A method, device and storage medium for dividing rail transit operation periods

Abstract

The present invention discloses a method, device and storage medium for dividing rail transit operating periods based on an FCM clustering algorithm. The method first samples historical data of urban rail transit within the line operating hours and divides the historical data according to a set time granularity. After obtaining the target data set, passenger flow analysis indicators are calculated, and then passenger flow analysis indicators with a greater impact on passenger flow peaks are selected through screening. The FCM fuzzy clustering algorithm is then used for clustering, and the defuzzified clustering results are used as boundary points for operating period division. Finally, the optimal operating period division scheme is obtained, which can more effectively reflect the actual changing patterns of passenger flow and solve the problems of strong subjectivity and low accuracy of existing operating period division methods.

Description

Technical Field

[0001] The present invention belongs to the field of transportation, and in particular to a method for dividing rail transit operation time periods based on an FCM clustering algorithm. Background Art

[0002] Currently, urban rail transit has become the preferred mode of transportation due to its high punctuality, high speed, and large capacity. The gradual expansion of the rail transit network has led to an increasingly complex network structure, resulting in a more diverse and complex relationship between passenger flow distribution structure and pattern, as well as passenger demand and train capacity. Therefore, given the dynamic temporal and spatial distribution of passenger flow, subway operations management departments need to dynamically adjust subway operation plans based on passenger flow characteristics within operating hours. However, subway passenger flow exhibits significant variations across different time periods, placing higher demands on the rational division of operating hours and the formulation of appropriate time period parameters.

[0003] Currently, the division of subway operating periods relies primarily on the work experience of technical personnel, which is highly subjective and deviates from actual operating rules, easily resulting in unreasonable period division results. The mismatch between passenger demand and transportation capacity in some urban rail systems is prominent, especially during peak hours in the morning and evening. Passengers are often stranded on platforms and oversaturated carriages occur frequently, resulting in serious passenger congestion. Since operating parameters inevitably change over time, in actual applications, in order to keep operating parameters as stable as possible, it is necessary to divide operating periods and use different operating strategies for different operating periods. In order to formulate operational organization plans that accurately match passenger demand and improve the operational management level of urban rail, there is an urgent need for a scientific and reasonable method for dividing operating periods to guide plan formulation and achieve a precise match between transportation capacity and transportation capacity. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for dividing rail transit operating time periods based on the FCM clustering algorithm, so as to solve the problem that the current division of subway operating time periods mainly relies on the work experience of technical personnel, is highly subjective, deviates from the actual operating rules, and is very likely to produce unreasonable time period division results.

[0005] The technical solution to achieve the purpose of the present invention is: a method for dividing rail transit operation periods based on the FCM clustering algorithm, which adopts the following steps:

[0006] Step S1: Collect historical data of urban rail transit during the line operation time, and divide the original data into N minutes as the time granularity to obtain the target data set;

[0007] Step S2: Based on the target data set obtained in S1, calculate the passenger flow analysis indicators within each time granularity, where the passenger flow analysis indicators include two categories: cross-sectional passenger flow indicators and station passenger flow indicators. The cross-sectional passenger flow indicators include cross-sectional passenger flow, cross-sectional full load factor, and passenger standing density. The station passenger flow indicators include passenger volume, passenger arrival rate, and passenger flow imbalance coefficient.

[0008] Step S3: Calculate the correlation coefficients between the passenger flow analysis indicators in S2, and then select the two section passenger flow indicators and two station passenger flow indicators with the highest correlation coefficients;

[0009] Step S4: Perform FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in S3, and fuzzify the clustering results to divide the section operation period and station operation period;

[0010] Step S5: Merge the section operation period and the station operation period obtained in step S4 to obtain the division result of the line operation period.

[0011] Furthermore, the historical data includes operating data, cross-sectional passenger flow and station data. The operating data includes the number of train vehicles, car length, car width, total width of the cross-sectional area of ​​the car seating area, number of seats in the car, and the station data includes passenger flow entering the station and rail transit OD data.

[0012] Furthermore, the S1 also includes data cleaning of historical data and standardization of the target data set.

