CN108846514A - A kind of emergency event lower railway traffic passenger flow evacuation needing forecasting method - Google Patents
A kind of emergency event lower railway traffic passenger flow evacuation needing forecasting method Download PDFInfo
- Publication number
- CN108846514A CN108846514A CN201810580795.6A CN201810580795A CN108846514A CN 108846514 A CN108846514 A CN 108846514A CN 201810580795 A CN201810580795 A CN 201810580795A CN 108846514 A CN108846514 A CN 108846514A
- Authority
- CN
- China
- Prior art keywords
- website
- passenger flow
- emergency event
- demand
- rail traffic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013277 forecasting method Methods 0.000 title claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 39
- 239000011159 matrix material Substances 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 238000000513 principal component analysis Methods 0.000 claims abstract description 9
- 230000014759 maintenance of location Effects 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 24
- 230000002123 temporal effect Effects 0.000 claims description 3
- 239000000523 sample Substances 0.000 description 16
- 238000012546 transfer Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000009825 accumulation Methods 0.000 description 3
- 238000009795 derivation Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of emergency event lower railway traffic passenger flows to evacuate needing forecasting method, includes the following steps：1) according to rail traffic automatic fare collection system history ticket card data acquisition normality transport need sample set in short-term, and based on normality in short-term transport need sample set, multi-step prediction is carried out using the hybrid prediction model based on principal component analysis, obtains subway Travel Demand Forecasting result in short-term under normality；2) Rail traffic network digraph is established according to Rail traffic network topological relation；3) rail traffic travelling OD is allocated to obtain allocation matrix F；4) according to the feature of emergency event and prior information, in conjunction with prediction result, the calculation formula of each website evacuation demand of the whole network during obtaining outgoing event influence；5) temporal-spatial evolution of demand is evacuated during obtaining emergency event influence using emulation, and obtains peak retention.Compared with prior art, the present invention have many advantages, such as strong operability, it is practical, can directly apply, method it is complete.
Description
Technical field
The present invention relates to public transport sudden incidents report fields, hand over more particularly, to a kind of emergency event lower railway
Logical passenger flow evacuates needing forecasting method.
Background technique
All kinds of emergency events in recent years (including production event, operation accident, extreme weather, social public accident and terror
Attack etc.) frequently occur, in addition in fragile traffic system, will cause many adverse effects, including network traffic it is crowded,
It interrupts completely, different property loss of degree even casualties etc..This largely seriously reduces traffic system
Operational reliability and bearing capacity become the severe challenge of field of traffic at this stage.
Traffic Demand Forecasting is one of research hotspot of field of traffic in short-term, and prediction result can be applied to active traffic control
The fields such as system, traffic guidance, for its measure implement and formulate guidance is provided, thus be the key that intelligent transportation system composition because
Element.And under emergency event, the sustainable scientific emergency response of real-time, accurate demand forecasting information, it appears more crucial.It is existing
Forecast of Traffic Demand in short-term be all mostly predicted in normality using historical data, and it is abnormal under phase
Pass research is less, and one of Major Difficulties are that targetedly historical data is less.Generally, the abnormal thing that especially happens suddenly
When part leads to interruption of communication, data collection is difficult, and sample is sparse, and real time data source cannot even more be protected, if
Using with research method as normality lower class, i.e., modeled, solved simultaneously as sample set dependent on targetedly historical data
It is relatively difficult to achieve therefrom to probe into its rule.Currently, not retrieving the patent of invention that can be solved the above problems also.
By the retrieval to existing literature and patent, up to the present, dredged to the track traffic for passenger flow under emergency event
The more relevant content of scattered requirement forecasting mainly summarizes, it is as follows to comment：
(1) Ricardo Silva et al. exists《Predicting traffic volumes and estimating the
effects of shocks in massive transportation systems》(Proceedings of the
National Academy of Sciences of the United States of America,2015,112(18):
Using London track traffic accident database combination regression and statistical method to being predicted in the case of interruption in 5643-8).Its
Method is primarily limited in use：The evacuation demand under accident condition need to be carried out using more comprehensive historical record pre-
It surveys.
(2) Yang Li et al. people exists《Forecasting short-term subway passenger flow under
special events scenarios using multiscale radial basis function networks》
(Transportation Research Part C Emerging Technologies,2017,77:It is utilized in 306-328)
Beijing Rail Transit enter the station data under emergency event outbound traffic predict, using machine learning method according to real time data into
Row prediction, adaptability of the emphasis boosting algorithm to abnormal conditions.Its method is limited in use：Output result is emergency event
Under the outbound volume of the flow of passengers, directly evacuation work cannot be instructed, and need the real time data under the conditions of emergency event as mould
Type input.
