CN110458325A - A kind of traffic zone passenger flow forecasting and device in short-term - Google Patents
A kind of traffic zone passenger flow forecasting and device in short-term Download PDFInfo
- Publication number
- CN110458325A CN110458325A CN201910601311.6A CN201910601311A CN110458325A CN 110458325 A CN110458325 A CN 110458325A CN 201910601311 A CN201910601311 A CN 201910601311A CN 110458325 A CN110458325 A CN 110458325A
- Authority
- CN
- China
- Prior art keywords
- passenger flow
- prediction
- period
- data
- traffic zone
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims description 21
- 238000003062 neural network model Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 10
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 abstract description 19
- 230000002452 interceptive effect Effects 0.000 abstract description 4
- 230000015654 memory Effects 0.000 description 18
- 238000010586 diagram Methods 0.000 description 14
- 210000002569 neuron Anatomy 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000003860 storage Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 150000001875 compounds Chemical class 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 210000005036 nerve Anatomy 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Tourism & Hospitality (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Fuzzy Systems (AREA)
- Databases & Information Systems (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the invention discloses a kind of traffic zone passenger flow forecasting and devices in short-term, method includes: the prediction period for obtaining traffic zone passenger flow estimation, the corresponding multi-class data of prediction period is determined from history passenger flow information database, it is separately input to prediction period corresponding multi-class data to obtain multiple prediction results in multiclass prediction model, the multiple prediction result is integrated, determines the volume of the flow of passengers of traffic zone passenger flow estimation.The present invention by the changing rule of the analysis volume of the flow of passengers considers that model can be trained using more targeted feature the factor that result has an impact.By the structure for modifying neural network, full Connection Neural Network is changed into the splicing of multiple neural networks, eliminate interfering with each other for different types of data, to improve the accuracy of transport hub region passenger flow estimation in short-term, data support is provided for the whole control of hinge passenger flow state, the integrity service for improving hinge is horizontal.
Description
Technical field
The present invention relates to traffic zone a kind of in technical field of transportation more particularly to Transportation Terminal passenger flow estimations in short-term
Method and device.
Background technique
The core function of transport hub is that passenger flow is collecting and distributing and passenger transference, space division when all management works of hinge surround passenger flow
Cloth is carried out, and is capable of providing the following hinge of precognition and reaches/set out the volume of the flow of passengers, is the urgent need of hinge management.It is real-time based on hinge
Acquisition and history passenger flow data influence hinge passenger flow in conjunction with periphery socio-economic activity, climate change, constitute to hinge passenger flow
And formation mechenism is analysed in depth, and then constructs different Passenger flow forecast model, obtain passenger flow time change or when space division
The Evolution of cloth predicts the following hinge in advance and reaches/set out the volume of the flow of passengers, is hinge scheduling of resource, emergency evacuation, sector planning
Data supporting is provided.
In the hinge operation of reality, key area is such as in hinge: bus platform, taxi waiting point volume of the flow of passengers mistake
Big situation relies primarily on monitor video or administrative staff's personal experience's subjective judgement, to the scarce capacity of passenger flow estimation.Due to
Passenger flow influence factor randomness and complexity make judging result and passenger flow forecast situation there are biggish deviation, cause to dispatch
It is poor to manage real-time, effective temporary scheduling measure cannot be taken in time, bring security risk.Passenger's trip requirements can not have
Effect is met, and the effective operation and service level of traffic system are directly affected.Therefore using reasonable model algorithm to visitor
Stream, which carries out accurately prediction, seems particularly important, carries out Field Force, service setting layout according to passenger flow estimation on this basis,
Effectively passenger can be dredged.
Existing passenger flow forecasting is by using LSTM (Long Short-Term Memory, shot and long term memory network)
Feature correlation is analyzed, Lai Tigao prediction accuracy.The prediction of its considered passenger flow and history passenger flow have certain
Correlation improve essence by the way that as far as possible operation will be carried out in the model being added with the historical data before predicted time point
Degree.Although this method has some improvement traditional neural network model, it does not consider the particularity of traffic passenger flow.
Current passenger flow data is not only influenced by moment passenger flow data before, and passenger flow data also has according to the regular wave of week progress
It is dynamic, and daily synchronization passenger flow similitude with higher.Simultaneously by history passenger flow data can not to too far when
Between predicted.Therefore passenger flow data at the time of rely only on continuous can not compound passenger flow mechanical periodicity completely characteristic, by shadow
Ring the accuracy of final estimation result.Existing passenger flow forecasting uses the algorithm model of full Connection Neural Network, is connecting entirely
It connects in neural network, each input can be attached with the neuron of hidden layer.If input data attribute is different, and number
When larger according to difference, the different same neurons of input control, different types of data can have an impact the neuron, lead to mind
It is interfered with each other through first weight, to influence whole accuracy.
