A method for identifying key sections of a public transportation network, an electronic device, and a storage medium

Abstract

The present invention discloses a method for identifying key sections of a public transportation network, an electronic device, and a storage medium. The method extracts the running trajectories of each bus line based on the GPS data of buses within the research scope; obtains the urban road network within the research scope, and constructs a map matching algorithm based on a hidden Markov model; applies the constructed map matching algorithm to match the running trajectories of each bus line with the actual road network, and obtains the bus route information of each section in the road network; calculates the key index of each section in the road network based on the extracted bus route information, lane information, surrounding population and POI information, and the departure frequency of each bus line, and identifies the key sections in the public transportation network. The present invention takes into account the spatiotemporal integration of the public transportation network and the urban road network, accurately identifies the key sections in the public transportation network from the road network level, and provides decision support for the network planning, station layout, and transportation capacity allocation of public transportation.

Description

Technical Field

[0001] The present invention relates to a method for identifying a key road section of a public transportation network, electronic equipment and a storage medium, and belongs to the technical field of traffic control systems for road vehicles. Background Art

[0002] Some sections of the bus network have a high density of bus routes, so any emergencies (such as natural disasters or congestion) would significantly impact the normal operation of the entire bus system. Some sections are also densely populated, and the operation of bus routes on these sections has a significant impact on residents' daily travel. Furthermore, some sections have limited lanes, and due to the fixed nature of bus routes, road maintenance or traffic control on these sections can easily lead to paralysis of the bus network. The impact of emergencies on bus operations along these key sections can make it difficult to ensure timely travel for passengers, prompting more citizens to choose private cars as their means of transportation, further exacerbating urban road congestion.

[0003] Currently, research on identifying critical bus network links primarily focuses on the station level, while relatively little research has been conducted on the network level. Only a few studies have used bus speeds on a given route as a criterion for determining whether it is a critical link in the bus network. This has led to irrational bus network planning.

[0004] Therefore, how to identify key sections in the bus network is of vital importance for bus network planning. Summary of the Invention

[0005] Purpose: To overcome the deficiencies in the prior art, the present invention provides a method, electronic device, and storage medium for identifying key sections in a public transportation network. These methods take into account the spatiotemporal integration of public transportation networks and urban road networks, accurately identify key sections in a public transportation network from a road network perspective, and provide decision support for bus network planning, station layout, and capacity allocation.

[0006] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is:

[0007] In a first aspect, a method for identifying key road sections in a public transportation network specifically includes:

[0008] Step 1: Obtain the geographic information of buses within the study area, and extract the up and down running trajectories of each bus line based on the location data of the first and last bus stops.

[0009] Step 2: Obtain urban road network data within the study area and establish an urban road network.

[0010] Step 3: Construct a map matching algorithm based on the hidden Markov model to calculate the matching probability between the trajectory points of the up and down running trajectories of each bus line and the road sections in the urban road network.

[0011] Step 4: Considering the vector relationship between adjacent trajectory points in the bus route trajectory, calculate the transfer probability of the vehicle from one road segment to another at the intersection.

[0012] Step 5: Based on the matching probability and transition probability, set the parameters in the hidden Markov model to construct a map matching algorithm between bus routes and urban road networks, and obtain the bus route information passing through each section in the urban road network.

[0013] Step 6: Use the spatial connection method to obtain the population and POI information around each road section.

[0014] Step 7: Based on the extracted bus route information, lane information, surrounding population and POI information of each road section, as well as the departure frequency of each bus route, calculate the key index of each road section in the road network, and identify the key sections in the bus network based on the key index of each road section.

[0015] Optionally, step 1 specifically includes:

[0016] Step 1.1: Read bus route information data within a certain time period. The bus route information data includes but is not limited to: route name, number of vehicles, and trip number.

[0017] Step 1.2: Read the bus trajectory information within the research area. The bus trajectory information data includes but is not limited to: line name, up and down lines, starting station, end station, and trajectory coordinates.

[0018] Step 1.3: Based on the bus route information data, extract the travel trajectory G of each bus route in the up and down directions from the bus trajectory information data t , and count the total number of trips p of each line in the time period to obtain the geographic information of the bus lines within the research scope, which includes the line name, up and down lines, starting and ending stations, total number of trips, and longitude and latitude information.

[0019] Optionally, step 2 specifically includes:

[0020] Step 2.1: Based on the bus's trajectory G t , extract the longitude range [lon min , lon max ], latitude range [lat min , lat max ], define this range as the study area, where lon min, lon max Divided into the minimum and maximum values of longitude in all bus track points, lat min , lat max are the minimum and maximum latitudes of all bus trajectory points respectively.