(3) Chinese Patent Application No.：201710285974.2 patent name is：City rail is handed under a kind of emergency event
Logical passenger flow forecasting.Its patent is limited in actual use：Its patent exports result as under the influence of emergency event
Enter the station the volume of the flow of passengers, and needs according to the corresponding data under emergency event, detailed for carrying out to passenger choice behavior under emergency event
Thin effect assessment.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of emergency event lower rails
Traffic passenger flow in road evacuates needing forecasting method.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of emergency event lower railway traffic passenger flow evacuation needing forecasting method, includes the following steps：
1) according to rail traffic automatic fare collection system history ticket card data acquisition normality transport need sample set in short-term, and with
Normality based on transport need sample set, carries out multi-step prediction using the hybrid prediction model based on principal component analysis, obtains in short-term
Take under normality subway Travel Demand Forecasting result in short-term；
2) Rail traffic network digraph is established according to Rail traffic network topological relation；
3) rail traffic travelling OD is allocated to obtain allocation matrix F；
4) according to the feature of emergency event and prior information, in conjunction with gained requirement forecasting in step 1) as a result, being met accident
The calculation formula of each website evacuation demand of the whole network during part influences；
5) temporal-spatial evolution of demand is evacuated during obtaining emergency event influence using emulation, and obtains peak retention.
The step 1) specifically includes following steps：
11) rail traffic automatic fare collection system history ticket card data are pre-processed, including data cleansing, stroke
Match, by required time granularity carry out collection based on, obtain normality transport need sample set in short-term；
12) to normality, transport need sample set carries out principal component analysis in short-term, obtains principal component and corresponding load；
13) modeling and forecasting is carried out to load and its corresponding temporal characteristics, by the prediction gained corresponding principal component of load
It is combined as subway under normality Traffic Demand Forecasting result in short-term.
The predicted time window range of the hybrid prediction model is longer than the rail transportation ticket-card data back period.
In the step 3), stream that allocation matrix F is carried by each side of Rail traffic network digraph at all moment
Amount, calculating formula are：
FT=XT×U
Wherein, the OD trip matrix that X is orbit traffic full network one day, U are the matrix of shortest path.
In the step 4), the calculation formula of each website evacuation demand of the whole network is during event influences：
Wherein,For the evacuation demand of moment t website r, ft eThe flow on the e of side, and f are loaded for moment tt e∈ F, t0
For the interruption moment of website r.
In the step 5), temporal-spatial evolution includes multiple emulation cycles, and successively sequence includes multiplying to each emulation cycle
Visitor is entered the station process, process of getting off and process of getting on the bus.
The passenger enters the station process using orbit traffic full network one day OD trip matrix as input.
The process of getting off is specially：
511) path after obtaining the distribution of rail traffic travelling OD, and obtain either site in all paths after distribution
Next website；
512) to any side e in the set E on all sides of Rail traffic network digraph full figure, the directed to station of side e is obtained
Point vnAnd the upper a line e of side ep；
513) the Website Hosting V traversed in Rail traffic network digraph judges whether website d is side to either site d
The source site of e, if it is not, step 514) is then carried out, if so, carrying out step 516)；
514) it realizes and gets off, judge the following conditions：
Condition 1：With OD in corresponding path in shortest path allocation matrix U, the next stop of the source site e (o) of side e
Point is not vn, the OD is to the source site e (o) and website d for side e；
Condition 2：With OD in corresponding path in shortest path allocation matrix U, next website of the source site e (o) of e
Section V is interrupted under emergency eventdisrIt is interior；
Condition 3：Current time t belongs to the period T that burst is interrupted；
If condition 1 be true or condition 2 and condition 3 at the same be it is true, carry out step 515), otherwise number of people in car is constant, then t
+ 1 moment went to the passenger flow of website d in the train on the e of sideThe passenger flow of website d is gone in the train on the e of side with t momentIt is identical, i.e.,And carry out step 516)；
515) the passenger flow demand that t moment goes to d in website o wish is updatedI.e.Wherein,The passenger flow demand of d is gone in website o wish for the t-1 moment；
516) return step 513 if all websites not yet traverse end), next website is traversed, if all websites traverse
Terminate, carries out step 517)；
517) return step 512 if all sides not yet traverse end), lower one side is traversed, if all sides traversal terminates,
Terminate.
The process of getting on the bus is specially：
521) all affected OD pairs are obtained in the whole network according to interruption section, and is denoted as set disrupted_od；
522) ODs all to Rail traffic network the whole network update t moment in website o, wish goes to the visitor of d to traversing
Stream demand Wherein,It is the t-1 moment in website o, wish goes to the passenger flow demand of d,
It enters the station for t moment from o, wish goes to the transport need of d；
523) side es all to Rail traffic network digraph full figure are traversed, and obtain the direction website v of side en；
524) all websites of Rail traffic network digraph full figure are traversed, if either site d meets condition：
With OD in corresponding path in shortest path allocation matrix U, if the next point of the o point e (o) of side e is vn, then
The list board_list that gets on the bus is added in website d；
525) it traverses next website and carries out step 524) if all websites, which not yet traverse, to be terminated, if all websites traverse
Terminate, then carries out step 526)；
526) all websites of Rail traffic network digraph full figure are traversed, to any website d：
Obtain demand of always ridingWith total vehicle carrying capacitySmaller value w in the two.
If w > 0, the passenger flow that t moment goes to d in the train on the e of side is updated Otherwise it carries out
Step 527), wherein ceFor the transporting power of side e；
527) return step 526 if all websites not yet traverse end), next website is traversed, if all websites traverse
Terminate, carries out step 528)；
528) return step 522 if all sides not yet traverse end), lower one side is traversed, if all sides traversal terminates,
Terminate.