[0013] Furthermore, in said S2:

[0014] The calculation method of section full load factor α is as follows:

[0015]

[0016] Where, α is the section full load factor; Q 断面 is the passenger flow in the section within the unit time granularity; g is the number of train sets; p is the number of passengers in the train;

[0017] The formula for calculating passenger standing density is as follows:

[0018]

[0019] Where, ρ is the passenger standing density; q is the number of passengers on the train; δ is the carriage conversion factor; g is the number of train vehicles; φ is the train conversion factor; S is the number of seats on the train; L is the length of the carriage; B is the width of the carriage; b is the total cross-sectional width of the carriage seating area;

[0020] The calculation formula for passenger arrival rate is as follows:

[0021]

[0022] Where, λ is the passenger arrival rate by time; Q e is the passenger flow in the e-th time granularity during the line operation time, Q e+1 The passenger flow entering the station within the e+1th time granularity during the line operation time;

[0023] The imbalance coefficient of passenger flow characteristic index is calculated based on the passenger flow OD data in S1. The calculation formula is as follows:

[0024]

[0025] Where, β i is the passenger flow imbalance coefficient; V e is the maximum passenger flow in the e-th time granularity during the line operation time; V max is the maximum passenger flow within the line operation time; H is the total number of time granularities divided within the line operation time.

[0026] Furthermore, the S4 includes:

[0027] S401: Based on the two section passenger flow indices and the two station passenger flow indices selected in S3, two data sets T and T' are constructed, where T = {t1, t2, ..., t H}, T'={t'1,t'2,...,t' H}, t e =(x e ,y e ), t e '=(d e ,n e ), (x e ,y e ) represents the data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, (d e ,n e ) represents a data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, e = 1, 2, ..., H, H is the total number of time granularities divided within the line operation time;

[0028] S402: Perform FCM fuzzy clustering on T and defuzzify it, using the time dividing points of each category in the defuzzification result as the dividing boundary points of the section operation period to obtain the divided section operation period;

[0029] S403: Perform FCM fuzzy clustering on T′ and defuzzify it, and use the time dividing points of each category in the defuzzification result as the dividing boundary points of the site operation period to obtain the divided site operation period.

[0030] Furthermore, the results of fuzzy clustering are defuzzified according to the maximum membership principle.

[0031] Furthermore, in S5, the principles for merging the section operation period and the site operation period are as follows:

[0032]

[0033] in, The starting time of the line operation period; The start time of the site operation period; The starting time of the section operation period; The end time of the line operation period; The end time of the site operation period; The end time of the section operation period.

[0034] Furthermore, a rail transit operation period division device includes:

[0035] The data collection module is used to collect historical data of urban rail transit during the line operation period, then divide the historical data into the set time granularity and obtain the target data set;

[0036] The calculation module calculates the passenger flow analysis indicators within each time granularity based on the target data set obtained by the data acquisition module. The passenger flow analysis indicators include cross-section passenger flow indicators and station passenger flow indicators.

[0037] The analysis module performs correlation analysis on the passenger flow analysis indicators obtained in the calculation module, and then selects the two section passenger flow indicators and two station passenger flow indicators with the highest correlation coefficients;

[0038] The clustering module performs FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in the analysis module, defuzzifies the local results, and then divides the section operation period and station operation period;

[0039] The merging module merges the section operation period and station operation period divided by the clustering module to obtain the final line operation period division result.

[0040] Furthermore, a computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for dividing rail transit operating time periods as described in any one of claims 1 to 7.

[0041] Compared with the prior art, the present invention has the following significant advantages:

[0042] 1. Compared with the previous time period division method that directly uses passenger flow as a descriptive variable, this method divides time periods based on multiple passenger flow characteristic indicators, takes into account the volatility of other factors affecting passenger flow, and uses the Pearson correlation analysis method to extract passenger flow characteristic indicators with high correlation. This can better reflect the actual changes in passenger flow and the changing characteristics of passenger flow in different time periods;

[0043] 2. The model uses the FCM fuzzy clustering algorithm, which has a certain degree of flexibility. It avoids the strict classification of each object into each class like the hard-partitioned K-means clustering. When solving the partitioning scheme, it focuses on the similarity between different time periods, making the operation period division scheme more in line with actual operation conditions.

[0044] 3. By comprehensively considering both station and cross-section passenger flows, grasping the external distribution and internal flow of passenger flows, and dividing station and cross-section operating periods respectively, this will help subway operators more effectively address congestion during peak passenger flow periods and has strong practical significance.

