Method and device for planning emergency transfer of people in flood risk area

Abstract

The application discloses a flood risk area personnel emergency transfer planning method and device, and belongs to the technical field of emergency transfer path planning, which comprises the following steps: constructing a road network map of a target area based on three-dimensional geographic data, and obtaining the relative elevation index of the roadbed of each path in the road network map; determining a sensitive path set based on the road network map and historical disaster data, wherein the sensitive path set comprises a river-crossing bridge and a wading path; performing hydrological simulation on the target area by using a hydrological model; calculating the passing cost of each path in the road network map by using a cost function; and determining the optimal transfer path of a first user to be transferred at time t according to the passing cost of each path in the road network map; wherein if a path in the sensitive path set is flooded at time t, the passing cost of the flooded sensitive path at time t is set to infinity. The method makes the planned personnel transfer route executable under extreme rainstorm working conditions.

Description

Technical Field

[0001] This invention relates to the field of emergency evacuation route planning technology, and in particular to a method and device for emergency evacuation planning of people in flood-prone areas. Background Technology

[0002] Mountainous and hilly areas are prone to flash floods, characterized by their sudden onset, high peak volume, and significant destructive power, making them a natural disaster that can easily cause casualties. Scientific and rational planning of emergency evacuation routes in flash flood scenarios is a crucial technical step in reducing disaster losses and ensuring the safety of residents.

[0003] Currently, research on flood relocation planning mainly focuses on macro-scale risk zoning or shortest path analysis based on two-dimensional maps.

[0004] However, in specific applications involving complex terrain and extreme rainstorms, traditional emergency evacuation route planning is mostly based on static road network models. But in actual flash flood scenarios, there are situations where rising water levels due to heavy rain render key nodes such as bridges and flooded roads impassable. In such cases, emergency evacuation routes planned using the above methods cannot be used. Summary of the Invention

[0005] This invention provides a method and apparatus for planning emergency evacuation of people in flood-prone areas, enabling the planned evacuation routes to remain feasible even under extreme rainstorm conditions. The technical solution includes at least the following components: Firstly, a method for planning emergency evacuation of people in flood-prone areas is provided, comprising: acquiring high-precision three-dimensional geographic data, historical disaster data, and a hydrological model of the target area; constructing a road network map of the target area based on the three-dimensional geographic data, and acquiring the roadbed relative elevation index of each path in the road network map, wherein the roadbed relative elevation index is used to describe the micro-geomorphology of each path; determining a set of sensitive paths based on the road network map and the historical disaster data, wherein the set of sensitive paths includes cross-river bridges and flooded paths; performing hydrological simulation of the target area using the hydrological model; calculating the passage cost of each path in the road network map using a cost function, wherein the cost function is implemented based on the hydrological simulation results at time t and the roadbed relative elevation index; determining the optimal evacuation path for the first user to be evacuated at time t based on the passage cost of each path in the road network map, wherein the optimal evacuation path is the evacuation path with the minimum cumulative passage cost; wherein, at each time point of the hydrological simulation, it is determined whether any path in the set of sensitive paths is flooded, and if any path in the set of sensitive paths is flooded at time t, then the passage cost of the flooded sensitive path is set to infinity at time t.

[0006] Optionally, obtaining the roadbed relative elevation index of each path in the road network map includes: obtaining the elevation data of the background terrain on both sides of the i-th path, wherein the background terrain on both sides is located in two directions perpendicular to the path where the i-th path is located; obtaining the roadbed relative elevation index of the i-th path based on the elevation data of the i-th path and the elevation data of the background terrain on both sides of the i-th path; and determining the micro-topographic morphology of the i-th path based on the relationship between the roadbed relative elevation index of the i-th path and the concave threshold and the convex threshold, wherein the micro-topographic morphology includes concave water catchment sections, convex embankment sections and flat sections.

[0007] Optionally, the cost function is expressed by the following formula:

[0008] in, Let be the cost function value of the m-th path at time t, representing the travel cost of the m-th path at time t. Let m be the physical length of the m-th path. Let be the roadbed relative elevation index for the m-th route. As a micro-topographic risk penalty, this applies when the micro-topographic features of the m-th path are convex embankment sections and flat sections. In the case where the micro-topographical morphology of the m-th path is a concave water catchment section , Risk coefficients corresponding to different types of micro-geomorphic features. For real-time hydrodynamic penalty items, Let be the water depth of the m-th path at time t. Let be the flow velocity of the m-th path at time t. and The results were obtained from the hydrological simulation at time t. For the historical risk memory item of the m-th path, in the case that there are historical hydrological disasters on the m-th path. Used to impose punishment These are the weighting coefficients for different penalty items.

[0009] Optionally, determining the optimal transfer path for the first user to be transferred at time t based on the travel cost of each path in the road network map includes: obtaining a set of emergency resettlement points, which includes multiple emergency resettlement points used to resettle users to be transferred; using the location of the first user to be transferred as the starting point, each emergency resettlement point in the set of emergency resettlement points as the ending point, and the cumulative travel cost of the transfer path as the objective function, using an optimization algorithm to search for the optimal transfer path for the first user to be transferred from the road network map.

[0010] Optionally, the method further includes: if there are paths in the sensitive path set that are flooded at time t, performing a connectivity check on the road network map and removing road segments that cannot be connected to any emergency resettlement points due to the flooded paths.

[0011] Optionally, the method further includes: obtaining the cumulative passage cost of the optimal transfer path for each person to be transferred at time t, and filtering out the persons to be transferred whose cumulative passage cost is infinite, to obtain the predicted set of trapped persons at time t; sending a pre-emptive forced transfer command to the set of trapped persons at time t, so that the persons in the set of trapped persons are transferred to the emergency resettlement point before time t.

