate a risk level distribution map, and each level is rendered with different colors.

[0078] In S43, the flooding risk levels are as follows: 1) Low risk: R < 0.2; 2) Medium risk: 0.2 ≤ R < 0.4; 3) Higher risk: 0.4 ≤ R < 0.6; 4) High risk: 0.6 ≤ R < 0.8; 5) Extremely high risk: R ≥ 0.8;

[0079] S44. Integrate the results of the multi-scenario simulation in step S3 with the inundation risk variable factors to conduct a regional inundation risk level assessment, which includes single-scenario assessment and multi-scenario comprehensive assessment.

[0080] In S44, the single-scenario assessment process is as follows: 1) Extract the maximum water depth, flow velocity, and duration data of each grid cell output in step S3; 2) Overlay spatial layers of population density, economic density, and infrastructure exposure; 3) Calculate the consequence value C using the formula in step S42, and obtain the risk value R by combining the scenario probability L; 4) Divide the grid cell risk level according to the threshold in step S43.

[0081] The multi-scenario comprehensive assessment process is as follows: the risk levels of the design flood, dam break and extreme precipitation scenarios are weighted and averaged to eliminate single scenario bias. The assessment results are verified by historical disaster data to ensure that the spatial overlap rate between high-risk areas and actual severely affected areas is greater than or equal to the set value, and the average risk level error is less than or equal to the set value.

[0082] S45. Based on the assessment results of the regional inundation risk level assessment, construct an inundation risk database to achieve systematic storage and management of risk information;

[0083] In step S45, the database design adopts a three-level architecture of scenario-region-risk: 1) The scenario layer stores metadata such as scenario ID, type, and recurrence period; 2) The region layer includes administrative division code, geographical boundary and disaster-bearing body attributes; 3) The risk layer records the risk value, level and key disaster-causing factors of each region under different scenarios. The risk layer supports multi-dimensional queries and is linked with the standardized spatiotemporal sequence data in step S3 to realize a closed-loop data process of inference-assessment-query.

[0084] Furthermore, in step S5, a flood risk database is constructed based on the assessment results built in step S4. When developing the flood situation simulation system, the PostgreSQL+PostGIS risk database is used as the data core. By integrating three-dimensional terrain data with the dynamic flood spatiotemporal sequence generated in step S3, an engine is used to realize the three-dimensional visualization of the dynamic flood process. The risk level data in the risk database is overlaid, and the risk level thematic map is automatically rendered with different colors. The administrative divisions, important facility markings, and key disaster-causing factors are displayed simultaneously. For the flood evacuation route planning function, the system calls the spatial distribution data of high-risk areas in the risk database, couples the road network topology and refuge site capacity information, and generates the optimal flood evacuation route based on the improved A* algorithm or Dijkstra algorithm, while avoiding high-risk road sections with flood depth > 0.5m and flow velocity > 1m / s. The route planning results need to simultaneously display the estimated evacuation time, the risk level of the route, and the real-time capacity of the refuge site. It also supports batch planning with multiple starting points and multiple ending points and export of route feasibility assessment reports, ultimately realizing the integrated function of dynamic simulation, risk visualization, and route planning.

[0085] The beneficial effects of this invention are as follows: The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, as described in this invention, improves the simulation accuracy of the inundation range to over 90% based on multi-source data fusion and a dynamic coupling model, and controls the water depth error to ≤0.3m, achieving high-precision simulation. Simultaneously, by employing parallel computing technology, the simulation time for a 50-year flood is significantly reduced from 8 hours using traditional methods to 40 minutes, achieving efficient computation. Furthermore, by predicting the inundation process in real time, it provides a basis for optimizing evacuation routes and allocating resources for flood control command, forming dynamic decision support. Finally, by leveraging a risk matrix and visualization system, it achieves a closed-loop process of data-model-decision, intuitively expressing risks and effectively improving risk management efficiency. Attached Figure Description

[0086] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0087] Figure 1 This is a flowchart illustrating a method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model. Detailed Implementation

[0088] To make the technical solutions and advantages of the embodiments of the present invention clearer, the exemplary embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0089] refer to Figure 1 This embodiment describes a method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, specifically including the following steps:

[0090] S1. Collect relevant data on downstream inundation of the reservoir to form multi-source data for downstream inundation projection. Preprocess the data through coordinate unification, terrain correction and data standardization to obtain multi-source dataset for downstream inundation projection.

[0091] S2. Select and couple a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model, set the coupling nodes and time steps, clarify the control equations of the model, calibrate the parameters by assigning roughness coefficients to partitions and setting upstream and downstream boundary conditions, and construct a hydraulic model for inferring the inundation development trend of the downstream reservoir.

