Passenger flow refined analysis method based on train operation network

Abstract

The invention discloses a passenger flow refined analysis method based on a train operation network. The method comprises the following steps: fusing an urban rail transit physical network with train operation diagram data; constructing a train operation network, and constructing a passenger flow distribution simulation model based on train arrival and departure as driving events by taking a traditional multi-path sorting proportion as a probability of selecting paths by passengers; the number of passengers getting on and off a station of each train number in an urban rail transit network and the number of passengers in each interval are dynamically deduced through a simulation process, and the analysis of the train full load condition and the analysis of the transfer waiting time are performed based on a simulation result, so that a basis is provided for the compiling and adjustment of a network train working diagram and the connection optimization of trains between different lines.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban rail transit passenger flow analysis, and in particular relates to a refined passenger flow analysis method based on a train operation network. Background Art

[0002] Analyzing the spatiotemporal distribution characteristics of passenger flow in the urban rail transit network and evaluating the adaptability of the train operation diagram to the spatiotemporal distribution of passenger flow are the basis for adjusting the network train operation diagram and optimizing the connection between trains on different lines. The commonly used method is to use a multi-path sorting model to dynamically allocate passenger flow, and then statistically analyze and evaluate the adaptability of transport capacity and volume by calculating relevant passenger flow indicators. However, the traditional multi-path sorting model only distributes passenger flow to the physical spatiotemporal network, without associating it with specific train numbers, and cannot accurately evaluate the train load and the connection effect between trains on different lines. Summary of the invention

[0003] The purpose of the present invention is to provide a passenger flow refinement analysis method based on a train operation network, combine the urban rail transit physical network and train operation diagram data, construct an urban rail transit train operation network, and finely distribute the short-term and short-term historical data collected by the automatic ticket vending and checking system (Auto Fare Collection, referred to as AFC) to the train operation network, and calculate the number of passengers carried by each train in each section in the network, and then finely analyze the train load rate and the waiting time index of the passengers, so as to guide the preparation and adjustment of the network operation diagram, as well as the connection optimization of trains between different lines, and improve the refined and intelligent operation level of urban rail transit.

[0004] The above technical objectives of the present invention are achieved through the following technical solutions:

[0005] A refined passenger flow analysis method based on a train operation network, comprising:

[0006] Integrate the urban rail transit physical network with train operation diagram data;

[0007] Construct a train operation network, use the traditional multi-path clearing ratio as the probability of passengers choosing a path, and build a passenger flow allocation simulation model based on train arrival and departure as driving events;

[0008] Through the simulation process, the number of passengers getting on and off at stations and the number of passengers in each section of each train in the urban rail transit network are dynamically deduced. Based on the simulation results, the train load situation and transfer waiting time are analyzed, providing a basis for the preparation and adjustment of the network train operation diagram and the optimization of train connections between different lines.

[0009] Optionally, the train operation network is constructed, and the traditional multi-path clearing ratio is used as the probability of the passenger choosing a path, and a passenger flow allocation simulation model based on train arrival and departure as driving events is constructed, including:

[0010] According to the historical entry and exit ticket information, for all passenger IDs, match the passenger's travel OD information to obtain the passenger's historical travel OD information table;

[0011] The multi-path clearing ratio of the passenger's historical travel OD information table is used as the probability of the passenger choosing a path, and then each matched travel record is extracted into the time-varying passenger flow OD data;

[0012] A passenger flow distribution simulation model based on train arrival and departure as driving events is constructed. The simulation model takes time-varying OD passenger flow data, train operation diagram data and passengers' travel route selection data as input, and processes passenger flows of different OD and different time granularities in chronological order. The simulation obtains the number of passengers entering and leaving the station, the number of transfers and the number of waiting passengers at different stations at different times, as well as the number of passengers getting on and off at different stations for different trains.

[0013] Optionally, the historical entry and exit ticket information is pre-processed to remove passenger ID information with only entry or exit data.