Summary of the invention
The embodiment of the present invention provides a kind of traffic zone passenger flow forecasting and device in short-term, to solve in the prior art
Passenger flow forecasting in the comprehensive passenger transport hub inner region low problem of passenger flow estimation accuracy rate in short-term.
In a first aspect, the embodiment of the present invention provides a kind of traffic zone passenger flow forecasting in short-term, comprising:
Obtain the prediction period of traffic zone passenger flow estimation；
The corresponding multi-class data of the prediction period is determined from history passenger flow information database；
The corresponding multi-class data of the prediction period is separately input in multiclass prediction model, multiple prediction knots are obtained
Fruit；Wherein, the corresponding a kind of prediction model of a kind of data；The multiclass prediction model is according to the history passenger flow information database
In the training of history passenger flow data obtain；
The multiple prediction result is integrated, determines the volume of the flow of passengers of the traffic zone passenger flow estimation.
Above scheme, by analyzing the changing rule of the volume of the flow of passengers, consideration can be to the factor that result has an impact, using more
Targeted feature is trained model, according to the far and near adjustment predictive reference data of predicted time, gets over predicted time
Nearly referenced valid data are more, to improve the accuracy of algorithm.By modifying the structure of neural network, mind will be connected entirely
The splicing for changing multiple neural networks into through network eliminates interfering with each other for different types of data, to improve transport hub area
The accuracy of domain passenger flow estimation in short-term provides data and supports, improves the whole clothes of hinge for the whole control of hinge passenger flow state
Business is horizontal.
Optionally, the corresponding multi-class data of the prediction period include be located at the prediction period before and with the prediction
The history passenger flow data of adjacent N number of period period, it is M days first in period identical with the prediction period daily history passenger flow
Data, it is P weeks first in the prediction period on the same day and the history passenger flow data of period identical with the prediction period；Wherein,
M, N, P are positive integer.
Above scheme, it is contemplated that the particularity of traffic passenger flow, current passenger flow data is not only by moment passenger flow data before
It influences, passenger flow data also has according to the regular fluctuation of week progress, and daily synchronization passenger flow is with higher
Similitude.The time too far can not be predicted by history passenger flow data simultaneously.Therefore visitor at the time of relying only on continuous
Flow data can not completely compound passenger flow mechanical periodicity characteristic, will affect the accuracy of final prediction result.By by consecutive hours
The passenger flow data at quarter is screened, and input is located at before the prediction period and N number of period adjacent with the prediction period
History passenger flow data, it is M days first in daily the history passenger flow data of period identical with the prediction period, it is P weeks first in it is described
Prediction period is on the same day and the history passenger flow data of period identical with the prediction period trains prediction model, to improve prediction
As a result accuracy.
Optionally, the history passenger flow data training according in the history passenger flow information database obtains the multiclass
Prediction model, comprising:
Obtain multiple trained periods；
Corresponding a variety of data of each trained period are determined from the history passenger flow data；
According to each trained period corresponding a variety of data, are trained to preset multiple neural network models
It practises, obtains the multiclass prediction model.
Above scheme has more targetedly historical data training pattern according to prediction period by extracting, and will screen
Historical data afterwards is divided into different types, and same type of historical data trains corresponding prediction model, so that prediction result
Accuracy be improved.
Optionally, described to integrate the multiple prediction result, determine the visitor of the traffic zone passenger flow estimation
Flow, comprising:
The multiple prediction result is connected entirely, obtains the volume of the flow of passengers of the traffic zone passenger flow estimation.
Full Connection Neural Network is changed into the spelling of multiple neural networks by modifying the structure of neural network by above scheme
It connects, eliminates interfering with each other for different types of data, to improve the accuracy of transport hub region passenger flow estimation in short-term.
Optionally, before the prediction period for obtaining traffic zone passenger flow estimation, further includes:
Obtain the passenger flow information of the disengaging traffic zone of passenger flow acquisition equipment identification；
The passenger flow information of the traffic zone is cleaned；
The passenger flow information of the traffic zone after cleaning is stored in the history passenger flow information database.
Above scheme is cleaned by the passenger flow information to the traffic zone, prevents dirty data from influencing prediction result,
To the operation and the good decision support of management offer to transport hub.