[0092] S3. Based on the multi-source dataset of downstream reservoir inundation simulation and the hydraulic model of downstream reservoir inundation development trend simulation, set up a multi-scenario inundation trend simulation scheme, calculate the inundation range, water depth distribution and arrival time under different scenarios, and output the spatiotemporal sequence data and dynamic evolution animation of multi-scenario simulation results.

[0093] S4. Define inundation risk variable factors, classify risk levels by risk values, and complete the inundation risk level assessment of the downstream reservoir and construct an inundation risk database based on the multi-scenario simulation results of step S3.

[0094] S5. Based on the inundation risk database constructed in step S4, integrate the functions of dynamic inundation process 3D visualization, automatic rendering of risk level thematic maps and flood evacuation route planning to complete the simulation of the inundation development trend downstream of the reservoir.

[0095] Furthermore, step S1 includes the following steps:

[0096] S11. Collect topographic data, hydrological data, socio-economic data, and real-time monitoring data, i.e., data related to downstream inundation of the reservoir, to form multi-source data for downstream inundation projection of the reservoir;

[0097] In S11, the terrain data includes digital elevation model (DEM), contour line data, and administrative division vector map, which are derived from publicly available materials from regional surveying and mapping geographic information bureaus or natural resources surveying institutes.

[0098] Hydrological data includes historical flood hydrographs, river cross-section data, and reservoir operation rules, which are derived from publicly available materials from river basin hydrological bureaus, meteorological bureaus, or water conservancy project management units.

[0099] Socioeconomic data includes population distribution, housing construction, and infrastructure data, which are derived from publicly available materials from local statistical bureaus, urban planning departments, or government data platforms.

[0100] Real-time monitoring data includes reservoir water levels, river flow, and rain gauge data, which are derived from publicly available materials from water conservancy project operation and maintenance units or Internet of Things (IoT) monitoring systems.

[0101] S12. Extract the required information from the multi-source data of downstream inundation simulation of the reservoir and construct their respective databases;

[0102] In step S12, elevation values, resolution, and projection information fields are extracted from the digital elevation model to construct a terrain attribute database in .mdb format;

[0103] Extract the administrative code and area fields from the administrative division vector map to generate administrative division attribute data in .shp format;

[0104] Extract flood discharge, water level, and occurrence time fields from historical flood data obtained based on historical flood hydrographs, and construct a hydrological time series database in .xlsx format;

[0105] The scheduling curves and flood discharge capacity parameters are extracted from the reservoir scheduling rules to form a scheduling parameter attribute table;

[0106] Extract regional population size and density fields from population distribution data to construct a population attribute database;

[0107] Extract building type, number of floors, and coordinate fields from building vector data to generate a socioeconomic element layer in .shp format;

[0108] The monitoring time, water level, and flow fields are extracted from sensor data of reservoir water level, river flow, and rain gauge stations. A real-time monitoring database in .csv format is constructed and associated with the monitoring station ID and spatial coordinates.

[0109] S13. Perform data preprocessing on the respective databases generated in step S12 to obtain a multi-source dataset for downstream inundation simulation of the reservoir.

[0110] In step S13, the terrain attribute database, hydrological time series database, socio-economic element layer and real-time monitoring database from step S12 are imported into the geographic information system platform software (Esri ArcGIS), and high-resolution remote sensing image data in GeoTIFF format is loaded, and coordinate unification, terrain correction and data standardization are performed on them.

[0111] Coordinate unification: The National Geodetic Coordinate System 2000 (CGCS2000) is used to transform the coordinates of all spatial data to ensure consistent spatial reference for topographic, hydrological, and socio-economic data;

[0112] Terrain correction: Using the 3D Analyst tool in the geographic information system platform software, outliers in the digital elevation model data are removed to generate a smooth terrain surface;

[0113] Data standardization: Hydrological data is converted to a uniform time step, socioeconomic data is aggregated according to a set grid unit of 100m×100m, and real-time monitoring data format is converted to GeoJSON to support dynamic updates;

[0114] Finally, the integrated multi-source data was published using ArcGIS Server to provide map services, forming a standardized multi-source dataset for downstream inundation simulation of reservoirs.

[0115] Specifically, the multi-source data requirements for downstream inundation projection of the reservoir are shown in Table 1:

[0116] Table 1. Multi-source data requirements for downstream inundation simulation of the reservoir

[0117]

[0118] Furthermore, step S2 includes the following steps:

[0119] S21. Based on the coupling mechanism of wet and dry boundaries, a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model are selected for coupling. That is, the river flood process calculated by the one-dimensional model is used as the upstream inflow boundary of the two-dimensional model to realize the dynamic connection between the river and the inundation zone flow.

