Regional multi-modal rail transit network passenger flow distribution prediction method, medium and device

Abstract

The present application relates to the field of rail transit, and specifically provides a regional multi-mode rail transit network passenger flow distribution prediction method, medium and computer device, wherein the regional multi-mode rail transit network passenger flow distribution prediction method comprises the following steps: improving the K-shortest path search algorithm according to the hierarchical characteristics of the pre-constructed regional multi-mode rail transit network; constructing a set of feasible paths between any OD based on the pre-generated passenger flow OD matrix of the "region + urban area" and the improved K-shortest path search algorithm; and integrating the set of feasible paths between any OD into a travel path probability selection model to realize the passenger flow distribution of the regional multi-mode rail transit network. Through such a structure, more scientific integrated analysis of the passenger flow demand of the regional rail transit network can be provided, so as to meet the diversified travel demands of passengers in different regions, with different functions and different speeds.

Description

Technical Field

[0001] This invention relates to the field of rail transit passenger flow forecasting, and specifically provides a method for forecasting passenger flow distribution in a regional multimodal rail transit network, a computer-readable storage medium, and a computer device. Background Technology

[0002] Currently, my country's urbanization development has transitioned from single-center city development to regional development (such as metropolitan areas and urban agglomerations). Rail transit is a key infrastructure supporting the development of metropolitan areas / urban agglomerations. For rail transit networks within a region, those undertaking different functions typically exhibit a distinct hierarchical structure and thus multimodal characteristics, hence the term "regional multimodal rail transit network." Such regional multimodal rail transit networks include (but are not limited to) high-speed rail subnetworks, intercity rail subnetworks, urban (suburban) rail subnetworks, and urban rail transit subnetworks.

[0003] Accurately grasping the passenger flow demand distribution characteristics of rail transit networks is a prerequisite for scientifically designing transportation service products and formulating transportation organization plans. Passenger flow demand distribution characteristics are usually derived through passenger flow distribution forecasting methods. Existing passenger flow distribution forecasting methods are typically based on single-mode rail transit, such as relying on traditional transportation demand analysis theory or using the "four-stage method" framework for separate analysis after implementing the third stage of transportation mode division. Because different rail transit networks have significant differences in structure, transportation organization methods, passenger travel characteristics, and urban public transportation, individual analysis methods are insufficient to support the analysis of travel demand in multi-mode integrated rail transit networks that include multiple rail transit sub-networks. Therefore, passenger flow distribution forecasting for multi-mode rail transit networks is quite necessary. For example, passenger flow distribution forecasting methods should reflect both the travel characteristics of passengers traveling across different modes and the differences between the operation services of different rail transit modes, thereby providing a more scientific analysis of integrated travel.

[0004] Current research directions regarding the prediction of passenger flow distribution in multimodal rail transit include:

[0005] One approach involves adding virtual transfer stations to a single-layer network and constructing a three-layer, multi-modal rail transit topology model without implicit connections using the SPACEL method. This model utilizes Dijkstra's method to search for the K-shortest paths between origins and destinations, establishes a passenger familiarity function with the network, and calculates path selection probabilities using the Logit model. However, this method, which constructs a multi-modal rail transit topology network using the SPACEL method, fails to consider the differences in train operation organization modes among different rail transit sub-networks.

[0006] Secondly, some scholars have adopted a "four-stage" passenger flow forecasting approach, studying passenger flow distribution forecasting at both the urban area and city-wide levels, by constructing a two-layer maximum entropy distribution model. At the regional level, they propose a two-stage forecasting approach for rail passenger flow classification: initial classification and further classification. It can be seen that this scheme uses the "four-stage method" as a framework, analyzing passenger behavior separately after implementing transportation mode classification, without considering passenger behavior analysis under cross-modal travel conditions.

[0007] It can be seen that existing technologies for predicting passenger flow distribution in multimodal rail transit do not consider the differences in train operation modes of different rail transit sub-networks. The prediction methods are mostly based on the "four-stage method" framework and are analyzed separately after the implementation of the traffic mode division. The analysis theory and methods under cross-modal travel conditions are relatively weak.

[0008] Accordingly, a new technical solution is needed in this field to solve the above problems. Summary of the Invention

[0009] The present invention aims to solve at least part of the above-mentioned technical problems. Specifically, the present invention provides a method for predicting passenger flow distribution in a regional multimodal rail transit network. By starting from the perspective of passengers' intermodal travel across rail transit lines, it seeks to more scientifically support the analysis of passenger flow demand in the regional rail transit network, or to provide a more scientific integrated analysis of passenger flow demand in the regional rail transit network, thereby meeting the diverse travel needs of passengers in different regions, with different functions and at different speeds.

[0010] In a first aspect, the present invention provides a method for predicting passenger flow distribution in a regional multimodal rail transit network. The method includes the following steps: improving the K-short path search algorithm based on the hierarchical characteristics of a pre-constructed regional multimodal rail transit network; constructing a set of feasible paths between any ODs based on a pre-generated "regional + urban area" passenger flow OD matrix and the improved K-short path search algorithm; and integrating the set of feasible paths between any ODs into a pre-constructed cross-mode travel path selection probability model to achieve passenger flow allocation in the regional multimodal rail transit network.

[0011] This structure enables more scientific predictions of passenger flow distribution in regional multimodal rail transit networks.

[0012] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in a regional multimodal rail transit network, the method for constructing the regional multimodal rail transit network includes: constructing a transportation service subnetwork corresponding to the rail transit subnetwork based on the differences in train operation modes of different rail transit subnetworks; and connecting the topologies between different transportation service subnetworks based on a virtual node fusion method.