Second aspect, the embodiment of the present invention provide a kind of traffic zone passenger flow estimation device in short-term, comprising:
Module is obtained, for obtaining the prediction period of traffic zone passenger flow estimation；
Processing module, for determining the corresponding multi-class data of the prediction period from history passenger flow information database；It will
The corresponding multi-class data of the prediction period is separately input in multiclass prediction model, obtains multiple prediction results；Wherein, a kind of
The corresponding a kind of prediction model of data；The multiclass prediction model is according to the history passenger flow in the history passenger flow information database
Data training obtains；The multiple prediction result is integrated, determines the volume of the flow of passengers of the traffic zone passenger flow estimation.
Optionally, the corresponding multi-class data of the prediction period include be located at the prediction period before and with the prediction
The history passenger flow data of adjacent N number of period period, it is M days first in period identical with the prediction period daily history passenger flow
Data, it is P weeks first in the prediction period on the same day and the history passenger flow data of period identical with the prediction period；Wherein,
M, N, P are positive integer.
Optionally, the processing module is specifically used for:
The multiple prediction result is connected entirely, obtains the volume of the flow of passengers of the traffic zone passenger flow estimation.
Optionally, the processing module is also used to:
Before the prediction period for obtaining traffic zone passenger flow estimation, the disengaging institute of passenger flow acquisition equipment identification is obtained
State the passenger flow information of traffic zone；
The passenger flow information of the traffic zone is cleaned；
The passenger flow information of the traffic zone after cleaning is stored in the history passenger flow information database.
Optionally, the processing module is specifically used for:
Obtain multiple trained periods；
Corresponding a variety of data of each trained period are determined from the history passenger flow data；
According to each trained period corresponding a variety of data, are trained to preset multiple neural network models
It practises, obtains the multiclass prediction model.
The third aspect, the embodiment of the present invention provide a kind of computer equipment, comprising:
Memory, for storing program instruction；
Processor executes above-mentioned first according to the program of acquisition for calling the program instruction stored in the memory
Method described in aspect.
Fourth aspect, the embodiment of the present invention provides a kind of computer-readable non-volatile memory medium, including computer can
Reading instruction, when computer is read and executes the computer-readable instruction, so that computer executes described in above-mentioned first aspect
Method.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill in field, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of schematic diagram of system architecture provided in an embodiment of the present invention；
Fig. 2 is a kind of flow diagram of traffic zone provided in an embodiment of the present invention passenger flow forecasting in short-term；
Fig. 3 is a kind of schematic diagram of traditional neural network model provided in an embodiment of the present invention；
Fig. 4 is a kind of schematic diagram of parallel system neural network model provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of traffic zone provided in an embodiment of the present invention passenger flow estimation device in short-term.
Specific embodiment
In order to better understand the above technical scheme, below in conjunction with Figure of description and specific embodiment to above-mentioned
Technical solution is described in detail, it should be understood that the specific features in the embodiment of the present invention and embodiment are to skill of the present invention
The detailed description of art scheme, rather than the restriction to technical solution of the present invention, in the absence of conflict, the embodiment of the present invention
And the technical characteristic in embodiment can be combined with each other.
Fig. 1 illustratively shows a kind of system architecture that the embodiment of the present invention is applicable in, which can be clothes
Business device 100, including processor 110, communication interface 120 and memory 130.
Wherein, communication interface 120 is received and dispatched the information of terminal device transmission, is realized for being communicated with terminal device
Communication.
Processor 110 is the control centre of server 100, connects entire server 100 with route using various interfaces
Various pieces by running or execute the software program/or module that are stored in memory 130, and are called and are stored in storage
Data in device 130, the various functions and processing data of execute server 100.Optionally, processor 110 may include one
Or multiple processing units.
Memory 130 can be used for storing software program and module, and processor 110 is stored in memory 130 by operation
Software program and module, thereby executing various function application and data processing.Memory 130 can mainly include storage journey
Sequence area and storage data area, wherein storing program area can application program needed for storage program area, at least one function etc.；
Storage data area can store the data etc. created according to business processing.In addition, memory 130 may include high random access
Memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device or other are volatile
Property solid-state memory.
It should be noted that above-mentioned structure shown in FIG. 1 is only a kind of example, it is not limited in the embodiment of the present invention.
