Quasi-dynamic random user balanced allocation method, electronic equipment and storage medium

Abstract

The invention relates to the technical field of passenger flow distribution, in particular to a quasi-dynamic random user balanced distribution method, electronic equipment and a storage medium, and the method comprises the steps: firstly obtaining a line physical network of urban rail transit, then building a travel transfer network based on the line physical network, and then building a travel transfer network based on the travel transfer network; and constructing a quasi-dynamic random user balanced distribution model, and then solving the quasi-dynamic random user balanced distribution model according to the line physical network to obtain a target passenger flow distribution result. By solving the quasi-dynamic random user equilibrium distribution model, the target passenger flow distribution result of the line interval in each time period can be quickly obtained, so that the problem of complexity in dynamic passenger flow distribution of urban rail transit is effectively solved, and the overall operation efficiency and passenger satisfaction of a traffic system are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to but are not limited to the technical field of passenger flow distribution, and in particular to a quasi-dynamic random user balanced distribution method, an electronic device, and a computer-readable storage medium. Background Art

[0002] Urban rail transit plays a positive role in optimizing urban spatial structure, alleviating urban traffic congestion, and protecting the environment. It is a large-capacity, convenient, punctual, and efficient mode of transportation. The subway AFC system only records the entry and exit information of passengers, and cannot obtain the travel trajectory of passengers within the rail transit network. Therefore, the distribution of passenger flow on the rail transit network is usually calculated by the passenger flow allocation method. The rail transit passenger flow allocation method is borrowed from road traffic, mainly including static passenger flow allocation method and dynamic passenger flow allocation method. The static passenger flow allocation model is usually designed based on the departure frequency, and the dynamic passenger flow allocation model is usually designed based on the timetable. Si Bingfeng et al. designed a static passenger flow allocation model for the urban rail transit network based on the principle of random user equilibrium, and solved the model using the sequential average algorithm (MSA). Xu Dejie et al. considered the travel selection strategies of different passenger types and designed a static passenger flow allocation method for the urban rail transit network with multiple routes operating on the same line, and solved it using the passenger flow increment allocation algorithm. Nur it and Sh lomo considered that passengers can know the expected arrival time of the path through the network, and designed a frequency-based passenger flow allocation method for the urban transportation network. Xu et al. designed a superpath-based solution algorithm for the balanced flow distribution problem of urban bus networks. Ren et al. designed a new variational inequality to describe the user equilibrium passenger flow allocation problem based on route and node strategies. However, these methods can only obtain the cross-sectional passenger flow on each route interval, and cannot calculate the passenger flow distributed in each route interval in each time period.

[0003] In order to obtain the detailed spatiotemporal distribution characteristics of passenger flow on the overall transportation network, some scholars have designed dynamic passenger flow allocation methods. SchmS cker et al. designed a quasi-dynamic frequency-based passenger flow allocation model, designed a probability calculation method for passenger flow loading, and used MSA to allocate OD passenger flow in each time period to a line network with capacity constraints according to the time sequence. The passenger flow that cannot be allocated due to overcapacity in each time period is transferred to the next time period. Yang Dongchi et al. designed a dynamic passenger flow equilibrium allocation model for the rail transit network based on the timetable, and used the successive average algorithm to solve it. The experimental analysis was conducted on some lines of the Beijing Metro. Hu Jianpeng and Luo Xia proposed a simulation method for dynamic passenger flow allocation of urban rail transit considering the change of transfer path travel time with the spatiotemporal characteristics of passenger flow. Pi et al. designed a dynamic user equilibrium allocation method for passenger and vehicle flow for a transportation network composed of multiple transportation modes. However, these dynamic passenger flow allocation methods usually face the problems of large network scale, more decision-making links for passengers, and more complex calculation processes. Summary of the invention

[0004] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0005] The embodiments of the present application provide a quasi-dynamic random user balanced allocation method, an electronic device, and a computer-readable storage medium, which effectively solve the complex problems in the dynamic passenger flow allocation of urban rail transit and improve the overall operating efficiency of the transportation system and passenger satisfaction.

[0006] In a first aspect, an embodiment of the present application provides a quasi-dynamic random user balanced allocation method, comprising:

[0007] Obtain the physical network of urban rail transit lines;

[0008] Building a travel transfer network based on the line physical network;

[0009] Based on the travel transfer network, a quasi-dynamic random user equilibrium allocation model is constructed;

[0010] According to the line physical network, the quasi-dynamic random user equilibrium allocation model is solved to obtain a target passenger flow distribution result.

[0011] In combination with the first aspect, in one embodiment of the present application, the quasi-dynamic random user equilibrium allocation model is constructed based on the travel transfer network, including: obtaining the costs of various arcs of the travel transfer network at the departure time period t, and the actual time required to pass through the various arcs; obtaining the path selected by the passenger in the travel transfer network at the departure time period t; Travel needs in, represents the set of feasible paths L between the starting point r and the end point s on the transfer network rs The kth path in the equation; according to the cost of each type of arc, the actual time required to pass through each type of arc, and the demand Calculate the flow and passenger selection paths of various arcs flowing into the travel transfer network at the departure time t travel costs; select routes based on the flow and passengers flowing into each type of arc travel costs, and construct a quasi-dynamic random user equilibrium allocation model; wherein the quasi-dynamic random user equilibrium allocation model is used to select paths according to the flow and passengers flowing into the various arcs The travel cost is calculated to obtain the route selected by the passenger based on the travel cost and the flow of each type of arc segment at the departure time t. decision.

[0012] In combination with the first aspect, in an embodiment of the present application, the various types of arcs include waiting arcs, stop arcs, transfer arcs and operating arcs, and the cost of various types of arcs in the travel transfer network in the departure time period t, as well as the actual time required to pass through the various types of arcs, are obtained, including: calculating the cost of the waiting arc according to the average waiting time; calculating the cost of the stop arc according to the average stop time of the station; calculating the cost of the transfer arc according to the transfer time and transfer penalty fee; calculating the cost of the operating arc according to the time perceived by the passengers when the train is running; and obtaining the cost of each type of arc in the time period t from the cost of the waiting arc, the cost of the stop arc, the cost of the transfer arc and the cost of the operating arc.