[0012] Secondly, a flood-prone area emergency evacuation planning device is also provided, comprising: a first acquisition module for acquiring high-precision three-dimensional geographic data, historical disaster data, and a hydrological model of the target area; a second acquisition module for constructing a road network map of the target area based on the three-dimensional geographic data, and acquiring the roadbed relative elevation index of each path in the road network map, wherein the roadbed relative elevation index is used to describe the micro-topography of each path; a sensitive path determination module for determining a set of sensitive paths based on the road network map and the historical disaster data, wherein the set of sensitive paths includes river-crossing bridges and flooded paths; and a hydrological simulation module for using the hydrological model to simulate the target area. The system performs hydrological simulation; a cost calculation module is used to calculate the passage cost of each path in the road network map using a cost function, which is based on the hydrological simulation results at time t and the roadbed relative elevation index; a path planning module is used to determine the optimal transfer path for the first user to be transferred at time t based on the passage cost of each path in the road network map, where the optimal transfer path is the transfer path with the minimum cumulative passage cost; wherein, at each time point in the hydrological simulation, it is determined whether there is a path in the sensitive path set that is flooded, and if there is a path in the sensitive path set that is flooded at time t, then the passage cost of the flooded sensitive path is set to infinity at time t.

[0013] Optionally, the second acquisition module is further configured to acquire elevation data of the background terrain on both sides of the i-th path, wherein the background terrain on both sides is located in two directions perpendicular to the path in which the i-th path is located; acquire the roadbed relative elevation index of the i-th path based on the elevation data of the i-th path and the elevation data of the background terrain on both sides of the i-th path; and determine the micro-topographic morphology of the i-th path based on the relationship between the roadbed relative elevation index of the i-th path and the concave threshold and the convex threshold, wherein the micro-topographic morphology includes concave water catchment sections, convex embankment sections and flat sections.

[0014] Optionally, in the cost calculation module, the cost function is expressed by the following formula:

[0015] in, Let be the cost function value of the m-th path at time t, representing the travel cost of the m-th path at time t. Let m be the physical length of the m-th path. Let be the roadbed relative elevation index for the m-th route. As a micro-topographic risk penalty, this applies when the micro-topographic features of the m-th path are convex embankment sections and flat sections. In the case where the micro-topographical morphology of the m-th path is a concave water catchment section , Risk coefficients corresponding to different types of micro-geomorphic features. For real-time hydrodynamic penalty items, Let be the water depth of the m-th path at time t. Let be the flow velocity of the m-th path at time t. and The results were obtained from the hydrological simulation at time t. For the historical risk memory item of the m-th path, in the case that there are historical hydrological disasters on the m-th path. Used to impose punishment These are the weighting coefficients for different penalty items.

[0016] Optionally, the route planning module is further configured to obtain a set of emergency resettlement points, which includes multiple emergency resettlement points used to accommodate users to be transferred; taking the location of the first user to be transferred as the starting point, each emergency resettlement point in the set of emergency resettlement points as the ending point, and using the cumulative travel cost of the transfer path as the objective function, an optimization algorithm is used to search for the optimal transfer path of the first user to be transferred from the road network map.

[0017] Optionally, the route planning module is also used to perform connectivity verification on the road network map when there are submerged paths in the sensitive path set at time t, and to remove road segments that cannot be connected to any emergency resettlement points due to the submerged paths.

[0018] Optionally, the device further includes: an instruction issuing module, which is used to obtain the cumulative passage cost of the optimal transfer path for each person to be transferred at time t, and to filter out the persons to be transferred whose cumulative passage cost is infinite, thereby obtaining the predicted set of trapped persons at time t; and to send a pre-emptive forced transfer instruction to the set of trapped persons at time t, so that the persons in the set of trapped persons are transferred to the emergency resettlement point before time t.

[0019] Thirdly, a computer device is also provided, comprising: a memory and a processor, wherein the memory stores at least one computer program, the at least one computer program being loaded and executed by the processor to perform the emergency evacuation planning method for people in flood-prone areas described in the above embodiments.

[0020] Fourthly, a computer-readable storage medium is also provided, wherein at least one computer program is stored in the computer-readable storage medium, the at least one computer program being loaded and executed by a processor to perform the emergency evacuation planning method for people in flood-prone areas described in the above embodiments.

[0021] Fifthly, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0022] The beneficial effects of the technical solution provided by this invention include at least the following: In this embodiment, the relative elevation index of the roadbed enables the emergency evacuation planning process to distinguish different micro-topographical features, avoiding the "taking the main road only to be flooded" phenomenon caused by traditional planning based on macroscopic three-dimensional models, and significantly improving the physical safety of evacuation routes. By setting the passage cost of the flooded sensitive paths at time t to infinity, the algorithm mechanism eliminates the fatal risk of guiding personnel to wade across bridges under extreme conditions, ensuring the feasibility of evacuation routes planned under extreme rainstorm conditions. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this embodiment, the accompanying drawings used in the description of the embodiment will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart of an exemplary embodiment of the present invention for planning emergency evacuation of people in flood-prone areas is shown. Figure 2 These are schematic diagrams of various micro-landforms; Figure 3 This is a schematic diagram of the optimal evacuation route planned under normal and extreme rainstorm conditions; Figure 4 This diagram illustrates the structure of an emergency evacuation planning device for people in flood-prone areas provided by an exemplary embodiment of the present invention. Figure 5This is a schematic diagram of the structure of a computer device provided in an exemplary embodiment of the present invention. Detailed Implementation

[0025] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms “first,” “second,” “third,” and similar terms used in this patent application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “a” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising” or “including” and similar terms mean that the elements or objects preceding “comprising” or “including” encompass the elements or objects listed following “comprising” or “including” and their equivalents, and do not exclude other elements or objects. The terms “connected” or “linked” and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0027] Example 1.