[0021] Step 2.2: Download the map data for the study area and extract the map edge and node information. A node is the endpoint of two adjacent roads in the map, and its endpoint information includes the node number and node coordinates. An edge is a road in the map, and its road information includes the starting point number, end point number, road name, number of lanes, and edge coordinates.

[0022] Step 2.3: According to the node coordinate set G nodes and the edge coordinate set G edges , create a map object G o As a city road network, G o ={G nodes ,G edges}.

[0023] Optionally, step 3 specifically includes:

[0024] Step 3.1: Rasterize the urban road network data to obtain the node coordinate set G nodes The circle area within a certain range of the node in the center is used as the error area, and the radius of the circle is set to ir meters. The road sections with a distance greater than ir meters are eliminated.

[0025] Step 3.2: Use the projection method to determine the candidate points, calculate the actual distance d between the trajectory point in the bus trajectory information and all the matching road sections within a range of i meters from the node, and compare the calculation results to find the matching road section closest to the trajectory point.

[0026] Step 3.3: Set the positioning error E in the map matching algorithm d , positioning error E d The expression is as follows:

[0027] ,

[0028] in, is the standard deviation of bus trajectory points, is pi, is a natural constant.

[0029] Step 3.4: Extract the bus's driving direction information based on the bus's up and down information, and calculate the direction error E α , direction error E α The expression is as follows:

[0030] ,

[0031] Among them, α represents the angle between the direction of the bus trajectory point in its trajectory and the direction of the road section to be matched.

[0032] Step 3.5: According to the bus positioning error E d and the direction error between the bus GPS trajectory and the road section to be matched , calculate the matching probability E of the bus trajectory point and its to-be-matched road section as the actual driving road section of the bus. The expression of E is as follows:

[0033] .

[0034] Optionally, step 4 specifically includes:

[0035] Step 4.1: Get the Euclidean distance d between two adjacent points in the bus trajectory n-1，n The shortest distance D between the candidate points corresponding to the two trajectory points n-1，n , and calculate the similarity S between the bus trajectory and the road section to be matched r , similarity S r The expression is as follows:

[0036] ,

[0037] Step 4.2: Calculate the positioning change trend C between two adjacent trajectory points p , positioning change trend C p The expression is as follows:

[0038] ,

[0039] Among them, d n-1 and d n are the distances between two adjacent bus trajectory points and their corresponding matching road sections.

[0040] Step 4.3: Based on the similarity S between the bus trajectory and the road segment to be matched r and the positioning change trend C between two adjacent trajectory points p , calculate the transition probability T of a bus traveling from one road segment to another at an intersection. The expression of the transition probability T is as follows:

[0041] ,

[0042] Among them, t r is the similarity S between the bus trajectory and the road section to be matched r The coefficient of t p is the positioning change trend C between two adjacent trajectory points pThe coefficient of .

[0043] Optionally, step 5 specifically includes:

[0044] Step 5.1: Set the parameter id to specify the maximum distance between the bus trajectory and the road section to be matched, and set the parameter di max Used to specify the maximum distance between each trajectory point and the corresponding starting position, setting parameters Used to specify the standard deviation value of bus trajectory points, set the parameter st n It is used to specify whether bus trajectories are allowed to have parts that do not directly correspond to the road network, and set the parameter p stn It is used to stipulate the proportion of bus trajectories that do not directly correspond to the road network, and to construct a map matching algorithm between bus routes and the actual road network based on the hidden Markov model.

[0045] Step 5.2: Randomly select a bus trajectory sample, divide it into multiple small segments, and compare the extracted bus trajectory coordinates with the urban road network G o Projected into the same coordinate system, the trajectory is matched with the urban road network G using the map matching algorithm of bus routes and actual road networks. o The actual path in the is matched, the path matching result is extracted, and the path matching of a bus trajectory is completed.

[0046] Step 5.3: Repeat step 5.2 to match all bus trajectories within the study area with the actual road network to obtain the urban road network G o The bus route information of each road section in the urban road network includes the route name, up and down lines, number of departures, road name, road starting point number, road end point number, and number of lanes.

[0047] Step 5.4: Link the bus route information of each road section in the urban road network with the bus route information within the study area in time period q, and calculate the number of trips p of all bus routes on road section m in time period q. m , and the public transportation situation B on section m m .

[0048] Among them, the number of passes p m The expression is as follows:

[0049]

[0050] Among them, I is the total number of bus routes passing through a certain road section m, p id is the total number of bus departures for route id within time period q.

[0051] Public transport situation B m The expression is as follows:

[0052]

[0053] Among them, l m is the number of lanes in road section m.

[0054] Optionally, step 6 specifically includes:

[0055] Step 6.1: Take the trajectory points of each road section in the actual road network as the center of the circle, set the road influence radius to θ meters, and construct a road buffer zone.

