Flood avoidance path planning method based on population dynamic distribution and flood inundation risk

Abstract

The invention provides a flood avoidance path planning method based on population dynamic distribution and flood inundation risk, and relates to the technical field of flood control engineering. The population gathering area is determined by acquiring the position data of the mobile terminal in real time, the population short-time migration rule is analyzed, and the dominant direction of population migration is determined; historical road blocking events and real-time rainfall monitoring data are utilized, and an urban road drainage pipe network structure is combined to construct a road inundation evolution network for limiting a ponding propagation boundary; identifying path nodes with traffic conflict risks through spatial topology analysis, and determining a node avoidance sequence according to node connectivity stability characteristics; an ant colony search model fused with a local path direction guide mechanism is adopted to realize accurate planning of a flood avoidance evacuation path for avoiding a road flooding risk area, so that the instantaneity and the reliability of population evacuation path planning under a flood disaster scene are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of flood control engineering technology, and in particular to a method for planning flood avoidance routes based on dynamic population distribution and flood inundation risk. Background Technology

[0002] With the intensification of urbanization and climate change, the frequency of extreme weather events, especially urban flooding caused by torrential rains, is constantly increasing, posing a severe challenge to urban public safety. The high population density and frequent movement of people in cities make the timely and efficient evacuation of people and the protection of residents' lives and property in emergency situations a crucial issue in the field of public safety management.

[0003] Traditional urban flood control and evacuation methods rely heavily on static population distribution information and historical flood data for route planning, neglecting the real-time dynamic characteristics of population movement and the propagation and evolution of water accumulation in urban drainage systems during rainstorms. This limitation of technical means leads to poor actual emergency evacuation results and makes it difficult to meet the practical needs of rapid response and precise guidance.

[0004] Therefore, there is an urgent need for an evacuation route planning method that can comprehensively consider the real-time changes in population distribution and the dynamic evolution of floodwater propagation, in order to improve the emergency evacuation efficiency of urban floods and ensure the safety of residents. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a flood avoidance route planning method based on dynamic population distribution and flood inundation risk.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a flood avoidance route planning method based on dynamic population distribution and flood inundation risk, including: Population clusters are identified based on real-time location data from mobile terminals, and spatial convergence features of population flow directions between these clusters are extracted to generate dominant population migration directions that characterize short-term population migration patterns. ; Based on historical road flooding event data, road blockage areas are identified, and combined with real-time rainfall monitoring data and the topology of urban road drainage pipe network, a road flooding evolution network is constructed to define the boundary of water accumulation propagation. Establish the dominant direction of population migration The set of spatial topological relationships with the road flooding evolution network And identify the set of path nodes that are at risk of traffic conflict. ; Generate a node avoidance sequence based on the stability characteristics of the path node connectivity. ; According to the node avoidance sequence An ant colony search model integrating local path direction guidance mechanism is constructed, and flood avoidance and evacuation routes that avoid road flooding risk areas are output based on the ant colony search model.

[0007] Furthermore, the method for determining the population concentration area includes: Continuous spatiotemporal trajectories are constructed from the real-time location data of all mobile terminals within a preset time period to obtain the spatiotemporal activity trajectories of the mobile terminals. Based on the spatiotemporal activity trajectory, a density clustering algorithm is used to spatially cluster all mobile terminal location points to obtain the initial population location clustering region. Calculate the spatial overlap between the initial population location clustering area and the historical population clustering area. According to the aforementioned spatial overlap The comparison is performed based on the magnitude of the value with the preset matching threshold. If the spatial overlap If the population location clustering area is greater than or equal to the preset matching threshold, the initial population location clustering area will be spatially fused with the corresponding historical population clustering area and determined as the population clustering area. If the spatial overlap If the population location is less than the preset matching threshold or does not match a historical population clustering area, the initial population location clustering area will be directly determined as the population clustering area.

[0008] Furthermore, the dominant direction of population migration The generation process includes: Extract the locations of population agglomeration areas over multiple consecutive time periods, and construct a flow vector to characterize spatial flow paths based on the spatial location changes of agglomeration areas between any two consecutive time periods. ; According to the flow vector Population flow between any two densely populated areas Calculate the spatial convergence strength, which reflects the spatial convergence capacity. and the directional deviation angle reflecting the time stability of the flow direction ; The spatial convergence strength and the direction deviation angle Vector concatenation is performed to obtain the spatial convergence characteristics of inter-regional population flow; Based on the spatial distribution stability characteristics of the aforementioned spatial convergence features, the flow directions between highly stable regions are identified, and the dominant direction of population migration is determined. .

[0009] Furthermore, the method for determining the road obstruction area includes: Obtain spatial location data of road blockage events corresponding to multiple historical rainfall events, and construct an initial spatial distribution map of road blockage events; Using a spatial hotspot detection method, high-frequency occurrence areas of road blockage events are identified in the initial spatial distribution map to obtain road blockage hotspot areas; The road blockage area is determined based on the location stability and spatial morphological characteristics of the road blockage hotspot.

[0010] Furthermore, the construction process of the road inundation evolution network is as follows: Based on the spatial distribution characteristics of real-time rainfall monitoring data and the topology of urban road drainage networks, a road network waterlogging diffusion model is established. Based on the aforementioned road network water accumulation diffusion model Analyze the overflow status of each road node to obtain the corresponding overflow flow. ; The road blockage area is used as a correction constraint for water accumulation distribution, and combined with the overflow flow rate. The propagation range of water accumulation in the road network was determined using fluid dynamics simulation methods. Based on the water accumulation propagation range and road network connectivity, a road flooding evolution network that defines the water accumulation propagation boundary is obtained.