[0014] Optionally, the time-varying OD passenger flow can be divided into three categories for processing according to the number of transfers of the path selected by the time-varying OD passenger flow: processing of OD passenger flow without transfer, processing of OD passenger flow with one transfer and processing of OD passenger flow with multiple transfers.

[0015] Optionally, the processing of OD passenger flow without transfer: Assume that among the passenger flow from s1 to s2 generated in the time granularity t, there are p people who choose the path from s1 to s2 without transfer, the average entry travel time at s1 station is t1, and the average exit travel time at s2 station is t2;

[0016] ⑦Record the corresponding entry volume

[0017]

[0018] ⑧Calculate the time when you arrive at the platform

[0019]

[0020] ⑨ Infer the train number, waiting time and boarding and alighting time of this group of passengers based on the operation diagram

[0021] The passenger takes the next train no earlier than the passenger's arrival time at the platform t 到 The number of the train that left, the number of the train that took is k 乘 , the time from the drawing to the departure is

[0022] Therefore, the boarding time for passengers is:

[0023]

[0024] Therefore, the passengers' alighting time is:

[0025]

[0026] ⑩Record the number of people waiting for the bus at the corresponding time granularity

[0027]

[0028] Record the corresponding number of people getting on and off the bus

[0029]

[0030] Infer passenger exit times and record corresponding exit volumes

[0031]

[0032] Optionally, the processing of OD passenger flow with one transfer: Assume that the passenger flow from s3 to s5 generated in the time granularity t has p people choosing the path that needs to transfer from s4, the average entry time at s3 station is t3, the average transfer time at s4 is t4, and the average exit time at s5 is t5;

[0033] ⑨Record the corresponding entry volume

[0034]

[0035] ⑩Calculate the time when the passenger arrives at the platform (granularity)

[0036]

[0037] According to the operation diagram, the train number, waiting time, boarding time and getting off time of the passengers at the current station s3 are inferred; the passenger takes the last train no earlier than the passenger's arrival time t 到 The number of the train that left, the number of the train that took is k 乘 , and its scheduled departure time is

[0038] Therefore, the boarding time for passengers is:

[0039]

[0040] The time for passengers to get off at the transfer station is

[0041]

[0042] Record the number of people waiting for the bus at the corresponding time granularity

[0043]

[0044] Record the corresponding number of people getting on and off the bus

[0045]

[0046] Record the corresponding number of transfers

[0047]

[0048] Estimate the time to arrive at the target transfer station

[0049]

[0050] According to the current OD passenger flow, consider the non-transfer passenger flow from s4 to s5, skip ①, and let t in ② 到 =t 到换 ,

[0051] Start from ③ where there is no transfer passenger flow;

[0052] Optionally, the processing of OD passenger flow with multiple transfers: Assume that the passenger flow from s3 to s9 generated in the time granularity t has p people who choose a path that requires transfer at multiple stations s4, s5...s8 once each, the average entry time at s3 station is t3, the average transfer time at s4, s5...s8 is t4, t5...t8 respectively, and the average exit time at s9 is t9;

[0053] ①According to steps ①-⑤ of a transfer, process the process of passengers entering the station at S3 and getting off at S4;

[0054] ② From transfer station to transfer station, let t 到 =t 到换 , and then proceed as ③-⑦ in the OD passenger flow processing for one transfer;

[0055] ③ According to the current OD passenger flow, consider the passenger flow without transfer from the last transfer station s8 to s9, skip the first step, and start processing from step 3 of the passenger flow without transfer in the second step.

[0056] The beneficial effects of the present invention include: the present invention can, based on historical or predicted passenger flows, more finely analyze the number of passengers on all lines of trains in the rail transit network, the number of waiting people at stations, the full load of trains, the waiting time of passengers, the waiting time for transfers and other fine indicators, better evaluate the adaptability of the train operation diagram of the entire rail transit network to passenger flow demand, and provide important support for the optimization and adjustment of the train operation diagram of the urban rail transit network. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The invention provides a refined passenger flow analysis method based on a train operation network.

