Subway hub station surrounding road congestion prediction method based on timetable
By combining subway timetables and passenger density data with the KNN clustering model, the problem of accurately predicting traffic congestion around subway hubs has been solved, enabling advance traffic management and high-precision traffic control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CASCO SIGNAL LTD
- Filing Date
- 2022-09-27
- Publication Date
- 2026-06-19
AI Technical Summary
Existing technologies struggle to accurately predict traffic congestion around subway hubs, especially during peak passenger flow periods when it's difficult to manage congestion in advance.
By combining subway timetables and passenger density data with historical data, a KNN clustering model is established to predict future passenger flow density and transfer choices, comprehensively influencing ground traffic flow, accurately predicting passenger flow to station exits, and forecasting road congestion.
It achieves high-precision prediction of traffic congestion around subway hubs, supports advance traffic management, and improves the accuracy and real-time performance of predictions.
Smart Images
Figure CN115563761B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to train signal control systems, and more particularly to a method for predicting road congestion around subway hubs based on timetables. Background Technology
[0002] As urbanization rates rise, urban traffic congestion becomes increasingly prominent, with subway hubs being a major source of congestion. These hubs typically connect two to three subway lines, complementing the existing public transportation system for convenient transfers. The need for passengers to switch to surface transportation often leads to a surge in traffic at these hubs during peak hours, resulting in road congestion.
[0003] Traffic flow is highly unpredictable, and the exact time and location of large passenger flows are often difficult to foresee. Even with current intelligent transportation systems employing traffic light phase control, vehicle-to-infrastructure communication, and filtering mechanisms, traffic management often only takes effect after congestion occurs, making it difficult to proactively manage congestion during peak hours. Rail transit, due to its stable, timetable operation, allows for precise tracking of passenger arrival times at relevant stations. This clear, effective, and stable information is a crucial tool for predicting traffic congestion in the current context. Therefore, addressing road congestion around subway hubs has become a critical technical challenge. Summary of the Invention
[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a timetable-based method for predicting road congestion around subway hubs with high accuracy.
[0005] The objective of this invention can be achieved through the following technical solutions:
[0006] According to a first aspect of the present invention, a method for predicting traffic congestion around a subway hub station based on a timetable is provided. The method first predicts the flow of people arriving at the platform at future times by using the subway timetable and the crowding level of passengers in the train; secondly, it matches the current passenger flow density at key nodes inside the platform with a density model generated from historical data to determine the passenger flow arriving at a specific exit in the future; and finally, it matches historical passenger transfer tool models and models of the impact of different modes of transportation on ground traffic flow to determine the possible traffic congestion at future times.
[0007] As a preferred technical solution, the method specifically includes the following steps:
[0008] Step S1: Train passenger flow input;
[0009] Step S2: Construct an arrival passenger flow model;
[0010] Step S3: Optimize the model constructed in step S2;
[0011] Step S4: Construct a transfer selection model;
[0012] Step S5: Predict the congestion level of different road sections near the hub station.
[0013] As a preferred technical solution, step S1, the train passenger flow input, specifically involves:
[0014] It interacts with the urban rail ATS subsystem to obtain timetables for all lines passing through this platform and the density of people inside the subway, thereby estimating the incoming passenger flow.
[0015] As a preferred technical solution, the estimated input passenger flow specifically refers to:
[0016] By obtaining the subway timetables for lines arriving at this station, we can estimate the time it will take for arriving passengers to create traffic congestion.
[0017] By analyzing pedestrian density information inside subway cars and historical passenger numbers at this station, the expected passenger flow during peak periods can be predicted.
[0018] As a preferred technical solution, step S2, constructing the arrival passenger flow model, specifically involves:
[0019] A KNN clustering model was established based on historical data, and a model of the impact of different input flows on the exit flow was constructed.
[0020] As a preferred technical solution, step S3, optimizing the model constructed in step S2, specifically involves:
[0021] Key points for passenger entry and exit are selected for passenger flow density monitoring. The model constructed in step S2 is then optimized in real time by monitoring the current passenger flow.
