A flood early warning intelligent management system and method based on multi-source data fusion

Abstract

The application discloses a kind of flood early warning intelligent management system and method based on multi-source data fusion, it is related to flood early warning management technical field;The application collects the hydrological data of target area in river basin, constructs river basin simulation model;Target area is carried out grid classification processing, obtains river channel grid and soil grid;According to the hydrological data of soil grid, the real-time water storage state of each soil grid is analyzed, and the runoff generation data of each soil grid is determined;According to the hydrological data of river channel grid, the real-time flow data of each river channel grid is analyzed;Determine the soil grid associated with each river channel grid, according to the runoff generation data of associated soil grid, the runoff inflow data of each associated soil grid to river channel grid is analyzed;And according to the analysis data, the river channel grid with flood risk is located and warned;The application can accurately locate the river channel grid with flood risk, significantly improve the river flow monitoring accuracy and flood early warning timeliness.

Description

Technical Field

[0001] This invention relates to the field of flood early warning management technology, specifically a flood early warning intelligent management system and method based on multi-source data fusion. Background Technology

[0002] Currently, flood warnings in river basins mainly rely on real-time monitoring data such as river levels and flow velocities, combined with rainfall forecast information, and are determined through human experience or simple hydrological model extrapolation. However, under precipitation conditions, the dynamic changes in river flow are not only affected by direct rainfall replenishment, but also greatly constrained by the soil water storage status and the process of slope runoff inflow within the basin. Existing monitoring systems often separate river flow from soil hydrological processes, failing to effectively integrate multi-source information such as soil water storage changes, runoff generation mechanisms, and confluence paths. This leads to significant biases in monitoring precipitation and slope runoff. Especially under conditions of heavy rainfall, soil moisture content rapidly saturates, and a large amount of runoff flows into rivers, causing a sharp increase in river flow. However, traditional monitoring methods struggle to quantify the soil's contribution to river runoff in real time, resulting in delayed and incomplete river flow monitoring data that fails to accurately reflect the true evolution of floods. This monitoring blind spot directly weakens the timeliness and reliability of early warnings, leading to inaccurate flood risk assessments and significantly increasing the risk of flooding in riverside areas. Summary of the Invention

[0003] The purpose of this invention is to provide a flood early warning intelligent management system and method based on multi-source data fusion, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A flood early warning intelligent management method based on multi-source data fusion, the method includes the following steps: Determine the target area, collect watershed hydrological data within the target area, and construct a watershed simulation model; In the watershed simulation model, the target area is subjected to raster classification processing to obtain river raster and soil raster; Based on the hydrological data of the soil grid, the real-time water storage status of each soil grid is analyzed, and the runoff generation data of each soil grid is determined based on the real-time water storage status. Analyze the real-time flow data of each river grid based on the hydrological data of the river grid; determine the soil grid associated with each river grid; and analyze the runoff collection data of each associated soil grid to the river grid based on the runoff generation data of the associated soil grid. Based on the runoff data of the soil grids associated with each of the aforementioned river grids, locate and alert river grids that are at risk of flooding.

[0005] Furthermore, a target area is selected in the actual scenario, and data surveys are conducted on the watershed within the target area to collect watershed hydrological data; It should be noted that a watershed includes a flowing area that can support or affect human activities, as well as the surrounding soil area. The watershed hydrological data includes water level, flow velocity, soil moisture, and precipitation, etc. The watershed area in the target region is subjected to three-dimensional simulation processing using a simulation model, and a watershed simulation model of the target region is constructed by combining imported hydrological data. It should be noted that the watershed simulation model can simulate the virtual state of the watershed in the actual scene by mapping the watershed topographic data and hydrological data in the actual scene, which facilitates data observation and analysis.

[0006] Furthermore, based on the watershed simulation model, the river area and soil area in the target region are rasterized to obtain river raster and soil raster respectively. When dividing the river channel grid and soil grid, the equal-length river channel grid is used. The river channel width is used as the reference, and the river channel length is divided at equal intervals to obtain the river channel grid. When dividing the soil into grids, the division range should be determined first. The soil area within the river channel should be divided into grids of the same size.

[0007] It should be noted that using equal-length grids for river channels facilitates a holistic analysis of the river cross-section during subsequent flow analysis. If equal-size grids are used, the overall impact of soil runoff on river flow is not considered, and only grids not located at the river's edges are included, which can easily lead to analytical errors.

[0008] Furthermore, in the watershed simulation model, hydrological data from each soil grid is retrieved, and the water storage status of each soil grid is analyzed within a preset monitoring period to obtain the water storage capacity P of the soil grid within the current period. i,t Its calculation is as follows: ; Among them, P i,t Let k be the water storage of soil grid cell i during time period t; i P is the water storage decline coefficient for soil grid number i. i,t-1 p represents the water storage of soil grid cell i during the time period t-1; i,tLet represent the amount of precipitation received by soil grid cell i during time period t. It should be noted that the water storage decline coefficient of the soil grid cell represents the rate of soil moisture decline; the water storage capacity of the soil grid cell represents the total amount of water actually held in the soil pores, expressed as equivalent water depth.