[0120] Specifically, the one-dimensional model is constructed using the Saint-Venant equations, which are used to simulate the evolution of floods in the main channel of the river. It has the advantages of high computational efficiency and easy setting of boundary conditions. The two-dimensional model is constructed using the shallow water equations, which are used to simulate the planar inundation diffusion after flood overflow, and can accurately express the water depth distribution under complex terrain.

[0121] S22. Set up coupling nodes at key sections at the junction of the river channel and the inundation area, extract water level / discharge data from the one-dimensional model (i.e., the one-dimensional river hydrodynamic model) as the boundary input of the two-dimensional model (i.e., the two-dimensional hydrodynamic model), and simultaneously collect feedback on the inundation range calculated by the two-dimensional model to correct the downstream boundary conditions of the one-dimensional model.

[0122] In S22, an adaptive time step algorithm is adopted. The time step of the one-dimensional model is controlled within a set range (0.5-1.0 s) according to the Courant number (CFL). The time step of the two-dimensional model is dynamically adjusted based on the minimum grid size and water flow velocity to ensure the stability of the coupled calculation. The time steps of the two models are synchronized by interpolation algorithm to achieve data exchange.

[0123] S23. Based on steps S21 and S22, mathematical expressions are given for the one-dimensional river hydrodynamic model and the two-dimensional hydrodynamic model.

[0124] In S23, the mathematical expression of the one-dimensional river hydrodynamic model is as follows:

[0125]

[0126]

[0127] Where B is the width of the water surface (unit: mm), and Q is the flow rate (unit: ...). / s), Z: water level (unit: mm), q: lateral flow rate (unit: ... / s), where t is time (unit: s). A is the distance along the direction of water flow (unit: m), and A is the cross-sectional area of ​​the water passage (unit: ...). g is the acceleration due to gravity (taken as 9.81). R is the hydraulic radius (unit: mm), and C is the Chezy coefficient (unit: mm). );

[0128] The mathematical expression of a two-dimensional hydrodynamic model includes the continuity equation and the momentum equation;

[0129] The continuity equations of the two-dimensional hydrodynamic model are expressed as follows:

[0130]

[0131] The momentum equation for the two-dimensional hydrodynamic model is expressed as:

[0132]

[0133]

[0134] Where H is the water depth and Z is the water level. For the source and sink terms in the continuity equation, M and N are the vertical average unit width flow rates in the x and y directions, respectively, u and v are the components of the vertical average velocity in the x and y directions, n is the Manning roughness coefficient, and g is the gravitational acceleration.

[0135] S24. Based on land use type, assign roughness coefficient n to different areas. Use ArcGIS spatial analysis tools to convert the roughness coefficient assignment results into grid data of a two-dimensional model to realize spatial differentiation parameter input.

[0136] Specifically, for the main channel, the roughness coefficient n = 0.025-0.035 (0.025 for sandy riverbeds and 0.035 for pebble riverbeds).

[0137] For beach land / cultivated land, the roughness coefficient n = 0.035-0.050 (take the higher value when there is vegetation cover).

[0138] For urban areas, the roughness coefficient n = 0.040-0.060 (0.060 for densely built-up areas).

[0139] S25. The reservoir discharge flow process line or water level-discharge relationship curve is used as the inlet boundary of the one-dimensional model, and the river outlet water level process line or zero gradient condition is set as the outlet boundary of the one-dimensional model. The downstream boundary of the two-dimensional model adopts sponge boundary treatment to allow floodwater to flow freely out of the computational domain. The measured water level and discharge data of historical flood events are selected, and the model is calibrated by adjusting the roughness coefficient and the river roughness correction coefficient to ensure that the Nash efficiency coefficient (NSE) of the calculated value and the measured value is ≥0.85, so as to ensure that the accuracy of the two models meets the simulation requirements.

[0140] S26. Integrate the above parameters and model equations, realize data interaction between the one-dimensional model and the two-dimensional model through the model interface, construct a hydraulic model for the inundation development trend of the downstream reservoir, verify it using typical flood event data, output the simulation results of the inundation range and water depth distribution at different times, and compare it with the actual inundation range monitored by remote sensing, control the error within 10%, and complete the reliability verification of the hydraulic model for the inundation development trend of the downstream reservoir.

[0141] Furthermore, step S3 includes the following steps:

[0142] S31. Based on the preprocessed data from step S13 and the downstream inundation development trend of the reservoir, a hydraulic model is used to deduce three major scenarios: design flood, reservoir dam failure, and extreme precipitation.

[0143] In S31, the flood process line of a typical scenario is designed for flood simulation input, and the flood process and normal reservoir operation rules are determined by hydrological frequency analysis.

