CN112598177B

CN112598177B - Urban rail transit emergency online passenger flow prediction and simulation system

Info

Publication number: CN112598177B
Application number: CN202011535098.2A
Authority: CN
Inventors: 白云云; 汪波; 黄建玲; 吴欣然; 陈文�; 吕楠; 胡清梅; 韩庆龙
Original assignee: BEIJING TRANSPORTATION INFORMATION CENTER; Beijing Subway Operation Corp
Current assignee: BEIJING TRANSPORTATION INFORMATION CENTER; Beijing Subway Operation Corp
Priority date: 2020-12-22
Filing date: 2020-12-22
Publication date: 2023-10-24
Anticipated expiration: 2040-12-22
Also published as: CN112598177A

Abstract

The invention discloses an urban rail transit emergency online passenger flow prediction and simulation system, which comprises a real-time passenger flow monitoring module and an emergency passenger flow prediction module; the real-time passenger flow monitoring module is used for predicting the passenger flow evolution state in the future preset time length under the normal condition and generating passenger flow indexes of the urban rail transit real-time-sharing granularity; the predicted evolution state of the passenger flow is visually displayed, so that the real-time passenger flow is monitored; the emergency passenger flow prediction module is used for adjusting a train running schedule aiming at an emergency occurring in a road network; predicting the distribution condition of the passenger flow under the condition of being influenced by the event according to the adjusted timetable; and issues the affected degree information of each terminal to the passengers. The urban rail transit emergency online passenger flow prediction and simulation system can monitor passenger flow online in real time and automatically adjust train operation when accidents occur.

Description

Urban rail transit emergency online passenger flow prediction and simulation system

Technical Field

The invention relates to the technical field of urban rail transit emergency processing, in particular to an urban rail transit emergency online passenger flow prediction and simulation system.

Background

Under the condition of urban rail transit networked operation, the network structure is more and more complex, the correlation degree of each line is higher and higher, the emergency types are gradually diversified, the occurrence frequency is increased, the scope of spread is enlarged, and once the emergency occurs, the chain reaction can occur, which is a serious test for the urban rail transit system which runs at full load, the emergency can influence the normal operation of each train of the rail line, and if the emergency is not handled in time, the network transportation capacity is reduced and even paralysis occurs.

At present, the existing real-time passenger flow monitoring system cannot accurately reflect the passenger flow distribution state under the emergency condition. The passenger flow prediction method is suitable for the situation of small passenger flow fluctuation in a normal state, such as a four-stage method, a non-centralized meter model, a neural network and the like, and has the advantages of large sudden passenger flow fluctuation, small occurrence probability of the same event in the same situation and poor applicability; the data base is questionnaire investigation, and the artificial interference is serious based on a non-aggregate model of behavior analysis, so that the common passenger flow prediction method generally has a certain limit in the emergency passenger flow prediction.

Disclosure of Invention

The invention provides an urban rail transit emergency online passenger flow prediction and simulation system, which aims to solve the technical problem of poor applicability of the existing real-time passenger flow monitoring system under the emergency condition.

In order to solve the technical problems, the invention provides the following technical scheme:

an urban rail transit emergency online passenger flow prediction and simulation system comprises a real-time passenger flow monitoring module and an emergency passenger flow prediction module; wherein,

the real-time passenger flow monitoring module is used for predicting the passenger flow evolution state in the future preset time length under the normal condition and generating passenger flow indexes of the urban rail transit real-time-sharing granularity; the predicted evolution state of the passenger flow is visually displayed, so that the real-time passenger flow is monitored;

the emergency passenger flow prediction module comprises an emergency train operation plan adjustment and simulation unit, an emergency passenger flow prediction and simulation unit and an emergency information service unit; wherein,

the emergency train operation plan adjustment and simulation unit is used for adjusting a train operation schedule aiming at the emergency occurring in the road network; the emergency passenger flow prediction and simulation unit is used for predicting passenger flow distribution conditions under the condition of being influenced by the event according to the adjusted train operation schedule; the emergency information service unit is used for issuing the affected degree information of each terminal station to passengers.

Further, the passenger flow index includes: ingress and egress volume, transfer volume, passenger traffic volume, and section passenger traffic volume.

Further, when no emergency occurs in the road network, the real-time passenger flow monitoring module operates, and the emergency passenger flow prediction module stops; when an emergency occurs in the road network, the emergency passenger flow prediction module starts to operate according to an operation instruction of an operator, and the real-time passenger flow monitoring module stops.

Further, the emergency train operation plan adjustment and simulation unit is specifically configured to:

according to the planned train operation schedule data and the emergency information, the train operation schedule of the emergency line is adjusted by combining with a driving scheduling adjustment strategy under the emergency condition, and train operation simulation is carried out by using the adjusted operation schedule; inputting the adjusted operation schedule into the emergency passenger flow prediction and simulation unit; the emergency information comprises emergency occurrence time, emergency occurrence position, emergency type, emergency predicted duration, emergency extension time, emergency influence direction and train operation influence degree; the train operation schedule data comprises a train ID, an update date, a line number, a train number, a schedule type, a schedule number, a train direction, a station name, a station number, a station type, an arrival time, a departure time, a vehicle type, a train grouping and a train stator.

Further, the emergency passenger flow prediction and simulation unit is specifically configured to:

and predicting the OD passenger flow according to the emergency information, the historical passenger flow OD data and the road network basic data, adopting a preset emergency OD prediction model for the OD of the O/D on the emergency line, adopting a preset real-time passenger flow prediction model for the OD of the O/D on the non-emergency line, using the K short path set updated according to the emergency information, the adjusted train running schedule and the prediction result of the OD passenger flow for passenger flow distribution, outputting relevant passenger flow indexes and carrying out visual early warning.

Further, the emergency information service unit is specifically configured to:

and evaluating the affected degree of each station according to the OD shortest path set, the emergency information and the road network basic data under the normal and emergency conditions, and distributing the affected degree information from the station to other stations for the group.

Further, if it is determined that the duration of the emergency is changed within the predicted duration, the emergency train operation plan adjustment and simulation unit is specifically configured to:

when the duration of the emergency is prolonged, correspondingly readjusting the train operation schedule, and increasing the time step of passenger flow prediction according to the prolonged duration of the emergency; and when the duration of the emergency is shortened, reducing the time step of passenger flow prediction according to the shortened duration of the emergency.