Compared with prior art, the present invention has the following advantages that：
The present invention provides a kind of, and the emergency event lower railway traffic passenger flow based on normality historical data evacuates requirement forecasting
Method, present invention is primarily based on the historical datas under normality, such as subway ticket card data to carry out, and does not need to collect a large amount of emergency events
Under historical data, thus strong operability is practical；And the principle of the present invention is mathematical derivation, Computer Simulation, algorithm
Simplicity, resource cost is few, computational efficiency is high, and the present invention has originality.Existing background technique does not have specific aim prediction burst thing
Track traffic for passenger flow under part evacuates demand, and the present invention innovatively proposes a kind of burst thing based on normality historical data
Part lower railway traffic passenger flow evacuates needing forecasting method, and the present invention is carried out using two stages, and wherein the first stage carries out normal
Rail traffic Traffic Demand Forecasting under state, second stage carry out evacuation demand and its temporal-spatial evolution prediction under emergency event,
It can directly apply, method is complete.
Detailed description of the invention
Fig. 1 is the flow chart that a kind of emergency event lower railway traffic passenger flow of the present invention evacuates needing forecasting method.
Fig. 2 is to evacuate demand sample calculation in the case of single site (non-transfer stop) is interrupted.
Fig. 3 is evacuation demand sample calculation of breaking in the case of single site (transfer stop) is interrupted.
Fig. 4 is embodiment associated stations schematic diagram in specific embodiment.
Fig. 5 is the evacuation requirement forecasting result figure of embodiment 1 in specific embodiment.
Fig. 6 is the evacuation requirement forecasting result figure of embodiment 2 in specific embodiment.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment
The purpose of the present invention is in view of the deficiencies of the prior art, propose that a kind of emergency event lower railway traffic passenger flow evacuation needs
Prediction technique is sought, it is made to solve deficiency present in background technique, realizes emergency event lower railway traffic practical, convenient for operation
Passenger flow evacuates requirement forecasting, to instruct the optimal enforcement of formulation and the emergency evacuation measure of emergency preplan.
The present invention carries out pretreatment to rail traffic automatic fare collection system history ticket card data first and extracts that obtain normality short
When transport need sample set.Based on the sample set, it is pre- that multistep is carried out using the hybrid prediction model based on principal component analysis
It surveys, wherein predicted time window range is longer than the rail transportation ticket-card data back period；Secondly, being closed according to Rail traffic network topology
System establishes Rail traffic network digraph；Again, rail traffic travelling OD (Origin-Destination, OD) is allocated
Obtain allocation matrix；Later, according to the feature of emergency event and prior information, in conjunction under scientific hypothesis and aforementioned gained normality
Traffic Demand Forecasting is as a result, each website of the whole network evacuates the calculation formula of demand during deriving outgoing event influence in short-term；It is final to use
Emulation obtains the temporal-spatial evolution that demand is evacuated during emergency event influence.
In order to achieve the above objectives, solution of the invention is：A kind of emergency event lower railway traffic passenger flow evacuation demand
Prediction technique, to avoid the dependence to emergency event historical sample, the historical data being based primarily upon under normality, using number as far as possible
The technological means such as derivation, Computer Simulation are learned, being divided into two stages, totally five steps carry out.
First stage carries out normality lower railway traffic Traffic Demand Forecasting in short-term, is specifically divided into two steps.1) to rail traffic
Automatic fare collection system history ticket card data are pre-processed, including data cleansing, run length code matching, by required time granularity carry out
Collection meter, obtains normality transport need sample set in short-term；Based on the sample set, using the hybrid predicting based on principal component analysis
Model carries out multi-step prediction, and wherein predicted time window range should be longer than the rail transportation ticket-card data back update cycle.
Second stage carries out evacuation demand and its temporal-spatial evolution prediction under emergency event, is specifically divided into three steps.1) basis
Rail traffic network topological relation establishes Rail traffic network digraph；2) rail traffic travelling OD is allocated and is distributed
Matrix；3) classify to emergency event, according to emergency event time of origin, interrupt website number, whether the website containing transfer divides
Class.According to the feature of different type emergency event, in conjunction with to traveler behavior reasonable assumption and the first stage obtained in often
Under state in short-term subway Travel Demand Forecasting as a result, derive interrupt influence during in rail transit network each website evacuation demand meter
Calculate formula；4) according to the evacuation demand calculation formula of upper step, evacuated during obtaining emergency event using emulation and influencing demand when
Evacuation demand at sky evolution, i.e. any time, point, while providing key index peak stranded crowd number.
Specifically, the present invention is achieved by the following technical solutions.
Step 1：Multistep based on principal component analysis Traffic Demand Forecasting in short-term
Rail traffic automatic fare collection system history ticket card data are pre-processed, including data cleansing, run length code matching, are pressed
Required time granularity carries out collection meter, obtains normality transport need sample set in short-term；Based on the sample set, first it is carried out
Principal component analysis obtains principal component and corresponding load.Modeling and forecasting is carried out to load and its corresponding temporal characteristics later, can be adopted
With multiple regression or machine learning.The corresponding principal component of prediction gained load combines the subway traffic in short-term obtained under normality to need
Seek prediction result.In the case that the step is meant to ensure that acceptable precision of prediction, reliability, the more step numbers of forward prediction,
Multi-step prediction result can be it is abnormal under traffic administration control provide the longer decision-making time to more be of practical significance.
