Dynamic disaster evacuation path planning method and system based on network flow model

Abstract

The present invention relates to the technical field of emergency traffic evacuation and path planning, and provides a method and system for dynamic disaster evacuation path planning based on a network flow model, the method comprising: determining the evacuation area of ​​the evacuation event; obtaining the number of vehicles to be evacuated in the evacuation area, and determining the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated; generating a path network model according to vector geographic data and the evacuation traffic distribution, the path network model including network nodes composed of evacuation source points, evacuation junctions and road network intersections and edges composed of road network road segments; constructing an event dynamic evacuation model according to the path network model; searching for the minimum cost-maximum flow path from the evacuation source point to the evacuation junction point in the evacuation road network based on the event dynamic evacuation model, and allocating the vehicles to be evacuated of each evacuation source point according to the search results until the evacuation planning is completed. The present invention improves the evacuation efficiency, reduces the evacuation time, and realizes fast, orderly and safe evacuation.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency traffic evacuation and path planning, and in particular to a dynamic disaster evacuation path planning method and system based on a network flow model. Background Art

[0002] How to provide an effective dynamic disaster evacuation route planning method to avoid vehicle congestion and enhance its resilience and response capabilities is of great significance.

[0003] In evacuation route planning, the multi-objective planning method is commonly used, which helps to find a balance between multiple objectives, such as the shortest evacuation time, the highest path safety, the least evacuation resource consumption, etc. The method is based on the Dijkstra algorithm. The Dijkstra algorithm is an algorithm for finding the shortest path from a single source point to all other vertices in a weighted graph. This method still has the following shortcomings:

[0004] (1) Unable to cope with the problem of multiple sources and multiple sinks

[0005] Since the core idea of ​​the multi-objective programming method based on the Dijkstra algorithm is a greedy strategy, its basic principle is to take the starting point as the center and gradually explore the shortest paths to other vertices until the shortest path of each vertex is traversed. In this process, the method generates a set containing the vertices for which the shortest paths have been found, as well as a priority queue for selecting the next vertex to be processed. Due to the design of the above method, it can only be processed from one source point, which means that it cannot be directly used to solve the shortest path problem between all pairs of vertices. In the scenario of multiple sources and multiple sinks for emergency evacuation, it is necessary to consider the paths from multiple source points to multiple evacuation sinks, and this method cannot be directly applied.

[0006] (2) Unable to handle network path planning issues with capacity constraints

[0007] Since the multi-objective programming method mainly considers the weight of the edge (such as distance, cost or time, etc.) when calculating the shortest path, without considering the capacity limit in the network. This method assumes that each edge can be used an unlimited number of times without considering the actual capacity constraints. In emergency evacuation, the road capacity is limited. If this limitation is not taken into account and the shortest path is simply pursued, the traffic flow on some sections will far exceed its carrying capacity, causing traffic congestion and even more serious evacuation problems. Summary of the invention

[0008] In view of the above problems, the present invention is proposed to provide a method and system for dynamic disaster evacuation path planning based on a network flow model that overcomes the above problems or at least partially solves the above problems.

[0009] One aspect of the present invention provides a method for dynamic disaster evacuation path planning based on a network flow model, the method comprising:

[0010] Determine evacuation areas for evacuation events;

[0011] Obtain the number of vehicles to be evacuated in the evacuation area, and determine the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated. The evacuation traffic distribution includes the evacuation source point, the evacuation junction point, and the demand for vehicles to be evacuated at each evacuation source point in each evacuation period;

[0012] Generate a path network model according to preset vector geographic data and evacuation traffic distribution of the evacuation area, wherein the path network model includes network nodes composed of evacuation source points, evacuation junction points and intersections in the road network and edges composed of road segments in the road network;

[0013] Constructing an event dynamic evacuation model according to the path network model;

[0014] Based on the event dynamic evacuation model, the minimum cost-maximum flow path from the evacuation source point to the evacuation sink point in the evacuation road network is searched, and the vehicles to be evacuated at each evacuation source point are allocated and / or extended according to the search results until all the vehicles to be evacuated have completed the evacuation plan.

[0015] Optionally, obtaining the number of vehicles to be evacuated in the evacuation area includes:

[0016] Get the total number of residents to be evacuated in the evacuation area;

[0017] The product of the total number of residents to be evacuated and the preset conversion parameter is used as the number of vehicles to be evacuated, and the calculation formula is as follows:

[0018]

[0019] in, Indicates the number of vehicles to be evacuated in the evacuation area, represents the total number of residents to be evacuated, Represents the conversion parameters, The number of private cars in the evacuation area / statistical data of the permanent population in the evacuation area.

[0020] Optionally, determining the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated includes:

[0021] Mark the starting point of the evacuation starting point pre-selected in the evacuation area to the evacuation demand layer;

[0022] The vehicles to be evacuated are merged into the evacuation starting point that is closest to the path of the current vehicles to be evacuated. The evacuation starting point merged with the vehicles to be evacuated is used as the evacuation source point and is numbered and stored in the evacuation source point set. The evacuation starting point without evacuation demand is deleted.

[0023] If the current evacuation event is an evacuation event with warning, the number of evacuation vehicles in each period from the evacuation start time is determined according to the preset evacuation response curve, as follows:

[0024]

[0025] in, It's time The proportion of evacuation traffic at and are the preset model parameters;

[0026] Discretize the evacuation response curve to determine the dynamic evacuation vehicle quantity requirement for each period ;

[0027]

[0028]

[0029] in, express Evacuation source point of the evacuation area at any time The cumulative number of evacuation vehicles required; Indicates the evacuation source Number of evacuated vehicles in Indicates the source point Dynamic evacuation vehicle quantity requirements; is the length of time of a discretized period; Represents a time collection.

[0030] Optionally, generating a path network model according to preset vector geographic data and evacuation traffic distribution of the evacuation area includes:

[0031] Obtaining road network data corresponding to the evacuation area, and constructing a path network model according to road segments and intersections in the road network data, wherein the road network data includes vector road network, road type, road length, road width, road speed limit, direction attribute, intersection, traffic signal and / or traffic capacity;

[0032] The intersections of the path network model are taken as nodes of the network, the road segments are taken as edges of the network, and the evacuation source points and evacuation sink points in the evacuation traffic distribution are identified as source nodes and sink nodes of the network in the path network model, a node list and an edge list are created, the starting point and the end point of each edge are searched, and an adjacency matrix is ​​created to describe the connection relationship between the starting point and the end point of each edge, so as to identify whether the connection relationship between the starting point and the end point of each edge is a direct road;

[0033] Add road attribute descriptions to the edges marked as direct roads, add numbers and the number of vehicles to be evacuated to the source nodes consisting of evacuation source points, and the road attribute descriptions include traffic capacity, travel cost and direction attributes. The travel cost is determined based on the road length, road speed limit and / or road width.

[0034] Optionally, constructing an event dynamic evacuation model according to the path network model includes:

[0035] Establishing an evacuation road network model , create a node set according to the path network model and segment collection ,get , node set It includes a source node set, a sink node set and a common node set, wherein the source nodes correspond one-to-one to the evacuation source points of the path network model, the sink nodes correspond one-to-one to the evacuation sink points of the path network model, the common nodes correspond one-to-one to the intersections of the path network model, and the arc segment set E includes the edges between any two nodes in the node set N whose connection relationship is a direct road;

[0036] Determine the source nodes and sink nodes in the evacuation road network model, configure the numbers, names and / or capacity attributes of the source nodes and sink nodes, and generate virtual super start points and super end points, the super start points are connected to all source nodes in the source node set, and the distance and travel time of the connected arcs are both 0, and the super end points are connected to all sink nodes in the sink node set, and the distance and cost of the connected arcs are both 0;

[0037] Set the evacuation time window of the current evacuation event Discretize into multiple time intervals ;

[0038] A dynamic network flow evacuation model based on the fastest flow is established with the goal of minimizing the evacuation time window T. The dynamic network flow evacuation model includes an objective function and evacuation constraints of the objective function.

