Rail transit flow-limiting time evaluation method

Abstract

The application discloses a rail transit flow limiting time evaluation method, and relates to the technical field of rail transit operation management. The method comprises the following steps: collecting and preprocessing rail transit multi-source operation data to obtain standardized operation data; constructing a dynamic starting and ending point unit set according to the standardized operation data and performing starting and ending point-train probability distribution, and combining section passenger flow data to calibrate the distribution result; estimating potential arrival demand and actual release amount during the flow limiting period, and recursively calculating off-station queues and waiting time; generating a multi-scenario set and performing capacity-constrained dynamic loading simulation to calculate multi-dimensional evaluation indexes; and sorting and screening the optimal scenario according to the evaluation indexes to determine a recommended flow limiting time window. The application converts the flow limiting time judgment from experience judgment into a calculable and comparable data processing process, and can more clearly explain the influence of the flow limiting measures on the in-vehicle safety, off-station waiting and passing efficiency.

Description

Technical Field

[0001] This application relates to the field of rail transit operation and management, and in particular to a method for evaluating rail transit flow restriction time. Background Technology

[0002] During the morning rush hour, some urban rail transit stations need to control the pace of entry by means of limiting passenger flow, rerouting, diverting passengers, adding channels, and speeding up code verification in order to alleviate congestion at downstream train sections and passenger flow pressure on platforms.

[0003] In current practices, the start and end times of passenger flow restrictions largely rely on the experience of operations personnel, on-site queue observation, AFC (Automatic Fare Collection) entry volume, cross-sectional passenger flow statistics, or train occupancy rates. While these methods support daily scheduling, they still have shortcomings in refining the evaluation of the restriction time window: Firstly, AFC entry time records the time passengers swipe their cards to enter the station, which during the restriction period is closer to the time they are "allowed to enter the system," and cannot directly represent the actual arrival time of passengers at the station. If the AFC entry volume during the restriction period is directly taken as the actual demand, it is easy to misjudge the post-restriction release trough as a demand decline, thus affecting the judgment of the lifting time. Secondly, the granularity of data available to the operations side is inconsistent. AFC data may be the aggregated entry and exit volume of stations over 5 minutes, cross-sectional passenger flow may be the aggregated statistics over 15 minutes, and train timetables correspond to specific train arrival and departure events. Existing methods cannot unify these different granularities of data into the train-level cross-sectional load evaluation, and therefore cannot simultaneously explain the relationship between upstream entry pressure, downstream cross-sectional occupancy risk, and train supply capacity. Meanwhile, many evaluation methods tend to rely on single indicators, such as only looking at the number of passengers entering the station, the number of people queuing, or the occupancy rate. However, a decrease in the load inside the train does not necessarily mean that the solution is better; it may simply mean that more passengers are stranded outside the station, resulting in longer wait times and lower throughput. Actual on-site measures are not always simple, such as using gates to release passengers at a fixed rate. Barrier bypasses, channel expansion, zoned guidance, temporary code verification points, and manual crowd control all change the entry processing capacity and the release process. If the potential arrival demand, the actual release volume, and the train cross-sectional load are not distinguished, it is easy to misjudge changes in organizational capacity as changes in passenger demand.

[0004] Therefore, there is an urgent need for a rail transit flow restriction time evaluation method that can integrate multi-source operational data, correlate station entry and train cross-section passenger flow, distinguish between the flow restriction release capacity and actual demand, and comprehensively evaluate multiple dimensions to select flow restriction schemes. Summary of the Invention

[0005] The purpose of this application is to provide a method for evaluating the flow restriction time of rail transit, which can transform the judgment of flow restriction time from empirical judgment into a calculable data processing process, and more clearly illustrate the impact of flow restriction measures on in-vehicle safety, waiting outside stations, and traffic efficiency.

[0006] To achieve the above objectives, this application provides the following solution: A method for evaluating the time limit for rail transit flow control includes: Collect raw multi-source operation data of rail transit, and preprocess the raw multi-source operation data to obtain standardized operation data; the standardized operation data includes standardized AFC data, standardized train timetable data, standardized cross-sectional passenger flow data, standardized station topology data, and standardized flow restriction measure configuration data.

[0007] A dynamic origin-destination unit set is constructed based on the standardized AFC data. Then, origin-destination point matching and probability allocation are performed sequentially based on the dynamic origin-destination unit set, standardized train timetable data, and standardized station topology data to obtain the origin-destination point-train probability allocation result.

[0008] The estimated train cross-section load is calculated based on the origin-destination point-train probability allocation result, and the origin-destination point-train probability allocation result is calibrated by combining the cross-section passenger flow data to obtain the calibrated origin-destination point-train allocation result.

[0009] Based on the standardized AFC data and standardized flow control configuration data, the potential arrival demand and actual release volume during the flow control period are estimated. Based on the potential arrival demand and actual release volume, the off-site queue data and off-site waiting data of each station in each time window are calculated recursively in chronological order.