[0013] This structure enables the creation of a regional multimodal rail transit network that reflects real-world conditions.

[0014] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in a multi-modal rail transit network, the "constructing a transportation service sub-network corresponding to the rail transit sub-network based on the differences in train operation modes of different rail transit sub-networks" includes: constructing a high-speed rail and intercity rail train service sub-network based on train timetables; and / or constructing a suburban rail train service sub-network based on train operation plans; and / or constructing a physical sub-network of urban rail transit based on physical stations and sections.

[0015] This structure provides a possible way to construct the transportation service sub-network.

[0016] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in a multi-modal rail transit network, the method for generating the "region + urban area" passenger flow OD matrix includes: dividing the region into traffic zones of varying coarseness; predicting passenger flow in traffic zones of towns surrounding the region and between traffic zones of towns surrounding the region and the urban area based on a gravity model; predicting passenger flow within the urban area based on a system balance prediction model; and generating the "region + urban area" passenger flow OD matrix based on the prediction results.

[0017] This structure allows for the creation of a "regional + urban area" passenger flow OD matrix that comprehensively reflects the passenger flow of the regional multi-modal rail transit network.

[0018] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in a multi-modal rail transit network, the method for constructing the cross-modal travel path selection probability model includes: analyzing the entire travel process of passengers and constructing a cross-modal connecting "travel chain" for passengers; and constructing a cross-modal travel path selection probability model based on the cross-modal connecting "travel chain".

[0019] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in regional multimodal rail transit networks, the "improving the K-shortest path search algorithm based on the hierarchical characteristics of the pre-constructed regional multimodal rail transit network" includes: Let G... i '=(N i A i Let be the i-th level rail transit subnetwork, where i = 1, 2, and 3 represent the high-speed rail and intercity rail subnetwork, the suburban rail subnetwork, and the urban rail transit subnetwork, respectively. The K-shortest path from the starting node s to the ending node d of the i-th level rail transit subnetwork is... It can be represented as in This represents the length of the Kth shortest path from the first-level rail transit subnetwork to the (i-1)th-level rail transit subnetwork. This represents the length of the Kth shortest path of node d in the i-th level rail transit subnet; Where d k (n m S' represents the length of the Kth shortest path in the m-th level rail transit subnetwork; i D′ represents the connected node closest to the starting node S in the i-th level rail transit subnet. i This represents the connected node closest to the end node D in the i-th level rail transit subnetwork. It determines whether a high-speed rail and intercity rail subnetwork G1' exists. If G1' exists, the path is obtained using the K-shortest path search algorithm. in Determine if a subnetwork of the urban rail transit system G2' exists. If both G1' and G2' exist, obtain the path through a K-shortest path search. and If G1' does not exist but G2' does exist, the path is obtained using the K-shortest path search algorithm. Determine if an urban rail transit subnetwork G3' exists. If both G2' and G3' exist, obtain the K shortest path through path search. and If G2' does not exist, but G1' and G3' both exist, the K shortest paths are obtained through a path search algorithm. and If neither G1' nor G2' exists, the path is obtained using the K-shortest path search algorithm. Merge the paths of each sub-network.

[0020] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in a multi-modal rail transit network, the "constructing a set of feasible paths between any OD based on a pre-generated passenger flow OD matrix of 'region + city' and an improved K-shortest path search algorithm" includes: determining the set of alternative paths between ODs based on the principles of non-repetition of nodes and arcs, reasonable range of travel path selection, and limited number of transfers, and then selecting the set of feasible paths between ODs from the paths.

[0021] In one possible implementation of the above-mentioned method for predicting passenger flow distribution in regional multimodal rail transit networks, the step of "integrating the set of feasible paths between any origin-destination (OD) regions into a pre-constructed cross-modal travel path selection probability model to allocate passenger flow to the regional multimodal rail transit network" includes:

[0022] S810. Considering different types of passenger flow, determine the passenger flow between different OD pairs rs. Assume that the total passenger flow of the nth type of passenger flow between OD pairs rs is... A hierarchical K-shortest path search algorithm is used to generate candidate paths, and feasible paths between ODs are selected according to corresponding principles to obtain the corresponding set of effective paths. Determine the total generalized cost value for each path. And obtain the cost value of the path with the minimum cost.

[0023] S820. Calculate the probability that the kth effective path is selected for the nth type of passenger flow between OD and rs based on the cross-mode travel path probability selection model. Calculate the passenger flow on each path k based on the probability values ​​of each effective path selected by the nth type of passenger flow between OD pairs rs. in

[0024] S830, Based on the passenger flow of the nth type of passenger flow on the path k between OD pairs rs. Calculate the passenger flow in the interval (a, b) between rs. If path k is contained in the interval (a, b), then Otherwise, it is 0, where

[0025] S840, repeating S820 and S830, calculate the path passenger flow and interval passenger flow for all OD pairs of all passenger flow categories, and obtain the cross-sectional passenger flow between intervals (a, b). in

[0026] S850: The nth type of passenger flow is allocated sequentially. When all types of passenger flow have been allocated, proceed to S6; otherwise, return to S1 and continue the flow allocation.

[0027] S860. Once all types of passenger flow have been allocated, further calculate the interval flow and sum the path flows of different types of passengers to obtain the total flow between intervals (a, b). Passenger flow allocation is then complete.