[0028] Figure 1 A flowchart illustrating an exemplary embodiment of the present invention provides a method for emergency evacuation planning of people in flood-prone areas, which can be executed by a computer device. See also... Figure 1 The method includes: In step 101, high-precision three-dimensional geographic data, historical disaster data, and hydrological models of the target area are acquired.

[0029] The target area is a flood-prone area, such as a flood-prone area that requires emergency evacuation planning.

[0030] The high-precision 3D geographic data of the target area includes a digital elevation model (DEM), a digital surface model (DSM), and a digital orthophoto (DOM). For example, a UAV-borne lidar and oblique photogrammetry system can be used to acquire high-precision point cloud and imagery of the target area, which can then be registered, filtered, and resampled to obtain high-precision 3D geographic data of the target area.

[0031] Historical disaster data for the target area includes complete records of rainstorms and floods that have occurred in the target area within the past 10 years, including rainfall events, changes in river level / flow, disaster losses, and historical flood records.

[0032] The hydrological model of the target area is used to simulate the rainfall process in the target area. The hydrological model of the target area can be obtained using any of the relevant technologies. For example, firstly, the DOM data of the target area is classified into land features (forest, farmland, water bodies, buildings, roads, etc.) based on a deep learning algorithm. Then, according to the roughness reference table in Hydraulics, different Manning coefficients n are assigned to different land feature patches (e.g., farmland n=0.04, forest n=0.06, grassland n=0.05), thus constructing a heterogeneous surface roughness field.

[0033] Next, the collected historical disaster data and corresponding flood level data from multiple events are divided into two datasets according to a certain ratio (e.g., "70%:30%)": a calibration set and a validation set. Then, a hydrological model is constructed using a two-dimensional shallow water equation model. Historical measured rainfall data from the calibration set are input into the hydrological model, and unsteady flow numerical simulations are performed using topographic data, Manning coefficient, historical flood levels, and floodmark data to calibrate the hydrological model's parameters. After obtaining the optimal parameter set through calibration, the calibrated hydrological model is validated using data from the validation set. Only when all evaluation indicators in the validation set meet the preset accuracy requirements (e.g., Nash efficiency coefficient greater than 0.7) is the hydrological model considered successfully constructed; otherwise, the model input and boundary conditions are readjusted to debug the model.

[0034] In step 102, a road network map of the target area is constructed based on three-dimensional geographic data, and the relative elevation index of the roadbed for each path in the road network map is obtained.

[0035] The roadbed relative elevation index is used to describe the micro-topographic features of each route.

[0036] In this embodiment, step 102 essentially transforms the traditional two-dimensional road network into a three-dimensional risk road network with microscopic elevation attributes based on differential geometric analysis. Then, by calculating the concavity / convexity of the road relative to the background terrain (roadbed relative elevation index) point by point, the risk of rainstorm flooding is quantified.

[0037] Optionally, constructing a road network map of the target area includes: reading road network vector centerline data based on high-resolution DOM imagery of the target area, and discretizing continuous road lines into an ordered set of road network nodes at a fixed sampling step size (e.g., 5.0m). For the i-th road network node In other words, Let x be the x-coordinate of the i-th road network node. Let be the ordinate of the i-th road network node. Connect adjacent road network nodes in the order of road lines and road network nodes to obtain a road network map of the target area consisting of multiple road network nodes and road network edges. Each road network edge has two road network nodes at its two ends.

[0038] Optionally, step 102 includes steps 1021 to 1023 as follows.

[0039] Step 1021: Obtain the elevation data of the background terrain on both sides of the i-th path.

[0040] The background terrain on both sides is located in two directions perpendicular to the path containing the i-th path. The i-th path is any road network edge in the road network map.

[0041] In step 1021, we can first obtain the tangential vector of the i-th path and the normal vector perpendicular to the road direction, where the normal vector is calculated using a rotation matrix. Then, the directions of the background terrain on both sides are the directions parallel to the normal vector and the directions opposite to the normal vector. Let the direction parallel to the normal vector be the first direction, and the direction opposite to the normal vector be the second direction.

[0042] At this point, we can use the i-th path. Centered on a point, sampling points are extended M meters (M is typically 3.0m to 5.0m to cover the roadbed slope or ditch area) in both the first and second directions. This allows for the determination of a first sampling point in the first direction and a second sampling point in the second direction for the background terrain on both sides. Elevation data (i.e., absolute elevation values) of the first and second sampling points, as well as the i-th path, can be obtained from the 3D geographic data. The elevation data of the first and second sampling points constitute the elevation data of the background terrain on both sides.

[0043] Step 1022: Based on the elevation data of the i-th path and the elevation data of the background terrain on both sides of the i-th path, obtain the roadbed relative elevation index of the i-th path.

[0044] Optionally, the roadbed relative elevation index is calculated using formula (1).

[0045] (1) In formula (1), Let be the roadbed relative elevation index for the i-th path. Here is the elevation data for the i-th path. Let represent the elevation data of the first sampling point of the i-th path, and let represent the elevation data of the background terrain of the i-th path in the first direction. Let be the elevation data of the second sampling point of the i-th path, and let represent the elevation data of the background terrain of the i-th path in the second direction.

[0046] The positive or negative sign can indicate whether the road surface is convex or concave in a local area. For road sections with missing topographic data on one side or abrupt elevation changes (such as road sections near cliffs), one-sided comparison or neighborhood interpolation methods are used for correction.

[0047] The roadbed relative elevation index can eliminate the background of large-scale terrain slope, retaining only the local relative morphology, and reflecting the micro-geomorphic features of the path.