[0009] The current water storage capacity of each soil grid is compared with its maximum water storage capacity, and the runoff depth R of each soil grid in the current time period is determined based on the comparison results. i,t The analysis is as follows: ; Among them, R i,t P represents the runoff depth value of soil grid cell i within time period t; i,max The maximum water storage capacity of soil grid number i; Based on the runoff depth value R generated by each soil grid during the current time period i,t By combining the area specifications of the soil grid, the runoff volume V of each soil grid in the current time period is obtained. i,t The calculation is as follows: ; Among them, V i,t Let A be the runoff volume generated by soil grid cell i during time period t; i Let i be the area of ​​soil grid number i. In this embodiment, under normal circumstances, the runoff generated by the soil is expressed in terms of equivalent water depth. When expressing precipitation or runoff depth in terms of equivalent water depth, the unit is millimeters. Therefore, when calculating the runoff volume of the soil grid, unit conversion is required for subsequent calculation and analysis.

[0010] Furthermore, hydrological data corresponding to each river grid is retrieved separately. Based on the water level data of each river grid in the current time period, a dynamic flow model is constructed. The cross-sectional flow data of the river is analyzed at different water levels to determine the cross-sectional flow Q corresponding to the real-time water level of each river grid in the current time period. j,t The calculation is as follows:

[0011] Among them, Q j,t Let H be the cross-sectional flow rate of channel grid j during time period t; H be the water level of channel grid j during time period t; h be the water depth; W(h) be the river width of channel grid j corresponding to water depth h during time period t; n be the channel roughness of channel grid j; r H denoted as , where is the hydraulic radius of channel grid j at water level H; and is the water surface gradient of channel grid j.

[0012] In this embodiment, during the analysis of the cross-sectional flow of each river grid, since the river water is in a flowing state, its water level has a dynamic range of change. Therefore, when calculating the real-time cross-sectional flow of each river grid, it is not a unique fixed value. Thus, in the actual analysis process, the cross-sectional flow of a single river grid is often calculated multiple times, and the average value is taken as the real-time cross-sectional flow of the current time period.

[0013] Furthermore, using each river grid as the association base point, the association of each soil grid is determined; by analyzing the runoff direction of each soil grid, the runoff path of the soil grid is determined; and the soil grids whose runoff paths are directed to the river grid are marked as associated.

[0014] Specifically, the runoff direction analysis of the soil grid, to obtain the runoff path of the soil grid, involves: T1. Take any soil grid as the center grid and search and determine the soil grids adjacent to it; T2. Obtain the elevation values ​​of the center points of the central grid and each adjacent soil grid, as well as the horizontal distance between the center point of the central grid and the center points of each adjacent soil grid; using the central grid as the gradient reference, calculate the slope gradient value F for each adjacent soil grid. m The analysis is performed; the calculation is as follows: ; Among them, F m Z represents the slope gradient between adjacent soil grid cells and the central grid cell, where m is the grid number. i Z represents the elevation value of the center point of the central grid cell numbered i; m L is the elevation value of the center point of the adjacent soil grid cell numbered m; m The horizontal distance between the center points of adjacent soil grids numbered m and the center grid numbered i; T3. Based on the directional distribution of adjacent soil grids of the central grid, the runoff direction of the central grid is determined by numbering; and based on the slope gradient values ​​between each adjacent soil grid and the central grid, the runoff direction of the central grid is analyzed, and the calculation is as follows: ; Among them, D i The runoff direction is assigned to the central grid cell numbered i; T4. The flow direction of each central grid is determined by continuously analyzing the flow direction, and the flow path of each central grid is determined by connecting the flow directions. It should be noted that the runoff direction numbering is used to indicate the runoff direction, so that the algorithm can directly output the slope gradient between the central grid and the adjacent soil grid, and the corresponding runoff direction.

[0015] In this embodiment, the runoff direction numbering can adopt the ESRI standard coding, where 1=East, 2=Southeast, 4=South, 8=Southwest, 16=West, 32=Northwest, 64=North, and 128=Northeast. It should be noted that if there is a maximum (Fm) < 0 in the slope gradient values ​​between a central grid and an adjacent soil grid, the grid is identified as a depression and needs to be marked as an abnormal confluence. When there are multiple equal maximum values ​​in the slope gradient between the central grid and an adjacent soil grid, the grid is output in the highest order according to the input order of the runoff direction number. It should also be noted that when determining the runoff path of the central grid, there may be soil grids along the runoff path that do not generate runoff, which may result in runoff consumption. However, this invention takes into account the scenario where, under precipitation conditions, the soil around the river generates runoff and undergoes confluence. On the one hand, each soil grid still receives water storage from precipitation, and on the other hand, the flow velocity of the runoff is much higher than the rate at which water infiltrates the soil. Therefore, in this case, the consumption of runoff during the flow process is not considered; it is only regarded as the confluence and transport of runoff generated by each soil grid.

