CN116128160B - Method, system, equipment and medium for predicting peak passenger flow of railway station - Google Patents
Method, system, equipment and medium for predicting peak passenger flow of railway station Download PDFInfo
- Publication number
- CN116128160B CN116128160B CN202310353020.6A CN202310353020A CN116128160B CN 116128160 B CN116128160 B CN 116128160B CN 202310353020 A CN202310353020 A CN 202310353020A CN 116128160 B CN116128160 B CN 116128160B
- Authority
- CN
- China
- Prior art keywords
- peak
- station
- passenger flow
- proportion
- day
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 230000000737 periodic effect Effects 0.000 claims abstract description 20
- 238000010276 construction Methods 0.000 claims abstract description 19
- 238000003066 decision tree Methods 0.000 claims abstract description 16
- 230000003203 everyday effect Effects 0.000 claims abstract description 9
- 230000015654 memory Effects 0.000 claims description 31
- 238000005457 optimization Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000010923 batch production Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 230000002776 aggregation Effects 0.000 description 7
- 238000004220 aggregation Methods 0.000 description 7
- 230000002354 daily effect Effects 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013450 outlier detection Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000012892 rational function Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000029305 taxis Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
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)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a railway station peak passenger flow prediction method, which comprises the following steps: the ticket selling amount proportion prediction model construction step: based on the periodic characteristics that the ticket selling amount before driving accounts for the proportion of all ticket selling amounts from a designated station in the pre-selling period of the designated station, constructing a ticket selling amount proportion prediction model based on LSTM so as to predict the proportion of the ticket selling amount before driving to all ticket selling amounts; the construction step of the peak period prediction model comprises the following steps: based on the time-period characteristics of peak passenger flow in a station starting from a specified station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the peak time periods; and (3) predicting peak passenger flow: and according to the proportion of the predicted ticket selling quantity before driving to the total ticket selling quantity and the peak time, the prediction of the passenger flow at the peak time in the railway station on the driving day is realized. The application also discloses a railway station peak passenger flow prediction system.
Description
Technical Field
The application relates to a passenger flow prediction method, in particular to a railway station peak passenger flow prediction method and a system thereof.
Background
Currently, large railway passenger stations are usually comprehensive transportation hubs in cities, and passenger flows generally comprise passenger flows which are combined by various transportation modes such as departure and arrival of railway stations, urban buses and subways of the railway stations and private cars, railway stations and taxis, and the like, and the interior and the periphery of the railway stations are often unsmooth in evacuation due to the intersection of various passenger flows; the lack of real-time and comprehensive research and judgment of various information such as the traffic operation state of the urban road related to the station, the arrival passenger flow rule of the railway departure, and the like, may cause unreasonable work arrangement such as waiting, ticket checking, and the like, and cause backlog of the passenger flow in the station, thereby affecting smooth travel of railway passengers. Therefore, the railway passenger station management department needs to make a station management plan in advance according to the passenger flow in the station, especially the peak passenger flow in the station, so as to avoid the situation that passengers travel unsmoothly due to backlog of the passenger flow in the station.
Regarding the prediction of railway peak passenger flows, the prior art scheme:
patent publication No.: CN113919577a, patent name: a short-time passenger flow prediction method and device for rail transit are disclosed, and a short-time passenger flow prediction model is constructed through a federal learning technology, so that safe collaborative training of multi-client data is realized.
Patent publication No.: CN113793015a, patent name: the system comprises a storage system, a monitoring system, an information transmission system, a central processing system, a passenger flow prediction module, a sharing rate module and a parking berth module, wherein the travel cost is obtained based on flow distribution of an information processing platform and is used for iterative calculation of sharing rate, and the parking berth is calculated by using the converged sharing rate.
Patent publication No.: CN113177586a, patent name: the method comprises the steps of calculating the difference degree of the existing station and a newly built station, selecting a plurality of similar stations to calculate the highest aggregation number of the existing station and the newly built station, obtaining a reasonable interval of the highest aggregation number of the newly built station, and finishing the estimation of the highest aggregation number of the high-speed railway station.
The paper name is: the calculation model study of the highest aggregation number of large-scale high-speed railway stations, published journal information is as follows: railway science and engineering report 2021,18 (12), based on passenger waiting time and ticket checking channel service rate, 4 kinds of function fitting effects such as logarithmic normal distribution, weibull distribution, composite negative index distribution, rational function distribution and the like are compared, and a calculation model of the highest aggregation number of the railway passenger station is constructed. The model in the paper builds a prediction model based on railway station full ticketing transaction data, and is only applicable to railway stations with larger passenger flow scale.
The passenger flow in the railway passenger station presents a specific periodicity and time period law under the mutual influence of various factors such as passenger travel habit, related urban traffic capacity and state, railway train operation plan, railway passenger station management method and the like, but the single prediction method in the prior art is difficult to capture the periodicity and time period characteristics of the passenger flow time sequence in the station at the same time, and the prediction model is constructed based on the passenger flow data in the off-line station far from the date to be detected, but the passenger flow data change in the railway passenger ticket selling process is ignored, so that the prediction accuracy is possibly unstable.
Therefore, it is needed to propose a railway station peak passenger flow prediction method capable of capturing the periodic and time-period characteristics of the passenger flow time sequence in the station at the same time, namely, based on the historical offline passenger flow data and the passenger ticket sales process data in the railway passenger station, the prediction method of fusion time sequence prediction and outlier detection is adopted, so that the peak gathering time period and the number of people in the railway passenger station can be predicted offline, the peak passenger flow in the station can be effectively identified, and the problems that the periodic and time-period characteristics of the passenger flow time sequence in the station cannot be captured at the same time and the prediction accuracy is unstable in the prediction method in the prior art are solved.