[0048] The precision of the 3D geographic data used in related technologies is generally no more than 12.5m by 12.5m per pixel (rural roads are generally less than 12m wide, making it impossible to know the elevations at both ends and in the middle of the road), resulting in the inability to obtain detailed features. If conventional precision 3D geographic data is used, its highest precision does not exceed 12.5m. Given the limited width of rural roads, the effect of elevation differences, such as the roadbed relative elevation index, is not significant on these low-precision 3D geographic data and cannot reflect micro-topographical features. In this embodiment, by acquiring high-precision 3D geographic data of the target area, the precise elevation data of each sampling point and path can be accurately extracted. This allows for the accurate calculation of the roadbed relative elevation index, which is then used to reflect the micro-topographical features of the path.

[0049] Step 1023: Determine the micro-topography of the i-th path based on the relationship between the relative elevation index of the roadbed and the concave and convex thresholds.

[0050] Micro-topography includes concave water catchment sections, convex embankment sections, and flat sections. Figure 2 These are schematic diagrams of various micro-landforms. They illustrate the different micro-landform morphologies... , , The relative differences between them are relatively large, so the roadbed relative elevation index calculated using this elevation value can be used to reflect different micro-topographical features.

[0051] In this embodiment, both the concave threshold and the convex threshold are preset empirical values. For example, the concave threshold is -0.15m and the convex threshold is +0.20m.

[0052] When the relative elevation index of the roadbed of a certain route is less than the concave threshold, it indicates that the route is a concave water catchment section; when the relative elevation index of the roadbed of a certain route is greater than the convex threshold, it indicates that the route is a convex embankment section. When the relative elevation index of the roadbed of a certain route is greater than or equal to the concave threshold and less than or equal to the convex threshold, it indicates that the route is a flat section.

[0053] In this embodiment, different risk coefficients are assigned to different micro-topographical features. (For subsequent calculation of traffic costs), since the risk of road blockage only exists in concave catchment sections, only the risk coefficient for concave catchment sections is considered. Meaningful, the risk factor of other micro-topographical features, such as convex embankment sections and flat sections, is also considered. For example, the risk factor of a concave water catchment section. This indicates that the concave water collection section poses a significant risk.

[0054] In step 103, a set of sensitive routes is determined based on the road network map and historical disaster data.

[0055] The sensitive path set includes bridges across rivers and flooded paths.

[0056] Each road edge in the road network map has attributes that describe the type of each road segment. Based on these attributes, bridges across rivers and flooded paths can be filtered out.

[0057] Historical disaster data is used to identify low-lying road sections along rivers within the flood inundation range of historical disasters. Road network edges belonging to these low-lying river sections in the road network map are also stored in the sensitive path set. In acquiring low-lying river sections, in addition to identification from historical disaster data, the flood control planning standards for mountain torrents and rivers in the target area (such as a 20-year return period) can be used to simulate the evolution of rainstorms and floods at specific frequencies using hydrological models. This yields road sections in the road network map that will be submerged; these sections are also low-lying river sections.

[0058] In step 104, a hydrological model is used to simulate the hydrology of the target area.

[0059] The hydrological simulation in this embodiment is used to simulate the hydrological evolution of a target area from the current moment to a future period (this future period must include the rainstorm phase, and the hydrological simulation process can even continue until the rain stops). The simulation needs to include, but is not limited to, changes in water level (depth), flow velocity, and flood spread range at each location on the road network map at various times during this period. The hydrological simulation uses the predicted rainfall changes for a future period as input.

[0060] Based on the results of hydrological simulation, the hydrological evolution trend from the current moment to a certain period in the future can be obtained. Based on this, emergency resettlement sites that will not be flooded, can avoid secondary disasters, and have sufficient capacity can be selected, thus obtaining a set of emergency resettlement sites.

[0061] Optionally, the following four steps can be used to obtain a set of emergency shelter locations.

[0062] The first step is to obtain the total number of people to be evacuated in the target area based on the hydrological simulation results.

[0063] The people to be evacuated here are those located within the flood zone of the target area.

[0064] In implementation, it is first necessary to use data such as grid management data, on-site surveys, and mobile phone signaling to spatially locate the distribution of people in the potential flood hazard areas on both sides of the river, and construct a candidate set that includes the location of the head of household, the number of permanent residents, and the characteristics of vulnerable groups (such as the elderly and children).

[0065] Then, based on the flood control planning standards for mountain torrents and rivers within the target area (such as a 20-year return period), a hydrological model is used to simulate the evolution of rainstorms and floods at a specific frequency, obtaining the maximum inundation range of the target area. Personnel within this maximum inundation range from the aforementioned candidate personnel set are considered to be evacuated. This yields the total number of personnel to be evacuated within the target area.

[0066] The second step is to eliminate the area with the largest inundation range and the area prone to geological disasters in the target area to obtain the initial safe zone.

[0067] Before proceeding to the second step, it is necessary to identify areas prone to geological disasters. This can be done through methods such as surveys, discussions, and field measurements. These areas include the impact zones of potential geological hazard sites such as landslides, collapses, and debris flows.

[0068] The target area, the maximum inundation range, and the geological hazard-prone areas can be established as three different layers under the same coordinate system. Then, in GIS software, spatial difference calculations are performed on the global geographic space of the target area, the maximum inundation range, and the geological hazard-prone areas to obtain the initial set of safe zones. This process is represented by formula (2).

[0069] (2) In formula (2), For the initial set of safe zones, The target area is the entire geographical region. This represents the maximum inundation range of the target area as simulated by the hydrological model. The target area is a geologically hazardous zone.

[0070] The third step is to calculate the suitability score of each window unit in the initial safe zone, and determine the candidate safe zones based on the suitability scores to obtain the candidate safe zone set.