[0039] The objective function is as follows:

[0040]

[0041] The evacuation constraints are as follows:

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048] Where T represents the evacuation time window of the evacuation event; m(T) represents the maximum number of evacuation vehicles evacuated within the evacuation time window T; Indicates the number of vehicles to be evacuated in the evacuation area; Represents the node set in the evacuation road network; Represents the set of sink nodes in the evacuation road network; represents the source node set in the evacuation road network; E represents the arc segment set in the evacuation road network; It’s a super starting point; It is the super endpoint; Represents any arc in the road network The travel time; express Leave the node at any time and in Arrival time node Number of evacuated vehicles; Indicates the number of vehicles to be evacuated within the source node; represents the capacity limit of the sink node; express arrive In time, arc of traffic flow.

[0049] Optionally, the event-based dynamic evacuation model searches for the minimum cost-maximum flow path from the evacuation source point to the evacuation junction point in the evacuation road network, and allocates and / or extends the allocation of vehicles to be evacuated at each evacuation source point according to the search results until all vehicles to be evacuated have completed the evacuation plan, including:

[0050] S51, searching for all feasible flows in the evacuation road network, and finding the minimum cost-maximum flow path from each source node in the current source point set to the evacuation sink based on the event dynamic evacuation model;

[0051] S52, determining whether the number of vehicles to be evacuated in each source node is to be allocated to an evacuation group;

[0052] S53: If the current source node has not been assigned an evacuation group, an initial evacuation group assignment is performed for the current source node according to the minimum cost-maximum flow path to obtain the evacuation group assignment result for the current evacuation period. , express The set of evacuation group allocation results at time, represents an evacuation group, is the starting node of the evacuation group, is the flow rate of the evacuation group, is the number of vehicles in the evacuation group;

[0053] S53', if it is determined that the number of vehicles to be evacuated in the current source node has been allocated, then the evacuation flow expansion allocation is performed on the evacuation group in the current source node according to the minimum cost-maximum flow path to obtain the evacuation group allocation result of the current evacuation period ;

[0054] S54, calculation The evacuation delay time and evacuation completion time of each evacuation group in the evacuation group are calculated using the following formula:

[0055]

[0056] in, Indicates evacuation group The evacuation completion time, Indicates evacuation group The delay time, Indicates evacuation group The journey takes time, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow;

[0057] Evacuation delay time calculation formula:

[0058]

[0059] in, represents the evacuation completion time of the benchmark group, represents the travel time of evacuation group k, and the benchmark group refers to the group that is adjacent to the current evacuation group and completes the evacuation before the current evacuation group;

[0060] S55, searching for the target evacuation group that has completed the evacuation earliest among the evacuation groups that have not completed the evacuation, and determining whether the evacuation source node corresponding to the target evacuation group exists in the source point set. If the evacuation source node corresponding to the target evacuation group does not exist in the source point set, repeating the current step, if the evacuation source node corresponding to the target evacuation group exists in the source point set, moving the target evacuation group to the benchmark group cache data set for determining the benchmark group of the next evacuation state, and moving the evacuation source node corresponding to the target evacuation group out of the source point set;

[0061] S56, determine whether the source point set is empty. If the source point set is not empty, return to repeatedly execute the above steps S51-S53 and S54-S55, or S51-S53' and S54-S55, until the evacuation plan is completed when the source point set is empty.

[0062] Optionally, performing evacuation flow expansion allocation for the evacuation group in the current source node according to the minimum cost-maximum flow path includes:

[0063] Calculate the difference between the evacuation group flow after the expansion allocation and the evacuation group flow before the expansion. If the current evacuation group flow has not increased, set the number of vehicles in the evacuation group at the next moment to 0; if the current evacuation group flow increases, estimate the number of vehicles in the evacuation group at the next moment according to the following formula: The calculated value is:

[0064]

[0065] in, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow;

[0066] If the calculated value of the number of vehicles in the evacuation group at the next moment is greater than or equal to the number of vehicles in the current source node, the evacuation group is fully expanded, and the number of vehicles in the evacuation group is set to the number of vehicles in the current source node; if the calculated value of the number of vehicles in the evacuation group at the next moment is less than the number of vehicles in the current source node, the evacuation group is partially expanded, and the number of vehicles in the evacuation group is set to the calculated value, and the number of vehicles in the current source node is updated.

[0067] Optionally, the method further comprises:

[0068] The redistribution process of the evacuation demand of the source node satisfies the following rule: for a certain source node, if Moment and At the moment when both flow distributions are in progress, there is a need for evacuation. The evacuation flow at time will be greater than or equal to Evacuation flow at any time.

[0069] Another aspect of the present invention also provides a dynamic disaster evacuation path planning system based on a network flow model, the system comprising a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor, the processor executing the computer program / instructions to implement the steps of the above-mentioned dynamic disaster evacuation path planning method based on a network flow model.

[0070] Another aspect of the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for dynamic disaster evacuation path planning based on a network flow model.

[0071] Another aspect of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned method for dynamic disaster evacuation path planning based on a network flow model.

[0072] The method and system for dynamic disaster evacuation path planning based on network flow model provided by the embodiment of the present invention searches for the minimum cost-maximum flow from the source point to the sink point as the static optimal traffic allocation based on the network flow theory, allocates the dynamic evacuation demand in the capacity network multiple times by iterative search, and avoids congestion by delaying evacuation, thereby realizing spatiotemporal path planning for evacuation demand. The present invention can effectively cope with the large evacuation demand, long evacuation time, and constantly changing road network traffic during the evacuation process when an event occurs, and is more efficient in alleviating traffic congestion and making full use of the evacuation road network, thereby improving evacuation efficiency.

[0073] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0075] Figure 1 A schematic diagram of a flow chart of a method for dynamic disaster evacuation path planning based on a network flow model provided in an embodiment of the present invention;

[0076] Figure 2 A schematic diagram of the structure of a path network model proposed in an embodiment of the present invention;

[0077] Figure 3 The present invention proposes a flowchart for implementing evacuation flow distribution calculation when performing evacuation planning;

[0078] FIG. 4 (a) is a diagram of an embodiment of the present invention. The flow distribution result of the evacuation group at time Schematic diagram;

[0079] FIG. 4( b ) is a diagram of an embodiment of the present invention. The flow distribution result of the evacuation group at time Schematic diagram. DETAILED DESCRIPTION

[0080] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0081] Those skilled in the art will understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined.

[0082] The present invention proposes a dynamic disaster evacuation path planning method based on a network flow model, which aims to solve the problem of urban road vehicle congestion caused by sudden disaster events. Before introducing the dynamic disaster evacuation path planning method based on a network flow model in an embodiment of the present invention, a brief introduction to the professional technical terms in the present invention is given first:

[0083] Network flow: The amount of resources flowing from a specified source point to one or more specified sink points in a network.

[0084] Maximum flow: The maximum possible flow from a source to a sink that satisfies network constraints.

[0085] Minimum cost-maximum flow: The traffic allocation with the lowest cost while meeting the maximum flow demand.

[0086] Evacuation needs: The number and nature of people or materials that need to be evacuated from a specific area in the event of a natural disaster, accident or other crisis.

[0087] SHP: The full name is Shapefile format, which is a geographic data format used in geographic information system (GIS).

[0088] TIFF: Tagged Image File Format for raster datasets.

[0089] BMP: Bitmap graphics raster dataset format.