[0045] 4. Based on the passenger flow characteristics of rail transit, the operating hours are divided into four categories: flat peak, transition period, peak period, and super peak period. This is more matching and adaptable to the passenger flow characteristics of different time periods, especially peak period, making the deployment of transport capacity more precise, reducing resource waste, and facilitating the formulation of train operation plans that are more in line with actual operations, which can alleviate traffic congestion and improve operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic diagram of the steps of a method for dividing rail transit operating time periods based on FCM clustering described in the present invention.

[0047] Figure 2 It is a schematic diagram of the division of rail transit into four operating periods in the present invention.

[0048] Figure 3 It is a flow chart of the FCM clustering algorithm used in the present invention.

[0049] Figure 4 It is a schematic diagram of the actual clustering result of a site embodiment in the present invention.

[0050] Figure 5 It is a schematic diagram of the actual clustering result of a cross-section interval embodiment in the present invention.

[0051] Figure 6 This is the clustering effect diagram of Tianruncheng Station in the present invention.

[0052] Figure 7 This is the clustering effect diagram of Liuzhou East Road Station in the present invention. DETAILED DESCRIPTION

[0053] In order to make the technical means, creative features, objectives and effects of the present invention easier to understand, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0054] like Figure 1 As shown, the present invention provides a method for dividing rail transit operation periods based on the FCM clustering algorithm, which adopts the following steps:

[0055] Step S1: Collect historical data of urban rail transit during the line operation time, and divide the original data into N minutes as the time granularity to obtain the target data set;

[0056] Step S2: Based on the target data set obtained in S1, calculate the passenger flow analysis indicators within each time granularity, where the passenger flow analysis indicators include two categories: cross-sectional passenger flow indicators and station passenger flow indicators. The cross-sectional passenger flow indicators include cross-sectional passenger flow, cross-sectional full load factor, and passenger standing density. The station passenger flow indicators include passenger volume, passenger arrival rate, and passenger flow imbalance coefficient.

[0057] Step S3: Calculate the correlation coefficients between the passenger flow analysis indicators in S2, and then select the two section passenger flow indicators and two station passenger flow indicators with the highest correlation;

[0058] Step S4: Perform FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in S3, and fuzzify the clustering results to divide the section operation period and station operation period;

[0059] Step S5: Merge the section operation period and the station operation period obtained in step S4 to obtain the division result of the line operation period.

[0060] The method was validated using the Nanjing Metro Line 3 in its upward direction. This line, which officially began operations in March 2015, has a relatively mature passenger flow pattern, and its passenger flow data can well reflect passenger flow patterns. Nanjing Metro Line 3 utilizes wide-body, drum-shaped A-type trains with a train set size g of six cars, a train car length L of 22.8 meters, and a train car width B of 3.2 meters. The seating area calculation includes the cross-sectional width of the seats (0.45 meters) and the area in front of the seats (0.2 meters). Therefore, the total cross-sectional width b of the car seating area is calculated as 0.7 meters, and the number of seats per car (S) is 56. Due to the space occupied by other onboard facilities, and considering the impact of uneven standing patterns and design margins, the car and train conversion factors δ and φ were set at 0.85. Daily passenger transaction data for Nanjing Metro Line 3 from June 10 to 14, 2019, covering five consecutive working days, was selected as the baseline data for this example.

[0061] The raw data required for the rail transit operation period division method include operation data, cross-section passenger flow data, and station data. By organizing and cleaning the raw data, and performing standardization preprocessing on the raw data, and dividing the raw data into 15-minute time granularity, the target data sets are obtained: cross-section passenger flow data and station passenger flow data. Then, the obtained target data sets are subjected to standardization preprocessing. The standardization formula is as follows:

[0062] Where Z r is the original data, X r The data are standardized.

[0063] When cleaning data, some sample data that deviate significantly from the rest of the values ​​are treated as outliers, and the outliers are set as missing values ​​and then removed.