Disclosure of Invention
The embodiment of the application provides a railway station peak passenger flow prediction method, which solves the problems that the prediction method in the prior art is single and the periodic and time-period characteristics of passenger flow time sequences in stations cannot be captured simultaneously.
In a first aspect, an embodiment of the present application provides a method for predicting peak passenger flow of a railway station, where the method includes:
the ticket selling amount proportion prediction model construction step: based on the periodic characteristics that the ticket selling amount before driving accounts for the proportion of all ticket selling amounts from a designated station in the pre-selling period of the designated station, constructing a ticket selling amount proportion prediction model based on LSTM so as to predict the proportion of the ticket selling amount before driving to all ticket selling amounts;
the construction step of the peak period prediction model comprises the following steps: based on the time-period characteristics of peak passenger flow in a station starting from a specified station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
and (3) predicting peak passenger flow: and according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of people starting all the day, the prediction of the passenger flow in the peak time of the railway station on the driving day is realized.
Preferably, the ticket selling amount before driving is ticket selling amount from the appointed station in the day before driving in the preset selling period of the appointed station, and all ticket selling amounts are all ticket selling amounts from the appointed railway station.
Preferably, the ticket selling amount proportion prediction model construction step includes:
constructing a data set of the number of people before driving: constructing a pre-driving number of people ratio data set which is in proportion to the total number of people to be started in the day before driving in the pre-selling period, wherein the pre-driving number of people ratio data set comprises: the method comprises the steps of starting date, ticket selling amount in the day before starting in the pre-selling period, total number of starts, proportion of total number of starts in the day before starting in the pre-selling period, and normalizing the front number of starts in the data set by adopting a linear normalization method;
constructing an LSTM model: aiming at a front people number duty ratio data set, constructing a characteristic data set based on the number of days of a specified sliding window, constructing an LSTM-based prediction model based on the characteristic data set, and determining LSTM prediction model parameters needing continuous optimization;
constructing a parameter optimization model: and (3) taking the average absolute percentage error as an optimization target, and performing iterative computation on the LSTM prediction model parameters by adopting a Bayesian optimization method to obtain an optimized parameter set of the LSTM prediction model parameters.
Preferably, the peak period prediction model construction step includes:
constructing an in-station people data set: dividing the departure day according to a set time interval, calculating the number of people in the departure day station, sorting the number of people in the station from big to small, marking whether a plurality of appointed time periods are peak number periods of the number of people in the station, and forming an in-station number data set of each time period of the departure day, wherein the in-station number data set comprises the departure date, the departure time period, the number of people in the departure of the plurality of appointed time periods, the number of people in the departure of the whole day, the proportion of the appointed time period to the number of people in the departure of the whole day and the appointed sorting number of the plurality of people;
the step of building an intra-station peak time identification model: aiming at the number data sets in the station, respectively extracting the data sets of the number ordering numbers of the appointed people in a plurality of time ranges of the departure day in the peak time, and constructing an identification model of the peak time in the station based on a decision tree;
the step of constructing a predicting model of the proportion of the peak number to the number of departure all the day: aiming at the number data sets in the station, respectively extracting the data sets of the appointed number ordering numbers in a plurality of time ranges of the departure day in the peak time period, and constructing a proportion prediction model of the peak number of people occupying the departure number of the whole day.
Preferably, the peak passenger flow prediction step includes:
Based on the prediction results of the in-station peak time identification model and the ratio prediction model of the peak number to the number of people in departure on the whole day, and the ratio of the ticket selling amount to the total ticket selling amount before driving, the number of people in the peak station on a plurality of designated time periods on the departure day is obtained through prediction calculation.
Preferably, the step of constructing the LSTM model includes:
constructing a characteristic data set: taking the proportion of the ticket selling quantity before driving of the appointed sliding window days before the date to be measured to the total ticket selling quantity as a characteristic, constructing a characteristic data set based on the appointed sliding window days, wherein the appointed sliding window days are determined by the periodic characteristic of the data set of the occupancy ratio of the number of people before driving;
constructing an LSTM prediction model: based on the characteristic data set, constructing a prediction model based on the proportion of ticket sales amount of the day before departure to total departure number based on LSTM, and predicting to obtain the proportion of the date to be predicted and the actual departure number of the day before driving in the day before driving;
confirming model parameter optimization: according to the LSTM prediction model, determining LSTM prediction model parameters needing to be optimized, wherein the LSTM prediction model parameters comprise: hidden layer size, feature dimension, number of batch processes, and number of training iterations.
Preferably, the peak passenger flow prediction step includes:
the peak passenger flow is predicted as follows: s is(s) y /(p d* q m ) Wherein s is y To sell ticket before driving, p d Q is the proportion of the ticket selling quantity accounting for the total ticket selling quantity before driving 1 to d days before the date to be measured m The peak number of m time periods on the same day of departure is the proportion of the number of departure people on the whole day.
In a second aspect, an embodiment of the present application provides a railway station peak passenger flow prediction system, which adopts the above railway station peak passenger flow prediction method, and includes:
the ticket selling amount proportion prediction model building module: based on the periodic characteristics that the ticket selling amount before driving accounts for the proportion of all ticket selling amounts from a designated station in the pre-selling period of the designated station, constructing a ticket selling amount proportion prediction model based on LSTM so as to predict the proportion of the ticket selling amount before driving to all ticket selling amounts;
the peak period prediction model building module: based on the time-period characteristics of peak passenger flow in a station starting from a specified station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
Peak passenger flow prediction module: and according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of departure people in the whole day, the prediction of the peak time passenger flow in the railway station in the driving day is realized.