[0028] It can be seen that, in the preferred embodiment of the regional multimodal rail transit network passenger flow distribution prediction method of the present invention, by considering the differences in the train operation modes of different rail transit sub-networks, it is possible to construct a more scientific transportation service sub-network, and based on this, further construct a multimodal rail transit integrated composite network containing multiple modes of rail transit sub-networks. By dividing traffic zones of different coarseness, and based on gravity models and system balance prediction models, the total demand and passenger flow OD matrix are predicted to meet passenger flow requirements at different levels. By constructing a cross-mode travel path selection probability model based on "travel chains," the analytical theory and methods under the condition of cross-modal combined travel (using a combination of multiple rail transit sub-networks) are supplemented. By improving the K-shortest path search algorithm based on the hierarchical characteristics of the regional multimodal rail transit network, it is expected to more accurately predict the passenger flow distribution of the regional multimodal rail transit network.

[0029] Specifically, in a preferred embodiment of the regional multimodal rail transit network passenger flow distribution prediction method of the present invention, compared with direction one mentioned in the background art, the present invention combines the characteristics of different modes and passenger flow features to construct a transportation service sub-network corresponding to the rail transit sub-network, and connects the topologies between different transportation service sub-networks based on the virtual node fusion method, effectively integrating the rail transit operation organization mode with network construction. Compared with direction two mentioned in the background art, the present invention analyzes the entire travel process of passengers in the regional multimodal rail transit system, and establishes a passenger cross-mode connecting travel path selection behavior model based on the entire process travel chain, which can effectively support the "integrated" analysis of the transportation demand of the regional multimodal rail transit composite network.

[0030] In a second aspect, the present invention provides a computer-readable storage medium including a memory adapted to store a plurality of program codes adapted to be loaded and run by a processor to perform the regional multimodal rail transit network passenger flow distribution prediction method described in any of the preceding claims.

[0031] It is understood that the computer-readable storage medium has all the technical effects of the regional multimodal rail transit network passenger flow distribution prediction method described in any of the foregoing claims, and will not be repeated here.

[0032] Those skilled in the art will understand that all or part of the processes in the regional multimodal rail transit network passenger flow distribution prediction method of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which includes, but is not limited to, program code for executing the aforementioned regional multimodal rail transit network passenger flow distribution prediction method. For ease of explanation, only the parts relevant to the present invention are shown. The computer program code can be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable storage medium can include: any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content contained in the computer-readable storage medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0033] In a third aspect, the present invention provides a computer device including a memory and a processor, the memory being adapted to store a plurality of program codes adapted to be loaded and run by the processor to perform the regional multimodal rail transit network passenger flow distribution prediction method described in any of the preceding claims.

[0034] It is understood that this device possesses all the technical effects of the aforementioned regional multimodal rail transit network passenger flow distribution prediction method, which will not be elaborated upon here. This device can be a computer-controlled device comprising various electronic devices. Attached Figure Description

[0035] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart illustrating a method for predicting passenger flow distribution in a regional multimodal rail transit network according to an embodiment of the present invention is shown.

[0037] Figure 2This invention illustrates a schematic diagram of the structure of a high-speed railway and intercity railway train spatiotemporal service subnetwork based on train timetables in a regional multimodal rail transit network passenger flow distribution prediction method according to an embodiment of the present invention.

[0038] Figure 3 This diagram illustrates the structure of a sub-network of urban (suburban) railway train services constructed based on train operation schemes in a regional multimodal rail transit network passenger flow distribution prediction method according to an embodiment of the present invention.

[0039] Figure 4 This diagram illustrates the structure of a method for predicting passenger flow distribution in a regional multimodal rail transit network according to an embodiment of the present invention, which uses a virtual node fusion strategy to solve the same-station and different-station transfers in a single-mode rail transit subnetwork.

[0040] Figure 5 This diagram illustrates the structure of same-station and different-station transfers in the regional multimodal rail transit network constructed using the passenger flow distribution prediction method for regional multimodal rail transit networks according to an embodiment of the present invention. Detailed Implementation

[0041] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0042] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0043] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or couplings. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0044] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0045] Furthermore, to better illustrate the present invention, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that the present invention can be implemented even without certain specific details. In some instances, well-known algorithmic principles are not described in detail to highlight the main points of the present invention.

[0046] This invention provides a method for predicting passenger flow distribution in a regional multimodal rail transit network. Based on the constructed regional multimodal rail transit network, the generated "regional + urban area" passenger flow OD matrix, and the constructed cross-mode travel route selection probability model, it presents a passenger flow distribution theory and methodology system that better matches the structural characteristics and demand features of the regional multimodal rail transit network.

[0047] Main reference Figure 1 , Figure 1 A flowchart illustrating a method for predicting passenger flow distribution in a regional multimodal rail transit network according to an embodiment of the present invention is shown. Figure 1 As shown, in one possible implementation, the method for predicting passenger flow distribution in a regional multimodal rail transit network mainly includes the following steps:

[0048] S110. Based on the hierarchical characteristics of a pre-constructed regional multi-modal rail transit network, improve the K-shortest path search algorithm;

[0049] S120. Based on the pre-generated "regional + urban area" passenger flow OD matrix and the improved K-shortest path search algorithm, construct a set of feasible paths between any ODs;

[0050] S130. Integrate the set of feasible paths between any O-modes into a pre-built cross-mode travel path selection probability model to achieve passenger flow allocation for the regional multi-mode rail transit network.

[0051] The following sections will first elaborate on the construction of regional multimodal rail transit networks, the generation of passenger flow OD matrices for "regional + urban areas", and the construction of cross-mode travel route selection probability models.

[0052] I. Construction of a Regional Multimodal Rail Transit Network:

[0053] In one possible implementation, the construction of a regional multimodal rail transit network mainly includes the following steps:

[0054] S11. Based on the differences in train operation modes of different rail transit sub-networks, construct a transportation service sub-network corresponding to the rail transit sub-network.