[0013] Costs of various arc segments in time period t It is expressed as follows:

[0014]

[0015] In time period t, the actual time required for passengers to pass through arc segment a is recorded as It is expressed as follows:

[0016]

[0017] in, represents the passenger flow into arc segment a, represents the average waiting time of passengers, is the peak factor, Indicates the operating line lm where the waiting arc a is located i and direction d i The train departure interval in time period t; It represents the average stop time at the station; represents the transfer time, PN represents the penalty fee for one transfer; represents the time perceived by passengers when the train is running, where It represents the average running time of trains in various arc sections. It represents the congestion coefficient of the arc segment in time period t, which is expressed as: in, It indicates the passing capacity of operating arc segment a within time period t.

[0018] In combination with the first aspect, in an embodiment of the present application, the cost of each type of arc segment, the actual time required to pass through each type of arc segment and the demand Calculate the flow and passenger selection paths of various arcs flowing into the travel transfer network at the departure time t The travel cost includes: according to the cost of each arc segment and the actual time required to pass through each arc segment, the travel cost of the passenger choosing the route at the departure time t is calculated. Travel, flow into various arc segments According to the requirements and the time periods of the inflow into various arc segments Calculate the flow rate into each arc segment in time period t; calculate the passenger selection path in time period t based on the flow rate into each arc segment travel expenses.

[0019] In combination with the first aspect, in an embodiment of the present application, the time period It is expressed as follows:

[0020]

[0021] The flow rate flowing into each type of arc segment The representation is as follows:

[0022]

[0023] Among them, when the arc segment a is located on the path above and When it flows into the arc segment a, it is in the time period t′, When the arc segment a is located on the path above and When it flows into the arc segment a, it is not in the time period t′,

[0024] The travel expenses It is expressed as follows:

[0025] In combination with the first aspect, in an embodiment of the present application, the quasi-dynamic random user balanced allocation model is expressed as follows:

[0026]

[0027] st

[0028]

[0029] Among them, θ is a non-negative parameter used to describe the random characteristics of the model, t is the time period variable, NT is the total number of time periods, To meet the travel needs of passengers.

[0030] In combination with the first aspect, in an embodiment of the present application, solving the quasi-dynamic random user equilibrium allocation model according to the line physical network to obtain the target passenger flow distribution result includes: obtaining a first valid path set according to the line physical network; mapping the paths in the first valid path set to the travel transfer network to obtain a second valid path set; selecting a path according to the costs of the various arc segments and the passengers; The travel cost is calculated to obtain the path cost of the second valid path set; the path flow is calculated according to the quasi-dynamic random user equilibrium allocation model; the path flow is converted into section flow, and the cost of the valid paths in the second valid path set is calculated based on the section flow; the additional path flow is calculated according to the cost of the valid paths in the second valid path set, and the path flow is updated based on the additional path flow; a convergence function is constructed according to the passenger travel demand and the updated path flow, and when the value of the convergence function meets the convergence accuracy, the target passenger flow distribution result is obtained.

[0031] In combination with the first aspect, in an embodiment of the present application, the step of constructing a travel transfer network based on the line physical network includes:

[0032] According to the station set N, line section set E, operating line set LM and operating direction d on the line physical network i , define the waiting arc Stop arc Transfer arc and running arc (n1,n2) run ;

[0033] According to the waiting arc Get set A wait , according to the stop arc Get set A stop , according to the transfer arc Get set A tran , according to the running arc segment (n1, n2) run Get set A run ;

[0034] Based on the set A wait , Set A atop , Set A tran and set A run Construct a travel transfer network;

[0035] Where n represents the station, n∈N, lm i represents the i-th operating line in the operating line set LM; d i =1 indicates the upward direction, d i =2 indicates the downlink direction; lmj Indicates transfer line; d j Indicates the transfer direction; (n1,n2)∈E.

[0036] In a second aspect, an embodiment of the present application further provides an electronic device, comprising: one or more processors; a memory on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the quasi-dynamic random user balanced allocation method as described above.

[0037] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the quasi-dynamic random user balanced allocation method as described above is implemented.

[0038] In the embodiment of the present application, the line physical network of the urban rail transit is first obtained, and then a travel transfer network is constructed based on the line physical network, and then a quasi-dynamic random user equilibrium allocation model is constructed based on the travel transfer network. Since the line physical network includes the station location, line direction, connection relationship, etc. of the urban rail transit network, the travel transfer network constructed based on the line physical network can clarify the key factors such as the transfer path, transfer time, and transfer cost between each station, and can intuitively show the feasibility and convenience of passengers transferring between different lines, and provide a basis for passengers to choose the optimal travel plan. The quasi-dynamic random user equilibrium allocation model constructed based on the travel transfer network can simulate the changes in passenger flow in different time periods based on the time dimension, so that the prediction results are closer to reality, thereby improving the adaptability and accuracy of the model. According to the line physical network, solving the quasi-dynamic random user equilibrium allocation model can ensure that under a large-scale network and many decision-making links, the optimal passenger flow distribution results (i.e., target passenger flow distribution results) of the line interval in each time period can be quickly obtained, thereby effectively solving the complex problems in the dynamic passenger flow distribution of urban rail transit, thereby improving the overall operation efficiency and passenger satisfaction of the transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of a quasi-dynamic random user balanced allocation method provided in an embodiment of the present application;

[0040] Figure 2 The embodiment of this application provides Figure 1 A specific flow chart of step 120;

[0041] Figure 3 is a schematic diagram of a transfer network path provided in an embodiment of the present application;

[0042] Figure 4 The embodiment of this application provides Figure 1A specific flow chart of step 130;

[0043] Figure 5 The embodiment of this application provides Figure 4 A specific flow chart of step 430;

[0044] Figure 6 The embodiment of this application provides Figure 1 A specific flow chart of step 140;

[0045] Figure 7 It is a valid path information diagram of some OD pairs provided by a detailed example of this application;

[0046] FIG. 8A to FIG. 8J It is a detailed example of this application that provides a partial OD passenger flow distribution on the effective path and the algorithm convergence diagram;

[0047] 9A to 9D It is a comparison chart of passenger flow in the up and down directions provided by a detailed example of this application;

[0048] Fig.10 It is a schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] It should be noted that, although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in a different order from that in the flowchart. The terms "first", "second", etc. in the specification, claims and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions that the present invention can be implemented, so they have no technical substantive significance. Any modification of the structure, change of the proportional relationship or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope of the technical content disclosed by the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of narration, and are not used to limit the scope of the implementation of the present invention. The change or adjustment of the relative relationship should also be regarded as the scope of the implementation of the present invention without substantially changing the technical content.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0052] Urban rail transit plays a positive role in optimizing urban spatial structure, alleviating urban traffic congestion and protecting the environment. By the end of 2023, there will be 59 cities in China operating rail transit, with a total operating mileage of 11,232.65 kilometers. In 2023, the annual urban rail transit passenger volume will reach 29.44 billion, an increase of 51.7% over 2022. In recent years, with the sharp increase in passenger travel demand, urban rail transit companies are facing new challenges. How to accurately and quickly grasp the spatial and temporal distribution of passenger flow on the rail transit network and provide reliable decision-making basis for operation organizers is an urgent problem that urban rail transit companies need to solve.

