CN109300310A - A kind of vehicle flowrate prediction technique and device - Google Patents
A kind of vehicle flowrate prediction technique and device Download PDFInfo
- Publication number
- CN109300310A CN109300310A CN201811415714.3A CN201811415714A CN109300310A CN 109300310 A CN109300310 A CN 109300310A CN 201811415714 A CN201811415714 A CN 201811415714A CN 109300310 A CN109300310 A CN 109300310A
- Authority
- CN
- China
- Prior art keywords
- prediction model
- vehicle flowrate
- prediction
- training dataset
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention is suitable for forecasting traffic flow technical field, provide a kind of vehicle flowrate prediction technique and device, the described method includes: obtaining multiple history vehicle flowrates with time series, obtain history data set, the history data set is adjusted separately according to N number of default adjustment collection, obtain N number of training dataset for having different distributions mode, N number of prediction model is constructed according to N number of training dataset, obtain current vehicle flow, it chooses from N number of prediction model and is predicted with the matched prediction model of the current vehicle flow, obtain vehicle flowrate predicted value, prediction model is constructed again by being adjusted to history data set, selection is predicted with the matched prediction model of current vehicle flow, so that training dataset is identical as the distribution of current vehicle flow, prediction model can be chosen according to current vehicle flow, so that prediction model with Real time data carry out dynamic adjustment, realize the vehicle flowrate prediction of more high accuracy, recommend more reasonable traffic path in time for driver.
Description
Technical field
The invention belongs to forecasting traffic flow technical field more particularly to a kind of vehicle flowrate prediction technique and devices.
Background technique
With the increase of car ownership and the magnitude of traffic flow, traffic congestion frequently occurs, real-time accurate forecasting traffic flow
It is the key that intellectual traffic control and to dredge, facilitates people and efficiently go on a journey.Vehicle flowrate prediction is divided into for a long time according to time span
Vehicle flowrate predicts that two kinds, especially prediction of short-term traffic volume have sudden and randomness in short-term, is always that domestic and international traffic is special
The hot spot of family and scholar's research.
Vehicle flowrate in short-term is predicted, traditional prediction method is to construct prediction model according to history vehicle flowrate, according to prediction
Model predicts current data.Since road traffic system is time-varying, a Nonstationary Stochastic System, due to weather because
The many reasons such as element, the psychological condition of driver, emergency event and traffic accident, cause vehicle flowrate in short-term have height not really
It is qualitative, cause the distribution of current vehicle flow no longer consistent with the distribution of history vehicle flowrate, if still using the pre- of traditional prediction method
It surveys model to be predicted, it is true to may cause vehicle flowrate forecasting inaccuracy.Simultaneously as real-time traffic condition has sudden, tradition
The prediction model of prediction technique cannot carry out dynamic adjustment with current real time data, will also result in prediction model and be no longer applicable in
Current data, can not Accurate Prediction vehicle flowrate, more reasonable traffic path can not be recommended in time for driver.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of vehicle flowrate prediction technique and device, with solve the prior art without
Method Accurate Prediction vehicle flowrate cannot recommend the problem of more reasonable traffic path in time.
The first aspect of the embodiment of the present invention provides a kind of vehicle flowrate prediction technique, comprising:
Multiple history vehicle flowrates with time series are obtained, history data set is obtained；
The history data set is adjusted separately according to N number of default adjustment collection, obtains N number of instruction for having different distributions mode
Practice data set, wherein N is positive integer；
N number of prediction model is constructed according to N number of training dataset；
Current vehicle flow is obtained, chooses from N number of prediction model and is carried out with the matched prediction model of the current vehicle flow
Prediction, obtains vehicle flowrate predicted value.
In a kind of possible implementation, the element numbers of all default adjustment collection are identical, and are equal to the historical data
The element number of collection；
The distribution pattern is determined by mean value and variance；
The history data set is adjusted separately according to N number of default adjustment collection, obtains N number of instruction for having different distributions mode
Practice data set, comprising:
N number of default adjustment collection is added or is multiplied with the element of the corresponding position of the history data set respectively, is obtained
To N number of training dataset, N number of training dataset has different distributions mode.