[0056] Step 6.2: Read the POI data within the research scope, project the POI data and the buffer zone of each road section in a unified coordinate system to obtain the POI information in the buffer zone of each road section, and perform statistical analysis on the number of POI information in each buffer zone. The number of POIs in the buffer zone corresponding to road section m is counted as poi m .

[0057] Step 6.3: Read the population data of each traffic zone in the study area. Based on the ratio of the area of each road buffer zone to the area of the traffic zone, calculate the residential population and working population data in the buffer zone. The residential population in the buffer zone corresponding to road segment m is calculated as lp. m , the number of working population is counted as wp m .

[0058] Optionally, step 7 specifically includes:

[0059] Step 7.1: Link the bus route information of each road section in the road network with the various POI data around each road section, as well as the resident population and working population data. Build a judgment matrix X with the bus traffic situation of each road section as the first column, the number of POIs in the corresponding buffer area as the second column, the number of residents in the corresponding buffer area as the third column, and the number of working population in the corresponding buffer area as the fourth column. Normalize the judgment matrix X to obtain the elements of the normalized judgment matrix X. .

[0060] Among them, the judgment matrix X expression is:

[0061] ,

[0062] Among them, x ij represents the element in the i-th row and j-th column of the judgment matrix X, m is the number of road sections in the study area, i represents the i-th road section, and j represents the column number.

[0063] Step 7.2: Calculate the dimensional entropy H of the four columns of indicators j , and its calculation formula is:

[0064] ,

[0065] Use the dimensional entropy of each indicator to define the weight of each indicator , whose expression is:

[0066] ,

[0067] And calculate the weighted matrix R to determine the optimal solution and the worst solution , the expression of the weighted matrix R is:

[0068] .

[0069] Among them, r ij represents the element in the i-th row and j-th column of the matrix R, .

[0070] Step 7.3: Calculate the critical index C of each road segment in the road network based on the optimal and worst solutions of the weighted matrix R i , key index C i The calculation formula is:

[0071] .

[0072] C i The larger the value, the more critical the road segment is in the public transportation network.

[0073] In a second aspect, a computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for identifying key sections of a public transportation network as described in any one of the first aspects.

[0074] According to a third aspect, a computer device includes:

[0075] Memory, used to store instructions.

[0076] The processor is configured to execute the instructions so that the computer device performs the operations of the method for identifying key sections in a public transportation network as described in any one of the first aspects.

[0077] Beneficial effects: The present invention provides a method, electronic device, and storage medium for identifying key sections of a public transportation network. The method extracts the operating trajectories of each bus line based on GPS data of buses within a research scope; obtains the urban road network within the research scope, and constructs a map matching algorithm based on a hidden Markov model; applies the constructed map matching algorithm to match the operating trajectories of buses on each line with the actual road network to obtain bus route information along each section in the road network; utilizes a spatial connection method to identify and obtain various POI (Point of Interest) data and population data around each section; and calculates the key index of each section in the road network based on the extracted bus route information, lane information, surrounding population and POI information along each road section, as well as the departure frequency of each bus line, using the entropy weight TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method, thereby identifying key sections in the public transportation network. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 The present invention is a flowchart of a method for identifying key sections of a public transportation network.

[0079] Figure 2 Schematic diagram of bus track in the embodiment of the method of the present invention.

[0080] Figure 3 Schematic diagram of a road network created in an embodiment of the method of the present invention.

[0081] Figure 4 Schematic diagram showing the importance of each road in the embodiment of the method of the present invention. DETAILED DESCRIPTION

[0082] The following is a clear and complete description of the technical solutions in the examples of the present invention, in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0083] The present invention will be further described below with reference to specific embodiments.

[0084] Example 1:

[0085] This embodiment introduces a method for identifying key sections of a public transportation network. Figure 1 As shown, specifically including:

[0086] Step 1: Obtain the geographic information of buses within the study area, and extract the up and down running trajectories of each bus line based on the location data of the first and last bus stops.

[0087] Step 2: Obtain urban road network data within the study area and establish an urban road network.

[0088] Step 3: Construct a map matching algorithm based on the hidden Markov model to calculate the matching probability between the trajectory points of the up and down running trajectories of each bus line and the road sections in the urban road network.

[0089] Step 4: Considering the vector relationship between adjacent trajectory points in the bus route trajectory, calculate the transfer probability of the vehicle from one road segment to another at the intersection.

[0090] Step 5: Based on the obtained matching probability and transition probability, set the parameters in the hidden Markov model to construct a map matching algorithm between bus routes and urban road networks, and obtain the bus route information passing through each section in the urban road network.

[0091] Step 6: Use the spatial connection method to obtain various POI data and population data around each road section.

[0092] Step 7: Based on the extracted bus route information, lane information, surrounding population and POI information of each road section, as well as the departure frequency of each bus line, the key index of each road section in the road network is calculated based on the entropy weight TOPSIS method, thereby identifying the key sections in the bus network.

