Urban rail transit network multi-station cooperative flow limiting model

Abstract

The invention discloses a multi-station cooperative flow limiting model of an urban rail transit network. The model comprises the following steps: (a) collecting card swiping data of station entering and station exiting of an urban rail transit AFC system, and train operation data, passenger flow survey data and road network information data of an ATS system; (b) constructing an urban rail transit passenger travel path selection model in a crowded state, and evaluating the passenger flow volume of each station of the urban rail transit network in a unit control time period; (c) constructing a multi-station cooperative flow limiting model, taking maximization of the number of passengers getting on the bus and minimization of the passenger delay time as control optimization objectives, giving out formulating steps of a multi-station cooperative flow limiting scheme of the rail transit network, and depicting an action relationship between a control strategy and a system state; and (d) providing a collaborative flow limiting optimization scheme of a plurality of stations on the whole network of the urban rail transit to an external system through the data interface.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban rail transit operation management, and in particular relates to a multi-station collaborative flow limiting model for an urban rail transit network. Background Art

[0002] In recent years, urban rail transit has achieved rapid development in my country with its large capacity, fast, punctual and safe features. The rail transit in major cities such as Beijing, Shanghai, Guangzhou, Shenzhen and Chengdu has entered a new stage of network operation and management. With the improvement of the urban rail transit network, passengers can travel more conveniently. However, while attracting a large number of passengers, the congestion of the urban rail transit system has become increasingly serious, which not only seriously affects the efficiency of network operation, but also greatly reduces the safety of operation.

[0003] Common methods used by urban rail transit systems to ease passenger congestion include limiting the number of passengers entering the station and limiting the speed of passengers entering the station by setting up fences and other measures outside the station, and controlling the number of passengers transferring and the speed of passengers transferring by setting up iron horses and other measures inside the station. However, most of these passenger flow control measures are targeted at a single station and are mainly formulated based on historical experience, lacking a more scientific and quantitative basis.

[0004] Existing studies rarely consider the dynamic and time-varying characteristics of passenger flow in urban rail transit systems, resulting in poor results of flow control schemes. When formulating flow control schemes, existing studies mostly focus on passenger congestion at a single station or a single line of urban rail transit. However, the networked operation of urban rail transit has strengthened the interconnection between the originally independently operated lines. The flow control scheme formulated according to a certain station or a certain line can alleviate the passenger flow pressure of a single station or a single line, and achieve the relief of local passenger congestion. However, due to the lack of consideration of the coordination between stations and road networks, it cannot ensure that the entire urban rail transit network operation efficiency is optimal. Summary of the invention

[0005] Based on this, the purpose of the present invention is to provide a multi-station collaborative flow limiting model for urban rail transit networks, make full use of the various data resources generated in the process of urban rail transit network operation, and systematically consider the matching relationship between the supply capacity of the urban rail transit system and travel demand on the basis of analyzing the relationship between passenger travel choices and the spatiotemporal distribution of passenger flow. A multi-station collaborative flow limiting model for urban rail transit networks based on passenger path selection is proposed. It will provide a theoretical basis for the formulation of multi-station collaborative passenger flow control plans, guide the spatiotemporal distribution of passenger flow through demand adjustment measures between stations, and achieve the purpose of optimizing the supply and demand relationship and alleviating passenger congestion.

[0006] The above technical objectives of the present invention are achieved through the following technical solutions:

[0007] A multi-station coordinated flow limiting model for urban rail transit network, including:

[0008] (a) Collecting the card swiping data of the urban rail transit AFC system, the train operation data of the ATS system, the passenger flow survey data, and the road network information data;

[0009] (b) Construct a travel route selection model for urban rail transit passengers under crowded conditions and evaluate the passenger flow of each station in the urban rail transit network within a unit control time period;

[0010] (c) A multi-station coordinated flow control model is constructed, with the control optimization objectives of maximizing the total number of passengers and minimizing the passenger delay time. The steps for formulating a multi-station coordinated flow control plan for the rail transit network are given, and the relationship between the control strategy and the system state is characterized.