[0144] The reservoir dam failure simulation scenario sets the location of the breach in the middle of the dam or on the left bank, with a breach width of 50m or 100m. The failure modes are divided into instantaneous total failure and gradual failure.

[0145] Extreme precipitation scenarios are coupled with meteorological data, and uniform or peak-shaped rainfall patterns are input to design rainfall patterns.

[0146] The above data are respectively derived from the multi-source data of the downstream inundation simulation of the reservoir in step S11;

[0147] S32. Use the hydraulic model to calculate parameters for three major scenarios of downstream inundation development of the reservoir;

[0148] In step S32, the inundation range is determined by the dry and wet grids of the two-dimensional model, and grid cells with water depth greater than the set value (0.1m) are extracted. Vector boundaries are generated through GIS spatial analysis. The water depth distribution is processed by inverse distance weighted interpolation to process the water depth values ​​of the grid cells and generate a plane contour map. The arrival time is recorded as the moment when the water depth of each grid first exceeds the set value (0.1m). Combined with the water flow velocity field data for correction, a spatial distribution map of the inundation arrival time is generated. During the calculation process, the influence of flow velocity changes on the arrival time is corrected simultaneously to ensure time accuracy.

[0149] S33. Based on the scenario parameters in step S32, output standardized spatiotemporal sequence data from the hydraulic model derived from the downstream inundation development trend of the reservoir.

[0150] In S33, the water level-discharge process lines of key sections such as town centers and important dikes are extracted from the time series and stored in .csv format, which includes fields such as scenario ID, time, section ID, water level, and discharge.

[0151] The spatial sequence outputs the inundation range in .shp format, the water depth in .tif format, and the arrival time raster data in .tif format at a set time step (10 minutes), and correlates the scenario parameters with the inundation development trend of the downstream reservoir and the metadata of the hydraulic model.

[0152] Based on time series and spatial series, a spatiotemporal database containing standardized spatiotemporal sequence data is constructed using PostgreSQL+PostGIS database. Scenario information table and spatiotemporal data table are designed to realize unified storage and query of multi-scenario data.

[0153] S34. Based on standardized spatiotemporal sequence data, create dynamic animations of the inundation situation projection results output by the hydraulic model for the downstream inundation development trend projection of the reservoir.

[0154] In S34, the preprocessing stage aligns multi-step data along the time axis, uses a hierarchical color scheme to map water depth, represents arrival time with grayscale gradient, generates single-frame images using Python software or ArcGIS Pro, synthesizes MP4 video at a frame rate of 10 frames per second, overlays administrative divisions, important facility markings and time axis controls, and supports 0.5x-2x playback speed adjustment and keyframe pause query.

[0155] S35. Select measured data from historical flood events to verify the inundation situation projection results, ensuring that the mean absolute error is greater than or equal to the set value (0.3m) and the Nash efficiency coefficient is greater than or equal to the set value (0.85). By comparing and analyzing the dynamic curve of scenario-inundated area with the heat map of maximum water depth-population density, the risk differences of different scenarios are analyzed to help identify high-risk areas and provide data support for emergency decision-making.

[0156] Furthermore, step S4 includes the following steps:

[0157] S41. Based on the flooding situation simulation results and the characteristics of the disaster-bearing body, two major categories of risk variables, namely disaster-causing factors and disaster-bearing body vulnerability factors, are selected and standardized.

[0158] In S41, the disaster-causing factors include flooding depth, flooding flow velocity, and flooding duration;

[0159] Vulnerability factors of disaster-bearing bodies include population density, economic density, and infrastructure exposure.

[0160] The flooding risk variable factor is standardized to the [0,1] interval by min-max, that is, standardized value = (actual value - minimum value) / (maximum value - minimum value).

[0161] S42. A two-dimensional risk matrix framework of probability-consequence is adopted. A 5×5 risk matrix is ​​constructed by combining the characteristics of inundation scenarios and the quantified risk value R of inundation risk variable factors. The probability L is determined based on the return period of flood scenarios: 10-year return = 0.2 (low), 20-year return = 0.5 (medium), 50-year return = 0.8 (high), and dam failure scenario = 1.0 (extremely high). The consequences C are calculated by coupling the disaster-causing factors and the vulnerability factors of the disaster-bearing body.

[0162] In S42, consequence C is represented as:

[0163]

[0164] in, The water depth in the flooded area, For the flow rate of the flood, For the duration of flooding, For population density, For economic exposure, Land use type;

[0165] S43. Based on the risk value R and the actual disaster bearing capacity of the region, the flooding risk is divided into 5 levels. The level division results are generated by GIS spatial overlay to generate a risk level distribution map, and each level is rendered with different colors.