[0016] Furthermore, soil graticles associated with each river graticle are integrated and a set of associated soil graticles, I, is constructed. A comprehensive runoff inflow analysis is performed on each associated soil graticle to determine the runoff inflow Q of each soil graticle associated with each river graticle at the current time point. j,I,t Its calculation is as follows: ; Among them, v j,i The runoff velocity is the soil velocity associated with the j-th channel grid (i); v0 is the baseline velocity; β j,i The slope of soil grid i associated with river grid j; y is an empirical index; θ(ΣV j,i ) represents the correction factor for the runoff moisture content after runoff convergence to the runoff velocity of the soil grid numbered i associated with the j-th channel grid; T j,i The time required for runoff from soil grid i associated with river grid j to reach the associated river grid; i∈path(I) represents the soil grids traversed by the path of runoff from soil grid i to the associated river grid; L j,i,i+1 Q represents the path of runoff from soil grid i associated with river grid j through the next soil grid; j,I,t The real-time runoff inflow rate of each soil cell in the associated soil cell set associated with cell j (channel cell j) during time period t; T j,i →t represents the soil grids in the associated soil grid set associated with the river grid j, where runoff arrives at the river grid in real time during the current time period t; V j,i,tThe runoff volume generated by soil grid number i associated with river grid number j during time period t.

[0017] In this embodiment, the reference flow velocity is manually set by the management personnel according to the utilization / coverage type of the soil grid; since the runoff generated by multiple soil grids will cause the runoff water content to increase after confluence, it will affect the runoff velocity, so a correction coefficient is introduced for compensation. Since there are multiple soil grid runoffs converging on the runoff path of each soil grid, the runoff velocities between different adjacent soil grids on the path are different. Therefore, it is necessary to analyze the runoff time between each soil grid on the runoff path in segments and finally sum them to obtain the time when the runoff from the corresponding soil grid reaches the river grid. Since the river grid is associated with multiple soil grids, and the runoff from different soil grids arrives at the river grid at different times, the runoff concentrations generated after their convergence are different. Therefore, it is necessary to calculate the runoff concentrations of the soil grid runoffs that arrive at the river grid at the same time.

[0018] Furthermore, by determining the current warning water level data of each river grid and importing it into the dynamic flow model, the warning value Q of the current river cross-sectional flow of each river grid is obtained. j,max ; Set the real-time cross-sectional flow value Q of each river grid to the current value. j,t The runoff sink volume Q of each associated soil grid j,I,t Sum the values ​​and compare the sum with the warning values; If it is greater than or equal to the warning value Q j,max If so, it is determined that the current river grid with number j is at risk of flooding; If it is less than the warning value Q j,max If so, then the current number j river channel grid is determined to be normal; Based on the comparison results, river grids with flood risk are located in the watershed simulation model and marked with risk warnings.

[0019] It should be noted that the warning value is used to alert that there may be a flood risk in the river channel, and does not mean that a flood will occur when the river channel flow reaches the warning value; therefore, when there is a flood risk, it can provide managers with accurate early location and timely remedial space.

[0020] Example 2: The present invention provides another technical solution: A flood early warning intelligent management system based on multi-source data fusion: The system includes a model simulation unit, a grid division unit, a runoff analysis unit, a river flow analysis unit, a correlation assessment unit, a confluence impact analysis unit, and a flood risk assessment unit.

[0021] Furthermore, the model simulation unit conducts data surveys on the watersheds in the target area, collects watershed hydrological data, and constructs a watershed simulation model for the target area.

[0022] Furthermore, the grid division unit, based on the watershed simulation model, performs rasterization processing on the river area and soil area in the target area to obtain river raster and soil raster respectively.

[0023] Furthermore, the runoff analysis unit retrieves the hydrological data of each soil grid, analyzes the water storage status of each soil grid within the current time period through a preset monitoring period, and determines the runoff depth value of each soil grid within the current time period by comparing the score with the maximum water storage capacity of the soil grid; and obtains the runoff volume of each soil grid within the current time period by combining the area specification data of the soil grid.

[0024] Furthermore, the river flow analysis unit retrieves the hydrological data corresponding to each river grid, and constructs a dynamic flow model based on the water level data of each river grid in the current time period to analyze the river cross-sectional flow corresponding to the real-time water level of each river grid in the current time period.

[0025] Furthermore, the association assessment unit uses each river grid as the association base point to determine the association of each soil grid; it determines the runoff path of each soil grid by analyzing the runoff direction of each soil grid; and marks the soil grids whose runoff paths lead to the river grids as having an association status.

[0026] Furthermore, the runoff impact analysis unit coordinates the soil grates that are marked as associated with each river grates and constructs a set of associated soil grates; it performs a comprehensive runoff runoff analysis on each associated soil grates to determine the runoff runoff volume of each soil grates associated with each river grates at the current time point.