[0064] The present invention mainly determines which time period a passenger's trip belongs to based on the incoming passenger flow, so based on the processed raw data, it then starts to calculate the passenger flow analysis indicators. The passenger flow analysis indicators in the present invention include: cross-sectional passenger flow indicators and station passenger flow indicators. Cross-sectional passenger flow indicators include cross-sectional passenger flow, cross-sectional full load rate and passenger standing seat density. Station passenger flow indicators include passenger volume, passenger flow arrival rate and passenger flow imbalance coefficient. The value of passenger volume in the passenger flow characteristic indicators adopts the incoming passenger flow, and the cross-sectional passenger flow can be obtained through data collection. Then the remaining passenger flow characteristic indicators are calculated by the following formula: cross-sectional full load rate, passenger standing seat density, passenger flow arrival rate, and passenger flow imbalance coefficient:

[0065] Section load rate:

[0066] Among them, α is the section full load rate; Q 断面 is the cross-section passenger flow within the unit time granularity; g is the number of train sets; P is the number of passengers in the train;

[0067] Passenger standing density:

[0068]

[0069] Where ρ is the passenger standing density; q is the number of passengers on the train; δ is the carriage conversion factor; g is the number of train cars; φ is the train conversion factor; S is the number of seats on the train; L is the length of the carriage; B is the width of the carriage; b is the total cross-sectional width of the carriage seating area;

[0070] Passenger arrival rate:

[0071]

[0072] Among them, λ is the passenger flow arrival rate by time; Q eQ is the passenger flow entering the station within the e-th time granularity during the line operation time; e+1 The passenger flow entering the station within the e+1th time granularity during the line operation time;

[0073] Passenger flow imbalance coefficient:

[0074]

[0075] Among them, β is the passenger flow imbalance coefficient; V e is the maximum passenger flow within the e-th time granularity during the line operation time; V max is the maximum passenger flow within the line operation time; H is the total number of time granularities divided within the operation time.

[0076] Based on the calculated values ​​of the corresponding multivariate passenger flow characteristic indicators, the Pearson correlation analysis method is used to analyze the correlation between the characteristic indicator values ​​in different time periods.

[0077]

[0078] Note: ** indicates significant correlation at the correlation coefficient < 0.01 level

[0079] Table 1 Correlation analysis between passenger flow characteristic indicators and passenger flow

[0080] According to the two section passenger flow indicators and two station passenger flow indicators screened in S3, two data sets T, T' are constructed, where T = {t1, t2, ..., t H}, T'={t'1,t'2,...,t' H}, t e =(x e ,y e ), t e '=(d e ,n e ), (x e ,y e ) represents the data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, (d e ,n e ) represents a data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, e=1,2,…,H, H is the total number of time granularities divided within the line operation time.

[0081] Perform FCM fuzzy clustering on T and T' respectively, such as Figure 3 As shown in Figure 2, based on the characteristics of rail transit passenger flow distribution, the number of clusters c is set to 7, and the fuzzy parameter m is determined to be 4. The fuzzy clustering analysis based on the objective function can be expressed as:

[0082]

[0083]

[0084] According to the FCM fuzzy clustering algorithm process:

[0085] 1. Randomly initialize the matrix U = (u ik ) c*96 ;

[0086] 2. Calculate the class center c through the iterative formula j :

[0087]

[0088] 3. Update the membership matrix U through the iterative formula:

[0089]

[0090] 4. If μ ij When the infinite norm of the change is less than the set threshold, the membership matrix U and clustering results (such as Figure 4 and Figure 5 As shown), and terminate the algorithm, otherwise jump to step 2;

[0091] like Figure 6 and Figure 7 As shown in the figure, the information of Tianruncheng Station and Liuzhou East Road Station is clustered by FCM fuzzy clustering for stations and sections respectively.

[0092] In practical applications, the results of fuzzy clustering can be defuzzified according to certain rules to obtain deterministic classification. Generally, the maximum membership principle is adopted, that is: if Then sample t k Belonging to Category I, the station operation period division and section operation period division are shown in Table 2 and Table 3 respectively.

[0093] Table 2 FCM fuzzy clustering results (sites)

[0094]

[0095] Table 3 FCM fuzzy clustering results (section: Tianruncheng-Liuzhou East Road)

[0096]

[0097] The results of the classification of individual points according to the maximum membership principle do not conform to the actual operation. For example, the traffic flow at Liuzhou East Road Station at 6:45 is classified into Class II and Class V with membership of 0.40132 and 0.40198. According to the maximum membership principle, it should be classified as Class V, but in fact there is not much difference between the two values. The actual operation of the line should be combined with the passenger flow status before and after, and the classification of the passenger flow index characteristic value of the point should be corrected. For the isolated points in the clustering results, the operating period is classified using the constraint conditions, that is, if they are located in the interval Moment within If the passenger flow re-enters the operating state within 15 minutes after exiting the previous operating period, it is considered The passenger flow during the period has always been within the operating period.