In a kind of possible implementation, the acquisition current vehicle flow, chosen from N number of prediction model with it is described current
The matched prediction model of vehicle flowrate is predicted, vehicle flowrate predicted value is obtained, comprising:
Vehicle flowrate was obtained before current time in first time period as test data on line, test data is used on the line
Predict the vehicle flowrate at current time；
The prediction that test data on the line carries out the vehicle flowrate at current time to N number of prediction model is inputted respectively,
Obtain N number of predicted value；
Obtain the monitor value of current time vehicle flowrate；
According to the monitor value, the accuracy rate of N number of predicted value is calculated separately；
The corresponding prediction model of predicted value that selection accuracy rate meets preset condition is matched as with the current vehicle flow
Prediction model；
The vehicle flowrate in second time period after current time is predicted according to the prediction model of selection, obtains vehicle flowrate
Predicted value.
In a kind of possible implementation, the prediction model according to selection is in second time period after current time
Vehicle flowrate is predicted, vehicle flowrate predicted value is obtained, comprising:
Judge the number for the prediction model chosen；
If the number is one, according to selection prediction model to the wagon flow in second time period after current time
Amount is predicted, the predicted value of vehicle flowrate is obtained；
If the number be it is multiple, according to multiple prediction models of selection respectively in second time period after current time
Vehicle flowrate is predicted, multiple intermediate predictors are obtained, and calculates the average value of the multiple intermediate predictor, is determined described average
Value is vehicle flowrate predicted value.
It is described that N number of prediction model is constructed according to N number of training dataset in a kind of possible implementation, comprising:
N number of training dataset is initialized, N number of initialization result collection is obtained；
According to shot and long term memory network LSTM training N number of initialization result collection, N number of prediction model is obtained.
In a kind of possible implementation, initialization N number of training dataset obtains N number of initialization result collection,
Include:
Tranquilization processing is carried out to N number of training dataset；
N number of training dataset by tranquilization processing is standardized, N number of initialization result is obtained
Collection.
In a kind of possible implementation, the tranquilization processing includes: difference processing；
It includes: that will pass through at tranquilization that the described pair of training dataset by tranquilization processing, which is standardized,
The training dataset of reason maps in [- 1,1] range.
The second aspect of the embodiment of the present invention provides a kind of vehicle flowrate prediction meanss, comprising:
Module is obtained, for obtaining multiple history vehicle flowrates with time series, obtains history data set；
Module is adjusted, for adjusting separately the history data set according to N number of default adjustment collection, is obtained N number of with difference
Distribution pattern training dataset, wherein N is positive integer；
Module is constructed, for constructing N number of prediction model according to N number of training dataset；
The acquisition module, is also used to obtain current vehicle flow；
Prediction module carries out in advance for choosing from N number of prediction model with the matched prediction model of the current vehicle flow
It surveys, obtains vehicle flowrate predicted value.
The third aspect of the embodiment of the present invention provides a kind of terminal device, comprising:
Memory, processor and storage are in the memory and the computer journey that can run on the processor
The step of sequence, the processor realizes method as described above when executing the computer program.
The fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, comprising:
The computer-readable recording medium storage has computer program, real when the computer program is executed by processor
The step of existing method described above.
The present invention provides a kind of vehicle flowrate prediction technique and devices, which comprises obtaining has time series
Multiple history vehicle flowrates, obtain history data set, adjust separately the history data set according to N number of default adjustment collection, obtain N number of
Have different distributions the training dataset of mode, wherein N is positive integer, constructs N number of prediction mould according to N number of training dataset
Type obtains current vehicle flow, chooses from N number of prediction model and predicted with the matched prediction model of the current vehicle flow,
Vehicle flowrate predicted value is obtained, constructs prediction model again by being adjusted to history data set, selection is matched with current vehicle flow
Prediction model predicted that, so that training dataset is identical as the distribution of current vehicle flow, prediction model can be according to current
Vehicle flowrate is chosen, so that prediction model carries out dynamic adjustment with real time data, realizes the vehicle of more high accuracy
Volume forecasting recommends more reasonable traffic path for driver in time.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the implementation process schematic diagram for the vehicle flowrate prediction technique that the embodiment of the present invention one provides；
Fig. 2 is the implementation process schematic diagram of vehicle flowrate prediction technique provided by Embodiment 2 of the present invention；
Fig. 3 is the implementation process schematic diagram for the vehicle flowrate prediction technique that the embodiment of the present invention three provides；
Fig. 4 is the schematic diagram for the vehicle flowrate prediction meanss that the embodiment of the present invention four provides；
Fig. 5 is the schematic diagram for the terminal device that the embodiment of the present invention five provides.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Fig. 1 is the implementation process schematic diagram for the vehicle flowrate prediction technique that the embodiment of the present invention one provides, as shown in Figure 1, this
The executing subject of embodiment is vehicle flowrate prediction meanss, and vehicle flowrate prediction technique provided in this embodiment includes:
Step 11 obtains multiple history vehicle flowrates with time series, obtains history data set.