In a third aspect, embodiments of the present application provide a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method of railway station peak passenger flow prediction as described above when executing the computer program.
In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements a railway station peak passenger flow prediction method as described above.
Compared with the related prior art, the method has the following outstanding beneficial effects:
1) The method for predicting the peak aggregate passenger flow of the railway passenger transportation hub constructed by the method can comprehensively consider the periodic and time-period characteristics of the passenger flow in the railway passenger transportation station, decompose the periodic and time-period characteristics of the passenger flow in the station by constructing characteristic data, comprehensively apply various theories such as deep learning, machine learning and optimization algorithm, and the like, and capture the periodic and time-period characteristics of the passenger flow in the station in a layering manner, so as to reasonably and accurately predict the number of the peak aggregate passenger flow in the railway passenger transportation station;
2) The method for predicting the peak accumulated passenger flow of the railway passenger transportation hub constructed by the method is an offline prediction method, and the peak accumulated passenger flow in the second day station is predicted by combining the historical passenger flow data in the station and the current actual ticket selling data in the pre-selling period every day, so that the method does not depend on the real-time passenger flow data of the station, is simple in calculation method and high in prediction precision, and provides decision basis for making management plans in advance for the railway passenger transportation station.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a schematic diagram of a method for predicting peak passenger flow of a railway station according to the present invention;
FIG. 2 is a flow chart of peak aggregate head count prediction in a railroad passenger terminal based on LSTM and decision tree in accordance with an embodiment of the present invention;
FIG. 3 shows an LSTM-based predictive model M according to an embodiment of the invention 1 A structural schematic;
FIG. 4 is a schematic diagram of a railway station peak passenger flow prediction system of the present invention;
fig. 5 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present application.
In the above figures:
10 ticket sales ratio prediction model construction module 20 peak period prediction model construction module
A peak passenger flow prediction module 30.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described and illustrated below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on the embodiments provided herein, are intended to be within the scope of the present application.
It is apparent that the drawings in the following description are only some examples or embodiments of the present application, and it is possible for those of ordinary skill in the art to apply the present application to other similar situations according to these drawings without inventive effort. Moreover, it should be appreciated that while such a development effort might be complex and lengthy, it would nevertheless be a routine undertaking of design, fabrication, or manufacture for those of ordinary skill having the benefit of this disclosure, and thus should not be construed as having the benefit of this disclosure.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the other features, objects, and advantages of the application.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly and implicitly understood by those of ordinary skill in the art that the embodiments described herein can be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar terms herein do not denote a limitation of quantity, but rather denote the singular or plural. The terms "comprising," "including," "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to only those steps or elements but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "connected," "coupled," and the like in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as used herein refers to two or more. "and/or" describes an association relationship of an association object, meaning that there may be three relationships, e.g., "a and/or B" may mean: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. The terms "first," "second," "third," and the like, as used herein, are merely distinguishing between similar objects and not representing a particular ordering of objects.
The invention aims to provide a method for predicting a peak aggregation time period and the number of people in a railway passenger station in an offline manner by integrating time sequence prediction and outlier detection, firstly, the method aims at the periodic characteristic that the proportion of the ticket selling amount (abbreviated as 'ticket selling amount before driving') from a specified station to all tickets selling (abbreviated as 'all ticket selling amount') from the railway station in a day before driving in a pre-selling period is calculated, and the proportion of the ticket selling amount before driving to all ticket selling amount can be effectively and accurately predicted by constructing a prediction model based on LSTM; secondly, aiming at the characteristic that peak passenger flow in a station starting from a designated station every day has relatively stable timeliness, after the sample data are divided and marked at certain time intervals, a classification model based on a decision tree is constructed to identify the peak time, and the peak passenger flow in the station on the current day of driving is estimated according to the proportion of the ticket selling quantity before driving to the total ticket selling quantity.
Meanwhile, the invention provides a method for predicting peak gathering passenger flow of a railway passenger station based on a specific periodicity and a time period law of passenger flow in the railway passenger station, which mainly comprises 2 parts: firstly, constructing a passenger flow data set of a railway passenger station departure, constructing a prediction model based on LSTM, and predicting the proportion of ticket selling quantity from a specified railway station to the total ticket selling quantity from the railway station in a day before driving in a pre-selling period; secondly, constructing a daily peak passenger flow data set in the railway passenger station, constructing a prediction model based on a decision tree, identifying the daily peak passenger flow time period of the specified railway station, and predicting the peak passenger flow of the specified railway station according to the result in the first part. The method considers the periodic and time-period characteristics of the daily departure passenger flow of the appointed railway station, adopts a distributed prediction and identification mode to complete the prediction of the peak passenger flow in the station in one day before driving, and provides decision basis for the advanced appointed management plan of the railway passenger station.
As shown in fig. 1, an embodiment of the present application provides a method for predicting peak passenger flow of a railway station, where the method includes:
the ticket selling amount proportion prediction model construction step S10: based on the periodic characteristics that the ticket selling amount before driving accounts for the proportion of all ticket selling amounts from a designated station in the pre-selling period of the designated station, constructing a ticket selling amount proportion prediction model based on LSTM so as to predict the proportion of the ticket selling amount before driving to all ticket selling amounts;
peak period prediction model construction step S20: based on the time-period characteristics of peak passenger flow in a station starting from a specified station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
peak passenger flow prediction step S30: and according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of people starting all the day, the prediction of the passenger flow in the peak time of the railway station on the driving day is realized.
Preferably, the ticket selling amount before driving is ticket selling amount from the appointed station in the day before driving in the preset selling period of the appointed station, and all ticket selling amounts are all ticket selling amounts from the appointed railway station.