[0098] Combined with the actual operation of Line 3, the clustering results of 29 stations and 28 sections along the entire line are summarized, and the weekday operation period division scheme of stations and sections of Line 3 is given as shown in Table 4 and Figure 2 Obviously, the operating period obtained by combining the actual operating conditions can better reflect the objective actual situation.

[0099] Table 4: Division of working hours on weekdays

[0100]

[0101] Since the passenger flow of a one-way section of rail transit is generated by the superposition of passengers departing from each station, the start time of the station period should be earlier than the start time of the section period, and the end time of the station period should be later than the end time of the section period. Therefore, the line operation period should be the union of the line operation period and the station operation period, that is: in, The starting time of the line operation period; The start time of the site operation period; The starting time of the section operation period; The end time of the line operation period; The end time of the site operation period; The end time of the section operation period. The line operation period division scheme of Nanjing Metro Line 3 is shown in Table 5.

[0102] Table 5 Line operation period division plan

[0103]

[0104] The present invention comprehensively analyzes the line operation period from two aspects: section and station operation period, and uses Pearson correlation analysis to extract characteristic indicators, which can more effectively reflect the actual change pattern of passenger flow and reflect the changing characteristics of passenger flow in different time periods. The FCM fuzzy clustering algorithm targets the correlation between different time periods, which can ensure that the operation period division scheme is more in line with the actual operation situation. At the same time, the division of four time periods into flat peak, transition section, peak and super peak is more compatible with passenger flow characteristics, which improves operational efficiency while reducing resource waste and making the deployment of transport capacity more precise. The division of station operation period and section operation period is conducive to the subway operation to solve the congestion problem during peak passenger flow period in a more targeted manner.

[0105] This method facilitates passenger flow control from a temporal perspective, such as adjusting the start and end times of commuter traffic at urban rail stations to achieve staggered travel times for pedestrians along different routes. Applying this method to the automatic division of rail transit operating hours overcomes the irrationality of manual time division and avoids the clustering errors in the transitional segments between peak and flat periods caused by the hard division of traditional clustering methods. The proposed division method can accurately divide operating hours, laying the foundation for further predicting peak-period passenger flow and developing operational scheduling plans with high punctuality rates.

[0106] The present application also provides a rail transit operation period division device, comprising:

[0107] The data collection module is used to collect historical data of urban rail transit during the line operation period, then divide the historical data into the set time granularity and obtain the target data set;

[0108] The calculation module calculates the passenger flow analysis indicators within each time granularity based on the target data set obtained by the data acquisition module. The passenger flow analysis indicators include cross-section passenger flow indicators and station passenger flow indicators.

[0109] The analysis module performs correlation analysis on the passenger flow analysis indicators obtained in the calculation module, and then selects the two section passenger flow indicators and two station passenger flow indicators with the highest correlation coefficients;

[0110] The clustering module performs FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in the analysis module, defuzzifies the clustering results, and then divides the section operation period and station operation period;

[0111] The merging module merges the section operation period and station operation period divided by the clustering module to obtain the final line operation period division result.

[0112] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for dividing rail transit operating periods. The computer-readable storage medium may include: a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, among other media capable of storing program code.