[0093] As a preferred solution, the step 1 specifically includes:

[0094] Step 1.1: Read bus route information data within a certain period of time (e.g., one month) within the research scope. The bus route information data includes but is not limited to three fields: route name, number of vehicles, and number of trips.

[0095] Step 1.2: Read the bus trajectory information within the research area. The bus trajectory information data includes but is not limited to five fields: line name, up and down lines, starting station, end station, and trajectory coordinates (a collection of trajectory points).

[0096] Step 1.3: Based on the bus route information data obtained in step 1.1, extract the travel trajectory G of each bus route in the up and down directions from the bus trajectory information data obtained in step 1.2. t , and count the total number of trips p of each line in the time period to obtain the geographic information of the bus lines within the research scope, where G tThis is a collection of bus trajectory points in the form of a LINESTRING (lon.lat,...), where lon is the longitude coordinate of the trajectory point and lat is the latitude coordinate of the trajectory point. The obtained bus route geographic information includes six fields: route name, up and down routes, starting and ending points, total number of trips, and longitude and latitude information.

[0097] As a preferred solution, the step 2 specifically includes:

[0098] Step 2.1: Based on the bus trajectory G obtained in step 1 t , extract the longitude range [lon min , lon max ], latitude range [lat min , lat max ], define this range as the study area, where lon min , lon max Divided into the minimum and maximum values of longitude in all bus track points, lat min , lat max are the minimum and maximum latitudes of all bus trajectory points respectively.

[0099] Step 2.2: Based on the study area obtained in Step 2.1, download the map data and extract the map edge and node information. A node is the endpoint of two adjacent roads on the map, and its information includes two fields: node number and node coordinates. An edge is a road on the map, and its information includes five fields: start number, end number, road name, number of lanes, and edge coordinates.

[0100] Step 2.3: Obtain the node coordinate set G according to step 2.2 nodes and the edge coordinate set G edges , create a map object G o As a city road network, G o ={G nodes ,G edges}. Create the map object G o Projected into the Mercator coordinate system for precise control of parameters. G nodes represents a set of nodes, G edges Represents a set of edges.

[0101] As a preferred solution, the step 3 specifically includes:

[0102] Step 3.1: Rasterize the urban road network data obtained in step 2.2. nodesThe circle area within a certain range of the node in the center is taken as the error area. According to the actual complexity of the bus line, the radius of the circle is set to ir meters. The sections with a distance greater than ir meters are eliminated to reduce the candidate sections and improve the matching efficiency.

[0103] Step 3.2: Use the projection method to determine the candidate points. Calculate the actual distances between the bus trajectory point obtained in step 1.2 and all the matching road segments within a range of meters from the node i. Compare the calculated results to find the matching road segment closest to the trajectory point, thereby obtaining the actual running trajectory of the bus in the road network. The formula for calculating the actual distance d between a certain trajectory point G in the bus trajectory information and the matching road segment is:

[0104] ,

[0105] Among them, x l 、y l x is the horizontal and vertical coordinates of the projection point L from the trajectory point G to the road section to be matched in the coordinate system, g 、y g are the horizontal and vertical coordinates of the trajectory point G in the coordinate system. The calculation formula for the coordinates of the projection point L from the trajectory point G to the road section AB to be matched is:

[0106] ,

[0107] Among them, x a 、y a x is the horizontal and vertical coordinates of the vertex A of the to-be-matched road segment AB in the coordinate system, and k is the slope of the to-be-matched road segment AB in the coordinate system. b 、y b are the horizontal and vertical coordinates of the vertex B of the road section AB to be matched in the coordinate system.

[0108] Step 3.3: According to the principle of GPS positioning error, the closer the road section to be matched is to the track point, the higher the probability that the road section is the actual running track of the bus. Set the positioning error E in this map matching algorithm d Satisfies the normal distribution, and its expression is:

[0109] ,

[0110] in, is the standard deviation of bus trajectory points, by setting the parameter Optimize the matching degree between bus trajectory and road network, is pi.

[0111] Step 3.4: Based on the bus up and down information obtained in step 1.3, extract the bus's driving direction information, introduce direction deviation into the map matching tool, and improve the matching degree between the bus trajectory and the road network. The direction error Eα The calculation formula is:

[0112] ,

[0113] Among them, α represents the angle between the direction of the bus trajectory point in its trajectory and the direction of the road section to be matched. Taking the trajectory point G and its matching road section AB as an example, its calculation formula is:

[0114] ,

[0115] Among them, k g is the orientation of trajectory point G in its trajectory.