[0011] (d) Provide external systems with coordinated flow-limiting optimization solutions for multiple stations on the entire urban rail transit network through data interfaces.

[0012] Optionally, the constructing of a travel path selection model for urban rail transit passengers under a congested state includes:

[0013] Based on massive rail transit data, the relationship between passengers' travel path selection and network topology and passenger flow status is explored using big data mining methods. A travel path selection model for urban rail transit passengers under crowded conditions is constructed, and the parameters in the model are calibrated using Bayesian and MCMC methods.

[0014] Optionally, based on massive rail transit data, the method of big data mining is used to explore the relationship between passenger travel path selection and network topology and passenger flow status, to construct a passenger travel path selection model for urban rail transit under crowded conditions, and to calibrate the parameters in the model using Bayesian and MCMC methods, specifically including:

[0015] (b1) Generate a valid path set for each passenger: For a passenger n of any OD pair (o, d), generate a valid path set K by combining AFC data and train operation information;

[0016] (b2) Construct a passenger route selection model under congestion: For any OD pair (o, d), the utility maximization theory is used to characterize the passenger's route selection behavior, and the generalized cost on the rth effective route is determined by travel time, transfer time and train congestion;

[0017] (b3) Based on the construction of the passenger route selection model, the passenger route selection behavior problem is transformed into a parameter estimation problem. The parameters in the route selection model are calibrated using MCMC and Bayesian methods, forming an estimation method for urban rail transit passenger route selection based on AFC data.

[0018] Optionally, step (c) specifically includes:

[0019] (c1) Construct a multi-station coordinated flow control model for urban rail transit, taking into account the interests of both passengers and operators, establish an objective function with the goal of maximizing the total number of passengers boarding and minimizing the total delay time, use passenger path selection behavior to characterize and deduce the passenger transfer process in the urban rail transit system, and constrain the passenger path selection behavior, passenger flow demand, passenger boarding and alighting, platform capacity, passenger entry, etc., to study the optimal passenger flow of each station in the urban rail transit network within the unit control time period;

[0020] (c2) A method for formulating a multi-station collaborative passenger flow control strategy is given to obtain the optimal flow limiting plan. Based on the multi-station collaborative flow limiting optimization model established in (c1), the optimal entry volume of each station during the flow limiting period is obtained. Based on the model solution results, a flow limiting plan is generated and specific flow limiting measures are implemented.

[0021] Optionally, the construction of the urban rail transit passenger travel path selection model under congestion is as follows:

[0022] Incorporating the passenger congestion coefficient into the path utility function and taking the minimum path impedance value as the standard, the improved Logit model is as follows:

[0023]

[0024] The steps of using the improved Logit model to distribute passenger flow are as follows: (1) Search for valid paths between the nth OD pairs in the rail transit network and obtain a valid path set K; (2) Solve the searched valid paths to obtain the path impedance c and find the minimum path impedance c among the paths in the path set. min ; (3) Solve the possible selected P of each path in the path set k ; (4) Allocate all passenger flow to each valid path according to this probability, then the passenger flow f allocated to the kth path in the rail transit network is k It can be expressed as:

[0025] f k =qP k (2)

[0026] q——the total passenger flow between the nth OD pair in the rail transit network;

[0027] Finally, calculate the passenger flow of each section in the rail transit network:

[0028]

[0029] ——The passenger flow ratio of the ath section of the kth path between the nth OD pair; ——The passenger flow allocated to the kth path between the nth OD pair.

[0030] Optionally, the multi-station collaborative flow limiting model is constructed as follows:

[0031] From the perspective of passengers, the goal is to minimize the delay time of passengers; from the perspective of operators, the goal is to transport passengers to the maximum extent with limited facilities and equipment. Taking into account the interests of both parties, the objective function is established as follows:

[0032] (1) Minimum overall delay time

[0033]

[0034] Δt——the length of unit control time; ——the optimal number of passengers entering the station i within the unit control time t; q i,t ——the entry demand at station i within the unit control time t; ——The passenger flow gathered at the station during the unit control period t because they failed to get on the train before; ——The passenger flow newly arriving at the station within the unit control period t; T——The set of flow-limited periods; I——The set of stations;