[0010] Based on the calibrated origin-destination train allocation results, potential arrival demand, actual release volume, off-station queue data, and preset scenario parameters, a set of scenarios to be evaluated is generated.

[0011] Capacity-constrained dynamic loading simulations were performed on each scenario to obtain the train loading results under each scenario. Based on the train loading results under each scenario and the waiting data outside the station, multi-dimensional evaluation indicators were calculated to obtain the scenario evaluation indicator table.

[0012] Based on the scenario evaluation index table, each scenario to be evaluated is sorted and filtered to obtain the result of determining the flow restriction time.

[0013] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method for evaluating passenger flow control time in rail transit. First, it collects raw multi-source operational data of rail transit and preprocesses it to obtain standardized operational data, enabling the originally scattered multi-source data to be processed within the same calculation flow. Second, it constructs a dynamic origin-destination unit set based on standardized AFC data. Then, based on the dynamic origin-destination unit set, standardized train timetable data, and standardized station topology data, it sequentially performs origin-destination point-train matching and probability allocation to obtain the origin-destination point-train probability allocation result, ensuring that station entry pressure and train operation events correspond at the data level. Next, it calculates the estimated train cross-sectional load based on the origin-destination point-train probability allocation result and calibrates the result using cross-sectional passenger flow data, obtaining a calibrated origin-destination point-train allocation result, ensuring that the reconstructed train cross-sectional load corresponds to the measured data in terms of aggregation. Next, based on standardized AFC (Automatic Train Capacity) data and standardized flow control measure configuration data, the potential arrival demand and actual release volume during the flow control period are estimated. Based on the potential arrival demand and actual release volume, the off-station queue data and off-station waiting data for each station in each time window are calculated recursively in chronological order. The AFC entry volume within the flow control time window is identified as the actual release volume rather than the real demand. The potential arrival demand is estimated by combining the passenger flow distribution during non-flow control periods or reference stations, reducing misjudgments of passenger flow demand caused by flow control. Afterwards, based on the calibrated origin-destination-train allocation results, potential arrival demand, actual release volume, off-station queue data, and preset scenario parameters, a set of scenarios to be evaluated is generated. Capacity-constrained dynamic loading simulation is performed on each scenario to obtain the train loading results under each scenario. Based on the train loading results and off-station waiting data under each scenario, multi-dimensional evaluation indicators are calculated to obtain a scenario evaluation indicator table. This ensures that different flow control schemes are not only compared based on the degree of congestion inside the train, but also comprehensively reflect the risks inside the train and the costs of waiting outside the station. Finally, based on the scenario evaluation index table, the scenarios to be evaluated are sorted and screened to obtain the results of the flow restriction time determination. This allows for direct comparison of different flow restriction start times, end times and changes in traffic capacity under a consistent standard, resulting in a quantifiable recommended flow restriction time scheme. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 An application environment diagram of a rail transit flow restriction time evaluation method provided in an embodiment of this application; Figure 2A flowchart illustrating a method for evaluating traffic flow restriction time in rail transit, provided as an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] The rail transit flow restriction time evaluation method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on other servers. Terminal 102 can send the raw multi-source operation data of rail transit to be processed to server 104. After receiving the raw multi-source operation data, server 104 preprocesses the raw multi-source operation data to obtain standardized operation data; constructs a dynamic origin-destination unit set based on standardized AFC data; and performs origin-destination point-train matching and probability allocation sequentially based on the dynamic origin-destination unit set, standardized train timetable data, and standardized station topology data to obtain the origin-destination point-train probability allocation result; calculates the train section load estimation result based on the origin-destination point-train probability allocation result, and calibrates the origin-destination point-train probability allocation result by combining it with section passenger flow data to obtain the calibrated origin-destination point-train... The system calculates train allocation results; based on standardized AFC data and standardized flow control configuration data, it estimates potential arrival demand and actual release volume during the flow control period. Based on these, it recursively calculates off-station queue data and off-station waiting data for each station in each time window, chronologically. Based on the calibrated origin-destination train allocation results, potential arrival demand, actual release volume, off-station queue data, and preset scenario parameters, it generates a set of scenarios to be evaluated. It performs capacity-constrained dynamic loading simulations on each scenario to obtain train loading results under each scenario. Based on the train loading results and off-station waiting data under each scenario, it calculates multi-dimensional evaluation indicators to obtain a scenario evaluation indicator table. Based on the scenario evaluation indicator table, it sorts and filters each scenario to obtain the flow control time determination result. Server 104 can feed back the obtained flow control time determination result to terminal 102. In addition, in some embodiments, the rail transit flow restriction time evaluation method can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly process the original multi-source operation data of rail transit to be processed, or the server 104 can obtain the original multi-source operation data of rail transit to be processed from the data storage system and process the original multi-source operation data of rail transit to be processed.

