Rail transit platform passenger flow prediction method and system, computer equipment and storage medium

Abstract

The invention provides a rail transit platform passenger flow prediction method and system, computer equipment and a storage medium. The method comprises the following steps: extracting directed OD pairs from entry and exit records of an automatic fare collection system based on a completely directed network; determining travel time of non-transfer and one-time transfer; determining the real-time position of the passenger according to the directed OD pair and the travel time; according to the real-time position of the passenger, determining the passenger flow volume of each station in the target time period; acquiring and processing weather factors; a CNN-BiLSTM-Attention passenger flow prediction model is constructed; and taking the passenger flow volume of each station in the target time period and the weather factors as input variables, and predicting the passenger flow volume of each station by using the CNN-BiLSTM-Attention passenger flow prediction model. According to the method, the influence of weather factors, time dependence and spatial features is considered, the attention on key features is enhanced, and the prediction precision of the platform passenger flow is improved.

Description

Technical Field

[0001] The present invention belongs to the field of rail transit technology, and in particular relates to a rail transit platform passenger flow prediction method, system, computer equipment and storage medium. Background Art

[0002] In recent years, urban rail transit has developed rapidly, and more and more cities have entered a new era of networked operations. Networked operations of urban rail transit have brought convenience to urban residents' travel, but as the scale of urban rail transit increases, the tasks it undertakes in urban transportation become heavier, and the passenger flow of rail transit in different scenarios becomes increasingly difficult to grasp.

[0003] At present, the passenger flow of the rail transit line is summarized by the number of people entering and leaving the station to obtain the passenger flow of the station. The passenger flow data is registered by swiping the card at the gate, and each transaction is used as a record to accumulate the number of people entering or leaving the station. For the transfer passenger flow between lines, the transfer passenger flow is calculated through the clearing algorithm, combined with factors such as the shortest path, cost, and clearing ratio. Since the number of passenger flow transaction records is usually very large and the data collection frequency is about 5 to 15 minutes, it is usually calculated uniformly after the rail transit stops operating every day to generate the passenger flow statistics for the day or history.

[0004] However, the existing passenger flow analysis is used for financial clearing and accounting of ticket revenue, and the statistical dimension is to conduct passenger flow data statistics based on the reception time after the gate machine collects data in a cycle. There is a certain gap between the passenger flow based on the reception time and the actual passenger flow in the line network, and it cannot truly reflect the dynamic distribution of the real-time passenger flow in the line network. The transfer passenger flow calculated by the clearing algorithm usually only counts the transfer passenger flow between lines, and it is impossible to accurately obtain the passenger flow of a certain line at each station in different directions from the transfer station to the transfer line. Since the passenger flow collection frequency is 5 minutes as the shortest interval, and tens of thousands of data will be generated within 5 minutes, the calculation scale of the transfer passenger flow divided into each destination is extremely large. Without real-time and accurate passenger flow analysis results, it is difficult to determine the passenger flow status distributed in each section of the line network. The results calculated according to the existing method will have a certain delay with the actual passenger flow distribution, so the passenger transportation organization measures formulated are difficult to achieve the expected effect. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method, system, computer equipment and storage medium for predicting passenger flow at a rail transit platform.

[0006] In a first aspect, the present invention provides a method for predicting passenger flow at a rail transit platform, comprising:

[0007] Based on the fully directed network, the directed OD pairs are extracted from the entry and exit records of the automatic ticket vending and checking system;

[0008] Determine the journey time without transfer and with one transfer; the journey time includes the time it takes for passengers to walk to the platform, the average waiting time without transfer, the time the train stays at the platform, the average walking time with transfer, the average waiting time with transfer, and the time it takes to walk from the platform to the exit gate;

[0009] Determine the real-time location of passengers based on directed OD pairs and journey time;

[0010] Determine the passenger flow of each platform within the target time period based on the real-time location of passengers;

[0011] Acquire and process weather factors; weather factors include wind speed, temperature and precipitation;

[0012] Build a CNN-BiLSTM-Attention passenger flow prediction model;

[0013] The passenger flow and weather factors of each platform during the target time period are taken as input variables, and the CNN-BiLSTM-Attention passenger flow prediction model is used to predict the passenger flow of each platform.