Vehicle flowrate prediction meanss obtain multiple history vehicle flowrates under line with time series as history data set.Example
Such as, obtain the vehicle flowrate of a certain traffic intersection: vehicle flowrate is source1 when 1~2, and vehicle flowrate is source2 ..., n when 2~3
Vehicle flowrate is sourcen when~n+1, obtains history data set [source1, source2 ..., sourcen], wherein n is positive
Integer.
Step 12 adjusts separately the history data set according to N number of default adjustment collection, obtains N number of have different distributions
The training dataset of mode.Wherein, N is positive integer.
Optionally, the element number of all default adjustment collection is identical, and is equal to the element number of the history data set.By
In default adjustment collection for adjusting history data set, therefore, the element number of all default adjustment collection is equal to history data set
Element number.
Optionally, the distribution pattern is determined by mean value and variance.Multiple mean value threshold values and multiple variance threshold values are preset, are pressed
It according to the distribution situation of history data set, is adjusted, obtains multiple training datasets with different distributions mode.Its
In, the number N of distribution pattern is determined by the number of mean value threshold value and the number of variance threshold values.For example, with two mean value threshold values (
One mean value threshold value and the second mean value threshold value) and two variance threshold values (first variance threshold value and second variance threshold value) for, definition
Mean value less than the first mean value threshold value is " mean value is small ", not less than the first mean value threshold value and be " mean value less than the second mean value threshold value
In ", not less than the second mean value threshold value be " mean value is big ", defining variance to be less than first variance threshold value is " variance is small ", not less than the
One variance threshold values and be less than second variance threshold value be " in variance ", be not less than second variance threshold value be " variance is big ", thus combine,
9 kinds of distribution patterns are obtained, it is specific as shown in table 1.
9 kinds of distribution patterns that table 1 is divided according to two mean value threshold values and two variance threshold values
| Mean value is small, variance is small | In mean value, variance it is small | Mean value is big, variance is small |
| Mean value is small, in variance | In mean value, in variance | Mean value is big, in variance |
| Mean value is small, variance is big | In mean value, variance it is big | Mean value is big, variance is big |
Wherein, the first mean value threshold value, the second mean value threshold value, first variance threshold value and second variance threshold value are according to history number
It is preset according to the mean value and variance of collection.
It still is exemplified as example with above-mentioned, the element number of all default adjustment collection of the traffic intersection is n, and the 1st pre-
If adjustment collection is [noise11, noise12 ..., noise1n], the 2nd default adjustment collection for [noise21, noise22 ...,
Noise2n] ..., the default adjustment collection of n-th is [noiseN1, noiseN2 ..., noiseNn].According to the default adjustment collection of i-th
Adjust history data set, obtain i-th training dataset [mergeI1, mergeI2 ..., mergeIn], wherein I be no more than
The positive integer of N.
Optionally, the history data set is adjusted separately according to N number of default adjustment collection, obtains N number of have different distributions
The training dataset of mode, can specifically include: by N number of default adjustment collection respectively with the corresponding position of the history data set
The element set is added, and obtains N number of training dataset, N number of training dataset has different distributions mode.
For example, using the default adjustment collection [noiseI1, noiseI2 ..., noiseIn] of i-th to history data set
[source1, source2 ..., sourcen] is adjusted, obtain the 1st training dataset [mergeI1, mergeI2 ...,
MergeIn], wherein mergeI1=source1+noiseI1, mergeI2=source2+noiseI2 ..., mergeIn=
sourcen+noiseIn.Specifically, assuming that the history data set obtained is [2.3,2.5,2.7,2.5,2.3], a default tune
It is whole collection be [0.1,0.2,0.4,0.2,0.1], the training dataset that the element of the two corresponding position is added be [2.4,
2.7,3.1,2.7,2.4].
Optionally, the history data set is adjusted separately according to N number of default adjustment collection, obtains N number of have different distributions
The training dataset of mode, can specifically include: by N number of default adjustment collection respectively with the corresponding position of the history data set
The element multiplication set, obtains N number of training dataset, and N number of training dataset has different distributions mode.