[0019] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0020] In one exemplary embodiment, such as Figure 2 As shown, a method for evaluating passenger flow control time in rail transit is provided. This method is executed by computer equipment, specifically by a terminal, server, workstation, rail transit passenger flow analysis platform, or other electronic devices with data processing capabilities, or by a terminal and server working together. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps A1 to A7. Wherein: Step A1: Collect raw multi-source operation data of rail transit, and preprocess the raw multi-source operation data to obtain standardized operation data; the standardized operation data includes standardized AFC data, standardized train timetable data, standardized cross-sectional passenger flow data, standardized station topology data, and standardized flow restriction measure configuration data.

[0021] Step A2: Construct a dynamic origin-destination unit set based on the standardized AFC data, and perform origin-destination point-train matching and probability allocation sequentially based on the dynamic origin-destination unit set, standardized train timetable data, and standardized station topology data to obtain the origin-destination point-train probability allocation result.

[0022] Step A3: Calculate the train section load estimation result based on the origin-destination point-train probability allocation result, and calibrate the origin-destination point-train probability allocation result in combination with the section passenger flow data to obtain the calibrated origin-destination point-train allocation result.

[0023] Step A4: Based on the standardized AFC data and standardized flow control configuration data, estimate the potential arrival demand and actual release volume during the flow control period. Based on the potential arrival demand and actual release volume, recursively calculate the off-site queue data and off-site waiting data of each station in each time window in chronological order.

[0024] Step A5: Based on the calibrated origin-destination train allocation results, potential arrival demand, actual release volume, off-station queue data, and preset scenario parameters, generate a set of scenarios to be evaluated.

[0025] Step A6: Perform capacity-constrained dynamic loading simulation for each scenario to be evaluated, obtain the train loading results under each scenario, and calculate multi-dimensional evaluation indicators based on the train loading results under each scenario and the waiting data outside the station to obtain the scenario evaluation indicator table.

[0026] Step A7: Based on the scenario evaluation index table, sort and filter each scenario to be evaluated to obtain the result of determining the flow restriction time.

[0027] In this embodiment, the standardized cross-sectional passenger flow data includes the cross-sectional operating day, cross-sectional direction, cross-sectional starting point, cross-sectional ending point, cross-sectional time window, and cross-sectional passenger flow. The origin and destination points can be simplified as OD (Origin-Destination).

[0028] Compared to methods that primarily rely on manual experience, on-site observation, AFC (Average Fare Collection) entry volume, or a single full load rate indicator to determine flow restriction time, this method, by implementing steps A1 to A7, can process previously scattered data within a single calculation process. After unified field mapping and time standardization, AFC entry / exit data, train timetables, cross-sectional passenger flow, station topology, and flow restriction measures configuration can be further transformed into dynamic OD (Original Demand) units, candidate train matching results, and train cross-sectional load results. Thus, upstream station entry pressure, train operation events, and downstream cross-sectional load are no longer separate indicators but can correspond to each other within the same data link. This application also clarifies the calculation process for train cross-sectional load. Instead of directly using station entry volume or cross-sectional statistics to roughly determine congestion, it first generates candidate trains for dynamic OD units based on train timetables and station topology, then calculates the probability of OD units being allocated to each candidate train based on waiting time and travel time residuals, thereby reconstructing the passenger capacity of trains at each cross-section. The cross-sectional passenger flow calibration process re-adjusts the allocation results using observed cross-sectional passenger flow while maintaining the total passenger volume of each dynamic origin-destination (OD) unit. This preserves the passenger flow source and ensures that the cross-sectional load corresponds to the observed data in the aggregation caliber. Therefore, the purpose of this method is not simply to provide a score, but to transform the judgment of flow restriction time from an empirical judgment into a calculable, traceable, and comparable data processing process. Through multi-source data fusion, dynamic OD construction, candidate train probability allocation, cross-sectional passenger flow calibration, potential arrival demand estimation, off-station queue recursion, and capacity constraint simulation, the impact of flow restriction measures on in-vehicle safety, off-station waiting, and traffic efficiency can be more clearly illustrated.

[0029] In another exemplary embodiment of this application, the preprocessing in step A1 above is replaced by the following steps B1 to B2: Step B1 involves accessing, aggregating, and labeling the source of the original multi-source operational data to obtain a multi-source operational data set.

[0030] Step B2 involves mapping fields, converting time, and validating data from the multi-source operational data set to obtain standardized operational data.

[0031] In this embodiment, specifically, the field mapping is achieved by the data standardization module mapping the station name, direction field, time field, and passenger flow field in different data tables to a unified field; the time conversion is to convert passenger entry time, passenger exit time, passenger entry time window, passenger exit time window, train arrival time, train departure time, flow restriction start time, flow restriction end time, flow restriction time window, and cross-sectional time window into a unified minute system or a unified time window format.

[0032] In another exemplary embodiment of this application, step A2 above, which constructs a dynamic origin-destination unit set based on the standardized AFC data, specifically includes: aggregating the standardized AFC data according to passenger operating day, passenger route direction, passenger entry station, passenger exit station, preset passenger entry time window, and preset passenger exit time window to form multiple dynamic origin-destination units, thus obtaining a dynamic origin-destination unit set; the standardized AFC data includes passenger operating day, passenger entry station, passenger exit station, passenger entry time, passenger exit time, and passenger number; the dynamic origin-destination unit includes passenger operating day, route direction, passenger entry station, passenger exit station, preset passenger entry time window, and preset passenger exit time window.