[0014] Optionally, the step of extracting directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network includes:

[0015] The single card swipe record collected by the automatic ticket vending and checking system is converted into an undirected OD pair, and the passenger records are grouped by ID number. The passenger records are sorted by time sequence and entry and exit status within the group, with entry first and exit later. The entire list is moved up one row and joined with the original list. Records with different ID numbers or mismatched entry and exit status are deleted to extract the station-based OD pairs within a day; each OD pair contains the passenger's departure station number and destination station number;

[0016] Renumber the stations based on the fully directed network in the format of "line + station number";

[0017] A completely directed subway network topology is constructed based on the records of transfers and non-transfers; where the OD pairs are on the same line, it is non-transfer, and the OD pairs are on different lines, it is transfer;

[0018] Obtain the number of stations and platform numbers that the non-transfer record passes through and determine the shortest path between OD pairs according to the topological structure of the completely directed subway network to obtain the first shortest path;

[0019] Obtain the number of stations and platform numbers passed by the transfer record and determine the shortest path between OD pairs based on the completely directed subway network topology with the goal of minimizing the number of transfers, and obtain the second shortest path;

[0020] Slice the stations that the transfer record passes through. Stations with the same route belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs;

[0021] Correct the non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, subtract the sequence number of the departure platform number in the vector from the sequence number of the terminal platform number in the vector, if the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their opposite numbers to obtain a directed OD pair.

[0022] Optionally, determining the journey time without transfer and with one transfer includes:

[0023] Calculate the non-transfer journey time using the following formula:

[0024]

[0025] in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations that the train passes; s is the index of the station that passenger k passes on line l;

[0026] Calculate the journey time with transfers according to the following formula:

[0027]

[0028] Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

[0029] Optionally, the passenger flow and weather factors of each platform in the target time period are used as input variables, and the CNN-BiLSTM-Attention passenger flow prediction model is used to predict the passenger flow of each platform, including:

[0030] Get the passenger flow of the transfer platform within the target time period is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series;

[0031] The wind speed, temperature and precipitation are interpolated to make them maintain the same time series length as the passenger flow in the corresponding observation period. t ={m 1 ,m 2 ,…,m n}, the temperature is C t ={c 1 ,c 2 ,…,c n}, precipitation is I t ={i 1 ,i 2 ,…,i n};m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window;

[0032] X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0033] In a second aspect, the present invention provides a rail transit platform passenger flow prediction system, comprising:

[0034] An extraction module, used for extracting directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network;

[0035] The first determination module is used to determine the journey time without transfer and with one transfer; wherein the journey time includes the time for passengers to walk to the platform, the average waiting time without transfer, the train stay time at the platform, the average walking time for transfer, the average waiting time for transfer and the time to walk from the platform to the exit gate;

[0036] A second determination module is used to determine the real-time location of the passenger based on the directed OD pair and the journey time;

[0037] The third determination module is used to determine the passenger flow of each platform within the target time period according to the real-time location of the passengers;

[0038] An acquisition module is used to acquire and process weather factors; wherein the weather factors include wind speed, temperature and precipitation;

[0039] Construction module, used to build CNN-BiLSTM-Attention passenger flow prediction model;

[0040] The prediction module is used to take the passenger flow and weather factors of each platform in the target time period as input variables, and use the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0041] Optionally, the extraction module includes:

[0042] The conversion unit is used to convert a single card swiping record collected by the automatic ticket vending and checking system into an undirected OD pair, group the passenger records by ID number, sort them by time sequence and entry and exit status within the group, first enter the station and then exit the station, move the entire list up one row and splice it with the original list, delete the records with different ID numbers or mismatched entry and exit status, so as to extract the OD pairs based on the station within a day; each OD pair contains the passenger's departure station number and destination station number;

[0043] Numbering unit, used to renumber stations based on a fully directed network, in the format of "line + station number";

[0044] A construction unit is used to construct a completely directed subway network topology structure according to the records of transfer and non-transfer; wherein, if the lines of the OD pairs are the same, it is non-transfer, and if the lines of the OD pairs are different, it is transfer;

[0045] A first acquisition unit is used to acquire the number of stations and platform numbers that the non-transfer record passes through and determine the shortest path between OD pairs according to the completely directed subway network topology structure to obtain a first shortest path;

[0046] The second acquisition unit is used to obtain the number of stations and platform numbers passed by the transfer record and determine the shortest path between OD pairs based on the completely directed subway network topology structure with the goal of minimizing the number of transfers to obtain a second shortest path;

[0047] The station slicing unit is used to slice the stations that the transfer record passes through. Stations with the same line belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs;

[0048] The extraction unit is used to correct non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, and subtract the serial number of the departure platform number in the vector from the serial number of the terminal platform number in the vector. If the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their respective opposite numbers to obtain a directed OD pair.