Preferably, the ticket selling amount proportion prediction model constructing step S10 includes:
constructing a data set of the number of people before driving: constructing a pre-driving number of people ratio data set which is in proportion to the total number of people to be started in the day before driving in the pre-selling period, wherein the pre-driving number of people ratio data set comprises: the method comprises the steps of starting date, ticket selling amount in the day before starting in the pre-selling period, total number of starts, proportion of total number of starts in the day before starting in the pre-selling period, and normalizing the front number of starts in the data set by adopting a linear normalization method;
constructing an LSTM model: aiming at a front people number duty ratio data set, constructing a characteristic data set based on the number of days of a specified sliding window, constructing an LSTM-based prediction model based on the characteristic data set, and determining LSTM prediction model parameters needing continuous optimization;
constructing a parameter optimization model: and (3) taking the average absolute percentage error as an optimization target, and performing iterative computation on the LSTM prediction model parameters by adopting a Bayesian optimization method to obtain an optimized parameter set of the LSTM prediction model parameters.
Preferably, the peak period prediction model construction step S20 includes:
constructing an in-station people data set: dividing the departure day according to a set time interval, calculating the number of people in the departure day station, sorting the number of people in the station from big to small, marking whether a plurality of appointed time periods are peak number periods of the number of people in the station, and forming an in-station number data set of each time period of the departure day, wherein the in-station number data set comprises the departure date, the departure time period, the number of people in the departure of the plurality of appointed time periods, the number of people in the departure of the whole day, the proportion of the appointed time period to the number of people in the departure of the whole day and the appointed sorting number of the plurality of people;
The step of building an intra-station peak time identification model: aiming at the number data sets in the station, respectively extracting the data sets of the number ordering numbers of the appointed people in a plurality of time ranges of the departure day in the peak time, and constructing an identification model of the peak time in the station based on a decision tree;
the step of constructing a predicting model of the proportion of the peak number to the number of departure all the day: aiming at the number data sets in the station, respectively extracting the data sets of the appointed number ordering numbers in a plurality of time ranges of the departure day in the peak time period, and constructing a proportion prediction model of the peak number of people occupying the departure number of the whole day.
Preferably, the peak passenger flow prediction step S30 includes:
based on the prediction results of the in-station peak time identification model and the ratio prediction model of the peak number to the number of people in departure on the whole day, and the ratio of the ticket selling amount to the total ticket selling amount before driving, the number of people in the peak station on a plurality of designated time periods on the departure day is obtained through prediction calculation.
Preferably, the step of constructing the LSTM model includes:
constructing a characteristic data set: taking the proportion of the ticket selling quantity before driving of the appointed sliding window days before the date to be measured to the total ticket selling quantity as a characteristic, constructing a characteristic data set based on the appointed sliding window days, wherein the appointed sliding window days are determined by the periodic characteristic of the data set of the occupancy ratio of the number of people before driving;
Constructing an LSTM prediction model: based on the characteristic data set, constructing a prediction model based on the proportion of ticket sales amount of the day before departure to total departure number based on LSTM, and predicting to obtain the proportion of the date to be predicted and the actual departure number of the day before driving in the day before driving;
confirming model parameter optimization: according to the basic structure of the LSTM prediction model (refer to LSTM detailed description, source: blog. Csdn. Net/m0_ 47891203/arc/details/122166634), determining LSTM prediction model parameters to be optimized, wherein the LSTM prediction model parameters comprise: hidden layer size, feature dimension, number of batch processes, and number of training iterations.
Preferably, the peak passenger flow prediction step S30 includes:
the peak passenger flow is predicted as follows: s is(s) y /(p d* q m ) Wherein s is y To sell ticket before driving, p d Q is the proportion of the ticket selling quantity accounting for the total ticket selling quantity before driving 1 to d days before the date to be measured m The peak number of m time periods on the same day of departure is the proportion of the number of departure people on the whole day.
Specific embodiments of the present invention are described in detail below with reference to the accompanying drawings:
the flow of the embodiment of the invention is shown in fig. 2:
the method for predicting the peak aggregation number in the railway passenger terminal station based on the LSTM and the decision tree mainly comprises two parts, wherein the first part is to predict the proportion of the ticket selling amount of the day before driving to the total number of the departure persons in the pre-selling period, the second part is to predict the peak period of the driving day, and the peak passenger flow in the driving station is predicted according to the proportion of the ticket selling amount before driving to the total ticket selling amount in a conversion mode.
(1) Step 1: construction dataset D 1 。
1) Constructing the total number of departure people in the pre-selling period until the day before drivingProportional data set D 1 The method comprises the steps of departure date, ticket selling amount after 1 day before starting in a pre-selling period, total departure number, proportion p (short for 'proportion p') of the total departure number when the ticket selling amount is sold in the pre-selling period after starting and before starting, and the like.
2) Data set D using a linear normalization method 1 Normalization processing is carried out, and the original numerical value is scaled to be 0,1]And respectively carrying out normalization processing on the training set and the test set in the interval, combining the maximum value and the minimum value in the training set with the input variable set of the date to be predicted, and normalizing the input variable set of the date to be predicted.
If the training set is [ x ] 1 ,x 2 ,...x i ,...,x t ]Maximum value and minimum value in training set are x respectively min And x max The method comprises the steps of carrying out a first treatment on the surface of the The input dataset for the date to be predicted is [ z 1 ,z 2 ,...z j ,...,z l ]The maximum value and the minimum value are respectively z min And z max . For x i And z j Normalized to obtainAnd->Then->
(2) Step 2: a long and short term memory network (LSTM) model is constructed.