[0033] In this embodiment, OD refers to Origin-Destination, and the dynamic origin-destination unit can be geometrically simplified as a dynamic OD unit; the dynamic OD unit can be represented as g. g=(d,o,r,t) e ,t x ), where d is the passenger operating day, o is the passenger entry station, r is the passenger exit station, and t is the passenger exit station. e For passenger entry time window, t x This represents the passenger exit time window. The number of passengers corresponding to the dynamic OD unit g is denoted as N. g .

[0034] In another exemplary embodiment of this application, the standardized train timetable data includes train operating days, train route direction, train number, train route through stations and sections, train arrival time, and train departure time; the standardized station topology data includes station route direction, station sequence, adjacent stations, and section numbers; the origin-destination point-train matching in step A2 above is replaced by steps C1 to C7: Step C1: For each train in the standardized train timetable data, determine whether the passenger route direction matches the train route direction to obtain a first determination result.

[0035] Step C2: If the first judgment result is yes, then determine whether the passenger's entry station and the passenger's exit station are both within the station section that the train passes through, and obtain the second judgment result; otherwise, exclude the train.

[0036] Step C3: If the second judgment result is yes, then determine whether the waiting time is greater than or equal to the preset entry preparation time to obtain the third judgment result; otherwise, exclude the train. The waiting time is the time difference obtained by subtracting the passenger entry time from the train departure time.

[0037] Step C4: If the third judgment result is yes, then determine whether the waiting time is less than or equal to the preset maximum waiting time to obtain the fourth judgment result; otherwise, exclude the train.

[0038] Step C5: If the fourth judgment result is yes, then determine whether the passenger's exit time is later than the train's arrival time to obtain the fifth judgment result; otherwise, exclude the train.

[0039] Step C6: If the fifth judgment result is yes, then the train is selected as a candidate train for the dynamic origin-destination unit.

[0040] Step C7: Associate each dynamic origin-destination unit with the corresponding candidate train to obtain the origin-destination-train candidate matching table.

[0041] In this embodiment, the matching method for candidate trains can be replaced. Besides calculating the weights of candidate trains using waiting time and travel time residuals, minimum waiting time, minimum travel residuals, normalized weights, or a learning-based matching model can also be used to assign dynamic OD units to corresponding candidate trains. As long as a correspondence between OD passenger flow and train operation events can be established, the train matching objective in this embodiment can be achieved.

[0042] In another exemplary embodiment of this application, the probability allocation in step A2 above is replaced by the following steps D1 to D2: Step D1: Calculate the candidate weight of each candidate train based on the waiting time and travel time residual of each candidate train; the travel time residual is the difference between the passenger exit time and the arrival time of each candidate train at the passenger exit station.

[0043] Step D2: Normalize the candidate weights of each candidate train within the same dynamic origin-destination unit to obtain the origin-destination train probability allocation result.

[0044] In this embodiment, specifically, the probability allocation module calculates the candidate weight w based on the waiting time and travel time residual of the candidate trains. gk The allocation probability P is obtained by normalization within the same dynamic OD unit. gk The candidate weights can be represented as w. gk =exp(-aW gk -bE gk ), where Wgk E is the waiting time corresponding to the selection of train k for the dynamic OD unit g. gk The residual travel time is calculated based on passenger exit time and train arrival time. 'a' and 'b' are preset weighting parameters, used to control the influence of candidate train waiting time and the residual travel time on the allocation probability. The specific values ​​of 'a' and 'b' are not limited and vary depending on the route, direction, and time of day. The allocation probability can be expressed as P. gk =w gk / sum(w gh ), where h represents any candidate train of dynamic OD unit g. After normalization, the sum of the probabilities of the same dynamic OD unit being assigned to each candidate train is 1.

[0045] In another exemplary embodiment of this application, step A3 is replaced by steps E1 to E3: Step E1: Based on the origin-destination point-train probability allocation results and standardized station topology data, determine the line sections traversed by each dynamic origin-destination point unit, sum the number of passengers assigned to the same train and the same section, and obtain the train section load estimation results.

[0046] Step E2: Calculate the calibration ratio between the estimated train section load and the standardized section passenger flow data.

[0047] Step E3: Adjust the origin-destination point-train probability allocation result according to the calibration ratio, and renormalize it in each dynamic origin-destination point unit to obtain the calibrated origin-destination point-train allocation result and the calibrated train section load result.

[0048] In this embodiment, specifically, the train section load reconfiguration module determines the line section traversed by each dynamic OD unit based on the station topology, and sums the number of passengers assigned to the same train and the same section. The number of passengers assigned to train k by dynamic OD unit g is X. gk =N g P gk , where N g For the number of passengers in the dynamic OD unit g, P gk The probability of assigning dynamic OD unit g to train k. The estimated load of train k on section l can be expressed as L. kl =sum(X gk ), where the summation range is the dynamic OD unit of all paths passing through section l and assigned to train k.