[0049] Optionally, the first determining module includes:

[0050] The first calculation unit is used to calculate the non-transfer journey time according to the following formula:

[0051]

[0052] in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations that the train passes; s is the index of the station that passenger k passes on line l;

[0053] The second calculation unit is used to calculate the transfer journey time according to the following formula:

[0054]

[0055] Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

[0056] Optionally, the prediction module includes:

[0057] The third acquisition unit is used to obtain the passenger flow of the transfer platform within the target time period is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series;

[0058] The interpolation unit is used to interpolate the wind speed, temperature and precipitation so that the wind speed, temperature and precipitation maintain the same time series length as the passenger flow in the corresponding observation time period; where the wind speed is M t ={m 1 ,m 2 ,…,m n}, the temperature is C t ={c 1 ,c 2 ,…,c n}, precipitation is I t ={i 1 ,i 2 ,…,i n};m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window;

[0059] Prediction unit, used to convert X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0060] In a third aspect, the present invention provides a computer device comprising a processor and a memory; wherein, when the processor executes a computer program stored in the memory, the steps of the rail transit platform passenger flow prediction method described in the first aspect are implemented.

[0061] In a fourth aspect, the present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the rail transit platform passenger flow prediction method described in the first aspect are implemented.

[0062] The present invention provides a method, system, computer device and storage medium for predicting passenger flow at a rail transit platform, wherein the method uses a fully bidirectional subway network model and data from a subway ticket sales and inspection system to calculate the specific passenger flow at the platform within a target time period; and constructs a typical architecture that considers the influence of weather factors, time dependence and spatial features. The CNN-BILSTM-Attention model combines the spatial feature extraction capability of a convolutional neural network with the sequence processing capability of a bidirectional long short-term memory network. By combining the attention mechanism, the model enhances the focus on key features and improves the accuracy of platform passenger flow prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 A schematic diagram of a flow chart of a method for predicting passenger flow at a rail transit platform provided by an embodiment of the present invention;

[0065] Figure 2 A time composition diagram of transfer and non-transfer journeys provided by an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of the structure of the CNN-BiLSTM-Attention passenger flow prediction model provided in an embodiment of the present invention;

[0067] Figure 4 A schematic structural diagram of a rail transit platform passenger flow prediction system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] Example 1

[0070] like Figure 1 As shown, this embodiment provides a rail transit platform passenger flow prediction method, including:

[0071] Step 101: extract directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network.

[0072] The extraction of directed OD pairs is actually the correction process of the start and end numbers of the automatic fare collection system (AFCS). The entry or exit record includes the passenger's identification number, timestamp (entry time or exit time), station number (entry number or exit number), status (entry or exit), and payment method. The OD pair represents a travel record including the departure and destination. To convert a single record into a station-based OD pair, first group the records by ID number, ensuring that records belonging to the same passenger are placed together. In each group, sort them by time order and status, ensuring that the entry record is in front and the exit record is in the back. Then, move the sorted record list up one row and connect it with the original list so that each exit record is preceded by a corresponding entry record. Delete records with different ID numbers or mismatched entry and exit status to extract station-based OD pairs within a day, each of which contains the passenger's departure and destination.

[0073] After converting the records into station-based OD pairs, it is necessary to determine whether each OD pair involves transfers based on the topology of the subway network. If the origin and destination stations are on the same route, it will be regarded as non-transfer; if they are different, they will be regarded as transfers. In addition, considering that passengers usually prefer trips with relatively few transfers, a network topology is constructed to minimize the number of transfers. The weight of the transfer station is set to a low value (such as 1), while the weight between connecting platforms is set to a high value (such as 10). After preliminary processing of traffic data, the station-based OD pairs are then converted into platform-based OD pairs. The platform is the actual location where passengers get on and off, which more accurately reflects their travel trajectory. For OD pairs involving transfers, it is necessary to divide them into several non-transfer trips. Each segmented trip involves only one route, and the segmented trips can be used for more detailed passenger flow analysis.

[0074] For non-transfer trips, all stations on the route can be sorted according to the natural direction of train travel (i.e., upstream direction) to form an ordered vector. Each station has a unique position index that represents the ordinal position of the station on the route. The position index of the starting point is subtracted from the position index of the end point to determine the direction of the trip. If the result is positive, the train will travel upward; otherwise, the train will travel in the downward direction. The departure and arrival platforms of the passengers are also determined. Use "+" to indicate the upward direction and "-" to indicate the downward direction. For example, if the starting point is +107, it means that the passenger departs from the platform in the upstream direction of Station 7 on Line 1.