[0071] The suitability score is derived from a combination of local slope and site flatness. In calculating the suitability score, a sliding window algorithm is used to divide the initial safe zone into multiple window units, and the local slope and site flatness of each window unit are calculated, thereby calculating the suitability score for each window unit. Window units with suitability scores greater than a suitability threshold are selected as candidate safe zones, resulting in a set of candidate safe zones. The suitability threshold is a preset value.

[0072] Local slope reflects the degree of inclination of the land parcel; slopes that are too steep are unsuitable for setting up tents or for personnel to stay on. The local slope of each window unit can be obtained directly from the three-dimensional geographic data.

[0073] Site flatness reflects the degree of surface roughness and fragmentation. Site flatness is calculated using the standard deviation of elevation within a window unit.

[0074] Based on this, the suitability score of any window unit is represented by the following formula (3).

[0075] (3) In formula (3), For suitability score, The maximum allowable slope threshold, This refers to the local slope of the window unit. The flatness sensitivity coefficient needs to be calibrated. The flatness of the site for the window unit. , For the preset normalized weights, , The fourth step is to obtain a set of emergency shelters from the set of candidate safe zones.

[0076] In this embodiment, the fourth step includes: performing connectivity analysis on each window unit in the candidate safety zone set, merging adjacent window units, and ensuring that the merged regions are not connected to each other. Each region is a candidate emergency resettlement point, resulting in multiple independent candidate emergency resettlement points.

[0077] Calculate the effective flat area of ​​each candidate emergency resettlement site. The effective flat area of ​​each candidate emergency resettlement site should meet the following constraints. .in, Let the effective flat area be the j-th candidate emergency resettlement site. Let be the maximum number of people that can be accommodated at the j-th candidate emergency resettlement site. Minimum area occupied per person (e.g.) / person). Since the effective flat area and minimum per capita area of ​​each candidate emergency resettlement site are fixed values, the maximum number of people that can be accommodated at each candidate emergency resettlement site can be calculated, i.e. .

[0078] Finally, calculate the Euclidean distance or path distance from the centroid of each candidate emergency shelter to the nearest road network node. Use this as supplementary reference and give priority to it when selecting the final emergency resettlement site. Areas smaller than the traffic coverage radius (e.g., 500m).

[0079] Finally, based on fitness score, maximum settlement population, and To determine the assembly point for emergency shelters.

[0080] For example, fitness scores, maximum accommodable population, and First, normalization is performed, followed by weighted fusion to obtain the first score for each candidate emergency resettlement site. These sites are then sorted according to their first scores, and the top K sites in the sorted list are selected as the set of emergency resettlement sites. The fitness score for each candidate emergency resettlement site is the average of the fitness scores of all grid cells within that site. Furthermore, the weights of these three indicators can be set according to actual needs. For example, when the total number of people to be relocated is large, the weight of the maximum resettlement population can be set to the maximum, or by default, the weights of all three can be set to the same.

[0081] The database can store detailed attribute parameters for each emergency shelter in the set of emergency shelters: shelter name, shelter coordinates, and maximum capacity for accepting people.

[0082] In step 105, the cost of each path in the road network map is calculated using a cost function.

[0083] The cost function is realized based on the hydrological simulation results at time t and the roadbed relative elevation index.

[0084] In this process, at each moment of the hydrological simulation, it is determined whether there is a path in the sensitive path set that is submerged. If there is a path in the sensitive path set that is submerged at time t, then the passage cost of the submerged sensitive path is set to infinity at time t.

[0085] Optionally, the cost function of the m-th path in the road network diagram is represented by the following formula (4). Here, a path in the road network diagram refers to a road network edge.

[0086] (4) In formula (4), Let be the cost function value of the m-th path at time t, representing the travel cost of the m-th path at time t. Let m be the physical length of the m-th path. Let be the roadbed relative elevation index for the m-th route. As a micro-topographic risk penalty, this applies when the micro-topographic features of the m-th path are convex embankment sections and flat sections. In the case where the micro-topographical morphology of the m-th path is a concave water catchment section , Risk coefficients corresponding to different types of micro-geomorphic features. For real-time hydrodynamic penalty items, Let be the water depth of the m-th path at time t. Let be the flow velocity of the m-th path at time t. and The results were obtained from the hydrological simulation at time t. For the historical risk memory item of the m-th path, in the case that there are historical hydrological disasters on the m-th path. Used to impose penalties, for example, in the case of a historical hydrological disaster on the m-th path. In the case that the m-th path does not have any historical hydrological disasters . The weighting coefficients for different penalty terms and .

[0087] Under normal circumstances, the cost function value of each path can be calculated at each time step. However, for any path, if it is a sensitive path and is determined to be flooded at time t, the passage cost of that path is set to infinity. This allows the subsequent algorithm to automatically abandon flooded paths during optimization. Therefore, if a path is determined to be flooded at time t, it is not necessary to use the above formula (4) to calculate the passage cost; instead, the passage cost can be set to infinity directly. In this way, the possibility of passing through the road segment can be forcibly blocked in the calculation logic without requiring real-time modification of the road network structure in the three-dimensional geographic data.

[0088] Here, each path in the sensitive path set corresponds to a flooding threshold (such as water level or rainfall). If, at time t, the water level or rainfall of a sensitive path exceeds the flooding threshold during the hydrological simulation, then the path is determined to be flooded at time t and cannot be passed at time t.

[0089] The aforementioned hydrological simulation process is used to simulate hydrological conditions from the current time to a future period, thus resulting in hydrological simulation results at multiple time points. If a path is submerged, then in the road network map, this path and all subsequent paths connected only to it are unreachable. In this case, these unreachable paths can be removed from the road network map, reducing the difficulty of subsequent optimization. For example, if there are submerged paths in the sensitive path set at time t, the connectivity of the road network map is checked, and road segments that cannot connect to any emergency resettlement points due to submersion are removed. It should be noted that the removal of nodes here essentially involves real-time modification of the fitted road network map, not modification of the road network structure in the 3D geographic data. Therefore, it only involves changes in the simulation process and algorithm optimization process, and does not involve changes to the actual 3D model.