[0011] Furthermore, the set of spatial topological relationships The establishment process includes: Based on the dominant direction of population migration Vector feature mapping is performed on road nodes within the target area to determine the direction of the guiding vector for each road node. ; The guiding vector direction of each road node Spatial topology matching is performed with the road flooding evolution network to identify the direction of the guiding vector. The spatial relative position between the guiding road node and the boundary of the water propagation; Based on the spatial relative positions, the topological relationships between nodes are established, resulting in a set of spatial topological relationships between the dominant population migration direction and the road flooding evolution network. .

[0012] Furthermore, the set of path nodes with the risk of traffic conflict. The identification methods include: According to the set of spatial topological relationships Identify the set of road nodes to be verified that are located within the guidance range of the dominant direction of population migration and are affected by the boundary envelope of the water accumulation propagation. ; Calculate the set of road nodes to be verified The probability of spatial conflict among nodes under the dual pressures of mass influx and approaching floodwaters ; Based on the spatial conflict probability, each path node is filtered, and the spatial conflict probability is... Greater than or equal to the preset risk probability threshold The corresponding path nodes are marked as having a risk of passage conflict, thus obtaining the set of path nodes with a risk of passage conflict. .

[0013] Furthermore, the node avoidance sequence The generation process includes: Based on the set of path nodes with potential traffic conflict. Construct a spatial adjacency network that reflects the spatial location and connectivity of nodes. ; According to the spatial adjacency network The impact of removing nodes from each path on network connectivity was analyzed using a node removal simulation method. ; Based on the degree of impact The avoidance priority of each node is determined based on the principle of minimizing connectivity loss after node removal, resulting in a node avoidance sequence. .

[0014] Furthermore, the construction process of the ant colony search model that integrates local path direction guidance mechanism includes: Starting from the densely populated area and ending at the preset safe shelter, the road segments to be searched in the road flooding evolution network are extracted to construct a basic path network. The node avoidance sequence Converted into a passage probability adjustment factor for nodes in the underlying path network And based on the dominant direction of population migration Establish a local direction consistency guiding function ; The local direction consistency guiding function With the passage probability adjustment factor By embedding the probability transition formula of the ant colony algorithm, an ant colony search model that integrates local path direction guidance mechanism is constructed and output.

[0015] Furthermore, the output process of the flood avoidance and evacuation path includes: Using the ant colony search model that integrates the local path direction guidance mechanism, multiple candidate node path sequences are obtained through iterative search in the road flooding evolution network. ; For the candidate node path sequence Perform a passability check and ensure that the set of path nodes that pose a risk of passability conflict is avoided. Under the premise of selecting risk cost from the candidate node path sequence Minimal node path sequence As a node path sequence to avoid risks; Based on the node sequence for risk avoidance, generate and output flood evacuation routes to avoid areas at risk of road flooding.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention captures and analyzes the dynamic distribution characteristics of the population in real time, accurately identifies the patterns of population gathering and migration, effectively solves the problem that traditional static evacuation route planning methods are difficult to adapt to real-time population changes, and improves the response speed of personnel evacuation in flood disaster scenarios.

[0017] This invention constructs a road waterlogging propagation evolution network that integrates rainfall processes and urban drainage pipe networks, enabling accurate prediction and dynamic updating of flood waterlogging propagation boundaries. This overcomes the problems of ambiguous waterlogging prediction boundaries and inability to provide timely feedback in existing technologies, thereby enhancing the dynamic adaptability and accuracy of path planning.

[0018] This invention establishes an ant colony search model that considers the direction of population migration and the dynamic propagation of water accumulation, accurately outputting the best evacuation route to avoid areas at risk of water accumulation. This effectively reduces the risk of traffic congestion and secondary disasters during the evacuation process and comprehensively improves the level of urban emergency management in flood disaster scenarios. Attached Figure Description

[0019] Figure 1 This is a flowchart of the flood avoidance route planning method based on dynamic population distribution and flood inundation risk in Example 1.

[0020] Figure 2 This is an architecture diagram of the flood avoidance route planning system based on dynamic population distribution and flood inundation risk in Example 2. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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.

[0022] Example 1 Please see Figure 1This invention provides a flood avoidance route planning method based on dynamic population distribution and flood inundation risk, the method comprising: Population clusters are identified based on real-time location data from mobile terminals, and spatial convergence features of population flow directions between these clusters are extracted to generate dominant population migration directions that characterize short-term population migration patterns. ; Based on historical road flooding event data, road blockage areas are identified, and combined with real-time rainfall monitoring data and the topology of urban road drainage pipe network, a road flooding evolution network is constructed to define the boundary of water accumulation propagation. Establish the dominant direction of population migration The set of spatial topological relationships with the road flooding evolution network And identify the set of path nodes that are at risk of traffic conflict. ; Generate a node avoidance sequence based on the stability characteristics of the path node connectivity. ; According to the node avoidance sequence An ant colony search model integrating local path direction guidance mechanism is constructed, and flood avoidance and evacuation routes that avoid road flooding risk areas are output based on the ant colony search model.

[0023] It should be noted that, in this embodiment, the real-time location data of the mobile terminal is obtained through wireless communication base stations within the target area or... GPS The positioning system collects data in real time; the collected data includes mobile terminal identifiers. timestamp and latitude and longitude coordinates ;in, Indicates the first A mobile terminal at any time Location, , They represent the first A mobile terminal at any time The longitude and latitude.

[0024] It is particularly important to note that, to ensure the physical accuracy of spatial distance calculations, the original latitude and longitude coordinates were first converted into Cartesian coordinates using the Gauss-Kruger projection algorithm before data processing. The transformed planar coordinate points are represented as: ;in, Indicates the first A mobile terminal at any time The planar position point, Represents the x-coordinate of the plane. Represents the vertical coordinate of the plane.