[0049] Specifically, the cross-section calibration module calculates the proportional relationship between the observed cross-section passenger flow and the estimated cross-section load, and uses this proportional relationship to adjust the OD-train allocation probability. For cross-section observation m, its calibration ratio can be expressed as R. m=Y m / Z m , where Y m For cross-sectional observation of passenger flow, Z m Estimate the cross-sectional load for the model. The cross-sectional calibration module uses R... m The relevant OD-train allocation probabilities are adjusted and renormalized within each dynamic OD unit, ensuring that the sum of allocation probabilities within the same dynamic OD unit remains 1, thus maintaining the total passenger volume of the dynamic OD unit unchanged. Different methods can be used for cross-sectional passenger flow calibration. Besides calibration based on the proportional relationship between observed cross-sectional passenger flow and estimated cross-sectional load, iterative proportional fitting, constrained least squares, and Bayesian calibration can also be used to ensure that the reconstructed cross-sectional load and observed cross-sectional passenger flow remain consistent within the corresponding time window, without changing the total passenger volume of each dynamic OD unit.

[0050] In another exemplary embodiment of this application, the standardized traffic restriction measure configuration data includes traffic restriction sites, traffic restriction start time, traffic restriction end time, measure type, and traffic capacity parameters; step A4 above is replaced by steps F1 to F5: Step F1: Determine the traffic restriction sites and traffic restriction time windows based on the standardized traffic restriction measures configuration data; the traffic restriction time window is from the start time of traffic restriction to the end time of traffic restriction.

[0051] Step F2: Select an unrestricted time window or a reference site that has not implemented flow restriction for the same flow-limited site as a reference for potential demand estimation.

[0052] Step F3: Identify the number of passengers in the standardized AFC data within the flow restriction time window as the actual release volume.

[0053] Step F4: Based on the reference passenger flow change trend estimated by the potential demand, estimate the potential arrival demand of the flow-limited station within the flow-limiting time window.

[0054] Step F5: Based on the potential arrival demand and the actual release volume, calculate the off-site queue data and off-site waiting data of each station in each time window in chronological order.

[0055] In this embodiment, the potential arrival demand is denoted as A. st The actual number of vehicles released is denoted as R. st Where s is the station and t is the time window. A st R is the quantity to be estimated. st This is a known quantity derived from the passenger count in the AFC standard data. Specifically, the off-site queue recursion module calculates the queue change for each station between adjacent time windows in chronological order. The off-site queue of station s in time window t+1 can be represented as Q. s,t+1 =max(0,Qst +A st -R st ), where Q st For the off-site queue of site s in time window t, A st For the potential arrival demand of site s in time window t, R st This represents the actual passenger flow at station s within time window t. Waiting time outside the station can be calculated based on the size of the external queue and the passenger flow capacity per unit time. The estimation method for potential arrival demand can be varied. Besides using non-restricted time windows, the concurrent time distribution of uncontrolled stations, and historical distributions of the same station for estimation, time series models, machine learning regression models, or manual setting of potential arrival demand by operators based on entrance capacity can also be used. The common goal of these methods is to distinguish between the AFC (Automatic Flow Control) entry volume during the flow restriction period and the actual passenger arrival demand. The capacity parameter is formed based on the station's existing flow restriction configuration and available channels.

[0056] Implementing steps F1 to F5 above can reduce misjudgments of passenger demand during flow control periods. During flow control, the AFC entry volume records the actual number of passengers swiping their cards to enter the system, which is closer to the release capacity than the actual demand for passengers arriving at the station. Existing methods, if they directly treat this AFC entry volume as demand, easily misinterpret the decrease in release capacity caused by flow control as a decrease in demand. This method, in the potential arrival demand estimation stage, treats the AFC entry volume within the flow control time window as the actual release capacity, and combines it with non-flow control time windows, concurrent distribution at uncontrolled stations, historical distribution at the same station, or prediction models to estimate potential arrival demand, thereby distinguishing between passengers who have "arrived but not yet entered the gate" and passengers who have "entered the gate."

[0057] In another exemplary embodiment of this application, the preset scenario parameters in step A5 above include the flow restriction start time, flow restriction end time, traffic capacity, organization measures, and vehicle interval parameters; the set of scenarios to be evaluated includes actual organization scenarios and counterfactual scenarios; the counterfactual scenarios include no flow restriction scenario, early lifting scenario, extended flow restriction scenario, traffic capacity adjustment scenario, and vehicle interval change scenario.

[0058] In this embodiment, the preset scenario parameters for the early release scenario and the capacity adjustment scenario also include a gradual release time. The gradual release time is derived from simulation results of different release times and gradually increasing capacity schemes, and is determined from candidate schemes that meet operational constraints such as train occupancy rate, station queues, and waiting time.

[0059] In another exemplary embodiment of this application, the capacity-constrained dynamic loading simulation of each scenario to be evaluated in step A6 above is replaced by the following steps G1 to G5: Step G1: Determine the remaining capacity of the train based on the difference between the train's capacity and the current passenger capacity; the train capacity is preset based on the train formation type and model.