[0166] In S43, the flooding risk levels are as follows: 1) Low risk: R < 0.2; 2) Medium risk: 0.2 ≤ R < 0.4; 3) Higher risk: 0.4 ≤ R < 0.6; 4) High risk: 0.6 ≤ R < 0.8; 5) Extremely high risk: R ≥ 0.8;

[0167] S44. Integrate the results of the multi-scenario simulation in step S3 with the inundation risk variable factors to conduct a regional inundation risk level assessment, which includes single-scenario assessment and multi-scenario comprehensive assessment.

[0168] In S44, the single-scenario assessment process is as follows: 1) Extract the maximum water depth, flow velocity, and duration data of each grid cell output in step S3; 2) Overlay spatial layers of population density, economic density, and infrastructure exposure; 3) Calculate the consequence value C using the formula in step S42, and obtain the risk value R by combining the scenario probability L; 4) Divide the grid cell risk level according to the threshold in step S43.

[0169] The multi-scenario comprehensive assessment process is as follows: the risk levels of the design flood, dam break, and extreme precipitation scenarios are weighted and averaged to eliminate single scenario bias. The assessment results are verified by historical disaster data to ensure that the spatial overlap rate between high-risk areas and actual severely affected areas is greater than or equal to the set value (80%), and the average risk level error is less than or equal to the set value (0.5).

[0170] S45. Based on the assessment results of the regional inundation risk level assessment, construct an inundation risk database to achieve systematic storage and management of risk information;

[0171] In step S45, the database design adopts a three-level architecture of scenario-region-risk: 1) The scenario layer stores metadata such as scenario ID, type, and recurrence period; 2) The region layer includes administrative division code, geographical boundary and disaster-bearing body attributes; 3) The risk layer records the risk value, level and key disaster-causing factors of each region under different scenarios. The risk layer supports multi-dimensional queries and is linked with the standardized spatiotemporal sequence data in step S3 to realize a closed-loop data process of inference-assessment-query.

[0172] Specifically, the classification table of probability L is shown in Table 2;

[0173] Table 2

[0174]

[0175] The consequences C classification table is shown in Table 3.

[0176] Table 3

[0177]

[0178] The risk value calculation and classification table is shown in Table 4;

[0179] Table 4

[0180]

[0181] The classification table of flood risk levels is shown in Table 5.

[0182] Table 5

[0183]

[0184] Furthermore, in step S5, when constructing an inundation risk database based on the assessment results built in step S4 and developing an inundation situation projection system, a PostgreSQL+PostGIS risk database is used as the data core. This is achieved by integrating three-dimensional terrain data (DEM / DOM) with the dynamic inundation spatiotemporal sequence generated in step S3, and using WebGL or Unity. The 3D engine enables three-dimensional visualization of the dynamic inundation process, allowing users to interactively view the flood evolution process under different scenarios in real time. It also overlays risk level data from the risk database, automatically rendering risk level thematic maps with different color gradients, and simultaneously displays administrative divisions, important facility markings, and key disaster-causing factors. For the flood evacuation route planning function, the system calls the spatial distribution data of high-risk areas in the risk database, coupled the road network topology and refuge site capacity information, and generates the optimal flood evacuation route based on the improved A* algorithm or Dijkstra algorithm, while avoiding high-risk road sections with inundation depth > 0.5m and flow velocity > 1m / s. The route planning results need to simultaneously display the estimated evacuation time, the risk level of the route, and the real-time capacity of the refuge sites, and support batch planning with multiple starting points and multiple ending points and export of route feasibility assessment reports. Ultimately, it realizes the integrated function of dynamic simulation, risk visualization, and route planning.

[0185] Although the invention has been described with reference to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and instructional purposes, and not for the purpose of interpreting or limiting the subject matter of the invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the invention is illustrative and not restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. A method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, characterized in that, Includes the following steps: S1. Collect relevant data on downstream inundation of the reservoir to form multi-source data for downstream inundation projection. Preprocess the data through coordinate unification, terrain correction and data standardization to obtain multi-source dataset for downstream inundation projection. S2. Select and couple a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model, set the coupling nodes and time steps, clarify the control equations of the model, calibrate the parameters by assigning roughness coefficients to partitions and setting upstream and downstream boundary conditions, and construct a hydraulic model for inferring the inundation development trend of the downstream reservoir. S3. Based on the multi-source dataset of downstream reservoir inundation simulation and the hydraulic model of downstream reservoir inundation development trend simulation, set up a multi-scenario inundation trend simulation scheme, calculate the inundation range, water depth distribution and arrival time under different scenarios, and output the spatiotemporal sequence data and dynamic evolution animation of multi-scenario simulation results. S4. Define inundation risk variable factors, classify risk levels by risk values, and complete the inundation risk level assessment of the downstream reservoir and construct an inundation risk database based on the multi-scenario simulation results of step S3. S5. Based on the inundation risk database constructed in step S4, integrate the functions of dynamic inundation process 3D visualization, automatic rendering of risk level thematic maps and flood evacuation route planning to complete the simulation of the inundation development trend downstream of the reservoir.