[0027] Furthermore, the flood risk assessment unit obtains the early warning value of the current river cross-sectional flow of each river grid; it sums the current real-time river cross-sectional flow value of each river grid with the current runoff inflow of each associated soil grid, and compares it with the early warning value; based on the comparison result, it locates the river grid with flood risk in the watershed simulation model and marks it with risk warning.

[0028] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention divides the watershed into river grids and soil grids, analyzes the real-time water storage status and runoff generation of the soil, and calculates the runoff concentration of the associated soil grids to the river grids. This achieves the organic integration of soil hydrological response and river dynamic monitoring, overcomes the shortcomings of traditional methods that separate the two, significantly improves the accuracy of river flow monitoring and the timeliness of flood warning, and effectively reduces the risk of missed flood reports due to runoff concentration estimation errors. 2. This invention analyzes runoff flow based on a grid scale, which can accurately locate specific river grid locations with flood risk, giving the early warning results clear spatial directionality. This provides a reliable basis for the precise allocation of flood control resources and the rapid activation of emergency response, effectively improving the refined management and control capabilities of watershed flood control and disaster reduction. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating an intelligent flood early warning management method based on multi-source data fusion according to the present invention. Detailed Implementation

[0030] 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.

[0031] Example 1: As Figure 1 As shown, the present invention provides a technical solution: A flood early warning intelligent management method based on multi-source data fusion, the method includes the following steps: Determine the target area, collect watershed hydrological data within the target area, and construct a watershed simulation model; In the watershed simulation model, the target area is subjected to raster classification processing to obtain river raster and soil raster; Based on the hydrological data of the soil grid, the real-time water storage status of each soil grid is analyzed, and the runoff generation data of each soil grid is determined based on the real-time water storage status. Analyze the real-time flow data of each river grid based on the hydrological data of the river grid; determine the soil grid associated with each river grid; and analyze the runoff collection data of each associated soil grid to the river grid based on the runoff generation data of the associated soil grid. Based on the runoff data of the soil grids associated with each of the aforementioned river grids, locate and alert river grids that are at risk of flooding.

[0032] Furthermore, a target area is selected in the actual scenario, and data surveys are conducted on the watershed within the target area to collect watershed hydrological data; It should be noted that a watershed includes a flowing water area that can support or affect human activities, as well as the surrounding soil area.

[0033] In the embodiments, the watershed in the target area of ​​the present invention is preferably a river and the soil area around the river located in or around an area of ​​human activity; wherein the area of ​​human activity can be an urban area or a village area; The watershed hydrological data includes water level, flow velocity, soil moisture, and precipitation, etc. In this embodiment, the state of the river can be determined by the river water level and river flow velocity in the target area; the water storage status of the soil around the river can be analyzed by precipitation and soil moisture to determine whether the generation of soil runoff has an impact on the river and causes flood risk.

[0034] The watershed area in the target region is subjected to three-dimensional simulation processing using a simulation model, and a watershed simulation model of the target region is constructed by combining imported hydrological data. It should be noted that the watershed simulation model can simulate the virtual state of the watershed in the actual scene by mapping the watershed topographic data and hydrological data in the actual scene, which facilitates data observation and analysis.

[0035] In this embodiment, terrain data of river areas in real-world scenarios can be collected using facilities such as drones / unmanned vessels and hydrological stations, including river width, river depth, river slope, and surrounding soil distribution and slope; and the hydrological data can be imported to realize data model simulation.

[0036] Furthermore, based on the watershed simulation model, the river area and soil area in the target region are rasterized to obtain river raster and soil raster respectively. When dividing the river channel grid and soil grid, the equal-length river channel grid is used. The river channel width is used as the reference, and the river channel length is divided at equal intervals to obtain the river channel grid. When dividing the soil into grids, the division range should be determined first. The soil area within the river channel should be divided into grids of the same size.

[0037] It should be noted that using equal-length grids for river channels facilitates a holistic analysis of the river cross-section during subsequent flow analysis. If equal-size grids are used, the overall impact of soil runoff on river flow is not considered, and only grids not located at the river's edges are included, which can easily lead to analytical errors.

[0038] In this embodiment, if the river in the current target area is 10 meters wide and 100 meters long, then when dividing the river grid, the 10-meter width should be used as the basis and the 100-meter length should be divided at equal intervals. For example, if the interval is 5 meters, the river grid obtained by the division will be 10 meters wide and 5 meters long. When dividing the soil into grids, the operators need to determine the soil area within the range on both sides of the river based on experience or data analysis. If the soil area within 5 meters on both sides of the river is selected, square grids of equal size, such as equal length and width, are used to divide the soil area within 5 meters on both sides of the river into grids. When dividing the river channel grid and soil grid, three-dimensional attributes need to be considered. Therefore, both the river channel grid and the soil grid are three-dimensional grid blocks. The river channel grid is determined by the actual water level, while the soil grid is divided according to preset specifications with equal height, width and length.