[0060] Step G2: Determine the current platform waiting demand of the station based on the actual passenger volume and the off-site queue data.

[0061] Step G3: Determine whether the waiting demand at the platform is less than or equal to the remaining capacity of the train to obtain the sixth determination result; if the sixth determination result is yes, then determine that the actual number of passengers boarding is equal to the waiting demand at the platform; otherwise, determine that the actual number of passengers boarding is equal to the remaining capacity of the train, and record the number of passengers who failed to board as the unboarded demand.

[0062] Step G4: Transfer the unloaded demand to subsequent trains and continue the simulation until all trains and all stations have been visited to obtain the train loading results under each scenario.

[0063] Step G5: Calculate multi-dimensional evaluation indicators based on the train loading results and the waiting data outside the station under each scenario to obtain a scenario evaluation indicator table; the multi-dimensional evaluation indicators include in-vehicle safety indicators, waiting data outside the station, station entry efficiency indicators, and fairness indicators.

[0064] In this embodiment, in-vehicle safety indicators include the maximum cross-sectional load factor and the number of overloaded trains; station waiting indicators include the average station waiting time, station waiting quantiles, and the maximum station queue; station entry efficiency indicators include total throughput and capacity loss; and fairness indicators include waiting differences between different stations. The capacity constraint simulation can be adjusted according to operational guidelines. Train capacity can be physical capacity, comfort capacity, vehicle-type capacity, or capacity thresholds set by operational strategies; passengers who fail to board can transfer to the next train, or the process can be accumulated by time window or recursively applied according to platform queue rules. These different processing methods all reflect boarding, waiting, and congestion situations when train capacity is limited. Scenario settings and indicator weights can also be adjusted. In addition to scenarios such as actual organization, no flow restriction, early release, extended flow restriction, unorganized optimization, and changes in train intervals, scenarios such as weather changes, equipment failures, temporary large passenger flows, different numbers of channels, or different personnel configurations can also be added. The weights of evaluation indicators can also be set according to safety priority, efficiency priority, or fairness priority. The above alternative schemes can still be used to compare different flow restriction times and organizational capacity schemes.

[0065] By implementing steps G1 to G5 above, this application can simultaneously reflect both in-train risks and waiting times outside the station. When only considering the load factor, a certain scheme may appear less crowded inside the train due to allowing fewer passengers to enter the station, but the actual cost is that passengers are left outside the station. To avoid this situation, this method, in addition to calculating the train cross-sectional load, also recursively calculates the queues and waiting times outside the station, and calculates the number of people waiting on the platform, the number of people who failed to board, and the passenger waiting time in the capacity constraint simulation. In this way, different flow control schemes are no longer compared solely based on "whether the train is emptier," but rather by simultaneously considering in-train safety, waiting times outside the station, and entry efficiency.

[0066] In another exemplary embodiment of this application, step A7 is replaced by steps H1 to H5: Step H1: Based on in-vehicle safety indicators, station waiting indicators, and station entry efficiency indicators, comprehensively evaluate and rank each scenario to be evaluated, and obtain the evaluation indicators and ranking results for each dimension of each scenario to be evaluated.

[0067] Step H2: Select the scenarios to be evaluated that meet the preset operational constraints and have the best overall score as recommended scenarios.

[0068] Step H3: Extract the start time, end time, and traffic capacity parameters of the recommended scenario, and use them as the recommended start time, end time, and traffic capacity, respectively.

[0069] Step H4: If the recommended scenario is an early termination scenario or a capacity adjustment scenario, then the gradual termination time of the recommended scenario is extracted as the recommended optional gradual termination time; otherwise, the recommended optional gradual termination time is empty.

[0070] Step H5: Extract the evaluation indicators of each dimension corresponding to the recommended scenario as the passenger flow organization evaluation result.

[0071] Implementing steps H1 to H5 facilitates the comparison of different organizational schemes. Scenarios such as actual organization, no flow restriction, early release, extended flow restriction, unorganized optimization, and changes in train intervals can all be calculated within the same data processing and simulation workflow. Because the input criteria, calculation process, and indicator criteria remain consistent, the start and end times of flow restriction, organizational capacity, and changes in train supply under different scenarios can be directly compared, thereby obtaining the recommended start and end times of flow restriction, as well as optional gradual release times.

[0072] This application also provides an application scenario in which the above-mentioned rail transit flow restriction time evaluation method is applied. Specifically, the rail transit flow restriction time evaluation method provided in this embodiment can be applied in urban rail transit operation and scheduling scenarios. This scenario includes a data acquisition stage, a data processing stage, and a flow restriction decision stage; raw multi-source operation data enters the data processing link from various business systems, and after processing steps such as data standardization, dynamic OD construction, probability allocation, cross-section calibration, and simulation evaluation, the flow restriction time determination result is obtained and output to the downstream flow restriction decision stage.