[0116] Step 3.5: The bus positioning error E obtained from steps 3.3 and 3.4 d and the direction error between the bus GPS trajectory and the road section to be matched , calculate the matching probability E of the bus trajectory point and its to-be-matched road section as the actual driving road section of the bus. The calculation formula of E is:

[0117] .

[0118] As a preferred solution, the step 4 specifically includes:

[0119] Step 4.1: Get the Euclidean distance d between two adjacent points in the bus trajectory n-1，n The shortest distance D between the candidate points corresponding to these two points n-1，n , and calculate the similarity S between the bus trajectory and the road section to be matched r , S r The calculation formula is:

[0120] .

[0121] Step 4.2: Calculate the positioning change trend C between two adjacent trajectory points p , ensuring that the two adjacent points in the bus trajectory described in step 4.1 are the two adjacent points in the actual driving trajectory. p The calculation formula is:

[0122] ,

[0123] Among them, d n-1 and d n is the distance between two adjacent bus trajectory points and their corresponding matching road segments.

[0124] Step 4.3: The similarity S between the bus trajectory and the road segment to be matched obtained according to steps 4.1 and 4.2 r and the positioning change trend C between two adjacent trajectory pointsp , calculate the transfer probability T of a bus traveling from one road section to another at an intersection. The calculation formula of T is:

[0125] ,

[0126] Among them, t r is the similarity S between the bus trajectory and the road section to be matched r The coefficient of t p is the positioning change trend C between two adjacent trajectory points p The coefficient of t r +t p = 1. By adjusting the map matching tool, the probability T of the matched path being the actual driving path is increased, and the matching degree between the actual driving trajectory of the bus and the road network when entering the intersection is improved.

[0127] As a preferred solution, the step 5 specifically includes:

[0128] Step 5.1: According to the method of rasterization and determination of candidate points in step 3, set the parameter id to specify the maximum distance between the bus trajectory and the road section to be matched, and set the parameter di max Used to specify the maximum distance between each trajectory point and the corresponding starting position, setting parameters Used to specify the standard deviation value of the bus trajectory point; set the parameter st according to the method in step 4 to improve the matching degree between the actual driving trajectory of the bus when entering the intersection and the road network n It is used to specify whether bus trajectories are allowed to have parts that do not directly correspond to the road network, and set the parameter p stn Used to specify the proportion of bus trajectories that do not directly correspond to the road network. By setting the above parameters, a map matching tool distancemacther is created to build a map matching algorithm between bus routes and the actual road network to enhance the matching accuracy of trajectories.

[0129] Step 5.2: Randomly select a bus track sample, divide it into multiple small segments, and compare the extracted bus track coordinates with the road network G created in step 2.3. o Project them into the same coordinate system, use the map matching algorithm constructed in step 5.1 to match the trajectory with the actual path in the road network, extract the path matching results, and complete the path matching of a bus trajectory.

[0130] Step 5.3: According to the matching method of a bus trajectory in step 5.2, all bus trajectories within the research scope are matched with the actual road network to obtain the bus route information of each section in the road network. The obtained information of each bus route section in the road network includes six fields: route name, up and down lines, number of departures, road name, road starting point number, road end point number, and number of lanes.

[0131] Step 5.4: Link the bus route information of each road section in the road network obtained in step 5.3 with the bus route information within the study area in time period q obtained in step 1.1. The number of trips p of all bus routes on road section m in time period q can be calculated. m , p m The calculation formula is:

[0132]

[0133] Among them, I is the total number of bus routes passing through a certain road section m, p id is the total number of bus departures within the time period q of route id. Further, the public transportation traffic situation B of route m can be calculated. m , B m The calculation formula is:

[0134]

[0135] Among them, l m is the number of lanes in road section m.

[0136] As a preferred solution, the step 6 specifically includes:

[0137] Step 6.1: With the trajectory points of each road section in the actual road network as the center of the circle, set the road influence radius to θ meters, and construct a road buffer zone. This buffer zone is used as the influence range for evaluating whether each road section is congested.

[0138] Step 6.2: Read the POI data within the research scope, project the POI and the buffer of each road section in a unified coordinate system to obtain the POI information in the buffer of each road section, and perform statistical analysis on the number of POIs in each buffer. The number of POIs in the buffer corresponding to road section m is counted as poi m .

[0139] Step 6.3: Read the population data of each traffic zone in the study area. Based on the ratio of the area of each road buffer zone to the area of the traffic zone, calculate the residential population and working population data in the buffer zone. The residential population in the buffer zone corresponding to road segment m is calculated as lp. m , the number of working population is counted as wp m .