2. The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, as described in claim 1, is characterized in that... S1 includes the following steps: S11. Collect topographic data, hydrological data, socio-economic data, and real-time monitoring data, i.e., data related to downstream inundation of the reservoir, to form multi-source data for downstream inundation projection of the reservoir; In S11, the terrain data includes digital elevation model, contour line data, and administrative division vector map; Hydrological data includes historical flood hydrographs, river cross-section data, and reservoir operation rules; Socioeconomic data includes population distribution, housing construction, and infrastructure data; Real-time monitoring data includes reservoir water levels, river flow, and rain gauge data; S12. Extract the required information from the multi-source data of downstream inundation simulation of the reservoir and construct their respective databases; In step S12, elevation values, resolution, and projection information fields are extracted from the digital elevation model to construct a terrain attribute database in .mdb format; Extract the administrative code and area fields from the administrative division vector map to generate administrative division attribute data in .shp format; Extract flood discharge, water level, and occurrence time fields from historical flood data obtained based on historical flood hydrographs, and construct a hydrological time series database in .xlsx format; The scheduling curves and flood discharge capacity parameters are extracted from the reservoir scheduling rules to form a scheduling parameter attribute table; Extract regional population size and density fields from population distribution data to construct a population attribute database; Extract building type, number of floors, and coordinate fields from building vector data to generate a socioeconomic element layer in .shp format; The monitoring time, water level, and flow fields are extracted from sensor data of reservoir water level, river flow, and rain gauge stations. A real-time monitoring database in .csv format is constructed and associated with the monitoring station ID and spatial coordinates. S13. Perform data preprocessing on the respective databases generated in step S12 to obtain a multi-source dataset for downstream inundation simulation of the reservoir. In step S13, the terrain attribute database, hydrological time series database, socio-economic element layer and real-time monitoring database from step S12 are imported into the geographic information system platform software. At the same time, high-resolution remote sensing image data in GeoTIFF format is loaded and subjected to coordinate unification, terrain correction and data standardization. Coordinate unification: The national 2000 geodetic coordinate system is used to transform the coordinates of all spatial data to ensure consistent spatial reference for topographic, hydrological, and socio-economic data; Terrain correction: Using the 3D Analyst tool in the geographic information system platform software, outliers in the digital elevation model data are removed to generate a smooth terrain surface; Data standardization: Hydrological data is converted to a uniform time step, socioeconomic data is aggregated according to a set grid unit, and real-time monitoring data format is converted to GeoJSON; Finally, the integrated multi-source data was published using ArcGIS Server to provide map services, forming a standardized multi-source dataset for downstream inundation simulation of reservoirs.