[0090] Figure 1 The flowchart of a method for dynamic disaster evacuation path planning based on a network flow model according to an embodiment of the present invention is schematically shown. Figure 1 The dynamic disaster evacuation path planning based on the network flow model of the embodiment of the present invention specifically includes the following steps:

[0091] S11. Determine the evacuation area of ​​the evacuation event.

[0092] In this embodiment, the evacuation area of ​​the evacuation event can be determined by obtaining the impact range and risk level of the event or disaster.

[0093] S12. Obtain the number of vehicles to be evacuated in the evacuation area, and determine the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated. The evacuation traffic distribution includes the evacuation source point, the evacuation junction point, and the demand for vehicles to be evacuated at each evacuation source point in each evacuation period.

[0094] In this embodiment, the number of residents to be evacuated in the disaster area can be calculated based on the WorldPop population grid data corrected by the Seventh National Population Census, and the evacuation area can be superimposed using a geographic information system. On the basis of calculating the number of people to be evacuated in the area, the number of evacuation vehicles is estimated based on the regional vehicle ownership rate. In order to carry out subsequent evacuation planning, it is necessary to divide the evacuation personnel into different areas and delineate the evacuation safety areas to form the evacuation starting point and evacuation end point, so as to grasp the number of people in the evacuation area and determine the number and location of vehicles loaded onto the traffic road network. Vehicles in the evacuation area will be planned to be evacuated in groups.

[0095] S13. Generate a path network model according to preset vector geographic data and evacuation traffic distribution of the evacuation area, wherein the path network model includes network nodes consisting of evacuation source points, evacuation junctions, and intersections in the road network, and edges consisting of road segments in the road network.

[0096] S14. Constructing an event dynamic evacuation model according to the path network model.

[0097] S15, based on the event dynamic evacuation model, searching for the minimum cost-maximum flow path from the evacuation source point to the evacuation junction point in the evacuation road network, and allocating and / or extending the allocation of vehicles to be evacuated at each evacuation source point according to the search results until all vehicles to be evacuated have completed the evacuation planning. Specifically, the present invention calculates dynamic flow allocation, delayed flow allocation and flow extension allocation based on the existing path network model and evacuation demand information, and determines the detailed evacuation results of each evacuation source point, and the results include the evacuation quantity, the selected evacuation path, and the departure time information of each evacuation group.

[0098] In one embodiment of the present invention, obtaining the number of vehicles to be evacuated in the evacuation area in step S12 specifically includes the following steps which are not shown in the drawings:

[0099] S121. Obtain the total number of residents to be evacuated in the evacuation area.

[0100] In a specific example, the steps for obtaining the total number of residents to be evacuated in the evacuation area are as follows:

[0101] Data acquisition: obtain the evacuation area determined in the evacuation warning stage, obtain population grid data, and vehicle ownership rate statistics from the traffic control office.

[0102] Data screening: remove or correct erroneous data, identify and delete duplicate records and outliers in the data set, and convert data in different formats into geographic database formats, including SHP, TIFF, and BMP formats.

[0103] Correcting population data: According to the ratio of the seventh census data and the population grid data, the population correction coefficient of each street is calculated. The correction coefficient is connected with the corresponding street population grid data, that is, pixel value × correction coefficient = new pixel value, to obtain the corrected population raster data.

[0104] Calculate the number of residents to be evacuated: extract the boundary coordinates of the evacuation area and the town and street attribute values ​​to clip the population data in the evacuation area, cluster the population data according to the population density attribute, and use the location of the cluster points to count the total number of residents to be evacuated, and generate an evacuation demand layer. The evacuation demand layer is a vector layer with geographic location, stored in SHP, TIFF, and BMP formats, and is used to store evacuation traffic demand distribution information, including evacuation demand location, number of residents to be evacuated, and evacuation starting point information.

[0105] S122, taking the product of the total number of residents to be evacuated and the preset conversion parameter as the number of vehicles to be evacuated, and the calculation formula is as follows:

[0106]

[0107] in, Indicates the number of vehicles to be evacuated in the evacuation area, represents the total number of residents to be evacuated, Represents the conversion parameters, The number of private cars in the evacuation area / statistical data of the permanent population in the evacuation area.

[0108] In one embodiment of the present invention, determining the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated in step S12 specifically includes the following steps not shown in the drawings:

[0109] S123. Mark the starting point of the evacuation starting point pre-selected in the evacuation area to the evacuation demand layer.

[0110] S124, merge the vehicles to be evacuated to the evacuation starting point closest to the path of the current vehicles to be evacuated, take the evacuation starting point merged with the vehicles to be evacuated as the evacuation source point and store its number in the evacuation source point set, and delete the evacuation starting point without evacuation demand. Specifically, when the evacuation demand is determined to be not allocated to the evacuation source point, merge the demand quantity to the evacuation source point closest to the demand path. When it is determined that the evacuation demand has been allocated to the evacuation source point, execute the next step. After deleting the starting point position without evacuation demand, store the starting point number in the evacuation source point set.

[0111] In this embodiment, for each evacuation source point in the evacuation source point set, the following judgment steps are performed by traversing the evacuation source points to calculate the evacuation traffic distribution for each evacuation source point in the evacuation area and obtain the dynamic evacuation demand:

[0112] Read the preset disaster type database to determine whether the evacuation type is an evacuation with warning. For evacuations without warning, perform the evacuation distribution steps without warning. For evacuations with warning, read the preset evacuation area type database to determine whether the evacuation range is a residential area. If it is determined to be a non-residential area, perform the evacuation distribution steps for non-residential areas with warning. If it is determined to be a residential area, perform the evacuation distribution steps for residential areas with warning.

[0113] Evacuation distribution without warning: In the absence of warning, evacuation will start immediately after receiving the notification, and the evacuation demand quantity is the total evacuation demand quantity in the node. Evacuation distribution of non-residential areas with warning: The evacuation quantity in each period of evacuation with warning conforms to the evacuation response curve. The number of people to be evacuated in non-residential areas is equal to the number of people in that period minus the number of people returning home. Evacuation distribution of residential areas with warning: The number of people to be evacuated in residential areas is equal to the number of people in that period plus the number of people returning home. Both non-residential areas with warning and residential areas with warning can use the preset evacuation response curve to determine the evacuation demand in each period from the start time of evacuation.

[0114] S125. If the current evacuation event is an evacuation event with warning, the number of evacuation vehicles in each time period from the evacuation start time is determined according to a preset evacuation response curve, as follows:

[0115]

[0116] in, It's time The proportion of evacuation traffic at and are the preset model parameters;

[0117] S126. Discretize the evacuation response curve to determine the number of dynamic evacuation vehicles required for each period. ;

[0118]

[0119]

[0120] in, express Evacuation source point of the evacuation area at any time The cumulative number of evacuation vehicles required; Indicates the evacuation source Number of evacuated vehicles in Indicates the source point Dynamic evacuation vehicle quantity requirements; is the length of time of a discretized period; Represents a time set, where n=1, 2, 3, ..., refers to the sequence number.

[0121] In this embodiment, the evacuation distribution calculation includes two parts: the spatial distribution calculation of the evacuation demand and the temporal distribution calculation of the evacuation demand. The spatial distribution of people in the evacuation area at different times of the day is obviously different. For evacuation events with early warning, most people will return to their place of residence to reunite with their families after receiving the evacuation notice, and then start to evacuate, so the place of residence can be used as the spatial distribution of the evacuation demand. The longer the evacuation preparation time, the higher the possibility that people will go home first, so the number of permanent residents in the evacuation area is similar to the number of people to be evacuated in the evacuation area.

[0122] Based on the regional resident population statistics and the number of private cars, we can roughly get the conversion parameter Kptc between the number of people to be evacuated and the number of private cars to be evacuated. The demand for private car evacuation in the disaster area can be obtained by general formula (1.1). The spatial distribution of private car evacuation demand is the same as the spatial distribution of population, and the population location coordinates can be directly obtained.