[0140] As a preferred solution, the step 7 specifically includes:

[0141] Step 7.1: Link the bus route information for each road section in the road network obtained in step 5.4 with the various POI data around each road section obtained in step 6.2 and the resident and working population data obtained in step 6.3. Construct a judgment matrix X with the bus traffic situation for each road section as the first column, the number of POIs in the corresponding buffer area as the second column, the number of residents in the corresponding buffer area as the third column, and the number of working people in the corresponding buffer area as the fourth column. Its expression is:

[0142] ,

[0143] Among them, x ij represents the element in the i-th row and j-th column of the matrix X, and m is the number of roads in the study area.

[0144] The judgment matrix is normalized, and the normalized calculation formula is:

[0145] ,

[0146] Among them, x max It is the maximum value under the same indicator.

[0147] Step 7.2: Calculate the dimensional entropy H of each indicator j , and its calculation formula is:

[0148] ,

[0149] Among them, j is the number of each indicator.

[0150] Use the dimensional entropy of each indicator to define the weight of each indicator , whose expression is:

[0151] ,

[0152] And calculate the weighted matrix R to determine the optimal solution and the worst solution , the expression of the weighted matrix R is:

[0153] .

[0154] Among them, r ij Represents the element in the i-th row and j-th column of the matrix R.

[0155] Step 7.3: Based on the optimal and worst solutions of the weighted matrix R obtained in step 7.2, calculate the critical index C of each road section in the road network. i , key index C i The calculation formula is:

[0156] .

[0157] C i The larger the value, the more critical the road segment is in the bus network. Finally, the key index C of each road segment is calculated. i Perform visualization.

[0158] Example 2:

[0159] This embodiment is used to further illustrate the technical solution of the present invention. The initial research object of this embodiment is 440 bus routes in a certain city. After screening the operating trajectory range and up and down travel of these 440 bus routes, 819 bus operation trajectories were finally selected as the final research objects for judging key sections.

[0160] 1. Obtain bus routes and geographic information (route name, up and down routes, starting and ending stations, number of departures, and latitude and longitude information of the running track) for 819 bus running tracks within the research scope for one month. The specific data format is shown in Table 1. The bus tracks obtained in this embodiment are as follows: Figure 2 shown.

[0161] Table 1 Bus route information

[0162]

[0163] 2. Based on the longitude and latitude of 819 bus trajectories within the research scope, the longitude range [118.361, 119.46] and the latitude range [31.85, 32.61] are extracted as the research area, and the map data is downloaded to extract the map node and edge information. In this embodiment, 37,290 pieces of node information are extracted, as shown in Table 2; 88,514 pieces of edge information are extracted, as shown in Table 3. Based on the extracted node and edge information, a map object is created to form a road network model and establish an urban road network. The road network created in this embodiment is as follows Figure 3 shown.

[0164] Table 2 Road network node information

[0165]

[0166] Table 3 Road network edge information

[0167]

[0168] 3. Based on the rasterization and candidate point determination methods, set the maximum distance between bus trajectories and their matching road segments to 500 meters, the maximum distance between each trajectory point and its corresponding starting position to 170 meters, and the standard deviation of bus trajectory points to 50. Furthermore, to improve the matching degree between the actual driving trajectory of buses entering intersections and the road network, allow for portions of bus trajectories that do not directly correspond to the road network. By setting these parameters, a map matching algorithm is constructed to enhance the matching accuracy between bus routes and the actual road network. Table 4 shows the parameter settings for the map matching tool created in this example.

[0169] Table 4 Map matching tool parameters

[0170]

[0171] 4. Match all bus trajectories within the study area with the actual road network to obtain bus route information for each section of the road network. The bus routes obtained in this example are shown in Table 5.

[0172] Table 5 Information on bus route sections

[0173]

[0174] 5. Link the bus route information with a month's worth of bus operation records to obtain the monthly travel times for all bus routes on each road segment. Combined with the number of lanes on each road segment, calculate the public transportation traffic conditions for each road segment. Set the road influence radius to 200 meters, construct road buffers, and count the number of POIs, residents, and workers within each buffer. This example reports public transportation traffic conditions for 13,857 roads within the study area, as well as the number of POIs, residents, and workers within the buffers, as shown in Table 6.

[0175] Table 6 Traffic conditions and buffer zone information for each road

[0176]

[0177] 6. According to the traffic conditions and buffer area information of each road shown in Table 3, the key index of each road section in the road network is calculated based on the entropy weight TOPSIS method, and the obtained key index is sorted from large to small. The judgment value of whether a road is a key section obtained in this example is shown in Table 7. The obtained judgment value is visualized as follows Figure 4 As shown in Figure 2, it can be found that the road critical index reaches its peak on some sections of Zhujiang Road.

[0178] Table 7 Road Criticality Index

[0179]

[0180] 7. After obtaining the key index of each road section, query the bus route information passing through the road section in Table 5 based on the road start and end point numbers. Based on the actual situation, evaluate whether there is functional duplication of these bus routes on the key sections, and then adjust and optimize the existing bus network to minimize the impact of emergencies on the key sections on the bus network.