[0022] As a preferred technical solution, the key nodes include transfer entrances, escalators, exits, and turnstiles.
[0023] As a preferred technical solution, step S4, constructing the transfer selection model, specifically involves:
[0024] A transfer choice model for people's possible modes of transportation is established by using historical data on time, weather, season, holidays, temperature, and current congestion levels.
[0025] As a preferred technical solution, the number of passengers entering and exiting each station during the expected time period is matched with a transfer selection model to obtain the number of different modes of transportation that will eventually enter the area.
[0026] By obtaining the possible entry routes of vehicles and the current traffic conditions along these routes, the system comprehensively predicts the pressure on current road traffic and the traffic congestion caused when the vehicles leave.
[0027] As a preferred technical solution, different types of transportation are input into the congestion impact model according to their quantity, and finally the time periods when congestion may occur in the future are obtained.
[0028] As a preferred technical solution, step S5, predicting the congestion level of different road sections near the hub station, specifically involves:
[0029] By obtaining historical statistics on the impact factors of people's choice of transportation at the current entrance and exit on platform traffic flow, and by using the transfer choice model of the population and the influencing factors, the congestion level of different road sections near the hub station can be predicted.
[0030] As a preferred technical solution, the prediction results are accurate to different exits; if the exits are in the same direction of travel and there are no traffic lights between the exits, they are combined for statistical analysis.
[0031] As a preferred technical solution, the congestion level prediction of the road segment takes into account the time for ground vehicles to stop and pick up passengers, and the congestion inside the platform causes longer exit time and longer waiting time.
[0032] As a preferred technical solution, this method takes into account the bus stop conditions near the relevant exits and the arrival time of buses. Under good weather conditions, buses can effectively alleviate the pressure of passenger flow, and timely bus arrival can prevent passengers from choosing to call a taxi, thus increasing the burden on the road.
[0033] According to a second aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described thereon.
[0034] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.
[0035] Compared with the prior art, the present invention has the following advantages:
[0036] 1) This invention uses the linkage of urban rail transit and road traffic data to accurately predict the arrival of large passenger flows.
[0037] 2) This invention improves the accuracy of prediction by continuously refining the passenger flow model through linkage with urban rail transit.
[0038] 3) Compared with traditional traffic model prediction models, this invention has a clear passenger arrival time, which can be used to achieve advance diversion.
[0039] 4) This invention can dynamically adjust the model based on the current passenger flow and also has a real-time guidance function. Attached Figure Description
[0040] Figure 1 This is a flowchart of the method of the present invention;
[0041] Figure 2 This is a schematic diagram of the two-dimensional KNN method;
[0042] Figure 3 A schematic diagram illustrating the logic of the multi-dimensional KNN method;
[0043] Figure 4 This is a schematic diagram of historical data examples. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0045] This invention, based on timetables, accurately obtains the arrival time of incremental passenger flow and precisely detects the number of passengers about to arrive by detecting passenger density inside subway carriages. By monitoring stable passenger flow arriving via the subway, a correlation model is established between this flow and the degree of surface traffic congestion, thereby enabling the prediction of surface traffic congestion. This invention targets specific traffic nodes, establishing models based on subway transfer hubs. Compared to traditional traffic congestion prediction models, it features strong stability, readily available data, and flexible usage. For different transfer hubs, it can be deployed in conjunction with existing prediction models or operated independently.
[0046] This invention provides a method for predicting traffic congestion around subway hubs based on timetables. By using subway timetables and the crowding levels of passengers in trains, it predicts the flow of people arriving at the platform at future times. Combined with the current passenger flow density at key nodes within the platform, it matches a density model generated from historical data to determine the passenger flow arriving at a specific exit in the future. Simultaneously, it matches historical passenger transfer tool models and models of the impact of different modes of transportation on ground traffic flow to ultimately determine the traffic congestion that may occur at future times.