[0073] Specifically, in the peak-hour traffic control organization of rail transit, the flow control stations and flow control time windows can be optimized and determined based on the collaborative approach of dynamic OD reconstruction and cross-section calibration, that is, to provide operators with scientific flow control start time, flow control end time and optional gradual release time.

[0074] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores the flow control time determination results. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program can implement a method for evaluating rail transit flow control time.

[0075] Figure 3 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0076] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0077] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0078] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with relevant regulations and be authorized by the owner of the corresponding device.

[0079] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0080] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0081] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for evaluating the time of passenger flow restriction in rail transit, characterized in that, include: Collect raw multi-source operation data of rail transit, and preprocess the raw multi-source operation data to obtain standardized operation data; the standardized operation data includes standardized AFC data, standardized train timetable data, standardized cross-sectional passenger flow data, standardized station topology data, and standardized flow restriction measure configuration data; A dynamic origin-destination unit set is constructed based on the standardized AFC data. Then, origin-destination point matching and probability allocation are performed sequentially based on the dynamic origin-destination unit set, standardized train timetable data, and standardized station topology data to obtain the origin-destination point-train probability allocation result. The estimated train cross-section load is calculated based on the origin-destination point-train probability allocation result, and the origin-destination point-train probability allocation result is calibrated by combining the cross-section passenger flow data to obtain the calibrated origin-destination point-train allocation result. Based on the standardized AFC data and standardized flow control configuration data, the potential arrival demand and actual release volume during the flow control period are estimated. Based on the potential arrival demand and actual release volume, the off-site queue data and off-site waiting data of each station in each time window are calculated recursively in chronological order. Based on the calibrated origin-destination train allocation results, potential arrival demand, actual release volume, off-station queue data, and preset scenario parameters, a set of scenarios to be evaluated is generated; Capacity-constrained dynamic loading simulations were performed on each scenario to obtain the train loading results under each scenario. Based on the train loading results under each scenario and the waiting data outside the station, multi-dimensional evaluation indicators were calculated to obtain the scenario evaluation indicator table. Based on the scenario evaluation index table, each scenario to be evaluated is sorted and filtered to obtain the result of determining the flow restriction time.

2. The method for evaluating rail transit flow restriction time according to claim 1, characterized in that, The preprocessing includes: The original multi-source operational data is accessed, collected, and its source is marked to obtain a multi-source operational data set. The multi-source operational data set is subjected to field mapping, time conversion, and data verification to obtain standardized operational data.

3. The method for evaluating rail transit flow restriction time according to claim 1, characterized in that, A dynamic origin-destination unit set is constructed based on the standardized AFC data, specifically including: Standardized AFC (Automatic Flow Coordination) data is aggregated according to passenger operating day, passenger route direction, passenger entry station, passenger exit station, preset passenger entry time window, and preset passenger exit time window to form multiple dynamic origin-destination unit sets, resulting in a dynamic origin-destination unit set. The standardized AFC data includes passenger operating day, passenger entry station, passenger exit station, passenger entry time, passenger exit time, and passenger number. The dynamic origin-destination unit includes passenger operating day, route direction, passenger entry station, passenger exit station, preset passenger entry time window, and preset passenger exit time window.

4. The method for evaluating rail transit flow restriction time according to claim 3, characterized in that, The standardized train timetable data includes train operating days, train route direction, train number, train station sections, train arrival time, and train departure time; the standardized station topology data includes station route direction, station sequence, adjacent stations, and section numbers; the origin-destination train matching specifically includes: For each train in the standardized train timetable data, determine whether the passenger's route direction matches the train's route direction to obtain a first determination result; If the first judgment result is yes, then it is determined whether the passenger's entry station and the passenger's exit station are both within the station section that the train passes through, and a second judgment result is obtained; otherwise, the train is excluded. If the second judgment result is yes, then it is determined whether the waiting time is greater than or equal to the preset entry preparation time, and a third judgment result is obtained; otherwise, the train is excluded; the waiting time is the time difference obtained by subtracting the passenger entry time from the train departure time. If the third judgment result is yes, then it is determined whether the waiting time is less than or equal to the preset maximum waiting time to obtain the fourth judgment result; otherwise, the train is excluded. If the fourth judgment result is yes, then it is determined whether the passenger's exit time is later than the train's arrival time, and the fifth judgment result is obtained; otherwise, the train is excluded. If the fifth judgment result is yes, then the train will be selected as a candidate train for the dynamic origin and destination unit; Each dynamic origin-destination unit is associated with the corresponding candidate train to obtain an origin-destination train candidate matching table.

5. The method for evaluating rail transit flow restriction time according to claim 4, characterized in that, The probability allocation specifically includes: The candidate weight of each candidate train is calculated based on the waiting time and travel time residual of each candidate train; the travel time residual is the difference between the passenger's exit time and the arrival time of each candidate train at the passenger's exit station; The candidate weights of each candidate train are normalized within the same dynamic origin-destination unit to obtain the origin-destination train probability allocation result.