[0181] Example 3:

[0182] This embodiment introduces a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for identifying key sections of a public transportation network as described in any one of the embodiments 1 is implemented.

[0183] Example 4:

[0184] This embodiment introduces a computer device, including:

[0185] Memory, used to store instructions.

[0186] The processor is used to execute the instructions so that the computer device performs the operations of the method for identifying key sections of a public transportation network as described in any one of the embodiments 1.

[0187] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0188] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0189] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0191] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for identifying key sections of a public transportation network, characterized by: Specifically include: Step 1: Obtain the geographic information of bus routes within the research scope, and extract the up and down running trajectories of each bus route based on the location data of the starting and ending stations of the bus routes; the geographic information of the bus routes includes the route name, up and down routes, starting and ending stations, total number of trips, and latitude and longitude information; Step 2: Obtain urban road network data within the research scope and establish an urban road network; Step 3: Build a map matching algorithm based on the hidden Markov model to calculate the matching probability between the trajectory points of the up and down running trajectories of each bus line and the road sections in the urban road network; Step 4: Considering the vector relationship between adjacent trajectory points in the bus route trajectory, calculate the transfer probability of the vehicle from one road segment to another at the intersection; Step 5: Based on the matching probability and transition probability, set the parameters in the hidden Markov model to construct a map matching algorithm between bus routes and urban road networks, and obtain the bus route information of each section in the urban road network; Step 6: Use the spatial connection method to obtain the population and POI information around each road section; Step 7: Based on the extracted bus route information, lane information, surrounding population and POI information of each road segment, as well as the departure frequency of each bus route, calculate the key index of each road segment in the road network, and identify the key road segments in the bus network based on the key index of each road segment; The step 7 specifically includes: Step 7.1: Link the bus route information of each road section in the road network with the various POI data around each road section, as well as the resident population and working population data. Build a judgment matrix X with the bus traffic situation of each road section as the first column, the number of POIs in the corresponding buffer area as the second column, the number of residents in the corresponding buffer area as the third column, and the number of working population in the corresponding buffer area as the fourth column. Normalize the judgment matrix X to obtain the elements of the normalized judgment matrix X. ; Among them, the judgment matrix X expression is: ， Among them, x ij represents the element in the i-th row and j-th column of the judgment matrix X, m is the number of road sections in the study area, i represents the i-th road section, and j represents the column number; Step 7.2: Calculate the dimensional entropy H of the four columns of indicators j , and its calculation formula is: ， Use the dimensional entropy of each indicator to define the weight of each indicator , whose expression is: ， And calculate the weighted matrix R to determine the optimal solution and the worst solution , the expression of the weighted matrix R is: ； Among them, r ij represents the element in the i-th row and j-th column of the matrix R, ; Step 7.3: Calculate the critical index C of each road segment in the road network based on the optimal and worst solutions of the weighted matrix R i , key index C i The calculation formula is: ； C i The larger the value, the more critical the road segment is in the public transportation network.

2. A method for identifying key sections of a public transportation network according to claim 1, characterized in that: The step 1 specifically includes: Step 1.1: Read bus route information data within a certain time period, including but not limited to: route name, number of buses, and trip times; Step 1.2: Read bus trajectory information within the research area, including but not limited to: route name, up and down routes, starting station, end station, and trajectory coordinates; Step 1.3: Based on the bus route information data, extract the travel trajectory G of each bus route in the up and down directions from the bus trajectory information t , and count the total number of trips p of each line in the time period to obtain the geographic information of the bus lines within the research scope.

3. A method for identifying key sections of a public transportation network according to claim 1, characterized in that: The step 2 specifically includes: Step 2.1: Based on the bus's trajectory G t , extract the longitude range [lon min , lon max ], latitude range [lat min ,lat max ], define this range as the study area, where lon min , lon max Divided into the minimum and maximum values of longitude in all bus track points, lat min , lat max are the minimum and maximum latitudes of all bus trajectory points respectively; Step 2.2: Download map data based on the study area and extract map edge and node information. Nodes are the endpoints of two adjacent roads in the map, and their information includes node numbers and node coordinates. Edges are roads in the map, and their information includes the starting number, ending number, road name, number of lanes, and edge coordinates. Step 2.3: According to the node coordinate set G nodes and the edge coordinate set G edges , create a map object G o As a city road network, G o ={G nodes ,G edges }.