Further, when the duration of the emergency is not longer than the preset duration, the emergency passenger flow prediction module still keeps running; and when the preset time is full after the emergency is over, the emergency passenger flow prediction module is completely stopped, and the real-time passenger flow monitoring module starts to normally operate.

acquiring the occurrence time, the occurrence position, the predicted duration, the extension time, the type, the influence direction, the influence degree and the train running interval of the train during the fault period of the emergency according to the manually input emergency information;

based on the current planned train operation schedule, by combining with driving adjustment rules under different emergency conditions, automatically generating a train operation schedule under the emergency, after the train operation schedule is adjusted, if the artificial judgment of the emergency is not finished and the emergency duration is changed, manually inputting the emergency extension time, and if the emergency duration is prolonged, automatically adjusting and generating a new train operation schedule; repeating the judging process until no new emergency is input for a prolonged time; if the extension time is not input, defaulting to a duration time point, and ending the emergency;

On the basis of the newly generated train operation schedule data, a visual train operation diagram is generated through a computer simulation technology, and the affected train range is displayed.

Further, the step of automatically generating the train operation schedule under the emergency by combining the driving adjustment rules under different emergency conditions based on the current planned train operation schedule comprises the following steps:

s1, adjusting an influence area in advance, which comprises the following steps:

s11, obtaining the initial train number sequence number of the pre-event influence area according to the train outbound time of the emergency interval initial station and the emergency initial time, and marking asThe expression is as follows:

wherein ,ud_i,s Representing departure time of train number i at S station in operation schedule before adjustment, S _fromLocation A starting station representing an emergency interval, and a from time representing a starting time of the emergency;

s12, obtaining the sequence number of the ending train number of the prior influence area according to the train outbound time and the fault start time of the current direction starting station, and marking asThe expression is as follows:

wherein ,S₁ Representing the origin of the train;

s13, calculating the time length of the prior influence area according to the starting time of the starting train number and the ending train number of the prior influence area at the starting time of the starting station

wherein ,starting train number indicating the influence area before the event, +.>An ending train number indicating the pre-event impact zone;

obtaining the serial number set of the line drawing train number as

According to the desired interval t of manual input ₁ and Q_before Number of middle train number determining number of line drawing of influence interval before issueWherein count () represents the number of passes in the set, +.>Representing downward rounding, and taking 1 as a result if the rounded result is less than 1;

s14, from Q _before Uniformly and randomly pumping N _before The number of vehicles, the extracted number of vehicles run the line to delete the part after the starting time of the emergency, and the part before the starting time of the emergency is reserved;

s2, adjusting an incident area, which comprises the following steps:

s21, determining the initial train number of the accident area adjustment wherein ,Representing the initial number of vehicles in the incident area;

s22, obtaining the sequence number of the ending train number of the accident area according to the train outbound time of the starting station in the current direction and the ending time of the accident

Wherein, toTime represents the ending time of the emergency;

s23, calculating the time length of the accident area according to the starting time of the starting time and the ending time of the accident area at the starting station wherein ,Indicating the ending number of vehicles in the accident area;

determining a line drawing train number set in an emergency region

According to the desired interval t of manual input ₁ and Q_fault The number of middle train number determines the number of line drawing in accident areaWherein count () represents the number of passes in the set, +.>Representing a downward rounding; if the rounded result is less than 1, the result is 1;

s24, from Q _fault Uniformly and randomly pumping N _fault The number of vehicles, the extracted running line of vehicles stops from the starting station, and the vehicles are stoppedThe subsequent train number runs according to the original train running schedule;

s3, adjusting the duration of the emergency event, which comprises the following steps:

if the predicted duration of the emergency is shortened before the emergency is finished, the adjusted train operation schedule is still maintained and is not changed;

if an input t for prolonging the duration of the emergency is received before the end time of the emergency _{Extension of} The operations S1 and S2 are carried out again on the basis of the adjusted train operation schedule, the new-round adjusted fromTime is the original total time, and the new-round adjusted total time is the original total time plus t _{Extension of} The expected train running interval of the new round of adjustment is the manually input accident prolonged train running interval t ₂ ；

If the influence direction of the emergency is unidirectional, executing S1-S3 to adjust the train number in the influence direction of the emergency; and if the influence direction of the emergency event is bidirectional, executing S1 to S3 to respectively adjust the train number of the upward train and the downward train.

The technical scheme provided by the invention has the beneficial effects that at least:

the invention realizes high-precision prediction of the future short-time passenger flow evolution state through the real-time passenger flow monitoring module, and generates the passenger flow index of the urban rail transit real-time granularity; the predicted evolution state of the passenger flow is visually displayed, so that the real-time passenger flow is monitored; the train running schedule is adjusted according to the emergency occurring in the road network through the emergency passenger flow prediction module; predicting the distribution condition of the passenger flow under the condition of being influenced by the event according to the adjusted timetable; and issues the affected degree information of each terminal to the passengers. Therefore, the passenger flow can be monitored online in real time, and the running condition of the train can be automatically adjusted when an emergency occurs.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a business flow of an urban rail transit emergency online passenger flow prediction and simulation system provided by an embodiment of the invention;

fig. 2 is a schematic business flow diagram of an emergency passenger flow prediction module according to an embodiment of the present invention;

FIG. 3 is a flowchart of an implementation of an emergency train operation adjustment and simulation process provided by an embodiment of the present invention;

FIG. 4 is a flow chart of predicting emergency passenger flow provided by an embodiment of the invention;

FIG. 5 is a flow chart of OD prediction of an emergency passenger flow provided by an embodiment of the invention;

FIG. 6 is a flow chart of an emergency passenger flow simulation provided by an embodiment of the invention;

fig. 7 is a flowchart of issuing guidance information for a group according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings.

The embodiment provides an urban rail transit emergency online passenger flow prediction and simulation system, which supports data by using historical card swiping transaction data, real-time packed data, road network basic data, planned train schedule data, emergency information and the like, and realizes business functions by using a model between data and big data operation, storage exchange and the like. The service function module mainly comprises a real-time passenger flow monitoring module and an emergency passenger flow prediction module.