[0047] This invention obtains the timetable of subway lines arriving at the station to predict the time when passenger flow will create traffic pressure. By analyzing pedestrian density information inside subway cars and historical passenger entry data for the station, it estimates the passenger flow during periods of pressure.
[0048] The historical model of this invention should consider time, date, and weather, and separately count boarding and alighting passengers, calculating their proportion of the total number of passengers arriving at the station. The historical model is continuously revised by monitoring passenger flow at key nodes within the platform to improve prediction accuracy. Key nodes include transfer entrances, escalator and stairwell entrances / exits, and turnstiles. The model is revised using node data to more accurately predict the number of passengers exiting the station.
[0049] The prediction results of this invention need to be accurate to different exits, because subway hubs contain multiple subway lines, and each exit has a different impact on road traffic. If exits are in the same direction of traffic and there are no traffic lights between them, then they are combined into a single statistical analysis.
[0050] This invention uses a transfer selection model to match the number of passengers entering and exiting each station within a projected time period, thus determining the final number of different modes of transportation entering the area. The transfer selection model considers the impact of factors such as weather, time, date, temperature, and current congestion levels on passenger choices.
[0051] This invention predicts the pressure on current road traffic by acquiring possible entry routes of vehicles and the current traffic conditions along those routes, and similarly predicts traffic congestion caused when related vehicles leave. The road congestion prediction takes into account the time for ground vehicles to stop and pick up passengers (affected by the degree of congestion within the platform), where congestion within the platform leads to longer exit times and thus longer waiting times.
[0052] By inputting different modes of transportation into the congestion impact model according to their quantities, the potential time periods of future congestion can be obtained. Taking into account the situation of bus stops near relevant exits and bus arrival times, under good weather conditions, buses can effectively alleviate passenger flow pressure and avoid slow passenger boarding and alighting due to overcrowding.
[0053] like Figure 1 As shown, the method of the present invention includes the following steps:
[0054] Step S1: The train inputs passenger flow and interacts with the urban rail ATS subsystem to obtain the timetables of all lines passing through this platform and the passenger density inside the subway.
[0055] Step S2, Arrival Passenger Flow Selection Model: Establish a KNN clustering model based on historical data to model the impact of different input passenger flows on exit passenger flow;
[0056] Step S3, Node monitoring model correction: Select the necessary nodes for passengers to enter and exit the station (escalators, turnstiles, exits, etc.), monitor the current passenger flow in real time, and optimize the model in S2 in real time.
[0057] Step S4, Crowd Selection Matching KNN Model: Build a selection model of possible modes of transportation for the crowd based on historical data such as time, weather, season, holidays, and temperature;
[0058] Step S5, Road Traffic Congestion Coefficient: By obtaining the influence factors of the choice of transportation by the current entrance and exit on the platform traffic flow through historical statistics, and by using the population choice matching model and the influence factors, the congestion degree of different road sections near the hub station is finally predicted. Specific Implementation
[0060] The method of this invention is as follows:
[0061] Step 1: Set a future time period T, and obtain the arrival times and time points T of trains within the time period T based on the timetable. 0， T1……T x , and the trains arriving at the corresponding times: TR0, TR1, ..., TR x By using in-vehicle cameras to obtain the passenger congestion level λ inside the platform, the formula for the number of people who may arrive at the station can be simplified into a time pulse function representing the incoming passenger flow (Num×λ0)×δ(T-T0).
[0062] Step 2: Based on historical data, establish a passenger boarding and alighting number model, and classify the sample based on the percentage of passengers getting off and the number of passengers getting on, according to date, weather, and time, to form clusters.