Step 2：The building of Rail traffic network digraph
It constructs Rail traffic network digraph D (V, E), the digraph that figure D is made of point set and side collection, wherein V right and wrong
Empty but limited vertex set, E are the set on the relationship side between description combines.Company of the website as the vertex in figure, between two stations
It connects as the side in figure, is bi-directionally connected as two sides.
For any side e, e ∈ E, e=(r, v), i.e. the starting point of e, terminal is respectively r, v for definition.Definition is for any top
Point v, v ∈ V, adjacent sites are denoted as r ∈ N (v).
Step 3：OD distribution in Rail traffic network
If orbit traffic full network one day OD trip matrix is X(m,t), allocation matrix is to describe the matrix of shortest path
U(m,n), the matrix F that is obtained after distribution(n,t)The flow carried by each side at all moment, while we define f ∈ F,For
Moment t loads the flow on the e of side.Wherein m=| OD |, n=| V |.
FT=XT×U (1)
Wherein the method that obtains of U includes：(1) critical path method (CPM), what is simplified thinks that each side right heavy phase is same, using Dijkstra
Algorithm solves to obtain；(2) under conditions of having metro operation related data, consider a possibility that secondary short-circuit, selected using multipath
Probabilistic method series methods are selected to obtain；(3) if or having more detailed data that can directly adopt such as the sorting routing table of actual use.
Step 4：The evacuation demand derivation of equation
Emergency event makes part station or section stop transport, and inevitably resulting in part passenger cannot be arrived by original plan path
Up to destination, therefore part passenger is detained, and produces evacuation demand.It makes the assumption that herein, if without sufficient emergency event
Relevant information, then most of traveler is still gone on a journey according to original plan.I.e. former trip route will supported by the passenger in interruption section
It gets off when station previous up to interruption section, generates delay.With the increase of break period, interrupts the forward and backward website that closes in section and evacuate
Demand is constantly accumulated.
Wherein t > t0,For the evacuation demand of moment t website r, f ∈ F.
Formula (2) is further described.For singly standing and be interrupted for non-transfer stop, such as Fig. 2, if website 1 when
It carves t0 to interrupt, passenger flow will be trapped in website 2 and website 3, be endpoint stations referred to here as website both ends adjacent sites are interrupted.Website
2 delays are the passenger that original plan passes through by path link 21 or arrive at website 1, and similarly, the delay of website 3 is in the original plan
Pass through or arrive at the passenger of website 1 by link 31.
Similar, if interruption website is transfer stop, it is detained several rail traffics that will concentrate mainly on through the website
On line, the website adjacent with the website.As shown in Figure 3.By taking website 4 as an example, because the website is two wires transfer stop, then share adjacent
Website 4, respectively：Website 5, website 6, website 7, website 8.
Can also calculate according to above-mentioned principle when interrupting website is multiple and when including transfer stop, the evacuation demand feelings of endpoint stations
Condition.In addition, other include interrupting website itself there may be the main of delay in addition to endpoint stations.Interrupt the evacuation of website itself
Demand is the passenger flow that original plan slave site is set out during interrupting, and can pass through aforementioned X(m,t)Matrix obtains.
Step 5：The emulation of evacuation demand temporal-spatial evolution
Simulation process is described as follows.In Rail traffic network, as the entering the station of passenger, train reach and are driven out to station
Dynamic pulsed variation occurs for the passengers quantity on point, each website and every train.The variation of its passenger flow has " week
Phase " property feature, using " passenger is entered the station, and --- train leaves --- train reaches " as an emulation cycle.Therefore simulation process is every
One period included three key steps：Passenger enters the station, process of getting off, process of getting on the bus, and executes by said sequence, i.e., every to update one
A moment t first completes the process that enters the station, then completes every train in the process of getting off a little that stops, and finally completes the stopped point of train
Process of getting on the bus.Enter next moment t+1 later.The process that enters the station is executed according to the normality OD of input, and process of getting off was got on the bus
Journey is described as follows respectively.
It is as follows to the variable-definition used in emulation
--- t moment enters the station from o, and wish goes to the transport need of d
xo--- the passenger flow that all t moments enter the station from o,
--- t moment goes to the passenger flow demand of d in station o, wish,
--- t moment goes to the passenger flow of d in the train on the e of side
ce--- the transporting power of side e,
The o point of e (o) --- side e
Vdisr--- the Website Hosting that burst is interrupted
T --- the period that burst is interrupted
It gets off process：
1. establishing function next_point for judging the next point in OD distribution rear path；
2. any side e in the set E on all sides of pair full figure seeks the direction website v of side enAnd a line e on the e of sidep；
3. traversing all website d in full figure Website Hosting V；
4. if d ≠ e (o), executes 5, otherwise executes 7；
5. judging the following conditions：
Condition 1:Matrix Um(e(o),d),1:nIn, the next point (next_point) of e (o) is not vn；
Condition 2:Matrix Um(e(o),d),1:nIn, in emergency event website area occurs for the next point (next_point) of e (o)
Between VdisrIn range；
Condition 3:Current time t ∈ T；
If condition 1 is true or (condition 2 is true and condition 3 is true), 6 are executed, otherwise
And execute 7；
6.