For example, using the default adjustment collection [noiseI1, noiseI2 ..., noiseIn] of i-th to history data set
[source1, source2 ..., sourcen] is adjusted, obtain the 1st training dataset [mergeI1, mergeI2 ...,
MergeIn], wherein mergeI1=source1*noiseI1, mergeI2=source2*noiseI2 ..., mergeIn=
sourcen*noiseIn.Specifically, assuming that the history data set obtained is [2.3,2.5,2.7,2.5,2.3], a default tune
It is whole collection be [0.1,0.2,0.4,0.2,0.1], the training dataset that the element multiplication of the two corresponding position is obtained be [0.23,
0.5,1.08,0.5,0.23].
Step 13 constructs N number of prediction model according to N number of training dataset.
N number of training dataset that step 12 is obtained is configured to N number of prediction model according to default processing step, for example, will
I-th training dataset [mergeI1, mergeI2 ..., mergeIn] is handled according to default processing step, obtains i-th
Prediction model is denoted as modelI.
Step 14 obtains current vehicle flow.
Step 15 is chosen from N number of prediction model and is predicted with the matched prediction model of the current vehicle flow, obtains
Vehicle flowrate predicted value.
Vehicle flowrate prediction meanss are chosen from N number of prediction model is distributed the prediction mould to match with the current vehicle flow obtained
Type is predicted using the prediction model of selection, obtains the vehicle flowrate predicted value for meeting current vehicle flow distribution.
For example, if the integrated distribution pattern that mean value is small, variance is small of current data, chooses and it in N number of prediction model
The prediction model matched, for construct the prediction model chosen training dataset is also small at mean value, the small distribution pattern of variance, if working as
In preceding data integration mean value, the distribution pattern that variance is big, the prediction model of selection be according in mean value, the distributed mode that variance is big
The prediction model of the training dataset building of formula guarantees that building prediction model is matched with current data set, so that prediction
Vehicle flowrate is more accurate.
Present embodiments provide a kind of vehicle flowrate prediction technique, comprising: obtain multiple history wagon flows with time series
Amount, obtains history data set, adjusts separately the history data set according to N number of default adjustment collection, obtains N number of with different points
The training dataset of cloth mode, wherein N is positive integer, constructs N number of prediction model according to N number of training dataset, front truck is worked as in acquisition
Flow is chosen from N number of prediction model and is predicted with the matched prediction model of the current vehicle flow, and vehicle flowrate prediction is obtained
Value, constructs prediction model by being adjusted to history data set again, chooses and carries out with the matched prediction model of current vehicle flow
Prediction, so that training dataset is identical as the distribution of current vehicle flow, prediction model can be chosen according to current vehicle flow,
So that prediction model carries out dynamic adjustment with real time data, the vehicle flowrate prediction of more high accuracy is realized, to drive
Member recommends more reasonable traffic path in time.
Fig. 2 is the implementation process schematic diagram of vehicle flowrate prediction technique provided by Embodiment 2 of the present invention, as shown in Fig. 2, this
The executing subject of embodiment is vehicle flowrate prediction meanss, and the present embodiment is step 13 shown in Fig. 1: according to N number of training dataset structure
A kind of possible implementation for building N number of prediction model, specifically includes:
Step 21, initialization N number of training dataset, obtain N number of initialization result collection.
Optionally, initialization includes tranquilization processing and standardization.
Since road traffic system is time-varying, a Nonstationary Stochastic System, the vehicle flowrate generally obtained is non-stationary
The data of change, the training dataset adjusted are also non-stationary data, right in order to study the variation tendency of the following vehicle flowrate
The vehicle flowrate data of non-stationary carry out tranquilization processing, then by tranquilization treated vehicle flowrate data normalization, to complete
Initialization will be trained in the vehicle flowrate data input network after initialization, obtain prediction model.
Specifically, carrying out tranquilization processing to N number of training dataset in the present embodiment, to by tranquilization processing
N number of training dataset is standardized, and obtains N number of initialization result collection.The described pair of institute by tranquilization processing
Stating training dataset and being standardized includes: that will map to [- 1,1] model by the training dataset of tranquilization processing
In enclosing.
Wherein, tranquilization processing include difference processing, be known as first-order difference as first difference, when using first-order difference also not
When training data can be made to be integrated into stationary time series, higher difference also can be used, to become stationary time series.
Step 22 trains N number of initialization result collection according to shot and long term memory network LSTM, obtains N number of prediction model.
By taking middle layer is the LSTM network of 100 dimensions as an example:
Input layer: N number of initialization result collection is inputted into LSTM network respectively first；
Into 1 layer of LSTM: the data of current input node are 1 dimension, and the data of current output node are 100 dimensions；
Into 2 layers of LSTM: the data of current input node are 100 dimensions, and the data of current output node are 100 dimensions；
Into 3 layers of LSTM: the data of current input node are 100 dimensions, and the data of current output node are 1 dimension；
Output layer: the predicted value of next time is exported.