[0025] Specifically, the method for determining the population concentration area includes: Continuous spatiotemporal trajectories are constructed from the real-time location data of all mobile terminals within a preset time period to obtain the spatiotemporal activity trajectories of the mobile terminals. In practice, the location data of each mobile terminal is first sorted in chronological order to construct the continuous spatiotemporal activity trajectory of the terminal in a planar coordinate system, as shown below: ; In the formula, Indicates the first The spatiotemporal trajectory of a mobile terminal within a preset time period. Indicates the first The total number of sampling points for each mobile terminal within a preset time period.

[0026] It should be noted that the preset time period can be 1 minute, 5 minutes, or even half an hour, depending on the actual needs.

[0027] Based on the spatiotemporal activity trajectory, a density clustering algorithm is used to spatially cluster all mobile terminal location points to obtain the initial population location clustering region. It should be understood that for any planar location point The method for calculating local spatial density in a Cartesian coordinate system is as follows: ; In the formula, Indicates location point Spatial density value, Indicates the spatial density search radius. This represents the total number of sampling points for all mobile terminals within a preset time period. Represents the characteristic function.

[0028] It should be noted that the indicator function In the middle, when When the value is 1, the value is 1; otherwise, the value is 0. The spatial density search radius... In this embodiment, the value is taken as 50m.

[0029] Subsequently, based on the spatial density of the location points The value will be greater than or equal to the preset density threshold. The location points are determined as cluster core points. Using these core points as seed points, the initial population location clustering regions are gradually expanded through neighborhood space search; among them, the spatial density threshold is... The top 10 percentile of the density values ​​at all locations can be taken.

[0030] Calculate the spatial overlap between the initial population location clustering area and the historical population clustering area; determine the relationship between the spatial overlap and a preset matching threshold: If the spatial overlap is greater than or equal to the preset matching threshold, the initial population location clustering area and the corresponding historical population clustering area are spatially fused and determined as the population clustering area. If the spatial overlap is less than the preset matching threshold or does not match a historical population cluster area, the initial population location cluster area will be directly determined as the population cluster area.

[0031] Understandably, the first step is to obtain the historical population clusters stored in the historical population distribution database, and then calculate the spatial overlap between the initial population location cluster and each historical population cluster. The calculation formula is as follows: ; in, Represents the initial population location clustering region. with historically densely populated areas Spatial overlap between them This represents the area calculation function.

[0032] Based on the relationship between the spatial overlap and the preset matching threshold (e.g., 0.65-0.80), this embodiment performs the following classification process: If the spatial overlap is greater than or equal to a preset matching threshold, it is determined that the initial population location clustering area and the historical population clustering area are highly consistent. At this time, the initial population location clustering area and the corresponding historical population clustering area are spatially fused. Specifically, spatial centroid weighting or envelope merging algorithms can be used to generate the fused boundary, which is used as the final output population clustering area. The processing method effectively filters out instantaneous spatial deviations caused by meteorological interference or positioning drift by introducing historical prior knowledge.

[0033] If the spatial overlap is less than the preset matching threshold, or if the initial population location clustering area does not find a matching historical population clustering area in the database (e.g., the area is a newly discovered evacuation hub or the first time population clustering area division is performed), then the area is determined to be a newly added clustering point with high timeliness; at this time, the fusion step is skipped, and the initial population location clustering area is directly determined as the population clustering area of ​​the current time period and output.

[0034] Furthermore, the spatial convergence features of the flow of people between the population-gathering areas are extracted to generate the dominant direction of population migration for characterizing the short-term migration patterns of the population.

[0035] In practice, the process of generating the dominant direction of population migration includes: Extract the locations of population agglomeration areas over multiple consecutive time periods, and construct a flow vector to characterize the spatial flow path based on the spatial location change relationship between any two consecutive time periods. In practice, the first step is to calculate the population density of each area at the sampling time. Plane geometric centroid coordinates The calculation formula is as follows: ; In the formula, Indicates the first In a densely populated area at any time The plane geometric centroid coordinates, Indicates the first The total number of location points contained within a population cluster area , They represent the first The first in a densely populated area The location of each mobile terminal at time The horizontal and vertical plane coordinates.

[0036] Subsequently, by comparing the time... Centroid of population concentration areas and the next adjacent time step set of centroids of population-gathering areas This determines the spatial location changes of population gathering areas at two consecutive adjacent moments.

[0037] It should be understood that the aforementioned spatial positional change relationship refers to the relationship over a time interval. Within, the population size center is located at coordinate point Move to coordinate point The geometric displacement mapping relationship.

[0038] Based on the aforementioned spatial positional change relationship, a flow vector is constructed. To characterize the spatial flow path, specifically represented as: ; In the formula, This indicates the preset sampling time interval. Indicates from time arrive During the period, from densely populated areas Pointing to densely populated areas The flow vector, vector magnitude It represents the physical displacement distance between the centroids of two population clusters.

[0039] Based on the population flow between any two population gathering areas in the flow vector, calculate the spatial convergence strength, which reflects the spatial convergence capacity, and the directional deviation angle, which reflects the temporal stability of the flow direction; By concatenating the spatial convergence strength and the directional deviation angle into vectors, the spatial convergence characteristics of inter-regional personnel flow can be obtained. It is understandable that the synchronous statistics in this embodiment are performed at time [time]. arrive During the time period, along the flow vector The number of unique mobile terminals moving along the corresponding path is defined as population flow. .

[0040] Based on this, this embodiment generates the spatial convergence feature by integrating personnel size, spatial orientation, and temporal stability. The specific calculation process includes: First, by utilizing the geometric distribution characteristics of multiple flow vectors, combined with the population flow... Perform weighted analysis to calculate and quantify the population concentration in the target population area. Spatial convergence strength of convergence Specifically, it is expressed as: ; In the formula, Indicates the target population concentration area At any moment Spatial convergence strength, Indicates all at time Targeting densely populated areas The initial region set, Indicates a densely populated area The centroid points to the target population concentration area The unit vector of the centroid.