[0058] Figure 2 This is a Nanchang rail transit network diagram according to an embodiment of the present invention.

[0059] Figure 3 This is a passenger flow simulation flow chart of an embodiment of the present invention.

[0060] Figure 4 This is a diagram of the adaptability analysis of the transport capacity and volume of Nanchang Metro Line 1 according to an embodiment of the present invention.

[0061] Figure 5 This is a visualization result of the passengers on trains and passengers waiting on platforms during the morning rush hour on Nanchang Metro Line 1 according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0063] The present invention provides a passenger flow refinement analysis method based on a train operation network, comprising:

[0064] Integrate the urban rail transit physical network with train operation diagram data;

[0065] Construct a train operation network, use the traditional multi-path clearing ratio as the probability of passengers choosing a path, and build a passenger flow allocation simulation model based on train arrival and departure as driving events;

[0066] Through the simulation process, the number of passengers getting on and off at stations and the number of passengers in each section of each train in the urban rail transit network are dynamically deduced. Based on the simulation results, the train load situation and transfer waiting time are analyzed, providing a basis for the preparation and adjustment of the network train operation diagram and the optimization of train connections between different lines.

[0067] Optionally, the train operation network is constructed, and the traditional multi-path clearing ratio is used as the probability of the passenger choosing a path, and a passenger flow allocation simulation model based on train arrival and departure as driving events is constructed, including:

[0068] According to the historical entry and exit ticket information, for all passenger IDs, match the passenger's travel OD information to obtain the passenger's historical travel OD information table;

[0069] The multi-path clearing ratio of the passenger's historical travel OD information table is used as the probability of the passenger choosing a path, and then each matched travel record is extracted into the time-varying passenger flow OD data;

[0070] Construct a passenger flow distribution simulation model based on train arrival and departure as driving events. The simulation model takes time-varying OD passenger flow data, train operation diagram data and passenger travel path selection data as input, processes passenger flows of different OD and different time granularities in chronological order, and obtains the number of passengers entering and leaving the station, the number of transfers and the number of waiting passengers at different stations at different times, and the number of passengers getting on and off at different stations for different trains. The simulation model can be a neural network model.

[0071] 1. Construction of basic data of urban rail transit

[0072] 1) Construction of urban rail transit topology network

[0073] a) Construction of connectivity between stations

[0074] Assuming there are N stations, the connectivity of the stations can be represented by an adjacency matrix a(i, j) with N rows and N columns.

[0075]

[0076] b) Length of line (section) between stations

[0077] The length of the interval between station i and station j is represented by a matrix d(i,j) with N rows and N columns. When i=j, d(i,j)=0; when i≠j and a(i,j)=1, the value of d(i,j) is the actual length of the interval; when i≠j and a(i,j)=0, the value of d(i,j) is ∞.

[0078] c) Station attributes

[0079] Station attributes include: station name, station number, line, transfer station, front station, back station, X coordinate and Y coordinate. The front and back stations refer to the front and back of the line in the upward direction.

[0080] Table 1 Example of station attribute table

[0081] Station name Station number Line Transfer Station Station ahead Rear Station X coordinate Y coordinate Station 1 1 Line 1 no 2 none x1 y1 Station 2 2 Line 1 no 3 1 x2 y2 Station 28 28 Line 2 yes 29 27 x28 y28 …… …… …… …… …… ……

[0082] 2) Construction of time-varying OD data

[0083] a) Structure of time-varying OD data

[0084] The structure of time-varying OD data contains three contents as shown in Table 2.

[0085] Table 2 Time-varying OD data content

[0086] content Passenger flow date Extraction time granularity (unit: s) Time-varying OD passenger flow Data Types Date Int32 key-value

[0087] Time-varying OD passenger flow is a key-value pair data dictionary with OD name as key and n rows and 2 columns of OD passenger flow array as value. Figure 2 Each row of the OD passenger flow array represents a time granularity. The first row contains the passenger flow occurring in the first time granularity, the first column records the incoming passenger flow occurring in the first time granularity, and the second column represents the outgoing passenger flow occurring in the first time granularity.