Based on foregoing description, Fig. 2 illustratively shows a kind of traffic zone provided in an embodiment of the present invention passenger flow in short-term
The process of prediction technique, the process can passenger flow estimation device executes in short-term by traffic zone, which can be located at above-mentioned Fig. 1
In the server, it is also possible to the server.
As shown in Fig. 2, the process specifically includes:
Step 201, the prediction period of traffic zone passenger flow estimation is obtained.
Specifically, before the prediction period for obtaining traffic zone passenger flow estimation, it is also necessary to obtain passenger flow acquisition equipment and know
The passenger flow information of other disengaging traffic zone, cleans the passenger flow information of traffic zone, by the traffic zone after cleaning
Passenger flow information is stored in history passenger flow information database.
It should be noted that all entrances of traffic zone to be predicted dispose passenger flow acquisition equipment, volume of the flow of passengers letter is obtained
The equipment that the mode of breath mainly passes through the recognizable volumes of the flow of passengers such as entrance passenger flow acquisition equipment, video analytics server.In region
Each entrance disposes passenger flow acquisition equipment, and carries out clock check to passenger flow capture setting, and unified data transmitting pin is arranged
Rate transmits data to backstage.The real-time passenger flow in the region is calculated by acquiring device data processing to passenger flow in central server
Measure information.
For example, table 1 shows the pattern of single passenger flow acquisition equipment transmission data.Fig. 3 is the acquisition of entrance passenger flow
The schematic diagram of equipment, as shown in figure 3, the region the a001 i period passes in and out volume of the flow of passengers calculating:
Si=Si s001+Si s002+Si s003
Wherein: Si: the region the a001 i period enters passenger flow quantity；
Si s001: passenger flow acquire s001 the i period acquire into passenger flow value；
Si s002: passenger flow acquire s002 the i period acquire into passenger flow value；
Si s003: passenger flow acquire s003 the i period acquire into passenger flow value；
Oi=Oi s001+Oi s002+Oi s003
Wherein: Oi: the region the a001 i period leaves passenger flow quantity；
Oi s001: passenger flow acquire s001 the i period acquire leave passenger flow value；
Oi s002: passenger flow acquire s002 the i period acquire leave passenger flow value；
Oi s003: passenger flow acquire s003 the i period acquire leave passenger flow value；
Table 1
| Device coding | Into | Out | Time |
| s001 | 12 | 24 | 2018-10-10 08:00:00 |
| s002 | 28 | 13 | 2018-10-10 08:00:00 |
| s003 | 32 | 10 | 2018-10-10 08:00:00 |
Table 2 shows the pattern of region passenger flow acquisition equipment transmission data.
Qi a001=qi0 a001-qi1 a001+Q(i-1) a001
Wherein: Qi: in i period a001 region passenger flow quantity；
qi0: enter the region a001 passenger flow quantity in the i period；
qi0: the region a001 passenger flow quantity is left in the i period；
Q(i-1) a001: in i last moment a001 region passenger flow quantity.
Table 2
| Region | Passenger flow quantity | Into | Out | Time |
| a001 | 25 | 72 | 47 | 2018-10-10 08:00:00 |
| a002 | 78 | 102 | 60 | 2018-10-10 08:00:00 |
| a003 | 71 | 86 | 42 | 2018-10-10 08:00:00 |
In addition, data cleansing is carried out to required history passenger flow while passenger flow estimation, to prevent dirty data from tying to prediction
The influence of fruit provides good decision support with management for hinge operation.
Step 202, the corresponding multi-class data of the prediction period is determined from history passenger flow information database.
The passenger flow information of equipment acquisition is acquired by passenger flow, and is stored in after history passenger flow information database, so that it may root
It is predicted that the period determines the corresponding multi-class data of prediction period, the corresponding multi-class data of the prediction period includes when being located at prediction
Before section and the history passenger flow data of N number of period adjacent with prediction period, it is M days first in period identical with prediction period daily
History passenger flow data, preceding P Zhou Zhongyu prediction period on the same day and the history passenger flow data of period identical with prediction period.Its
In, M, N, P are positive integer.
Step 203, the corresponding multi-class data of the prediction period is separately input in multiclass prediction model, is obtained multiple
Prediction result.
In embodiments of the present invention, which can correspond with the multi-class data of prediction period, wherein
The corresponding a kind of prediction model of a kind of data.The multiclass prediction model can be according to the history passenger flow in history passenger flow information database
Data training obtains, specifically, multiple trained periods can first be obtained, each training is then determined from history passenger flow data
Period corresponding a variety of data, finally according to corresponding a variety of data of each trained period, to preset multiple neural network moulds
Type is trained study, obtains multiclass prediction model.