Optionally, dropout layers are used in the present embodiment, in the training process of deep learning network, for nerve net
Network unit temporarily abandons it according to certain probability from network, to improve the speed of LSTM network, while can be to avoid instruction
Over-fitting occurs during practicing.Preferably, it is 20% that dropout layers, which temporarily abandon ratio, in the present embodiment.
A kind of vehicle flowrate prediction technique provided in this embodiment, initializes N number of training dataset, according to LSTM net
The N number of initialization result collection of network training, obtains N number of prediction model, realizes the training dataset building of multiple and different distribution patterns
Multiple and different prediction models adapts to the vehicle flowrate number of a variety of different distributions it was predicted that choosing matched pre- with current vehicle flow
Survey model predicted so that prediction model with real time data carry out dynamic adjustment, compared with the existing technology in only with one
Kind prediction model carries out vehicle flowrate prediction, improves the accuracy of vehicle flowrate prediction, can recommend in time for driver more rationally
Traffic path.
Fig. 3 is the implementation process schematic diagram for the vehicle flowrate prediction technique that the embodiment of the present invention three provides, as shown in figure 3, this
The executing subject of embodiment is vehicle flowrate prediction meanss, and the present embodiment is that one kind of step 14 shown in Fig. 1 and step 15 is possible
Implementation specifically includes:
Step 31 obtained before current time in first time period vehicle flowrate as test data on line, tested on the line
Data are used to predict the vehicle flowrate at current time.
In the present embodiment, the data of acquisition current time first time period first are tested.For example, obtaining current time
Vehicle flowrate in nearest 1 hour is as test data on line.
Step 32 inputs the vehicle flowrate that test data on the line carries out current time to N number of prediction model respectively
Prediction, obtain N number of predicted value.
Test data on line is inputted into N number of prediction model, obtains the current time vehicle flowrate predicted value of N number of prediction.
Step 33, the monitor value for obtaining current time vehicle flowrate.
Vehicle flowrate prediction meanss obtain the monitor value of current time vehicle flowrate again, which is the actual measurement of vehicle flowrate
Value.
Step 34, according to the monitor value, calculate separately the accuracy rate of N number of predicted value.
Contrastive detection value and N number of predicted value, according toThe accurate of i-th predicted value is calculated
Rate, wherein y is monitor value, and x is i-th predicted value.A number of the accuracy rate rate between [0,1], rate are said closer to 1
Bright accuracy rate is higher, at this point, the x of prediction is closer to monitor value y.
Step 35, choose accuracy rate meet the corresponding prediction model of predicted value of preset condition as with the current wagon flow
Flux matched prediction model.
The corresponding prediction model of predicted value that accuracy rate meets preset condition is determined as prediction model.Preset condition is preparatory
Setting is more than or equal to 0.9 as accuracy rate for example, setting preset condition.
Step 36 predicts the vehicle flowrate in second time period after current time according to the prediction model of selection, obtains
To vehicle flowrate predicted value.
Wherein, the vehicle flowrate in second time period after current time is predicted according to the prediction model of selection, is obtained
Vehicle flowrate predicted value specifically includes, and judges the number for the prediction model chosen；If the number is one, according to the one of selection
Prediction model predicts the vehicle flowrate in second time period after current time, obtains the predicted value of vehicle flowrate；If described
It is multiple for counting, and is predicted respectively the vehicle flowrate in second time period after current time according to multiple prediction models of selection,
Multiple intermediate predictors are obtained, the average value of the multiple intermediate predictor is calculated, determine the average value for vehicle flowrate prediction
Value.If then using the prediction model specifically, meeting prediction model of the preset condition accuracy rate more than or equal to 0.9 only has 1
Predicted value of the predicted value of prediction as vehicle flowrate has 3 when meeting prediction model of the preset condition accuracy rate more than or equal to 0.9
When, then predicted value of the average value for 3 predicted values predicted using 3 prediction models as vehicle flowrate.Optionally, when meeting
When the prediction model of preset condition has multiple, weight can be matched for each calculating value distribution according to order of accuarcy, accuracy is high,
Weight is big, and accuracy is low, and weight is small, then calculates average value again, so that the predicted value accuracy calculated is higher, it can be to drive
The person of sailing recommends more reasonable traffic path in time.