[0088]

[0089] Figure 2 Composition of time-varying OD passenger flow data

[0090] b) Time-varying OD data extraction

[0091] ACF data includes passenger entry and exit data. Both entry and exit data contain four fields: passenger ID, transaction date, transaction time, and transaction station. Based on these four fields, AFC data is preprocessed to remove passenger ID information that only has entry or exit data.

[0092] Initialize the time-varying OD passenger flow. Add N (N-1) time-varying OD passenger flow key-value pairs to the time-varying OD passenger flow. The key name is in the form of "station 1-station 2". Use "-" to connect the two station names. The key value is an array with n rows and two columns and all elements are 0. n needs to take a value of appropriate size, which can be the sum of the number of time granularities contained in a day and the number of time granularities from 0:00 to the start of operation every day.

[0093] According to the ticket information of entering and exiting the station, for all passenger IDs, the OD information of the passenger trip is matched to obtain the passenger trip OD information table in the form of Table 3, and then each matched trip record is extracted into the time-varying passenger flow OD data. For example, the first data in the table, taking the time granularity of 60s as an example, the entry time is 11:25:11, which falls within the 686th time granularity of a single day, and the entry time is 11:35:11, which falls within the 696th time granularity of the same day. Let the OD(686,1)=OD(686,1)+1 and OD(686,2)=OD(686,2)+1 of the key name "Station 8-Station 2".

[0094] Table 3 Example of passenger travel OD information table

[0095] Passenger ID Pit time Pit stop location Departure time Exit location 674772359 2024 / 3 / 1 / 11:25:11 Station 8 2024 / 3 / 1 / 11:35:11 Station 2 672353353 2024 / 3 / 1 / 11:31:32 Station 2 2024 / 3 / 1 / 12:15:11 Station 9 ……

[0096] 6) Travel route data construction

[0097] The travel path data is a key-value pair data with the OD name as the key and the path information list as the value ( Figure 3 ). The path information list includes the stations that the path passes through, the length of the path and the selection probability of the path. Table 4 shows the path information list from station 1 to station 6. The list lists two paths from station 1 to station 6 and the corresponding selection probabilities and path lengths.

[0098]

[0099] Figure 3 Travel route data composition

[0100] Table 4 Example of path information list

[0101] path Probability Path length Station 1 - Station 2 - Station 3 - Station 4 - Station 5 - Station 6 0.82 5.9 Station 1 - Station 13 - Station 14 - Station 15 - Station 5 - Station 6 0.18 5.5

[0102] 7) Run graph data construction

[0103] The timetable data records the arrival and departure times of trains at various stations (as shown in Table 5). At any time, the location of the train can be inferred based on the timetable data, and the train that the passengers will take after entering the station can also be inferred.

[0104] Table 5 Example of operation chart data

[0105]

[0106]

[0107] 4.2 Passenger flow simulation model

[0108] The passenger flow simulation model is a travel simulation model that simulates the entire process from the passengers passing through the urban rail transit entrance gate to arriving at the destination station and exiting the station. It also takes into account factors such as passengers waiting for the train, getting on the train, getting off the train, walking in the station, and transferring. It can simulate the entry, exit, and transfer volume indicators of different stations at different times (granularity) throughout the day, as well as the number of passengers getting on and off the train at different stations and other passenger indicators. Based on these indicators, the operational indicators related to the urban rail transit passenger flow can be analyzed in detail.

[0109] The simulation model takes the time-varying OD passenger flow data, train operation diagram data and passengers' travel route selection data as input, and processes the passenger flow of different OD and different time granularity in chronological order ( Figure 3 ), the simulation obtains data such as the number of passengers entering and leaving the station, the number of transfer passengers, and the number of waiting passengers at different stations at different times, as well as the number of passengers getting on and off at different stations for different trains.