[0035] (2) The largest overall passenger volume

[0036] Z2=max∑ t∈T ∑ l∈L Q l,t (5)

[0037] Q l,t ——The cross-sectional passenger flow of interval l in the tth flow-limiting period; T——the set of flow-limiting periods; L——the set of intervals;

[0038] The above two objective functions are correlated and relative, and it is impossible to find the optimal solution to achieve the optimal results for both objectives at the same time. First, the two objective functions are processed and converted into single objective functions. According to the importance of the two objectives, the weight coefficients of the two objectives are set to 0.6 and 0.4 respectively, and the satisfaction function is obtained:

[0039]

[0040] P(Z)=max (0.6 P(Z1)+0.4P(Z2)) (8)

[0041] The particle swarm is used to solve the problem. The flow-restricted stations, flow-restricted time periods, and flow control rates are obtained through calculation of the model. The flow control rate is the ratio of the passenger flow restricted from entering a station to the actual passenger flow per unit time. The formula is:

[0042]

[0043] Due to the adoption of the above-mentioned technical scheme, the beneficial effects of the present invention include: being able to generate a complete multi-station coordinated flow limiting scheme without changing the train operation plan, providing a complete flow control workflow for the rail transit system, thereby improving the capacity utilization of interval trains and the balance of boarding ratios, and reducing the loss of line capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is an implementation flow chart of the present invention.

[0045] Figure 2 It is the algorithm flow chart of the present invention.

[0046] Figure 3 It is a system structure diagram of an embodiment of the present invention.

[0047] Figure 4 1 is a calculation example road network diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0049] With the improvement of the scientific and technological level in the field of transportation and the rapid development of related technologies, a large amount of data resources that can be used for passenger flow analysis have been accumulated in the process of urban rail transit operation and management: the AFC system (Auto Fare Collection, automatic ticket vending and checking system) records the entry and exit information of passengers, including the time of card swiping, which can truly reflect the travel behavior of passengers between OD and OD. It is an important basis for inferring the passenger flow in and out of the station and the cross-sectional passenger flow, and is the basis for formulating flow control plans. The automatic train monitoring (ATS) system saves the operation information of all trains on the network, as well as the path information calculated according to the network structure, the travel time and train schedule information obtained according to the on-site investigation, etc., providing important information on the flow of passenger flow between stations and providing the prerequisite for formulating multi-station coordinated flow control plans.

[0050] The present invention makes full use of various data resources generated during the network operation of urban rail transit. On the basis of analyzing the relationship between passenger travel choices and the spatiotemporal distribution of passenger flow, the system considers the matching relationship between the supply capacity of the urban rail transit system and travel demand, and proposes a multi-station collaborative flow limiting model for the urban rail transit network based on passenger path selection. The invention results will provide a theoretical basis for the formulation of multi-station collaborative passenger flow control plans, guide the spatiotemporal distribution of passenger flow through demand adjustment measures between stations, and achieve the purpose of optimizing the supply and demand relationship and alleviating passenger congestion.

[0051] The present invention extracts urban rail transit passenger OD travel data and rail transit network basic data based on AFC data. Collecting and storing massive AFC data and constructing and maintaining the urban rail transit network are two important prerequisites of the present invention.

[0052] like Figure 3 As shown in the figure, the system consists of a road network data management module, an AFC data storage module, an operation plan management module, a network passenger flow distribution module, and a multi-station collaborative flow limiting module. These software modules are independent and can be installed on one machine or multiple machines according to actual conditions. In addition, it can also be integrated into the urban rail transit operation management platform as a subsystem of the urban rail transit operation management auxiliary decision support system.

[0053] Now combined Figure 3 Briefly explain the functions of the above modules and their data relationship with the multi-station coordinated flow control module. Figure 1 The implementation process and results of the method of the present invention are described.