Fig. 4 is the schematic diagram for the vehicle flowrate prediction meanss that the embodiment of the present invention four provides, as shown in figure 4, the embodiment
Vehicle flowrate prediction meanss include:
Module 41 is obtained, for obtaining multiple history vehicle flowrates with time series, obtains history data set.
Module 42 is adjusted, for adjusting separately the history data set according to N number of default adjustment collection, is obtained N number of with not
The training dataset of same distribution pattern, wherein N is positive integer.
Module 43 is constructed, for constructing N number of prediction model according to N number of training dataset.
The acquisition module 41, is also used to obtain current vehicle flow.
Prediction module 44 is carried out for choosing from N number of prediction model with the matched prediction model of the current vehicle flow
Prediction, obtains vehicle flowrate predicted value.
A kind of vehicle flowrate prediction meanss provided in this embodiment, for realizing vehicle flowrate prediction side described in embodiment one
Method, wherein the function of modules can be referred to and be described accordingly in embodiment of the method, and it is similar that the realization principle and technical effect are similar,
Details are not described herein again.
Fig. 5 is the schematic diagram for the terminal device that the embodiment of the present invention five provides, as shown in figure 5, the terminal of the embodiment is set
Standby 5 include: processor 50, memory 51 and are stored in the meter that can be run in the memory 51 and on the processor 50
Calculation machine program 52, such as vehicle flowrate Prediction program.The processor 50 is realized above-mentioned each when executing the computer program 52
Step in vehicle flowrate prediction technique embodiment, such as step 11 shown in FIG. 1 is to 15.Alternatively, the processor 50 executes institute
The function of each module in above-mentioned each vehicle flowrate prediction meanss embodiment, such as mould shown in Fig. 4 are realized when stating computer program 52
The function of block 41 to 44.
Illustratively, the computer program 52 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 51, and are executed by the processor 50, to complete the present invention.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program 52 in the terminal device 5 is described.For example, the computer program 52 can be divided
It is cut into and obtains module, adjustment module, building module and prediction module (unit module in virtual bench), each module concrete function
It is as follows:
Module is obtained, for obtaining multiple history vehicle flowrates with time series, obtains history data set；
Module is adjusted, for adjusting separately the history data set according to N number of default adjustment collection, is obtained N number of with difference
Distribution pattern training dataset, wherein N is positive integer；
Module is constructed, for constructing N number of prediction model according to N number of training dataset；
The acquisition module, is also used to obtain current vehicle flow；
Prediction module carries out in advance for choosing from N number of prediction model with the matched prediction model of the current vehicle flow
It surveys, obtains vehicle flowrate predicted value.
The terminal device 5 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The terminal device 5 may include, but be not limited only to, processor 50, memory 51.It will be understood by those skilled in the art that figure
5 be only the example of terminal device 5, does not constitute the restriction to terminal device 5, may include than illustrating more or fewer portions
Part perhaps combines certain components or different components, such as the terminal device 5 can also include input-output equipment, net
Network access device, bus etc..
Alleged processor 50 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 51 can be the internal storage unit of the terminal device 5, such as the hard disk or interior of terminal device 5
It deposits.The memory 51 is also possible to the External memory equipment of the terminal device 5, such as be equipped on the terminal device 5
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card dodge
Deposit card (Flash Card) etc..Further, the memory 51 can also both include the storage inside list of the terminal device 5
Member also includes External memory equipment.The memory 51 is for storing needed for the computer program and the terminal device 5
Other programs and data.The memory 51 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of the terminal device is divided into different functional unit or module, to complete
All or part of function described above.Each functional unit in embodiment, module can integrate in one processing unit,
It is also possible to each unit to physically exist alone, can also be integrated in one unit with two or more units, above-mentioned collection
At unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function
Unit, module specific name be also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.Above-mentioned system
The specific work process of unit in system, module, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed device/terminal device and method, it can be with
It realizes by another way.For example, device described above/terminal device embodiment is only schematical, for example, institute
The division of module or unit is stated, only a kind of logical function partition, there may be another division manner in actual implementation, such as
Multiple units or components can be combined or can be integrated into another system, or some features can be ignored or not executed.Separately
A bit, shown or discussed mutual coupling or direct-coupling or communication connection can be through some interfaces, device
Or the INDIRECT COUPLING or communication connection of unit, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated module/unit be realized in the form of SFU software functional unit and as independent product sale or
In use, can store in a computer readable storage medium.Based on this understanding, the present invention realizes above-mentioned implementation
All or part of the process in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program generation
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..The computer-readable medium
It may include: any entity or device, recording medium, USB flash disk, mobile hard disk, magnetic that can carry the computer program code
Dish, CD, computer storage, read-only memory (Read-Only Memory, ROM), random access memory (Random
Access Memory, RAM), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that the meter
The content that calculation machine readable medium includes can carry out increase and decrease appropriate according to the requirement made laws in jurisdiction with patent practice,
It such as does not include electric carrier signal and telecommunications according to legislation and patent practice, computer-readable medium in certain jurisdictions
Signal.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of vehicle flowrate prediction technique characterized by comprising
Multiple history vehicle flowrates with time series are obtained, history data set is obtained；
The history data set is adjusted separately according to N number of default adjustment collection, obtains N number of training number for having different distributions mode
According to collection, wherein N is positive integer；
N number of prediction model is constructed according to N number of training dataset；
Current vehicle flow is obtained, chooses from N number of prediction model and is predicted with the matched prediction model of the current vehicle flow,
Obtain vehicle flowrate predicted value.