[0041] Understandably, this is achieved through population flow. The product of the vector projection coefficient quantifies the spatial aggregation capacity of large-scale crowds.

[0042] To assess the persistence of a specific flow path over time, this embodiment calculates the flow vector for the current sampling period. Flow vector of the previous adjacent sampling period The geometric angle between them is defined as the direction deviation angle, specifically expressed as: ; In the formula, Indicates the directional deviation angle of the flow path within a continuous time period. Indicates the vector length.

[0043] Finally, the obtained spatial convergence strength The deviation angle with respect to the direction Vector concatenation is performed to generate spatial convergence features that characterize the patterns of inter-regional population flow.

[0044] Based on the spatial distribution stability characteristics of the spatial convergence features, the flow direction between highly stable regions is identified, and the dominant direction of population migration is determined.

[0045] In specific implementation, this embodiment identifies flow directions with spatial distribution stability characteristics based on the parameters in the spatial convergence features. The specific determination logic is as follows: if the spatial convergence strength corresponding to a certain flow path is high within multiple consecutive sampling times... The traffic coverage threshold is consistently greater than or equal to the preset threshold (indicating strong spatial convergence capability), and the corresponding directional deviation angle is... If the flow direction remains below a preset directional stability threshold (e.g., 30°, indicating that the flow direction is stable over time), then the flow direction is determined to have a highly stable spatial distribution characteristic.

[0046] Finally, the identified high-stability flow vectors are normalized and weighted to obtain the dominant direction of population migration, which characterizes the short-term migration pattern, as follows: ; In the formula, This indicates the dominant direction of normalized population migration.

[0047] In one embodiment, the method for determining the road blockage area includes: Obtain spatial location data of road blockage events corresponding to multiple historical rainfall events, and construct an initial spatial distribution map of road blockage events; It should be understood that this embodiment first establishes a road blockage event database containing historical road flooding event data. The information stored in the database includes, but is not limited to, the spatial coordinates of the road blockage event, the specific time when the blockage occurred, the duration of rainfall, and the intensity of rainfall.

[0048] It should be noted that the spatial location data of road blockage events is obtained from historical records collected by traffic management departments or urban emergency management departments. This data includes, but is not limited to, the start and end spatial location data of road blockages during each rainfall event, specifically represented as follows: ; In the formula, This represents a dataset of road blockage events. Indicates the first The spatial location plane coordinates of a road blockage event This represents the total number of road blockage events collected.

[0049] By spatially overlaying and statistically analyzing road blockage event data corresponding to multiple historical rainfall events, an initial spatial distribution map of road blockage events was obtained.

[0050] A hotspot detection method based on spatial statistics identifies high-frequency occurrence areas of road blockage events in the initial spatial distribution map, thereby obtaining road blockage hotspot areas. It should be understood that this embodiment employs a space hotspot detection method (e.g., Getis-OrdG (Hotspot analysis method) to identify hotspot areas of road blockage events in the initial spatial distribution map: Specifically, the formula for hotspot analysis is as follows: ; In the formula, Indicates the first Hotspot index of each spatial unit Indicates the first The number of road blockage events within a spatial unit This represents the average number of blocking events occurring within all space cells. Representing spatial units and Spatial weight function between, This represents the standard deviation of the number of road blockage events across all spatial units. This represents the total number of location points in all spatial units.

[0051] It should be noted that when the distance between two location points is less than a preset spatial threshold... (For example, at 500 meters) The value is 1 if it is not 1, otherwise the value is 0.

[0052] Using the above method, when the hotspot index of a spatial unit... When the value exceeds a preset significance threshold (e.g., the critical value corresponding to a significance level of 95%), the spatial unit is determined to be a road blockage hotspot area.

[0053] The road blockage area is determined based on the location stability and spatial morphological characteristics of the road blockage hotspot area.

[0054] In practice, location stability is quantified by the consistency of spatial location of hotspot areas across multiple historical events. This involves calculating the percentage frequency of each hotspot area across multiple rainfall events, specifically expressed as: ; In the formula, Indicates the first Percentage of location stability frequency in each hotspot area This indicates the number of times a hotspot area has appeared in historical rainfall events. This represents the total number of historical rainfall events.

[0055] Spatial morphological characteristics are determined by the spatial geometric features of hotspot areas, such as area, aspect ratio, or compactness, and are specifically expressed as follows: ; In the formula, Indicates the first The compactness of each hotspot area Indicates the first The area of ​​each hotspot region Indicates the first The perimeter of the boundaries of each hotspot area.

[0056] In this embodiment, the road blockage area is screened and determined based on threshold conditions such as the percentage of location stability frequency (e.g., greater than or equal to 70%) and the area compactness (e.g., greater than or equal to 0.5).

[0057] Subsequently, this embodiment combines the spatial distribution characteristics of real-time rainfall monitoring data to construct a road inundation evolution network that defines the boundary of water accumulation propagation. The construction process of the road inundation evolution network is as follows: The spatial distribution characteristics of real-time rainfall monitoring data are obtained, and a road network waterlogging diffusion model is established in combination with the topology of urban road drainage pipe network. Based on the road network waterlogging diffusion model, the overflow status of each road node is analyzed to obtain the corresponding overflow flow. It should be noted that the real-time rainfall monitoring data is obtained in real time through an array of rainfall monitoring sensors deployed in the target area. Specifically, it includes: the real-time rainfall intensity at each monitoring point, the topology of the urban road drainage network including road nodes, pipeline connections, and corresponding hydraulic discharge capacity information.