[0055] S12. Based on the virtual node fusion method, solve the topology connection problem between different transportation service sub-networks to achieve effective fusion between different transportation service sub-networks, thereby constructing a regional multi-modal rail transit network.

[0056] In one possible implementation, S11 includes:

[0057] S11(1) Construct a high-speed railway and intercity railway train service sub-network based on train timetables.

[0058] Main reference Figure 2 Because high-speed trains and intercity trains have low departure frequencies and long stop times, train operation plans are quite complex. When passengers choose their travel routes, they are actually choosing the "train" rather than the "travel route." Therefore, it is necessary to construct a train spatiotemporal service subnetwork by combining train timetable information.

[0059] In one possible implementation, a railway physical network G1 = (V1, E1) is defined, where V1 is the set of railway physical stations and E1 is the set of railway physical sections; a train set Ω = {T = (V1, E1)} is defined. T D T A T )}, where V T This refers to the sequence of stops along the route for train T. v Ti ∈V,D T This represents the departure time sequence of train T at each stop along its route. A T This represents the arrival time sequence of train T at each stop along its route. n(T) represents the number of stops for train T. The train spatiotemporal service subnetwork G1' = (N, A) is defined, where N is the spatiotemporal node set, i.e., the train stop node set, which includes the train departure node set, the train stop node set, and the train arrival node set. A is the spatiotemporal arc segment set, mainly including four subsets: boarding arc segment set, running arc segment set, transfer arc segment set, stopping arc segment set, and alighting arc segment set. The specific process of constructing the high-speed railway and intercity railway train spatiotemporal service subnetwork based on train timetables is as follows:

[0060] (1) Based on the train timetable, the physical station nodes that each train stops at are extended on the time axis. The spatiotemporal network nodes represent the spatiotemporal points where the train actually runs. The train's arrival spatiotemporal points and departure spatiotemporal points are represented by the train stopping node (train terminal node) and the train departure node, respectively. For example, all three types of nodes are ternary quantities including the train information to which the node belongs, the station information to which the node belongs, and the time information contained in the node.

[0061] (2) Connect the arrival and departure nodes of the same train on the same line at the same physical node through the train stop arc, connect the departure and arrival nodes of the same train on the same line at adjacent stations through the train operation arc, connect a physical node with the departure node of a train on a line passing through the station through the boarding arc, and connect a physical node with the arrival node of a train on a line passing through the station through the alighting arc.

[0062] (3) Based on the timetable information, select any transferable train after the train arrives at the station. The time interval between the two trains (i.e., the time difference between the departure time of the later train and the arrival time of the earlier train) satisfies the entire transfer process for passengers. Connect the arrival and departure times of different trains at the same physical station through same-station transfer arcs. Given the maximum walking transfer radius that passengers can tolerate, and using this distance as the search radius, conduct transfer searches around the station. Connect the arrival and departure times of different trains at different physical stations through different-station transfer arcs.

[0063] S11(2) Construct a sub-network of urban (suburban) railway train services based on train operation plans.

[0064] Main reference Figure 3 Because suburban railways possess characteristics of both railways and urban rail transit, with high-density, public transport-like operations, expanding the suburban railway subnetwork using timetables would result in an overly complex network structure due to the excessive number of nodes and arcs. Therefore, a train service network representation based on operational plans is adopted to construct the suburban railway train service subnetwork. This approach reflects the service characteristics of trains while simplifying the network scale.

[0065] In one possible implementation, the urban (suburban) railway physical network G2 = (V2, E2) is defined, where V2 is the set of urban (suburban) railway physical stations and E2 is the set of urban (suburban) railway physical sections; V is defined as... R For the stop sequence along the route of trains operating from station R, V R =(v R1 v R2 , ..., v R(n)Define the service network G2' = (N, A), where N is the spatiotemporal node set mentioned above, and A is the spatiotemporal arc set mentioned above. The specific process of constructing the urban (suburban) railway train service subnetwork based on the train operation scheme is as follows:

[0066] (1) Based on the train operation plan, the physical stations covered by the same route are expanded to generate virtual stop nodes. If the node belongs to a binary quantity including the station to which the node belongs and the route to which the node belongs.

[0067] (2) Connect adjacent stopping nodes of the same route through the train running arc, connect the physical node with the stopping node of the route at the station through the boarding arc, and connect the physical node with the stopping node of the route at the station through the alighting arc.

[0068] (3) Construct same-station transfer arcs between different routes at the same station; by giving the maximum tolerable walking transfer radius and using this distance as the search radius, conduct transfer search around the station to construct different-station transfer arcs between different routes at different physical stations.

[0069] S11(3) Construct a physical sub-network of urban rail transit based on physical stations and sections.

[0070] Because urban rail transit trains are characterized by high departure frequency and simple operation plans, passengers' travel choices often involve choosing a "travel route" rather than a specific "train." Therefore, the urban rail transit subnetwork can be characterized by a physical subnetwork.

[0071] In one possible implementation, the urban rail transit physical sub-network is defined as G3 = (V, E, L), where V is the set of physical stations, E is the set of physical sections, and L is the set of lines. Network G3' = (N, A) is defined, where N is the set of spatiotemporal nodes mentioned above, and A is the set of spatiotemporal arcs mentioned above. The specific process of constructing the urban rail transit physical sub-network based on physical stations and sections is as follows:

[0072] Each station node in urban rail transit is a physical station in the rail transit physical subnetwork. Connecting adjacent physical nodes generates interval arcs, and the line corresponding to one or more interval arcs is the actual line in the rail transit physical subnetwork.