[0039] Furthermore, in the watershed simulation model, hydrological data from each soil grid is retrieved, and the water storage status of each soil grid is analyzed within a preset monitoring period to obtain the water storage capacity P of the soil grid within the current period. i,t Its calculation is as follows: ; Among them, P i,t Let k be the water storage of soil grid cell i during time period t; i P is the water storage decline coefficient for soil grid number i. i,t-1 p represents the water storage of soil grid cell i during the time period t-1; i,t The amount of precipitation received by soil grid cell number i during time period t.

[0040] It should be noted that the water storage decline coefficient of the soil grid is expressed as the rate of soil moisture decline; the water storage capacity of the soil grid is expressed as the total amount of water actually held in the soil pores, expressed as equivalent water depth.

[0041] In this embodiment, the precipitation received by the soil grid is the equivalent water depth of the precipitation at the corresponding soil grid, which represents the precipitation data at the soil grid and is not the same as the water storage capacity of the soil grid.

[0042] The current water storage capacity of each soil grid is compared with its maximum water storage capacity, and the runoff depth R of each soil grid in the current time period is determined based on the comparison results. i,t The analysis is as follows: Among them, R i,t P represents the runoff depth value of soil grid cell i within time period t; i,max The maximum water storage capacity of soil grid number i.

[0043] Based on the runoff depth value R generated by each soil grid during the current time period i,t By combining the area specifications of the soil grid, the runoff volume V of each soil grid in the current time period is obtained. i,t Its calculation is as follows: ; Among them, V i,t Let A be the runoff volume generated by soil grid cell i during time period t; i Let i be the area of ​​the soil grid with number i.

[0044] In this embodiment, under normal circumstances, the runoff generated by the soil is expressed in terms of equivalent water depth. When expressing precipitation or runoff depth in terms of equivalent water depth, the unit is millimeters. Therefore, when calculating the runoff volume of the soil grid, unit conversion is required for subsequent calculation and analysis.

[0045] Furthermore, hydrological data corresponding to each river grid is retrieved separately. Based on the water level data of each river grid in the current time period, a dynamic flow model is constructed. The cross-sectional flow data of the river is analyzed at different water levels to determine the cross-sectional flow Q corresponding to the real-time water level of each river grid in the current time period. j,t Its calculation is as follows: ; Among them, Q j,t Let H be the cross-sectional flow rate of channel grid j during time period t; H be the water level of channel grid j during time period t; h be the water depth; W(h) be the river width of channel grid j corresponding to water depth h during time period t; n be the channel roughness of channel grid j; r H denoted as , where is the hydraulic radius of channel grid j at water level H; and is the water surface gradient of channel grid j.

[0046] In this embodiment, during the analysis of the cross-sectional flow of each river grid, since the river water is in a flowing state, its water level has a dynamic range of change. Therefore, when calculating the real-time cross-sectional flow of each river grid, it is not a unique fixed value. Thus, in the actual analysis process, the cross-sectional flow of a single river grid is often calculated multiple times, and the average value is taken as the real-time cross-sectional flow of the current time period.

[0047] Furthermore, using each river grid as the association base point, the association of each soil grid is determined; by analyzing the runoff direction of each soil grid, the runoff path of the soil grid is determined; and the soil grids whose runoff paths are directed to the river grid are marked as associated.

[0048] Specifically, the runoff direction analysis of the soil grid, to obtain the runoff path of the soil grid, involves: T1. Take any soil grid as the center grid and search and determine the soil grids adjacent to it; T2. Obtain the elevation values ​​of the center points of the central grid and each adjacent soil grid, as well as the horizontal distance between the center point of the central grid and the center points of each adjacent soil grid; using the central grid as the gradient reference, calculate the slope gradient value F for each adjacent soil grid. m The analysis is performed; the calculation is as follows: ; Among them, F m Z represents the slope gradient between adjacent soil grid cells and the central grid cell, where m is the grid number. i Z represents the elevation value of the center point of the central grid cell numbered i; m L is the elevation value of the center point of the adjacent soil grid cell numbered m; m The horizontal distance between the center points of adjacent soil grids numbered m and the center grid numbered i; T3. Based on the directional distribution of adjacent soil grids of the central grid, the runoff direction of the central grid is determined by numbering; and based on the slope gradient values ​​between each adjacent soil grid and the central grid, the runoff direction of the central grid is analyzed, and the calculation is as follows: ; Among them, D i The runoff direction is assigned to the central grid cell numbered i; T4. The flow direction of each central grid is determined by continuously analyzing the flow direction, and the flow path of each central grid is determined by connecting the flow directions. It should be noted that the runoff direction numbering is used to indicate the runoff direction, so that the algorithm can directly output the slope gradient between the central grid and the adjacent soil grid, and the corresponding runoff direction.