[0053] In the related art, the subway AFC system only records the passenger's entry and exit information, and cannot obtain the passenger's travel trajectory within the rail transit network. Therefore, the distribution state of passenger flow on the rail transit network is usually calculated by the passenger flow allocation method. The rail transit passenger flow allocation method is borrowed from road traffic, mainly including static passenger flow allocation method and dynamic passenger flow allocation method. The static passenger flow allocation model is usually designed based on the departure frequency, and the dynamic passenger flow allocation model is usually designed based on the timetable. Si Bingfeng et al. designed a static passenger flow allocation model for the urban rail transit network based on the principle of random user equilibrium, and solved the model using the sequential average algorithm (MSA). Xu Dejie et al. considered the travel selection strategies of different passenger types and designed a static passenger flow allocation method for the urban rail transit network with multiple routes operating on the same line, and solved it using the passenger flow increment allocation algorithm. Nur it and Sh lomo considered that passengers can know the expected arrival time of the path through the network, and designed a frequency-based urban transportation network passenger flow allocation method. Xu et al. designed a super path-based solution algorithm for the equilibrium flow distribution problem of the urban bus network. Ren et al. designed a new variational inequality to describe the user equilibrium passenger flow allocation problem based on route and node strategies. However, these methods can only obtain the cross-sectional passenger flow on each route interval, and cannot calculate the passenger flow distributed in each time period of the route interval.

[0054] In order to obtain the detailed spatiotemporal distribution characteristics of passenger flow on the overall transportation network, some scholars have designed dynamic passenger flow allocation methods. SchmS cker et al. designed a quasi-dynamic frequency-based passenger flow allocation model, designed a probability calculation method for passenger flow loading, and used MSA to allocate OD passenger flow in each time period to a line network with capacity constraints according to the time sequence. The passenger flow that cannot be allocated due to overcapacity in each time period was transferred to the next time period. Yang Dongchi et al. designed a dynamic passenger flow equilibrium allocation model for rail transit network based on the timetable, and used the successive average algorithm to solve it. The experimental analysis was conducted on some lines of Beijing Metro. Hu Jianpeng and Luo Xia proposed a simulation method for dynamic passenger flow allocation of urban rail transit considering the change of transfer path travel time with the spatiotemporal characteristics of passenger flow. Pi et al. designed a dynamic user equilibrium allocation method for passenger and vehicle flow for a transportation network composed of multiple transportation modes. Although these methods have made important progress in theory, they still face many challenges in practical applications due to the huge network scale, complex passenger decision-making process and heavy computational burden.

[0055] In view of this, the embodiment of the present application provides a quasi-dynamic random user balanced allocation method, electronic device and computer storage medium. In this method, the line physical network of urban rail transit is first obtained, and then a travel transfer network is constructed based on the line physical network, and then a quasi-dynamic random user balanced allocation model is constructed based on the travel transfer network. Since the line physical network includes the station location, line direction, connection relationship, etc. of the urban rail transit network, the travel transfer network constructed based on the line physical network can clarify the key factors such as the transfer path, transfer time, and transfer cost between each station, and can intuitively show the feasibility and convenience of passengers transferring between different lines, and provide a basis for passengers to choose the optimal travel plan. The quasi-dynamic random user balanced allocation model constructed based on the travel transfer network not only considers the travel cost of passengers (such as time and cost), but also introduces the time dimension, which can simulate the changes in passenger flow in different time periods, such as morning and evening peaks, flat peaks, etc., so that the prediction results are closer to reality and the adaptability and accuracy of the model are improved. According to the line physical network, solving the quasi-dynamic random user balanced allocation model can ensure that the optimal passenger flow distribution result can be quickly obtained under complex network scale and many decision-making links. It provides more reliable data support for operation management and optimization. In addition, through more accurate passenger flow distribution, it reduces the waiting time and travel time of passengers in the transportation system, improves passenger travel satisfaction and the service level of the overall transportation system.

[0056] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0057] See also Figure 1 , Figure 1 : is a flow chart of the quasi-dynamic random user balanced allocation method provided in the embodiment of the present application. Figure 1 As shown, the calculation method includes but is not limited to steps 110 to 140.

[0058] Step 110: Acquire the line physical network of urban rail transit;

[0059] Step 120: constructing a travel transfer network based on the line physical network;

[0060] Step 130: constructing a quasi-dynamic random user equilibrium allocation model based on the travel transfer network;

[0061] Step 140: Solve the quasi-dynamic random user equilibrium allocation model according to the line physical network to obtain the target passenger flow distribution result.

[0062] In one embodiment, the line physical network of urban rail transit refers to the actual physical space network composed of a series of rail lines, stations and their mutual connection relationships. Among them, the rail line is the skeleton of the rail transit system, which is used for train travel. The station is an important node in the rail transit system, mainly for passengers to get on and off, transfer and wait. The connection relationship includes the connection between stations (such as station intervals), the connection between stations and lines (such as entry and exit passages), and the transfer connection between different lines (such as transfer passages, transfer platforms, etc.). These connection relationships constitute the complex network structure of the urban rail transit system, ensuring that trains and passengers can move and transfer smoothly.

[0063] In one embodiment, the physical network of the rail transit line can be easily obtained by visiting the official website of the city's rail transit system. In addition, modern map services (such as Google Maps, Baidu Maps, Amap, etc.) usually integrate information on urban rail transit. By searching for the city name on these maps and switching to the rail transit layer, you can intuitively see the physical network of the rail transit line and the location of the station. It should be noted that the method of obtaining the physical network of the urban rail transit line is not limited to the above examples, and the specific method of obtaining the physical network of the line can be selected according to actual needs. It is worth noting that due to the complexity and dynamism of the urban rail transit system, the obtained line physical network information may change over time. Therefore, when obtaining and using the information of the line physical network, it is necessary to pay attention to its timeliness and accuracy.