4. A method for identifying key sections of a public transportation network according to claim 1, characterized in that: The step 3 specifically includes: Step 3.1: Rasterize the urban road network data to obtain the node coordinate set G nodes The circle area within a certain range of the node in the center is used as the error area, and the radius of the circle is set to ir meters. The road sections with a distance greater than ir meters are eliminated; Step 3.2: Use the projection method to determine the candidate points, calculate the actual distance d between the bus trajectory point and all the matching road segments within a range of i meters from the node, and compare the calculated results to find the matching road segment closest to the trajectory point; Step 3.3: Set the positioning error E in the map matching algorithm d , positioning error E d The expression is as follows: ， in, is the standard deviation of bus trajectory points, is pi, is a natural constant; Step 3.4: Extract the bus's driving direction information based on the bus's up and down information, and calculate the direction error E α , direction error E α The expression is as follows: ， Among them, α represents the angle between the direction of the bus trajectory point in its trajectory and the direction of the road section to be matched; Step 3.5: According to the bus positioning error E d and the direction error between the bus GPS trajectory and the road section to be matched , calculate the matching probability E of the bus trajectory point and its to-be-matched road section as the actual driving road section of the bus. The expression of E is as follows: 。 5. The method for identifying key sections of a public transportation network according to claim 1, wherein: The step 4 specifically includes: Step 4.1: Get the Euclidean distance d between two adjacent points in the bus trajectory n-1，n The shortest distance D between the candidate points corresponding to the two trajectory points n-1，n , and calculate the similarity S between the bus trajectory and the road section to be matched r , similarity S r The expression is as follows: ， Step 4.2: Calculate the positioning change trend C between two adjacent trajectory points p , positioning change trend C p The expression is as follows: ， Among them, d n-1 and d n are the distances between two adjacent bus trajectory points and their corresponding matching road sections; Step 4.3: Based on the similarity S between the bus trajectory and the road segment to be matched r and the positioning change trend C between two adjacent trajectory points p , calculate the transition probability T of a bus traveling from one road section to another at an intersection; the expression of the transition probability T is as follows: ， Among them, t r is the similarity S between the bus trajectory and the road section to be matched r The coefficient of t p is the positioning change trend C between two adjacent trajectory points p The coefficient of .

6. A method for identifying key sections of a public transportation network according to claim 1, characterized in that: The step 5 specifically includes: Step 5.1: Set the parameter id to specify the maximum distance between the bus trajectory and the road section to be matched, and set the parameter di max Used to specify the maximum distance between each trajectory point and the corresponding starting position, setting parameters Used to specify the standard deviation value of bus trajectory points, set the parameter st n It is used to specify whether bus trajectories are allowed to have parts that do not directly correspond to the road network, and set the parameter p stn It is used to specify the proportion of bus trajectories that do not directly correspond to the road network and to construct a map matching algorithm between bus routes and the actual road network based on the Hidden Markov Model; Step 5.2: Randomly select a bus trajectory sample, divide it into multiple small segments, and compare the extracted bus trajectory coordinates with the urban road network G o Projected into the same coordinate system, the trajectory is matched with the urban road network G using the map matching algorithm of bus routes and actual road networks. o Match the actual path in the image, extract the path matching result, and complete the path matching of a bus trajectory; Step 5.3: Repeat step 5.2 to match all bus trajectories within the study area with the actual road network to obtain the urban road network G o The bus route information of each road section in the urban road network includes the route name, up and down lines, number of departures, road name, road starting point number, road end point number, and number of lanes; Step 5.4: Link the bus route information of each road section in the urban road network with the bus route information within the study area in time period q, and calculate the number of trips p of all bus routes on road section m in time period q. m , and the public transportation situation B on section m m ; Among them, the number of passes p m The expression is as follows: ； Among them, I is the total number of bus routes passing through a certain road section m, p id is the total number of departures of bus route id within time period q; Public transport situation B m The expression is as follows: ； Among them, l m is the number of lanes in road section m.

7. The method for identifying key sections of a public transportation network according to claim 1, characterized in that: The step 6 specifically includes: Step 6.1: Take the trajectory points of each road section in the actual road network as the center of the circle, set the road influence radius to θ meters, and construct a road buffer zone; Step 6.2: Read the POI data within the research scope, project the POI data and the buffer zone of each road section in a unified coordinate system to obtain the POI information in the buffer zone of each road section, and perform statistical analysis on the number of POI information in each buffer zone. The number of POIs in the buffer zone corresponding to road section m is counted as poi m ; Step 6.3: Read the population data of each traffic zone in the study area. Based on the ratio of the area of each road buffer zone to the area of the traffic zone, calculate the residential population and working population data in the buffer zone. The residential population in the buffer zone corresponding to road segment m is calculated as lp. m , the number of working population is counted as wp m .

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, a method for identifying key sections of a public transportation network as claimed in any one of claims 1 to 7 is implemented.

9. A computer device, characterized in that: include: a memory for storing instructions; The processor is configured to execute the instructions so that the computer device performs the operations of the method for identifying key sections in a public transportation network as described in any one of claims 1 to 7.

Images