[0049] In this embodiment, the runoff direction numbering can adopt the ESRI standard coding, where 1=East, 2=Southeast, 4=South, 8=Southwest, 16=West, 32=Northwest, 64=North, and 128=Northeast. It should be noted that if there is a maximum (Fm) < 0 in the slope gradient values ​​between a central grid and an adjacent soil grid, the grid is identified as a depression and needs to be marked as an abnormal confluence. When there are multiple equal maximum values ​​in the slope gradient between the central grid and an adjacent soil grid, the grid is output in the highest order according to the input order of the runoff direction number. It should also be noted that when determining the runoff path of the central grid, there may be soil grids along the runoff path that do not generate runoff, which may result in runoff consumption. However, this invention takes into account the scenario where, under precipitation conditions, the soil around the river generates runoff and undergoes confluence. On the one hand, each soil grid still receives water storage from precipitation, and on the other hand, the flow velocity of the runoff is much higher than the rate at which water infiltrates the soil. Therefore, in this case, the consumption of runoff during the flow process is not considered; it is only regarded as the confluence and transport of runoff generated by each soil grid.

[0050] Furthermore, soil graticles associated with each river graticle are integrated and a set of associated soil graticles, I, is constructed. A comprehensive runoff inflow analysis is performed on each associated soil graticle to determine the runoff inflow Q of each soil graticle associated with each river graticle at the current time point. j,I,t The calculation is as follows: ; Among them, v j,i The runoff velocity is the soil velocity associated with the j-th channel grid (i); v0 is the baseline velocity; β j,i The slope of soil grid i associated with river grid j; y is an empirical index; θ(ΣV j,i ) represents the correction factor for the runoff moisture content after runoff convergence to the runoff velocity of the soil grid numbered i associated with the j-th channel grid; T j,i The time required for runoff from soil grid i associated with river grid j to reach the associated river grid; i∈path(I) represents the soil grids traversed by the path of runoff from soil grid i to the associated river grid; L j,i,i+1 Q represents the path of runoff from soil grid i associated with river grid j through the next soil grid; j,I,t The real-time runoff inflow rate of each soil cell in the associated soil cell set associated with cell j (channel cell j) during time period t; T j,i →t represents the soil grids in the associated soil grid set associated with the river grid j, where runoff arrives at the river grid in real time during the current time period t; V j,i,t The runoff volume generated by soil grid number i associated with river grid number j during time period t.

[0051] In this embodiment, the reference flow velocity is manually set by the management personnel according to the utilization / coverage type of the soil grid; since the runoff generated by multiple soil grids will cause the runoff water content to increase after confluence, it will affect the runoff velocity, so a correction coefficient is introduced for compensation. Since there are multiple soil grid runoffs converging on the runoff path of each soil grid, the runoff velocities between different adjacent soil grids on the path are different. Therefore, it is necessary to analyze the runoff time between each soil grid on the runoff path in segments and finally sum them to obtain the time when the runoff from the corresponding soil grid reaches the river grid. Since the river grid is associated with multiple soil grids, and the runoff from different soil grids arrives at the river grid at different times, the runoff concentrations generated after their convergence are different. Therefore, it is necessary to calculate the runoff concentrations of the soil grid runoffs that arrive at the river grid at the same time.

[0052] Furthermore, by determining the current warning water level data of each river grid and importing it into the dynamic flow model, the warning value Q of the current river cross-sectional flow of each river grid is obtained. j,max ; Set the real-time cross-sectional flow value Q of each river grid to the current value. j,t The runoff sink volume Q of each associated soil grid j,I,t Sum the values ​​and compare the sum with the warning values; If greater than or equal to the warning value Q j,max If so, it is determined that the current river grid with number j is at risk of flooding; If it is less than the warning value Q j,max If so, then the current number j river channel grid is determined to be normal; Based on the comparison results, river grids with flood risk are located in the watershed simulation model and marked with risk warnings.

[0053] It should be noted that the warning value is used to alert that there may be a flood risk in the river channel, and does not mean that a flood will occur when the river channel flow reaches the warning value; therefore, when there is a flood risk, it can provide managers with accurate early location and timely remedial space.

[0054] Example 2: The present invention provides another technical solution: A flood early warning intelligent management system based on multi-source data fusion: The system includes a model simulation unit, a grid division unit, a runoff analysis unit, a river flow analysis unit, a correlation assessment unit, a confluence impact analysis unit, and a flood risk assessment unit.

[0055] Furthermore, the model simulation unit conducts data surveys on the watershed in the target area, collects watershed hydrological data, and constructs a watershed simulation model for the target area.

[0056] Furthermore, the grid division unit, based on the watershed simulation model, performs rasterization processing on the river area and soil area in the target area to obtain river raster and soil raster respectively.

[0057] Furthermore, the runoff analysis unit retrieves the hydrological data of each soil grid, analyzes the water storage status of each soil grid within the current time period through a preset monitoring period, and determines the runoff depth value of each soil grid within the current time period by comparing the score with the maximum water storage capacity of the soil grid; and obtains the runoff volume of each soil grid within the current time period by combining the area specification data of the soil grid.