[0054] 1. Road network data management module

[0055] This module provides management, maintenance and update of various basic data required by the multi-station collaborative current limiting module by establishing a basic database of the urban rail transit network. It can also realize the digitization of the urban rail transit network and achieve unified and standardized management of various basic information of the network. This module uses graphical and visual methods to uniformly manage various basic data information. The basic data includes: basic line information, basic station information, train operation plan, station interval facilities and equipment and other on-site infrastructure configuration information.

[0056] 2. AFC data storage module

[0057] This module has a data interface with the Urban Rail Transit ACC (Automated Clearing Centre, referred to as ACC) to obtain the AFC data provided by ACC and clean, organize and save the original AFC data. The data structure of AFC data is:

[0058] Table 1 AFC data structure for entering the station by swiping the card

[0059]

[0060]

[0061] Table 2 Outbound card swiping AFC data structure

[0062]

[0063] 3. Operation plan management module

[0064] This module has a data interface with the urban rail transit operation planning system, which is used for the urban rail transit operation planning data required by the network passenger flow distribution and multi-station coordinated flow limiting module, including: full-day driving plan data of the road network, connection plan data of the first / last trains between lines, etc.

[0065] 4. Network passenger flow distribution module

[0066] This module provides a method for generating effective path sets and allocating passenger flow based on massive AFC data in the actual context of the automatic ticket collection system (AFC) of the Rail Transit Clearing and Management Center (ACC) in various cities in my country. Combining AFC data and train operation data, based on the K-shortening path algorithm, absolute thresholds and relative thresholds are set to judge the rationality of K paths, thereby generating an effective path set, and calibrating the model parameters using Bayesian and MCMC methods. The passenger flow allocation results generated by this module will be fed back to the multi-station collaborative flow limiting module.

[0067] 5. Multi-station collaborative flow control module

[0068] The multi-station collaborative flow limiting module is the core module of the system and also the main embodiment of the achievements of the present invention. This module is based on the basic data of the urban rail transit network provided by the network data management module, the train operation plan data provided by the operation plan management module, and the network passenger flow distribution data provided by the network passenger flow distribution module. It uses the multi-station collaborative flow limiting model and particle swarm algorithm to provide a time-divided flow limiting solution.

[0069] The implementation process can be specifically explained through the following example:

[0070] (a) Example application data

[0071] The basic road network of the example is taken as an example of Nanchang urban rail transit in 2024 (such as Figure 4 The network includes 4 rail transit lines with a total length of 128.4 kilometers, 94 stations and 9 transfer stations. The AFC data and network information data used in the example application are shown in the following table.

[0072] Table 3 Part of the interval information data

[0073]

[0074] Table 4 Information data of some transfer stations

[0075]

[0076] Table 5 Passenger flow in some time periods

[0077]

[0078] (b) Passenger flow distribution

[0079] Incorporating the passenger congestion coefficient into the path utility function and taking the minimum path impedance value as the standard, the improved Logit model is as follows:

[0080]

[0081] The steps of using the improved Logit model to distribute passenger flow are as follows: (1) Search for valid paths between the nth OD pairs in the rail transit network and obtain the valid path set K. (2) For the searched valid paths, solve the path impedance c and find the minimum path impedance c among the paths in the path set. min (3) Solve the possible selected P of each path in the path set k (4) Allocate all passenger flow to each valid path according to this probability, then the passenger flow f allocated to the kth path in the rail transit network is k It can be expressed as:

[0082] f k =qP k (2)

[0083] q is the total passenger flow between the nth OD pair in the rail transit network.

[0084] Finally, calculate the passenger flow of each section in the rail transit network:

[0085]

[0086] ——The passenger flow ratio of the ath section of the kth path between the nth OD pair; ——The passenger flow allocated to the kth path between the nth OD pair.

[0087] The passenger flow distribution results are shown in Table 6:

[0088] Table 6 Partial passenger flow distribution results

[0089]

[0090] (c) Coordinated flow control at multiple stations

[0091] A multi-station collaborative flow control model is established based on AFC data, road network information data, and passenger flow distribution result data. From the perspective of passengers, the goal is to minimize the delay time of passengers; from the perspective of operators, the goal is to transport passengers to the maximum extent with limited facilities and equipment. Taking into account the interests of both, the objective function is established as follows:

[0092] (1) Minimum overall delay time

[0093]

[0094] Δt——the length of unit control time; ——the optimal number of passengers entering the station i within the unit control time t; q i,t ——the entry demand at station i within the unit control time t; ——The passenger flow gathered at the station during the unit control period t because they failed to get on the train before; ——The passenger flow newly arriving at the station within the unit control period t; T——The collection of flow-limited periods; I——The collection of stations.