7. returning to 3 if all websites not yet traverse end, next website d is traversed；If all website traversals terminate, execute
8；
8. returning to 2 if all sides not yet traverse end, lower one side e is traversed；If all side traversals terminate, terminate.
It gets on the bus process：
1. being denoted as set disrupted_od according to interrupting in interval computation the whole network all affected OD pairs；
It is traversed 2. pair the whole network owns (o, d)
3. all side e of pair full figure are traversed, the direction website v of side e is soughtn；
4. all website d of pair full figure are traversed, if meeting condition：
Matrix Um(e(o),d),1:nIn, the next point (next_point) of e (o) is vn
Then board_list+=d；
5. traversing next website d, 4 are executed if all websites not yet traverse end；If all website traversals terminate, execute
6；
6. all website d of pair full figure are traversed；
W=min (total_demand, total_space)
If w > 0Otherwise 7 are directly executed；
7. returning to 6 if all websites not yet traverse end, next website d is traversed；If all website traversals terminate, execute
8；
8. returning to 2 if all sides not yet traverse end, lower one side e is traversed；If all side traversals terminate, terminate
Furthermore it is worth noting that, peak retention refers to interrupts, in evacuation event procedure entire, may in network-wide basis
The delay maximum value of appearance.The value appear at the time of evacuating pressure maximum in whole event and place and security risk most
Seriously, the node most possibly collapsed to other modes of transportation propagation congestion, at first.It is emulated in the temporal-spatial evolution to evacuation demand
In the process, while peak retention is provided.
Embodiment 1：
Example background：Using Shanghai rail transit live network information and operation data as input, it is assumed that rail traffic
Website 39,40,41,42 interrupts under emergency event, and emergency event time of origin is morning peak 8:00, the duration 30 divides
Clock.It is predicted using the evacuation demand of algorithm of the invention to the whole network under emergency event, obtains result as shown in figure 5, Fig. 5 is retouched
Generation, accumulation and dispersal curve that the whole network under the scene is mainly detained at website demand of evacuating are drawn.It is stagnant that peak has been marked in figure
It stays value and is detained the end time.
From figure 5 it can be seen that after emergency event occurs, it is detained more serious, that is, evacuates that demand intensity is highest to include
Endpoint stations (website 12,43,117,219) are interrupted website itself (website 39,41,42), and interrupt the transfer that section is more closed on
It stands (website 38).In addition, terminating after transport power restores in emergency event, the evacuation demand of part website can still rise, such as website 38,
The reason for this is that the evacuation demand of its uplink website accumulation will preferentially be met, so as to cause the delay of downlink website.
Embodiment 2：
Example background：Using Shanghai rail transit live network information and operation data as input, it is assumed that rail traffic
Website 39,40,41,42 interrupts under emergency event, and emergency event time of origin is non-peak period 12:00, the duration
30 minutes.Predicted using the evacuation demand of algorithm of the invention to the whole network under emergency event, obtain result as shown in fig. 6,
Fig. 6 depicts generation, accumulation and the dispersal curve that the whole network under the scene is mainly detained at website demand of evacuating.Height has been marked in figure
Peak retention and delay end time.
From fig. 6 it can be seen that after emergency event occurs, it is detained more serious, that is, evacuates that demand intensity is highest to include
Endpoint stations (website 12,43,117,88,204) are secondly interruption website itself (website 39,40,41,42), and more close on
Transfer stop (website 38).
In addition, embodiment 1 and embodiment 2, which is comprehensively compared, to be obtained, emergency event is betided to evacuate caused by peak period and be needed
It asks higher in off-peak period intensity than occurring, is embodied in whole delay and peak retention significantly increases, and event terminates
Evacuation demand evanishment afterwards is slower.Embodiment 1 and embodiment 2 illustrate the prediction burst thing that the present invention can be scientific and reasonable
Part lower railway traffic passenger flow evacuates demand and its temporal-spatial evolution.
This hair can be understood and applied the above description of the embodiments is intended to facilitate those skilled in the art
It is bright.Person skilled in the art obviously easily can make various modifications to these embodiments, and described herein
General Principle is applied in other embodiments without having to go through creative labor.Therefore, the present invention is not limited to implementations here
Example, those skilled in the art's announcement according to the present invention, the improvement made for the present invention and modification all should be of the invention
Within protection scope.
Claims (9)
1. a kind of emergency event lower railway traffic passenger flow evacuates needing forecasting method, which is characterized in that include the following steps：
1) according to rail traffic automatic fare collection system history ticket card data acquisition normality transport need sample set in short-term, and with normality
In short-term based on transport need sample set, multi-step prediction is carried out using the hybrid prediction model based on principal component analysis, is obtained normal
Subway Travel Demand Forecasting result in short-term under state；
2) Rail traffic network digraph is established according to Rail traffic network topological relation；
3) rail traffic travelling OD is allocated to obtain allocation matrix F；
4) according to the feature of emergency event and prior information, in conjunction with gained requirement forecasting in step 1) as a result, obtaining outgoing event shadow
The calculation formula of each website evacuation demand of the whole network during sound；
5) temporal-spatial evolution of demand is evacuated during obtaining emergency event influence using emulation, and obtains peak retention.