[0073] In one possible implementation, S12 mainly includes:

[0074] S12(1) Construct a single-mode rail transit subnet based on the virtual node fusion strategy.

[0075] Main reference Figure 4In one possible implementation, the specific process of constructing a single-mode rail transit subnet based on a virtual node fusion strategy is as follows:

[0076] (1) Generate same-station transfer arcs. For all stations on the urban rail transit physical subnetwork, traverse all the rail lines passing through them and count the number of lines. Expand the original stations into a corresponding number of virtual nodes, with each virtual node corresponding to one line. A pair of virtual nodes corresponding to the same station is called a co-location node. Connect co-location nodes corresponding to different lines in pairs to generate same-station transfer arcs. If the number of co-location nodes is 1, it means that only one line passes through the station, and in this case, there is no need to introduce same-station transfer arcs.

[0077] (2) Node search. Given the maximum walking transfer radius that passengers can tolerate, and using this distance as the search radius, a transfer search is conducted around the station to find all nodes that allow transfers between different lines. The searched nodes must be virtual nodes on different lines because: rational travelers will not transfer on the same line, and the searched nodes cannot be the same as virtual nodes, which would be considered as same-station transfers.

[0078] (3) Generate inter-station transfer arcs. For all virtual nodes of each station, a series of virtual nodes within the walking transfer distance are obtained through node search, such as the number of virtual nodes being m. When m≥1, the virtual node is connected to the m nodes obtained from the search respectively; if m=0, no inter-station transfer lines can be found for this station, and no inter-station transfer arc is introduced.

[0079] S12(2) Constructing a regional multi-mode rail transit network based on virtual node fusion strategy.

[0080] Main reference Figure 5 In one possible implementation, the specific process of constructing a regional multimodal rail transit network based on a node fusion strategy is as follows:

[0081] (4) Generate same-station transfer arcs between different modes. Connect the nodes of the urban rail transit physical sub-network with the extended nodes of different rail transit systems (train arrival and departure nodes of high-speed railway and intercity railway, stop nodes of urban (suburban) railway, and virtual co-location nodes of urban rail transit) to generate same-station transfer arcs.

[0082] (5) Node search. Given the maximum walking transfer radius that passengers can tolerate, and using this distance as the search radius, perform a transfer search around the station to find all transfer nodes between different modes. If the searched nodes are extended nodes between different modes, they must be the nodes that can be transferred between different modes.

[0083] (6) Generate inter-station transfer arcs between different modes. For all extended nodes of each station, the extended nodes between different modes within the walking transfer distance are obtained through node search, and the extended node is connected to the extended nodes between different modes obtained in the search respectively; if the number of extended nodes is 0, no extended nodes between different modes can be found at the station, and no inter-station transfer arc is introduced.

[0084] II. Generation of the "Regional + Urban Area" Passenger Flow OD Matrix:

[0085] Main reference Figure 3 In one possible implementation, the generation of the "regional + urban area" passenger flow OD matrix mainly includes the following steps:

[0086] S21. Divide the area into traffic zones of varying coarseness;

[0087] S22. Based on the system equilibrium prediction model and the gravity model, predict the passenger flow (e.g., total demand) of the region. In this example, the system equilibrium model is used for passenger flow prediction within the city, while the gravity model is used for passenger flow prediction within the traffic zones of the surrounding towns and between the traffic zones of the surrounding towns and the city.

[0088] S23. Based on the prediction results, generate a passenger flow OD matrix for "region + city".

[0089] The system equilibrium model is shown in the following formula:

[0090]

[0091] Where X ij T represents the travel volume distribution from traffic zone i to traffic zone j. i U represents the number of trips generated in traffic zone i. j f(t) represents the travel attraction of traffic zone i, n represents the number of traffic zones, and f(t) represents the number of travel zones i. ij ) represents the impedance function from traffic zone i to traffic zone j, t ij This is the impedance parameter.

[0092] The gravity model is shown in the following formula:

[0093] X ij =A i ·B j ·T i ·U j ·f(t ij (2)

[0094]

[0095]

[0096] Where X ij T represents the traffic distribution from traffic zone i to traffic zone j. i U represents the traffic generation in traffic zone i. j Let f(t) represent the traffic attraction of traffic zone j, n represent the number of traffic zones, and f(t) represent the traffic attraction of traffic zone j. ij ) represents the impedance function from traffic zone i to traffic zone j, t ij This is the impedance parameter.

[0097] In one possible implementation, S23 specifically includes:

[0098] Based on the prediction results, an OD matrix is ​​further generated for each prediction type. The final composite "regional + urban area" passenger flow OD matrix distribution is shown in Table 1 below.

[0099] Table 1 Passenger Flow Origin-Destination Distribution by Region + Urban Area

[0100]

[0101]

[0102] Among them, a ij b represents the traffic generation from traffic zone i to j in the main urban area A. ij This represents the traffic generation volume from traffic zone i to j in the outer town (cluster) B. AB This represents the OD distribution of passenger flow from the main urban area A to the surrounding town (cluster) B, and its OD matrix is ​​expressed as follows: W BA This represents the OD distribution of passenger flow from the peripheral town (cluster) B to the main urban area A, and its OD matrix is ​​expressed as follows:

[0103] III. Construction of a cross-modal travel route selection probability model:

[0104] In one possible implementation, determining the probability model for travel route selection mainly includes the following steps:

[0105] S31. Analyze the entire travel process of passengers and construct a cross-system intermodal "travel chain" for passengers;

[0106] S32. Based on the cross-system intermodal "travel chain", construct a cross-system travel route selection probability model (construct a cross-system travel route selection probability model based on the "travel chain").