[0095] (2) The largest overall passenger volume

[0096] Z2=max∑ t∈T ∑ l∈L Q l,t (5)

[0097] Q l,t ——The cross-sectional passenger flow of interval l during the tth flow-limited period; T——the set of flow-limited periods; L——the set of intervals.

[0098] The station collaborative control model is a multi-objective linear programming model. There is correlation and relativity between the two objectives, and it is impossible to find the optimal solution to achieve both objectives at the same time. First, the multi-objective function is processed and converted into a single-objective function. According to the importance of the two objectives, the weight coefficients of the two objectives are set to 0.6 and 0.4 respectively, and the satisfaction function is obtained.

[0099]

[0100] P(Z)=max (0.6 P(Z1)+0.4P(Z2)) (8)

[0101] At the same time, the present invention takes into account the actual situation and the conditions of equipment and facilities in the station, combines the road network information, the entry and exit information mined and calculated from the AFC data, the section passenger flow information obtained by network passenger flow distribution, etc., and establishes constraints from several aspects such as passenger flow demand, flow limiting intensity, section transportation capacity, station throughput capacity, and platform accommodation capacity.

[0102] The present invention uses particle swarm to solve the problem, and the algorithm flow is as follows: Figure 2 As shown. By calculating the model, we can obtain the flow-restricted stations, flow-restricted time periods and flow control rate. The flow control rate is the ratio of the passenger flow restricted from entering a station to the actual passenger flow per unit time, and its formula is:

[0103]

[0104] Table 7 Results of multi-station time-limited current

[0105]

[0106] The above-mentioned related explanations and descriptions of the embodiments are for the convenience of those skilled in the art to understand and apply the present invention. It is obvious that those familiar with the art can easily make various modifications to these contents and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above-mentioned related explanations and descriptions of the embodiments. Improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the protection scope of the present invention.

Claims

1. A multi-station coordinated flow limiting model for urban rail transit network, characterized in that: include: (a) Collect the entry and exit card swiping data of the urban rail transit AFC system, the train operation data of the ATS system, passenger flow survey data, and road network information data; (b) Construct a travel route selection model for urban rail transit passengers under crowded conditions and evaluate the passenger flow of each station in the urban rail transit network within a unit control time period; (c) A multi-station coordinated flow control model is constructed, with the control optimization objectives of maximizing the total number of passengers and minimizing the passenger delay time. The steps for formulating a multi-station coordinated flow control plan for the rail transit network are given, and the relationship between the control strategy and the system state is characterized. (d) Provide external systems with coordinated flow-limiting optimization solutions for multiple stations on the entire urban rail transit network through data interfaces.

2. The multi-station coordinated flow limiting model for urban rail transit network according to claim 1 is characterized in that: The construction of the urban rail transit passenger travel path selection model under congestion includes: Based on massive rail transit data, the relationship between passengers' travel path selection and network topology and passenger flow status is explored using big data mining methods. A travel path selection model for urban rail transit passengers under crowded conditions is constructed, and the parameters in the model are calibrated using Bayesian and MCMC methods.

3. The multi-station coordinated flow limiting model for urban rail transit network according to claim 2 is characterized in that: Based on massive rail transit data, the method of big data mining is used to explore the relationship between passenger travel path selection and network topology and passenger flow status, and a passenger travel path selection model for urban rail transit under crowded conditions is constructed. The parameters in the model are calibrated using Bayesian and MCMC methods, including: (b1) Generate a valid path set for each passenger: For a passenger n of any OD pair (o, d), generate a valid path set K by combining AFC data and train operation information; (b2) Construct a passenger route selection model under congestion: For any OD pair (o, d), the utility maximization theory is used to characterize the passenger's route selection behavior, and the generalized cost on the rth effective route is determined by travel time, transfer time and train congestion; (b3) Based on the construction of the passenger route selection model, the passenger route selection behavior problem is transformed into a parameter estimation problem. The parameters in the route selection model are calibrated using MCMC and Bayesian methods, forming an estimation method for urban rail transit passenger route selection based on AFC data.