1) The construction is based on a feature dataset that specifies a sliding window as d days. For dataset D 1 Characterized by the proportion p of 1 to d days before the date to be measured, the ticket selling quantity s of 1 day before departure based on LSTM is constructed y Predictive model M of total number of departure p 1 The ratio of the predicted date to be predicted is p 1 The actual departure number of the day of driving is s, which is predicted to be the day of driving before driving y /p 1 The method comprises the steps of carrying out a first treatment on the surface of the The D value is represented by data set D 1 Periodic feature decisions of (1), e.g. the day of the week to be predicted can be calculated separatelyThe pearson correlation coefficient or other types of correlation coefficients between the day proportion p and the proportion p of the date to be predicted, the d value corresponding to the maximum correlation coefficient value is selected, and the d value can be specified according to service experience.
2) As shown in FIG. 3, an LSTM based prediction framework is constructed. For example, characterized by a proportion p of the first 1 to 21 days of the date to be predicted, a three-layer structure LSTM prediction model M consisting of an LSTM layer, a dropout layer and a dense layer (fully connected layer) is constructed 1 The dark marked neurons represent that the neurons are deleted in the training process, and the deleted neurons are no longer involved in data transmission;indicating the time sequence. Remaining neurons->、/>、……、/>Transmitting to dense layer, realizing latitude conversion and arrangement calculation of multiple output results, and finally obtaining predicted result p d 。
3) Model parameters that need to be optimized are determined. According to the deep learning prediction framework of fig. 3, a predicted value is obtained after a plurality of variables are input, and parameters to be determined in an LSTM layer comprise a hidden layer size, a characteristic dimension, the number of batch processing and the number of training iterations, wherein the characteristic dimension is the sliding window d, and other parameters are obtained through the calculation of the Bayesian optimization model in the step 3; the parameters to be determined in the dropout layer are deleted neuron proportions, and the deleted neuron proportions are also calculated by the Bayesian optimization model in the step 3.
(3) Step 3: and constructing a Bayesian parameter optimization model.
According to the traditional Bayesian optimization method (see Bayesian optimization of machine learning optimization algorithm, sources: zhuanlan. Zhihu. Comp/146329121) with MAPE (Mean Absolute Percentage Error ) as the optimization target for LSTM model M in step 2 1 Iterative calculation is carried out for T times by adopting a Bayesian optimization method, and candidate parameter sets of the model are respectively obtained，/>. The T value is generally determined according to the data scale and experience, if the T value can be set to be 100, MAPE value change is observed in the iterative process, and when the MAPE value tends to be stable, the calculation process can be finished in advance; according to the MAPE sorting from small to large, k represents the sorting value of MAPE, s k 、r k 、b k 、t k The hidden layer size, the deleted neuron proportion, the batch number and the training iteration number parameters of the model M1 when the MAPE sequence is k are respectively represented. Taking n as the sample number, y i As an actual value, p i For the predicted value, MAPE is calculated as follows.
1) Inputting a group of super parameters, determining the search range of the super parameters according to the characteristic analysis result, the training set length, the service experience and the like, taking d=21 as an example, wherein the search range of the hidden layer size is [1,100], the search range of the deleted neuron proportion is [1,50], the search range of the batch data number is [2,50], and the search range of the training times is [50,200];
2) Constructing a surrogate function comprising an LSTM predictive model M 1 Is a complete training process;
3) Constructing an objective function by taking MAPE as an evaluation index, and recording and outputting the MAPE in the optimization process;
4) Iterative T times of sequential computation 1) to 3), selecting a super-parameter set corresponding to the first k smaller MAPE values;
5) And respectively substituting k hyper-parameter sets into the verification set, training the LSTM model again, and selecting the hyper-parameter set corresponding to the minimum MAPE value to obtain a corresponding prediction result.
(4) Step 4: constructing daily on-site peak passenger flow data set D 2 。
The departure day is set according to a certain time intervalDividing, calculating the number of people in the station on the departure day, sorting the number of people in the station from big to small, marking whether the time period is the peak time period of the number of people in the station on the whole day, the morning, the afternoon and the evening of the departure day respectively, if the time period is the peak time period of the number of people in the whole day when the number of people in the whole day is sorted and numbered 1, the time period is the peak time period of the number of people in the station on the morning when the number of people in the morning is sorted and numbered 1, and forming a data set D of the number of people in the station in each time period of the driving day 2 The field content comprises departure date, departure time, departure number of the whole day, proportion s of the time to the departure number of the whole day c A number of people ordering throughout the day, a number of people ordering in the morning, a number of people ordering in the afternoon, and a number of people ordering in the evening.
Wherein, according to a certain time intervalDividing the whole day time into a time period sequence t 1 ,t 2 ,...t i ,...,t L ]At t i The number of people who get in the period is +.>The number of ticket checking persons is->The number of people in the station->Subtracting the number of people in the current time period from the sum of the number of people in the current time period and the number of people waiting in the current time period but not starting ticket checking, calculating the number of people in the station based on a data structure in the current system, and defining the number of people in the station. Namely:
(5) Step 5: and constructing an intra-station peak time and passenger flow prediction model.
For dataset D 2 The method comprises the steps of respectively extracting data sets with the number of people in the peak period of all days, the morning, the afternoon and the evening of the departure day being numbered as 1, constructing an in-station peak period identification model based on a decision tree and a proportion prediction model of the peak number of people occupying all days of the departure day, wherein in the specific embodiment of the method, similar prediction effects can be achieved by adopting in-station peak period and passenger flow prediction models based on classical random forests, XGBoost and lightGBM, and the method preferably adopts default parameters for prediction. The invention is not limited in this regard and other predictive models may be employed.