[0107] In one possible implementation, S31 mainly includes:

[0108] Analyze the entire passenger travel process and define travel links to construct a passenger "travel chain" across rail transit systems. For example, the "travel chain" can include cross-system transfers between high-speed rail / intercity rail and suburban rail, cross-system transfers between high-speed rail / intercity rail and urban rail transit, and cross-system transfers between suburban rail and urban rail transit.

[0109] Based on the constructed "travel chain", the entire travel process of a passenger including one cross-system transfer includes: departure station — (purchase / collect ticket) — ticket inspection and security check — waiting — (ticket inspection) — boarding — arriving at the transfer station — getting off — ticket inspection and exit — transfer — (purchase / collect ticket) — security check and ticket inspection — waiting — boarding — arriving at the destination station — getting off — ticket inspection and exit.

[0110] In one possible implementation, S32 mainly includes:

[0111] The route selection problem for passenger travel is a decision-making problem. The probability of a passenger's route selection depends on the generalized cost of the effective route and the distribution of the random term. Assuming that the random error term of the route cost estimate follows an independent Gumbel distribution, the probability model for cross-modal travel route selection is as follows:

[0112]

[0113] in, Let n represent the probability that the k-th effective path between OD and rs is selected for the n-th type of passenger flow. The path selection probability is calculated using the relative cost difference. n = 1, 2, 3, 4, which represent long-distance trunk flow, inter-regional flow within the city, short-distance flow in the suburbs, and intra-city flow, respectively. This represents the total generalized cost of the k-th effective path for the nth type of passenger flow between OD and rs, which is composed of the five spatiotemporal service arcs of high-speed and intercity trains (represented by the train operation arc). Transfer arc Stop arc Upward arc and disembarkation arc ), and the four types of service arcs for urban (suburban) trains (namely, train operation arcs) Transfer arc Upward arc and disembarkation arc ) and two types of arc segments in urban rail transit (namely, section arcs) Transfer arc Linear structure.

[0114] IV. Forecast of Passenger Flow Distribution in Regional Multimodal Rail Transit Networks:

[0115] The regional multimodal rail transit network passenger flow distribution prediction method of the present invention is mainly based on the aforementioned constructed regional multimodal rail transit network, the generated "regional + urban area" passenger flow OD matrix, and the determined travel route selection probability model. Specifically, the regional multimodal rail transit network passenger flow distribution prediction method of the present invention improves the K-shortest path search algorithm based on the hierarchical characteristics of the regional multimodal rail transit network, constructs a set of feasible paths between any OD, incorporates the aforementioned cross-mode travel route selection probability model, and realizes passenger flow allocation of the regional multimodal rail transit network.

[0116] In one possible implementation, the method for predicting passenger flow distribution in a regional multimodal rail transit network mainly includes the following steps:

[0117] S41. An improved K-short path search algorithm is proposed based on the hierarchical characteristics of a regional multi-modal rail transit network. This algorithm can be decomposed into path searches for three rail transit sub-networks to generate alternative paths. The specific algorithm is as follows:

[0118] Let G i '=(N i A i Let N be the i-th level rail transit subnetwork. i With A i Let G' represent the node set and arc set of the i-th level rail transit subnetwork, respectively, and G' = G1', G2', G3' = N1, A1, N2, A2, (N3, A3)). Let i = 1, 2, and 3 represent the high-speed rail and intercity rail subnetwork, the urban (suburban) rail subnetwork, and the urban rail transit subnetwork, respectively. The K-shortest path from the starting node s to the ending node d of the i-th level rail transit subnetwork is... in This represents the length (impedance) of the Kth shortest path from the first-level rail transit subnetwork to the (i-1)th-level rail transit subnetwork. Let represent the length (impedance) of the Kth shortest path of node d in the i-th level rail transit subnet. Where d k (n m S' represents the length (impedance) of the Kth shortest path in the m-th level rail transit subnetwork; i D′ represents the connected node closest to the starting node S in the i-th level rail transit subnet. i It represents the connected node closest to the end node D in the i-th level rail transit subnet.

[0119] S41(1) Determine whether there is a high-speed railway and intercity railway subnetwork G1';

[0120] If G1' exists, search for the K-shortest path from the starting point S to D in G1' = (N1, A1). On the high-speed and intercity railway subnetwork, find the connected node S′1 closest to the starting node S and the connected node D′1 closest to the ending node D. Obtain the path using the K-shortest path search algorithm. in

[0121] S41(2) Determine whether there is a suburban (urban) railway subnetwork G2';

[0122] (a) If both G1' and G2' exist, then on the network G2' = (N2, A2), find the connected node S'2 that is closest to the start node S'1 and the connected node D'2 that is closest to the end node D'1. The path is obtained through K-shortest path search. and in

[0123] (b) If G1' does not exist but G2' exists, then search for the K-shortest path from the starting point S to D in G2' = (N2, A2). On the urban (suburban) railway subnetwork, find the connected node S′2 closest to the starting node S and the connected node D′2 closest to the ending node D. Obtain the path using the K-shortest path search algorithm.