4. The multi-station coordinated flow limiting model for urban rail transit network according to claim 1 is characterized in that: Step (c) specifically comprises: (c1) Construct a multi-station coordinated flow control model for urban rail transit, taking into account the interests of both passengers and operators, establish an objective function with the goal of maximizing the total number of passengers boarding and minimizing the total delay time, use passenger path selection behavior to characterize and deduce the passenger transfer process in the urban rail transit system, and constrain the passenger path selection behavior, passenger flow demand, passenger boarding and alighting, platform capacity, passenger entry, etc., to study the optimal passenger flow of each station in the urban rail transit network within the unit control time period; (c2) A method for formulating a multi-station collaborative passenger flow control strategy is given to obtain the optimal flow limiting plan. Based on the multi-station collaborative flow limiting optimization model established in (c1), the optimal entry volume of each station during the flow limiting period is obtained. Based on the model solution results, a flow limiting plan is generated and specific flow limiting measures are implemented.

5. The multi-station coordinated flow limiting model for urban rail transit network according to claim 1 is characterized in that: The model for selecting the travel path of urban rail transit passengers under congestion is as follows: Incorporating the passenger congestion coefficient into the path utility function and taking the minimum path impedance value as the standard, the improved Logit model is as follows: The steps of using the improved Logit model to distribute passenger flow are as follows: (1) Search for valid paths between the nth OD pairs in the rail transit network and obtain a valid path set K; (2) Solve the searched valid paths to obtain the path impedance c and find the minimum path impedance c among the paths in the path set. min ; (3) Solve the possible selected P of each path in the path set k ; (4) Allocate all passenger flow to each valid path according to this probability, then the passenger flow f allocated to the kth path in the rail transit network is k It can be expressed as: f k =qP k (2) q——the total passenger flow between the nth OD pair in the rail transit network; Finally, calculate the passenger flow of each section in the rail transit network: ——The passenger flow ratio of the ath section of the kth path between the nth OD pair; ——The passenger flow allocated to the kth path between the nth OD pair.

6. The multi-station coordinated flow limiting model for urban rail transit network according to claim 1 is characterized in that: The multi-station collaborative flow limiting model is constructed as follows: From the perspective of passengers, the goal is to minimize the delay time of passengers; from the perspective of operators, the goal is to transport passengers to the maximum extent with limited facilities and equipment. Taking into account the interests of both parties, the objective function is established as follows: (1) Minimum overall delay time Δt——the length of unit control time; ——the optimal number of passengers entering the station i within the unit control time t; q i,t ——the entry demand at station i within the unit control time t; ——The passenger flow gathered at the station during the unit control period t because they failed to get on the train before; ——The passenger flow newly arriving at the station within the unit control period t; T——The set of flow-limited periods; I——The set of stations; (2) The largest overall passenger volume Z2=max∑ t∈T ∑ l∈L Q l,t (5) Q l,t ——The cross-sectional passenger flow of interval l in the tth flow-limiting period; T——the set of flow-limiting periods; L——the set of intervals; The above two objective functions are correlated and relative, and it is impossible to find the optimal solution to achieve the optimal results for both objectives at the same time. First, the two objective functions are processed and converted into single objective functions. According to the importance of the two objectives, the weight coefficients of the two objectives are set to 0.6 and 0.4 respectively, and the satisfaction function is obtained: P(Z)=max (0.6 P(Z1)+0.4P(Z2)) (8) The particle swarm is used to solve the problem. The flow-restricted stations, flow-restricted time periods, and flow control rates are obtained through calculation of the model. The flow control rate is the ratio of the passenger flow restricted from entering a station to the actual passenger flow per unit time. The formula is:

Images