[0064] In one embodiment, the travel transfer network refers to a network model that includes passenger travel and transfer relationships, which can be used to describe the travel process of passengers on the urban rail transit network and the cost calculation method of the corresponding arc segment. Among them, the corresponding arc segment is generated based on the line physical network. For example, the waiting arc segment, transfer arc segment, etc. can be generated according to the station nodes of the line physical network. Specifically, in the travel transfer network, by analyzing the information such as stations and lines in the line physical network, it can be determined which stations are transfer stations and the transfer relationship between them, so as to obtain the corresponding arc segments. These arc segments constitute the feasible path of the travel transfer network.

[0065] See also Figure 2 In one embodiment, the process of constructing a travel transfer network based on a line physical network includes but is not limited to steps 210 to 230.

[0066] Step 210: Based on the station set N, line section set E, operating line set LM and operating direction d on the line physical network i , define the waiting arc Stop arc Transfer arc and running arc (n1,n2) run ;

[0067] Step 220: According to the waiting arc Get set A wait , according to the stop arc Get set A stop , according to the transfer arc Get set A tran , according to the running arc (n1,n2) run Get set A run ;

[0068] Step 230: Based on set A wait , Set A stop , Set A tran and set A run Construct a travel transfer network.

[0069] Where n represents the station, n∈N, lm i represents the i-th operating line in the operating line set LM; d i =1 indicates the upward direction, d i =2 indicates the downlink direction; lm j Indicates transfer line; d j Indicates the transfer direction; (n1,n2)∈E.

[0070] The following is an exemplary description of the process of constructing a travel transfer network based on a line physical network based on steps 210 to 230.

[0071] After obtaining the physical network of urban rail transit lines, the line physical network is recorded as (N, E), where N is the station set and E is the line section set. The network is connected. The set of operating lines on the network is recorded as LM, and the i-th operating line is recorded as lm i , the operation direction is denoted as d i , uplink direction d i =1, downlink direction d i =2. Divide the operating time of a day into several time periods, with a time period length of Δt, a total number of time periods denoted as NT, and a time period variable denoted as t, then t = 1, 2, …, NT. Any two different stations r, s∈N constitute an OD (origin-destination) pair (r, s) of passenger travel, where r represents the passenger's departure station and s represents the passenger's final destination station. Let W be the set of all OD pairs. The total demand between OD pairs (r, s) is denoted as q rs , the demand corresponding to the departure time period t is recorded as So

[0072] The travel transfer network constructed based on the line physical network can be recorded as (N, Α), where Α is a set of arcs including waiting arcs, stop arcs, transfer arcs and running arcs. It should be noted that the waiting arcs, stop arcs and transfer arcs are all generated at station nodes, while the running arcs are generated on the line sections. For each station n∈N and the operating line lm where it is located i and direction d i , define the waiting arc The set of waiting arcs is denoted as A wait ; Define the stop arc The set of arc segments consisting of stop stations is denoted as A stop For each transfer station n∈N, the operating line lm i and direction d i Transfer to line lm j and direction d j , define the transfer arc The set of transfer arcs is denoted as A tran For each line section (n1,n2)∈E, define the running arc (n1,n2) run , the set of running arcs is recorded as A run Then the arc set Α=A wait ∪A stop ∪A tran ∪A run .

[0073] It is worth noting that the feasible path on the travel transfer network starts from the waiting arc, goes through several running arcs and stop arcs (if it includes transfer, the transfer arc and waiting arc corresponding to the transfer station are added), and ends with the running arc, which corresponds to the path on the line physical network one by one. For example, suppose a path on the line physical network (N, E) The route passes through the stations in sequence where n x3 It is a transfer station, connecting line lm i and lm j ,station Located on line lm i Up, Station Located on line lm j Then the path 1 on the travel transfer network (N, Α) corresponding to the path R is as follows Figure 3 Then, the path 1 on the transfer network (N, Α) can be expressed as follows:

[0074]

[0075] Among them, the symbol "+" represents the connection relationship between the front and back lines of the arc segment.

[0076] In one embodiment, there may be multiple paths between the OD pair (r, s), denoted by L rs is the set of feasible paths between OD pairs (r, s) on the travel transfer network (N, Α), where the kth path is recorded as Passengers can choose between multiple routes according to their needs and comprehensively evaluate the cost of each route. It is understandable that the travel route selection behavior of passengers is affected by many factors and has a certain degree of randomness. It is worth noting that since the route fares between the same travel OD pair are the same, passengers mainly consider the time, congestion and convenience of the route when choosing a travel route. The time of the route can be estimated based on the existing train timetable and rail transit network conditions, the congestion can be measured by the passenger flow on the route, and the convenience can be determined based on the transfer situation.

[0077] See also Figure 4 Exemplarily, the process of constructing a quasi-dynamic random user equilibrium allocation model based on the travel transfer network in step 130 includes but is not limited to steps 410 to 440.

[0078] Step 410: Obtain the costs of various arcs of the travel transfer network at the departure time period t, and the actual time required to pass through the various arcs;

[0079] Step 420: Obtain the path of the travel transfer network selected by the passenger at the departure time t Travel needs in, represents the set of feasible paths L between the starting point r and the end point s on the transfer network rs The kth path in ;

[0080] Step 430: Based on the cost of each arc segment, the actual time required to pass through each arc segment and the demand Calculate the flow and passenger choice paths of various arcs flowing into the travel transfer network at the departure time t travel expenses;

[0081] Step 440: Selecting routes based on the flow and passengers flowing into each arc segment A quasi-dynamic random user equilibrium allocation model is constructed based on the travel cost.

[0082] Among them, the quasi-dynamic random user equilibrium allocation model can select paths according to the flow and passengers flowing into various arcs. The travel cost is calculated, and the passenger chooses the path based on the travel cost and the flow of various arcs at the departure time t. decision.

[0083] In one embodiment, the various arcs of the travel transfer network include waiting arcs, stop arcs, transfer arcs, and operating arcs. The process of obtaining the costs of various arcs of the travel transfer network at departure time period t and the actual time required to pass through various arcs in step 410 includes: calculating the cost of the waiting arc according to the average waiting time; calculating the cost of the stop arc according to the average stop time of the station; calculating the cost of the transfer arc according to the transfer time and the transfer penalty fee; calculating the cost of the operating arc according to the time perceived by the passengers when the train is running; and obtaining the costs of various arcs and the actual time required to pass through various arcs in time period t from the costs of the waiting arc, the stop arc, the transfer arc, and the operating arc. Specifically, for various arcs a of the travel transfer network, the passenger flow into arc a is recorded as Where t represents the time period when the arc flows into the segment, and the corresponding segment cost is recorded as

[0084] In the embodiment of the present application, based on different arc segments a, different calculation methods are used to obtain the costs of various arc segments a and the actual time required to pass through each arc segment.