[0090] The degree of inundation along each route may change depending on the hydrological simulation results at each moment (because the inundated area will gradually expand as rainfall continues). If the above-mentioned operation of eliminating inaccessible routes is performed, the road network map at each moment may be different. When planning routes in the future, it is also necessary to simulate based on the road network map at different times to ensure that the calculated route absolutely avoids the identified high-risk bridges and inundated road sections.

[0091] In step 106, the optimal transfer path for the first user to be transferred at time t is determined based on the travel cost of each path in the road network map.

[0092] The optimal transfer path is the transfer path with the minimum cumulative travel cost. The first user to be transferred can be any user.

[0093] The aforementioned step 104 has already obtained the set of emergency resettlement points. Based on this, step 106 includes: taking the location of the first user to be transferred as the starting point, each emergency resettlement point in the set of emergency resettlement points as the ending point, and using the cumulative travel cost of the transfer path as the objective function, an optimization algorithm is used to search for the optimal transfer path of the first user to be transferred from the road network map.

[0094] The road network diagrams at different times described above are essentially graph-theoretic structures. Therefore, graph optimization search algorithms can be used to find candidate paths from the first user to be transferred to each emergency shelter. These candidate paths are also the paths with the lowest travel costs. For example, the path with the lowest cumulative travel cost from the first user to any emergency shelter is a candidate path. The optimization algorithm can obtain candidate paths from the first user to each emergency shelter, and then select the path with the lowest cumulative travel cost among these candidate paths as the optimal transfer path.

[0095] It should be noted that since the road network map is different at different times, the optimal transition path is also different at different times. Therefore, the optimization algorithm will output the optimal transition path at each time as guidance.

[0096] For example, the optimization algorithm is an improved A-star heuristic search algorithm or Dijkstra's algorithm.

[0097] Figure 3 This diagram illustrates the optimal transfer path planned under normal and extreme rainstorm conditions. It shows that under normal conditions, the bridge is not submerged, and the optimal transfer path includes the bridge. Under extreme rainstorm conditions, the bridge is submerged. During the optimization algorithm's search process, because the passage cost of the submerged path was set to infinity in step 105, and a high penalty was imposed on concave roads in the cost function, the optimization algorithm automatically discards easily submerged paths such as "bridge shortcuts" and "low-lying ditches," generating an asymmetric safe path that bypasses higher ground on the same bank or a convex embankment.

[0098] Optionally, a rolling update cycle (e.g., 10 minutes) can be set for periodic calibration. At the current moment, the system automatically accesses the latest meteorological monitoring data and future short-term rainfall forecast data. Next, the relative deviation rate between the forecast rainfall of the previous period and the current measured rainfall is calculated. If the relative deviation rate is greater than a deviation threshold (e.g., 20%), it indicates that there is an error in the meteorological forecast data. In this case, a recalculation is triggered, and steps 104 to 106 are re-executed. Generally, it is not necessary to re-execute the emergency resettlement site set planning process in step 104, unless the original resettlement site becomes unavailable. This can solve the scheduling problem caused by the uncertainty of meteorological forecasts.

[0099] Furthermore, the difficulty and priority of rescuing each person to be evacuated can be determined by the traffic resistance of their optimal transfer path. Since the optimal transfer path for each person to be evacuated may change at different times, if the cumulative traffic cost of the optimal transfer path for any person to be evacuated at any future time is infinite, it indicates that the person to be evacuated is trapped and unable to evacuate independently, and must wait for rescue. In this case, a warning message can be sent to the person to be evacuated in advance to prevent them from becoming trapped.

[0100] This embodiment uses time t as an example (where time t is a future time, not the current time). However, for each future predicted time other than time t, the following operation is required to determine the trapped personnel and issue an early warning. Although the bridge is not physically broken at this time, the algorithm predicts that it will soon fail and become impassable. Therefore, the current time window is used to force personnel to evacuate in advance, reflecting a "preemptive" strategy. This method includes the following two steps: The first step is to obtain the cumulative travel cost of the optimal transfer path for each person to be transferred at time t, and then filter out the persons to be transferred whose cumulative travel cost is infinite, thus obtaining the predicted set of trapped persons at time t.

[0101] The second step is to send a pre-emptive forced transfer order to the group of trapped people at time t, so that the people in the group of trapped people can be transferred to the emergency resettlement point before time t.

[0102] The pre-emergency forced evacuation instruction has the highest sending priority. For example, the pre-emergency forced evacuation instruction is: Heavy rain may cause the bridge to be impassable, please evacuate immediately, and please arrive at the planned emergency shelter before time t.

[0103] In addition, the emergency forced evacuation order also includes the optimal evacuation route for each trapped person at the current moment. Under normal circumstances, the current moment is a light rain condition and there are no trapped persons. All persons to be evacuated only need to evacuate within the specified time to avoid becoming trapped persons.

[0104] For other individuals awaiting evacuation outside the designated group of trapped personnel, they can evacuate to the emergency shelter on foot. Therefore, it is sufficient to provide them with the optimal evacuation route for each time period and instructions for evacuation as needed. An example of an instruction for evacuation as needed would be: "Please evacuate in an orderly manner according to the planned route, which avoids areas that may be flooded."

[0105] Finally, using mobile communication networks or emergency broadcasting systems, the pre-evacuation or on-demand evacuation instructions for each person to be evacuated are precisely pushed to the corresponding mobile terminals of the individuals to be evacuated. Simultaneously, the system continuously receives real-time location data transmitted from user terminals and calculates the offset distance between the user's real-time location and the current optimal evacuation path. If the offset distance exceeds a safety threshold (e.g., 50m), the system determines that the user has "lost their way," immediately generates a correction alarm, and pushes it to on-site command personnel, thus forming a complete dynamic closed loop of "perception-decision-action-feedback."