[0058] Based on this, a road network waterlogging diffusion model is established using a hydrological model that couples rainfall infiltration and surface runoff (such as the SWMM model), specifically expressed as follows: ; In the formula, Indicates time Nodes in the road network With nodes The water flow rate between Indicates the real-time rainfall intensity distribution. This represents the runoff resistance coefficient between road nodes. This represents the drainage capacity coefficient between road nodes. A hydraulic coupling calculation function representing the water flow rate.

[0059] It should be understood that the hydraulic coupling calculation function It can be constructed based on the hydraulic logic of Manning's formula or Saint-Venant's equations, where the runoff resistance coefficient... With drainage capacity coefficient The roughness parameters can be obtained by referring to the table in the "Outdoor Drainage Design Code" and the specific road surface roughness parameters.

[0060] Furthermore, when the water flow rate Greater than the drainage capacity coefficient At that time, the overflow flow rate of each drainage node is calculated, specifically expressed as: ; In the formula, Indicates the first Each node at time... Overflow flow Indicates at time Entering the Total inbound traffic to each node Indicates the first The hydraulic discharge load threshold for each node.

[0061] It should be noted that the total inbound traffic It consists of two parts: first, the internal flow of the pipe network flowing into the node from the upstream adjacent node according to the topology of the urban road drainage pipe network; second, the flow of the drainage node flowing into the drainage node through surface runoff formed by real-time rainfall within the catchment area covered by the node.

[0062] The road blockage area is used as a correction constraint for water accumulation distribution, and the water accumulation propagation range of the road network is determined by using fluid dynamics simulation methods based on the overflow parameters. In practice, the road blockage area serves as a rigid boundary for water propagation. A two-dimensional shallow water wave equation is used for fluid dynamics simulation to calculate the water propagation range. ; In the formula, Indicates the net source term intensity of surface water. Indicates the water depth at the node location. , The symbols represent the water flow at the nodes in the plane. direction and Flow velocity in the direction, Indicates simulation time. , Represents coordinates in planar space.

[0063] In the simulation, the road obstruction area is mapped onto a discrete grid of the two-dimensional shallow water wave equation, and the physical properties of its corresponding grid cells are set to rigid boundary conditions of zero permeability and impassability (e.g., by setting an infinite damping coefficient or making its terrain elevation much higher than that of adjacent grids), thereby simulating the obstruction behavior of water accumulation in a real road environment; when the water depth during the solution process... and flow rate , When evolving to the boundary grid of the blocking region, the boundary flux is forced to zero through a numerical discretization scheme, which induces the simulated flow field to change direction or accumulate, thereby accurately defining the spatial boundary range of water propagation.

[0064] It should be noted that the net source term strength Overflow flow rates at each node calculated from the initial road network water diffusion model The transformation yields; specifically, when the node When an overflow occurs, As the flow injection value at this spatial coordinate point, it drives the increase in surface water depth and the diffusion of runoff.

[0065] It should also be noted that the solution of the two-dimensional shallow water wave equation is performed by discretization simulation using the finite volume method or the finite difference method. By iteratively updating the hydraulic parameters of each grid node within the time step, the dynamic tracking of the water accumulation evolution process is achieved.

[0066] Based on the water accumulation propagation range and road network connectivity, a road flooding evolution network that defines the water accumulation propagation boundary is obtained.

[0067] Understandably, the boundary of the water accumulation range is used as a constraint, ensuring that the water depth is less than or equal to a preset maximum allowable passage depth threshold. Road nodes are marked as passable nodes, while those with water depth exceeding a preset maximum allowable passability depth threshold are marked as passable nodes. Road nodes are marked as impassable nodes; further, based on the topological connectivity between road nodes, a road inundation evolution network is obtained that defines the boundary of water accumulation propagation, which serves as the basic support for subsequent flood avoidance path planning and implementation.

[0068] In practical application scenarios, this embodiment first establishes a spatial topological relationship between the dominant direction of population migration and the road flooding evolution network. Then, based on the spatial topological relationship, it further identifies the traffic conflict risks existing in the path nodes and determines the avoidance node sequence based on the connectivity stability characteristics of the path nodes.

[0069] Specifically, the process of establishing the set of spatial topological relationships includes: Based on the dominant direction of population migration, vector feature mapping is performed on road nodes within the target area to determine the guiding vector direction of each road node; In practice, the guiding vector direction of a road node is first defined as the spatial projection vector of the dominant population migration direction at the road node, specifically expressed as: ; In the formula, Represents road nodes The direction of the guiding vector, Indicates pointing to a node The set of all regions from which the dominant direction of population migration originates. Indicates the area Pointing to node Population flow Indicates the area Pointing to node The flow vector.

[0070] It is understandable that the above vector mapping can accurately reflect the dominant directional characteristics of population migration carried by road nodes.

[0071] Spatial topology matching is performed between the guiding vector direction of each road node and the road flooding evolution network to identify the spatial relative position between the road node guided by the guiding vector direction and the water accumulation propagation boundary; It should be understood that the formula for calculating the spatial relative position index between road nodes and the boundary of water accumulation propagation is as follows: ; In the formula, Represents road nodes The shortest spatial distance to the boundary of water accumulation. This represents the set of nodes on the boundary of water propagation. Represents a node Boundary nodes of water accumulation propagation The Euclidean distance between them.

[0072] It is understandable that, based on the above spatial relative position indicators, it is determined whether each node is guided by the dominant direction of population migration and is located near the boundary of water accumulation propagation.

[0073] Based on the spatial relative positions, the topological relationships between nodes are established to obtain the set of spatial topological relationships between the dominant direction of population migration and the road flooding evolution network.