[0085] When arc a is a waiting arc, its cost can be represented by the average waiting time of passengers at the station. Since the passenger flow distribution based on the physical network of urban rail transit does not describe the specific riding behavior of passengers, it is impossible to accurately calculate the waiting time of passengers. Therefore, for the waiting arc a and time period t, a peak coefficient can be designed The average waiting time of passengers is Indicates the operating line lm where the waiting arc a is located i and direction d i The train departure interval in time period t). During the off-peak period, the coefficient It is usually set to zero. During peak passenger flow periods, the coefficient can be adjusted according to the passenger flow of the station. Set to greater than zero.

[0086] When arc segment a is a stop arc segment, its cost can be expressed by the average stop time of the station, which is expressed as

[0087] When arc a is a transfer arc, its cost is represented by the transfer time and transfer penalty fee of the passengers in the transfer station. The transfer time is recorded as It represents the time it takes a passenger to get off a train on one line and walk to the platform of another line at a transfer station. It is usually determined by dividing the transfer distance of the station by the average walking speed of the passenger. The penalty fee for a transfer is recorded as PN.

[0088] When arc a is a running arc, its cost can be expressed by the time perceived by passengers when the train is running, which is expressed as in It represents the average running time of the train in the running arc a (line section), It represents the congestion coefficient of the arc segment in time period t, which is expressed as follows:

[0089]

[0090] in, It represents the capacity of the running arc a in the time period t. When the passenger flow on the running arc exceeds the capacity, congestion occurs in the carriage, and the time perceived by passengers exceeds the running time of the train.

[0091] Therefore, in time period t, the cost of each arc segment is It is expressed as follows:

[0092]

[0093] In time period t, the actual time required for the passenger to pass through arc segment a (target arc segment) is recorded as So

[0094]

[0095] For Path An arc segment a on For path A subpath of the path At the departure time t, OD selects the path for the passenger (r, s) The time period of traveling into arc segment a is recorded as So

[0096]

[0097] See also Figure 5 , for example, in step 430, according to the cost of each type of arc segment, the actual time required to pass through each type of arc segment, and the demand Calculate the flow and passenger choice paths of various arcs flowing into the travel transfer network at the departure time t The process of calculating travel expenses may include but is not limited to steps 510 to 530.

[0098] Step 510: Calculate the time required for the passenger to choose a route at the departure time t according to the cost of each arc segment and the actual time required to pass through each arc segment. Travel, flow into various arc segments

[0099] Step 520: According to the requirements and the time periods of inflow into various arc segments Calculate the flow rate into each type of arc segment in time period t;

[0100] Step 530: Calculate the passenger selection path in time period t based on the flow into each type of arc segment travel expenses.

[0101] In one embodiment, at the departure time t, the OD pair (r, s) of passengers selects the path The travel demand is recorded as If arc segment a is on the path Up and flow When it flows into the arc segment, it is in the time period t′, then otherwise Then, in time period t, the flow into arc segment a can be expressed as follows:

[0102]

[0103] At departure time t, the OD pair (r, s) of passengers chooses the path The travel expenses are recorded as It can be expressed as follows:

[0104]

[0105] In one embodiment, it is understandable that the passenger flow distribution on the rail transit network fluctuates greatly with the time of day, so the cost of the route also varies with the time of day. Generally speaking, passengers always want to choose the route with the lowest cost, and there is a certain degree of randomness, which leads to certain differences in the selection behavior of passengers at different departure times. As passengers accumulate travel experience, the route selection behavior of passengers departing at different time periods gradually tends to be stable, that is, No longer changes. At this point, the passenger flow distribution in each time period on the rail transit network tends to be balanced, reaching a random user equilibrium state. Based on this feature, in step 440, the flow and passengers flowing into each arc segment are used to select the path. The travel cost is 200,000, and the quasi-dynamic random user equilibrium allocation model is constructed, which can be expressed as follows:

[0106]

[0107] Among them, θ is a non-negative parameter used to describe the random characteristics of the model.

[0108] The optimal solution of the above model satisfies the first-order condition:

[0109]

[0110] From the above first-order conditions, we can get

[0111]

[0112] Equivalent to

[0113]

[0114] So

[0115]

[0116] therefore,

[0117]

[0118] It can be seen from formula (16) that when the value of parameter θ tends to be smaller, passengers show higher randomness in choosing travel routes, which leads to a more dispersed distribution of passenger flow among multiple optional routes, without significant concentration on a specific route. On the contrary, as the value of parameter θ increases, the randomness of passengers in choosing routes is significantly reduced, and passenger flow tends to be more concentrated on routes with relatively low travel costs. Parameter θ can actually be regarded as an indicator of passengers' familiarity with travel routes: the more familiar passengers are with the overall transportation network, the less random they are in choosing routes, and accordingly, the larger the value of parameter θ.

[0119] make

[0120]

[0121] So

[0122]

[0123] Select a path for the passenger of OD pair (r,s) at departure time t The probability of traveling. Since the distribution of passenger flow on the arc fluctuates with the change of time period, the arc cost on the path changes accordingly. Therefore, this probability changes with the change of departure time period t, causing the passenger's choice behavior to adjust with the change of departure time period, reflecting the dynamic characteristics of the above passenger flow distribution model. However, in each time period, the probability of passengers choosing a travel path is certain, that is, a random user equilibrium state is reached. Since the passenger flow in each section of the rail transit network is composed of the passengers departing from each OD pair in each time period, the passengers departing in each time period are not independent of each other, but influence and restrain each other. Therefore, the passengers of the OD pair (r,s) choose the path in the departure time period t The probability of traveling is When , the rail transit network as a whole reaches a dynamic random user equilibrium state.

[0124] It is worth noting that the key to solving the quasi-dynamic random user equilibrium allocation model is to determine the passenger's travel path and load the passenger flow of the corresponding period onto the path. For urban rail transit networks with more station nodes, there are more paths between each OD pair, but passengers generally only choose a few of them to travel, and not every path will have passengers. For example, some paths with longer detours or more transfers are rarely chosen by passengers. Therefore, when solving the quasi-dynamic random user equilibrium allocation model based on the line physical network, it is not necessary to calculate all the paths between each OD pair. It is only necessary to determine the paths commonly used by passengers, which can improve the efficiency of solving the model.

[0125] See also Figure 6 In one embodiment, the process of solving the quasi-dynamic random user equilibrium allocation model according to the line physical network to obtain the target passenger flow distribution result may include but is not limited to steps 610 to 670.