[0075] The transfer trip is divided into multiple non-transfer trips according to the passenger's entry and exit records and the subway network topology. Then, for each segmented trip, the up and down directions are determined according to the same method of the non-transfer trip. It should be noted that for the special case where the starting point is a transfer station, the end point is a transfer station, or the starting point and the end point are both transfer stations, the specific platform for entering and exiting the station should be determined. For example, if a passenger swipes a card from the No. 5 transfer station on Line 1 and Line 4, according to the numbering rule, the station includes No. 105 and No. 405. Assume that the passenger's entry station is 105. When the transfer trip is divided into non-transfer trips, if the passenger does have 105 people entering the station to board, then 105 people will be divided into trips with a length greater than 2. On the contrary, if the length is 1, only 105 is included, which is an invalid ride behavior. At this time, the trip with a length of 1 is deleted, and the first number of the first trip is redefined as the starting point to confirm the specific station where the passenger departs. Then, the specific departure platform is determined by judging the direction, whether it is +105, -105, +405 or -405. The same applies when the destination is a transfer station or the starting point and the destination are both transfer stations.

[0076] Exemplarily, this step includes:

[0077] The single card swiping record collected by the automatic ticket vending and checking system is converted into an undirected OD pair, and the passenger records are grouped by ID number. The group is sorted by time sequence and entry and exit status, with entry first and exit later. The entire list is moved up one row and joined with the original list, and records with different ID numbers or mismatched entry and exit status are deleted to extract station-based OD pairs within a day; each OD pair contains the passenger's departure station number and destination station number.

[0078] The stations are renumbered based on the fully directed network in the format of "line + station number".

[0079] A completely directed subway network topology is constructed based on the records of transfers and non-transfers; among them, if the OD pairs are on the same line, it is non-transfer, and if the OD pairs are on different lines, it is transfer.

[0080] The number of stations and platform numbers passed by the non-transfer records are obtained, and the shortest path between OD pairs is determined according to the topological structure of the completely directed subway network to obtain the first shortest path.

[0081] The number of stations and platform numbers passed by the transfer records are obtained and based on the completely directed subway network topology, the shortest path between OD pairs is determined with the goal of minimizing the number of transfers, and the second shortest path is obtained.

[0082] Slice the stations that the transfer record passes through, and the stations with the same line belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs. If the station is always a transfer station, the correct number will be included in a trip with a length greater than 2 platform numbers. If the length is 1, it is an invalid record and deleted.

[0083] Correct the non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, subtract the sequence number of the departure platform number in the vector from the sequence number of the terminal platform number in the vector, if the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their opposite numbers to obtain a directed OD pair.

[0084] Step 102, determining the travel time without transfer and with one transfer; wherein the travel time includes the time it takes for passengers to walk to the platform, the average waiting time without transfer, the train's stay time at the platform, the average walking time with transfer, the average waiting time with transfer, and the time it takes to walk from the platform to the exit gate.

[0085] In order to identify the real-time location of passengers during their journey, in addition to extracting directed OD pairs, it is also necessary to analyze the time spent by passengers at each stage of their journey. The time composition of transfer and non-transfer journeys is as follows: Figure 2 As shown, for non-transfer, the journey process can be divided into walking to the platform, waiting, boarding and walking out of the station.

[0086] For transfer journeys, there are more transfer processes compared to non-transfer journeys, which include the time of walking, waiting and riding.

[0087] Exemplarily, the non-transfer journey time is calculated according to the following formula:

[0088]

[0089] in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations that the train passes; s is the index of the station that passenger k passes on line l;

[0090] Calculate the journey time with transfers according to the following formula:

[0091]

[0092] Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

[0093] This embodiment is to establish a general model to capture the movement of passengers in the station to count the actual number of platform passengers in a certain time interval, assuming the following points:

[0094] 1) Assume that passengers do not make irrational round trips;

[0095] 2) Passengers have the same walking time entering and exiting the station;

[0096] 3) The waiting time of the passenger with the shortest travel time between any stations is 0;

[0097] For the shortest non-transfer journey, assuming there is no waiting time, the non-transfer journey time is calculated according to the following formula:

[0098]

[0099] After extracting the record with the shortest travel time between any stations, the travel time ratio between stations is determined based on the straight-line distance between stations, and the number of train stops is confirmed based on the shortest travel path. Based on multiple linear regression, the walking time, the running time between stations, and the train stop time can be calculated according to the above formula. Then, for the journey record with 1 transfer and the shortest travel time between any stations, assuming that its waiting time is also 0, the walking time required for the transfer process can be calculated, and then the average waiting time for the transfer can be calculated.