[0106] The scheduling problems caused by the lag in personnel movement can be solved by using the aforementioned preemptive forced transfer orders or on-demand transfer orders.

[0107] In this embodiment, the relative elevation index of the roadbed enables the emergency evacuation planning process to distinguish different micro-topographical features, avoiding the "taking the main road only to be flooded" phenomenon caused by traditional planning based on macroscopic three-dimensional models, and significantly improving the physical safety of evacuation routes. By setting the passage cost of the flooded sensitive paths at time t to infinity, the algorithm mechanism eliminates the fatal risk of guiding personnel to wade across bridges under extreme conditions, ensuring the feasibility of evacuation routes planned under extreme rainstorm conditions.

[0108] Example 2.

[0109] The following are device embodiments of this application. For details not described in detail in the device embodiments, please refer to the above method embodiments.

[0110] Figure 4 A schematic diagram of the structure of an emergency evacuation planning device for people in flood-prone areas provided by an exemplary embodiment of the present invention is shown. See also Figure 4 The flood-prone area emergency evacuation planning device 400 includes: a first acquisition module 401, a second acquisition module 402, a sensitive path determination module 403, a hydrological simulation module 404, a cost calculation module 405, a path planning module 406, and an instruction issuance module 407.

[0111] The first acquisition module 401 is used to acquire high-precision three-dimensional geographic data, historical disaster data and hydrological models of the target area; The second acquisition module 402 is used to construct a road network map of the target area based on three-dimensional geographic data, and to acquire the roadbed relative elevation index of each path in the road network map. The roadbed relative elevation index is used to describe the micro-topography of each path. The sensitive path determination module 403 is used to determine the set of sensitive paths based on the road network map and historical disaster data. The set of sensitive paths includes cross-river bridges and flooded paths. Hydrological simulation module 404 is used to perform hydrological simulation of the target area using a hydrological model; The cost calculation module 405 is used to calculate the passage cost of each path in the road network diagram using a cost function. The cost function is based on the hydrological simulation results at time t and the roadbed relative elevation index. The route planning module 406 is used to determine the optimal transfer path for the first user to be transferred at time t based on the toll cost of each path in the road network map. The optimal transfer path is the transfer path with the minimum cumulative toll cost. In this process, at each moment of the hydrological simulation, it is determined whether there is a path in the sensitive path set that is submerged. If there is a path in the sensitive path set that is submerged at time t, then the passage cost of the submerged sensitive path is set to infinity at time t.

[0112] Optionally, the second acquisition module 402 is further configured to acquire the elevation data of the background terrain on both sides of the i-th path, wherein the background terrain on both sides is located in two directions perpendicular to the path in which the i-th path is located; based on the elevation data of the i-th path and the elevation data of the background terrain on both sides of the i-th path, acquire the roadbed relative elevation index of the i-th path; and based on the relationship between the roadbed relative elevation index of the i-th path and the concave threshold and the convex threshold, determine the micro-topographic morphology of the i-th path, wherein the micro-topographic morphology includes concave water catchment sections, convex embankment sections and flat sections.

[0113] Optionally, in the cost calculation module 405, the cost function is expressed using the following formula:

[0114] in, Let be the cost function value of the m-th path at time t, representing the travel cost of the m-th path at time t. Let m be the physical length of the m-th path. Let be the roadbed relative elevation index for the m-th route. As a micro-topographic risk penalty, this applies when the micro-topographic features of the m-th path are convex embankment sections and flat sections. In the case where the micro-topographical morphology of the m-th path is a concave water catchment section , Risk coefficients corresponding to different types of micro-geomorphic features. For real-time hydrodynamic penalty items, Let be the water depth of the m-th path at time t. Let be the flow velocity of the m-th path at time t. and The results were obtained from the hydrological simulation at time t. For the historical risk memory item of the m-th path, in the case that there are historical hydrological disasters on the m-th path. Used to impose punishment These are the weighting coefficients for different penalty items.

[0115] Optionally, the route planning module 406 is also used to obtain a set of emergency resettlement points, which includes multiple emergency resettlement points used to resettle users to be transferred; taking the location of the first user to be transferred as the starting point, each emergency resettlement point in the set of emergency resettlement points as the ending point, and the cumulative travel cost of the transfer path as the objective function, an optimization algorithm is used to search for the optimal transfer path of the first user to be transferred from the road network map.

[0116] Optionally, the route planning module 406 is also used to perform connectivity verification on the road network map when there are flooded paths in the sensitive path set at time t, and to remove road segments that cannot be connected to any emergency resettlement points due to flooded paths.

[0117] Optionally, the device further includes: an instruction issuing module 407, which is used to obtain the cumulative passage cost of the optimal transfer path for each person to be transferred at time t, and to filter out the persons to be transferred whose cumulative passage cost is infinite, thereby obtaining the predicted set of trapped persons at time t; and to send a pre-emptive forced transfer instruction to the set of trapped persons at time t, so that the persons in the set of trapped persons are transferred to the emergency resettlement point before time t.

[0118] It should be noted that the above-described embodiment of the flood-prone area emergency evacuation planning device is only illustrated by the division of the functional modules described above. In practical applications, the functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the flood-prone area emergency evacuation planning device and the flood-prone area emergency evacuation planning method embodiment are based on the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0119] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. Furthermore, the functional modules in each embodiment of the invention can be integrated into a single processor, exist as separate physical entities, or consist of two or more modules integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0120] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a terminal device (which may be a personal computer, mobile phone, or communication device, etc.) or processor to execute all or part of the steps of the method of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0121] Figure 5 This is a schematic diagram of the structure of a computer device provided in an exemplary embodiment of the present invention. For example... Figure 5 As shown, the computer device 500 includes a processor 501 and a memory 502.