[0123] The time distribution of evacuation demand refers to the time when evacuees choose to evacuate after receiving the evacuation notice. By recording and counting the evacuation response records of evacuees in previous historical events, the empirical distribution of evacuation demand in the time dimension is obtained. The evacuation vehicles in the evacuation area begin to evacuate one after another according to the judgment of the evacuees, and the number of vehicles entering the traffic network changes with time. The judgment time of the evacuees is considered to be the response time of the evacuation demand. The evacuation decision of the public can be regarded as an orderly discrete problem. The evacuation response rate can be calculated by statistical data or demand model. The Logistic regression curve in formula (1.2) can be used as an evacuation response curve to describe the evacuation ratio of people at different times. The evacuation response curve reflects the change process of the cumulative departure volume in the total evacuation demand over time. The focus here is on the impact of changes in evacuation demand, rather than the exact departure time selection. By discretizing the evacuation response curve, the cumulative increment of the cumulative evacuation demand in each time interval can be further obtained, thereby determining the dynamic evacuation demand of each evacuation interval. Formula (1.3) represents the cumulative number of evacuation vehicles at a certain evacuation source point at a certain time. Discretizing equation (1.3) yields equation (1.4), which represents the relationship between the number of evacuated vehicles and time at a certain evacuation source point.

[0124] In one embodiment of the present invention, generating a path network model according to preset vector geographic data and evacuation traffic distribution of the evacuation area in step S13 specifically includes the following steps not shown in the drawings:

[0125] S131, obtaining road network data corresponding to the evacuation area, and constructing a path network model according to road segments and intersections in the road network data, wherein the road network data includes vector road network, road type, road length, road width, road speed limit, direction attribute, intersection, traffic signal and / or traffic capacity;

[0126] S132, using the intersection of the path network model as the node of the network, the road segment as the edge of the network, and marking the evacuation source point and the evacuation sink point in the evacuation traffic distribution as the source node and sink node of the network in the path network model, creating a node list and an edge list, searching the starting point and the end point of each edge, and creating an adjacency matrix to describe the connection relationship between the starting point and the end point of each edge, so as to identify whether the connection relationship between the starting point and the end point of each edge is a direct road;

[0127] S133. Add road attribute descriptions to the edges marked as direct roads, add numbers and the number of vehicles to be evacuated to the source nodes consisting of evacuation source points, and the road attribute descriptions include traffic capacity, travel cost and direction attributes. The travel cost is determined based on the road length, road speed limit and / or road width.

[0128] In this embodiment, establishing a real path network model requires obtaining vector geographic data and processing data. Specifically, it includes the following contents:

[0129] Vector geographic data collection: Obtain road network data from the Digital Earth open platform, including 9 categories: vector road network, road type, length, width, speed limit, one-way, intersection, traffic signal, and traffic capacity.

[0130] Data preprocessing: import vector geographic data, clean, integrate and convert the data.

[0131] Data cleaning: Identify and delete duplicate roads in the data, fill in missing values, delete abnormal road nodes or isolated points in the data, and standardize data formats such as unified date and time units.

[0132] Data integration: When it is determined that the data comes from multiple data sources, the data is integrated and the data sources are merged and aligned, such as road names and road numbers.

[0133] Data conversion: extract road segment and intersection information and construct network diagram.

[0134] Data meets the accuracy and consistency judgment: Geographic data is cleaned, integrated and converted, and the data is judged. When it is judged that the data is missing or wrong, the data is retrieved again.

[0135] Network construction: Define the intersections in the road network as nodes of the network, define the road segments as edges, add evacuation source points and evacuation sink points as source nodes and sink nodes of the network, create node lists and edge lists, search the starting and ending points of each edge, and create an adjacency matrix to describe the connection relationship between them. Specific creation steps: determine the number of nodes N, initialize the N*N matrix, and fill the matrix; if there is a direct road between node i and node j, A[i][j] can be set to the weight of the road (such as distance, time, cost, etc.). If there is no direct road, A[i][j] can be set to a large number (indicating infinity, that is, unreachable), or a specific mark value, such as 0 or -1.

[0136] Topological analysis: Analyze and process dangling points, pseudo nodes, and overlapping errors in the road network. Dangling points: The endpoints of a line are not connected to other lines, forming road islands; pseudo nodes: False nodes, such as two roads that do not intersect but have nodes; Overlapping errors: The same road cannot overlap itself, and it cannot overlap other roads.

[0137] Attribute definition: Create a road attribute table to define the traffic capacity, travel time, and direction attributes of each edge. Traffic capacity represents the traffic capacity, and travel time represents the travel cost.

[0138] In this embodiment, the road network data used to construct the real path network model mainly includes but is not limited to 9 categories, including vector road network, road type, length, width, speed limit, one-way (i.e., direction attribute), intersection, traffic signal, and traffic capacity. The collected road network data is cleaned, integrated and converted, missing values ​​are filled, and outliers are processed. Data preprocessing operations such as processing ensure the accuracy and consistency of the data. Network construction is to define the intersections in the road network as nodes of the network, and the road segments are defined as edges. It is also necessary to mark the evacuation source points and evacuation sink points in the evacuation traffic distribution as source nodes and sink nodes of the network in the path network model, and establish a connection relationship between them. The purpose of topological analysis is to ensure the connectivity of the network and handle topological errors such as hanging points, pseudo nodes, and overlaps. Finally, traffic capacity, travel time, and direction attributes are assigned to each edge in the network. The capacity and travel cost are calculated by length, speed limit, and road width.

[0139] In a specific example, if Figure 2 As shown in the figure, the path network model consists of 8 nodes and 7 arcs, and the calculation conditions are as follows:

[0140] is the set of evacuation nodes;

[0141] is a set of evacuation sink nodes, which has one element ;

[0142] is the set of evacuation source nodes;

[0143] It is a collection of road segments.

[0144] The evacuation source node is marked with the number of evacuation needs, that is, the number of vehicles to be evacuated, separated by commas and periods. The road sections in the conditions are all bidirectional. The capacity of the road section, that is, the traffic capacity and the travel cost, are obtained through the length, speed limit, and road width attributes. The values ​​in brackets are used on the arcs. Taking the road section from N1 to N2 in the figure as an example, (5,2) means that the traffic capacity of the arc is 5 and the travel time is 2.

[0145] In this embodiment, the number of evacuation demands and evacuation flow rates at different evacuation starting points are different, resulting in different evacuation times at different starting points. Therefore, the evacuation demand and road network flow rate are both dynamically changing. Generally, the static evacuation network flow model only solves the evacuation problem once, and cannot solve the dynamic evacuation demand or obtain the evacuation process. Therefore, the static network flow model is used in the evacuation time window. It is expanded to a dynamic network flow model, divided into discrete time intervals for The dynamic network flow model has the same network structure as the static network flow model, the difference lies in whether a relationship function between time and flow is established for each edge.

[0146] The goal of emergency evacuation is to transfer all vehicles to be evacuated in the area in the shortest time, that is, the fastest evacuation flow problem. The goal of the fastest flow problem is to find the shortest evacuation time window. , transfer a certain number of evacuees. Equation (1.5) is the objective function of the fastest flow problem, a single time interval The flow conservation constraint and arc capacity constraint can be expressed by equations (1.6) and (1.7). The sum of the number of transfers in each time interval should be equal to the total number of evacuees to be transferred.

[0147]

[0148]

[0149]

[0150] in, Representing Arc Traffic on Representing Arc Capacity; Represents the set of edges in the network; is the source node in the specified network, is a sink node; Is a non-negative number.