[0074] In specific implementation, the topological association logic is manifested as establishing a set of topological adjacency relationships, specifically represented as follows: ; In the formula, Represents a node The set of topologically related nodes, This indicates a preset spatial proximity threshold. Represents a node With nodes The direct connectivity between them on the road network.

[0075] It should be noted that the preset spatial proximity threshold Used to determine whether nodes are adjacent; in this embodiment, the value is set to 100 meters. When a node... , The direct connectivity relationship when there is a direct road segment connection The value is 1 if it is not 1, otherwise the value is 0.

[0076] Furthermore, the method for identifying the set of path nodes with potential traffic conflict includes: Based on the set of spatial topological relationships, identify a set of road nodes to be verified that are located within the guidance range of the dominant direction of population migration and are affected by the envelope of the water accumulation propagation boundary; It should be understood that the set of road nodes to be verified can be represented as: ; In the formula, This represents the set of road nodes to be verified. This represents the spatial threshold for risk assessment (e.g., 50 meters). This represents the angle between the node's guiding vector and the direction the node points towards the water accumulation boundary. This indicates the threshold for the directional angle (e.g., 45°). Indicates from node A unit vector pointing to the boundary of water accumulation propagation.

[0077] The spatial conflict probability method was used to calculate the spatial conflict probability of each node in the set of road nodes to be verified under the dual pressure of mass influx and flood approach. It should be noted that before calculating the spatial conflict probability, the mapping relationship between population flow from region pairs to path nodes must first be established: based on each flow vector The corresponding preset shortest path will determine the population flow. Assigned to the corresponding road nodes according to the route. Thus, the nodes are summarized. Number of people entering within a unit of time period The spatial conflict probability The calculation formula is as follows: ; In the formula, Represents a node The probability of spatial conflict, This represents the conflict probability adjustment factor (e.g., 0.5). Represents a node The number of people entering within a unit of time period. Represents a node The design capacity of the passageway Represents a node The depth of the water at the location, This indicates the maximum permissible water depth threshold (e.g., 0.3 meters).

[0078] Based on the spatial conflict probability of each node, path nodes with a risk of passage conflict are selected.

[0079] In practice, based on the preset risk probability threshold (e.g., 0.6), determine if the condition is met. The nodes are the set of nodes that pose a risk of passage conflict. .

[0080] Specifically, the process of generating the node avoidance sequence includes: Based on the set of path nodes with potential traffic conflict, a spatial adjacency network reflecting the spatial location and connectivity of the nodes is constructed. Understandably, in order to analyze the structural characteristics of the risk area, this embodiment first targets the set of nodes with the risk of passage conflicts. Construct a spatial risk adjacency network that reflects the spatial location and local connectivity of nodes. ;in The set of nodes that are at risk of traffic conflict. For direct adjacency relationships between risk nodes, the matrix elements can be represented as: ; By constructing the spatial risk adjacency network It can quantify the degree of clustering of risk nodes in local space and the strength of topological association.

[0081] Based on the spatial adjacency network, the impact of removing each path node on network connectivity is analyzed using a node removal simulation method. It should be understood that this embodiment calculates the impact of node removal on network connectivity before and after node removal using a node removal method. Specifically, it is expressed as: ; In the formula, This represents the overall connectivity measure of the original network (e.g., the maximum number of nodes in the connected subgraph). Represents a node Network connectivity measure after removal.

[0082] Based on the degree of impact, the avoidance priority of each node is determined according to the principle of minimizing connectivity loss after node removal, thus obtaining the node avoidance sequence.

[0083] It should be understood that the degree of impact Sort the nodes from smallest to largest to form a priority sequence for node avoidance, specifically represented as follows: .

[0084] It should be noted that, while ensuring the overall connectivity of the evacuation network, and following the principles of protecting core hubs and prioritizing the scheduling of redundant paths, the impact on network connectivity after node removal will be minimized, provided that the node's contribution to overall network connectivity is low (i.e., the degree of impact on network connectivity after node removal is minimized). (Smaller) and spatial conflict probability Higher-risk nodes are prioritized for avoidance to prevent systemic paralysis caused by excessive congestion of core evacuation routes, thereby determining the avoidance priority of each node and obtaining a node avoidance sequence.

[0085] In practical applications, this embodiment utilizes the global optimization characteristics of the ant colony algorithm, combining node avoidance sequences with the dominant direction of population migration to accurately construct an ant colony search model that integrates local path direction guidance mechanisms, and uses this model to generate and output flood avoidance and evacuation paths.

[0086] In implementation, the construction process of the ant colony search model that integrates local path direction guidance mechanism includes: Starting from the densely populated area and ending at the preset safe shelter, the road segments to be searched in the road flooding evolution network are extracted to construct a basic path network. In practical implementation, the population agglomeration area is a densely populated area obtained by spatial clustering. To facilitate the transition of the algorithm from isal clusters to linear road networks, the centroid of the population agglomeration area is calculated. The Euclidean distance between each road node in the road flooding evolution network is used to select the road node closest to the population gathering area as the starting node; the safe shelter is a pre-defined refuge area, and similarly, the road node closest to the geometric center of the safe shelter is determined as the ending point. This embodiment defines the basic path network as a directed graph. ,in This represents the set of all passable road nodes in the road flooding evolution network that are not located within the water accumulation propagation boundary. This represents the set of passable road segments between the road nodes that are not covered by water accumulation.

[0087] The node avoidance sequence is converted into a passage probability adjustment factor for nodes in the basic path network, and a local direction consistency guiding function is established based on the dominant direction of population migration. In practice, the probability adjustment factor is used to adjust the node passage probability, and the specific calculation method is as follows: First, define the node avoidance priority as follows: The node avoidance sequence is assigned a numerical ranking from high to low, with smaller values ​​indicating higher avoidance priority; subsequently, it is normalized into a node passage probability adjustment factor, specifically expressed as: ; In the formula, Represents a node The passage probability adjustment factor, This represents the maximum value among all node avoidance priorities. This represents a preset minimum bias constant (e.g., 0.01) used to prevent the node's selection probability from becoming completely zero in the initial iteration due to an excessively high node avoidance level.