[0122] Processor 501 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 501 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 501 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 501 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 501 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0123] The memory 502 may include one or more computer-readable storage media, which may be non-transitory. The memory 502 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 502 is used to store at least one instruction, which is executed by the processor 501 to implement the flood hazard zone personnel emergency evacuation planning method provided in this embodiment of the invention.

[0124] Those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on the computer device 500, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0125] This invention also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of a computer device, enables the computer device to execute the flood hazard zone emergency evacuation planning method provided in this invention.

[0126] This invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the flood hazard zone emergency evacuation planning method provided in this invention.

[0127] The above description is merely an optional embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for planning the emergency evacuation of people in flood-prone areas, characterized in that, The method includes: Acquire high-precision 3D geographic data, historical disaster data, and hydrological models of the target area; Based on the three-dimensional geographic data, a road network map of the target area is constructed, and the roadbed relative elevation index of each path in the road network map is obtained. The roadbed relative elevation index is used to describe the micro-topography of each path. Based on the road network map and the historical disaster data, a set of sensitive routes is determined, which includes river-crossing bridges and flooded routes; The hydrological model is used to perform hydrological simulation on the target area; The cost of each path in the road network map is calculated using a cost function, which is based on the hydrological simulation results at time t and the roadbed relative elevation index. Based on the toll cost of each path in the road network map, determine the optimal transfer path for the first user to be transferred at time t. The optimal transfer path is the transfer path with the minimum cumulative toll cost. In the hydrological simulation, at each time point, it is determined whether there is a path in the sensitive path set that is flooded. If there is a path in the sensitive path set that is flooded at time t, then the passage cost of the flooded sensitive path is set to infinity at time t.

2. The method according to claim 1, characterized in that, The process of obtaining the relative elevation index of the roadbed for each path in the road network map includes: Obtain the elevation data of the background terrain on both sides of the i-th path, wherein the background terrain on both sides is located in two directions perpendicular to the path in which the i-th path is located; Based on the elevation data of the i-th path and the elevation data of the background terrain on both sides of the i-th path, the relative elevation index of the roadbed of the i-th path is obtained. Based on the relationship between the relative elevation index of the roadbed of the i-th path and the concave threshold and convex threshold, the micro-topographic morphology of the i-th path is determined. The micro-topographic morphology includes concave water catchment sections, convex embankment sections and flat sections.

3. The method according to claim 2, characterized in that, The cost function is expressed by the following formula: in, Let be the cost function value of the m-th path at time t, representing the travel cost of the m-th path at time t. Let m be the physical length of the m-th path. Let be the roadbed relative elevation index for the m-th route. As a micro-topographic risk penalty, this applies when the micro-topographic features of the m-th path are convex embankment sections and flat sections. In the case where the micro-topographical morphology of the m-th path is a concave water catchment section , Risk coefficients corresponding to different types of micro-geomorphic features. For real-time hydrodynamic penalty items, Let be the water depth of the m-th path at time t. Let be the flow velocity of the m-th path at time t. and The results were obtained from the hydrological simulation at time t. For the historical risk memory item of the m-th path, in the case that there are historical hydrological disasters on the m-th path. Used to impose punishment These are the weighting coefficients for different penalty items.

4. The method according to any one of claims 1 to 3, characterized in that, The step of determining the optimal transfer path for the first user to be transferred at time t based on the travel cost of each path in the road network map includes: Obtain a set of emergency shelters, which includes multiple emergency shelters and are used to accommodate users awaiting relocation; Starting from the location of the first user to be transferred, and taking each emergency resettlement point in the set of emergency resettlement points as the endpoint, and using the cumulative travel cost of the transfer path as the objective function, an optimization algorithm is used to search for the optimal transfer path of the first user to be transferred from the road network map.

5. The method according to any one of claims 1 to 3, characterized in that, The method further includes: If, at time t, there are paths in the set of sensitive paths that are flooded, the connectivity of the road network map is checked, and road segments that cannot be connected to any emergency resettlement points due to flooding are removed.

6. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Obtain the cumulative passage cost of the optimal transfer path for each person to be transferred at time t, and filter out the persons to be transferred whose cumulative passage cost is infinite, thus obtaining the predicted set of trapped persons at time t. A forced transfer command is sent to the group of trapped people at time t so that the people in the group of trapped people are transferred to the emergency resettlement point before time t.

7. A device for planning the emergency evacuation of people in flood-prone areas, characterized in that, The device includes: The first acquisition module is used to acquire high-precision three-dimensional geographic data, historical disaster data and hydrological models of the target area; The second acquisition module is used to construct a road network map of the target area based on the three-dimensional geographic data, and to acquire the roadbed relative elevation index of each path in the road network map. The roadbed relative elevation index is used to describe the micro-topography of each path. The sensitive path determination module is used to determine a set of sensitive paths based on the road network map and the historical disaster data, wherein the set of sensitive paths includes bridges across rivers and flooded paths; The hydrological simulation module is used to perform hydrological simulation on the target area using the hydrological model. The cost calculation module is used to calculate the passage cost of each path in the road network map using a cost function, which is based on the hydrological simulation results at time t and the roadbed relative elevation index. The route planning module is used to determine the optimal transfer path for the first user to be transferred at time t based on the toll cost of each path in the road network map. The optimal transfer path is the transfer path with the minimum cumulative toll cost. In the hydrological simulation, at each time point, it is determined whether there is a path in the sensitive path set that is flooded. If there is a path in the sensitive path set that is flooded at time t, then the passage cost of the flooded sensitive path is set to infinity at time t.

8. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to implement the method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images