[0063] Step 3: Predict passenger arrivals and departures using existing clusters. This step uses the KNN model, a basic classification and regression method commonly used in supervised learning. The k-nearest neighbors algorithm assumes a training dataset with predefined instance categories. During classification, for a new instance, prediction is made based on the categories of its k nearest neighbors (training instances) through methods such as majority voting. The majority vote represents the closest proximity to a particular cluster. For example, on Tuesday, with an estimated time of 6 PM and rainy weather, the expected percentage of passengers disembarking (N) is calculated. Different times and weather conditions will form different clusters. During prediction, the current weather (rainy) and time (6 PM) are used to categorize the passengers. The categorization method uses... Lp ( w i , x j )=(∑ n i=1 =∣w i (i) - x j (l) | p ) 1 / p The distance between the vector to be predicted and different clusters is used to determine which cluster it belongs to. The distance p is chosen as the Euclidean distance, and the overall dimension p = 2. See details... Figure 2 In two-dimensional KNN, the classification selection is based on time T; in multi-dimensional cases, the next KNN model is selected accordingly. Figure 3 As shown, the next layer of clustering, composed of the number of passengers getting off and the date, will further classify the passengers to obtain the percentage N of possible passengers getting off. The number of passengers arriving at the station is obtained through the impulse function in step 1, thus obtaining the number of passengers getting off at this station.
[0064] Step 4: Repeat steps 2 and 3 to obtain the number of passengers boarding. The only change is that the percentage of passengers boarding is replaced with the percentage of passengers alighting. The prediction accuracy will be lower than that for passengers alighting. Step 5: Using the same method as steps 2 and 3, build a model for passengers choosing their transfer methods. Because subway hubs have a large flow of people with transfer needs, the model is similar to the previous models, including time, date, weather, and temperature. Figure 4 As shown in the figure. The final selection ratio of different transfer routes is obtained, and the number of people getting off the train obtained in step 3 is used to obtain the possible number of people exiting the station.
[0065] Step 6: Subtract the number of transfer passengers from the number of passengers boarding during time period T to obtain the estimated number of passengers arriving during time period T.
[0066] Step 7: Predict the flow of people arriving at different exits by establishing a traffic flow prediction model. When establishing the model, consider the existing passenger flow model as V(new) = aV + (1-a)V(old), where V(new) is T+T. m Passenger flow at the exit during the time period, T m V(old) represents the average time it takes for passengers to reach a certain exit, where V is the time taken in the past T+T. m The pedestrian flow within a time period, V, is the predicted new pedestrian flow that will arrive at the exit in the future. The expected passenger flow for boarding and alighting obtained in steps 5 and 6 is adjusted by the parameter 'a' so that the pedestrian flow at each exit is within T+T. m The passenger flow during a given time period is matched with the predicted number of passengers getting on and off the bus to obtain the passenger flow at each exit.
[0067] Step 8: Based on the different hub settings, consider whether the relevant input pedestrian flow needs to be entered separately or combined for each hub. For example, if two exits are located on the same side and there is no traffic light between them, then they should be combined for calculation.
[0068] Step 9: Using weather, time, date, temperature, and congestion level, establish KNN clustering for passenger ground transportation selection at each entrance / exit, using the same method as steps 2 and 3.
[0069] Step 10: Assign a traffic flow impact coefficient N to different entrances and exits. The coefficient N for each entrance / exit is... x The sum of the whole is 1.
[0070] Step 11: Assign different coefficients to each mode of transportation to affect traffic flow. When setting the coefficients, consider the pressure J1 on traffic flow when different modes of transportation leave, the pressure J2 on traffic flow when they pick up and drop off passengers, and the reduction M in the inflow of people.
[0071] Step 12: Model the traffic flow pressure of non-motorized vehicles and buses separately. Let the total number of shared bicycles deployed in the surrounding area be B, and BN (the number of people who selected shared bicycles) be the number of people that can be diverted.
[0072] Step 13: For buses, it is necessary to compare their timetables. If bus route X arrives at this exit within time period T, then the total number of people that can be diverted is X / total number of routes × Num (number of people choosing buses). The remaining people need to be brought into the selection model again to confirm their possibility of choosing different modes of transportation in the future.
[0073] Step 14: Based on the coefficients and the number of people (shared bicycles do not cause congestion, and bus arrivals reduce passenger flow), the final congestion index is obtained. Based on the congestion results of different road sections caused by historical congestion coefficients, the final congestion level of different road sections is obtained.