[0110] 1) Simulation method for time-varying passenger flow OD

[0111] In simulation, for the same OD, passengers may choose different paths. The number of people choosing different paths is first allocated according to probability, and then the corresponding OD passenger flow is processed according to different paths. In simulation, the time-varying passenger flow OD can be divided into three categories according to the number of transfers of the path selected by the time-varying OD passenger flow: the processing of OD passenger flow without transfer, the processing of OD passenger flow with one transfer, and the processing of OD passenger flow with multiple transfers.

[0112] There are some common symbols in the simulation processing steps, as shown in Table 6.

[0113] Table 6 General Symbols

[0114]

[0115] a) Processing of OD passenger flow without transfer

[0116] Assume that among the passenger flow from s1 to s2 generated in time granularity t, there are p people who choose the route from s1 to s2 without transfer. The average entry time at s1 station is t1, and the average exit time at s2 station is t2.

[0117] Record the corresponding inbound volume

[0118]

[0119] Calculate the time when arriving at the platform (granularity)

[0120]

[0121] Infer the train number, waiting time and boarding and alighting time of this group of passengers based on the operation diagram (granularity)

[0122] The passenger takes the next train no earlier than the passenger's arrival time at the platform (t 到 ) The number of the train that left, and the number of the train that was taken is k 乘 , the time from the drawing to the departure is

[0123] Therefore, the boarding time for passengers is:

[0124]

[0125] Therefore, the passengers' alighting time is:

[0126]

[0127] Record the number of people waiting for the bus at the corresponding time granularity

[0128]

[0129] Record the corresponding number of people getting on and off the bus

[0130]

[0131] Infer passenger exit times and record corresponding exit volumes

[0132]

[0133] b) Processing of OD passenger flow during a single transfer

[0134] Assume that among the passenger flow from s3 to s5 generated in the time granularity of t, there are p people who choose the path that requires transfer from s4. The average travel time for entering the station at s3 is t3, the average travel time for transfer at s4 is t4, and the average travel time for leaving the station at s5 is t5.

[0135] Record the corresponding inbound volume

[0136]

[0137] Calculate the time when the passenger arrives at the platform (granularity)

[0138]

[0139] According to the operation diagram, the train number, waiting time, boarding time (granularity) and the time (granularity) of getting off at the transfer station of this batch of passengers at the current station (s3) are inferred.

[0140] The passenger takes the next train no earlier than the passenger's arrival time at the platform (t 到 ) The number of the train that left, and the number of the train that was taken is k 乘 , and its scheduled departure time is

[0141] Therefore, the boarding time for passengers is:

[0142]

[0143] The time for passengers to get off at the transfer station is

[0144]

[0145] Record the number of people waiting for the bus at the corresponding time granularity

[0146]

[0147] 21Record the corresponding number of people getting on and off the bus

[0148]

[0149] 22Record the corresponding number of transfer passengers

[0150]

[0151] 23 Estimate the time to arrive at the target transfer station

[0152]

[0153] 24 According to the current OD passenger flow, consider the non-transfer passenger flow from s4 to s5, skip ①, and let t in ② 到 =t 到换 , start processing from ③ where there is no transfer passenger flow.

[0154] c) Processing of OD passenger flow with multiple transfers

[0155] Assume that there are p passengers in the passenger flow from s3 to s9 generated in the time granularity of t who choose a route that requires transferring once at multiple stations such as s4, s5...s8. The average travel time into the s3 station is t3, the average travel time for transfers at s4, s5...s8 is t4, t5...t8 respectively, and the average travel time for leaving the s9 station is t9.

[0156] ①Process the process of passengers entering the station at S3 and getting off at S4 according to the steps of one transfer (①-⑤)

[0157] ② From transfer station to transfer station, let t 到 =t 到换 , and then proceed according to ③-⑦ in the OD passenger flow processing for one transfer.