2. a kind of emergency event lower railway traffic passenger flow according to claim 1 evacuates needing forecasting method, feature exists
In the step 1) specifically includes following steps：
11) rail traffic automatic fare collection system history ticket card data are pre-processed, including data cleansing, run length code matching, pressed
Required time granularity carries out collection meter, obtains normality transport need sample set in short-term；
12) to normality, transport need sample set carries out principal component analysis in short-term, obtains principal component and corresponding load；
13) modeling and forecasting is carried out to load and its corresponding temporal characteristics, the corresponding principal component of prediction gained load is combined
As the Traffic Demand Forecasting result in short-term of the subway under normality.
3. a kind of emergency event lower railway traffic passenger flow according to claim 1 evacuates needing forecasting method, feature exists
In the predicted time window range of the hybrid prediction model is longer than the rail transportation ticket-card data back period.
4. a kind of emergency event lower railway traffic passenger flow according to claim 1 evacuates needing forecasting method, feature exists
In, in the step 3), the flow that allocation matrix F is carried by each side of Rail traffic network digraph at all moment,
Calculating formula is：
FT=XT×U
Wherein, the OD trip matrix that X is orbit traffic full network one day, U are the matrix of shortest path.
5. a kind of emergency event lower railway traffic passenger flow according to claim 1 evacuates needing forecasting method, feature exists
In in the step 4), the calculation formula of each website evacuation demand of the whole network is during event influences：
Wherein,For the evacuation demand of moment t website r, ft eThe flow on the e of side, and f are loaded for moment tt e∈ F, t0 are station
The interruption moment of point r.
6. a kind of emergency event lower railway traffic passenger flow according to claim 1 evacuates needing forecasting method, feature exists
In in the step 5), temporal-spatial evolution includes multiple emulation cycles, and successively sequence enters the station each emulation cycle including passenger
Process, process of getting off and process of getting on the bus.
7. a kind of emergency event lower railway traffic passenger flow according to claim 6 evacuates needing forecasting method, feature exists
In the passenger enters the station process using orbit traffic full network one day OD trip matrix as input.
8. a kind of emergency event lower railway traffic passenger flow according to claim 6 evacuates needing forecasting method, feature exists
In the process of getting off is specially：
511) obtain rail traffic travelling OD distribution after path, and obtain distribution after all paths in either site it is next
Website；
512) to any side e in the set E on all sides of Rail traffic network digraph full figure, the direction website v of side e is obtainednAnd
The upper a line e of side ep；
513) the Website Hosting V traversed in Rail traffic network digraph judges whether website d is side e to either site d
Source site, if it is not, step 514) is then carried out, if so, carrying out step 516)；
514) it realizes and gets off, judge the following conditions：
Condition 1：With OD in corresponding path in shortest path allocation matrix U, next website of the source site e (o) of side e is not
For vn, the OD is to the source site e (o) and website d for side e；
Condition 2：With OD in corresponding path in shortest path allocation matrix U, next website of the source site e (o) of e is prominent
Section V is interrupted under hair eventdisrIt is interior；
Condition 3：Current time t belongs to the period T that burst is interrupted；
If condition 1 be true or condition 2 and condition 3 at the same be it is true, carry out step 515), otherwise number of people in car is constant, then when t+1
The passenger flow of website d is gone in the train being engraved on the e of sideThe passenger flow of website d is gone in the train on the e of side with t moment
It is identical, i.e.,And carry out step 516)；
515) the passenger flow demand that t moment goes to d in website o wish is updatedI.e.Wherein,
The passenger flow demand of d is gone in website o wish for the t-1 moment；
516) return step 513 if all websites not yet traverse end), next website is traversed, if all websites traversal terminates,
Carry out step 517)；
517) return step 512 if all sides not yet traverse end), lower one side is traversed, if all sides traversal terminates, is tied
Beam.