[0074] The above is an introduction to the method embodiments. The following embodiments using electronic devices and storage media will further illustrate the solution of the present invention.
[0075] The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.
[0076] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0077] The processing unit performs the various methods and processes described above, such as the methods of the present invention. For example, in some embodiments, the methods of the present invention may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of the methods of the present invention described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute the methods of the present invention by any other suitable means (e.g., by means of firmware).
[0078] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.
[0079] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.
[0080] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0081] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for predicting road congestion around a subway hub station based on a timetable, characterized by, This method first predicts the flow of people arriving at the platform at future times by using subway timetables and the crowding levels in trains; secondly, it matches the current flow density at key nodes within the platform with a density model generated from historical data to determine the flow of people arriving at the exit in the future; at the same time, it matches historical passenger transfer selection models and the impact of different modes of transportation on ground traffic to finally determine the traffic congestion that will occur in the future. The method specifically includes the following steps: Step S1: Train passenger flow input; Step S2: Construct an arrival passenger flow model; Step S3: Optimize the model constructed in step S2; Step S4: Construct a transfer selection model; Step S5: Predict the congestion level of different road sections near the hub station; Step S1, the train passenger flow input specifically involves: interacting with the urban rail ATS subsystem to obtain the timetables of all lines passing through this platform and the passenger density inside the subway, thereby estimating the input passenger flow; the estimated input passenger flow specifically involves: obtaining the timetables of subway lines arriving at this station to predict the time when the arriving passenger flow will create traffic pressure; and estimating the input passenger flow during the pressure period by using information on pedestrian density inside the subway car and information on the historical number of people entering this station. Step S4, constructing the transfer selection model, specifically involves: establishing a transfer selection model for people choosing modes of transportation based on historical data of time, weather, season, holidays, temperature, and current congestion level; The number of passengers entering and exiting each station during the expected time period is matched with the transfer selection model to obtain the number of different modes of transportation that will eventually enter the area. By obtaining the entry routes of the vehicles and the current traffic conditions of these routes, the pressure on the current road traffic is comprehensively predicted, and the traffic congestion caused when the relevant vehicles leave is also predicted. Step S5, predicting the congestion level of different road sections near the hub station, specifically involves: obtaining the influence factors of the choice of transportation by the current entrance and exit on the platform traffic flow through historical statistics; and finally predicting the congestion level of different road sections near the hub station through the transfer choice model of the crowd and the influence factors. 2.The subway hub station surrounding road congestion prediction method based on a timetable according to claim 1, wherein, Step S2, constructing the arrival passenger flow model, specifically involves: A KNN clustering model was established based on historical data, and a model of the impact of different input flows on the exit flow was constructed.
3. The method for predicting road congestion around a subway hub station based on a timetable, as described in claim 1, is characterized in that... Step S3, optimizing the model constructed in step S2, specifically involves: Key points for passenger entry and exit are selected for passenger flow density monitoring. The model constructed in step S2 is then optimized in real time by monitoring the current passenger flow.
4. The method for predicting road congestion around a subway hub station based on a timetable, as described in claim 3, is characterized in that... The key nodes include transfer entrances, escalators, exits, and turnstiles.
5. The method for predicting road congestion around a subway hub station based on a timetable, as described in claim 1, is characterized in that... By inputting different types of transportation into the congestion impact model according to their quantity, the time periods during which future congestion will occur can be obtained.
6. The method for predicting road congestion around a subway hub station based on a timetable, as described in claim 1, is characterized in that... The prediction results are accurate to different exits; if the exits are in the same direction of traffic and there are no traffic lights between them, they are combined into the statistics.
7. The method for predicting road congestion around a subway hub station based on a timetable, as described in claim 1, is characterized in that... This method takes into account the bus stop conditions near the relevant exits and the arrival time of buses. Under good weather conditions, buses can effectively alleviate the pressure of passenger flow, and timely bus arrival can prevent passengers from choosing to call a taxi, thus increasing the burden on the road.
8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 7.