[0158] ③ According to the current OD passenger flow, consider the non-transfer passenger flow from s8 (the station waiting for the last transfer) to s9, skip the first step, and let t 到 =t 到换 , start processing from step 3 of the passenger flow without transfer.

[0159] 2) Passenger flow index analysis method

[0160] Based on the simulation results, combined with basic data such as operation diagram data, station attribute data, and line network topology data, after defining the time and space scope of the analysis, some simple calculations can achieve a refined analysis of passenger flow indicators. At the station level, the number of inbound, outbound, and transfer volumes can be analyzed; at the line level, the section passenger flow and section full load rate can be analyzed in a refined manner; and at the road network level, the average transport distance and passenger turnover can be analyzed in a refined manner.

[0161] There are many indicators that can be analyzed, and here are a few simple examples.

[0162] a) Calculation method of cross-sectional passenger flow

[0163] For the s1-s2 section, the passenger flow from s1 to s2 Calculation method:

[0164]

[0165] K 经 , the set of vehicles from s1 to s2 passing through the s1-s2 section during the statistical period

[0166] S 后 , on the route S1 to S2, meet at the station behind S1

[0167] b) Calculation of the number of passengers entering, exiting and transferring.

[0168] According to the statistical needs, the corresponding Sum up and get your entry, exit, and transfer volume indicators for the corresponding time and space range. For example, the entry volume of the entire network is Where S is the set of stations in the entire network, and T is the set of all time granularities.

[0169] c) Waiting time calculation

[0170] According to the statistical needs, Sum and multiply by the time granularity t val For example, the average waiting time for the entire network is Where S is the set of stations in the entire network, and T is the set of all time granularities.

[0171] In addition, by combining the station distance in the station topology data, we can also conduct a detailed analysis of the average transport distance and passenger turnover. By combining the train capacity and the train's cross-sectional passenger flow, we can conduct a detailed analysis of the train's full load rate... and so on.

[0172] Example Application

[0173] Nanchang Metro is an urban rail transit system serving Nanchang City, Jiangxi Province, China. Its first line was opened for passenger trial operation on December 26, 2015, making Nanchang the first city in Jiangxi Province to open rail transit. As of May 6, 2024, Nanchang Metro has opened 4 lines ( Figure 2), namely Nanchang Metro Line 1, Line 2, Line 3 and Line 4, with a total of 94 stations (including 9 transfer stations, not counted repeatedly), and a total operating line length of 128.45km. Among them, Metro Line 4 has an operating mileage far exceeding the first three metro lines, with a total length of 40km. It is an important line for Nanchang Metro to achieve network operation. At present, the average daily passenger volume of the line network exceeds 1 million, and the maximum daily passenger volume of the line network is as high as 2.2 million (May 1, 2024).

[0174] The compressed inbound AFC data of 6.81MB and the outbound AFC data of 6.80MB were processed by OD, and 2.56MB of 30S granularity OD time-varying passenger flow data were extracted. The OD passenger flow of any OD was randomly selected for visualization.

[0175] The time-varying passenger flow OD is processed by the passenger flow simulation model, and the boarding and alighting conditions of each train at different stations on the operation diagram are initially obtained. and Number of people entering each station at different times (granularity) Number of people leaving the station and number of passengers transferring By performing some simple arithmetic operations on these data, we can achieve a detailed analysis of urban rail transit passenger flow.

[0176] The analysis results are shown in Table 7: Analysis results of passenger transport indicators of Metro Building Station during morning peak hours, Table 8: Refined analysis results of passenger transport indicators of Nanchang Metro Line 1 during morning and evening peak hours, Table 9: Analysis results of the maximum granularity section and its full load rate in the upward direction of Nanchang Metro Line 1 in 15 minutes, Figure 4 :Analysis on the adaptability of transportation capacity and volume of Nanchang Metro Line 1, Figure 5 : Visualization results of train passengers and platform waiting passengers during the morning rush hour on Nanchang Metro Line 1.