[0100] Step 103, determining the real-time location of the passenger based on the directed OD pair and the journey time.

[0101] Taking passengers as the research object, the journey process is divided into stages such as walking, waiting, riding, getting off, and walking. According to the specific time taken in each stage, the approximate position of the passengers at each time point can be inferred more accurately, thereby capturing the passengers' spatial needs during the journey.

[0102] Step 104, determining the passenger flow of each platform within the target time period according to the real-time location of the passengers.

[0103] The capture of space demand in step 103 is crucial for the operation of subway or rail transit systems. For ordinary platforms, the platform demand can usually be considered as the number of passengers who enter the station directly by swiping their cards. However, when it comes to transfer stations, the situation is more complicated. The demand of transfer stations cannot be simply equated with the number of passengers who swipe their cards to enter the station, because the passenger flow from other lines to this line also needs to be considered. These transfer passengers also need to occupy platform space, so they are also an important part of the platform demand.

[0104] The number of passengers at the normal platform at the time when they start waiting is the capacity demand of the platform X p,s,l (t,d)=K p,s,l (t,d),K p,s,l (t, d) is the set of passengers waiting at platform p in station s on line l during time slot (t, d).

[0105] Passenger flow at the transfer platform during the target time period The capacity requirement of the station is the set of passengers who transfer from other platforms in station s on line l to platform p during the time slot (t, d).

[0106] Step 105, obtaining and processing weather factors; wherein the weather factors include wind speed, temperature and precipitation.

[0107] Step 106: construct a CNN-BiLSTM-Attention passenger flow prediction model.

[0108] CNN-BILSTM-Attention is a model that combines convolutional neural network (CNN), bidirectional long short-term memory neural network (BILSTM) and attention mechanism to handle natural language processing tasks such as text classification and sentiment analysis. CNN captures local patterns and features in the data, while BILSTM can handle long-term dependencies in data capture. The attention mechanism enables the model to focus on key information, thereby improving prediction accuracy. According to the characteristics of CNN, BILSTM and Attention mechanism, a passenger flow prediction model based on CNN-BILSTM-Attention is established, such as Figure 3 shown.

[0109] One-dimensional convolutional neural networks (1D CNNs) are widely used to process time series data in machine learning, especially in natural language processing (NLP), speech recognition, and time series forecasting. The model usually consists of multiple convolutional layers, activation functions, pooling layers, and fully connected layers, and is trained using the back-propagation algorithm. Convolutional neural networks for time series data have strong feature extraction capabilities, large-scale data processing capabilities, and flexibility, which means that they can identify temporal patterns and trends that are useful for the target variable. Convolutional neural networks extract features from input data by stacking multiple convolutional layers and pooling layers. Each convolutional layer contains multiple convolution kernels to capture different features. On the other hand, the pooling layer is used to reduce the dimensionality of the data, reduce the computational load, and enhance the robustness of the model.

[0110] Step 107, taking the passenger flow and weather factors of each platform in the target time period as input variables, and using the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0111] In this step, the passenger flow of the transfer platform within the target time period is obtained. is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series.

[0112] The wind speed, temperature and precipitation are interpolated to make them maintain the same time series length as the passenger flow in the corresponding observation period. t ={m 1 ,m 2 ,…,m n}, the temperature is C t ={c 1 ,c 2 ,…,c n}, precipitation is I t ={i 1 ,i 2 ,…,i n};m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window.

[0113] X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0114] In summary, this embodiment provides a method for predicting passenger flow at rail transit platforms, which uses a fully bidirectional subway network model and data from the subway ticket sales and inspection system to calculate the specific passenger flow at the platform during the target time period; and constructs a typical architecture that takes into account the influence of weather factors, time dependence, and spatial features. The CNN-BILSTM-Attention model combines the spatial feature extraction capability of a convolutional neural network with the sequence processing capability of a bidirectional long short-term memory network. By combining the attention mechanism, the model enhances the focus on key features and improves the accuracy of platform passenger flow prediction.