9. a kind of emergency event lower railway traffic passenger flow according to claim 8 evacuates needing forecasting method, feature exists
In the process of getting on the bus is specially：
521) all affected OD pairs are obtained in the whole network according to interruption section, and is denoted as set disrupted_od；
522) ODs all to Rail traffic network the whole network update t moment in website o, wish goes to the passenger flow of d to need to traversing
It asks Wherein,It is the t-1 moment in website o, wish goes to the passenger flow demand of d,For t
Moment enters the station from o, and wish goes to the transport need of d；
523) side es all to Rail traffic network digraph full figure are traversed, and obtain the direction website v of side en；
524) all websites of Rail traffic network digraph full figure are traversed, if either site d meets condition：
With OD in corresponding path in shortest path allocation matrix U, if the next point of the o point e (o) of side e is vn, then will station
The list board_list that gets on the bus is added in point d；
525) it traverses next website and carries out step 524) if all websites, which not yet traverse, to be terminated, if all websites traversal knot
Beam then carries out step 526)；
526) all websites of Rail traffic network digraph full figure are traversed, to any website d：
Obtain demand of always ridingWith total vehicle carrying capacitySmaller value w in the two.If w
> 0 then updates the passenger flow that t moment goes to d in the train on the e of side Otherwise step is carried out
527), wherein ceFor the transporting power of side e；
527) return step 526 if all websites not yet traverse end), next website is traversed, if all websites traversal terminates,
Carry out step 528)；
528) return step 522 if all sides not yet traverse end), lower one side is traversed, if all sides traversal terminates, is tied
Beam.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810580795.6A CN108846514B (en) | 2018-06-07 | 2018-06-07 | Method for predicting demand of rail transit passenger flow evacuation in emergency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810580795.6A CN108846514B (en) | 2018-06-07 | 2018-06-07 | Method for predicting demand of rail transit passenger flow evacuation in emergency |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108846514A true CN108846514A (en) | 2018-11-20 |
| CN108846514B CN108846514B (en) | 2020-07-28 |
Family
ID=64211439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810580795.6A Active CN108846514B (en) | 2018-06-07 | 2018-06-07 | Method for predicting demand of rail transit passenger flow evacuation in emergency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108846514B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544927A (en) * | 2018-12-10 | 2019-03-29 | 中南大学 | A kind of multitiered network collaboration restricted driving method |
| CN110427402A (en) * | 2019-07-24 | 2019-11-08 | 上海工程技术大学 | A kind of delay of rail traffic failure, which propagates and involves range, estimates system |
| CN110782070A (en) * | 2019-09-25 | 2020-02-11 | 北京市交通信息中心 | Urban rail transit emergency passenger flow space-time distribution prediction method |
| CN110852650A (en) * | 2019-11-19 | 2020-02-28 | 交通运输部公路科学研究所 | Comprehensive passenger transport hub group network modeling method based on dynamic graph hybrid automaton |
| CN111881548A (en) * | 2020-06-16 | 2020-11-03 | 广东工业大学 | Computer simulation method for intelligent transportation |
| CN112686417A (en) * | 2019-10-18 | 2021-04-20 | 深圳先进技术研究院 | Subway large passenger flow prediction method and system and electronic equipment |
| CN113190935A (en) * | 2020-11-19 | 2021-07-30 | 西南交通大学 | Regional rail transit station bearing capacity simulation system and method based on layered Petri network |
| CN114912233A (en) * | 2022-04-19 | 2022-08-16 | 华北科技学院(中国煤矿安全技术培训中心) | Method and system for determining and cooperatively managing and controlling influence range of road network transportation capacity reduction |
| CN114937366A (en) * | 2022-07-22 | 2022-08-23 | 深圳市城市交通规划设计研究中心股份有限公司 | Traffic flow calculation method based on multi-scale traffic demand and supply conversion |
| CN115620525A (en) * | 2022-12-16 | 2023-01-17 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169606A (en) * | 2010-02-26 | 2011-08-31 | 同济大学 | Method for predicting influence of heavy passenger flow of urban rail transit network |
| CN103984993A (en) * | 2014-05-13 | 2014-08-13 | 东南大学 | Rail transit passenger flow OD distribution real-time speculation method |
| CN104239726A (en) * | 2014-09-22 | 2014-12-24 | 北京交通大学 | Passenger flow estimation system under condition of urban rail road network emergency |
| CN106503315A (en) * | 2016-10-10 | 2017-03-15 | 北京市交通信息中心 | The determination method of the passenger's trip characteristicses ratio that enters the station under track traffic accident |
| CN107194497A (en) * | 2017-04-27 | 2017-09-22 | 北京交通大学 | Urban track traffic passenger trip route planing method under a kind of accident |
| CN107274000A (en) * | 2017-04-27 | 2017-10-20 | 北京交通大学 | Urban track traffic section passenger flow forecasting under a kind of accident |
-
2018
- 2018-06-07 CN CN201810580795.6A patent/CN108846514B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102169606A (en) * | 2010-02-26 | 2011-08-31 | 同济大学 | Method for predicting influence of heavy passenger flow of urban rail transit network |
| CN103984993A (en) * | 2014-05-13 | 2014-08-13 | 东南大学 | Rail transit passenger flow OD distribution real-time speculation method |
| CN104239726A (en) * | 2014-09-22 | 2014-12-24 | 北京交通大学 | Passenger flow estimation system under condition of urban rail road network emergency |
| CN106503315A (en) * | 2016-10-10 | 2017-03-15 | 北京市交通信息中心 | The determination method of the passenger's trip characteristicses ratio that enters the station under track traffic accident |
| CN107194497A (en) * | 2017-04-27 | 2017-09-22 | 北京交通大学 | Urban track traffic passenger trip route planing method under a kind of accident |