2. the method as described in claim 1, which is characterized in that the element number of all default adjustment collection is identical, and is equal to institute
State the element number of history data set；
The distribution pattern is determined by mean value and variance；
The history data set is adjusted separately according to N number of default adjustment collection, obtains N number of training number for having different distributions mode
According to collection, comprising:
N number of default adjustment collection is added or is multiplied with the element of the corresponding position of the history data set respectively, is obtained N number of
Training dataset, N number of training dataset have different distributions mode.
3. the method as described in claim 1, which is characterized in that the acquisition current vehicle flow is chosen from N number of prediction model
It is predicted with the matched prediction model of the current vehicle flow, obtains vehicle flowrate predicted value, comprising:
Obtained before current time that vehicle flowrate is used as test data on line in first time period, test data is used to predict on the line
The vehicle flowrate at current time；
The prediction that test data on the line carries out the vehicle flowrate at current time to N number of prediction model is inputted respectively, obtains N
A predicted value；
Obtain the monitor value of current time vehicle flowrate；
According to the monitor value, the accuracy rate of N number of predicted value is calculated separately；
It chooses accuracy rate and meets the corresponding prediction model of predicted value of preset condition as matched pre- with the current vehicle flow
Survey model；
The vehicle flowrate in second time period after current time is predicted according to the prediction model of selection, obtains vehicle flowrate prediction
Value.
4. method as claimed in claim 3, which is characterized in that the prediction model according to selection is to after current time second
Vehicle flowrate in period is predicted, vehicle flowrate predicted value is obtained, comprising:
Judge the number for the prediction model chosen；
If the number be one, according to selection prediction model to the vehicle flowrate in second time period after current time into
Row prediction, obtains the predicted value of vehicle flowrate；
If the number be it is multiple, according to multiple prediction models of selection respectively to the wagon flow in second time period after current time
Amount is predicted, is obtained multiple intermediate predictors, is calculated the average value of the multiple intermediate predictor, determines that the average value is
Vehicle flowrate predicted value.
5. the method as described in claim 1, which is characterized in that it is described that N number of prediction model is constructed according to N number of training dataset,
Include:
N number of training dataset is initialized, N number of initialization result collection is obtained；
According to shot and long term memory network LSTM training N number of initialization result collection, N number of prediction model is obtained.
6. method as claimed in claim 5, which is characterized in that initialization N number of training dataset obtains N number of first
Beginningization result set, comprising:
Tranquilization processing is carried out to N number of training dataset；
N number of training dataset by tranquilization processing is standardized, N number of initialization result collection is obtained.
7. method as claimed in claim 6, which is characterized in that the tranquilization processing includes: difference processing；
It includes: that will handle by tranquilization that the described pair of training dataset by tranquilization processing, which is standardized,
The training dataset maps in [- 1,1] range.
8. a kind of vehicle flowrate prediction meanss characterized by comprising
Module is obtained, for obtaining multiple history vehicle flowrates with time series, obtains history data set；
Module is adjusted, for adjusting separately the history data set according to N number of default adjustment collection, is obtained N number of with different points
The training dataset of cloth mode, wherein N is positive integer；
Module is constructed, for constructing N number of prediction model according to N number of training dataset；
The acquisition module, is also used to obtain current vehicle flow；
Prediction module is predicted with the matched prediction model of the current vehicle flow for choosing from N number of prediction model, is obtained
To vehicle flowrate predicted value.