[0115] Example 2

[0116] Based on the same inventive concept as Example 1, this embodiment provides a rail transit platform passenger flow prediction system. Since the principle of solving the problem by this system is similar to the aforementioned rail transit platform passenger flow prediction method, the implementation of this system can refer to the implementation of the rail transit platform passenger flow prediction method.

[0117] like Figure 4 As shown, the rail transit platform passenger flow prediction system includes:

[0118] The extraction module 10 is used to extract directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network.

[0119] The first determination module 20 is used to determine the journey time without transfer and with one transfer; wherein the journey time includes the time it takes for passengers to walk to the platform, the average waiting time without transfer, the train stay time at the platform, the average walking time for transfer, the average waiting time for transfer and the time to walk from the platform to the exit gate.

[0120] The second determination module 30 is used to determine the real-time location of the passenger according to the directed OD pair and the journey time.

[0121] The third determination module 40 is used to determine the passenger flow of each platform within the target time period according to the real-time position of the passengers.

[0122] The acquisition module 50 is used to acquire and process weather factors; wherein the weather factors include wind speed, temperature and precipitation.

[0123] The construction module 60 is used to construct a CNN-BiLSTM-Attention passenger flow prediction model.

[0124] The prediction module 70 is used to take the passenger flow and weather factors of each platform in the target time period as input variables, and use the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0125] Exemplarily, the extraction module includes:

[0126] The conversion unit is used to convert a single card swiping record collected by the automatic ticket vending and checking system into an undirected OD pair, group the passenger records by ID number, sort them by time sequence and entry and exit status within the group, first enter the station and then exit the station, move the entire list up one row and splice it with the original list, delete the records with different ID numbers or mismatched entry and exit status, so as to extract the station-based OD pairs within a day; wherein each OD pair contains the passenger's departure station number and destination station number.

[0127] The numbering unit is used to renumber the stations based on the fully directed network. The format is "line + station number".

[0128] The construction unit is used to construct a completely directed subway network topology structure according to the records of transfer and non-transfer; wherein, if the lines of the OD pairs are the same, it is non-transfer, and if the lines of the OD pairs are different, it is transfer.

[0129] The first acquisition unit is used to acquire the number of stations and platform numbers that the non-transfer record passes through and determine the shortest path between OD pairs according to the completely directed subway network topology structure to obtain a first shortest path.

[0130] The second acquisition unit is used to obtain the number of stations and platform numbers passed by the transfer record and determine the shortest path between OD pairs based on the completely directed subway network topology structure with the goal of minimizing the number of transfers to obtain a second shortest path.

[0131] The station slicing unit is used to slice the stations that the transfer record passes through. Stations with the same line belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs.

[0132] The extraction unit is used to correct non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, and subtract the serial number of the departure platform number in the vector from the serial number of the terminal platform number in the vector. If the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their respective opposite numbers to obtain a directed OD pair.

[0133] Exemplarily, the first determining module includes:

[0134] The first calculation unit is used to calculate the non-transfer journey time according to the following formula:

[0135]

[0136] in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations passed by the train; s is the index of the station passed by passenger k on line l.

[0137] The second calculation unit is used to calculate the transfer journey time according to the following formula:

[0138]

[0139] Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

[0140] Exemplarily, the prediction module includes:

[0141] The third acquisition unit is used to obtain the passenger flow of the transfer platform within the target time period is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series.

[0142] The interpolation unit is used to interpolate the wind speed, temperature and precipitation so that the wind speed, temperature and precipitation maintain the same time series length as the passenger flow in the corresponding observation time period; where the wind speed is M t ={m 1 ,m 2 ,…,m n}, the temperature is C t ={c 1 ,c 2 ,…,c n}, precipitation is I t ={i 1 ,i 2 ,…,i n};m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window.

[0143] Prediction unit, used to convert X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

[0144] For more specific working processes of the above modules, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here.

[0145] Example 3

[0146] This embodiment provides a computer device, including a processor and a memory; wherein, when the processor executes a computer program stored in the memory, the steps of the rail transit platform passenger flow prediction method described in Example 1 are implemented.

[0147] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here.

[0148] Example 4

[0149] This embodiment provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the rail transit platform passenger flow prediction method described in Example 1 are implemented.

[0150] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here.

[0151] Example 5

[0152] This embodiment provides a computer program product, including computer executable instructions or a computer program. When the computer executable instructions or the computer program are executed by a processor, the steps of the rail transit platform passenger flow prediction method described in Example 1 are implemented.

[0153] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here.

[0154] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the systems, devices, storage media, and computer program products disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0155] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention or some parts of the embodiments.