[0126] Step 610: Obtain a first valid path set according to the line physical network;

[0127] Step 620: Map the paths in the first valid path set to the travel transfer network to obtain a second valid path set;

[0128] Step 630: Select a route based on the cost and passengers of each segment The travel cost of the second valid path set is calculated;

[0129] Step 640: Calculate the path flow according to the quasi-dynamic random user equilibrium allocation model;

[0130] Step 650: converting the path flow into the segment flow, and calculating the cost of the valid path in the second valid path set based on the segment flow;

[0131] Step 660: Calculate the additional path flow according to the cost of the valid path in the second valid path set, and update the path flow based on the additional path flow;

[0132] Step 670: Construct a convergence function based on passenger travel demand and updated path flow. When the value of the convergence function meets the convergence accuracy, the target passenger flow distribution result is obtained.

[0133] In one embodiment, on the physical network of urban rail transit lines, the paths commonly used by passengers between any two stations can be called effective paths, and the characteristics of these effective paths include at least: not too many transfers, not too long detours, and no detours. It is worth noting that since more transfers or long detours will increase the cost of the path, passengers always hope to choose the path with the lowest cost. In general, the cost of effective paths between the same OD pair is not much different, which is equivalent to the minimum cost. In particular, the possibility of passengers choosing a path with a large cost difference is very small and can be ignored.

[0134] In one embodiment, the determination conditions of the effective path can be obtained according to the characteristics of the effective path, including: the maximum number of transfers of the path does not exceed TM; the mileage of the path does not exceed λ times the mileage of the shortest path (also called λ as the non-shortest path coefficient); there are no repeated nodes on the path. According to the determination conditions of the effective path, multiple effective paths can be extracted from the line physical network (N, E) to obtain a first effective path set. It should be noted that each effective path in the first effective path set meets the above-mentioned characteristics of the effective path.

[0135] In one embodiment, in the process of extracting multiple valid paths from the line physical network (N, E) according to the determination condition of the valid path to obtain the first valid path set, an effective path search algorithm (such as Dijkstra algorithm, etc.) can be used to extract the valid path set ER between each OD pair (r, s) on the line physical network (N, E) based on the determination condition. rs , (r,s)∈W.

[0136] In one embodiment, since the line physical network (N, E) facilitates the search for paths, and the transfer network (N, Α) can describe the travel process of passengers in detail, it is convenient to calculate the path cost. Therefore, after determining the valid path (first valid path set) of the passenger based on the line physical network (N, E), the path of the first valid path set can be mapped to the travel transfer network (N, Α) to obtain the second valid path set, and then calculate the cost of the second valid path set. Specifically, after obtaining the first valid path set, According to formula (1), ER rs The paths in the travel transfer network (N, Α) are mapped one by one to determine the effective path set L of passengers on the travel transfer network (N, Α) rs , and then calculate the effective path set L rs The cost of (the second valid path set).

[0137] In one embodiment, the route is selected based on the cost and passengers of each segment. The travel cost is calculated to obtain the second effective path set (effective path set L rs ) route cost, you can first set the road flow Then, according to formula (3) and formula (7), the set L is calculated rs The cost of each path in (i.e., the path cost of the second valid path set).

[0138] In one embodiment, when calculating the path flow according to the quasi-dynamic random user equilibrium allocation model, the optimal solution of the model can be used to derive the path of the OD pair (r, s) passengers choosing the travel transfer network (N, Α) at the departure time t. Travel needs Then, according to the departure time t, OD pairs (r, s) passengers select the path Probability of traveling and demand Specifically, the OD pair (r, s) can be calculated according to formulas (17) and (18) to select the path in the departure period t. Initial route flow of the trip (i.e., path flow),

[0139] In one embodiment, in step 550, the path flow rate can be calculated according to formula (6). Calculate road traffic Then according to formula (3) and formula (7), based on the road flow Calculate the set L rs The cost of the effective path That is, the cost of the valid paths in the second valid path set.

[0140] In one embodiment, after calculating the cost of the effective path, After that, the additional path flow can be calculated according to formulas (17) and (18): Then based on the additional path flow Update route flow Among them, the update path flow The calculation formula is

[0141] It should be noted that the quasi-dynamic random user equilibrium allocation model aims to obtain the passenger flow distribution result of the user equilibrium state. The user equilibrium state means that no user can reduce his generalized cost by unilaterally changing his path selection. This equilibrium state is usually not obvious, and it is necessary to gradually adjust the flow on each path through an iterative process until equilibrium is reached. In addition, the passenger flow allocation problem in the transportation network is usually complex and nonlinear. This is because the cost of the path (such as travel time) depends not only on the characteristics of the path itself (such as road conditions and length), but also on the flow on other paths (such as increased travel time caused by congestion). Therefore, in this embodiment, by setting a convergence function and its convergence condition (such as the ratio of the iterative step size decrease is less than a certain threshold, etc.), it can be ensured that the iterative process stops when it reaches a level close enough to the true solution, thereby ensuring the stability and reliability of the passenger flow distribution results. Specifically, when the ratio of the iterative step size decrease is less than a certain threshold as the convergence condition, when the initial path flow is calculated, After that, we can first set the number of iterations j = 1 and initialize the iteration step α j , set the initial value E(0) of the convergence function to a larger value.

[0142] In one embodiment, according to the travel needs of passengers and the updated path flow The convergence function E(j) can be constructed, and the convergence condition of the convergence function E(j) is expressed as follows:

[0143]

[0144] Among them, ∈ is a positive number, which is the convergence accuracy. If the convergence condition is met, the algorithm is terminated; otherwise, the variable scale ratio descent method is used to update the step size α j =η j α j , where η j To reduce the proportion, the change rules are as follows:

[0145]

[0146] Among them, [ηmin ,η max ] is η j Let j = j + 1, continue to execute step 550 and subsequent steps until the convergence condition is met, and obtain the target passenger flow distribution result (ie, the target passenger flow distribution result).

[0147] The following is a detailed example, which aims to illustrate the actual operation process of the quasi-dynamic random user balanced allocation method proposed in the embodiment of the present application.