For example, being observed by right history passenger flow data, discovery day passenger flow variation was showed with (7 days) week for week
Phase carries out circulation change, and the hour volume of the flow of passengers, which is showed, carries out circulation change with day for the period in (24 hours).Integrally become for passenger flow
Law, former Zhou Zhongyu prediction periods are on the same day and the history passenger flow data of period identical with prediction period is as a kind of defeated
Enter.In view of influence factors such as similar summits, the influence taken several days can be generated to the volume of the flow of passengers, and it is identical that the volume of the flow of passengers can be made to generate
Variation tendency, therefore using the history passenger flow data of period identical with prediction period is inputted as one kind daily in a few days ago.
In view of corresponding day it is possible that the factors such as rainy weather influence, and these influence factors can be arrived in preceding several hours with regard to reflection
In the variation of same day passenger flow, therefore before prediction period being located at and the history visitor of the several periods adjacent with prediction period
Flow data is as a category feature.
Step 204, the multiple prediction result is integrated, determines the passenger flow of the traffic zone passenger flow estimation
Amount.
Specifically, multiple prediction results can be connected entirely, the volume of the flow of passengers of traffic zone passenger flow estimation is obtained.Also
It is to say, multiple prediction results is attached by a full Connection Neural Network, so that it may obtain traffic zone passenger flow
The volume of the flow of passengers of prediction.
It should be noted that the embodiment of the present invention in full Connection Neural Network structure basis, proposes that one kind " is divided and controlled
It " strategy, a neural network is divided into three neural network modules, to solve to hide caused by input type is different
The problem of layer weight interferes with each other.Three kinds of factors are realized by parallel network structure, utilize three traditional neural networks
The processing of model independently these three input information, one kind be used to extract former Zhou Zhongyu prediction periods on the same day and with prediction when
The passenger flow feature of the history passenger flow data of section identical period, one kind are used to extract identical with prediction period daily in a few days ago
The passenger flow feature of the history passenger flow data of period, it is a kind of for extract be located at prediction period before and it is adjacent with prediction period several
The passenger flow feature of the history passenger flow data of a period, this namely so-called " segmentation " process；Finally by the output of three kinds of networks
Integration is carried out to obtain final prediction result.Improved model is called parallel system neural network model by us.By changing
Into, it allows the isolated operation in the neural network of oneself of different type number, prevents different types of data from interfering with each other, it may be effectively
Raising forecasting accuracy.
Embodiment in order to preferably explain the present invention, below according to Fig. 2 involved in the embodiment of the present invention will be further explained
The schematic diagram of traditional neural network model, as shown in Fig. 2, being the schematic diagram of traditional neural network model.
The short-term passenger flow estimation in region is predicted that neural network is made of neuron using neural network model, neuron
Output formula be formula (1):
Wherein, P=[P1,P2,…,PR]TFor the input vector of neuron；W=[w1,w2,…,wR]TExpression is input to nerve
The intensity of first R connection, i.e. weight vector, the value of each element can just be born, and reinforcement or inhibiting effect are respectively indicated；θ is
The threshold values of neuron, if the weighted input of neuron and being greater than θ, neuron is activated；The input and output of f expression neuron
Function, that is, conventional function.
As shown in figure 3, a kind of schematic diagram of the parallel system neural network model proposed for the embodiment of the present invention.
Parallel system neural network mainly includes three independent modules: be located at prediction period before and with prediction period phase
The history passenger flow data module of adjacent N number of period, it is M days first in period identical with prediction period daily history passenger flow data mould
Block, preceding P Zhou Zhongyu prediction period are on the same day and the history passenger flow data module of period identical with prediction period.Each module
It is to be made of traditional neural network model.Using the passenger flow data of three classes as the input factor of model.In parallel system nerve
In network model, this 9 inputs are divided into three classes, as shown in figure 3, X1, X2, X3 be one kind, X4, X5, X6 be one kind, X7, X8,
X9 is one kind respectively as the input for being each independent prediction module.After three modules are completed in building, by the pre- of three modules
Output data is surveyed to be integrated to obtain final prediction result.
Based on the same technical idea, Fig. 4 illustratively shows the embodiment of the present invention and provides a kind of traffic zone in short-term
The structure of passenger flow estimation device, the device can execute the process of traffic zone passenger flow estimation in short-term, which can be located at upper
It states in server described in Fig. 1, is also possible to the server.