3. The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, as described in claim 2, is characterized in that... S2 includes the following steps: S21. Based on the coupling mechanism of wet and dry boundaries, a one-dimensional river hydrodynamic model and a two-dimensional hydrodynamic model are selected for coupling. That is, the river flood process calculated by the one-dimensional model is used as the upstream inflow boundary of the two-dimensional model to realize the dynamic connection between the river and the inundation zone flow. S22. Set up coupling nodes at key sections at the junction of the river channel and the inundation area, extract water level / discharge data from the one-dimensional model (i.e., the one-dimensional river hydrodynamic model) as the boundary input of the two-dimensional model (i.e., the two-dimensional hydrodynamic model), and simultaneously collect feedback on the inundation range calculated by the two-dimensional model to correct the downstream boundary conditions of the one-dimensional model. In S22, an adaptive time step algorithm is adopted. The time step of the one-dimensional model is controlled within a set range according to the Courant number, and the time step of the two-dimensional model is dynamically adjusted based on the minimum grid size and water flow velocity to ensure the stability of coupled calculation. The time steps of the two models are synchronized by interpolation algorithm to achieve data exchange. S23. Based on steps S21 and S22, mathematical expressions are given for the one-dimensional river hydrodynamic model and the two-dimensional hydrodynamic model. In S23, the mathematical expression of the one-dimensional river hydrodynamic model is as follows: ； ； Where B is the width of the water surface, Q is the flow rate, Z is the water level, q is the lateral flow rate, and t is the time. Where is the distance along the direction of water flow, A is the cross-sectional area of ​​the water passage, g is the acceleration due to gravity, R is the hydraulic radius, and C is the Chezy coefficient; The mathematical expression of a two-dimensional hydrodynamic model includes the continuity equation and the momentum equation; The continuity equations of the two-dimensional hydrodynamic model are expressed as follows: ； The momentum equation for the two-dimensional hydrodynamic model is expressed as: ； Where H is the water depth and Z is the water level. For the source and sink terms in the continuity equation, M and N are the vertical average unit width flow rates in the x and y directions, respectively, u and v are the components of the vertical average velocity in the x and y directions, n is the Manning roughness coefficient, and g is the gravitational acceleration. S24. Assign values ​​to the roughness coefficient n based on land use type, and use ArcGIS spatial analysis tools to convert the assigned values ​​of the roughness coefficient into grid data of a two-dimensional model; S25. The reservoir discharge flow process line or water level-discharge relationship curve is used as the inlet boundary of the one-dimensional model, and the river outlet water level process line or zero gradient condition is set as the outlet boundary of the one-dimensional model. The downstream boundary of the two-dimensional model adopts sponge boundary treatment to allow floodwater to flow freely out of the computational domain. The measured water level and discharge data of historical flood events are selected, and the model is calibrated by adjusting the roughness coefficient and the river roughness correction coefficient to ensure that the Nash efficiency coefficient of the calculated value and the measured value is ≥0.85, so as to ensure that the accuracy of the two models meets the simulation requirements. S26. Integrate the above parameters and model equations, realize data interaction between the one-dimensional model and the two-dimensional model through the model interface, construct a hydraulic model for the inundation development trend of the downstream reservoir, verify it using typical flood event data, output the simulation results of the inundation range and water depth distribution at different times, and compare it with the actual inundation range monitored by remote sensing, control the error within 10%, and complete the reliability verification of the hydraulic model for the inundation development trend of the downstream reservoir.

4. The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model according to claim 3, characterized in that, S3 includes the following steps: S31. Based on the preprocessed data from step S13 and the downstream inundation development trend of the reservoir, a hydraulic model is used to deduce three major scenarios: design flood, reservoir dam failure, and extreme precipitation. In S31, the flood process line of a typical scenario is designed for flood simulation input, and the flood process and normal reservoir operation rules are determined by hydrological frequency analysis. The reservoir dam failure simulation scenario sets the location of the breach in the middle of the dam or on the left bank, and the failure mode is divided into instantaneous total failure and gradual failure. Extreme precipitation scenarios are coupled with meteorological data, and rainfall patterns are designed by inputting uniform rainfall or peak rainfall. S32. Use the hydraulic model to calculate parameters for three major scenarios of downstream inundation development of the reservoir; In step S32, the inundation range is determined by the dry and wet grids of the two-dimensional model, and grid cells with water depths greater than a set value are extracted. Vector boundaries are generated through GIS spatial analysis. The water depth distribution is processed by inverse distance weighted interpolation to process the water depth values ​​of the grid cells and generate a plane contour map. The arrival time is recorded as the moment when the water depth of each grid first exceeds the set value. Combined with the water flow velocity field data for correction, a spatial distribution map of the inundation arrival time is generated. During the calculation process, the influence of flow velocity changes on the arrival time is corrected synchronously to ensure time accuracy. S33. Based on the scenario parameters in step S32, output standardized spatiotemporal sequence data from the hydraulic model derived from the downstream inundation development trend of the reservoir. In step S33, the water level-flow process line of key sections is extracted from the time series and stored in .csv format, which includes scenario ID, time, section ID, water level, and flow field; The spatial sequence outputs the inundation range in .shp format, the water depth in .tif format, and the arrival time raster data in .tif format according to the set time step, and correlates the scenario parameters with the inundation development trend of the downstream reservoir and the metadata of the hydraulic model. Based on time series and spatial series, a spatiotemporal database containing standardized spatiotemporal sequence data is constructed using PostgreSQL+PostGIS database. Scenario information table and spatiotemporal data table are designed to realize unified storage and query of multi-scenario data. S34. Based on standardized spatiotemporal sequence data, create dynamic animations of the inundation situation projection results output by the hydraulic model for the downstream inundation development trend projection of the reservoir. In S34, the preprocessing stage aligns multi-step data according to the time axis, uses a hierarchical color scheme to map water depth, represents arrival time with grayscale gradient, generates single-frame images using Python software or ArcGIS Pro, synthesizes MP4 videos, and overlays administrative divisions, important facility labels, and time axis controls. S35. Select measured data from historical flood events to verify the inundation situation projection results, ensuring that the mean absolute error is greater than or equal to the set value and the Nash efficiency coefficient is greater than or equal to the set value. By comparing and analyzing the risk differences of different scenarios through the dynamic curve of scenario-inundated area and the heat map of maximum water depth-population density, we can help identify high-risk areas and provide data support for emergency decision-making.