[0156] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0157] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0158] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0159] The present invention has been described in detail above in conjunction with specific implementations and exemplary examples, but these descriptions cannot be understood as limiting the present invention. Those skilled in the art understand that, without departing from the spirit and scope of the present invention, a variety of equivalent substitutions, modifications or improvements may be made to the technical solution of the present invention and its implementation methods, all of which fall within the scope of the present invention. The scope of protection of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting passenger flow at a rail transit platform, characterized in that: include: Based on the fully directed network, the directed OD pairs are extracted from the entry and exit records of the automatic ticket vending and checking system; Determine the journey time without transfer and with one transfer; the journey time includes the time it takes for passengers to walk to the platform, the average waiting time without transfer, the time the train stays at the platform, the average walking time with transfer, the average waiting time with transfer, and the time it takes to walk from the platform to the exit gate; Determine the real-time location of passengers based on directed OD pairs and journey time; Determine the passenger flow of each platform within the target time period based on the real-time location of passengers; Acquire and process weather factors; weather factors include wind speed, temperature and precipitation; Build a CNN-BiLSTM-Attention passenger flow prediction model; The passenger flow and weather factors of each platform during the target time period are taken as input variables, and the CNN-BiLSTM-Attention passenger flow prediction model is used to predict the passenger flow of each platform.

2. The method for predicting passenger flow at a rail transit platform according to claim 1, characterized in that: The method of extracting directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network includes: The single card swipe record collected by the automatic ticket vending and checking system is converted into an undirected OD pair, and the passenger records are grouped by ID number. The passenger records are sorted by time sequence and entry and exit status within the group, with entry first and exit later. The entire list is moved up one row and joined with the original list. Records with different ID numbers or mismatched entry and exit status are deleted to extract the station-based OD pairs within a day; each OD pair contains the passenger's departure station number and destination station number; Renumber the stations based on the fully directed network in the format of "line + station number"; A completely directed subway network topology is constructed based on the records of transfers and non-transfers; where the OD pairs are on the same line, it is non-transfer, and the OD pairs are on different lines, it is transfer; Obtain the number of stations and platform numbers that the non-transfer record passes through and determine the shortest path between OD pairs according to the topological structure of the completely directed subway network to obtain the first shortest path; Obtain the number of stations and platform numbers passed by the transfer record and determine the shortest path between OD pairs based on the completely directed subway network topology with the goal of minimizing the number of transfers, and obtain the second shortest path; Slice the stations that the transfer record passes through. Stations with the same route belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs; Correct the non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, subtract the sequence number of the departure platform number in the vector from the sequence number of the terminal platform number in the vector, if the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their opposite numbers to obtain a directed OD pair.

3. The method for predicting passenger flow at a rail transit platform according to claim 1, characterized in that: Determining the journey time without transfer and with one transfer includes: Calculate the non-transfer journey time using the following formula: in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations that the train passes; s is the index of the station that passenger k passes on line l; Calculate the journey time with transfers according to the following formula: Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

4. The rail transit platform passenger flow prediction method according to claim 1, characterized in that: The passenger flow and weather factors of each platform in the target time period are used as input variables, and the CNN-BiLSTM-Attention passenger flow prediction model is used to predict the passenger flow of each platform, including: Get the passenger flow of the transfer platform within the target time period is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series; The wind speed, temperature and precipitation are interpolated to make them maintain the same time series length as the passenger flow in the corresponding observation period. t ={m 1 ,m 2 ,…,m n }, the temperature is C t ={c 1 ,c 2 ,…,c n }, precipitation is I t ={i 1 ,i 2 ,…,i n };m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window; X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

5. A rail transit platform passenger flow prediction system, characterized in that: include: An extraction module, used for extracting directed OD pairs from the entry and exit records of the automatic ticket vending and checking system based on a completely directed network; The first determination module is used to determine the journey time without transfer and with one transfer; wherein the journey time includes the time for passengers to walk to the platform, the average waiting time without transfer, the train stay time at the platform, the average walking time for transfer, the average waiting time for transfer and the time to walk from the platform to the exit gate; A second determination module is used to determine the real-time location of the passenger based on the directed OD pair and the journey time; The third determination module is used to determine the passenger flow of each platform within the target time period according to the real-time location of the passengers; An acquisition module is used to acquire and process weather factors; wherein the weather factors include wind speed, temperature and precipitation; Construction module, used to build CNN-BiLSTM-Attention passenger flow prediction model; The prediction module is used to take the passenger flow and weather factors of each platform in the target time period as input variables, and use the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