[0113] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0114] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for dividing rail transit operation periods based on the FCM clustering algorithm, characterized by: Use the following steps: Step S1: Collect historical data of urban rail transit during the line operation time, and divide the original data into N minutes as the time granularity to obtain the target data set; Step S2: Based on the target data set obtained in S1, calculate the passenger flow analysis indicators within each time granularity, where the passenger flow analysis indicators include two categories: cross-sectional passenger flow indicators and station passenger flow indicators. The cross-sectional passenger flow indicators include cross-sectional passenger flow, cross-sectional full load factor, and passenger standing density. The station passenger flow indicators include passenger volume, passenger arrival rate, and passenger flow imbalance coefficient. Step S3: Calculate the correlation coefficients between the passenger flow analysis indicators in S2, and then select the two section passenger flow indicators and two station passenger flow indicators with the highest correlation coefficients; Step S4: Perform FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in S3, and fuzzify the clustering results to divide the section operation period and station operation period; Step S5: Merge the section operation period and the station operation period obtained in step S4 to obtain the division result of the line operation period; The S4 includes: S401: Based on the two section passenger flow indices and the two station passenger flow indices selected in S3, two data sets T and T' are constructed, where T = {t1, t2, ..., t H }, T'={t'1,t'2,...,t' H }, t e =(x e ,y e ), t e '=(d e ,n e ), (x e ,y e ) represents the data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, (d e ,n e ) represents a data pair consisting of the calculated values ​​of the passenger flow indicators of the two stations selected in S3 within the e-th time granularity, e = 1, 2, ..., H, H is the total number of time granularities divided within the line operation time; S402: Perform FCM fuzzy clustering on T and defuzzify it, using the time dividing points of each category in the defuzzification result as the dividing boundary points of the section operation period to obtain the divided section operation period; S403: Perform FCM fuzzy clustering on T′ and defuzzify it. Use the time dividing points of each category in the defuzzification result as the boundary points for dividing the site operation period to obtain the divided site operation period; In S5, the principles for merging the section operation period and the station operation period are as follows: in, The starting time of the line operation period; The start time of the site operation period; The starting time of the section operation period; The end time of the line operation period; The end time of the site operation period; The end time of the section operation period.

2. The method for dividing rail transit operation periods based on the FCM clustering algorithm according to claim 1 is characterized by: The historical data includes operation data, cross-sectional passenger flow and station data. Operation data includes the number of train vehicles, carriage length, carriage width, total width of the cross-sectional area of ​​the carriage seating area, number of seats in the car. Station data includes passenger flow entering the station and rail transit OD data.

3. The method for dividing rail transit operation periods based on the FCM clustering algorithm according to claim 1 is characterized in that: The S1 also includes data cleaning of historical data and standardization of target data sets.

4. The method for dividing rail transit operation periods based on the FCM clustering algorithm according to claim 2 is characterized in that: In S2: The calculation method of section full load factor α is as follows: Where, α is the cross-section full load factor; Q 断面 is the passenger flow in the section within the unit time granularity; g is the number of train sets; p is the number of passengers in the train; The formula for calculating passenger standing density is as follows: Where, ρ is the passenger standing density; q is the number of passengers on the train; δ is the carriage conversion factor; g is the number of train vehicles; φ is the train conversion factor; S is the number of seats on the train; L is the length of the carriage; B is the width of the carriage; b is the total cross-sectional width of the carriage seating area; The calculation formula for passenger arrival rate is as follows: Where, λ is the passenger arrival rate by time; Q e is the passenger flow in the e-th time granularity during the line operation time, Q e+1 The passenger flow entering the station within the e+1th time granularity during the line operation time; The imbalance coefficient of passenger flow characteristic index is calculated based on the passenger flow OD data in S1. The calculation formula is as follows: Where, β i is the passenger flow imbalance coefficient; V e is the maximum passenger flow in the e-th time granularity during the line operation time; V max is the maximum passenger flow within the line operation time; H is the total number of time granularities divided within the line operation time.

5. The method for dividing rail transit operation periods based on the FCM clustering algorithm according to claim 1 is characterized in that: The results of fuzzy clustering are defuzzified according to the maximum membership principle.

6. A rail transit operation period division device, the device adopts the rail transit operation period division method according to any one of claims 1 to 5 to implement operation period division, characterized in that: include: The data collection module is used to collect historical data of urban rail transit during the line operation period, then divide the historical data into the set time granularity and obtain the target data set; The calculation module calculates the passenger flow analysis indicators within each time granularity based on the target data set obtained by the data acquisition module. The passenger flow analysis indicators include cross-section passenger flow indicators and station passenger flow indicators. The analysis module performs correlation analysis on the passenger flow analysis indicators obtained in the calculation module, and then selects the two section passenger flow indicators and two station passenger flow indicators with the highest correlation coefficients; The clustering module performs FCM fuzzy clustering on the section passenger flow indicators and station passenger flow indicators selected in the analysis module, defuzzifies the clustering results, and then divides the section operation period and station operation period; The merging module merges the section operation period and station operation period divided by the clustering module to obtain the final line operation period division result.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for dividing rail transit operating periods as described in any one of claims 1 to 5.

Images