5. The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model according to claim 4, characterized in that, S4 includes the following steps: S41. Based on the flood situation simulation results and the characteristics of the disaster-bearing body, two major categories of risk variables, namely disaster-causing factors and disaster-bearing body vulnerability factors, are selected and standardized. In S41, the disaster-causing factors include flooding depth, flooding flow velocity, and flooding duration; Vulnerability factors of disaster-bearing bodies include population density, economic density, and infrastructure exposure. The flooding risk variable factor is standardized to the [0,1] interval by min-max, that is, standardized value = (actual value - minimum value) / (maximum value - minimum value). S42. A two-dimensional risk matrix framework of probability-consequence is adopted. The risk matrix is ​​constructed by combining the characteristics of the inundation scenario with the quantified risk value R of the inundation risk variable factors. The probability L is determined based on the return period of the flood scenario, and the consequence C is calculated by coupling the disaster-causing factors and the vulnerability factors of the disaster-bearing body. In S42, consequence C is represented as: ； in, The water depth in the flooded area, For the flow rate of the flood, For the duration of flooding, For population density, For economic exposure, Land use type; The risk value R is represented as: R=L×C S43. Based on the risk value R and the actual disaster bearing capacity of the region, the flooding risk is divided into 5 levels. The level division results are generated by GIS spatial overlay to generate a risk level distribution map, and each level is rendered with different colors. In S43, the flooding risk levels are as follows: 1) Low risk: R < 0.2; 2) Medium risk: 0.2 ≤ R < 0.4; 3) Higher risk: 0.4 ≤ R < 0.6; 4) High risk: 0.6 ≤ R < 0.8; 5) Extremely high risk: R ≥ 0.8; S44. Integrate the results of the multi-scenario simulation in step S3 with the inundation risk variable factors to conduct a regional inundation risk level assessment, which includes single-scenario assessment and multi-scenario comprehensive assessment. In S44, the single-scenario assessment process is as follows: 1) Extract the maximum water depth, flow velocity, and duration data of each grid cell output in step S3; 2) Overlay spatial layers of population density, economic density, and infrastructure exposure; 3) Calculate the consequence value C using the formula in step S42, and obtain the risk value R by combining the scenario probability L; 4) Divide the grid cell risk level according to the threshold in step S43. The multi-scenario comprehensive assessment process is as follows: the risk levels of the design flood, dam break and extreme precipitation scenarios are weighted and averaged to eliminate single scenario bias. The assessment results are verified by historical disaster data to ensure that the spatial overlap rate between high-risk areas and actual severely affected areas is greater than or equal to the set value, and the average risk level error is less than or equal to the set value. S45. Based on the assessment results of the regional inundation risk level assessment, construct an inundation risk database to achieve systematic storage and management of risk information; In step S45, the database design adopts a three-level architecture of scenario-region-risk: 1) The scenario layer stores metadata such as scenario ID, type, and recurrence period; 2) The region layer includes administrative division code, geographical boundary and disaster-bearing body attributes; 3) The risk layer records the risk value, level and key disaster-causing factors of each region under different scenarios. The risk layer supports multi-dimensional queries and is linked with the standardized spatiotemporal sequence data in step S3 to realize a closed-loop data process of inference-assessment-query.

6. The method for predicting the downstream inundation development trend of a reservoir based on a hydraulic model, as described in claim 5, is characterized in that... In step S5, an inundation risk database is constructed based on the assessment results built in step S4. When developing the inundation situation simulation system, the PostgreSQL+PostGIS risk database is used as the data core. By integrating three-dimensional terrain data with the dynamic inundation spatiotemporal sequence generated in step S3, an engine is used to realize the three-dimensional visualization of the dynamic inundation process. The risk level data in the risk database is overlaid, and the risk level thematic map is automatically rendered with different colors. The administrative divisions, important facility markings, and key disaster-causing factors are displayed simultaneously. For the flood evacuation route planning function, the system calls on the spatial distribution data of high-risk areas in the risk database, coupled the road network topology and refuge site capacity information, and generates the optimal flood evacuation route based on the improved A* algorithm or Dijkstra algorithm, while avoiding high-risk road sections with flood depth > 0.5m and flow velocity > 1m / s. The route planning results need to simultaneously display the estimated evacuation time, the risk level of the route, and the real-time capacity of the refuge sites, and support batch planning with multiple starting points and multiple ending points and export of route feasibility assessment reports, ultimately realizing the integrated function of dynamic simulation, risk visualization and route planning.

Images