(6) Step 6: and predicting and converting peak passenger flow in the station.
Inputting the ratio of peak number and peak number of the front 1 day of starting to the number of the departure people in the whole day, the morning, the afternoon and the eveningPredicting the number of people in peak stations of the departure day, the morning, the afternoon and the evening to be respectively、/>、/>、/>。
In a second aspect, an embodiment of the present application provides a railway station peak passenger flow prediction system, as shown in fig. 4, using the above railway station peak passenger flow prediction method, where the railway station peak passenger flow prediction system includes:
ticket sales ratio prediction model construction module 10: based on the periodic characteristics that the ticket selling amount before driving accounts for the proportion of all ticket selling amounts from a designated station in the pre-selling period of the designated station, constructing a ticket selling amount proportion prediction model based on LSTM so as to predict the proportion of the ticket selling amount before driving to all ticket selling amounts;
rush hour prediction model construction module 20: based on the time-period characteristics of peak passenger flow in a station starting from a specified station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
Peak passenger flow prediction module 30: and according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of people starting all the day, the prediction of the passenger flow in the peak time of the railway station on the driving day is realized.
In a third aspect, embodiments of the present application provide a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the railway station peak passenger flow prediction method as described above when executing the computer program.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium having stored thereon a computer program that, when executed by a processor, implements a railway station peak passenger flow prediction method as described above.
In addition, the railway station peak passenger flow prediction method of the embodiment of the present application described in connection with fig. 1 may be implemented by a computer device. Fig. 5 is a schematic diagram of a hardware structure of a computer device according to an embodiment of the present application.
The computer device may include a processor 81 and a memory 82 storing computer program instructions.
In particular, the processor 81 may include a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), or may be configured to implement one or more integrated circuits of embodiments of the present application.
Memory 82 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 82 may comprise a Hard Disk Drive (HDD), floppy Disk Drive, solid state Drive (Solid State Drive, SSD), flash memory, optical Disk, magneto-optical Disk, tape, or universal serial bus (Universal Serial Bus, USB) Drive, or a combination of two or more of the foregoing. The memory 82 may include removable or non-removable (or fixed) media, where appropriate. The memory 82 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 82 is a Non-Volatile (Non-Volatile) memory. In a particular embodiment, the Memory 82 includes Read-Only Memory (ROM) and random access Memory (Random Access Memory, RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, abbreviated PROM), an erasable PROM (ErasableProgrammable Read-Only Memory, abbreviated EPROM), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, abbreviated EEPROM), an electrically rewritable ROM (ElectricallyAlterable Read-Only Memory, abbreviated EAROM), or a FLASH Memory (FLASH), or a combination of two or more of these. The RAM may be Static Random-Access Memory (SRAM) or dynamic Random-Access Memory (Dynamic Random Access Memory DRAM), where the DRAM may be a fast page mode dynamic Random-Access Memory (Fast Page Mode Dynamic Random AccessMemory FPMDRAM), extended data output dynamic Random-Access Memory (Extended Date Out Dynamic Random Access Memory EDODRAM), synchronous dynamic Random-Access Memory (Synchronous Dynamic Random-Access Memory SDRAM), or the like, as appropriate.
Memory 82 may be used to store or cache various data files that need to be processed and/or communicated, as well as possible computer program instructions for execution by processor 81.
The processor 81 reads and executes the computer program instructions stored in the memory 82 to implement any one of the rail station peak passenger flow prediction methods of the above embodiments.
In some of these embodiments, the computer device may also include a communication interface 83 and a bus 80. As shown in fig. 5, the processor 81, the memory 82, and the communication interface 83 are connected to each other through the bus 80 and perform communication with each other.
The communication interface 83 is used to implement communications between various modules, devices, units, and/or units in embodiments of the present application. Communication port 83 may also enable communication with other components such as: and the external equipment, the image/data acquisition equipment, the database, the external storage, the image/data processing workstation and the like are used for data communication.
Bus 80 includes hardware, software, or both, coupling components of the computer device to each other. Bus 80 includes, but is not limited to, at least one of: data Bus (Data Bus), address Bus (Address Bus), control Bus (Control Bus), expansion Bus (Expansion Bus), local Bus (LocalBus). By way of example, and not limitation, bus 80 may include a graphics acceleration interface (Accelerated Graphics Port), abbreviated AGP, or other graphics Bus, an enhanced industry standard architecture (ExtendedIndustry Standard Architecture, abbreviated EISA) Bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an industry standard architecture (Industry Standard Architecture, ISA) Bus, a wireless bandwidth (InfiniBand) interconnect, a Low Pin Count (LPC) Bus, a memory Bus, a micro channel architecture (Micro Channel Architecture, abbreviated MCa) Bus, a peripheral component interconnect (Peripheral Component Interconnect, abbreviated PCI) Bus, a PCI-Express (PCI-X) Bus, a serial advanced technology attachment (Serial Advanced Technology Attachment, abbreviated SATA) Bus, a video electronics standards association local (VideoElectronics Standards Association Local Bus, abbreviated VLB) Bus, or other suitable Bus, or a combination of two or more of the foregoing. Bus 80 may include one or more buses, where appropriate. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.