[0177] Table 7 Analysis results of passenger transport indicators at Metro Building Station during morning peak hours

[0178] time Station volume Outbound volume Transfer volume Passenger volume 7:00-7:15 175 69 665 425 7:15-7:30 201 117 809 544 7:30-7:45 279 203 1154 886 7:45-8:00 375 382 1666 1296 8:00-8:15 481 847 2210 1598 8:15-8:30 471 1706 3346 2446 8:30-8:45 411 1855 3230 2400 8:45-9:00 288 2010 2007 1599

[0179] Table 8 Results of detailed analysis of passenger transport indicators for Nanchang Metro Line 1 during morning and evening peak hours

[0180] time Passenger volume Passenger Capacity Station volume Outbound volume Transfer volume Transfer rate Transfer coefficient Turnover 7:00-7:15 3645 3124 2546 1298 1099 0.3 1.43 20484 7:15-7:30 4489 4113 3290 1942 1199 0.27 1.36 25900 7:30-7:45 6625 5746 4576 2549 2049 0.31 1.45 36442 7:45-8:00 9120 7822 6202 3759 2918 0.32 1.47 50295 8:00-8:15 12607 10827 7919 6481 4688 0.37 1.59 69447 8:15-8:30 13641 11984 7746 9224 5895 0.43 1.76 76131 8:30-8:45 11389 11119 6865 10298 4524 0.4 1.66 75757 8:45-9:00 7591 6688 4169 10657 3422 0.45 1.82 51041 17:00-17:15 11720 10777 8646 6052 3074 0.26 1.36 63028 17:15-17:30 12813 11482 8823 7228 3990 0.31 1.45 75988 17:30-17:45 16288 13755 11862 8070 4426 0.27 1.37 84759 17:45-18:00 16472 15103 11255 9805 5217 0.32 1.46 95427 18:00-18:15 16288 13989 10938 10481 5350 0.33 1.49 92322 18:15-18:30 14948 14170 9892 10981 5056 0.34 1.51 92579 18:30-18:45 12824 10618 8172 10079 4652 0.36 1.57 75810 18:45-19:00 9778 9720 6772 8795 3006 0.31 1.44 64950

[0181] Table 9 Analysis results of the maximum granularity section and its full load rate in the upward direction of Nanchang Metro Line 1 within 15 minutes

[0182]

[0183]

[0184]

[0185] The present invention can accurately extract time-varying OD passenger flow from AFC data, thus achieving data lightweighting. Based on the time-varying OD passenger flow and combined with the operation diagram, the line network data and passenger travel path data highly simulate and restore the passengers' rides and passenger flow migration throughout the day. Based on the simulation results, the operation of the urban rail transit system can be analyzed in detail from three levels: stations, lines, and lines. The present invention can accurately grasp the operation dynamics of trains, stations, lines, and lines, and accurately analyze the operation indicators of trains, stations, lines, and lines, providing important support for the optimization and adjustment of the train operation diagram of the urban rail transit network.

[0186] 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 refined passenger flow analysis method based on a train operation network, characterized in that: include: Integrate the urban rail transit physical network with train operation diagram data; Construct a train operation network, use the traditional multi-path clearing ratio as the probability of passengers choosing a path, and build a passenger flow allocation simulation model based on train arrival and departure as driving events; Through the simulation process, the number of passengers getting on and off at stations and the number of passengers in each section of each train in the urban rail transit network are dynamically deduced. Based on the simulation results, the train load situation and transfer waiting time are analyzed, providing a basis for the preparation and adjustment of the network train operation diagram and the optimization of train connections between different lines.