6. The method for evaluating rail transit flow restriction time according to claim 1, characterized in that, The estimated train cross-section load is calculated based on the origin-destination point-train probability allocation result, and the origin-destination point-train probability allocation result is calibrated by combining the cross-section passenger flow data to obtain the calibrated origin-destination point-train allocation result, specifically including: Based on the origin-destination point-train probability allocation results and standardized station topology data, the line sections traversed by each dynamic origin-destination point unit are determined, and the number of passengers allocated to the same train and the same section is summed to obtain the train section load estimation results. Calculate the calibration ratio between the estimated train section load and the standardized section passenger flow data; The origin-destination point-train probability allocation result is adjusted according to the calibration ratio, and then re-normalized within each dynamic origin-destination point unit to obtain the calibrated origin-destination point-train allocation result and the calibrated train section load result.

7. The method for evaluating rail transit flow restriction time according to claim 1, characterized in that, The standardized flow control configuration data includes flow control stations, flow control start time, flow control end time, measure type, and throughput parameters. Based on the standardized AFC data and standardized flow control configuration data, the potential arrival demand and actual throughput during the flow control period are estimated. Based on the potential arrival demand and actual throughput, the off-site queue data and off-site waiting data for each station in each time window are calculated recursively in chronological order, specifically including: The traffic restriction sites and traffic restriction time windows are determined based on the standardized traffic restriction measures configuration data; the traffic restriction time window is from the start time of traffic restriction to the end time of traffic restriction. Select the non-limited time window or the reference site that has not implemented flow restriction at the same flow restriction site as a reference for potential demand estimation; The number of passengers in the standardized AFC data within the flow restriction time window is identified as the actual release volume; Based on the reference passenger flow trend estimated by the potential demand, the potential arrival demand of the flow-limited station within the flow-limiting time window is estimated; Based on the potential arrival demand and the actual release volume, the off-site queue data and off-site waiting data of each station in each time window are calculated recursively in chronological order.

8. The method for evaluating rail transit flow restriction time according to claim 1, characterized in that, The preset scenario parameters include the start time of traffic restriction, the end time of traffic restriction, traffic capacity, organizational measures, and vehicle interval parameters; the set of scenarios to be evaluated includes actual organizational scenarios and counterfactual scenarios; the counterfactual scenarios include scenarios without traffic restriction, scenarios with early lifting of traffic restriction, scenarios with extended traffic restriction, scenarios with adjusted traffic capacity, and scenarios with changes in vehicle interval.

9. The method for evaluating rail transit flow restriction time according to claim 7, characterized in that, Capacity-constrained dynamic loading simulations are performed for each scenario to be evaluated, specifically including: For any scenario to be evaluated, the number of passengers disembarking at each station is determined based on the calibrated origin-destination train allocation results, and the current passenger capacity of the train is updated based on the number of passengers disembarking; the current passenger capacity of the train is the cumulative difference between the initial number of passengers boarding the train when it departs from the origin station and the number of passengers boarding and alighting at each station along the route. The remaining capacity of the train is determined based on the difference between the train's capacity and its current passenger capacity; the train's capacity is preset based on the train formation type and model. The platform waiting demand at the current station is determined based on the actual passenger volume and the off-site queue data. Determine whether the waiting demand at the platform is less than or equal to the remaining capacity of the train to obtain a sixth determination result; if the sixth determination result is yes, then determine that the actual number of passengers boarding is equal to the waiting demand at the platform; otherwise, determine that the actual number of passengers boarding is equal to the remaining capacity of the train, and record the number of passengers who failed to board as the unboarded demand. The unloaded demand is transferred to subsequent trains, and the simulation continues until all trains and all stations have been visited to obtain the train loading results under each scenario. Based on the train loading results and the waiting data outside the station under each scenario, multi-dimensional evaluation indicators are calculated to obtain the scenario evaluation indicator table; the multi-dimensional evaluation indicators include in-vehicle safety indicators, waiting data outside the station, station entry efficiency indicators, and fairness indicators.

10. The method for evaluating rail transit flow restriction time according to claim 9, characterized in that, The results of determining the flow restriction time include: recommended start time, recommended end time, recommended capacity, recommended optional gradual lifting time, and passenger flow organization evaluation results; based on the scenario evaluation index table, each scenario to be evaluated is sorted and filtered to obtain the results of determining the flow restriction time, specifically including: Based on in-vehicle safety indicators, station waiting indicators, and station entry efficiency indicators, each scenario to be evaluated is comprehensively evaluated and ranked, resulting in evaluation indicators and ranking results for each dimension of each scenario to be evaluated. The scenario to be evaluated that meets the preset operational constraints and has the best overall score is selected as the recommended scenario. Extract the start time, end time, and traffic capacity parameters of the recommended scenario, and use them as the recommended start time, end time, and traffic capacity, respectively. If the recommended scenario is an early termination scenario or a capacity adjustment scenario, then the gradual termination time of the recommended scenario is extracted as the recommended optional gradual termination time; otherwise, the recommended optional gradual termination time is empty. The evaluation indicators for each dimension corresponding to the recommended scenario are extracted as the evaluation results of passenger flow organization.

Images