Compared with the prior art, the method provided by the invention is based on specific periodicity and time period rules presented by passenger flow in the railway passenger station, and the method for predicting the peak accumulated passenger flow of the railway passenger station mainly comprises 2 parts: firstly, constructing a passenger flow data set of a railway passenger station departure, constructing a prediction model based on LSTM, and predicting the proportion of ticket selling quantity from a specified railway station to the total ticket selling quantity from the railway station in a day before driving in a pre-selling period; secondly, constructing a daily peak passenger flow data set in the railway passenger station, constructing a prediction model based on a decision tree, identifying the daily peak passenger flow time period of the specified railway station, and predicting the peak passenger flow of the specified railway station according to the result in the first part. The method considers the periodic and time-period characteristics of the daily departure passenger flow of the appointed railway station, adopts a distributed prediction and identification mode to complete the prediction of the peak passenger flow in the station in one day before driving, and provides decision basis for the advanced appointed management plan of the railway passenger station.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (10)
1. The railway station peak passenger flow prediction method is characterized by comprising the following steps of:
the ticket selling amount proportion prediction model construction step: based on the periodic characteristics of the proportion of the ticket selling quantity before driving in a preset station selling period to the total ticket selling quantity from the preset station, constructing a ticket selling quantity proportion prediction model based on LSTM to predict the proportion of the ticket selling quantity before driving to the total ticket selling quantity;
the construction step of the peak period prediction model comprises the following steps: based on the time-period characteristics of peak passenger flow in the station starting from the appointed station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
and (3) predicting peak passenger flow: according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of departure people in the whole day, the prediction of the passenger flow in the peak time period in the railway station in the driving day is realized;
the ticket selling amount proportion prediction model construction step further comprises the following steps:
constructing a data set of the number of people before driving: constructing a data set of the ratio of the number of people before driving to the ratio of the number of people before driving in the day before driving in the pre-selling period;
Constructing an LSTM model: aiming at the data set of the number of people before driving, constructing a characteristic data set based on the number of days of a specified sliding window, constructing an LSTM-based prediction model based on the characteristic data set, and determining LSTM prediction model parameters needing continuous optimization;
constructing a parameter optimization model: and taking the average absolute percentage error as an optimization target, and performing iterative computation on the LSTM prediction model parameters by adopting a Bayesian optimization method to obtain an optimized parameter set of the LSTM prediction model parameters.
2. The method for predicting peak passenger flow of railway station according to claim 1, wherein the ticket selling amount before driving is the ticket selling amount from the specified station in the day before driving in the pre-selling period of the specified station, and the total ticket selling amount is the total ticket selling amount from the specified station.
3. The method for predicting peak passenger flow of railway station as set forth in claim 2, wherein the pre-drive population ratio data set comprises: the method comprises the steps of starting date, ticket selling amount before starting in a pre-selling period, total number of starting persons, proportion of total number of starting persons in the pre-selling period after starting, and normalizing the data set of the number of starting persons before starting by a linear normalization method.
4. The railway station peak passenger flow prediction method according to claim 2, wherein the peak time prediction model construction step includes:
constructing an in-station people data set: dividing the departure day according to a set time interval, calculating the number of people in the departure day station, sorting the number of people in the station from big to small, marking whether a plurality of appointed time periods are peak number periods of the number of people in the station, and forming an in-station number data set of each time period of the driving day, wherein the in-station number data set comprises departure date, departure time period, the number of people in the plurality of appointed time periods, the number of people in the whole day, the proportion of the number of people in the whole day in the appointed time period and the appointed sorting number of the plurality of people;
the step of building an intra-station peak time identification model: respectively extracting data sets of appointed number of people ordering numbers in a plurality of time ranges of a departure day in a peak time according to the number data sets in the station, and constructing an identification model of the peak time in the station based on a decision tree;
the step of constructing a predicting model of the proportion of the peak number to the number of departure all the day: and respectively extracting the data sets of the appointed number of people ordering numbers in a plurality of time ranges of the departure day in the peak time aiming at the number data sets in the station, and constructing a proportion prediction model of the peak number of people occupying the departure number of the whole day.
5. The method for predicting peak passenger flow of a railway station of claim 4, wherein the step of predicting peak passenger flow comprises:
and predicting and calculating the number of people in the peak station in a plurality of appointed time slots on the departure day based on the prediction result of the recognition model of the peak time slot in the station and the ratio prediction model of the number of people in the peak time slot accounting for the departure number on the whole day on the basis of the prediction result of the day before the departure and the ratio of the ticket selling amount accounting for the whole ticket amount before the departure.
6. A method of predicting peak passenger flow of a railway station as set forth in claim 3, wherein the constructing an LSTM model step includes:
constructing a characteristic data set: taking the proportion of the ticket selling quantity before driving of the appointed sliding window number of days before the date to be measured to the total ticket selling quantity as a characteristic, constructing a characteristic data set based on the appointed sliding window number of days, wherein the appointed sliding window number of days is determined by the periodic characteristic of the data set of the proportion of the number of people before driving;
constructing an LSTM prediction model: based on the characteristic data set, constructing a prediction model based on the proportion of ticket sales amount of the day before departure to total departure number based on LSTM, and predicting to obtain the proportion of the date to be predicted and the actual departure number of the day before driving on the day of predicting driving;
Confirming model parameter optimization: determining LSTM prediction model parameters to be optimized according to the LSTM prediction model, wherein the LSTM prediction model parameters comprise: hidden layer size, feature dimension, number of batch processes, and number of training iterations.
7. The method for predicting peak passenger flow of a railway station of claim 5, wherein the step of predicting peak passenger flow comprises: the peak passenger flow is predicted as follows: s is(s) y /(p d* q m ) Wherein s is y To sell ticket before driving, p d Q is the proportion of the ticket selling quantity accounting for the total ticket selling quantity before driving 1 to d days before the date to be measured m The peak number of m time periods on the same day of departure is the proportion of the number of departure people on the whole day.