[0148] 1. Example Setup

[0149] Taking the Guangzhou Metro in 2023 as an example for analysis, the example includes 253 stations, including 38 transfer stations, 278 line sections, and a total mileage of 602 kilometers. The OD passenger flow between [6:00, 23:00] on May 22, 2023 (Monday) (obtained through the AFC system, a total of 4534431 passengers) is selected for allocation, and the time interval is 30 minutes. The operating time, capacity and train departure interval of the line section are obtained according to the actual operating train timetable. The transfer time in the transfer station is calculated according to the actual situation of the station, and the penalty fee for one transfer is set to 10 minutes. The maximum number of transfers for the effective path TM = 4, the maximum number of effective paths between each OD pair is 4, and the non-shortest coefficient λ of the effective path is 1.8. The convergence accuracy of the algorithm ∈ = 1×10 -5 , the iterative step size decreases by a factor of η j The value range is [0.85,0.95].

[0150] 2. Analysis of calculation results

[0151] Using C# programming, the passenger flow distribution calculation was performed for three sets of values ​​of the random parameter θ for path selection, θ = 0.001, 0.003, and 0.005. The effective path information of some OD pairs is as follows: Figure 7 As shown, Figure 7 In the figure, the superscript * indicates that the path is the minimum cost path; Line 2 and Line 3 North are the northern extension of Line 3. The results of passenger flow allocation on the effective path are shown in FIG. 8A to FIG. 8IAs shown. The path with the lowest comprehensive cost is not necessarily the shortest mileage, and passengers are more inclined to choose a path with fewer transfers. When θ = 0.001, the passengers' path selection is more random, and a certain proportion of passengers choose each valid path. This situation occurs when the passengers are not familiar with the path. When θ = 0.005, the passengers' path selection is less random, and most passengers choose the path with the lowest cost to travel, and very few passengers choose the path that does not have the lowest cost to travel. When θ = 0.003, the passengers' path selection is more moderately random, which means that most passengers are familiar with the path, and a small number of passengers are not familiar with the path. These passengers do not often travel in the city, which is more in line with reality. When θ = 0.003, the algorithm converges as shown in the figure. Figure 8J As shown in the figure, the algorithm converges quickly and reaches the convergence accuracy requirement within a small number of iterations.

[0152] When θ = 0.003, the temporal and spatial distribution of passenger flow on Line 2, Line 5, Line 3 and Line 18 is as follows: Figures 9A to 9D As shown in the figure, the line sections are numbered from small to large in the upward direction. The temporal and spatial distribution characteristics of passenger flow on the line are analyzed as follows:

[0153] Fig. 9A In the middle, Line 2 runs north-south, starting from Guangzhou South Station in the south and ending at Jiahe Wanggang in the north, passing through Panyu District, Haizhu District, Yuexiu District and Baiyun District. The section from Line 11 to Line 22 is part of the line from Haizhu Square to Jiahe Wanggang, located in Yuexiu District and Baiyun District, and there are old urban areas along the way. The passenger flow distribution on this section of the line has obvious characteristics of morning and evening peaks. During the morning peak, the passenger flow mainly flows from north to south, and during the evening peak, the passenger flow mainly flows from south to north, with obvious tidal phenomena of urban commuter flow. The section from Line 1 to Line 10 is part of the line from Guangzhou South Station to Haizhu Square, located in Panyu District and Haizhu District. The passenger flow on this section of the line has obvious characteristics of morning peaks, mainly from south to north; during the evening peak, the passenger flow increases but does not form a significant peak characteristic, mainly from north to south. The main reason for this distribution feature is that Guangzhou South Railway Station is a large passenger flow distribution center. During the morning peak period, passengers flow from Guangzhou South Railway Station to the urban area. This part of the passenger flow is mainly out-of-city passengers and does not return during the evening peak period. Therefore, the degree of passenger flow increase during the evening peak period is weakened.

[0154] Fig. 9BIn the middle, Line 5 runs east-west, starting from Jiaokou in the west and ending at Wenchong in the east. It passes through Liwan District, Yuexiu District, Tianhe District and Huangpu District, and runs through the old city center and Tianhe CBD. The section from Line 14 to Line 18 passes through Tianhe CBD. The passenger flow distribution of this section has obvious morning and evening peak characteristics. During the morning peak period, the passenger flow mainly flows from east to west, and during the evening peak period, the passenger flow mainly flows from west to east, which is a tidal phenomenon of urban commuting passenger flow. The sections of Lines 12 and 13 are located in the core area of ​​Zhujiang New Town. The passenger flow has obvious morning and evening peak characteristics, and the two-way flow is relatively balanced. The passenger flow on the section from Line 5 to Line 11 has obvious morning peak characteristics, mainly flowing from west to east. During the evening peak period, the passenger flow increases, mainly from east to west, but no significant peak characteristics are formed, which shows that the peak passenger flow on this section of the line is not the urban commuting passenger flow. The analysis found that the morning peak passenger flow in the section of Line 6 is the largest. This part of the passenger flow flows from Guangzhou Railway Station to Xiaobei and then to Tianhe CBD, mainly business passengers from outside the city. Overall, the temporal and spatial distribution characteristics of passenger flow in the up and down directions of Line 5 are somewhat similar to those of Line 2.

[0155] Fig. 9C In the middle, Line 3 runs north-south, starting from Panyu Square in the south and ending at Tianhe Bus Station in the north, passing through Panyu District, Haizhu District and Tianhe District. Passenger flow is mainly concentrated in the section between Lines 6 and 12 (Tiyuxi Road to Dashi), especially in the section between Lines 7 and 8 (Zhujiang New Town to Kecun), which is located in the core area of ​​Zhujiang New Town and has the most concentrated passenger flow. During the morning rush hour, passenger flow mainly flows from south to north, rushing to Zhujiang New Town; during the evening rush hour, passenger flow gathers in Zhujiang New Town and flows southward to Panyu District, with obvious tidal phenomenon.

[0156] Fig.9D Line 18 runs north-south and was opened on September 28, 2021 from Wanqingsha Station to Xiancun Station. It passes through Tianhe District, Haizhu District, Panyu District, and Nansha District. It is adjacent to Line 3 and has a good diversion effect, alleviating the overcrowding of Line 3. The temporal and spatial distribution characteristics of passenger flow on Line 18 are somewhat similar to those of Line 3 (the up and down directions of Line 18 are opposite to those of Line 3, that is, the up direction of Line 18 corresponds to the down direction of Line 3).

[0157] It can be seen from this detailed example that according to the quasi-dynamic random user equilibrium allocation method provided in the embodiment of the present application, the spatiotemporal distribution of passenger flow on the line section can be obtained to conform to the travel habits of passengers and the line distribution characteristics (target passenger flow distribution results), thereby effectively solving the complex problems in the dynamic passenger flow distribution of urban rail transit, and thus improving the overall operation efficiency of the transportation system and passenger satisfaction.