[0058] Furthermore, the river flow analysis unit retrieves the hydrological data corresponding to each river grid, and constructs a dynamic flow model based on the water level data of each river grid in the current time period to analyze the river cross-sectional flow corresponding to the real-time water level of each river grid in the current time period.

[0059] Furthermore, the association assessment unit uses each river grid as the association base point to determine the association of each soil grid; it determines the runoff path of each soil grid by analyzing the runoff direction of each soil grid; and marks the soil grids whose runoff paths lead to the river grids as having an association status.

[0060] Furthermore, the runoff impact analysis unit coordinates the soil grates that are marked as associated with each river grates and constructs a set of associated soil grates; it performs a comprehensive runoff runoff analysis on each associated soil grates to determine the runoff runoff volume of each soil grates associated with each river grates at the current time point.

[0061] Furthermore, the flood risk assessment unit obtains the early warning value of the current river cross-sectional flow of each river grid; it sums the current real-time river cross-sectional flow value of each river grid with the current runoff inflow of each associated soil grid, and compares it with the early warning value; based on the comparison result, it locates the river grid with flood risk in the watershed simulation model and marks it with risk warning.

[0062] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A flood early warning intelligent management method based on multi-source data fusion, characterized in that: The method includes the following steps: Determine the target area, collect watershed hydrological data within the target area, and construct a watershed simulation model; In the watershed simulation model, the target area is subjected to raster classification processing to obtain river raster and soil raster; Based on the hydrological data of the soil grid, the real-time water storage status of each soil grid is analyzed, and the runoff generation data of each soil grid is determined based on the real-time water storage status. Analyze the real-time flow data of each river grid based on the hydrological data of the river grid; determine the soil grid associated with each river grid; and analyze the runoff collection data of each associated soil grid to the river grid based on the runoff generation data of the associated soil grid. Based on the runoff data of the soil grids associated with each of the aforementioned river grids, locate and alert river grids that are at risk of flooding.

2. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: In the watershed simulation model, hydrological data from each soil grid is retrieved, and the water storage status of each soil grid is analyzed within a preset monitoring period to obtain the water storage capacity P of the soil grid within the current period. i,t Its calculation is as follows: ; Among them, P i,t Let k be the water storage of soil grid cell i during time period t; i P is the water storage decline coefficient for soil grid number i. i,t-1 p represents the water storage of soil grid cell i during the time period t-1; i,t The amount of precipitation received by soil grid cell number i during time period t; The current water storage capacity of each soil grid is compared with its maximum water storage capacity, and the runoff depth R of each soil grid in the current time period is determined based on the comparison results. i,t The analysis is as follows: ; Among them, R i,t P represents the runoff depth of soil grid cell i during time period t; i,max The maximum water storage capacity of soil grid number i; Based on the runoff depth value R generated by each soil grid during the current time period i,t By combining the area specifications of the soil grid, the runoff volume V of each soil grid in the current time period is obtained. i,t Its calculation is as follows: ; Among them, V i,t Let A be the runoff volume generated by soil grid cell i during time period t; i Let i be the area of ​​the soil grid with number i.

3. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: Hydrological data corresponding to each river grid is retrieved separately. Based on the water level data of each river grid in the current time period, a dynamic flow model is constructed. The cross-sectional flow data of the river is analyzed at different water levels to determine the cross-sectional flow Q corresponding to the real-time water level of each river grid in the current time period. j,t Its calculation is as follows: ; Among them, Q j,t Let H be the cross-sectional flow rate of channel grid j during time period t; H be the water level of channel grid j during time period t; h be the water depth; W(h) be the river width of channel grid j corresponding to water depth h during time period t; n be the channel roughness of channel grid j; r H denoted as , where is the hydraulic radius of channel grid j at water level H; and is the water surface gradient of channel grid j.

4. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: Using each river grid as the association base point, the association of each soil grid is determined; It determines the runoff path of each soil grid by analyzing the runoff direction of each soil grid, and marks the soil grids whose runoff paths are directed to the river grid as associated. Specifically, the runoff direction analysis of the soil grid, to obtain the runoff path of the soil grid, involves: T1. Take any soil grid as the center grid and search and determine the soil grids adjacent to it; T2. Obtain the elevation values ​​of the center points of the central grid and each adjacent soil grid, as well as the horizontal distance between the center point of the central grid and the center points of each adjacent soil grid; using the central grid as the gradient reference, calculate the slope gradient value F for each adjacent soil grid. m The analysis is performed; the calculation is as follows: ; Among them, F m Z represents the slope gradient between adjacent soil grid cells and the central grid cell, where m is the grid number. i Z represents the elevation value of the center point of the central grid cell numbered i; m L is the elevation value of the center point of the adjacent soil grid cell numbered m; m The horizontal distance between the center points of adjacent soil grids numbered m and the center grid numbered i; T3. Based on the directional distribution of the adjacent soil grids of the central grid, the runoff direction of the central grid is determined by numbering; and based on the slope gradient values ​​between each adjacent soil grid and the central grid, the runoff direction of the central grid is analyzed, and the calculation is as follows: ; Among them, D i The runoff direction is assigned to the central grid cell numbered i; T4. The flow direction of each central grid is determined by continuously analyzing the flow direction, and the flow path of each central grid is determined by connecting the flow directions.

5. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: Soil graticules that are associated with each river graticule are aggregated and a set of associated soil graticules, I, is constructed. A comprehensive runoff inflow analysis is performed on each associated soil graticule to determine the runoff inflow Q of each soil graticule associated with each river graticule at the current time point. j,I,t Its calculation is as follows: ; Among them, v j,i The runoff velocity is the soil velocity associated with the j-th channel grid (i); v0 is the baseline velocity; β j,i The slope of soil grid i associated with river grid j; y is an empirical index; θ(ΣV j,i ) represents the correction factor for the runoff moisture content after runoff convergence to the runoff velocity of the soil grid numbered i associated with the j-th channel grid; T j,i The time required for runoff from soil grid i associated with river grid j to reach the associated river grid; i∈path(I) represents the soil grids traversed by the path of runoff from soil grid i to the associated river grid; L j,i,i+1 Q represents the path of runoff from soil grid i associated with river grid j through the next soil grid; j,I,t The real-time runoff inflow rate of each soil cell in the associated soil cell set associated with cell j (channel cell j) during time period t; T j,i →t represents the soil grids in the associated soil grid set associated with the river grid j, where runoff arrives at the river grid in real time during the current time period t; V j,i,t The runoff volume generated by soil grid number i associated with river grid number j during time period t.

6. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: By determining the current warning water level data for each river grid and importing it into the dynamic flow model, the warning value Q of the current river cross-sectional flow for each river grid is obtained. j,max ; Set the real-time cross-sectional flow value Q of each river grid to the current value. j,t The runoff sink volume Q of each associated soil grid j,I,t Sum the values ​​and compare the sum with the warning values; If it is greater than or equal to the warning value Q j,max If so, it is determined that the current river grid with number j is at risk of flooding; If it is less than the warning value Q j,max If so, then the current number j river channel grid is determined to be normal; Based on the comparison results, river grids with flood risk are located in the watershed simulation model and marked with risk warnings.

7. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: In a real-world scenario, a target area is selected, and data surveys are conducted on the watershed within that target area to collect watershed hydrological data. The target area's watershed is simulated in three dimensions using a simulation model, and combined with imported hydrological data, a watershed simulation model of the target area is constructed.

8. The intelligent flood early warning management method based on multi-source data fusion according to claim 1, characterized in that: Based on the watershed simulation model, the river area and soil area in the target region are rasterized to obtain river raster and soil raster respectively. When dividing the river channel grid and soil grid, the equal-length river channel grid is used. The river channel width is used as the reference, and the river channel length is divided at equal intervals to obtain the river channel grid. When dividing the soil into grids, the division range should be determined first. The soil area within the river channel should be divided into grids of the same size.

9. A system for implementing the intelligent flood early warning management method based on multi-source data fusion as described in any one of claims 1-8, characterized in that: The system includes a model simulation unit, a grid division unit, a runoff analysis unit, a river flow analysis unit, a correlation assessment unit, a confluence impact analysis unit, and a flood risk assessment unit; The model simulation unit conducts data surveys of the watershed in the target area and collects watershed hydrological data; And construct a watershed simulation model for the target area; The grid division unit, based on the watershed simulation model, performs rasterization processing on the river area and soil area in the target area to obtain river raster and soil raster respectively. The runoff analysis unit retrieves hydrological data from each soil grid, analyzes the water storage status of each soil grid within the current time period through a preset monitoring period, and determines the runoff depth value of each soil grid within the current time period by comparing the score with the maximum water storage capacity of the soil grid; and obtains the runoff volume of each soil grid within the current time period by combining the area specification data of the soil grid.

10. The system according to claim 9, characterized in that: The river flow analysis unit retrieves the hydrological data corresponding to each river grid, and constructs a dynamic flow model based on the water level data of each river grid in the current time period to analyze the river cross-sectional flow corresponding to the real-time water level of each river grid in the current time period. The association assessment unit uses each river grid as the association base point to determine the association between each soil grid; It determines the runoff path of each soil grid by analyzing the runoff direction of each soil grid, and marks the soil grids whose runoff paths are directed to the river grid as associated. The runoff impact analysis unit coordinates the soil gratings that are marked as associated with each river grating and constructs a set of associated soil gratings; it performs a comprehensive runoff runoff analysis on each associated soil grating to determine the runoff runoff volume of each soil grating associated with each river grating at the current time point. The flood risk assessment unit obtains the early warning value of the current river cross-sectional flow of each river grid; it sums the current real-time river cross-sectional flow value of each river grid with the current runoff inflow of each associated soil grid, and compares it with the early warning value; based on the comparison result, it locates the river grid with flood risk in the watershed simulation model and marks it with risk warning.

Images