[0124] S41(3) Determine whether there is an urban rail transit subnetwork G3′;

[0125] If both G2' and G3' exist, then in the network G3' = (N2, A2), find the connected node S'3 that is closest to the start node S'2, and find the connected node D'3 that is closest to the end node D'2. The K shortest path is then obtained through path search. and ,in ,

[0126] If G2' does not exist, but G1' and G3' both exist, then on G3' = (N2, A2), find the nearest connected node S'3 to the start node S'1, and find the nearest connected node D'3 to the end node D'1. The K shortest path is then obtained using a path search algorithm. and in

[0127] If neither G1' nor G2' exists, then search for the K-shortest path from the starting point S to D in G3' = (N2, A2). On the urban (suburban) railway network, find the nearest connected node S′3 to the starting node S and the nearest connected node D′3 to the ending node D. The path is then obtained using the K-shortest path search algorithm.

[0128] S41(4) merges the paths corresponding to each sub-network to generate alternative paths. The alternative paths are mainly divided into the following 7 cases:

[0129] (1) If G′1, G′2, and G′3 all exist, then the complete alternative path is:

[0130] (2) If G1' does not exist, but both G2' and G3' exist, then the complete alternative path is:

[0131] (3) If G2' does not exist, but both G1' and G3' exist, then the complete alternative path is:

[0132] (4) If G3' does not exist, but both G1' and G2' exist, then the complete alternative path is:

[0133] (5) If G1' exists and neither G2' nor G3' exists, then the complete alternative path is:

[0134] (6) If G2' exists and neither G1' nor G3' exists, then the complete alternative path is:

[0135] (7) If G3' exists and neither G1' nor G2' exists, then the complete alternative path is:

[0136] S42. Based on the principles of non-repetition of nodes and arcs, reasonable range of travel route selection, and limited number of transfers, determine the set of alternative routes between ODs, and filter out the set of feasible routes between ODs from the complete set of alternative routes. Specifically:

[0137] definition This represents the k-th valid path for the nth type of passenger flow between OD pairs rs. Let represent the generalized cost of the k-th effective path for the n-th type of passenger flow between OD pairs rs. This indicates that the l-th effective path for the nth type of passenger flow between OD pairs rs minimizes the generalized cost. Let represent the set of all nodes of the k-th valid path between OD pairs rs for the nth type of passenger flow. Let represent the set of all arc segments of the k-th valid path between OD pairs rs for the nth type of passenger flow, then have All are positive numbers.

[0138] The principle of non-repetition of nodes in S42(1) is:

[0139] The principle of non-repetition of arc segments in S42(2) is:

[0140] S42(3) The reasonable range of travel route selection can also be called the relative threshold and absolute threshold restriction principle. It mainly means that the passenger's travel route selection fluctuates within a certain range of the minimum path, which can reasonably determine the effective route range.

[0141]

[0142]

[0143] The principle for limiting the number of transfers in S42(4) is that the number of transfers should be less than the maximum number of transfers, i.e.

[0144] S43. Integrate passenger intermodal travel route selection probability models to achieve integrated passenger flow distribution prediction for regional multimodal rail transit networks. The specific steps are as follows:

[0145] S43.1. Considering different types of passenger flow, determine the passenger flow between different OD pairs rs. Assume the total passenger flow of the nth type of passenger flow between OD pairs rs is... A hierarchical K-shortest path search algorithm is used to generate candidate paths, and feasible paths between ODs are selected according to corresponding principles to obtain the corresponding set of effective paths. Determine the total generalized cost value of each path in the set of valid paths. And obtain the cost value of the path with the minimum total generalized cost value.

[0146] S43.2. Based on the cross-mode travel path probability selection model, calculate the probability that the kth effective path is selected for the nth type of passenger flow between OD pairs rs. Calculate the passenger flow on each path k based on the probability values ​​of each effective path selected by the nth type of passenger flow between OD pairs rs. in

[0147] S43.3, Based on the passenger flow of the nth type of passenger flow on the path k between OD pairs rs Calculate the passenger flow in the interval (a, b) between rs. If path k is contained in the interval (a, b), then Otherwise, it is 0, where

[0148] S43.4 Repeat steps S43.2 and S43.3 to calculate the path passenger flow and interval passenger flow for all OD pairs rs for all categories of passenger flow, and obtain the cross-sectional passenger flow between intervals (a, b). in

[0149] S43.5. Assign passenger flow to the nth type sequentially, where n = 1, 2, 3, 4, representing long-distance trunk line passenger flow, inter-district passenger flow, suburban short-distance passenger flow, and intra-city passenger flow, respectively. Once all types of passenger flow have been assigned, proceed to S43.6; otherwise, return to step one (S43.1) and continue with flow allocation.

[0150] S43.6. Once all types of passenger flow have been allocated, further calculate the interval flow. Specifically, sum the path flows of different types of passengers to obtain the total flow between intervals (a, b), and the passenger flow allocation ends.

[0151] It can be seen that in the regional multi-modal rail transit network passenger flow distribution prediction method of the present invention, the rail transport service sub-network that considers the differences between the train operation modes of different rail transit sub-networks, the complete "regional + urban area" passenger flow OD matrix, and the cross-mode travel path selection probability model based on "travel chain" can more scientifically support the analysis of passenger flow demand of regional rail transit networks.

[0152] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effects of the present invention, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously or in other orders, and some steps can be added, replaced or omitted.

[0153] It should be noted that although the method for predicting passenger flow distribution in a regional multimodal rail transit network, as described above, has been presented as an example, those skilled in the art will understand that the present invention is not limited thereto. In fact, users can flexibly adjust the relevant steps and parameters according to actual application scenarios and other factors.