[0158] Reference Fig.10 The embodiment of the present application further discloses an electronic device, the electronic device 1000 comprising:

[0159] One or more processors 1001;

[0160] The memory 1002 stores one or more programs. When the one or more programs are executed by the one or more processors 1001, the one or more processors 1001 implement the quasi-dynamic random user balanced allocation method in any of the previous embodiments.

[0161] In addition, an embodiment of the present application further discloses a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the quasi-dynamic random user balanced allocation method in any of the previous embodiments is implemented.

[0162] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A quasi-dynamic random user balanced allocation method, characterized in that: include: Obtain the physical network of urban rail transit lines; Building a travel transfer network based on the line physical network; Based on the travel transfer network, a quasi-dynamic random user equilibrium allocation model is constructed; According to the line physical network, the quasi-dynamic random user equilibrium allocation model is solved to obtain a target passenger flow distribution result.

2. The quasi-dynamic random user balanced allocation method according to claim 1, characterized in that: The step of constructing a quasi-dynamic random user equilibrium allocation model based on the travel transfer network includes: Obtain the costs of various arcs of the travel transfer network at the departure time period t, and the actual time required to pass through the various arcs; Get the path selected by the passenger in the travel transfer network during the departure time period t Travel needs in, represents the set of feasible paths L between the starting point r and the end point s on the transfer network rs The kth path in ; According to the cost of each type of arc segment, the actual time required to pass through each type of arc segment and the demand Calculate the flow and passenger selection paths of various arcs flowing into the travel transfer network at the departure time t travel expenses; Select routes based on the flow and passengers flowing into each type of arc The travel expenses are calculated and a quasi-dynamic random user equilibrium allocation model is constructed; The quasi-dynamic random user equilibrium allocation model is used to select paths according to the flow and passengers flowing into each type of arc segment. The travel cost is calculated to obtain the route selected by the passenger based on the travel cost and the flow of each type of arc segment at the departure time t. decision.

3. The quasi-dynamic random user balanced allocation method according to claim 2, characterized in that: The various arcs include waiting arcs, stop arcs, transfer arcs and running arcs. The cost of various arcs in the travel transfer network at the departure time period t and the actual time required to pass through the various arcs are obtained, including: calculating the cost of the waiting arc according to the average waiting time; calculating the cost of the stop arc according to the average stop time of the station; calculating the cost of the transfer arc according to the transfer time and the transfer penalty fee; calculating the cost of the running arc according to the time perceived by the passengers when the train is running; and obtaining the cost of various arcs in the time period t from the cost of the waiting arc, the cost of the stop arc, the cost of the transfer arc and the cost of the running arc. Costs of various arc segments in time period t It is expressed as follows: In time period t, the actual time required for passengers to pass through arc segment a is recorded as It is expressed as follows: in, represents the passenger flow into arc segment a, represents the average waiting time of passengers, is the peak factor, Indicates the operating line lm where the waiting arc a is located i and direction d i The train departure interval in time period t; represents the average stop time at a station; represents the transfer time, PN represents the penalty fee for one transfer; represents the time perceived by passengers when the train is running, where It represents the average running time of trains in various arc sections. It represents the congestion coefficient of the arc segment in time period t, which is expressed as: in, It indicates the passing capacity of operating arc segment a within time period t.

4. The quasi-dynamic random user balanced allocation method according to claim 3, characterized in that: The cost of each type of arc segment, the actual time required to pass through each type of arc segment, and the demand Calculate the flow and passenger selection paths of various arcs flowing into the travel transfer network at the departure time t The travel cost includes: according to the cost of each arc segment and the actual time required to pass through each arc segment, the travel cost of the passenger choosing the route at the departure time t is calculated. Travel, flow into various arc segments According to the requirements and the time periods of the inflow into various arc segments Calculate the flow rate into each arc segment in time period t; calculate the passenger selection path in time period t based on the flow rate into each arc segment travel expenses.

5. The quasi-dynamic random user balanced allocation method according to claim 4, characterized in that: The period It is expressed as follows: The flow rate flowing into each type of arc segment The representation is as follows: Among them, when the arc segment a is located on the path above and When it flows into the arc segment a, it is in the time period t′, When the arc segment a is located on the path above and When it flows into the arc segment a, it is not in the time period t′, The travel expenses It is expressed as follows:

6. The quasi-dynamic random user balanced allocation method according to claim 5, characterized in that: The quasi-dynamic random user equilibrium allocation model is expressed as follows: st Among them, θ is a non-negative parameter used to describe the random characteristics of the model, t is the time period variable, NT is the total number of time periods, To meet the travel needs of passengers.

7. The quasi-dynamic random user balanced allocation method according to claim 6, characterized in that: The method of solving the quasi-dynamic random user equilibrium allocation model according to the line physical network to obtain the target passenger flow distribution result includes: obtaining a first valid path set according to the line physical network; mapping the paths in the first valid path set to the travel transfer network to obtain a second valid path set; and selecting a path according to the costs of the various arc segments and the passengers. The travel cost is calculated to obtain the path cost of the second valid path set; the path flow is calculated according to the quasi-dynamic random user equilibrium allocation model; the path flow is converted into section flow, and the cost of the valid paths in the second valid path set is calculated based on the section flow; the additional path flow is calculated according to the cost of the valid paths in the second valid path set, and the path flow is updated based on the additional path flow; a convergence function is constructed according to the passenger travel demand and the updated path flow, and when the value of the convergence function meets the convergence accuracy, the target passenger flow distribution result is obtained.

8. The quasi-dynamic random user balanced allocation method according to claim 1, characterized in that: The constructing of a travel transfer network based on the line physical network includes: According to the station set N, line section set E, operating line set LM and operating direction d on the line physical network i , define the waiting arc Stop arc Transfer arc and running arc (n1,n2) run ; According to the waiting arc Get set A wait , according to the stop arc Get set A stop , according to the transfer arc Get set A tran , according to the running arc segment (n1, n2) run Get set A run ; Based on the set A wait , Set A stop , Set A tran and set A run Construct a travel transfer network; Where n represents the station, n∈N, lm i represents the i-th operating line in the operating line set LM; d i =1 indicates the upward direction, d i =2 indicates the downlink direction; lm j Indicates transfer line; d j Indicates the transfer direction; (n1,n2)∈E.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When at least one of the programs is executed by at least one of the processors, the quasi-dynamic random user balanced allocation method as described in any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program executable by a processor is stored therein, and when the computer program executable by the processor is executed by the processor, it is used to implement the quasi-dynamic random user balanced allocation method as described in any one of claims 1 to 8.

Images