As shown in fig. 1, the real-time passenger flow monitoring module is configured to predict a future short-time passenger flow evolution state with high precision by using various basic data such as real-time and historical data, generate real-time granularity OD passenger flow (OD: origin-destination point of transportation trip, O refers to departure point of trip, D refers to destination of trip), section passenger flow, station-in and station-out amount, transfer amount, etc., allocate the predicted passenger flow, perform hierarchical early warning, and realize real-time passenger flow monitoring by visually displaying the predicted passenger flow state. When an emergency occurs, an operator manually switches an interface to the emergency passenger flow prediction module, inputs emergency information, the real-time passenger flow monitoring module stops running, and the emergency passenger flow prediction module starts running.

The real-time passenger flow monitoring module mainly comprises the following functions:

1) Road network level real-time passenger flow monitoring

2) Line-level real-time passenger flow monitoring

3) Interval level real-time passenger flow monitoring

4) Station-level real-time passenger flow monitoring

The emergency passenger flow prediction module is used for aiming at emergency information occurring in a road network, firstly, adjusting a train running schedule of an emergency line, respectively predicting OD passenger flows of O/D on the emergency line and OD on a non-emergency line, then, carrying out passenger flow distribution by applying the adjusted train running schedule, a k short path set under the emergency and a passenger flow prediction result under the emergency, outputting related results and carrying out visual early warning display, and simultaneously, issuing arrival information including influence degree information of each arrival station and the like for passengers. And after the emergency is finished, the emergency passenger flow prediction module continues to operate, and the passenger flow distribution is carried out by using the k short path set under the normal condition. And after the emergency is over for 30min, the emergency passenger flow prediction module is completely stopped, and the real-time passenger flow monitoring module continues to normally operate. The K short path set refers to more than one path between any two points in the road network, the first few paths are sorted from small to large according to path impedance, namely the K short path set is used for distributing OD passenger flow to specific paths to obtain section passenger flow.

The emergency passenger flow prediction module mainly comprises the following functions:

1) Train operation planning and simulation

2) Passenger flow prediction and simulation

3) Information service

Specifically, as shown in fig. 2, after entering the emergency passenger flow prediction module, the emergency train operation schedule adjustment and simulation unit adjusts the train operation schedule of the emergency line according to the scheduled train schedule and the emergency information, and performs train operation simulation by using the adjusted schedule, and the adjusted train operation schedule is input into the emergency passenger flow prediction and simulation unit. The emergency passenger flow prediction and simulation unit applies the input emergency information and the historical passenger flow OD data and invokes road network basic data to predict the OD passenger flow, an emergency OD prediction model is adopted for the OD of the O/D on the incident line, a real-time passenger flow prediction model is adopted for the OD of the O/D on the non-incident line, the k short path set updated according to the emergency information, the adjusted train schedule and the OD prediction result are used for passenger flow distribution, relevant passenger flow indexes are output, and visual early warning is carried out. Finally, the emergency information service unit evaluates the affected degree of each station according to the OD shortest path set, the emergency information and the road network basic data under the normal and emergency conditions, and distributes the affected degree information from the station to other stations for the group.

If the accident duration is judged to be changed within the predicted duration, the duration of the event can be manually prolonged or shortened by a certain duration through a system interface, when the duration of the emergency is prolonged, the module correspondingly readjusts the timetable, and correspondingly increases the time step of passenger flow prediction according to the prolonged duration of the time duration; when the duration of the emergency is shortened, the schedule is not required to be readjusted, and only the time step of passenger flow prediction is correspondingly reduced.

When the duration of the event is not longer than the preset time (30 min in this embodiment), the emergency passenger flow prediction module still keeps running, and the k short path set for passenger flow distribution is recovered to the k short path set of the normal condition. When the preset time is full after the emergency is finished, the emergency passenger flow prediction module is completely stopped, and the real-time passenger flow monitoring module starts to normally operate.

Next, each functional unit of the emergency passenger flow prediction module in this embodiment will be specifically described.

1. Emergency train operation plan adjusting and simulating unit

The emergency train operation plan adjustment and simulation unit is based on planned train schedule data, constructs an intelligent man-machine interaction system module for emergency train operation plan adjustment according to emergency information (comprising time, place, reason, type and influence degree of emergency occurrence) and combining with an emergency running train dispatching adjustment strategy, realizes automatic adjustment of a train operation schedule of an emergency line, develops a visualized train operation simulation system module under the emergency condition by utilizing a computer simulation technology on the basis of emergency train operation plan adjustment, automatically lays a train operation diagram according to the adjusted train operation schedule, displays the affected train range, simulates the whole process of all train operation time-space changes on the emergency line in the emergency time period, and provides auxiliary decision support for actual operation.

When the emergency information is judged to be abnormal manually, the early warning is triggered manually, and related accident information is input into the man-machine interaction system for providing basis for train operation plan adjustment. Mainly comprises the following indexes: the emergency occurrence time, the emergency occurrence position, the emergency type, the predicted duration of the emergency, the emergency extension time, the emergency influence direction and the train operation influence.

The train operation schedule records information such as the train operation line, arrival time and departure time of stations along the way, is a basis for grasping the train operation track, and provides basic data support for train operation plan adjustment under emergencies. Mainly comprises the following indexes: train ID, update date, line number, train number, schedule type, schedule number, train direction, station name, station number, station type, arrival time, departure time, vehicle type, train consist, and train stator.

1.1 workflow description

Based on the above, the workflow of the emergency train operation plan adjustment and simulation unit of the present embodiment is shown in fig. 3, and specifically includes the following steps:

1) Manual input of emergency information

Firstly, acquiring train emergency occurrence time, emergency occurrence position, predicted duration, emergency extension time, emergency type, influence direction, train operation influence and train operation interval during faults according to manually input emergency information, wherein the explanation of each index is as follows:

the emergency occurrence time refers to the time when the fault which is manually judged and input occurs.

The emergency occurrence position refers to a fault occurrence place which is manually judged and input, the emergency occurrence position is a station or a section, and if the station type is selected, an accident station of an accident line is input; if the section type is selected, an issuing section of the issuing line is input.

The emergency type refers to the type of equipment and facility faults in burst, and comprises signal faults, vehicle faults, shielding door faults, foreign object invasion and the like.

The predicted duration refers to a duration in which the emergency is manually determined to be likely to last from the occurrence to the end.

The emergency extension time refers to the accident predicted change time input by manpower if the accident duration is manually judged to be changed within the predicted duration. The duration of the accident can be prolonged and shortened according to the actual situation. The first time an incident is triggered, the item defaults to 0.

The impact direction refers to the impact of an emergency on the uplink and downlink of the incident line, and may be uplink, downlink or uplink and downlink.

The train operation influence refers to the influence of the emergency on the train operation, and comprises stopping operation in a fault interval, degrading operation in the fault interval and degrading operation of the fault train.

The train running interval during the accident refers to the minimum tracking time of the train allowed in the accident time which is judged manually.

2) Train operation plan adjustment initial calculation

After the indexes are obtained, the train operation schedule under the emergency is automatically generated and written into a database according to measures such as car buckling, turning back, stopping, adding and starting by combining driving adjustment rules under different emergency conditions based on the current planned train schedule. The response is completed within 30 seconds from the manual triggering of the emergency early warning to the generation of the adjusted train operation schedule data.

3) Train operation plan adjustment correction

After the train operation plan is adjusted for the first time, a train operation plan correction process is required, and the flow is as follows:

after the train operation plan is adjusted, if the accident is judged to be not finished by people and the accident duration time is changed, the emergency event is input by people for prolonging the time, and if the accident duration time is prolonged, the system automatically adjusts and generates a new train operation timetable; if the incident duration is shortened, no readjustment of the schedule is required. The system repeats the above-mentioned judgement process until there is no new emergency and the time is prolonged. If the extension time is not input, defaulting to the duration time point, and ending the accident.

4) Train operation plan adjustment visualization

Finally, based on the new train operation schedule data, a visual train operation diagram is generated through a computer simulation technology, and the affected train range is displayed.

1.2 model Algorithm description

The algorithm for adjusting the running train running chart by combining the original train running schedule according to the manually input emergency information by the emergency train running schedule adjusting and simulating unit of the embodiment is specifically as follows:

the input of the algorithm is train operation schedule data and emergency information data, and the output is the adjusted train schedule. The adjusted new train operation schedule data is provided to the emergency passenger flow prediction module through an interface form to serve as input, and is synchronously stored in a database.

Specifically, the train operation schedule data includes an ID, an update date, a line number, a train number, a schedule type, a schedule number, a direction, a station name, a station number, a station type, an arrival time, a departure time, a vehicle type, a train consist, and a train stator. The emergency information data comprises emergency occurrence time, expected duration, emergency occurrence position and influence direction. The emergency occurrence time needs to be manually input, the time of the occurrence of the event is manually input, the accuracy is high, and the system calculates the occurrence time point by taking the node 15min closest to the occurrence time point as the system according to the principle of 'nearest rounding'. The expected duration needs to be manually input, and is generally based on the duration of the event published by the rail operation enterprises. The emergency occurrence position needs to be manually input and is generally divided into two parts of a station or a section, and if the emergency occurrence position is the station, a station name is input; if the section is the section, the section of the incident published by the rail operation enterprise needs to be input.

Based on the above, the algorithm model is calculated as follows:

first, for convenience of explanation, variables related to the following calculation process will be explained as shown in table 1.

Table 1 operation schedule adjustment variable specification table

If the accident affecting direction is unidirectional, the train number in the direction is adjusted according to the following steps (1) - (3). If the accident affecting direction is bidirectional, the train number of the upward train and the downward train is respectively adjusted according to the following steps (1) - (3). The specific steps are shown in (1) - (3).

(1) Pre-event impact zone adjustment

a. And determining the initial train number adjusted by the pre-event impact area.

Obtaining the initial train number sequence number of the influence area before occurrence according to the train outbound time of the initial station of the fault interval and the fault initial time, and marking as

b. Determining an ending number of passes for a pre-event impact zone adjustment

Obtaining the sequence number of the ending train number of the influence area before departure according to the train outbound time and the fault starting time of the direction starting station, and marking as

c. Drawing wire

The time length of the prior influence area is the time difference between the starting time and the ending time of the prior influence area at the starting time of the starting station,

the serial number set of the line drawing train is

In order to make the adjusted train number of the influence area before occurrence run according to the expected interval of manual input and simultaneously reduce the influence on the running train number as much as possible, the running plans of the stations where the train is located and the subsequent stations when some train numbers are at the fault starting time are cancelled, and the part of the running line of the train number on the running diagram which is shown as the running line of the train number after the fault starting time is deleted, so that the running interval of the train after the line drawing is close to t ₁ . According to the desired interval t of manual input ₁ and Q_before And determining the number of the middle train number and the number of the line drawing of the influence interval before occurrence. The number of lines drawn in advance affected areacount () represents the number of passes in the set, +.>Representing a rounding down. If the result after rounding is less than 1, the result is 1.

Examples: t (T) ₁ ＝6min，t ₁ ＝5min，count(Q _before ) =2, then

d. According to the number N of the drawing lines _before Drawing wire

From Q _before Uniformly and randomly pumping N _before The number of vehicles, the extracted number of vehicles run the line to delete the part after the fault starting time, and the part before the fault starting time is reserved.

(2) Adjustment of the area of incidence

a. Determining initial train number of event zone adjustment

Initial train number of train with accident area adjustment

b. Determining an ending number of passes for incident area adjustment

Obtaining the sequence number of the ending train number of the accident area according to the train outbound time and the fault ending time of the direction starting station, and marking as

c. Drawing wire

The time length of the event zone is the time difference between the starting time of the event zone and the starting time of the ending time of the vehicle at the starting station,

accident area line drawing train number collection

In order to make the adjusted train number of the accident area run according to the expected interval, some train numbers are stopped from the starting station in the direction, and the running line represented by the train number on the running chart is deleted, so that the train running interval after the line drawing is close to t ₁ . According to the desired interval t of manual input ₁ and Q_fault And determining the number of the train number in the accident area. The number of the strings in the accident areacount () represents the number of passes in the set, +.>Representing a rounding down. If the result after rounding is less than 1, the result is 1.

Examples: t (T) ₂ ＝30min，t ₁ ＝5min，count(Q _fault ) =6, then

d. According to the number N of the drawing lines _fault Drawing wire

From Q _fault Uniformly and randomly pumping N _fault The number of vehicles, the extracted running line of vehicles stops from the starting station, and the vehicles are stoppedAnd running the following train number according to the original train schedule.

(3) Accident duration adjustment

If the expected duration of the incident is shortened before the fault ends, the adjusted schedule is maintained and is not modified.

If the system receives an input extending the duration of the incident before the end of the fault, operations (1) and (2) are re-performed based on the adjusted schedule, the new adjusted fromTime is the original toTime, and the new adjusted toTime is the original totime+t _{Extension of} The expected train running interval of the new round of adjustment is the manually input accident prolonged train running interval t ₂ 。

2. Emergency passenger flow prediction and simulation unit

The emergency passenger flow prediction is based on AFC (automatic fare collection) passenger flow data under historical emergency, OD prediction and distribution are carried out based on the occurrence time of the emergency and passenger flow rules within 30min after the emergency is finished, the passenger flow prediction under the emergency is realized through an emergency passenger flow OD prediction and distribution model, indexes such as time granularity OD passenger flow, section passenger flow, interval full rate, station in-out quantity, transfer quantity and OD shortest path and the like are generated in the time range of the emergency and 30min after the emergency is finished, the affected range including affected stations, the number of people affected and the affected time is calculated, and data support and basis are provided for passenger flow early warning, information service and passenger flow induction measures under the emergency.

The historical passenger flow OD data mainly comprises four parts of a starting station, a terminal station, an OD trip starting time and an OD trip passenger flow. The road network basic data mainly comprises four parts, namely site-to-line matching, site spacing, OD travel shortest route and line classification. And providing a basic basis for calculating the OD passenger flow prediction space parameters after the emergency occurs. The emergency information provides a basic basis for calculating the OD passenger flow prediction time parameter after the emergency, and comprises the following indexes: time of emergency, location of emergency, expected duration, time of emergency extension. The output result of the train operation plan adjusting and simulating unit is the train operation schedule adjusted after being influenced by the emergency, and the train operation schedule is obtained by the train operation plan adjusting and simulating unit through an interface form, so that a basic basis is provided for passenger flow distribution and corresponding index calculation.

2.1 workflow description

Based on the above, the workflow of the emergency passenger flow prediction and simulation unit of the present embodiment is shown in fig. 4, and specifically includes the following steps:

1) Sudden event OD passenger flow prediction period specification

Considering that the emergency can cause longer influence on the passenger flow, the OD passenger flow prediction period of the emergency is set to be 30min after the occurrence time and the end of the emergency.

2) Emergency passenger flow prediction flow

Based on a related emergency passenger flow prediction model and a related emergency passenger flow prediction method, the passenger flow OD data is predicted by combining the passenger flow data of the historical emergency. Because the emergency can continuously influence the passenger flow, OD prediction is carried out on the passenger flow of 30min after the occurrence time and the end of the emergency. And for the situation that the O/D is on the accident line, carrying out OD prediction by adopting a passenger flow OD prediction model under the emergency, and for the situation that the O/D is not on the accident line, carrying out OD prediction by adopting a real-time passenger flow prediction model. And (3) carrying out distribution and calculation of each index by utilizing a passenger flow distribution algorithm according to the predicted road network OD and the adjusted train operation schedule.

(1) OD prediction flow of emergency passenger flow

Aiming at the passenger flow prediction of O/D on an incident line, firstly, utilizing emergency information and road network basic data to calculate time parameters and space parameters of a prediction model, and inputting historical OD passenger flow data into an emergency passenger flow OD prediction model to carry out OD passenger flow prediction; and predicting the passenger flow of the O/D which is not on the accident line by adopting a passenger flow prediction model under a real-time passenger flow prediction module. The predicted output is OD passenger flow at 15min granularity.

The duration of the accident can be adjusted in the prediction process, the prediction step length is correspondingly reduced or increased by the model every time the duration of the accident is adjusted according to the actual situation. In the prediction process, whether the accident is ended or not needs to be judged, and the prediction is continued for 30 minutes after the accident is ended, so that whether the prediction is ended or not is judged. The OD prediction flow of the emergency passenger flow is shown in fig. 5, and the specific flow is as follows:

(1) incident information input

The required emergency information data containing indexes include: time of emergency, location of emergency, expected duration, time of emergency extension.

The emergency occurrence time needs to be manually input, and the time of the emergency occurrence is manually input, so that the accuracy is high.

(2) Road network basic data call

The road network basic data mainly comprises four parts, namely site-to-line matching, site spacing, OD travel shortest route and line classification. And providing a basic basis for calculating the OD passenger flow prediction space parameters after the emergency occurs.

(3) Historical incident OD passenger flow data input

The historical passenger flow OD data mainly comprises four parts, namely a starting station, a finishing station, an OD trip starting time and an OD trip passenger flow.

(4) OD passenger flow prediction process

After the historical incident OD passenger flow data, road network basic data and incident information are input, the model calculates time parameters according to incident occurrence time, predicted duration time and incident extension time; and calculating space parameters according to the emergency position, the matching of the station and the line, the spacing between the stations, the shortest distance of OD trip and the line classification. The method comprises the steps that a passenger flow OD prediction model under an emergency is adopted for the passenger flow OD prediction of the O/D on an incident line; and a real-time passenger flow prediction model is adopted for the passenger flow OD prediction of the O/D which is not on the incident line.

(5) OD passenger flow prediction duration correction

And manually judging whether the duration of the emergency and the OD passenger flow prediction time length need to be adjusted. If the OD passenger flow is prolonged, the OD passenger flow result with the increased time length is required to be correspondingly output; if the time is shortened, the accident end time is predicted to be added with the subsequent 30 minutes.

(6) And judging that the event is ended.

After the system predicts the OD passenger flow data with the new 15min granularity, whether the event is ended or not is judged, whether the predicted time length reaches the occurrence time length of the emergency and 30min after the ending is finished, and if the predicted time length does not reach the requirement, the prediction is continued; if the request is met, the prediction is exited.

(2) Emergency passenger flow simulation flow

Firstly, the OD passenger flow data obtained through prediction needs to be input in the passenger flow simulation of the emergency, meanwhile, k short path sets under the emergency generated after path impedance is calculated by utilizing the train operation schedule obtained through adjustment are utilized to carry out passenger flow distribution, but the passenger flow distribution is carried out by adopting a normal k short path set before change aiming at the OD in the road network when the emergency occurs, and similarly, when the duration of the emergency is finished, k short path sets for passenger flow distribution are restored to be k short path sets of normal conditions, but the passenger flow distribution is carried out by adopting the k short path sets of the emergency of a change sign aiming at the OD in the road network when the emergency is finished. The passenger flow simulation comprises two output aspects, namely calculation of various passenger flow indexes, and arrival quantity, transfer quantity, section passenger flow and section full rate under the granularity of 5 min. Obtaining an OD shortest path in a passenger flow distribution process; on the other hand, the calculation of the affected area is respectively the affected station, the number of people affected and the affected time. The simulation flow of the emergency passenger flow is shown in fig. 6, and the specific flow is as follows:

(1) Input of parameters

And (3) inputting the OD passenger flow data obtained through prediction in the previous step, inputting an adjusted train operation schedule, updating the impedance according to the time information of the adjusted train operation plan, generating a k short path set under an emergency and inputting the k short path set.

(2) Index generation

Passenger flow distribution index: and calculating the passenger flow simulation result to obtain the inbound quantity, outbound quantity, transfer quantity, section passenger flow and section full load rate under the granularity of 5 minutes.

Other indexes: the passenger flow distribution process obtains an OD shortest path.

Affected area: the calculation of the affected range comprises three parts, namely an affected station, an affected person number and an affected time.

(3) Prediction step adjustment

If the accident is not finished and the accident duration is artificially judged to be changed, the accident duration can be prolonged or shortened. When the duration of the accident is prolonged, the train running schedule is required to be readjusted, the step length of OD passenger flow prediction is correspondingly increased, and passenger flow distribution is carried out by utilizing the readjusted schedule, the predicted OD passenger flow after the step length is increased and a k short path set under the emergency; when the accident duration is shortened, the schedule does not need to be adjusted again, the OD prediction step length is shortened, and the OD prediction result is input into passenger flow distribution.

(4) Module stop

And when the emergency is finished and the time is less than 30 minutes after the emergency is finished, the k short path set of the passenger flow distribution application is recovered to the k short path set of the normal condition, and the emergency passenger flow prediction and simulation module keeps running. And stopping the operation of the emergency passenger flow prediction and simulation module when the emergency is over for 30min after the emergency is over.

2.2 model algorithm description:

the emergency passenger flow prediction and simulation module mainly comprises two important links: in the two links, the OD passenger flow prediction applies a passenger flow OD prediction model, the passenger flow distribution link needs to update the impedance of each path of a k short path set under normal conditions, and then the model is applied to distribute the passenger flow of each path, and the model mainly applied in the link is a k short path set generation model and a passenger flow distribution model.

1) Passenger flow OD prediction model

(1) Model input

The input parameters of the passenger flow OD prediction model are time parameters and space parameters. The time parameters include: the time point of occurrence, the estimated duration, the duration of the event, and the event influencing parameters. The spatial parameters include: the accident point, the shortest distance between the O/D and the accident point, the shortest distance between the OD and the trip, and the space influence parameters.

(1) Time parameter

a. Time point of occurrence: the time point of the event needs to be manually input, the time of the event is manually input, the accuracy is minute, and the system takes the node 15min closest to the time point of the event as the system calculation time point of the event according to the principle of 'nearest rounding'.

b. Estimating the duration: the estimated occurrence time of the event needs to be manually input by a human operator, and the manual input time of the dispatcher is taken as the reference.

c. Event duration influence time

The emergency duration impact time may take a fixed default length of time, typically 30 minutes.

d. Time-influencing parameters

And automatically adjusting the time to be the nearest 15min node according to the event occurrence time input manually by the system, and converting the time into system time data which is Arabic numeral 1. And marking time data by Arabic numerals every 15min in sequence in the pre-estimated time length of occurrence and the default duration influence time to obtain time influence parameters.

(2) Spatial parameters

a. Place of occurrence: the accident site needs to be manually input and is generally divided into two parts of a station or a section, and if the accident site is the station, a station name is input; if the section is the case, the event section needs to be input.

b.O/D shortest distance from the incident point: the system calculates the shortest distance of O or D from the departure station or section on the departure line according to the inputted departure point.

Od travel shortest: the shortest path of OD going out is the quantity that needs to be imported into the system in advance, and is generally unchanged, and when there is a new line or site and an emergency occurs, the corresponding update needs to be made.

d. Spatial influencing parameters: and according to the basic data link space distance data, calculating the distance between O or D and the accident station or the section and the OD shortest distance to obtain space influence parameters.

(2) Model calculation

in the formula ,-j period, spatial influencing parameter +.>The OD passenger flow of (1) is offset percentage of the normal passenger flow mean value;

j-time influencing parameters;

-j time period, the space between the origin and destination od affects the parameters;

t _e -a predicted duration of the event;

t _ed -the sum of the estimated duration of the accident and the duration of the impact;

a, B, C, D, E, F, J, H, I, J, K-model parameters.

(3) Model output

The model outputs the OD passenger flow deviation percentage, and the OD passenger flow predicted value under the emergency can be calculated by the following formula by means of the deviation percentage.

in the formula ,V_od -OD passenger flow prediction value in case of emergency;

-OD passenger flow means in normal state;

-j period, spatial influencing parameter +.>The OD passenger flow of (c) is offset by a percentage from the normal passenger flow mean.

2) k short path set update model

(1) Model input

And updating the impedance of the k short path set under normal conditions by using the train running interval according to the position of the k short path set under the emergency and reordering according to the updating result.

Location of emergency occurrence: the manually input emergency occurrence position is a station or an interval where the emergency occurs.

Train operation interval during incident: and the minimum tracking time of the train allowed in the accident time is judged manually.

Normally k short path sets: under the condition of no emergency, the system calculates the k short path set among the ODs obtained after the impedance is calculated according to the algorithm.

(2) Model calculation

In the formula, R (i, j) -the impedance on a path R taking an i station as a starting point and a j station as an end point under the emergency; r is R ₀ (i, j) -normally the impedance on path r with i station as start j station as end;

h, train operation interval in accident period;

d—a set of paths affected by the event.

According to the above equation, only the path affected by the emergency is updated, and the unaffected path impedance is kept consistent with the path impedance under normal conditions.

(3) Model output

The model outputs the impedance of each path, and the k short path set can be obtained according to the calculated path impedance and the re-ordering according to the calculated impedance.

3) Passenger flow distribution model

(1) Model input

The model input is to predict OD passenger flow, k short path set and adjust the train operation schedule.

And (5) predicting OD passenger flow: and a prediction result output in the passenger flow OD prediction link.

K short path set: the k short path set generates a result output by the model and corresponding impedance of each path.

Adjusting the resulting train schedule: and adjusting the emergency train operation plan and simulating the train operation schedule under the output emergency.

(2) Model calculation

The probability that a path is selected at the corresponding impedance is represented by a logic model as follows:

in the formula ,C_m,i -a cost function of factors influencing passenger path selection;

β _m,i -the weight of the term factor;

P _i the probability that path i in the set of paths is selected,

constructing a multipath probability distribution model, combining the characteristics of the determined path distribution model and the random multipath distribution model, and constructing a multipath probability distribution model based on user balance, wherein the multipath probability distribution model is as follows:

wherein, θ: the randomness of the model is described.

(3) Model output

And the probability of each path in the path set being selected can be used for obtaining the passenger flow on each path according to the multi-path probability distribution model.

2.3 output of results

1) Prediction result of OD passenger flow of emergency: and adding 15min granularity OD passenger flow data within a 30min range after the emergency occurrence time is over.

2) Passenger flow indexes obtained by passenger flow simulation of emergency events: the section passenger flow index comprises section passenger flow volume with granularity of 5min and section full load rate; the station passenger flow index comprises 5min granularity inbound amount, outbound amount and transfer amount.

(1) Section passenger flow volume: and the number of passengers passing through a section of the subway line in the same direction within a certain time.

(2) Interval full load rate: and in unit time, the ratio of the passenger flow volume of the unidirectional section of the operation line to the transport capacity of the corresponding section reflects the congestion condition of the train of the section in unit time of the train. The calculation method comprises the following steps:

(3) Station arrival amount: and calculating the number of passengers entering the station according to the predicted passenger flow OD data.

(4) Station outbound volume: the number of outbound passengers is calculated from the predicted OD data of the passenger flow.

(5) Transfer amount of transfer station: and in the counting period, the number of passengers is transferred in all directions between transfer station lines.

3) Affected range obtained by emergency passenger flow simulation

The calculation of the affected range comprises three parts, namely an affected station, an affected person number and an affected time.

The affected stations are all stations contained in the emergency occurrence line.

The number of the affected people is the sum of the historical synchronous incoming and outgoing amounts of the incident line in the incident period.

The affected time is the occurrence time of the emergency and 30min after the end of the emergency.

4) Other outputs: OD shortest path for emergency scenario.

3. Emergency information service unit

The emergency information service is based on road network basic data, basic parameters and emergency information, takes an emergency passenger flow prediction module as a basic data source, compares an OD shortest path set under normal conditions with an OD shortest path set after the occurrence of an emergency, and OD travel time under normal conditions with the OD travel time after the occurrence of the emergency based on road network changes in different emergency scenes, and judges whether the travel time reaching other destination sites changes or not according to each site. The method comprises the steps of issuing influence degree information of paths between ODs marked by red and green colors for a passenger group, and simultaneously storing individual path guidance information of the passengers, namely recommended path information between input origin and destination points of the passengers. And providing data support for related functions of the passenger inquiry system and the mobile client network operation information inquiry.

The road network basic data mainly comprises four parts of site-to-line matching, site spacing, OD travel shortest route and line classification. And one of important basic data for inducing information formulation during road network basic data. The emergency information data is used to present the affected time period. The method comprises the following indexes: the time of occurrence of the emergency, the location of occurrence of the emergency, the type of emergency, the expected duration, the time of extension of the emergency, the direction of influence and the influence of train operation. And (5) normally, an OD shortest path set and OD travel time management. The OD shortest path set and the OD travel time under normal conditions are data tables for recording the shortest path from each station to each arrival station and the corresponding travel time under normal conditions without emergency. The method is used for comparing with an OD shortest path data set and OD travel time under emergency conditions and making induction information. Under normal conditions, the OD shortest path set and the OD travel time comprise the following indexes: start station number, end station number, shortest path, travel time. And (5) after the emergency occurs, an OD shortest path set and OD travel time are managed. The OD shortest path set and the OD travel time after the emergency is a data table for recording the shortest path from each station to each reachable station and the corresponding travel time under the condition of recording the topology change of the road network after the emergency. The data is obtained from a passenger flow prediction module. The OD shortest path set and OD travel time after the emergency is as follows: start station number, end station number, shortest path, travel time. The emergency information data is input to the man-machine interaction system when the alarm is triggered manually; and after the emergency occurs, the OD shortest path set and the OD travel time are compared with the OD shortest path set and the OD travel time stored in the system under the normal condition so as to evaluate the affected degree of the paths between the ODs.

3.1 workflow description

Based on the above, the workflow of the emergency information service unit of the present embodiment is as follows:

1) The information release period specifies: after the emergency occurs, information release is started until the emergency is finished and all stations recover to the 'no influence' state, and information release is finished.

2) Information service flow management: when an emergency occurs, the information release module is started. The flow of the guidance information release for the group is shown in fig. 7. The information service includes two aspects: generating influence degree information of the self station to other stations of the group, and storing personalized path guidance information of the passengers of the individual.

(1) Generating influence degree information aiming at the local station to other stations of the group: and comparing the OD shortest path set and the OD travel time after the occurrence of the normal condition and the emergency according to the input event information, the OD shortest path set and the OD travel time after the occurrence of the emergency, road network basic data called by the module, the OD shortest path set and the OD travel time under the normal condition.

Judging whether the shortest path and travel time between the ODs are changed under normal conditions and after an emergency, and if the shortest path and travel time are unchanged, the final arrival station is not affected by the accident; if the shortest path or trip time length changes, the final arrival station is affected by the emergency.

The link outputs the influence degree information of the local station to other stations for the group. This information may be published through in-station PIS systems, mobile clients, and passenger inquiry systems.

(2) Personalized path guidance information preservation for individual passengers: the passengers input the start station and the end station according to personal demands, the system centrally selects the shortest path between the corresponding ODs from the OD shortest paths after the emergency occurs, stores the shortest paths, and does not visually display the personalized path guidance information of the individual passengers.

3.2 output of results

The module only outputs the influence degree information from the station to other stations for the group.

1) The influence degree information of the local station to other stations of the group is a data table for recording the influence degree among OD, and the specific indexes include a time stamp, a starting station number, a final station number, an influenced grade and a rendering color.

(1) Timestamp: and when the affected degree information of each station is updated, recording the time stamp.

(2) Start station number: and (5) when evaluating the influence degree, taking the station number as a starting point.

(3) Terminating station number: and (5) carrying out the affected degree evaluation, wherein the terminal station number is the station number.

(4) Affected rating: the evaluation of the degree of influence of the emergency among certain ODs is divided into two stages: unaffected, affected. Specific evaluation procedures are described in the section above.

(5) Color rendering: the corresponding relation of the color rendering according to the affected degree level is shown in table 2.

TABLE 2 affected class and color rendering significance for stops

In summary, the urban rail transit emergency online passenger flow prediction and simulation system of the embodiment realizes high-precision prediction of the passenger flow evolution state in the short future under normal conditions through the real-time passenger flow monitoring module, and generates the passenger flow index of the urban rail transit real-time granularity; the predicted evolution state of the passenger flow is visually displayed, so that the real-time passenger flow is monitored; the train running schedule is adjusted according to the emergency information occurring in the road network through the emergency passenger flow prediction module; predicting the distribution condition of the passenger flow under the condition of being influenced by the event according to the adjusted timetable; and issues the affected degree information of each terminal to the passengers. Therefore, the passenger flow can be monitored on line and in real time for normal conditions and emergency conditions.

Furthermore, it should be noted that the present invention can be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.

It is finally pointed out that the above description of the preferred embodiments of the invention, it being understood that although preferred embodiments of the invention have been described, it will be obvious to those skilled in the art that, once the basic inventive concepts of the invention are known, several modifications and adaptations can be made without departing from the principles of the invention, and these modifications and adaptations are intended to be within the scope of the invention. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Claims

1. The urban rail transit emergency online passenger flow prediction and simulation system is characterized by comprising a real-time passenger flow monitoring module and an emergency passenger flow prediction module; wherein,

the emergency train operation plan adjustment and simulation unit is used for adjusting a train operation schedule aiming at the emergency occurring in the road network; the emergency passenger flow prediction and simulation unit is used for predicting passenger flow distribution conditions under the condition of being influenced by the event according to the adjusted train operation schedule; the emergency information service unit is used for issuing the affected degree information of each terminal station to the passenger;

the emergency train operation plan adjusting and simulating unit is specifically used for:

According to the planned train operation schedule data and the emergency information, the train operation schedule of the emergency line is adjusted by combining with a driving scheduling adjustment strategy under the emergency condition, and train operation simulation is carried out by using the adjusted operation schedule; inputting the adjusted operation schedule into the emergency passenger flow prediction and simulation unit; the emergency information comprises emergency occurrence time, emergency occurrence position, emergency type, emergency predicted duration, emergency extension time, emergency influence direction and train operation influence degree; the train operation schedule data comprises a train ID, an update date, a line number, a train number, a schedule type, a schedule number, a train direction, a station name, a station number, a station type, arrival time, departure time, a vehicle type, a train grouping and a train stator;

the emergency passenger flow prediction and simulation unit is specifically used for:

according to the emergency information, historical passenger flow OD data and road network basic data, carrying out OD passenger flow prediction, adopting a preset emergency OD prediction model for the OD of an O/D on an emergency line, adopting a preset real-time passenger flow prediction model for the OD of the O/D on an emergency line, and according to a k short path set updated by the emergency information, an adjusted train running schedule and an OD passenger flow prediction result, carrying out passenger flow distribution, outputting relevant passenger flow indexes and carrying out visual early warning;

The emergency information service unit is specifically configured to:

evaluating the affected degree of each station according to the OD shortest path set, the emergency information and the road network basic data under the normal and emergency conditions, and distributing the affected degree information from the station to other stations for the group;

if the duration of the emergency is determined to be changed within the predicted duration, the emergency train operation plan adjustment and simulation unit is specifically configured to:

when the duration of the emergency is prolonged, correspondingly readjusting the train operation schedule, and increasing the time step of passenger flow prediction according to the prolonged duration of the emergency; when the duration of the emergency is shortened, reducing the time step of passenger flow prediction according to the shortened duration of the emergency;

when the duration of the emergency is not longer than the preset duration, the emergency passenger flow prediction module still keeps running; when the preset time is full after the emergency is finished, the emergency passenger flow prediction module is completely stopped, and the real-time passenger flow monitoring module starts to normally operate;

Based on the current planned train operation schedule, by combining with driving adjustment rules under different emergency conditions, automatically generating a train operation schedule under the emergency, after the train operation schedule is adjusted, if the artificial judgment of the emergency is not finished and the emergency duration is changed, manually inputting the emergency extension time, and if the emergency duration is prolonged, automatically adjusting and generating a new train operation schedule; manually judging whether the emergency is ended or not and whether the duration of the emergency is changed or not again until no new emergency is input for a prolonged time; if the extension time is not input, defaulting to a duration time point, and ending the emergency;

based on the newly generated train operation schedule data, generating a visual train operation diagram through a computer simulation technology, and displaying the affected train range;

the method for automatically generating the train operation schedule under the emergency by combining the driving adjustment rules under different emergency conditions based on the current planned train operation schedule comprises the following steps:

s11, obtaining the initial train number sequence number of the pre-event influence area according to the train outbound time of the emergency interval initial station and the emergency initial time, and marking as The expression is as follows:

wherein ,ud_i,s Representing departure time of train number i at S station in operation schedule before adjustment, S _fromLocation A starting station representing an emergency interval, and a from time representing the starting time of the emergency;

s12, according to the train outbound time and fault start of the current direction start stationTime, obtain the end train number of the affected area before issue, record asThe expression is as follows:

wherein ,S₁ Representing the origin of the train;

obtaining the serial number set of the line drawing train number as

S2, adjusting an incident area, which comprises the following steps:

Wherein, toTime represents the ending time of the emergency;

determining a line drawing train number set in an emergency region

2. The urban rail transit emergency online passenger flow prediction and simulation system according to claim 1, wherein the passenger flow index comprises: ingress and egress volume, transfer volume, passenger traffic volume, and section passenger traffic volume.

3. The urban rail transit emergency online passenger flow prediction and simulation system according to claim 1, wherein the real-time passenger flow monitoring module operates when no emergency occurs in the road network, and the emergency passenger flow prediction module stops; when an emergency occurs in the road network, the emergency passenger flow prediction module starts to operate according to an operation instruction of an operator, and the real-time passenger flow monitoring module stops.