As shown in figure 5, the apparatus may include:
Module 501 is obtained, for obtaining the prediction period of traffic zone passenger flow estimation；
Processing module 502, for determining the corresponding multi-class data of the prediction period from history passenger flow information database；
The corresponding multi-class data of the prediction period is separately input in multiclass prediction model, multiple prediction results are obtained；Wherein, one
The corresponding a kind of prediction model of class data；The multiclass prediction model is according to the history visitor in the history passenger flow information database
Flow data training obtains；The multiple prediction result is integrated, determines the passenger flow of the traffic zone passenger flow estimation
Amount.
Optionally, the corresponding multi-class data of the prediction period include be located at the prediction period before and with the prediction
The history passenger flow data of adjacent N number of period period, it is M days first in period identical with the prediction period daily history passenger flow
Data, it is P weeks first in the prediction period on the same day and the history passenger flow data of period identical with the prediction period；Wherein,
M, N, P are positive integer.
Optionally, the processing module 502 is specifically used for:
The multiple prediction result is connected entirely, obtains the volume of the flow of passengers of the traffic zone passenger flow estimation.
Optionally, the processing module 502 is also used to:
Before the prediction period for obtaining traffic zone passenger flow estimation, the disengaging institute of passenger flow acquisition equipment identification is obtained
State the passenger flow information of traffic zone；
The passenger flow information of the traffic zone is cleaned；
The passenger flow information of the traffic zone after cleaning is stored in the history passenger flow information database.
Optionally, the processing module 502 is specifically used for:
Obtain multiple trained periods；
Corresponding a variety of data of each trained period are determined from the history passenger flow data；
According to each trained period corresponding a variety of data, are trained to preset multiple neural network models
It practises, obtains the multiclass prediction model.
Based on the same technical idea, the embodiment of the invention also provides a kind of calculating equipment, comprising:
Memory, for storing program instruction；
Processor executes above-mentioned traffic according to the program of acquisition for calling the program instruction stored in the memory
Region passenger flow forecasting in short-term.
Based on the same technical idea, the embodiment of the invention also provides a kind of computer-readable non-volatile memories to be situated between
Matter, including computer-readable instruction, when computer is read and executes the computer-readable instruction, so that computer executes
State traffic zone passenger flow forecasting in short-term.
Finally, it should be noted that it should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, be
System or computer program product.Therefore, the present invention can be used complete hardware embodiment, complete software embodiment or combine software
With the form of the embodiment of hardware aspect.Moreover, it wherein includes that computer can use journey that the present invention, which can be used in one or more,
The computer implemented in the computer-usable storage medium (including but not limited to magnetic disk storage, optical memory etc.) of sequence code
The form of program product.
The present invention be referring to according to the method for the present invention, the flow chart of equipment (system) and computer program product and/or
Block diagram describes.It should be understood that each process that can be realized by computer program instructions in flowchart and/or the block diagram and/or
The combination of process and/or box in box and flowchart and/or the block diagram.It can provide these computer program instructions to arrive
General purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor to generate one
Machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for realizing flowing
The device for the function of being specified in one process of journey figure or multiple and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
Obviously, various changes and modifications can be made to the invention without departing from model of the invention by those skilled in the art
It encloses.In this way, if these modifications and changes of the present invention is within the scope of the claims of the present invention and its equivalent technology, then
The present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of traffic zone passenger flow forecasting in short-term, which is characterized in that the described method includes:
Obtain the prediction period of traffic zone passenger flow estimation；
The corresponding multi-class data of the prediction period is determined from history passenger flow information database；
The corresponding multi-class data of the prediction period is separately input in multiclass prediction model, multiple prediction results are obtained；Its
In, the corresponding a kind of prediction model of a kind of data；The multiclass prediction model is according in the history passenger flow information database
The training of history passenger flow data obtains；
The multiple prediction result is integrated, determines the volume of the flow of passengers of the traffic zone passenger flow estimation.
2. the method according to claim 1, wherein the corresponding multi-class data of the prediction period includes being located at institute
State before prediction period and the history passenger flow data of the N number of period adjacent with the prediction period, it is M days first in daily with it is described pre-
Survey identical period period history passenger flow data, it is P weeks first in and the prediction period it is on the same day and identical as the prediction period
Period history passenger flow data；Wherein, M, N, P are positive integer.