[0154] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for predicting passenger flow distribution in a regional multimodal rail transit network, characterized in that, The prediction method includes the following steps: Based on the hierarchical characteristics of a pre-constructed regional multimodal rail transit network, the K-shortest path search algorithm is improved; specifically, it includes: make For the first Level rail transit sub-network, These represent the high-speed railway and intercity railway sub-network, the urban (suburban) railway sub-network, and the urban rail transit sub-network, respectively; starting from the first node. To the Level rail transit subnet ending node The K shortest path between them is , can be represented as ,in This indicates the first-level rail transit subnetwork to the second-level rail transit subnetwork. The length of the Kth shortest path in the Class I rail transit subnetwork. Indicates the first Level rail transit sub-network node The length of the Kth shortest path; ,in Indicates the first The length of the Kth shortest path in the Class II rail transit subnetwork; Indicates the first Level rail transit subnetwork from the starting node The nearest connected node, Indicates the first Level rail transit subnetwork from the end node The nearest connected node, Determine whether a high-speed railway and intercity railway sub-network exists. ,like The path exists and can be obtained using the K-shortest path search algorithm. ,in ; Determine whether a sub-network of urban railways exists. ;like and Both exist; the path is obtained through K-shortest path search. and ;like It does not exist, but The path exists and can be obtained using the K-shortest path search algorithm. , ; Determine if an urban rail transit subnetwork exists ;like and All exist; the K shortest paths are obtained through path search. and ;like Does not exist and All exist, and the K shortest paths are obtained through a path search algorithm. and ;like and If none of them exist, the path is obtained using the K-shortest path search algorithm. , ; Merge the paths of each sub-network; Based on the pre-generated "regional + urban area" passenger flow OD matrix and the improved K-shortest path search algorithm, a set of feasible paths between any OD is constructed. By integrating the set of feasible paths between any origin-destination (OD) points into a pre-constructed cross-modal travel path selection probability model, passenger flow allocation for regional multimodal rail transit networks can be achieved; specifically including: S810. Based on considering different types of passenger flow, determine different right Passenger flow between, assuming the first Customer flow right The total passenger flow between them is A hierarchical K-shortest path search algorithm is used to generate candidate paths, which are then selected based on relevant principles. Among the feasible paths, the corresponding set of effective paths is obtained. Determine the total generalized cost value for each path. And obtain the cost value of the path with minimum cost. ; S820, Calculate the first based on the cross-mode travel route probability selection model. Customer flow right Between The probability of a valid path being selected According to the first Customer flow right The probability value of each valid path being selected is calculated for each path. Customer traffic ,in ; S830, according to the Customer flow right Path between Customer traffic ,calculate Between Passenger traffic If the path Included in the interval In the middle, then Otherwise, it is 0, where ; S840, repeat S820 and S830, calculate all passenger flows for all categories. Given the correct route passenger flow and interval passenger flow, the interval passenger flow is obtained. Cross-sectional passenger flow ,in ; S850, sequentially for the first The system allocates passenger flow by type. Once all types of passenger flow have been allocated, it proceeds to S6; otherwise, it returns to S1 to continue allocating passenger flow. S860. Once all types of passenger flow have been allocated, further calculate the interval flow by summing the flow of different passenger paths to obtain the interval flow. The full flow of passengers between them has been allocated.

2. The method for predicting passenger flow distribution in a regional multi-modal rail transit network according to claim 1, characterized in that, The method for constructing the regional multimodal rail transit network includes: Based on the differences in train operation modes among different rail transit sub-networks, a transportation service sub-network corresponding to the rail transit sub-network is constructed. Based on the virtual node fusion method, the topology between different transportation service sub-networks is connected.

3. The method for predicting passenger flow distribution in a regional multi-modal rail transit network according to claim 2, characterized in that, The aforementioned "constructing a transportation service subnetwork corresponding to the rail transit subnetwork based on the differences in train operation modes of different rail transit subnetworks" includes: Based on train timetables, construct a train service sub-network for high-speed railways and intercity railways; Based on the train operation plan, construct a sub-network for urban railway train services; Based on physical stations and sections, construct a physical sub-network for urban rail transit.

4. The method for predicting passenger flow distribution in a regional multi-modal rail transit network according to claim 1, characterized in that, The method for generating the "regional + urban area" passenger flow OD matrix includes: The area is divided into traffic zones of varying degrees of detail. Based on the gravity model, passenger flow in the transportation zones of peripheral towns and between the transportation zones of peripheral towns and the city interior is predicted. Based on the system balance prediction model, the passenger flow within the city is predicted; Based on the prediction results, a passenger flow OD matrix of "region + city" is generated.

5. The method for predicting passenger flow distribution in a regional multi-modal rail transit network according to claim 1, characterized in that, The method for constructing the cross-mode travel route selection probability model includes: Analyze the entire travel process of passengers and construct a cross-system intermodal "travel chain" for passengers; Based on the cross-system intermodal "travel chain", a cross-system travel route selection probability model is constructed.

6. The method for predicting passenger flow distribution in a regional multi-modal rail transit network according to claim 1, characterized in that, The phrase "constructing a set of feasible paths between any ODs based on a pre-generated "regional + urban area" passenger flow OD matrix and an improved K-shortest path search algorithm" includes: Based on the principles of non-repetition of nodes and arcs, reasonable range of travel routes, and limited number of transfers, the set of alternative routes between ODs is determined, and the set of feasible routes between ODs is selected from the routes.

7. A computer-readable storage medium comprising a memory adapted to store a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the regional multimodal rail transit network passenger flow distribution prediction method according to any one of claims 1 to 6.

8. A computer device, the device comprising a memory and a processor, the memory being adapted to store a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to perform the regional multimodal rail transit network passenger flow distribution prediction method according to any one of claims 1 to 6.

Images