2. The method for refined passenger flow analysis based on a train operation network according to claim 1 is characterized in that: The train operation network is constructed, the traditional multi-path clearing ratio is used as the probability of passengers choosing a path, and a passenger flow allocation simulation model based on train arrival and departure as driving events is constructed, including: According to the historical entry and exit ticket information, for all passenger IDs, match the passenger's travel OD information to obtain the passenger's historical travel OD information table; The multi-path clearing ratio of the passenger's historical travel OD information table is used as the probability of the passenger choosing a path, and then each matched travel record is extracted into the time-varying passenger flow OD data; A passenger flow distribution simulation model based on train arrival and departure as driving events is constructed. The simulation model takes time-varying OD passenger flow data, train operation diagram data and passengers' travel route selection data as input, and processes passenger flows of different OD and different time granularities in chronological order. The simulation obtains the number of passengers entering and leaving the station, the number of transfers and the number of waiting passengers at different stations at different times, as well as the number of passengers getting on and off at different stations for different trains.

3. The method for refined passenger flow analysis based on train operation network according to claim 2 is characterized in that: Pre-process the historical entry and exit ticket information to remove passenger ID information that only has entry or exit data.

4. The method for refined passenger flow analysis based on a train operation network according to claim 2 is characterized in that: According to the number of transfers of the path selected by the time-varying OD passenger flow, the time-varying passenger flow OD can be divided into three categories during simulation: the processing of OD passenger flow without transfer, the processing of OD passenger flow with one transfer and the processing of OD passenger flow with multiple transfers.

5. The method for refined passenger flow analysis based on train operation network according to claim 4 is characterized in that: 1) Processing of OD passenger flow without transfer: When p passengers in the passenger flow from s1 to s2 generated in the time granularity t choose the route from s1 to s2 without transfer, the average entry time at s1 station is t1, and the average exit time at s2 station is t2; ①Record the corresponding entry volume ②Calculate the time when you arrive at the platform ③ Infer the train number, waiting time and boarding and alighting time of this group of passengers based on the operation diagram The passenger takes the next train no earlier than the passenger's arrival time at the platform t 到 The number of the train that left, the number of the train that took is k 乘 , the time from the drawing to the departure is Therefore, the boarding time for passengers is: Therefore, the passengers' alighting time is: ④Record the number of people waiting for the bus at the corresponding time granularity ⑤Record the corresponding number of people getting on and off the bus ⑥ Infer the passenger exit time and record the corresponding exit volume 2) Processing of OD passenger flow during one transfer: When p passengers in the passenger flow from s3 to s5 generated in the time granularity t choose a route that requires a transfer from s4, the average travel time for entering the station at s3 is t3, the average travel time for transferring at s4 is t4, and the average travel time for leaving the station at s5 is t5; ①Record the corresponding entry volume ②Calculate the time when the passenger arrives at the platform (granularity) ③ According to the operation diagram, infer the train number, waiting time, boarding time and getting off time of the passengers at the current station s3; The passenger takes the next train no earlier than the passenger's arrival time at the platform t 到 The number of the train that left, the number of the train that took is k 乘 , and its scheduled departure time is Therefore, the boarding time for passengers is: The time for passengers to get off at the transfer station is ④Record the number of people waiting for the bus at the corresponding time granularity ⑤Record the corresponding number of people getting on and off the bus ⑥Record the corresponding number of transfers ⑦ Estimate the time to arrive at the target transfer platform ⑧ Consider the current OD passenger flow and the non-transfer passenger flow from s4 to s5, skip ①, and let t in ② 到 =t 到换 , start processing from ③ where there is no transfer passenger flow; 3) Processing of OD passenger flow with multiple transfers: When p passengers from s3 to s9 in the time granularity t choose a route that requires transfers at multiple stations s4, s5...s8, the average entry time at s3 is t3, the average transfer time at s4, s5...s8 is t4, t5...t8, and the average exit time at s9 is t9; ①According to steps ①-⑤ of a transfer, process the process of passengers entering the station at S3 and getting off at S4; ② From transfer station to transfer station, let t 到 =t 到换 , and then proceed as ③-⑦ in the OD passenger flow processing for one transfer; ③ According to the current OD passenger flow, consider the passenger flow without transfer from the last transfer station s8 to s9, skip the first step, and start processing from step 3 of the passenger flow without transfer in the second step.

Images