A method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling

Abstract

The present invention discloses a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling. First, the mobile phone signaling data is cleaned, and then the mobile phone signaling data of the selected base stations along the high-speed rail line are divided into high-speed rail travel railway section travel chains; the actual high-speed rail passenger mobile phone signaling data is screened according to the average speed of each section of the high-speed rail travel chain, and the mobile phone signaling high-speed rail passenger identification results are adjusted and expanded by using the mobile phone signaling data through the passenger volume direction ratio analysis, and the high-speed rail inter-city commuting type is determined according to the cross-city flow through the travel direction of the high-speed rail passengers, and the cross-city section travel law and cross-city travel frequency are obtained according to the cross-city flow, and the passenger flow and distribution characteristics of various commuting passengers are obtained after analysis, so as to realize the real-time dynamic identification of high-speed rail inter-city travel and commuting behavior from massive mobile phone signaling data. The method of the present invention can promote the integration of the comprehensive transportation system and the national land space planning, and can serve the integrated development of urban transportation planning and regional transportation.

Description

Technical Field

[0001] The present invention belongs to the field of traffic planning and computer technology, and specifically relates to a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling data. Background Art

[0002] With the in-depth implementation of the strategy of building a strong transportation nation and the continuous improvement of intercity high-speed railway services, the trend of cross-city commuting is becoming increasingly obvious in response to the rising cost of living in the core cities of metropolitan areas. Researching the scale of cross-city commuting in metropolitan areas and its influencing factors is of great significance for promoting the land space planning and regional integration of metropolitan areas oriented by high-speed transportation.

[0003] Traditional survey methods generally use passenger sampling travel questionnaire surveys or railway passenger flow data. The questionnaire survey method has the disadvantages of involving many people, high costs, and difficulty in organization and implementation. The survey results are also difficult to accurately reflect the cross-city commuting situation due to the sampling rate problem; railway passenger flow data can only reflect the overall passenger flow of the section, and cannot reflect the travel patterns and distribution of commuting passengers, and the data is not easy to obtain. In the era of mobile communication big data, mobile phones have become a must-have tool for people. The mobile phone signaling data field contains time and spatial location attributes, as well as call and information records. The user's travel trajectory can be inferred through the association of the above information, which is a good data source for studying the behavior and spatial distribution of urban residents.

[0004] Therefore, a technical method is needed to accurately identify the cross-sectional passenger flow and commuter passenger flow characteristics of intercity high-speed railways based on mobile phone signaling big data, promote the integration of the comprehensive transportation system and national land space planning, improve cross-city commuting efficiency, and promote urbanization. Summary of the invention

[0005] In view of the above-mentioned prior art, the present invention provides a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling, which is used to mine the characteristics of high-speed rail inter-city travel and inter-city commuting passengers from massive mobile phone signaling data. Based on this, it can serve urban transportation planning and regional transportation integrated development.

[0006] In order to solve the above technical problems, the present invention proposes a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling, comprising the following steps:

[0007] S1. Obtain mobile phone signaling data of the target city from the operator for at least 2 weeks, sort the mobile phone signaling data of each user in chronological order, and clean the data to remove duplicate, drifting and ping-pong data;

[0008] S2. Screening of base stations along the high-speed railway: First, the vector data of the high-speed railway lines is obtained from the national railway network vector dataset, and then the base stations RS along the high-speed railway lines that meet the following conditions are screened out:

[0009] Condition 1: The distance between the base station and the railway station along the high-speed railway line must be greater than 10km to avoid the impact of changes in the speed of high-speed trains entering and leaving the station:

[0010] Condition 2: The distance between the high-speed railway section where the base station is located and other railway lines along the high-speed railway must be greater than 3 km to prevent other lines from affecting the calculation results;

[0011] Condition 3: The curvature of the high-speed railway section where the base station is located is less than 0.0001 for at least 10 km, so as to facilitate the calculation of passenger speed later;

[0012] S3, dividing the mobile phone signaling data of the base stations RS along the high-speed railway line selected in step S2 into high-speed railway segment travel chains with time interval T0 as the dividing point;

[0013] S4, calculate the average speed V1 of each high-speed rail travel section, according to the set speed threshold V rail Filter out the mobile phone signaling data of those preliminarily identified as high-speed rail passengers;

[0014] S5, calculating the ratio R1 of the high-speed rail passenger volume identified by the mobile phone signaling data filtered out in step S4 to the actual high-speed rail passenger volume, R1 = Q1 / Q0, where Q1 represents the high-speed rail passenger volume corresponding to the high-speed rail passengers identified according to the mobile phone signaling data, and Q0 represents the actual high-speed rail passenger volume obtained according to ticketing statistics;

[0015] S6, judging the direction of high-speed rail travel according to the time and position of the two end points of the travel chain of each high-speed rail travel section, and calculating the ratio R2 of the high-speed rail passenger volume in each direction identified by the mobile phone signaling data filtered out in step S4;

[0016] S7. Determine the difference between the mobile phone signaling data recognition result and the actual high-speed rail passenger volume, and adjust and expand the mobile phone signaling high-speed rail passenger recognition result; including:

[0017] If the difference between the ratio R1 and the operator market share R01 is less than 5%, and the average difference between the ratio R2 and the actual railway passenger volume ratio R02 is less than 10%, then the expanded mobile phone signaling data is obtained according to the operator market share R01, and then step S8 is executed; otherwise, it is determined whether the ratio R1 is less than the operator market share R01; if the conditions are met, the speed threshold V is reduced. rail ; Otherwise, increase the speed threshold V rail ; Return to step S4;

[0018] S8, counting the cross-city cross-section flow of high-speed rail by date and time period, and continuously and dynamically monitoring the operation status and rules of the cross-city cross-section of high-speed rail; including: counting the cross-city cross-section passenger flow of high-speed rail after the sample expansion according to the method of step S6;

[0019] S9. Judging from the travel frequency of high-speed rail passengers, the type of cross-city commuting of high-speed rail passengers is one of the following three situations:

[0020] Scenario 1: Weekly commuting, Scenario 2: Daily commuting, Scenario 3: Other commuting;

[0021] Label the expanded mobile phone signaling data of each high-speed rail passenger; including: counting the number of cross-city trips per user per week and defining commuting classification rules:

[0022] Daily commuters: those who travel across cities within two weeks and have round trips at least three days a week;

[0023] Weekly commuters: those who travel across cities within two weeks and make fewer than three round trips per week;

[0024] Other commuting groups: except for weekly commuting and daily commuting;

[0025] The classified users are marked with commuting types in the expanded signaling data;

[0026] S10. Analyze the passenger flow and distribution characteristics of various types of commuter passengers, including:

[0027] 1) Statistics on cross-city high-speed rail commuting traffic in different directions by date and time period, and continuously and dynamically monitor the operation status and rules of cross-city high-speed rail commuting traffic every day within a week;

[0028] 2) Count the number of weekly commuters, daily commuters and other commuters respectively, and continuously and dynamically monitor the changes in the number of cross-city commuters on high-speed rail;

[0029] 3) Identify the stops of high-speed rail commuters before and after entering and exiting the station, and dynamically monitor the spatial distribution of different types of people commuting across cities by high-speed rail.

[0030] Furthermore, the high-speed rail inter-city travel and commuting behavior identification and monitoring method of the present invention includes:

[0031] In the step S1, the mobile phone signaling data of each user is sorted in chronological order and the data is cleaned, including:

[0032] S11, removing duplicate mobile phone signaling data;

[0033] S12, remove drift data: for the same user, obtain the time difference t between the current signaling and the previous signaling and the difference S in the longitude and latitude spherical distance between the two, and calculate the speed V of two consecutive points according to the speed formula: V = S / t. If V is greater than the speed threshold Vmax, remove the current mobile phone signaling record;

[0034] S13, remove ping-pong signaling data: for the same user, if the base stations of 3 or 4 consecutive signalings are the same, then remove the signaling records of other base stations except the first and last base stations.

[0035] The specific content of step S3 includes: extracting signaling data passing through the base station RS along the high-speed rail line from the mobile phone signaling data. For the same mobile phone user, if the time difference between two consecutive mobile phone signaling data records belonging to the high-speed rail base station is greater than T0, it is judged as 2 trips, and the high-speed rail travel railway segment travel chain is divided from here, where the time interval T0 is 2 hours.

[0036] The specific contents of step S4 include: calculating the time difference t between the first and last signaling records of each high-speed railway segment travel chain rail The distance difference S from the sphere rail , find the average speed V1 of the high-speed rail travel chain, if the average speed V1 is greater than the speed threshold V rail , then the mobile phone user is a high-speed rail passenger.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] (1) The high-speed rail inter-city passengers identified by mobile phone signaling data are verified by the total number of inter-city high-speed rail passengers and the proportion of passenger travel directions, and the high-speed rail operating speed threshold V is dynamically adjusted. rail Through detailed verification steps, the number of non-high-speed rail passengers identified incorrectly is reduced, the accuracy and reliability of data cleaning and recognition results are ensured, and the recognition results are more consistent with reality. Traditional railway passenger flow statistics are not easy to obtain, and the total volume and commuting characteristics are not accurate. This method considers the two dimensions of total volume and direction ratio, and uses traditional railway passenger flow statistics data for verification, which effectively improves the accuracy of high-speed rail passenger identification and provides more accurate basic data for subsequent high-speed rail cross-city passenger flow analysis.

[0039] (2) The present invention provides a full-process high-speed rail inter-city passenger identification and analysis algorithm, which can quickly realize the analysis and statistics of inter-city high-speed rail passenger flow and inter-city high-speed rail commuter passenger flow, and realize dynamic monitoring of daily time-divided inter-city section flow, the number of inter-city commuters, the inter-city commuter section flow, and the spatial activity distribution of inter-city commuters in the city. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1is a flow chart of the method of the present invention;

[0041] Figure 2 This is a diagram of the mobile phone signal cleaning process. The dots are the data selected after cleaning, and the circles and dots are all the data before cleaning.

[0042] Figure 3 This is a schematic diagram of the selection of base stations along the Beijing-Tianjin Intercity High-speed Railway. The circles are the selected base stations along the line, the curve is the Tianjin railway line, and the line where the circle is located is the Beijing-Tianjin Intercity High-speed Railway line;

[0043] Figure 4 This is an example of trip chain segmentation and speed calculation for high-speed rail trips;

[0044] Figure 5 It is a schematic diagram of the ratio of mobile phone signaling passenger flow to actual passenger flow;

[0045] Figure 6 is the passenger flow ratio from Beijing to Tianjin / from Tianjin to Beijing;

[0046] Figure 7-1 is the travel volume between Tianjin and Beijing in different time periods;

[0047] Figure 7-2 is the travel volume between Beijing and Tianjin in different time periods;

[0048] Figure 8-1 is the commuting volume between Tianjin and Beijing in different time periods;

[0049] Figure 8-2 is the commuting volume between Beijing and Tianjin by time period;

[0050] Fig. 9 It is the distribution map of commuter passengers between Beijing and Tianjin. The curve is the Tianjin railway line, and the heat map is the distribution heat map of commuter passengers between Beijing and Tianjin.

[0051] Fig.10 is the daily travel volume of Beijing-Tianjin intercity railway by type;

[0052] Fig.11 It is the ratio of daily inter-city commuting between Beijing and Tianjin to the commuting population in each administrative district of Tianjin. DETAILED DESCRIPTION

[0053] The present invention proposes a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling. The design idea is: first, clean the mobile phone signaling data, and then divide the user signaling data of the selected base stations along the high-speed rail into the high-speed rail travel railway segment travel chain; according to the average speed of each high-speed rail travel chain, the actual high-speed rail passenger user signaling data is screened, and the user signaling data is used to adjust and expand the mobile phone signaling high-speed rail passenger identification results through passenger volume direction ratio analysis. According to the travel direction of high-speed rail passengers, the cross-city section travel rules and cross-city travel frequency are obtained according to the cross-city flow to determine the type of high-speed rail inter-city commuting, and the passenger flow and distribution characteristics of various commuting passengers are obtained after analysis, so as to realize the real-time dynamic identification of high-speed rail inter-city travel and commuting behavior from massive mobile phone signaling data. The method of the present invention can promote the integration of the comprehensive transportation system and national land space planning, and can serve the integrated development of urban transportation planning and regional transportation.

[0054] In the method of the present invention, the high-speed rail inter-city passengers identified by the mobile phone signaling data are checked by two dimensions: the total number of high-speed rail inter-city passengers and the proportion of passenger travel directions, and the high-speed rail running speed threshold V is dynamically adjusted. rail ; Through detailed verification steps, the number of non-high-speed rail passengers identified incorrectly is reduced, ensuring the accuracy and reliability of data cleaning and identification results; effectively improving the accuracy of high-speed rail passenger identification, providing more accurate basic data for subsequent high-speed rail inter-city passenger flow analysis. The high-speed rail inter-city travel proposed in the present invention can be processed quickly, and dynamic monitoring of daily time-divided inter-city section flow, the number of inter-city commuters, the inter-city commuter section flow, and the spatial activity distribution of inter-city commuters in the city can be achieved; a full-process high-speed rail inter-city passenger identification and analysis algorithm is provided, which can quickly realize the analysis and statistics of inter-city high-speed rail passenger flow and inter-city high-speed rail commuter passenger flow; through detailed verification and sample expansion steps, the accuracy and reliability of the analysis results are ensured.

[0055] The design ideas of the technical solutions of the present invention are described below.

[0056] like Figure 1 As shown, the present invention proposes a method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling, comprising the following steps:

[0057] Step S1, obtaining the mobile phone signaling data of the target city for at least 2 weeks from the operator, sorting the mobile phone signaling data of each user in chronological order, and cleaning the data to remove duplicate, drift and ping-pong data; including:

[0058] Step S1: obtain the mobile phone signaling data of the target city for a period of time (at least 2 weeks) from the operator, sort the signaling data of each user in chronological order, and clean the data to remove duplicate, drift and ping-pong data, including:

[0059] Step S11: sort all the data of all users according to the time sequence of each user's mobile phone signaling, and remove duplicate mobile phone signaling data.

[0060] Step S12, remove signaling drift data with unreasonable speed. For the same user, obtain the time difference t between the current signaling and the previous signaling and the difference S in the longitude and latitude spherical distance between the two. According to the speed formula: V=S / t, obtain the speed V of two consecutive points. If V is greater than the speed threshold Vmax, remove the current mobile phone signaling record.

[0061] Step S13, remove ping-pong signaling data: For the same user, if the base stations of 3 or 4 consecutive signalings are the same, then remove the signaling records of base stations other than the first and last base stations. For example: if the base station number of 3 consecutive signalings is in the form of "aba", then remove the signaling record of the middle base station numbered "b"; if the base station number of 4 consecutive signalings is in the form of "abca", then remove the signaling records of the middle base stations numbered "b" and "c".

[0062] S2. Screening of base stations along the high-speed railway: including:

[0063] First, we obtain the vector data of the high-speed rail lines from the national railway network vector dataset. Then, we select the base stations along the high-speed rail lines that are suitable for calculating the speed. The selection criteria are the base station RS along the high-speed rail lines that meet the following conditions at the same time:

[0064] Condition 1: The selected base stations along the high-speed railway need to be far away from the railway stations, that is, the distance between the base stations along the high-speed railway and the railway stations needs to be greater than 10km to avoid the influence of the speed changes of high-speed trains entering and leaving the station:

[0065] Condition 2: The high-speed railway section where the base stations along the screened high-speed railway are located cannot be interfered by other lines, that is, the distance from other railway lines must be greater than 3km to prevent other lines from affecting the calculation results;

[0066] Condition 3: The high-speed railway section where the base station is located is as flat as possible. For example, the curvature of the high-speed railway section is less than 0.0001 for at least 10 km, so that the passenger speed can be calculated more accurately later.

[0067] Step S3, dividing the mobile phone signaling data of the base stations RS along the high-speed railway selected in step S2 into high-speed railway segment travel chains with time interval T0 as the dividing point; comprising:

[0068] The signaling data of the base stations RS along the high-speed rail line are extracted from the mobile phone signaling data. For the same mobile phone user, if the time difference between two consecutive mobile phone signaling data records belonging to the high-speed rail base station is greater than T0, it is judged as two trips, and the high-speed rail travel railway segment travel chain is divided from here, where the time interval T0 is 2 hours.

[0069] Step S4: Calculate the average speed V1 of each high-speed railway segment travel chain, according to the set speed threshold V rail Filter out the mobile phone signaling data of those who are initially identified as high-speed rail passengers; including:

[0070] Calculate the time difference t between the first and last signaling records of each high-speed rail travel chain rail The distance difference S from the sphere rail , find the average speed V1 of the high-speed rail travel chain, if the average speed V1 is greater than the speed threshold V rail , then the mobile phone user is a high-speed rail passenger.

[0071] Step S5, calculating the ratio R1 of the high-speed rail passenger volume identified by the mobile phone signaling data screened in step S4 to the actual high-speed rail passenger volume, R1=Q1 / Q0, wherein Q1 represents the high-speed rail passenger volume corresponding to the high-speed rail passengers identified according to the mobile phone signaling data, and Q0 represents the actual high-speed rail passenger volume obtained according to ticketing statistics;

[0072] S6, judging the direction of high-speed rail travel according to the time and position of the two end points of the travel chain of each high-speed rail travel section, and calculating the ratio R2 of the high-speed rail passenger volume in each direction identified by the mobile phone signaling data filtered out in step S4;

[0073] Step S7) determining the difference between the mobile phone signaling recognition result and the actual high-speed rail passenger volume, and adjusting and expanding the mobile phone signaling high-speed rail passenger recognition result; including:

[0074] Step S7, determine the difference between the mobile phone signaling data recognition result and the actual high-speed rail passenger volume, adjust and expand the mobile phone signaling high-speed rail passenger recognition result; if the ratio R1 is close to the operator market share R01 and the ratio R2 is close to the actual railway direction passenger volume ratio R02, then the expanded high-speed rail passenger mobile phone signaling data is obtained according to the operator market share; otherwise, readjust the speed threshold V rail, return to step S4, and repeat steps S5 to S7. Through detailed verification steps, the number of non-high-speed rail passengers identified incorrectly is reduced, ensuring the accuracy and reliability of data cleaning and identification results. Specifically, if the difference between the ratio R1 and the operator market share R01 is less than 5%, and the average difference between the ratio R2 and the actual railway passenger volume ratio R02 is less than 10%, then the expanded mobile phone signaling data is obtained according to the operator market share R01, and then step S8 is executed; otherwise, it is determined whether the ratio R1 is less than the operator market share R01; if the conditions are met, the speed threshold V is reduced. rail ; Otherwise, increase the speed threshold V rail ; Return to step S4.

[0075] Step S8, counting the cross-city high-speed rail cross-direction traffic by date and time period, and continuously and dynamically monitoring the operation status and rules of the cross-city high-speed rail sections; including: according to the method of step S6, counting the cross-city high-speed rail passenger flow in different directions after the sample expansion.

[0076] Step S9, judging the inter-city commuting type (weekly commuting, daily commuting, other commuting) of high-speed rail passengers according to the travel frequency of high-speed rail passengers, and labeling the expanded mobile phone signaling data of each high-speed rail passenger; including:

[0077] Count the number of cross-city trips each user makes each week and define commuting classification rules:

[0078] Daily commuters: those who travel across cities within two weeks and have round trips at least three days a week;

[0079] Weekly commuters: those who travel across cities within two weeks and have less than three round trips per week, and the trips are not concentrated on the same day;

[0080] Other commuting groups: people other than weekly commuters and daily commuters;

[0081] The classified users are marked with commuting types in the expanded signaling data;

[0082] Step S10: analyzing the passenger flow and distribution characteristics of various types of commuter passengers, including:

[0083] 1) Statistics on cross-city high-speed rail commuting traffic in different directions by date and time period, and continuously and dynamically monitor the operation status and rules of cross-city high-speed rail commuting traffic every day within a week;

[0084] 2) Count the number of weekly commuters, daily commuters and other commuters respectively, and continuously and dynamically monitor the changes in the number of cross-city commuters on high-speed rail;

[0085] 3) Identify the stops of high-speed rail commuters before and after entering and exiting the station, and dynamically monitor the spatial distribution of different types of people commuting across cities by high-speed rail.

[0086] Example:

[0087] Taking the Beijing-Tianjin intercity high-speed railway as an example, the method of the present invention is used to realize the recognition and monitoring of high-speed rail intercity travel and commuting behavior. The process is as follows:

[0088] Step 1) obtain mobile phone signaling data in Tianjin over a period of time from the operator, sort the signaling data of each user in chronological order, and clean the data to remove duplicate, drift and ping-pong data. Figure 2 The figure shows the process of cleaning the mobile phone signaling, where the dots are the data selected after cleaning, and the circles and dots are all the data before cleaning.

[0089] Step 2) Taking the Beijing-Tianjin Intercity High-Speed ​​Railway as an example, select base stations along the high-speed railway. Figure 3 It shows the screening results of base stations along the Beijing-Tianjin Intercity High-Speed ​​Railway, where the circles are the screened base stations along the line, the curve is the Tianjin railway line, and the line where the circle is located is the Beijing-Tianjin Intercity High-Speed ​​Railway line.

[0090] Step 3) Filter the user signaling data of the base station RS along the high-speed railway obtained in step 2), and divide the high-speed railway travel chain into sections with T0 as the time interval; calculate the average speed of each section of the high-speed railway travel chain, and use the speed threshold V rail Filter out actual high-speed rail passenger user signaling data; Figure 4 An example of high-speed rail travel railway segment travel chain segmentation and speed calculation in this embodiment is shown, wherein each row of data represents the high-speed rail travel segment information of each mobile phone user, the "User-Date" represents the user number and date information, the date month is March 2023, the "Split Travel" represents the high-speed rail travel segment grouping of the user in a certain day, and the "High-speed Rail Travel Chain Speed" represents the speed of a certain high-speed rail travel segment, and its unit is: meters per second.

[0091] Step 4) Calculate the ratio of high-speed rail passenger volume identified by mobile phone signaling to the actual high-speed rail passenger volume R1≈58%; calculate the ratio of high-speed rail passenger volume identified by mobile phone signaling by direction R2; determine the difference between the mobile phone signaling recognition result and the actual high-speed rail passenger volume, and adjust the speed threshold V rail =150km / h, adjust and expand the results of high-speed rail passenger recognition using mobile phone signaling; Figure 5 The figure shows the ratio of mobile signaling passenger flow to actual passenger flow. It can be seen that the ratio of daily high-speed rail passenger flow identified by mobile signaling to actual high-speed rail passenger flow is about 60%, with an average share of 58%, which is basically consistent with the mobile operator share of about 57.6%. Figure 6 The passenger flow ratio from Beijing to Tianjin / Tianjin to Beijing is shown. It can be seen that the average error between the ratio of mobile phone signaling recognition and the actual passenger flow ratio is only 6.75%.

[0092] Step 5) Statistics of cross-city high-speed rail cross-city cross-section flow by date and time period, and continuous dynamic monitoring of cross-city high-speed rail cross-section operation and regularity; in terms of time-divided operation on weekdays, from Tianjin to Beijing, the Monday morning peak (7:00-9:00), Friday evening peak (19:00-20:00) and multiple afternoon periods are in saturated operation; the saturated operation periods from Beijing to Tianjin are mainly concentrated in the Monday morning peak (8:00-9:00), Friday afternoon (15:00-17:00) and multiple evening peak periods (18:00-12:00, 21:00-23:00). In terms of time-divided operation on Saturdays and Sundays, Tianjin to Beijing direction is concentrated in Sunday afternoon and evening, and Beijing to Tianjin direction is concentrated in Saturday morning, and the corresponding time periods are all in saturated operation periods. Figure 7-1 It shows the travel volume (passengers / hour) in the Tianjin-Beijing direction by time period. Figure 7-2 The figure shows the travel volume (passengers / hour) in the Beijing-Tianjin direction by time period.

[0093] Step 6) Determine the commuting type (weekly commuting, daily commuting, other groups) by the travel frequency of high-speed rail passengers, and label the expanded signaling data of each user. In this embodiment, among the railway commuting population, the weekly commuting population accounts for a relatively high proportion. According to the headcount statistics, about 69,000 people were detected from the mobile phone signaling for two weeks to travel on a weekly basis, and about 7,200 people were detected to travel on a daily basis. The number of weekly commuters is 9.6 times that of daily commuters. Fig.10 The figure shows the comparison between the daily and weekly commuting trips for each day of the week. It can be seen that the weekly commuting trips on Friday, Sunday and Monday are much greater than the daily commuting trips, and the daily commuting trips on Saturday and Sunday are the least.

[0094] Step 7) Analyze the passenger flow and distribution characteristics of various types of commuter passengers.

[0095] (1) Monitoring of total commuter passenger flow

[0096] In terms of time-divided operation on weekdays, the trains from Tianjin to Beijing are in full operation during the Monday morning peak (7:00-9:00), Friday evening peak (19:00-20:00) and multiple periods in the afternoon; the saturated operation periods from Beijing to Tianjin are mainly concentrated in the Monday morning peak (8:00-9:00), Friday afternoon (15:00-17:00) and multiple periods in the evening peak (18:00-12:00, 21:00-23:00). In terms of time-divided operation on Saturdays and Sundays, the trains from Tianjin to Beijing are mostly concentrated in the afternoon and evening of Sundays, and the trains from Beijing to Tianjin are mostly concentrated in the morning of Saturdays, and the corresponding periods are all in full operation. Fig.10 The figure shows the daily travel volume of Beijing-Tianjin intercity railway by type (10,000 people / day).

[0097] (2) Beijing-Tianjin intercity commuter traffic by time period

[0098] The Beijing-Tianjin Intercity Railway carries about 95% of the railway commuter traffic between Beijing and Tianjin. From the perspective of the peak commuting hours of the Beijing-Tianjin Intercity Railway in different directions: from Tianjin to Beijing, the morning peak operation period on Monday is extended to two hours (7:00-9:00), and then moved forward to Sunday evening (19:00-22:00); from Beijing to Tianjin, the peak period is concentrated on Friday evening, and the period is extended to 18:00-22:00. The Beijing-Tianjin Intercity Railway is basically in a saturated operation state during the corresponding period. Figure 8-1 The figure shows the commuting volume (person-times / hour) in the Tianjin-Beijing direction by time period. Figure 8-2 The figure shows the commuting volume (passengers / hour) in the Beijing-Tianjin direction by time period.

[0099] (3) Distribution of commuters between Beijing and Tianjin

[0100] In terms of the ratio of daily commuters to the total number of commuters, Wuqing District, which is close to Beijing and has a railway hub, has the highest daily commuting proportion of over 20%, followed by Hedong District, Hebei District, Hongqiao District, Xiqing District and other areas close to the hub. Fig. 9 The figure shows the distribution map of commuter passengers between Beijing and Tianjin. The lines represent the Tianjin railway lines, and the heat map shows the distribution of commuter passengers between Beijing and Tianjin. Fig.11 The figure shows the ratio of daily inter-city commuting and commuting population between Beijing and Tianjin in each administrative region.

[0101] Although the present invention has been described above in conjunction with the accompanying drawings, the present invention is not limited to the above-mentioned specific embodiments, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can make many improvements and changes without departing from the purpose of the present invention, which are all within the protection of the present invention.

Claims

1. A method for identifying and monitoring high-speed rail inter-city travel and commuting behavior based on mobile phone signaling, characterized in that: The following steps are involved: S1. Obtain mobile phone signaling data of the target city from the operator for at least 2 weeks, sort the mobile phone signaling data of each user in chronological order, and clean the data to remove duplicate, drifting and ping-pong data; S2. Screening of base stations along the high-speed railway: First, obtain the vector data of the high-speed railway line from the national railway network vector dataset, and then screen out the base stations RS along the high-speed railway line that meet the following conditions: Condition 1: The distance between the base station and the railway station along the high-speed railway line must be greater than 10km to avoid the impact of changes in the speed of high-speed trains entering and leaving the station: Condition 2: The distance between the high-speed railway section where the base station is located and other railway lines along the high-speed railway must be greater than 3 km to prevent other lines from affecting the calculation results; Condition 3: The curvature of the high-speed railway section where the base station is located is less than 0.0001 for at least 10 km, so as to facilitate the calculation of passenger speed later; S3, dividing the mobile phone signaling data of the base stations RS along the high-speed railway line selected in step S2 into high-speed railway segment travel chains with time interval T0 as the dividing point; S4, calculate the average speed V1 of each high-speed rail travel section, according to the set speed threshold V rail Filter out the mobile phone signaling data of those preliminarily identified as high-speed rail passengers; S5, calculating the ratio R1 of the high-speed rail passenger volume identified by the mobile phone signaling data filtered out in step S4 to the actual high-speed rail passenger volume, R1 = Q1 / Q0, where Q1 represents the high-speed rail passenger volume corresponding to the high-speed rail passengers identified according to the mobile phone signaling data, and Q0 represents the actual high-speed rail passenger volume obtained according to ticketing statistics; S6, judging the direction of high-speed rail travel according to the time and position of the two end points of the travel chain of each high-speed rail travel section, and calculating the ratio R2 of the high-speed rail passenger volume in each direction identified by the mobile phone signaling data filtered out in step S4; S7. Determine the difference between the mobile phone signaling data recognition result and the actual high-speed rail passenger volume, and adjust and expand the mobile phone signaling high-speed rail passenger recognition result; including: If the difference between the ratio R1 and the operator market share R01 is less than 5%, and the average difference between the ratio R2 and the actual railway passenger volume ratio R02 is less than 10%, then the expanded mobile phone signaling data is obtained according to the operator market share R01, and then step S8 is executed; otherwise, it is determined whether the ratio R1 is less than the operator market share R01; if the conditions are met, the speed threshold V is reduced. rail ; Otherwise, increase the speed threshold V rail ; Return to step S4; S8, counting the cross-city cross-section flow of high-speed rail by date and time period, and continuously and dynamically monitoring the operation status and rules of the cross-city cross-section of high-speed rail; including: counting the cross-city cross-section passenger flow of high-speed rail after the sample expansion according to the method of step S6; S9. Judging from the travel frequency of high-speed rail passengers, the type of cross-city commuting of high-speed rail passengers is one of the following three situations: Scenario 1: Weekly commuting, Scenario 2: Daily commuting, Scenario 3: Other commuting; Label the expanded mobile phone signaling data of each high-speed rail passenger; including: counting the number of cross-city trips per user per week and defining commuting classification rules: Daily commuters: those who travel across cities within two weeks and have round trips at least three days a week; Weekly commuters: those who travel across cities within two weeks and make fewer than three round trips per week; Other commuting groups: except for weekly commuting and daily commuting; The classified users are marked with commuting types in the expanded signaling data; S10. Analyze the passenger flow and distribution characteristics of various types of commuter passengers, including: 1) Statistics on cross-city high-speed rail commuting traffic in different directions by date and time period, and continuously and dynamically monitor the operation status and rules of cross-city high-speed rail commuting traffic every day within a week; 2) Count the number of weekly commuters, daily commuters and other commuters respectively, and continuously and dynamically monitor the changes in the number of cross-city commuters on high-speed rail; 3) Identify the stops of high-speed rail commuters before and after entering and exiting the station, and dynamically monitor the spatial distribution of different types of people commuting across cities by high-speed rail.

2. The high-speed rail inter-city travel and commuting behavior identification and monitoring method according to claim 1 is characterized in that: In the step S1, the mobile phone signaling data of each user is sorted in chronological order and the data is cleaned, including: S11, removing duplicate mobile phone signaling data; S12, remove drift data: for the same user, obtain the time difference t between the current signaling and the previous signaling and the difference S in the longitude and latitude spherical distance between the two, and calculate the speed V of two consecutive points according to the speed formula: V = S / t. If V is greater than the speed threshold Vmax, remove the current mobile phone signaling record; S13, remove ping-pong signaling data: for the same user, if the base stations of 3 or 4 consecutive signalings are the same, then remove the signaling records of other base stations except the first and last base stations.

3. The high-speed rail inter-city travel and commuting behavior identification and monitoring method according to claim 1 is characterized in that: The specific contents of step S3 include: extracting the signaling data passing through the base station RS along the high-speed rail line from the mobile phone signaling data. For the same mobile phone user, if the time difference between two consecutive mobile phone signaling data records belonging to the high-speed rail base station is greater than T0, it is judged as 2 trips, and the high-speed rail travel railway segment travel chain is divided from here, where the time interval T0 is 2 hours.

4. The high-speed rail inter-city travel and commuting behavior identification and monitoring method according to claim 1 is characterized in that: The specific contents of step S4 include: calculating the time difference t between the first and last signaling records of each high-speed railway segment travel chain rail The distance difference S from the sphere rail , find the average speed V1 of the high-speed rail travel chain, if the average speed V1 is greater than the speed threshold V rail , then the mobile phone user is a high-speed rail passenger.

Images