3. according to the method described in claim 2, it is characterized in that, described according to going through in the history passenger flow information database
The training of history passenger flow data obtains the multiclass prediction model, comprising:
Obtain multiple trained periods；
Corresponding a variety of data of each trained period are determined from the history passenger flow data；
According to each trained period corresponding a variety of data, study is trained to preset multiple neural network models,
Obtain the multiclass prediction model.
4. being determined the method according to claim 1, wherein described integrate the multiple prediction result
The volume of the flow of passengers of the traffic zone passenger flow estimation out, comprising:
The multiple prediction result is connected entirely, obtains the volume of the flow of passengers of the traffic zone passenger flow estimation.
5. method according to any one of claims 1 to 4, which is characterized in that in the acquisition traffic zone passenger flow estimation
Prediction period before, further includes:
Obtain the passenger flow information of the disengaging traffic zone of passenger flow acquisition equipment identification；
The passenger flow information of the traffic zone is cleaned；
The passenger flow information of the traffic zone after cleaning is stored in the history passenger flow information database.
6. a kind of traffic zone passenger flow estimation device in short-term characterized by comprising
Module is obtained, for obtaining the prediction period of traffic zone passenger flow estimation；
Processing module, for determining the corresponding multi-class data of the prediction period from history passenger flow information database；It will be described
The corresponding multi-class data of prediction period is separately input in multiclass prediction model, obtains multiple prediction results；Wherein, a kind of data
Corresponding one kind prediction model；The multiclass prediction model is according to the history passenger flow data in the history passenger flow information database
What training obtained；The multiple prediction result is integrated, determines the volume of the flow of passengers of the traffic zone passenger flow estimation.
7. device according to claim 6, which is characterized in that the corresponding multi-class data of the prediction period includes being located at
Before the prediction period and the history passenger flow data of the N number of period adjacent with the prediction period, it is M days first in daily with it is described
The history passenger flow data of prediction period identical period, it is P weeks first in the prediction period on the same day and with the prediction period phase
The history passenger flow data of same period；Wherein, M, N, P are positive integer.
8. device according to claim 7, which is characterized in that the processing module is specifically used for:
The multiple prediction result is connected entirely, obtains the volume of the flow of passengers of the traffic zone passenger flow estimation.
9. device according to claim 6, which is characterized in that the processing module is also used to:
Before the prediction period for obtaining traffic zone passenger flow estimation, the disengaging friendship of passenger flow acquisition equipment identification is obtained
The passenger flow information in logical region；
The passenger flow information of the traffic zone is cleaned；
The passenger flow information of the traffic zone after cleaning is stored in the history passenger flow information database.
10. according to the device any in claim 6 to 9, which is characterized in that the processing module is specifically used for:
Obtain multiple trained periods；
Corresponding a variety of data of each trained period are determined from the history passenger flow data；
According to each trained period corresponding a variety of data, study is trained to preset multiple neural network models,
Obtain the multiclass prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910601311.6A CN110458325A (en) | 2019-07-03 | 2019-07-03 | A kind of traffic zone passenger flow forecasting and device in short-term |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910601311.6A CN110458325A (en) | 2019-07-03 | 2019-07-03 | A kind of traffic zone passenger flow forecasting and device in short-term |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110458325A true CN110458325A (en) | 2019-11-15 |
Family
ID=68482259
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910601311.6A Pending CN110458325A (en) | 2019-07-03 | 2019-07-03 | A kind of traffic zone passenger flow forecasting and device in short-term |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110458325A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768093A (en) * | 2020-06-23 | 2020-10-13 | 中国工商银行股份有限公司 | Network point configuration method and device based on time sequence |
| CN112380925A (en) * | 2020-10-27 | 2021-02-19 | 北京城建智控科技有限公司 | Data processing method, device, equipment and storage medium |
| 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 |
| CN113822458A (en) * | 2020-11-18 | 2021-12-21 | 京东城市(北京)数字科技有限公司 | Prediction method, training method, device, electronic equipment and readable storage medium |