[0151] In one embodiment of the present invention, the step S14 of constructing the event dynamic evacuation model according to the path network model specifically includes the following steps which are not shown in the drawings:

[0152] S141. Establish evacuation road network model , create a node set according to the path network model and segment collection ,get , node set It includes a source node set, a sink node set and an ordinary node set. The source nodes correspond one-to-one to the evacuation source points of the path network model, the sink nodes correspond one-to-one to the evacuation sink points of the path network model, the ordinary nodes correspond one-to-one to the intersections of the path network model, and the arc set E includes the edges between any two nodes in the node set N whose connection relationship is a direct road.

[0153] Emergency evacuation planning involves multiple source nodes and sink nodes. It is a dynamic evacuation planning problem, and the number of evacuation demands within the source node must also be considered. The fastest flow usually studies the evacuation problem from a single starting point to a single end point. Therefore, this application expands the static network flow model and adds super starting points and super end points to build a dynamic evacuation model based on the fastest flow. Before building the event dynamic evacuation model, the topological structure, node characteristics, and arc characteristics in the path network model are abstracted into a capacity network to establish an evacuation road network model. The evacuation road network model consists of a set of nodes. and segment collection There are three types of network nodes:

[0154] (1) Source node, i.e. the starting point of the evacuation cell. The source node defines the serial number, name, and capacity attributes. All vehicles in the evacuation network enter from the source node, but vehicles in the network cannot enter the source node. Set s as the super starting point, s is connected to all actual source nodes, and the distance and cost of the connecting arc are both 0;

[0155] (2) Sink node, i.e., evacuation exit point, has similar properties to source node. Unlike source node, vehicles entering sink node are considered to have completed evacuation and are not allowed to return to the evacuation network. Set d as the super end point, which is connected to all actual sink nodes, and the distance and cost of the connecting arc are both 0. It should be noted that the capacity of sink node is limited in some cases;

[0156] (3) Ordinary nodes, i.e., intersections in the evacuation network, have no capacity and do not allow vehicles to be stranded.

[0157] S142. Determine the source nodes and sink nodes in the evacuation road network model, configure the numbers, names and / or capacity attributes of the source nodes and sink nodes, and generate virtual super starting points and super end points. The super starting point is connected to all source nodes in the source node set, and the distance and travel time of the connected arcs are both 0. The super end point is connected to all sink nodes in the sink node set, and the distance and cost of the connected arcs are both 0.

[0158] S143: Set the evacuation time window of the current evacuation event Discretize into multiple time intervals .

[0159] S144, establishing a dynamic network flow evacuation model based on the fastest flow with the goal of minimizing the evacuation time window T, the dynamic network flow evacuation model including an objective function and evacuation constraints of the objective function;

[0160] The objective function is as follows:

[0161]

[0162] The evacuation constraints are as follows:

[0163] The constraints on the evacuation requirements of the source node before the evacuation begins are as follows:

[0164] ;

[0165] The evacuation demand from the source node is not greater than the evacuation demand within the node:

[0166]

[0167] The constraints on the number of evacuees received by each evacuation node are as follows:

[0168]

[0169] The road capacity constraint (the maximum traffic flow that each road or arc segment in the traffic network can withstand) is as follows:

[0170]

[0171]

[0172] The flow conservation constraint that the flow into and out of any node is equal is as follows:

[0173]

[0174] Where T represents the evacuation time window of the evacuation event; m(T) represents the maximum number of evacuation vehicles evacuated within the evacuation time window T, that is, the maximum number of evacuation demands that can be evacuated within the time window T; It represents the total evacuation demand, that is, the number of vehicles to be evacuated in the evacuation area; Represents the node set in the evacuation road network; Represents the set of sink nodes in the evacuation road network; represents the source node set in the evacuation road network; E represents the arc segment set in the evacuation road network; It’s a super starting point; It is the super endpoint; Represents any arc in the road network The toll, i.e. the travel time; express Leave the node at any time and in Arrival time node evacuation flow (number of evacuated vehicles); It represents the evacuation demand in the source node, that is, the number of vehicles to be evacuated; represents the capacity limit of the sink node; express arrive In time, arc The traffic capacity is the traffic capacity. Indicates leaving the super starting point at time t and The evacuation flow arriving at node n at time .

[0175] In one embodiment of the present invention, in step S15, the minimum cost-maximum flow path from the evacuation source point to the evacuation sink point in the evacuation road network is searched based on the event dynamic evacuation model, and the vehicles to be evacuated at each evacuation source point are allocated and / or extended according to the search results until all the vehicles to be evacuated have completed the evacuation planning, including the following steps not shown in the accompanying drawings:

[0176] S51, searching for all feasible flows in the evacuation road network, and finding the minimum cost-maximum flow path from each source node in the current source point set to the evacuation sink based on the event dynamic evacuation model.

[0177] Specifically, by searching all feasible flows in the road network, when it is determined that there is an augmenting path, the augmenting path is continued to be searched, and when it is determined that there is no augmenting path, the non-augmented feasible flow with the minimum cost is the minimum cost-maximum flow path.

[0178] S52: Determine whether the number of vehicles to be evacuated in each source node is to be allocated to an evacuation group.

[0179] Specifically, according to the minimum cost-maximum flow path, the vehicles to be evacuated in the source node are initially allocated to determine the path of the evacuation group. When it is determined that the vehicles in the source node have been allocated, the vehicles are reallocated.

[0180] S53: If the current source node has not been assigned an evacuation group, an initial evacuation group assignment is performed for the current source node according to the minimum cost-maximum flow path to obtain the evacuation group assignment result for the current evacuation period. , express The set of evacuation group allocation results at time, represents an evacuation group, is the starting node of the evacuation group, is the flow rate of the evacuation group, is the number of vehicles in the evacuation group;

[0181] S53', if it is determined that the number of vehicles to be evacuated in the current source node has been allocated, then the evacuation flow expansion allocation is performed on the evacuation group in the current source node according to the minimum cost-maximum flow path to obtain the evacuation group allocation result of the current evacuation period ;

[0182] S54, calculation The evacuation delay time and evacuation completion time of each evacuation group in the evacuation group are calculated using the following formula:

[0183]

[0184] in, Indicates evacuation group The evacuation completion time, Indicates evacuation group The delay time, Indicates evacuation group The journey takes time, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow;

[0185] Evacuation delay time calculation formula:

[0186]

[0187] in, represents the evacuation completion time of the benchmark group, It represents the travel time of evacuation group k, and the benchmark group refers to the group adjacent to the current evacuation group and completed the evacuation before the current evacuation group; wherein k=1, 2, 3…, refers to the sequence number.

[0188] S55. Search for the target evacuation group that has completed the evacuation earliest among the evacuation groups that have not yet completed the evacuation, and determine whether the evacuation source node corresponding to the target evacuation group exists in the source point set. If the evacuation source node corresponding to the target evacuation group does not exist in the source point set, repeat the current step. If the evacuation source node corresponding to the target evacuation group exists in the source point set, move the target evacuation group to the benchmark group cache data set for determining the next evacuation state, and move the evacuation source node corresponding to the target evacuation group out of the source point set; wherein, the evacuation group that has not yet completed the evacuation refers to an evacuation group in the allocated evacuation groups whose evacuation completion time is greater than the evacuation completion time of the benchmark group in the current time period.

[0189] S56, determine whether the source point set is empty. If the source point set is not empty, return to repeatedly execute the above steps S51-S53 and S54-S55, or S51-S53' and S54-S55, until the evacuation plan is completed when the source point set is empty.