[0088] It should be noted that the passage probability adjustment factor The range of values ​​is [ [1], the smaller the value, the higher the priority of the node to be avoided.

[0089] Furthermore, a local directional consistency guidance function based on the dominant direction of population migration is constructed to enhance the directional guidance of path search, specifically expressed as: ; In the formula, Represents a node To the node The directional consistency guiding coefficient, Represents a node The direction of the guiding vector, Represents a node To the node unit vector, Indicates the vector magnitude.

[0090] The local direction consistency guidance function and the passage probability adjustment factor are embedded into the probability transition formula of the ant colony algorithm to construct and output an ant colony search model that integrates the local path direction guidance mechanism.

[0091] It should be noted that the probability transition formula for the ant colony algorithm is defined as follows: ; In the formula, Ants At any moment From node Transfer to node The probability, Represents a node With nodes The concentration of pheromones between them. Ants The next set of nodes that can be accessed. , , These are the pheromone importance coefficient, the direction guidance importance coefficient, and the passage regulation factor importance coefficient, respectively. Their value ranges can be determined according to actual needs. In this embodiment... The value is 1. The value is 2. The value is 3.

[0092] Using the above method, an ant colony search model that integrates local path direction guidance mechanism is output to provide accurate guidance for determining flood avoidance and evacuation routes.

[0093] Furthermore, the output process of the flood avoidance and evacuation path includes: Using the ant colony search model that integrates the local path direction guidance mechanism, multiple candidate node path sequences are obtained through iterative search in the road flooding evolution network. Understandably, this embodiment utilizes the ant colony algorithm for multiple iterative calculations, repeatedly updating the pheromone concentration to generate the candidate node sequence, resulting in the following path set: ; In the formula, Represents the set of candidate node path sequences. Indicates the first Candidate node path sequence, This indicates the number of preset candidate node paths (e.g., 10).

[0094] The candidate node path sequences are tested for traversability, and the node path sequence with the lowest risk cost is selected from the candidate node path sequences as the node path sequence to avoid the risk, while ensuring that the path nodes with the risk of traversal conflict are avoided. In specific implementation, this embodiment considers candidate node paths Performing a passability check is specifically represented as follows: ; In the formula, Indicate candidate path The verification results Represents a node At any time The depth of the accumulated water, This indicates the threshold for the depth of standing water that allows passage.

[0095] Understandably, when This indicates that the path did not exceed the risk threshold for waterborne transmission.

[0096] The path risk cost function is further calculated and specifically expressed as follows: ; In the formula, Indicate candidate path The risks and costs, Represents a node The probability of spatial conflict, Represents a node To the node The actual length of the road segment between them.

[0097] Based on the above risk cost calculation, risk costs are selected from the paths that meet the verification criteria. The smallest node sequence is output as the node sequence that avoids risk.

[0098] Based on the node sequence for risk avoidance, generate and output flood evacuation routes to avoid areas at risk of road flooding.

[0099] It is understandable that this embodiment generates the final evacuation path step by step based on the node sequence for risk avoidance. Specifically, it is expressed as: ; in, Indicates the starting node of a population-concentrated area. Indicates a safe haven node. This represents the second intermediate node passed through in the node sequence for risk avoidance.

[0100] The above evacuation routes have been compiled and output to ensure that the route planning scheme is rigorous and reliable and meets the actual flood avoidance and evacuation needs.

[0101] Those skilled in the art should understand that the thresholds involved in the above steps (such as...) , , All of these (etc.) can be adaptively adjusted according to different urban road grades, drainage design standards, and historical rainfall recurrence periods, and should not be regarded as a limitation on the scope of protection of this invention.

[0102] Example 2 like Figure 2 As shown in the example, the parts not detailed in this embodiment are as shown in Example 1. This embodiment discloses a flood avoidance route planning system based on dynamic population distribution and flood inundation risk, including: Population Agglomeration Analysis Module: This module identifies population agglomeration areas based on real-time location data from mobile terminals, extracts spatial convergence features of population flow directions between these areas, and generates dominant population migration directions to characterize short-term migration patterns. ; Road flooding evolution module: Used to determine road blockage areas based on historical road flooding event data, and to construct a road flooding evolution network that defines the boundaries of flooding propagation by combining real-time rainfall monitoring data and the topology of urban road drainage pipe network; Risk Node Identification Module: Used to establish the dominant direction of population migration. The set of spatial topological relationships with the road flooding evolution network And identify the set of path nodes that are at risk of traffic conflict. ; Generate a node avoidance sequence based on the stability characteristics of the path node connectivity. ; Ant colony path planning module: used to plan the path based on the node avoidance sequence. An ant colony search model integrating local path direction guidance mechanism is constructed, and flood avoidance and evacuation routes that avoid road flooding risk areas are output based on the ant colony search model.

[0103] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0104] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A flood avoidance route planning method based on dynamic population distribution and flood inundation risk, characterized in that, include: Population clusters are identified based on real-time location data from mobile terminals, and spatial convergence features of population flow directions between these clusters are extracted to generate dominant population migration directions that characterize short-term population migration patterns. ; Based on historical road flooding event data, road blockage areas are identified, and combined with real-time rainfall monitoring data and the topology of urban road drainage pipe network, a road flooding evolution network is constructed to define the boundary of water accumulation propagation. Establish the dominant direction of population migration The set of spatial topological relationships with the road flooding evolution network And identify the set of path nodes that are at risk of traffic conflict. ; Generate a node avoidance sequence based on the stability characteristics of the path node connectivity. ; According to the node avoidance sequence An ant colony search model integrating local path direction guidance mechanism is constructed, and flood avoidance and evacuation routes that avoid road flooding risk areas are output based on the ant colony search model.

2. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The methods for determining the population concentration areas include: Continuous spatiotemporal trajectories are constructed from the real-time location data of all mobile terminals within a preset time period to obtain the spatiotemporal activity trajectories of the mobile terminals. Based on the spatiotemporal activity trajectory, a density clustering algorithm is used to spatially cluster all mobile terminal location points to obtain the initial population location clustering region. Calculate the spatial overlap between the initial population location clustering area and the historical population clustering area. According to the aforementioned spatial overlap The comparison is performed based on the magnitude of the value with the preset matching threshold. If the spatial overlap If the population location clustering area is greater than or equal to the preset matching threshold, the initial population location clustering area will be spatially fused with the corresponding historical population clustering area and determined as the population clustering area. If the spatial overlap If the population location is less than the preset matching threshold or does not match a historical population clustering area, the initial population location clustering area will be directly determined as the population clustering area.

3. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The dominant direction of population migration The generation process includes: Extract the locations of population agglomeration areas over multiple consecutive time periods, and construct a flow vector to characterize spatial flow paths based on the spatial location changes of agglomeration areas between any two consecutive time periods. ; According to the flow vector Population flow between any two densely populated areas Calculate the spatial convergence strength, which reflects the spatial convergence capacity. and the directional deviation angle reflecting the time stability of the flow direction ; The spatial convergence strength and the direction deviation angle Vector concatenation is performed to obtain the spatial convergence characteristics of inter-regional population flow; Based on the spatial distribution stability characteristics of the aforementioned spatial convergence features, the flow directions between highly stable regions are identified, and the dominant direction of population migration is determined. .

4. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The method for determining the road blockage area includes: Obtain spatial location data of road blockage events corresponding to multiple historical rainfall events, and construct an initial spatial distribution map of road blockage events; Using a spatial hotspot detection method, high-frequency occurrence areas of road blockage events are identified in the initial spatial distribution map to obtain road blockage hotspot areas; The road blockage area is determined based on the location stability and spatial morphological characteristics of the road blockage hotspot.

5. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The construction process of the road inundation evolution network is as follows: Based on the spatial distribution characteristics of real-time rainfall monitoring data and the topology of urban road drainage networks, a road network waterlogging diffusion model is established. Based on the aforementioned road network water accumulation diffusion model Analyze the overflow status of each road node to obtain the corresponding overflow flow. ; The road blockage area is used as a correction constraint for water accumulation distribution, and combined with the overflow flow rate. The propagation range of water accumulation in the road network was determined using fluid dynamics simulation methods. Based on the water accumulation propagation range and road network connectivity, a road flooding evolution network that defines the water accumulation propagation boundary is obtained.

6. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The set of spatial topological relationships The establishment process includes: Based on the dominant direction of population migration Vector feature mapping is performed on road nodes within the target area to determine the direction of the guiding vector for each road node. ; The guiding vector direction of each road node Spatial topology matching is performed with the road flooding evolution network to identify the direction of the guiding vector. The spatial relative position between the guiding road node and the boundary of the water propagation; Based on the spatial relative positions, the topological relationships between nodes are established, resulting in a set of spatial topological relationships between the dominant population migration direction and the road flooding evolution network. .

7. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The set of path nodes with potential traffic conflict The identification methods include: According to the set of spatial topological relationships Identify the set of road nodes to be verified that are located within the guidance range of the dominant direction of population migration and are affected by the boundary envelope of the water accumulation propagation. ; Calculate the set of road nodes to be verified The probability of spatial conflict among nodes under the dual pressures of mass influx and approaching floodwaters ; Based on the spatial conflict probability, each path node is filtered, and the spatial conflict probability is... Greater than or equal to the preset risk probability threshold The corresponding path nodes are marked as having a risk of passage conflict, thus obtaining the set of path nodes with a risk of passage conflict. .

8. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The node avoidance sequence The generation process includes: Based on the set of path nodes with potential traffic conflict. Construct a spatial adjacency network that reflects the spatial location and connectivity of nodes. ; According to the spatial adjacency network The impact of removing nodes from each path on network connectivity was analyzed using a node removal simulation method. ; Based on the degree of impact The avoidance priority of each node is determined based on the principle of minimizing connectivity loss after node removal, resulting in a node avoidance sequence. .

9. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The construction process of the ant colony search model that integrates local path direction guidance mechanism includes: Starting from the densely populated area and ending at the preset safe shelter, the road segments to be searched in the road flooding evolution network are extracted to construct a basic path network. The node avoidance sequence Converted into a passage probability adjustment factor for nodes in the underlying path network And based on the dominant direction of population migration Establish a local direction consistency guiding function ; The local direction consistency guiding function With the passage probability adjustment factor By embedding the probability transition formula of the ant colony algorithm, an ant colony search model that integrates local path direction guidance mechanism is constructed and output.

10. The flood avoidance route planning method based on dynamic population distribution and flood inundation risk according to claim 1, characterized in that, The output process of the flood avoidance and evacuation route includes: Using the ant colony search model that integrates the local path direction guidance mechanism, multiple candidate node path sequences are obtained through iterative search in the road flooding evolution network. ; For the candidate node path sequence Perform a passability check and ensure that the set of path nodes that pose a risk of passability conflict is avoided. Under the premise of selecting risk cost from the candidate node path sequence Minimal node path sequence As a node path sequence to avoid risks; Based on the node sequence for risk avoidance, generate and output flood evacuation routes to avoid areas at risk of road flooding.

Images