| CN114757447A (en) * | 2022-06-08 | 2022-07-15 | 广东工业大学 | Multi-model mixed passenger transport hub station passenger flow prediction method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104517159A (en) * | 2014-12-18 | 2015-04-15 | 上海交通大学 | Method for predicting short-time passenger flow of bus |
| US20170178044A1 (en) * | 2015-12-21 | 2017-06-22 | Sap Se | Data analysis using traceable identification data for forecasting transportation information |
| CN107451599A (en) * | 2017-06-28 | 2017-12-08 | 青岛科技大学 | A kind of traffic behavior Forecasting Methodology of the collective model based on machine learning |
| CN109034449A (en) * | 2018-06-14 | 2018-12-18 | 华南理工大学 | Short-term bus passenger flow prediction technique based on deep learning and passenger behavior mode |
| CN109376935A (en) * | 2018-10-31 | 2019-02-22 | 东南大学 | A kind of bus passenger flow neural network based combination forecasting method at times |
-
2019
- 2019-07-03 CN CN201910601311.6A patent/CN110458325A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104517159A (en) * | 2014-12-18 | 2015-04-15 | 上海交通大学 | Method for predicting short-time passenger flow of bus |
| US20170178044A1 (en) * | 2015-12-21 | 2017-06-22 | Sap Se | Data analysis using traceable identification data for forecasting transportation information |
| CN107451599A (en) * | 2017-06-28 | 2017-12-08 | 青岛科技大学 | A kind of traffic behavior Forecasting Methodology of the collective model based on machine learning |
| CN109034449A (en) * | 2018-06-14 | 2018-12-18 | 华南理工大学 | Short-term bus passenger flow prediction technique based on deep learning and passenger behavior mode |
| CN109376935A (en) * | 2018-10-31 | 2019-02-22 | 东南大学 | A kind of bus passenger flow neural network based combination forecasting method at times |
Non-Patent Citations (1)
| Title |
|---|
| 宋万清等: "《数据挖掘》", 31 January 2019 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111768093A (en) * | 2020-06-23 | 2020-10-13 | 中国工商银行股份有限公司 | Network point configuration method and device based on time sequence |
| CN112380925A (en) * | 2020-10-27 | 2021-02-19 | 北京城建智控科技有限公司 | Data processing method, device, equipment and storage medium |
| CN113822458A (en) * | 2020-11-18 | 2021-12-21 | 京东城市(北京)数字科技有限公司 | Prediction method, training method, device, electronic equipment and readable storage medium |
| 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 |
| CN114757447A (en) * | 2022-06-08 | 2022-07-15 | 广东工业大学 | Multi-model mixed passenger transport hub station passenger flow prediction method and system |
| CN114757447B (en) * | 2022-06-08 | 2022-10-11 | 广东工业大学 | Multi-model mixed passenger transport hub station passenger flow prediction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110458325A (en) | A kind of traffic zone passenger flow forecasting and device in short-term | |
| CN103632212B (en) | System and method for predicating time-varying user dynamic equilibrium network-evolved passenger flow | |
| US9639807B2 (en) | Method and system for forecasting future events | |
| Farouk El Gayar et al. | An integrated framework for advanced hotel revenue management | |
| CN109872535A (en) | A kind of current prediction technique of wisdom traffic, device and server | |
| CN107608862A (en) | Monitoring alarm method, monitoring alarm device and computer-readable recording medium | |
| CN105224992A (en) | To waiting for the method and system predicted of ridership and evaluation method and system | |
| CN108133619A (en) | Method, apparatus, storage medium and the terminal device of Parking prediction | |
| CN107679576A (en) | The fault monitoring method and device of vehicle | |
| CN107392389A (en) | Taxi dispatching processing method based on ARIMA models | |
| US10339620B2 (en) | Method of managing transportation fare, server performing the same and system performing the same | |
| CN111192090A (en) | Seat allocation method and device for flight, storage medium and electronic equipment | |
| CN116187555A (en) | Traffic flow prediction model construction method and prediction method based on self-adaptive dynamic diagram | |
| CN110363329A (en) | One kind being based on the matched net of the bilateral satisfaction of supply and demand about vehicle worksheet processing method | |
| CN112948412A (en) | Flight inventory updating method, system, electronic equipment and storage medium | |
| CN109345373A (en) | Check and write off method for prewarning risk, device, electronic equipment and computer-readable medium | |
| CN113837383B (en) | Model training method and device, electronic equipment and storage medium | |
| CN110263136B (en) | Method and device for pushing object to user based on reinforcement learning model | |
| CN113269402A (en) | Flight space control method and device and computer equipment | |
| CN109919675A (en) | Communication user upshift prediction probability recognition methods neural network based and system | |
| Langerudi et al. | Health and transportation: small scale area association | |
| CN108629381A (en) | Crowd's screening technique based on big data and terminal device | |
| CN116485589A (en) | Business analysis method and system applied to travel enterprise management | |
| CN110910241A (en) | Cash flow evaluation method, apparatus, server device and storage medium | |
| CN110210645A (en) | A kind of development trend data capture method, device and readable storage medium storing program for executing |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191115 |