8. A railway station peak passenger flow prediction system employing the railway station peak passenger flow prediction method as set forth in claims 1 to 7, characterized in that the railway station peak passenger flow prediction system includes:
the ticket selling amount proportion prediction model building module: based on the periodic characteristics of the proportion of the ticket selling quantity before driving in a preset station selling period to the total ticket selling quantity from the preset station, constructing a ticket selling quantity proportion prediction model based on LSTM to predict the proportion of the ticket selling quantity before driving to the total ticket selling quantity;
The peak period prediction model building module: based on the time-period characteristics of peak passenger flow in the station starting from the appointed station every day, dividing peak passenger flow sample data according to time intervals and marking peak time periods, constructing a peak time period prediction model based on a decision tree so as to predict and identify the proportion of peak time periods and peak people to the number of people starting all the day;
peak passenger flow prediction module: and according to the predicted proportion of the ticket selling quantity before driving to the total ticket selling quantity and the proportion of the peak number to the number of departure people in the whole day, the prediction of the peak time passenger flow in the railway station in the driving day is realized.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the railway station peak passenger flow prediction method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the railway station peak passenger flow prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310353020.6A CN116128160B (en) | 2023-04-04 | 2023-04-04 | Method, system, equipment and medium for predicting peak passenger flow of railway station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310353020.6A CN116128160B (en) | 2023-04-04 | 2023-04-04 | Method, system, equipment and medium for predicting peak passenger flow of railway station |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116128160A CN116128160A (en) | 2023-05-16 |
| CN116128160B true CN116128160B (en) | 2023-07-28 |
Family
ID=86306569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310353020.6A Active CN116128160B (en) | 2023-04-04 | 2023-04-04 | Method, system, equipment and medium for predicting peak passenger flow of railway station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116128160B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116564120A (en) * | 2023-07-05 | 2023-08-08 | 深圳市京视智能科技有限公司 | Station information intelligent display method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010061321A (en) * | 2008-09-03 | 2010-03-18 | Railway Technical Res Inst | Passenger flow prediction system |
| CN115660921A (en) * | 2022-10-21 | 2023-01-31 | 北京经纬信息技术有限公司 | Method, system, equipment and medium for automatically generating railway station passenger flow organization strategy |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897815B (en) * | 2017-01-17 | 2021-03-12 | 北京万相融通科技股份有限公司 | Method for predicting passenger flow trend of station in real time |
| CN111178598A (en) * | 2019-12-16 | 2020-05-19 | 中国铁道科学研究院集团有限公司 | Passenger flow prediction method and system for railway passenger station, electronic device and storage medium |
| CN112598182B (en) * | 2020-12-25 | 2022-11-29 | 卡斯柯信号有限公司 | Intelligent scheduling method and system for rail transit |
| CN113792907A (en) * | 2021-08-06 | 2021-12-14 | 上海申铁信息工程有限公司 | Passenger flow prediction method and device for railway passenger station and storage medium |
| CN114529032A (en) * | 2021-12-27 | 2022-05-24 | 浙江浙大网新众合轨道交通工程有限公司 | Subway section passenger flow prediction method based on LSTM network |
-
2023
- 2023-04-04 CN CN202310353020.6A patent/CN116128160B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010061321A (en) * | 2008-09-03 | 2010-03-18 | Railway Technical Res Inst | Passenger flow prediction system |
| CN115660921A (en) * | 2022-10-21 | 2023-01-31 | 北京经纬信息技术有限公司 | Method, system, equipment and medium for automatically generating railway station passenger flow organization strategy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116128160A (en) | 2023-05-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009972B (en) | Multi-mode travel O-D demand estimation method based on multi-source data check | |
| CN107194491A (en) | A kind of dynamic dispatching method based on Forecasting of Travel Time between bus passenger flow and station | |
| CN113077090A (en) | Passenger flow prediction method, system and computer readable storage medium | |
| CN116128160B (en) | Method, system, equipment and medium for predicting peak passenger flow of railway station | |
| CN115018148A (en) | Urban rail transit network passenger flow distribution prediction method and system based on digital twin model | |
| CN111063204B (en) | Expressway vehicle speed prediction model training method based on toll station flow | |
| CN112507624B (en) | Inter-city road trip mode identification model construction and identification method and device | |
| CN110363329A (en) | One kind being based on the matched net of the bilateral satisfaction of supply and demand about vehicle worksheet processing method | |
| CN112562377A (en) | Passenger vehicle real-time scheduling method based on random opportunity constraint | |
| CN116663811A (en) | Scheduling matching method and device for reciprocating dynamic carpooling of inter-city passenger transport | |
| CN116933946A (en) | Rail transit OD passenger flow prediction method and system based on passenger flow destination structure | |
| CN110222892B (en) | Passenger getting-off station prediction method and device | |
| CN112258029A (en) | Demand prediction method for shared bicycle around subway station | |
| CN116128172A (en) | Air-iron intermodal route generation method, system and equipment and storage medium | |
| CN112669595B (en) | Network taxi booking flow prediction method based on deep learning | |
| CN109118412B (en) | Urban rail transit network passenger flow online control system | |
| CN117523900A (en) | Parking lot occupancy prediction method and system | |
| CN110796301B (en) | Passenger flow prediction method and device based on IC card data | |
| CN116564551B (en) | Data-knowledge driven urban rail transit risk identification method | |
| CN112926796A (en) | Get-off point recommendation method and device based on specific scene | |
| CN109447352B (en) | Bus trip OD matrix reverse-thrust combination method | |
| CN115994787A (en) | Car pooling demand prediction matching method based on neural network | |
| CN114707284B (en) | High-speed railway section throughput simulation computing system | |
| CN114492967A (en) | Urban rail station passenger flow prediction method and medium based on CEEMDAN and BLSTM combined model | |
| CN114444795A (en) | Single-line bus passenger travel data generation method |
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 |