6. The rail transit platform passenger flow prediction system according to claim 5, characterized in that: The extraction module comprises: The conversion unit is used to convert a single card swiping record collected by the automatic ticket vending and checking system into an undirected OD pair, group the passenger records by ID number, sort them by time sequence and entry and exit status within the group, first enter the station and then exit the station, move the entire list up one row and splice it with the original list, delete the records with different ID numbers or mismatched entry and exit status, so as to extract the OD pairs based on the station within a day; each OD pair contains the passenger's departure station number and destination station number; Numbering unit, used to renumber stations based on a fully directed network, in the format of "line + station number"; A construction unit is used to construct a completely directed subway network topology structure according to the records of transfer and non-transfer; wherein, if the lines of the OD pairs are the same, it is non-transfer, and if the lines of the OD pairs are different, it is transfer; A first acquisition unit is used to acquire the number of stations and platform numbers that the non-transfer record passes through and determine the shortest path between OD pairs according to the completely directed subway network topology structure to obtain a first shortest path; The second acquisition unit is used to obtain the number of stations and platform numbers passed by the transfer record and determine the shortest path between OD pairs based on the completely directed subway network topology structure with the goal of minimizing the number of transfers to obtain a second shortest path; The station slicing unit is used to slice the stations that the transfer record passes through. Stations with the same line belong to a non-transfer sub-record, so as to divide a transfer OD pair into multiple non-transfer OD pairs; The extraction unit is used to correct non-transfer OD pairs and multiple non-transfer OD pairs with transfer into directed OD pairs, construct the subway line stations into a vector according to the subway running direction, and subtract the serial number of the departure platform number in the vector from the serial number of the terminal platform number in the vector. If the difference is positive, the terminal platform number and the departure station number remain unchanged; if the difference is negative, the terminal platform number and the departure station number are changed to their respective opposite numbers to obtain a directed OD pair.

7. The rail transit platform passenger flow prediction system according to claim 5, characterized in that: The first determining module comprises: The first calculation unit is used to calculate the non-transfer journey time according to the following formula: in, is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) without transfer; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) without transfer; is the time it takes for passenger k to walk from the gate at station s of line l to platform p or the time it takes for passenger k to walk from platform p of line l to the gate at exit station s; is the minimum waiting time for a passenger k at platform p in station s on line l without transfer; α s,l is the time it takes for a train to run from station s to station s+l; β s,l is the train’s dwelling time at station s; is the number of stations that the train passes; s is the index of the station that passenger k passes on line l; The second calculation unit is used to calculate the transfer journey time according to the following formula: Where t is the number of time slots in a day; d is the index of the day of the week; is the time when passenger k swipes his card to exit platform p at station s on line l during the time slot (t, d) in the transfer case; is the time when passenger k swipes his card to enter platform p in station s on line l during the time slot (t, d) in the transfer case; is the minimum waiting time for passenger k at platform p in station s of line l when transferring; q k is the number of passengers’ transfers; t q The average time spent on the transfer process; k is the total number of routes that passenger k has passed; i is the index of all routes that passenger k has passed.

8. The rail transit platform passenger flow prediction system according to claim 5, characterized in that: The prediction module comprises: The third acquisition unit is used to obtain the passenger flow of the transfer platform within the target time period is the passenger flow of platform p at station s on line l in the nth observation time window; n is the total length of the time series; The interpolation unit is used to interpolate the wind speed, temperature and precipitation so that the wind speed, temperature and precipitation maintain the same time series length as the passenger flow in the corresponding observation time period; where the wind speed is M t ={m 1 ,m 2 ,…,m n }, the temperature is C t ={c 1 ,c 2 ,…,c n }, precipitation is I t ={i 1 ,i 2 ,…,i n };m n is the wind speed in the nth observation time window; c n is the temperature of the nth observation time window; i n is the precipitation in the nth observation time window; Prediction unit, used to convert X p,s,l 、M t , C t and I t Input into the CNN-BiLSTM-Attention passenger flow prediction model to predict the passenger flow of each platform.

9. A computer device, characterized in that: It comprises a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the steps of the rail transit platform passenger flow prediction method described in any one of claims 1-4 are implemented.

10. A computer-readable storage medium, characterized in that: Used to store computer programs; when the computer programs are executed by the processor, the steps of the rail transit platform passenger flow prediction method described in any one of claims 1-4 are implemented.

Images