[0190] Furthermore, in step S53', the evacuation flow expansion allocation is performed for the evacuation group in the current source node according to the minimum cost-maximum flow path, specifically including:

[0191] Calculate the difference between the evacuation group flow after the expansion allocation and the evacuation group flow before the expansion. If the current evacuation group flow has not increased, set the number of vehicles in the evacuation group at the next moment to 0; if the current evacuation group flow increases, estimate the number of vehicles in the evacuation group at the next moment according to the following formula: The calculated value is:

[0192]

[0193] in, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow;

[0194] If the calculated value of the number of vehicles in the evacuation group at the next moment is greater than or equal to the number of vehicles in the current source node, the evacuation group is fully expanded, and the number of vehicles in the evacuation group is set to the number of vehicles in the current source node; if the calculated value of the number of vehicles in the evacuation group at the next moment is less than the number of vehicles in the current source node, the evacuation group is partially expanded, and the number of vehicles in the evacuation group is set to the calculated value, and the number of vehicles in the current source node is updated.

[0195] When performing evacuation planning in the embodiment of the present invention, the evacuation flow distribution calculation process is as follows: Figure 3 As shown in the figure, first input the road network, source nodes and node sinks. Then search for the minimum cost-maximum flow from the source node to the sink node, that is, the distribution of traffic on each edge, obtain the specific values ​​of inflow and outflow of each edge, allocate or expand the allocation of evacuation vehicles according to the search results, and calculate the delay time and completion time. Since the number of evacuation needs in the source point is limited, all evacuation vehicles in a certain source point are evacuated, and the road network traffic state changes. At this time, search for the minimum cost-maximum flow from the remaining source points to the sink point, and allocate evacuation vehicles. Iterate the search and allocation process until all vehicles are evacuated.

[0196] In order to minimize the evacuation time window and complete the evacuation of all vehicles, spatiotemporal path planning is performed for the evacuation needs. The planning is divided into two steps: initial allocation and redistribution. Initial allocation refers to the first traffic allocation of vehicles in an evacuation group, and redistribution refers to the re-allocation of traffic to the group of vehicles in order to maximize the use of road capacity after the traffic status of the road network changes. This method consists of three key algorithms: dynamic flow allocation algorithm, evacuation group delay algorithm, and flow expansion allocation algorithm. Among them, the dynamic flow allocation algorithm is responsible for flow allocation and iteration in evacuation planning; the evacuation group delay algorithm is used to calculate the delayed evacuation time and evacuation completion time of the evacuation group; the flow expansion allocation algorithm is used to redistribute the evacuation group.

[0197] Variables involved in the algorithm:

[0198] Represents the evacuation road network, including a set of nodes and segment collection ;

[0199] represents the set of nodes in the road network, Represents a node in the road network;

[0200] represents the set of exit nodes, represents the exit node of the road network, and The relationship can be expressed as ;

[0201] represents the starting node set, represents one of the starting nodes, and the number of vehicles to be evacuated in the node is , and The relationship can be expressed as ;

[0202] represents the set of arcs in the road network. Represents an arc segment in a collection;

[0203] is the process variable of the algorithm, representing the set of evacuation groups, represents an evacuation group, is the starting node of the evacuation group, is the flow rate of the evacuation group, is the number of vehicles in the evacuation group;

[0204] Indicates different network status moments;

[0205] is the process variable of the algorithm, indicating The set of evacuation group allocation results at the time;

[0206] It is the process variable of the algorithm and is used to store the evacuation group.

[0207] The dynamic traffic allocation algorithm is as follows:

[0208] Table 1 Dynamic traffic allocation algorithm

[0209]

[0210] In Algorithm 1, steps 1-2 search for the minimum cost-maximum flow. The augmentation algorithm searches for the minimum cost-maximum flow from the source to the sink. The principle is to continuously search for the minimum cost augmenting path from the source to the sink in the residual network until there is no augmenting path, at which point the network reaches the maximum flow. Step 3 assigns flow and paths to the source, and assigns evacuees at the source to evacuees; Step 4 finds the group that has completed evacuation in the current state, and marks the benchmark group for the next evacuation state. Steps 5-6 determine whether the algorithm is finished. The dynamic flow allocation algorithm achieves the maximum road network traffic volume in a time state during the evacuation process, reducing , which is the time the exit is occupied. When dealing with multi-source and multi-sink path search problems, super sinks and super sources can greatly improve the search efficiency.

[0211] The evacuation group delay algorithm is as follows:

[0212] Table 2 Evacuation group delay algorithm

[0213]

[0214] The spatial level path planning is a static traffic allocation under a certain state of evacuation demand. Due to the capacity constraints of the road network in time and space, Algorithm 2 calculates the departure time of the evacuation group to minimize , which is the interval time between evacuation groups. Let the evacuation group wait at the starting point to avoid congestion during the evacuation. Step 1 in Algorithm 2 is achieved by traversing Determine the benchmark group. The benchmark group is the group that is adjacent to the evacuation group and completes evacuation before the evacuation group. It is used to calculate the delay time of the evacuation group. Step 2-3 calculates the delayed departure time and completion time of the evacuation group according to the formula. Assume that there is an evacuation group Evacuate from the source to the sink, The evacuation completion time is composed of three parts: delay time, journey time, and evacuation group time. Evacuation completion time calculation formula:

[0215]

[0216] in, express The evacuation completion time, express The delay time, express The journey takes time, express The number of vehicles in express evacuation flow.

[0217] The calculation of the evacuation group delay time is based on the evacuation completion time of the baseline group. The benchmark group is , The delay time calculation formula is:

[0218]

[0219] in, represents the evacuation completion time of the benchmark group, represents the travel time of the evacuation group. If Equation (1.9) is calculated, If the value of is less than zero, then .

[0220] According to steps 5-6 of Algorithm 1, if all vehicles in a starting node are evacuated, the starting node will be regarded as a normal node. Different time points are defined according to the evacuation completion status of the starting node in the network. The evacuation status is calculated and the flow distribution at each moment is calculated. Different evacuation groups depart from the same source node at different times. The traffic allocated to these evacuation groups may be different. It is necessary to discuss the number of evacuation vehicles and completion time of these groups in different situations.

[0221] The redistribution process of the evacuation demand of the source node has the following rules: for a certain source node, if Moment and At the moment when both flow distributions are in progress, there is a need for evacuation. The evacuation flow at time will be greater than or equal to As shown in Figure 4 (a) and (b), Figure 4 (a) shows The flow distribution result of the evacuation group at time Schematic diagram, Figure 4(b) shows The flow distribution result of the evacuation group at time Schematic diagram. There is always From the source node Set off. Time source node The vehicles in the evacuation area are evacuated, and some path capacity is available in the evacuation network. According to the augmenting path principle in the minimum cost-maximum flow, the nearest Whether there is a feasible flow to the sink, then search for the farther If from There is a feasible flow to the sink, and the available road capacity is allocated to the source nodes. Departure ,so Finally, from the source node Departure Compared to The flow rate increases.

[0222] The flow expansion allocation algorithm is as follows:

[0223] Table 3 Traffic expansion allocation algorithm

[0224]

[0225] The flow expansion allocation can calculate the number of evacuation vehicles and the evacuation completion time of the reallocated evacuation group. If the reallocated evacuation flow remains unchanged, the flow of the evacuation group at the current moment is maintained, and the number of evacuation vehicles of the evacuation group at the next moment is considered to be 0. If the flow increases, it means that the path capacity is not fully utilized, and the source node needs to be expanded. , we need to subtract the number of vehicles evacuated at the current moment from the total number of vehicles in the starting node at the current moment, as shown in formula (1.10), and derive formula (1.12). The evacuation group is allocated at the next moment The number of evacuation vehicles is determined by the current evacuation group The evacuation completion status will be determined. The benchmark group When the evacuation is completed, the evacuation group at the current moment There are three evacuation situations. Partial evacuation means that some vehicles in the evacuation group have completed evacuation at the current moment. Less than , the next moment the group starting from the starting node is partially expanded, The number of vehicles is equal to .if Evacuation started or not evacuated, indicating that no vehicle in the evacuation group has completed evacuation at the current moment. Greater than or equal to , the next moment the group starting from the starting node is fully expanded (representing that in the current period, Not a single vehicle was evacuated.) The number of vehicles is equal to Finally, the evacuation group was reassigned The calculation of the number of evacuated vehicles is shown in formula (1.13), and the calculation of the evacuation completion time is shown in formula (1.8).

[0226] The calculation formula for the number of vehicles to be evacuated in the starting node at the next moment is:

[0227]

[0228] Simplifying, we get:

[0229]

[0230] in, represents the number of vehicles to be evacuated in the starting node at the next moment, Indicates the number of vehicles in the starting node at the current time. express The evacuation flow, yes The benchmark group The evacuation completion time. Because all vehicles in the starting node at the current moment are assigned to the evacuation group, , the calculation formula for the number of vehicles to be evacuated in the starting node at the next moment can be simplified as:

[0231]

[0232] According to the evacuation situation of the evacuation group at the current moment, the final The calculation formula is:

[0233]

[0234] express The number of vehicles in express evacuation flow.

[0235] As shown in Table 4, the calculation of evacuation path planning based on network flow is an iterative search process. First, the algorithm searches for the minimum cost-maximum flow from the source node to the sink node, and obtains a flow consisting of two paths N1-N2-N6-N8 and N4-N5-N6-N8 with a flow rate of 5. Second, two evacuation teams are assigned , determine the flow and required number of evacuation groups, calculate The delayed evacuation time and completion time of the group. Then, find The first group to complete evacuation , remove it , and added to .Bundle The source node from , and get the table The program variables at that moment. Then, The flow paths N3-N2-N6-N8 and N4-N5-N6-N8 are obtained by searching at this moment. Since the flow and path allocated from the source point N4 have not changed, the demand number is 0. There is only one evacuation group at any time Calculate the delay time and completion time of the group and get the Finally, the iterative search continues until There is no element in the algorithm, and the algorithm ends. It should be noted that the example The flow path N4-N5-N6-N8 is obtained by searching at this moment. At this time, the flow allocated from source point N4 increases. According to Algorithm 3, the source point N4 is expanded and allocated. It is calculated that the group is partially expanded, and the expansion demand quantity is 45, so The actual demand quantity within the group is 55.

[0236] Table 4 Program variables in the algorithm calculation process

[0237] time Traffic distribution path Evacuation Group Program variables <![CDATA[T1]]> <![CDATA[GP Ti ={g1,g2}]]> <![CDATA[g1:(N1-N2-N6-N8)g2:(N4-N5-N6-N8)]]> <![CDATA[g1=(s1,5,20)g2=(s3,5,100)]]> <![CDATA[GP={g2}S={s2,s3,s4}GC={g1}]]> <![CDATA[T2]]> <![CDATA[GP Ti ={g3}]]> <![CDATA[g3:(N3-N2-N6-N8)]]> <![CDATA[g3=(s2,5,50)]]> <![CDATA[P={g2}S={s3,s4}GC={g1,g . }]]> <![CDATA[T3]]> <![CDATA[GP Ti ={g4}]]> <![CDATA[g4:(N4-N5-N6-N8)]]> <![CDATA[g4=(s3,10,45)]]> <![CDATA[GP={}S={s4}GC={g1,g3,g4}]]> <![CDATA[T4]]> <![CDATA[GP Ti ={g5}]]> <![CDATA[g5:(N7-N5-N6-N8)]]> <![CDATA[g5=(s4,10,20)]]> <![CDATA[GP={}S={}GC={g1,g3,g4,g5}]]>

[0238] The final output evacuation planning results are shown in Table 5.

[0239] Table 5 Evacuation path planning results

[0240]

[0241] The dynamic disaster evacuation path planning method based on the network flow model provided by the embodiment of the present invention takes into account the capacity limitation of the road network, that is, each road or road section has its maximum capacity to carry vehicles. By planning the evacuation path with capacity limitation, it is possible to effectively avoid excessive concentration of vehicles in certain sections, thereby avoiding traffic congestion. The method of the present invention adjusts the evacuation route for other evacuation vehicles according to the road capacity, improves the road utilization rate, and ensures the rationality and effectiveness of the evacuation path. The evacuation path planned by the present invention avoids traffic congestion, and the evacuation path planning is more reasonable. It can more effectively utilize the existing road network resources, effectively improve the evacuation efficiency, reduce the evacuation time, and achieve rapid, orderly, and safe evacuation.

[0242] The present invention constructs an event dynamic evacuation model, and combines the characteristics of large evacuation demand, long evacuation time, and constantly changing road network traffic during the evacuation process, to propose an efficient and congestion-avoiding evacuation path planning method. This method realizes dynamic traffic distribution based on three key algorithms: dynamic traffic distribution, evacuation group delay, and traffic expansion distribution. Compared with the existing evacuation planning methods that fail to consider maximizing the use of the evacuation road network while avoiding congestion, it is more efficient in alleviating traffic congestion and making full use of the evacuation road network.

[0243] Another aspect of the present invention also provides a dynamic disaster evacuation path planning system based on a network flow model, the system comprising a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor, the processor executing the computer program / instructions to implement the steps of the above-mentioned dynamic disaster evacuation path planning method based on a network flow model.

[0244] According to a third aspect of the present invention, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of the dynamic disaster evacuation path planning method based on the network flow model are implemented.

[0245] According to a fourth aspect of the present invention, a computer program product is provided, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned method for dynamic disaster evacuation path planning based on a network flow model.

[0246] The method and system for dynamic disaster evacuation path planning based on network flow model provided by the embodiment of the present invention searches for the minimum cost-maximum flow from the source point to the sink point as the static optimal traffic allocation based on the network flow theory, allocates the dynamic evacuation demand in the capacity network multiple times by iterative search, and avoids congestion by delaying evacuation, thereby realizing spatiotemporal path planning for evacuation demand. The present invention can effectively cope with the large evacuation demand, long evacuation time, and constantly changing road network traffic during the evacuation process when an event occurs, and is more efficient in alleviating traffic congestion and making full use of the evacuation road network, thereby improving evacuation efficiency.

[0247] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, any one of the claimed embodiments may be used in any combination.

[0248] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dynamic disaster evacuation path planning method based on a network flow model, characterized in that: The method comprises: Determine evacuation areas for evacuation events; Obtain the number of vehicles to be evacuated in the evacuation area, and determine the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated. The evacuation traffic distribution includes the evacuation source point, the evacuation junction point, and the demand for vehicles to be evacuated at each evacuation source point in each evacuation period; Generate a path network model according to preset vector geographic data and evacuation traffic distribution of the evacuation area, wherein the path network model includes network nodes composed of evacuation source points, evacuation junction points and intersections in the road network and edges composed of road segments in the road network; Constructing an event dynamic evacuation model according to the path network model; Based on the event dynamic evacuation model, the minimum cost-maximum flow path from the evacuation source point to the evacuation sink point in the evacuation road network is searched, and the vehicles to be evacuated at each evacuation source point are allocated and / or extended according to the search results until all the vehicles to be evacuated have completed the evacuation plan; The constructing of the event dynamic evacuation model according to the path network model comprises: Establishing an evacuation road network model , create a node set according to the path network model and segment collection ,get , node set It includes a source node set, a sink node set and a common node set, wherein the source nodes correspond one-to-one to the evacuation source points of the path network model, the sink nodes correspond one-to-one to the evacuation sink points of the path network model, the common nodes correspond one-to-one to the intersection nodes of the path network model, and the arc segment set E includes the edges between any two nodes in the node set N whose connection relationship is a direct road; Determine the source nodes and sink nodes in the evacuation road network model, configure the numbers, names and / or capacity attributes of the source nodes and sink nodes, and generate virtual super start points and super end points, the super start points are connected to all source nodes in the source node set, and the distance and travel time of the connected arcs are both 0, and the super end points are connected to all sink nodes in the sink node set, and the distance and cost of the connected arcs are both 0; Set the evacuation time window of the current evacuation event Discretize into multiple time intervals ; A dynamic network flow evacuation model based on the fastest flow is established with the goal of minimizing the evacuation time window T. The dynamic network flow evacuation model includes an objective function and evacuation constraints of the objective function. Among them, the objective function is as follows: The evacuation constraints are as follows: Where T represents the evacuation time window of the evacuation event; m(T) represents the maximum number of evacuation vehicles evacuated within the evacuation time window T; Indicates the number of vehicles to be evacuated in the evacuation area; Represents the node set in the evacuation road network; Represents the set of sink nodes in the evacuation road network; Represents the source node set in the evacuation road network; E represents the set of arcs in the evacuation road network; It’s a super starting point; It is the super endpoint; Represents any arc in the road network The travel time; express Leave the node at any time and in Arrival time node Number of evacuated vehicles; Indicates the number of vehicles to be evacuated within the source node; represents the capacity limit of the sink node; express arrive In time, arc Traffic flow; The event-based dynamic evacuation model searches for the minimum cost-maximum flow path from the evacuation source point to the evacuation sink point in the evacuation road network, and allocates and / or extends the allocation of vehicles to be evacuated at each evacuation source point according to the search results until all vehicles to be evacuated have completed the evacuation plan, including: S51, searching for all feasible flows in the evacuation road network, and finding the minimum cost-maximum flow path from each source node in the current source point set to the evacuation sink based on the event dynamic evacuation model; S52, determining whether the number of vehicles to be evacuated in each source node is to be allocated to an evacuation group; S53: If the current source node has not been assigned an evacuation group, an initial evacuation group assignment is performed for the current source node according to the minimum cost-maximum flow path to obtain the evacuation group assignment result GP for the current evacuation period. Ti , express The set of evacuation group allocation results at time, represents an evacuation group, is the starting node of the evacuation group, is the flow rate of the evacuation group, is the number of vehicles in the evacuation group; S54, Calculate GP Ti The evacuation delay time and evacuation completion time of each evacuation group in the evacuation group are calculated using the following formula: in, Indicates evacuation group The evacuation completion time, Indicates evacuation group The delay time, Indicates evacuation group The journey takes time, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow; Evacuation delay time calculation formula: in, represents the evacuation completion time of the benchmark group, It represents the travel time of the evacuation group. The benchmark group refers to the group that is adjacent to the current evacuation group and completes the evacuation before the current evacuation group. S55, searching for the target evacuation group that has completed the evacuation earliest among the evacuation groups that have not completed the evacuation, and determining whether the evacuation source node corresponding to the target evacuation group exists in the source point set. If the evacuation source node corresponding to the target evacuation group does not exist in the source point set, repeating the current step, and if the evacuation source node corresponding to the target evacuation group exists in the source point set, moving the target evacuation group to the benchmark group cache data set for determining the benchmark group of the next evacuation state, and moving the evacuation source node corresponding to the target evacuation group out of the source point set; S56, judging whether the source point set is empty, if the source point set is not empty, returning to repeatedly execute the above steps S51 to S55, until the evacuation planning is completed when the source point set is empty.

2. The method according to claim 1, characterized in that The method of obtaining the number of vehicles to be evacuated in the evacuation area includes: Get the total number of residents to be evacuated in the evacuation area; The product of the total number of residents to be evacuated and the preset conversion parameter is used as the number of vehicles to be evacuated. The calculation formula is as follows: in, Indicates the number of vehicles to be evacuated in the evacuation area, represents the total number of residents to be evacuated, Represents the conversion parameters, The number of private cars in the evacuation area / statistical data of the permanent population in the evacuation area.

3. The method according to claim 1, characterized in that Determining the evacuation traffic distribution of the evacuation area according to the distribution of the vehicles to be evacuated includes: Mark the starting point of the evacuation starting point pre-selected in the evacuation area to the evacuation demand layer; The vehicles to be evacuated are merged into the evacuation starting point that is closest to the path of the current vehicles to be evacuated. The evacuation starting point merged with the vehicles to be evacuated is used as the evacuation source point and is numbered and stored in the evacuation source point set. The evacuation starting point without evacuation demand is deleted. If the current evacuation event is an evacuation event with warning, the number of evacuation vehicles in each period from the evacuation start time is determined according to the preset evacuation response curve, as follows: in, It's time The proportion of evacuation traffic at and are the preset model parameters; Discretize the evacuation response curve to determine the dynamic evacuation vehicle quantity requirement for each period ; in, express Evacuation source point of the evacuation area at any time The cumulative number of evacuation vehicles required; Indicates the evacuation source Number of evacuated vehicles in Indicates the source point Dynamic evacuation vehicle quantity requirements; is the length of time of a discretized period; Represents a time collection.

4. The method according to claim 1, characterized in that: The generating of the path network model according to the preset vector geographic data and the evacuation traffic distribution of the evacuation area comprises: Obtain the road network data corresponding to the evacuation area, and construct a path network model based on the road segments and intersections in the road network data; The intersections of the path network model are taken as nodes of the network, the road segments are taken as edges of the network, and the evacuation source points and evacuation sink points in the evacuation traffic distribution are identified as source nodes and sink nodes of the network in the path network model, a node list and an edge list are created, the starting point and the end point of each edge are searched, and an adjacency matrix is ​​created to describe the connection relationship between the starting point and the end point of each edge, so as to identify whether the connection relationship between the starting point and the end point of each edge is a direct road; Add road attribute descriptions to the edges marked as direct roads, add numbers and the number of vehicles to be evacuated to the source nodes composed of evacuation source points, and the road attribute descriptions include traffic capacity, travel cost, and direction attributes.

5. The method according to claim 1, characterized in that The method further comprises: S53', if it is determined that the number of vehicles to be evacuated in the current source node has been allocated, then the evacuation flow expansion allocation is performed on the evacuation group in the current source node according to the minimum cost-maximum flow path, and the evacuation group allocation result GP of the current evacuation period is obtained. Ti .

6. The method according to claim 5, characterized in that The method of performing evacuation flow expansion allocation for the evacuation group in the current source node according to the minimum cost-maximum flow path includes: Calculate the difference between the evacuation group flow after the expansion allocation and the evacuation group flow before the expansion. If the current evacuation group flow has not increased, set the number of vehicles in the evacuation group at the next moment to 0; if the current evacuation group flow increases, estimate the number of vehicles in the evacuation group at the next moment according to the following formula: The calculated value is: in, Indicates evacuation group The number of vehicles in Indicates evacuation group evacuation flow; It indicates the evacuation completion time of the benchmark group to be reallocated to the evacuation group at the next moment; If the calculated value of the number of vehicles in the evacuation group at the next moment is greater than or equal to the number of vehicles in the current source node, the evacuation group is fully expanded, and the number of vehicles in the evacuation group is set to the number of vehicles in the current source node; if the calculated value of the number of vehicles in the evacuation group at the next moment is less than the number of vehicles in the current source node, the evacuation group is partially expanded, and the number of vehicles in the evacuation group is set to the calculated value, and the number of vehicles in the current source node is updated.

7. The method according to claim 5, characterized in that The method further comprises: The redistribution process of the evacuation demand of the source node satisfies the following rule: for a certain source node, if Moment and At the moment when both flow distributions are in progress, there is a need for evacuation. The evacuation flow at the moment will be greater than or equal to Evacuation flow at any time.

8. A dynamic disaster evacuation path planning system based on a network flow model, the system comprising a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, characterized in that: The processor executes the computer program / instructions to implement the steps of the method according to any one of claims 1-7.

Images