| CN107274000A (en) * | 2017-04-27 | 2017-10-20 | 北京交通大学 | Urban track traffic section passenger flow forecasting under a kind of accident |
Non-Patent Citations (4)
| Title |
|---|
| 周玮腾: "拥塞条件下的城市轨道交通网络流量分配演化建模及疏导策略研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 安实等: "基于最优化与仿真的区域疏散交通管理研究", 《交通运输系统工程与信息》 * |
| 李伟等: "突发事件下地铁网络乘客出行行为仿真模型", 《华东交通大学学报》 * |
| 邢淋丽: "应急交通组织下的公交车辆疏散路径规划模型", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109544927B (en) * | 2018-12-10 | 2019-11-29 | 中南大学 | A kind of multitiered network collaboration restricted driving method |
| CN109544927A (en) * | 2018-12-10 | 2019-03-29 | 中南大学 | A kind of multitiered network collaboration restricted driving method |
| CN110427402A (en) * | 2019-07-24 | 2019-11-08 | 上海工程技术大学 | A kind of delay of rail traffic failure, which propagates and involves range, estimates system |
| CN110427402B (en) * | 2019-07-24 | 2021-10-08 | 上海工程技术大学 | Rail transit fault delay propagation and spread range estimation system |
| CN110782070A (en) * | 2019-09-25 | 2020-02-11 | 北京市交通信息中心 | Urban rail transit emergency passenger flow space-time distribution prediction method |
| CN110782070B (en) * | 2019-09-25 | 2022-04-22 | 北京市交通信息中心 | Urban rail transit emergency passenger flow space-time distribution prediction method |
| CN112686417B (en) * | 2019-10-18 | 2024-03-19 | 深圳先进技术研究院 | Subway large passenger flow prediction method, system and electronic equipment |
| CN112686417A (en) * | 2019-10-18 | 2021-04-20 | 深圳先进技术研究院 | Subway large passenger flow prediction method and system and electronic equipment |
| WO2021072959A1 (en) * | 2019-10-18 | 2021-04-22 | 深圳先进技术研究院 | Method and system for large passenger flow forecasting of metros, and electronic device |
| CN110852650A (en) * | 2019-11-19 | 2020-02-28 | 交通运输部公路科学研究所 | Comprehensive passenger transport hub group network modeling method based on dynamic graph hybrid automaton |
| CN110852650B (en) * | 2019-11-19 | 2021-11-02 | 交通运输部公路科学研究所 | Comprehensive passenger transport hub group network modeling method based on dynamic graph hybrid automaton |
| CN111881548A (en) * | 2020-06-16 | 2020-11-03 | 广东工业大学 | Computer simulation method for intelligent transportation |
| CN113190935A (en) * | 2020-11-19 | 2021-07-30 | 西南交通大学 | Regional rail transit station bearing capacity simulation system and method based on layered Petri network |
| CN114912233A (en) * | 2022-04-19 | 2022-08-16 | 华北科技学院(中国煤矿安全技术培训中心) | Method and system for determining and cooperatively managing and controlling influence range of road network transportation capacity reduction |
| CN114937366A (en) * | 2022-07-22 | 2022-08-23 | 深圳市城市交通规划设计研究中心股份有限公司 | Traffic flow calculation method based on multi-scale traffic demand and supply conversion |
| CN114937366B (en) * | 2022-07-22 | 2022-11-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Traffic flow calculation method based on multi-scale traffic demand and supply conversion |
| CN115620525A (en) * | 2022-12-16 | 2023-01-17 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
| CN115620525B (en) * | 2022-12-16 | 2023-03-10 | 中国民用航空总局第二研究所 | Short-time traffic passenger demand prediction method based on time-varying dynamic Bayesian network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108846514B (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846514A (en) | A kind of emergency event lower railway traffic passenger flow evacuation needing forecasting method | |
| CN111371904B (en) | Cloud-side-end-coordinated highway cloud control system and control method | |
| CN108550098B (en) | Urban rail transit network passenger flow limiting method | |
| CN107194497B (en) | Method for planning travel path of urban rail transit passenger in emergency | |
| CN110245806A (en) | A kind of urban track traffic for passenger flow management-control method and system | |
| CN109033718B (en) | Dynamic emergency evacuation method for urban rail transit line failure | |
| CN110866649A (en) | Method and system for predicting short-term subway passenger flow and electronic equipment | |
| CN102169606B (en) | Method for predicting influence of heavy passenger flow of urban rail transit network | |
| CN110149593B (en) | Road network passenger flow state identification method based on mobile station signaling | |
| WO2016045195A1 (en) | Passenger flow estimation method for urban rail network | |
| CN104239726A (en) | Passenger flow estimation system under condition of urban rail road network emergency | |
| CN104157132A (en) | Self-adaptive dynamic optimization method for bus dispatching timetable | |
| CN111882867A (en) | City wisdom traffic early warning system based on thing networking | |
| CN112990648A (en) | Rail transit network operation stability assessment method | |
| Zhang et al. | Coupling analysis of passenger and train flows for a large-scale urban rail transit system | |
| Chen et al. | Resilience assessment of an urban rail transit network under short-term operational disturbances | |
| CN104734770A (en) | Distributed constellation network routing method based on context | |
| Bi et al. | Energy-aware navigation in large-scale evacuation using g-networks | |
| CN111367900B (en) | Method for calculating normal current limiting intensity of urban rail transit network based on AFC data | |
| Minciardi et al. | A decentralized optimal control scheme for route guidance in urban road networks | |
| Di Febbraro et al. | INTRANET: A new simulation tool for intermodal transportation systems | |
| CN109842522A (en) | Communicate the configuration method and system of route resource | |
| Mejdoubi et al. | A distributed predictive road traffic management system in urban VANETs | |
| CN107195192A (en) | A kind of recognition methods of vehicle-mounted orientation running status | |
| Yang et al. | Research on Coordinated Passenger Inflow Control for the Urban Rail Transit Network Based on the Station‐to‐Line Spatial‐Temporal Relationship |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |