Freight clearing method for rail transit and computer equipment

Abstract

The invention is suitable for the technical field of traffic information, and provides a rail transit freight clearing method and computer equipment, and the method comprises the steps: determining a specific value of an element time vector of a rail transit network based on an element time calculation model of the rail transit network; for any candidate path of any traffic starting and ending point, determining path time of the candidate path based on the specific value of the element time vector; determining a path allocation probability of the candidate path based on the path time of the candidate path, the transfer time of the candidate path, the transfer frequency of the candidate path and a path allocation model; and carrying out freight clearing based on the actual passenger flow volume of the traffic starting and ending point, the path distribution probability of each candidate path of the traffic starting and ending point and the line information of each candidate path, and obtaining the clearing share of each operation main body. According to the invention, the right and interest ratio of each operation main body can be accurately defined, and the accuracy of clearing the freight of the rail transit network is improved.

Description

Technical Field

[0001] The present application belongs to the field of transportation information technology, and in particular, relates to a freight clearing method and computer equipment for rail transportation. Background Art

[0002] After the urban rail transit realizes network operation, passengers may have multiple routes to choose from between different stations. For multi-operating network with one-ticket barrier-free transfer, the paths of passengers traveling in the network cannot be counted through the transfer gates, which makes it impossible to clearly define the rights and interests of each operating entity providing passenger services in terms of freight, and it is difficult to ensure the accuracy of freight settlement, which has a huge impact on the management of urban rail transit by each operating entity. Summary of the invention

[0003] In view of this, the embodiments of the present application provide a method and computer device for clearing freight in rail transit, which can accurately define the equity ratio of each operating entity and improve the accuracy of freight clearing in the rail transit network.

[0004] A first aspect of an embodiment of the present application provides a method for clearing freight for rail transit, comprising: Determining specific values ​​of element time vectors of the rail transit network based on an element time calculation model of the rail transit network; For any candidate path of any traffic starting and ending point, determining the path time of the candidate path based on the specific value of the element time vector; Determining a path allocation probability of the candidate path based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path, and a path allocation model; Based on the actual passenger flow of the transportation starting and ending points, the path allocation probability of each candidate path of the transportation starting and ending points, and the line information of each candidate path, freight settlement is performed to obtain the settlement share of each operating entity.

[0005] In an implementation of the first aspect, for any candidate path of any traffic starting and ending point, determining the path time of the path based on the specific value of the element time vector includes: For any candidate path of any traffic starting and ending point, determine the element time subset of the candidate path; Selecting, from the element time vector, a specific value of each element time of the element time subset; The specific value of each element time of the element time subset is summed to obtain the path time of the candidate path.

[0006] In an implementation of the first aspect, the method further includes: Determine the driving path of any traffic starting and ending point based on the rail transit network and the interval travel time between any two adjacent stations; Based on the element time vector of the rail transit network and the driving path of each transportation start and end point in the rail transit network, determine the element time matching matrix; The travel time vector composed of the travel time of the starting and ending points of the traffic is used as the dependent variable, the element time vector is used as the independent variable, and the element time matching matrix is ​​combined to perform linear fitting to obtain a preliminary fitting model; Based on the current value of the element time vector, the preliminary fitting model is adjusted to obtain the element time extrapolation model.

[0007] In an implementation of the first aspect, the determining, based on the element time extrapolation model of the rail transit network, a specific value of the element time vector of the rail transit network includes: Based on the variance of travel time at each traffic starting point and end point and the current value of each element time, a weight matrix is ​​constructed; Performing matrix fitting calculation on the element time estimation model based on the weight matrix to obtain an estimated value of the element time vector; Based on the estimated value of the element time vector and the actual value of the interval travel time of each adjacent station, the estimated travel time of each traffic starting point and ending point is determined; Determine the time error based on the estimated travel time between the transportation start and end points and the actual travel time between the transportation start and end points; With the goal of minimizing the duration error, the element time vector is iteratively updated until the duration error converges to obtain a specific value of the element time vector.

[0008] In an implementation of the first aspect, the method further includes: Based on the travel data of the rail transit network, determining the travel time and travel time variance of each traffic starting and ending point in the rail transit network; If there are abnormal traffic start-end points with abnormal travel time variance, a baseline coefficient of variation is determined based on the mean of the travel time of other traffic start-end points except the abnormal traffic start-end point and the travel time variance of the other traffic start-end points; Based on the reference coefficient of variation and the travel time of the abnormal traffic start and end points, a corrected travel time variance of the abnormal traffic start and end points is determined. In an implementation of the first aspect, the method further includes: Based on the path utility characteristics and utility coefficients of the path, the path utility model of the path is constructed; Based on the path utility of each candidate path of each traffic origin and destination, constructing a total utility index model of the traffic origin and destination; Based on the path utility model and the total utility index model, constructing a path allocation model for the traffic start and end points; Optimizing the model parameters of the path allocation model by means of vector updating and matrix calculation to obtain the optimal parameters of the path allocation model; Among them, the model parameter is the utility coefficient, the path utility characteristics include path time, the number of transfers of the path and the transfer time of the path, and the utility coefficient includes path time sensitivity, transfer number sensitivity and transfer time sensitivity.

[0009] In an implementation of the first aspect, the optimizing the model parameters of the path allocation model by means of vector updating and matrix calculation to obtain the optimal parameters of the path allocation model includes: Determining a path utility vector for a candidate path set based on a path utility feature matrix of the candidate path set and the path utility model; the candidate path set includes each candidate path in the rail transit network, and each candidate path belongs to a traffic starting and ending point; Determine the path allocation probability vector of the candidate path set based on the path allocation matrix of the candidate path set, the path utility vector and the total utility index vector of the traffic origin and destination set, wherein the path allocation matrix represents the allocation relationship between the candidate paths and the traffic origin and destination; Determine a predicted route flow vector for the route set based on the route allocation matrix of the candidate route set and the route probability vector, wherein each element of the predicted route flow vector represents the predicted route flow of a single route; Determining a flow error based on an actual line flow vector of the line set and the predicted line flow vector; With the goal of minimizing the flow error, the model parameters of the path allocation model are iteratively updated, and finally a path allocation model with converged flow error is obtained.

[0010] In an implementation of the first aspect, determining the path allocation probability vector of the candidate path set based on the path allocation matrix of the candidate path set, the path utility vector, and the total utility index vector of the traffic origin and destination set includes: Eliminate abnormal candidate paths in the path allocation matrix of the candidate path set to obtain a path matching matrix of the candidate path set; Based on the path matching matrix and the path utility vector, a total utility index vector of a traffic origin and destination set is obtained; Based on the path allocation matrix and the total utility index vector, a path allocation probability vector of the candidate path set is determined.

[0011] In an implementation of the first aspect, performing freight clearing based on the actual passenger flow of the transportation starting and ending points, the path allocation probability of each candidate path of the transportation starting and ending points, and the line information of each candidate path to obtain the clearing share of each operating entity includes: Determine the path passenger flow of each candidate path of the traffic starting and ending point based on the actual passenger flow of the traffic starting and ending point and the path allocation probability of each candidate path of the traffic starting and ending point; Based on the route information on each candidate path, determining the proportion of sites to which each route on the candidate path belongs; Based on the passenger flow of each candidate route and the proportion of stations on each line on the candidate route, freight settlement is carried out to obtain the settlement share of each operating entity.

[0012] A second aspect of the embodiment of the present application provides a rail transit freight fee clearing device, comprising: A time determination module, used to determine the specific value of the element time vector of the rail transit network based on the element time calculation model of the rail transit network; A time determination module is used to determine the path time of any candidate path of any traffic starting and ending point based on the specific value of the element time vector; A probability determination module, configured to determine the path allocation probability of the candidate path based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path, and a path allocation model; The freight settlement module is used to settle freight based on the actual passenger flow of the transportation starting and ending points, the path allocation probability of each candidate path of the transportation starting and ending points, and the line information of each candidate path to obtain the settlement share of each operating entity.

[0013] A third aspect of an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the computer device implements the method described in the first aspect above.

[0014] A fourth aspect of the embodiments of the present application provides a computer program product, characterized in that it includes a computer program, and when the computer program is run, the method described in the first aspect above is executed.

[0015] A fifth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0016] In the first aspect of the embodiment of the present application, based on the element time calculation model of the rail transit network, the specific value of the element time vector of the rail transit network is determined. Then, for any candidate path of any traffic starting and ending point, based on the specific value of the element time vector, the path time of the candidate path is determined, which can provide accurate path time information for freight clearing. Then, based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path and the path allocation model, the path allocation probability of each candidate path is accurately determined. Finally, based on the actual passenger flow of the traffic starting and ending points, the path allocation probability of each candidate path of the traffic starting and ending points, and the line information of each candidate path, freight clearing is performed to obtain the clearing share of each operating entity and realize freight clearing. In this way, the equity ratio of each operating entity can be accurately defined, and the accuracy of realizing freight clearing of the rail transit network can be improved.

[0017] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 It is a flow chart of a method for clearing freight charges in a rail transit system provided in an embodiment of the present application; Figure 2 It is a flowchart of a method for determining path time provided in an embodiment of the present application; Figure 3 It is a flowchart of a method for constructing an element time calculation model provided in an embodiment of the present application; Figure 4 is a flow chart of a method for determining a specific value of an element time vector provided in an embodiment of the present application; Figure 5 It is a schematic flow chart of a method for correcting the variance of travel time provided in an embodiment of the present application; Figure 6 It is a flowchart of a method for constructing a path allocation model provided in an embodiment of the present application; Figure 7 It is a flowchart of an iterative updating method of a path allocation model provided in an embodiment of the present application; Figure 8 is a flow chart of a method for determining a path allocation probability vector provided by an embodiment of the present application; Fig. 9It is a flow chart of a specific implementation method of freight clearing provided in an embodiment of the present application; Fig.10 is a schematic diagram of a rail transit freight fee clearing device provided in an embodiment of the present application; Fig.11 It is a schematic diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0021] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0022] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0023] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0024] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0025] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0026] In one embodiment, Figure 1 As shown, a method for clearing freight for rail transit is provided, and the method is described by taking the method applied to a computer device as an example, including the following steps S101 to S104: Step S101, based on the element time estimation model of the rail transit network, determining the specific value of the element time vector of the rail transit network.

[0027] In applications, when performing functional calculations related to rail transit networks, directed graphs are usually used to represent rail transit networks, namely, rail transit network graphs. Rail transit network graphs are directed graphs. Computer equipment constructs directed graphs based on the line network interval travel time data of the rail transit network, with stations as vertices, travel intervals as directed edges, and interval travel time as the weight of directed edges, as rail transit network graphs. The origin and destination of a trip refer to the origin station Origin and the destination station Destination, called OD (Origin-Destination). A rail transit network graph includes multiple ODs. Each OD can include at least one valid path in the rail transit network, and the origin and destination stations of these valid paths are the same, but there is at least one different intermediate station. Each valid path may be selected by passengers. The ability to assign a valid path to a corresponding OD as the target path selected by the passenger on the OD is related to the path time of the valid path. Therefore, for any OD, the path time of the corresponding valid path must be determined.

[0028] In the rail transit network, according to the path definition, the path time of each valid path can be calculated in combination with the station interval travel time / station stop time. In the application, in the rail transit network, the path time of any path is determined by multiple element times, and the element time at least includes the train travel time, the waiting time of the entry station, the entry time of the entry station, and the exit time of the exit station. For non-direct candidate paths, the element time that constitutes the path time also includes the transfer time of at least one transfer channel. Among them, the waiting time and entry time of the same entry station can be combined into the entry time of the entry station. Therefore, the element time used to determine the path time includes at least the train travel time, the entry time of the entry station (including the waiting time) and the exit time of the exit station. For non-direct candidate paths, the element time that constitutes the path time can also include the transfer time of at least one transfer channel.

[0029] In the application, the number of all stations and the number of all transfer channels in the rail transit network are known, so the element time vector of the entire rail transit network can be obtained. Assuming that the rail transit network includes k transfer channels, p entry stations, the number of exit stations is the same as the entry stations, p exit stations, and q adjacent train passages, the element time vector of the entire rail transit network is =[transfer time; entry time; exit time; train section travel time], The dimension is m=k+2p+q, where k, p, q, and m are all positive integers. The specific form of the element time vector is as follows:

[0030] It should be noted that, in the rail transit network, at least one element time required for the path time of any path is a subset of the element time vector.

[0031] Therefore, for the one-stop ticket purchase scenario, since only the passenger's OD can be determined, it is not possible to determine which path the passenger has chosen for the OD. Therefore, the freight can be better cleared by determining the path allocation probability. Generally, the shorter the path time, the greater the path allocation probability (that is, the probability of being selected by the passenger as the OD is greater). The process of determining the path time can be converted into the process of determining the specific value of each element of the element time vector.

[0032] In the application, the specific value of the element time vector can be solved through the element time calculation model of the rail transit network. The element time calculation model is obtained by fitting the matching relationship between the rail transit card swiping travel data, path time and element time within a period of time in the rail transit network. The element time calculation model can characterize the linear relationship between the path time and the element time of the driving paths of all ODs in the entire rail transit network. The driving path of OD refers to the path with the shortest path time among the valid paths between ODs. The specific solution process for the element time vector will be explained later.

[0033] Step S102: for any candidate path of any traffic starting and ending point, the path time of the candidate path is determined based on the specific value of the element time vector.

[0034] In the application, for any candidate path of any traffic starting and ending point, the computer device can determine the element time subset of the path relative to the element time vector, and determine the path time of the candidate path according to the specific value of each element time in the element time subset. The calculation formula of path time is: path time = train travel time + transfer time + entry time (including waiting time) + exit time.

[0035] Among them, for each candidate path, the train travel time can be decomposed into the sum of the interval travel time between each two adjacent stations from the starting station to the end station. According to the train operation timetable, the interval travel time between any two adjacent stations and the train's stop time at the station are pre-planned, and the interval travel time between any two adjacent stations can be directly calculated, that is, the interval travel time between two adjacent stations = the train's stop time at the departure station / 2+the train's travel time between the two stations+the train's stop time at the arrival station / 2.

[0036] For any specific path, except for the train travel time, other element times (transfer time, station entry time (including waiting time), and station exit time) cannot be directly calculated. Therefore, the specific value of the element time vector of the rail transit network can be determined based on the element time vector model, and then the transfer time of all transfer channels required for the path, the station entry time required for the station, and the station exit time required for the station are read, and the train travel time of the path is obtained by combining the train travel time of q adjacent stations in the existing rail transit network, thereby determining the path time of the path.

[0037] Step S103, determining the path allocation probability of the candidate path based on the path duration of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path, and the path allocation model.

[0038] In the application, the path allocation probability of a candidate path refers to the probability that the candidate path is allocated to the OD of the transportation origin and destination. That is, the probability that the passenger chooses the candidate path as the actual travel path for the OD trip. The computer device obtains the path allocation model of the rail transit network, inputs the path time of the candidate path, the transfer time of the candidate path, and the number of transfers of the candidate path into the path allocation model, and the path allocation probability of the candidate path can be obtained.

[0039] In the application, the path time, the transfer time and the number of transfers of the path are path attributes used to determine the utility of the path, which can be called path utility characteristics. The path allocation model is used to determine the path allocation probability of the path being allocated to the corresponding OD. The model parameters of the path allocation model are determined based on the utility coefficient of the path utility characteristic. The difference in utility coefficient directly affects the probability of the candidate path being allocated. The utility coefficient includes the passenger's sensitivity to transfer time, the passenger's sensitivity to travel time, and the passenger's sensitivity to the number of transfers. The specific form of the path allocation model will be described in detail later.

[0040] Step S104, based on the actual passenger flow at the transportation starting and ending points, the path allocation probability of each candidate path at the transportation starting and ending points, and the line information of each candidate path, the freight is settled to obtain the settlement share of each operating entity.

[0041] In the application, the actual passenger flow to be cleared in the rail transit network within a period of time is grouped to determine the actual passenger flow of each traffic origin and destination OD. For any OD, the passenger flow allocation ratio of each valid candidate path can be determined by combining the path allocation probability of the valid candidate path of the OD.

[0042] In the application, each candidate path is associated with at least one line information. According to the line information of each station in each candidate path, the passenger flow distribution passing through each line is determined, and then the OD trip and the settlement amount of each operating entity are obtained through the relationship between the line and the operating entity.

[0043] In this embodiment, firstly, based on the element time calculation model of the rail transit network, the specific value of the element time vector of the rail transit network is determined, and then for any candidate path of any traffic starting and ending point, based on the specific value of the element time vector, the path time of the candidate path is determined, and then based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path and the path allocation model, the path allocation probability of each candidate path is accurately determined, and finally based on the actual passenger flow of the traffic starting and ending points, the path allocation probability of each candidate path of the traffic starting and ending points, and the line information of each candidate path, the freight is cleared, and the clearing share of each operating entity is obtained to realize the freight clearing. In this way, the equity ratio of each operating entity can be accurately defined, and the accuracy of realizing the freight clearing of the rail transit network can be improved.

[0044] In one embodiment, Figure 2 As shown, the implementation process of step 102 includes the following steps S201 to S203: Step S201, for any candidate path of any traffic starting and ending point, determine the element time subset of the candidate path.

[0045] In the application, the rail transit network includes multiple transportation start and end points, each transportation start and end point includes multiple valid candidate paths, each candidate path corresponds to an element time subset, and the element time subset includes at least one element time in the element vector.

[0046] Step S202: Select a specific value of each element time of the element time subset from the element time vector.

[0047] In the application, the computer device can obtain the element time vector through the element time calculation model. The element time vector includes the specific values ​​of all element times in the rail transit network. The computer device selects the specific value of each element time of the element time subset of the candidate path from the element time vector.

[0048] Step S203, summing up the specific value of each element time of the element time subset to obtain the path time of the candidate path.

[0049] In the application, the specific values ​​of the element times that make up the path time are summed up, and the sum result is the path time of the candidate path.

[0050] In this embodiment, the method of selecting the element time required for the path from the specific values ​​of the solved element time vectors to determine the path time can greatly improve the calculation accuracy of the path time.

[0051] In one embodiment, Figure 3 As shown, the construction process of the element time estimation model includes the following steps S301 to S304: Step S301, based on the rail transit network and the interval travel time between any two adjacent stations, determine the driving path of any traffic starting and ending point.

[0052] In the application, the Dijkstra algorithm is first used to calculate the shortest path among the valid paths of any traffic starting and ending points based on the rail transit network and the interval travel time of any two adjacent stations as the driving path between the starting and ending points.

[0053] Step S302: determining an element time matching matrix based on the element time vector of the rail transit network and the driving path of each transportation start and end point in the rail transit network.

[0054] In the application, according to the element time associated with the driving path of each traffic starting and ending point in the rail transit network, a matching matrix A of the path and the element time is obtained, which can be called the element time matching matrix.

[0055] Step S303, taking the travel time vector composed of the travel time of the traffic starting and ending points as the dependent variable and the element time vector as the independent variable, linear fitting is performed in combination with the element time matching matrix to obtain a preliminary fitting model.

[0056] In the application, the data set used to build the factor time estimation model refers to the travel data within a period of time, and the travel data comes from the passenger card swiping data. The travel data can be cleaned to obtain a high-quality data set, thereby improving the accuracy of the path time model.

[0057] Among them, the data cleaning for travel data is to remove non-routine data, among which non-routine data includes the following categories: 1) passenger travel data with transaction card type of employee card and work permit. 2) abnormal data of card swiping station (entry and exit are the same). 3) abnormal data of entry and exit station time (exit time is earlier than entry time). 4) data with travel time less than 1 minute and greater than 210 minutes, where the travel time is the difference between the exit time and the entry time.

[0058] In the application, considering the large number of passengers traveling during the morning and evening peak hours on weekdays, the number of passengers transferring is large and the transfer time is long compared to the off-peak hours. Therefore, in order to ensure the accuracy of the data, the data set can also be divided into three time period data sets according to the passenger travel time: morning peak (07:00-09:00 on weekdays), evening peak (17:30-19:30 on weekdays), and off-peak (other time periods), and the path time model is constructed for each time period data set.

[0059] Among them, for a single passenger, passenger travel time = (entry time + travel duration / 2), travel duration = (exit time - entry time).

[0060] In the application, for a data set including L records in a single time period, it is grouped according to the entry site O and the exit site D to obtain the travel data group of each OD, and the average travel time and travel time variance of each OD are determined according to the travel data group as the travel time of OD and the travel time variance of OD.

[0061] It should be noted that in order to ensure data uniformity, the travel data in each OD travel data group can be sorted in descending order according to travel duration, and the smallest 24.9% and the largest 25.1% (rounded off) of the travel data can be eliminated. Then, for the remaining travel data in the group, the average travel duration and the travel duration variance can be calculated as the OD travel duration and OD travel duration variance.

[0062] In the application, according to the rail transit network diagram, travel data, line site data and line network interval travel time operation planning data, all transfer channels, entry stations, exit stations, adjacent station travel intervals, station line information and other data in the rail transit network can be cleaned. Assume that for the rail transit network diagram D, the total number of ODs is n (n is a positive integer), the number of transfer channels is k, the number of entry stations is p, the number of exit stations is the same as the entry stations, and the number of adjacent train travel intervals is q. The number of element times generated is m= (k+2p+q). The element time includes transfer time, entry time, exit time, and train interval travel time. If the number of transfer channels is k, there are k transfer times in total. If the number of entry stations is p, there are p entry times. If the number of exit stations is p, there are p exit times. The train interval travel time is composed of the interval travel time of at least two adjacent stations. The number of adjacent train travel intervals is q, and the following element time vector is obtained : Element time vector , with a dimension of m=(k+2p+q). The travel time of each OD can be represented by the element time vector The element time matching matrix A is used to represent the inclusion of element time in the OD driving path. The element time matching matrix A is a sparse matrix with elements of 0 or 1. The dimension of A is n*m, and the element in the i-th row and j-th column is The value of is 0 or 1, indicating that the path time of the driving path of the i-th OD does not include or includes the j-th element time. The matrix form of the element time matching matrix A is:

[0063] In the application, the travel time vector composed of all OD travel times is used as the dependent variable, and the element time vector of the rail transit network is used as the dependent variable. Combined with the element time matching matrix, linear fitting is performed to obtain a preliminary fitting model. The expression of the preliminary fitting model is shown in the following formula (1): (1) in, is the OD travel time vector, is the travel time of the i-th OD, vector The dimension is the total number of ODs n. is the element time vector, represents the value of the i-th element time, and A is the element time matching matrix. The specific form of the above formula (1) is as follows: (2) In the above formula (2), is the travel time vector of OD , is the travel time of the i-th OD.

[0064] Step S304, based on the current value of the element time vector, the preliminary fitting model is adjusted to obtain the element time estimation model.

[0065] In the application, since a route has both outbound and inbound traffic, the rank of the element time matching matrix A is less than m, and the equations of the preliminary fitting model cannot be solved. Based on this, the current value of the element time vector is introduced (exist The train travel time of adjacent stations in the equation uses the known real value, and the initial values ​​of the other element times in the first iteration are all set to 5 minutes) to adjust the preliminary fitting model and obtain the element time calculation model. The formula of the element time calculation model is shown in (3): (3) In the above formula (3), =[ ; ] is a column vector with dimension (n+m), and the first n elements are the travel time vector of OD , after m dimension is the current value of the element time vector; X=[A;I] is a matrix formed by concatenating the element time matching matrix A and the unit matrix I. The element time matching matrix A is an n*m matrix, and the size of the unit matrix I is m*m, so the matrix (n+m)*m size matrix, element time vector is m-dimensional, Multiplying together gives a (n+m)-dimensional vector.

[0066] In the application, the vector matrix representation of the factor time estimation model is as follows: (4) In the above formula, Indicates the current value of the i-th element time of the element time vector.

[0067] The method of determining the element time calculation model in this embodiment can improve the accuracy of the element time calculation model.

[0068] In one embodiment, Figure 4 As shown, the implementation process of determining the specific value of the element time vector based on the element time calculation model includes the following steps S401 to S405: Step S401: construct a weight matrix based on the travel time variance of each traffic starting point and end point and the current value of each element time.

[0069] Step S402, performing matrix fitting calculation on the element time estimation model based on the weight matrix to obtain an estimated value of the element time vector.

[0070] In the application, in order to determine the minimum error of the path time model, the estimated value of each element time in the element time vector is continuously updated through the weight matrix to obtain an estimated travel time that is closer to the actual OD travel time.

[0071] In the application, a set of suitable element time vectors can be found by performing matrix fitting calculation on the element time model. , so that the error between the estimated travel time of OD and the actual travel time is minimized, the computer equipment performs matrix fitting calculation on the factor time estimation model through the weight matrix and the factor time matching matrix to obtain the factor time vector After the matrix fitting calculation, the element time vector is obtained. The expression is as shown in the following formula (5-1): (5-1) (5-2) (5-3) in, is the weight matrix, is the inverse matrix of V, which is a diagonal matrix composed of matrices C and B. The dimension of V is (n+m)*(n+m). The dimension of matrix C is n*n, and the dimension of B is m*m. C=diag(OD trip duration variance) is the first diagonal matrix composed of OD trip duration variance. B is the second diagonal matrix composed of the processed element time.

[0072] Step S403, based on the estimated value of the element time vector and the actual value of the interval travel time of each adjacent station, the estimated travel time of each traffic starting point and ending point is determined.

[0073] In the application, the estimated travel time of each traffic starting and ending point can be determined based on the estimated value of the element time vector and the actual value of the interval travel time of each adjacent station.

[0074] Step S404, determining a duration error based on the estimated travel duration of the traffic start and end points and the actual travel duration of the traffic start and end points.

[0075] In the application, there is a difference between the estimated travel time of each OD and the travel time of the OD, and the computer device calculates the error between the two through a preset error determination method.

[0076] In the application, the Normalized Root Mean Square Error (RMSN) can be used as the error indicator of the factor time estimation model, and the time error can be determined based on this error mode. . The calculation method of is shown in the following formula (6): (6) in, Each element in represents the estimated travel time of the corresponding OD, Each element in represents the actual travel time of the corresponding OD.

[0077] Step S405 , with the goal of minimizing the duration error, iteratively calculate the element time vector until the duration error converges to obtain a specific value of the element time vector.

[0078] In the application, minimizing the duration error is the adjustment goal, and the implementation process of iteratively updating the value of the element time vector is as follows: Each round of iterative calculation requires updating the vector in formula (5-1) and the matrix B in (5-3). The vector In the above, the specific value of the train section travel time is calculated by the method mentioned in the above step S102. It should be noted that, although the train travel time of adjacent stations is also predicted, it is actually known and has been calculated. Therefore, although the train travel time has a predicted value, the actual value of the train travel time is still used as the current value and final specific value of the train travel time in each iterative calculation and final determination. In addition, the initial value of the entry time, exit time and transfer time of the transfer channel is set to 5 minutes, and the vector is updated for each iterative calculation. The values ​​of other element times (entry time, exit time and transfer time of transfer channel) except the train section travel time in the last m items (i.e. the current value of the element time) are the values ​​of the corresponding element times calculated last time.

[0079] In the application, the elements in the diagonal matrix B are processed as follows: Matrix B is a diagonal matrix composed of the processed element time, and the values ​​and vectors of the diagonal elements The element times represented by the last m items correspond one to one. Initially, the elements of the train interval travel time part of the matrix B are processed as follows: (baseline variation coefficient * specific value of the corresponding train interval travel time) 2 ; The element processing method of the transfer time part is: (benchmark variation coefficient * initial value of corresponding transfer time * 10) 2; The element processing method of the pit stop time part is: (benchmark variation coefficient * initial value of corresponding pit stop time * 10) 2 ; The element processing method of the exit time part is: (benchmark variation coefficient * initial value of corresponding exit time * 10) 2 .

[0080] In the application, each iteration calculation must update the elements of the transfer / entry and exit time part of matrix B as (benchmark variation coefficient * the last calculation result of the corresponding element time) 2 .

[0081] In this embodiment, the efficiency and accuracy of determining the estimated value of the element time are improved based on the matrix fitting calculation method.

[0082] In one embodiment, Figure 5 As shown, the implementation method of the OD travel time variance includes the following steps S501 to S503: S501, based on the travel data of the rail transit network, determine the travel time and travel time variance of each transportation starting and ending point in the rail transit network.

[0083] In the application, the travel data of the rail transit network within a period of time (morning, afternoon, evening, etc.) is obtained, and the unconventional data is eliminated. The data is grouped according to the entry station O and the exit station D in the OD to obtain the travel data group of each OD. The travel time of a single OD is the mean of all the travel times in the travel data group of the OD. The travel time variance of a single OD is the variance of all the travel times in the travel data group of the OD.

[0084] It should be noted that in order to ensure data uniformity, the travel data in each OD travel data group can be sorted in descending order according to travel time, and the smallest 24.9% and the largest 25.1% (rounded off) of the travel data can be eliminated. Then, for the remaining travel data in the group, the average travel time and the travel time variance can be calculated as the OD travel time and OD travel time variance.

[0085] S502: If there is an abnormal traffic origin and destination with an abnormal travel time variance, a reference coefficient of variation is determined based on the mean travel time of other traffic origins and destinations except the abnormal traffic origin and destination and the mean travel time variance of the other traffic origins and destinations.

[0086] In the application, when the amount of OD trip data is 1, the variance of the trip time will be abnormal. Therefore, the OD with abnormal variance of the trip time is eliminated, and the baseline coefficient of variation of the OD is calculated for the remaining ODs.

[0087] (7) in, The function is used to find the mean. To eliminate the mean variance of the trip duration of abnormal OD, It is the mean OD travel time after excluding abnormal OD.

[0088] S503: Determine a corrected travel time variance of the abnormal traffic start and end points based on the reference variation coefficient and the travel time of the abnormal traffic start and end points.

[0089] In the application, based on the benchmark coefficient of variation, the OD with abnormal travel time variance is processed with the variance abnormal value: .

[0090] In this embodiment, the accuracy of data determination can be improved by correcting the OD with abnormal travel time using the reference variation coefficient.

[0091] In one embodiment, Figure 6 As shown, the implementation process of constructing the path allocation model includes the following steps S601 to S604: Step S601: construct a path utility model based on the path utility characteristics and utility coefficients.

[0092] In application, the path utility model characterizes the linear relationship between the path utility characteristics of a path and the path utility.

[0093] In the application, for any OD, the probability of its associated path assignment (i.e., the path assigned to the OD, which is selected by the passenger as the actual travel path from O to D) is related to the path utility. The higher the path utility, the greater the probability of being selected. The path utility model is constructed through the path utility characteristics and utility coefficients. The expression of the path utility model is as follows: (8) In the above formula (8), is the path utility of the path, which is used to characterize the linear relationship between the path utility characteristics, utility coefficient, and path utility. The path utility characteristic of the w-th candidate path is the path time , Transfer time , Number of transfers , Indicates whether the number of transfers reaches i times, 0.5 if reached, -0.5 if not reached. In traffic research, the threshold of the number of transfers is usually set to 5 times, i.e. i=1, 2, 3, 4, 5. Then the path utility feature of the w-th candidate path is specifically { , , , , , ,}. The utility coefficient includes { , , , , , , }, the extracted model parameters are { , , , },in, is the passenger's sensitivity to transfer time, μ is the passenger's sensitivity to travel time, , Passengers' sensitivity to the number of transfers.

[0094] Step S602: constructing a total utility index model of the traffic origin and destination based on the path utility of each candidate path of each traffic origin and destination.

[0095] In the application, the total utility index model represents the sum of the exp indexes of the path utilities of all valid candidate paths between the starting and ending points of the traffic.

[0096] Step S603: constructing a path allocation model for traffic origins and destinations based on the path utility model and the total utility index model.

[0097] In the application, the path allocation model is used to obtain the path allocation probability, and the path allocation probability of a single path is determined based on the path utility of the single path and the total utility index of the OD to which the single path belongs.

[0098] Therefore, based on the path utility model shown in formula (8), a total utility index model is constructed to characterize the total utility of the OD to which a single path belongs. Finally, a path allocation model is constructed based on the path utility model and the total utility index model. The specific form of the path allocation model is shown in the following formula (9): (9) In the above formula (9), is the utility mapping relationship of the path, The value of is the path utility of a path. Is the path The exp exponential form of the utility is, It is the sum of the exp utilities of all candidate paths of the same OD and is used to normalize the path assignment probability to ensure that the sum of the path assignment probabilities of all candidate paths of the same OD is 1.

[0099] Step S604: Optimize the model parameters of the path allocation model by means of vector updating and matrix calculation to obtain the optimal parameters of the path allocation model.

[0100] The model parameter is the utility coefficient, the path utility characteristics include the path time, the number of transfers of the path and the transfer time of the path, and the utility coefficient includes the sensitivity of the path time, the sensitivity of the number of transfers and the sensitivity of the transfer time.

[0101] In the application, there is { , , , } Four model parameters. Any change in any of these four model parameters will affect the path utility, and then affect the path allocation probability (that is, the probability that the path is selected by the passenger). Therefore, it is necessary to find the optimal solution of these four parameters to minimize the error of the path allocation model. If grid search is used, the grid space dimension of the four parameters grows at the level of the fourth power of x. Since there are tens of thousands of ODs in the rail transit network, the ordinary calculation method takes a long time.

[0102] Therefore, the model parameters can be updated based on the vector update and matrix calculation method to obtain the optimal solution of the model parameters. The vector update and matrix calculation method actually combines the OD travel data of the specified time period of the rail transit network to convert the tuning process of the model parameters of the path allocation model into a matrix calculation process based on multiple matrices and multiple vectors to solve the optimal solution of the model parameters. The vector update method can support global updates and greatly improve the calculation efficiency.

[0103] In the application, the matrices involved in the vector update and matrix calculation method of the path allocation model at least include the path utility feature matrix of the candidate path set, the path allocation matrix representing the allocation relationship between the candidate path and the OD, and the line allocation matrix representing the allocation relationship between the candidate path and the line. The vectors involved include the utility coefficient vector, the path utility vector, etc. The specific vector update method is described in detail below and will not be repeated here.

[0104] For example, assuming that the rail transit network N includes g lines, n ODs, and a maximum of 5 candidate paths are set for each OD, and the path utility features include 7 features such as path time, transfer time, and number of transfers, then the candidate path set of the rail transit network includes 5n candidate paths, each candidate path belongs to only one OD, and the path allocation matrix E is a sparse matrix with elements of 0 or 1, and the dimension of E is 5n*n. It means that the i-th candidate path belongs to the j-th OD, also called the i-th candidate path is assigned to the j-th OD. The path utility feature matrix C includes the path utility features of all candidate paths in the candidate path set. The dimension of C is 5n*7. represents the jth path utility feature of the i-th candidate path. The line allocation matrix A is a sparse matrix whose elements are 0 or 1, and the dimension of A is 5n*g, It means that the i-th candidate path passes through the j-th line. It should be noted that the abnormal paths in the candidate paths assigned by OD need to be eliminated. Eliminating abnormal paths is essentially to remove the allocation relationship, that is, to set the value of the abnormal path assigned to OD in the path allocation matrix E to 0. The path allocation matrix after eliminating abnormal paths is recorded as the path matching matrix R. R is a sparse matrix with elements of 0 or 1, and the dimension of R is 5n*n. Indicates that the i-th candidate path matches the j-th OD.

[0105] In this embodiment, the parameter tuning of the path allocation model of the whole rail transit network diagram is improved by means of matrix calculation, thereby improving the updating efficiency of the path allocation.

[0106] In one embodiment, Figure 7 As shown, the implementation process of step S603 includes the following steps S701 to S705: Step S701, based on the path utility feature matrix and path utility model of the candidate path set, determine the path utility vector for the candidate path set, the candidate path set includes each candidate path in the rail transit network, and each candidate path belongs to a transportation start and end point.

[0107] In the application, a candidate path set is used to represent all candidate paths in the rail transit network, and each candidate path can be assigned to a traffic start and end point. Based on the path utility model included in the path allocation model (shown in the above formula 8), the process of determining the path utility of each candidate path in the candidate path set is converted into a method based on the path utility matrix C and the utility coefficient vector Determine the path utility vector The path utility vector includes the path utility of each candidate path, that is, From the above path utility model, we can know that , the model parameters of the path allocation model are { , , , }, define the parameter vector , so we can Converted into the following parameter vector representation:

[0108] Step S702, based on the path allocation matrix of the candidate path set and the path utility vector, determine the path allocation probability vector of the candidate path set, the path allocation matrix represents the allocation relationship between the candidate paths and the traffic origin and destination.

[0109] In the application, the computer device can determine the path utility vector according to the path allocation matrix E of the candidate path set. The path utility of the candidate paths assigned to the same OD in the , so that the sum of the exp utilities of each OD can be obtained, the exp index of the path utility of the candidate path is divided by the sum of the exp utilities of the ODs to which it belongs, and the path assignment probability of the candidate path is obtained. Based on the sum of the exp utilities of each OD, the total utility index vector for the OD set is formed , The dimension is n, thus obtaining the path assignment probability vector , , where “. / ” is the division of the corresponding positions of the vectors, The dimension is 5n.

[0110] Step S703 : determining a predicted route flow vector for the route set based on the route allocation matrix and the route probability vector of the candidate route set, wherein each element of the route flow vector represents the predicted route flow of a single route.

[0111] In the application, obtain the OD trip quantity vector of the rail transit network , Each element in represents the actual passenger flow of a single OD, and the line allocation matrix A in Indicates that the i-th candidate path passes through the j-th route. Based on the OD trip quantity vector and the path probability vector Determine the path passenger flow vector corresponding to the candidate path set , The dimension is 5n. It should be noted that if the candidate path set has been identified as an abnormal path, the path passenger flow vector , operator symbol Multiply the corresponding positions of the vectors. Then use the transposed matrix of the route allocation matrix A and the path passenger flow vector By multiplying them together, we can get the predicted line flow of each line, thus getting the predicted line flow vector for the line set , predict line flow vector , predict line flow vector The dimension of is g.

[0112] Step S704: determining a flow error based on the actual line flow vector and the predicted line flow vector of the line set.

[0113] In the application, the computer device obtains the actual line flow vector corresponding to the line set , flow error is determined based on the standardized root mean square error RMSN method , flow error The calculation formula is shown in formula (10): (10) Step S705 , with the goal of minimizing the flow error, iteratively update the model parameters of the path allocation model to obtain the path allocation model.

[0114] In application, to minimize As the target, iteratively update the model parameters of the path allocation model, that is, the parameter vector , and iteratively execute steps S601-S604 to finally obtain a path allocation model.

[0115] In this embodiment, the calculation process of determining the path allocation probability of the candidate path is converted into a process based on matrix and vector operations, which can realize parameter tuning under the rail transit network diagram of the entire network, greatly improve the parameter update speed, and improve the efficiency of building the path allocation model.

[0116] In one embodiment, Figure 8 As shown, the implementation process of step S702 includes the following steps S801 to S803: Step S801 , eliminating abnormal candidate paths in the path allocation matrix of the candidate path set to obtain a path matching matrix of the candidate path set.

[0117] In the application, the computer device identifies the abnormal paths in the candidate path set according to the preset abnormal path judgment conditions, and removes the abnormal candidate paths from the path allocation matrix of the candidate path set to obtain the path matching matrix R of the candidate path set. The operation of obtaining the path matching matrix based on the path allocation matrix E is essentially to modify the element values ​​corresponding to the abnormal candidate paths in the path allocation matrix E to 0, and obtain the path matching matrix R. R is also a 0-1 sparse matrix with a dimension of 5n*n. Indicates that the i-th candidate path matches the j-th OD. An OD has at least one path matching it.

[0118] In the application, the abnormal path judgment conditions include at least one of the following conditions: path time > the time of the path with the shortest total time * 2 and path time > the time of the path with the shortest total time + 10 minutes, path time > the time of the path with the shortest total time + 20 minutes.

[0119] Step S802: obtaining a total utility index vector of a traffic origin and destination set based on a path matching matrix and a path utility vector.

[0120] In the application, for the path utility vector, the sum of the exp indexes of the path utilities of the remaining candidate paths after removing the abnormal paths under the same OD can be determined by multiplying it with the transposed matrix of the path matching matrix R to obtain the total exp utility of the OD. The sum of the exp utilities of each OD constitutes the total utility index vector , with dimension n.

[0121] Step S803: determining a path allocation probability vector of the candidate path set based on the path allocation matrix and the total utility index vector.

[0122] In the application, the computer device determines the path allocation probability of each candidate path in the candidate path set according to the path allocation matrix and the total utility index vector, thereby obtaining the path allocation probability vector of the candidate path set. , " " is the division of the corresponding positions of the vectors, The dimension is 5n.

[0123] This embodiment can effectively improve the accuracy of route selection, eliminate abnormal routes and reduce data noise, making route matching more accurate, thereby more truly reflecting the actual travel needs and preferences of passengers.

[0124] In one embodiment, Fig. 9 As shown, the implementation process of step S104 includes the following steps S901 to S903: Step S901 , based on the actual passenger flow at the transportation start and end points and the path allocation probability of each candidate path at the transportation start and end points, the path passenger flow of each candidate path at the transportation start and end points is determined.

[0125] Step S902: Based on the line information on each path, determine the proportion of sites to which each line on each path belongs.

[0126] Step S903, based on the passenger flow of each route and the proportion of stations on each line on the route, freight charges are settled to obtain the settlement share of each operating entity.

[0127] The present application also provides a rail transit freight fee clearing device, which is used to execute the steps in the rail transit freight fee clearing method embodiment. Fig.10 As shown, the rail transit freight fee clearing device 1000 provided in the embodiment of the present application includes: The time determination module 1010 is used to determine the specific value of the element time vector of the rail transit network based on the element time calculation model of the rail transit network. The time determination module 1020 is used to determine the path time of any candidate path between any traffic starting and ending points based on the specific value of the element time vector.

[0128] The probability determination module 1030 is used to determine the path allocation probability of the candidate path based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path and the path allocation model.

[0129] The freight clearing module 1040 is used to clear freight based on the actual passenger flow at the transportation starting and ending points, the path allocation probability of each candidate path at the transportation starting and ending points, and the line information of each candidate path to obtain the clearing share of each operating entity.

[0130] In one embodiment, the time determination module is also used to determine the element time subset of any candidate path of any traffic starting and ending point; select the specific value of each element time of the element time subset from the element time vector; sum the specific value of each element time of the element time subset to obtain the path time of the path.

[0131] In one embodiment, the time determination module is also used to determine the driving path of any traffic starting and ending point based on the interval travel time of the rail transit network and any two adjacent stations; determine the element time matching matrix based on the element time vector of the rail transit network and the driving path of each traffic starting and ending point in the rail transit network; use the travel time vector composed of the travel time of the traffic starting and ending points as the dependent variable, and the element time vector as the independent variable, combined with the element time matching matrix, to perform linear fitting to obtain a preliminary fitting model; based on the current value of the element time vector, adjust the preliminary fitting model to obtain the element time calculation model.

[0132] In one embodiment, the time determination module is also used to construct a weight matrix based on the travel time variance of each traffic starting and ending point and the current value of each element time; perform matrix fitting calculation on the element time calculation model based on the weight matrix to obtain an estimated value of the element time vector; determine the estimated travel time of each traffic starting and ending point based on the estimated value of the element time vector and the actual value of the interval travel time of each adjacent station; determine the time error based on the estimated travel time of the traffic starting and ending point and the actual travel time of the traffic starting and ending point; and iteratively update the element time vector with the goal of minimizing the time error until the time error converges to obtain a specific value of the element time vector. In one embodiment, the time determination module is also used to determine the travel time and travel time variance of each traffic starting and ending point in the rail transit network based on the travel data of the rail transit network; if there are abnormal traffic starting and ending points with abnormal travel time variance, then based on the mean travel time of other traffic starting and ending points except the abnormal traffic starting and ending points and the travel time variance of the other traffic starting and ending points, a baseline coefficient of variation is determined; based on the baseline coefficient of variation and the travel time of the abnormal traffic starting and ending points, a corrected travel time variance of the abnormal traffic starting and ending points is determined.

[0133] In one embodiment, the probability determination module is further used to construct a path utility model of the path based on the path utility characteristics and utility coefficient of the path; construct a total utility index model of the traffic start and end point based on the path utility of each candidate path of each traffic start and end point; construct a path allocation model of the traffic start and end point based on the path utility model and the total utility index model; optimize the model parameters of the path allocation model through vector update and matrix calculation to obtain the optimal parameters of the path allocation model; wherein the model parameter is the utility coefficient, the path utility characteristics include path time, the number of transfers of the path and the transfer time of the path, and the utility coefficient includes path time sensitivity, transfer number sensitivity and transfer time sensitivity.

[0134] In one embodiment, the probability determination module is also used to determine the path utility vector for the candidate path set based on the path utility feature matrix of the candidate path set and the path utility model; the candidate path set includes each candidate path in the rail transit network, and each candidate path belongs to a traffic origin and destination; based on the path allocation matrix of the candidate path set, the path utility vector and the total utility index vector of the traffic origin and destination set, the path allocation probability vector of the candidate path set is determined, and the path allocation matrix represents the allocation relationship between the candidate paths and the traffic origin and destination; based on the line allocation matrix of the candidate path set and the path probability vector, the predicted line flow vector for the line set is determined, and each element of the predicted line flow vector represents the predicted line flow of a single line; based on the actual line flow vector of the line set and the predicted line flow vector, the flow error is determined; with the goal of minimizing the flow error, the model parameters of the path allocation model are iteratively updated to finally obtain a path allocation model with converged flow error.

[0135] In one embodiment, the probability determination module is also used to eliminate abnormal candidate paths in the path allocation matrix of the candidate path set to obtain the path matching matrix of the candidate path set; based on the path matching matrix and the path utility vector, obtain the total utility index vector of the traffic starting and ending point set; based on the path allocation matrix and the total utility index vector, determine the path allocation probability vector of the candidate path set.

[0136] In one embodiment, the path clearing module is also used to determine the path passenger flow of each candidate path of the transportation starting and ending points based on the actual passenger flow of the transportation starting and ending points and the path allocation probability of each candidate path of the transportation starting and ending points; determine the proportion of stations belonging to each line on the path based on the line information on each path; and perform freight clearing based on the path passenger flow of each path and the proportion of stations belonging to each line on the path to obtain the clearing share of each operating entity.

[0137] In application, each module in the freight clearing device of rail transit can be a software program module, or can be implemented by different logic circuits integrated in a processor, or can be implemented by multiple distributed processors.

[0138] Fig.11 A schematic diagram of the structure of a computer device for implementing a method for clearing freight charges for rail transit provided in an embodiment of the present application. Fig.11 As shown, the computer device 1100 of this embodiment includes: at least one processor 1110 ( Fig.11 Only one is shown in the figure) a processor, a memory 1120, and a computer program 1130 stored in the memory 1120 and executable on the at least one processor 1110, and when the processor 1110 executes the computer program 1130, the steps in any of the above-mentioned exception handling method embodiments are implemented.

[0139] The computer device 1100 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 1100 may include, but is not limited to, a processor 1110 and a memory 1120. Those skilled in the art will appreciate that Fig.11 The computer device 1100 is merely an example and does not constitute a limitation on the computer device 1100 . The computer device 1100 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device 1100 may also include input and output devices, network access devices, etc.

[0140] The processor 1110 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0141] In some embodiments, the memory 1120 may be an internal storage unit of the computer device 1100, such as a hard disk or memory of the computer device 1100. In other embodiments, the memory 1120 may also be an external storage device of the computer device 1100, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 1100. Further, the memory 1120 may also include both an internal storage unit of the computer device 1100 and an external storage device. The memory 1120 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 1120 may also be used to temporarily store data that has been output or is to be output.

[0142] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0143] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0144] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0145] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0146] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the device / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0147] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0148] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0149] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0150] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for clearing freight for rail transit, characterized in that: include: Determining specific values ​​of element time vectors of the rail transit network based on an element time calculation model of the rail transit network; For any candidate path of any traffic starting and ending point, determining the path time of the candidate path based on the specific value of the element time vector; Determining a path allocation probability of the candidate path based on the path time of the candidate path, the transfer time of the candidate path, the number of transfers of the candidate path, and a path allocation model; Based on the actual passenger flow at the transportation starting and ending points, the path allocation probability of each candidate path at the transportation starting and ending points, and the line information of each candidate path, freight settlement is performed to obtain the settlement share of each operating entity.

2. The method for clearing freight for rail transit according to claim 1, characterized in that: The determining of the path time of any candidate path between any traffic starting and ending points based on the specific value of the element time vector comprises: For any candidate path of any traffic starting and ending point, determine the element time subset of the candidate path; Selecting, from the element time vector, a specific value of each element time of the element time subset; The specific value of each element time of the element time subset is summed to obtain the path time of the candidate path.

3. The method for clearing freight for rail transit according to claim 1, characterized in that: The method further comprises: Determine the driving path of any traffic starting and ending point based on the rail transit network and the interval travel time between any two adjacent stations; Based on the element time vector of the rail transit network and the driving path of each transportation start and end point in the rail transit network, determine the element time matching matrix; The travel time vector composed of the travel time of the starting and ending points of the traffic is used as the dependent variable, the element time vector is used as the independent variable, and the element time matching matrix is ​​combined to perform linear fitting to obtain a preliminary fitting model; Based on the current value of the element time vector, the preliminary fitting model is adjusted to obtain the element time extrapolation model.

4. The method for clearing freight for rail transit according to claim 1, characterized in that: The element time calculation model based on the rail transit network determines the specific value of the element time vector of the rail transit network, including: Based on the variance of travel time at each traffic starting point and end point and the current value of each element time, a weight matrix is ​​constructed; Performing matrix fitting calculation on the element time estimation model based on the weight matrix to obtain an estimated value of the element time vector; Based on the estimated value of the element time vector and the actual value of the interval travel time of each adjacent station, the estimated travel time of each traffic starting point and ending point is determined; Determine the time error based on the estimated travel time between the transportation start and end points and the actual travel time between the transportation start and end points; With the goal of minimizing the duration error, the element time vector is iteratively updated until the duration error converges to obtain a specific value of the element time vector.

5. The method for clearing freight for rail transit as claimed in claim 4, characterized in that: The method further comprises: Based on the travel data of the rail transit network, determine the travel time and travel time variance of each traffic starting and ending point in the rail transit network; If there are abnormal traffic start and end points with abnormal travel time variance, the benchmark coefficient of variation is determined based on the mean of the travel time of other traffic start and end points except the abnormal traffic start and end point and the mean of the travel time variance of the other traffic start and end points; Based on the reference coefficient of variation and the travel time of the abnormal traffic start and end points, a corrected travel time variance of the abnormal traffic start and end points is determined.

6. The method for clearing freight for rail transit as claimed in claim 1, characterized in that: The method further comprises: Based on the path utility characteristics and utility coefficients of the path, the path utility model of the path is constructed; Based on the path utility of each candidate path of each traffic origin and destination, a total utility index model of traffic origin and destination is constructed; Based on the path utility model and the total utility index model, constructing a path allocation model for the traffic start and end points; Optimizing the model parameters of the path allocation model by means of vector updating and matrix calculation to obtain the optimal parameters of the path allocation model; Among them, the model parameter is the utility coefficient, the path utility characteristics include path time, the number of transfers of the path and the transfer time of the path, and the utility coefficient includes path time sensitivity, transfer number sensitivity and transfer time sensitivity.

7. The method for clearing freight for rail transit as claimed in claim 6, characterized in that: The method of optimizing the model parameters of the path allocation model by means of vector updating and matrix calculation to obtain the optimal parameters of the path allocation model includes: Determining a path utility vector for a candidate path set based on a path utility feature matrix of a candidate path set and the path utility model, wherein the candidate path set includes each candidate path of the rail transit network, and each candidate path belongs to a transportation origin and destination; Determine the path allocation probability vector of the candidate path set based on the path allocation matrix of the candidate path set, the path utility vector and the total utility index vector of the traffic origin and destination set, wherein the path allocation matrix represents the allocation relationship between the candidate paths and the traffic origin and destination; Determine a predicted route flow vector for the route set based on the route allocation matrix of the candidate route set and the route probability vector, wherein each element of the predicted route flow vector represents the predicted route flow of a single route; Determining a flow error based on an actual line flow vector of the line set and the predicted line flow vector; With the goal of minimizing the flow error, the model parameters of the path allocation model are iteratively updated, and finally a path allocation model with converged flow error is obtained.

8. The method for clearing freight for rail transit as claimed in claim 7, characterized in that: The determining of the path allocation probability vector of the candidate path set based on the path allocation matrix of the candidate path set, the path utility vector and the total utility index vector of the traffic origin and destination set comprises: Eliminate abnormal candidate paths in the path allocation matrix of the candidate path set to obtain a path matching matrix of the candidate path set; Based on the path matching matrix and the path utility vector, a total utility index vector of a traffic origin and destination set is obtained; Based on the path allocation matrix and the total utility index vector, a path allocation probability vector of the candidate path set is determined.

9. The method for clearing freight for rail transit as claimed in claim 1, characterized in that: The method of performing freight clearing based on the actual passenger flow at the transportation starting and ending points, the path allocation probability of each candidate path at the transportation starting and ending points, and the line information of each candidate path to obtain the clearing share of each operating entity includes: Determine the path passenger flow of each candidate path of the traffic starting and ending point based on the actual passenger flow of the traffic starting and ending point and the path allocation probability of each candidate path of the traffic starting and ending point; Based on the route information on each candidate path, determining the proportion of sites to which each route on the candidate path belongs; Based on the passenger flow of each candidate route and the proportion of stations on each line on the candidate route, freight settlement is carried out to obtain the settlement share of each operating entity.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the computer device is caused to implement the method according to any one of claims 1 to 9.

Images