9. a kind of terminal device, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 7 when executing the computer program
The step of any one the method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In when the computer program is executed by processor the step of any one of such as claim 1 to 7 of realization the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811415714.3A CN109300310B (en) | 2018-11-26 | 2018-11-26 | Traffic flow prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811415714.3A CN109300310B (en) | 2018-11-26 | 2018-11-26 | Traffic flow prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109300310A true CN109300310A (en) | 2019-02-01 |
| CN109300310B CN109300310B (en) | 2021-09-17 |
Family
ID=65143432
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811415714.3A Active CN109300310B (en) | 2018-11-26 | 2018-11-26 | Traffic flow prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109300310B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109859469A (en) * | 2019-02-15 | 2019-06-07 | 重庆邮电大学 | A kind of vehicle flowrate prediction technique based on integrated LSTM neural network |
| CN110009429A (en) * | 2019-04-10 | 2019-07-12 | 金瓜子科技发展（北京）有限公司 | A kind of method, apparatus and computer equipment of predicted flow rate data |
| CN110222739A (en) * | 2019-05-23 | 2019-09-10 | 中国银行股份有限公司 | Model construction, site cash Demand Forecast method, apparatus and storage medium |
| CN111091708A (en) * | 2019-12-13 | 2020-05-01 | 中国科学院深圳先进技术研究院 | Vehicle track prediction method and device |
| CN111401609A (en) * | 2020-03-04 | 2020-07-10 | 平安科技（深圳）有限公司 | Prediction method and prediction device for traffic flow time series |
| CN111679952A (en) * | 2020-06-08 | 2020-09-18 | 中国银行股份有限公司 | Alarm threshold generation method and device |
| CN111833594A (en) * | 2019-04-23 | 2020-10-27 | 杭州海康威视系统技术有限公司 | Traffic flow prediction method, traffic flow prediction device, electronic device, and storage medium |
| CN112149901A (en) * | 2020-09-23 | 2020-12-29 | 华北电力科学研究院有限责任公司 | Method and device for predicting low-pressure final-stage pressure field of steam turbine |
| WO2021012342A1 (en) * | 2019-07-23 | 2021-01-28 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for traffic prediction |
| CN112652172A (en) * | 2021-01-19 | 2021-04-13 | 东南大学 | Road section traffic volume analysis method based on vehicle GPS track |
| CN112884190A (en) * | 2019-11-29 | 2021-06-01 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
| CN113222266A (en) * | 2021-05-22 | 2021-08-06 | 深圳市创仁科技有限公司 | Prediction model matching method based on large daily water meter data |
| WO2021169174A1 (en) * | 2020-02-29 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Road congestion degree prediction method, apparatus, computer device, and readable storage medium |
| CN114596702A (en) * | 2021-11-12 | 2022-06-07 | 南方科技大学 | Traffic state prediction model construction method and traffic state prediction method |
| CN114999182A (en) * | 2022-05-25 | 2022-09-02 | 中国人民解放军国防科技大学 | Vehicle flow prediction method, device and equipment based on LSTM feedback mechanism |
| CN115311842A (en) * | 2021-05-07 | 2022-11-08 | 杭州海康威视数字技术股份有限公司 | Traffic flow prediction model training and traffic flow prediction method, device and electronic equipment |
| CN117475633A (en) * | 2023-11-07 | 2024-01-30 | 西南交通大学 | Event-oriented traffic flow prediction method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120023043A1 (en) * | 2010-07-21 | 2012-01-26 | Ozgur Cetin | Estimating Probabilities of Events in Sponsored Search Using Adaptive Models |
| CN103646534A (en) * | 2013-11-22 | 2014-03-19 | 江苏大学 | A road real time traffic accident risk control method |
| CN105701571A (en) * | 2016-01-13 | 2016-06-22 | 南京邮电大学 | Short-term traffic flow prediction method based on nerve network combination model |
| CN105788249A (en) * | 2014-12-16 | 2016-07-20 | 高德软件有限公司 | Traffic flow prediction method, prediction model generation method and device |
| CN106033425A (en) * | 2015-03-11 | 2016-10-19 | 富士通株式会社 | A data processing device and a data processing method |
| CN108229724A (en) * | 2017-12-06 | 2018-06-29 | 华南理工大学 | A kind of transport data stream Forecasting Methodology in short-term based on Spatial-temporal Information Fusion |
-
2018
- 2018-11-26 CN CN201811415714.3A patent/CN109300310B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120023043A1 (en) * | 2010-07-21 | 2012-01-26 | Ozgur Cetin | Estimating Probabilities of Events in Sponsored Search Using Adaptive Models |
| CN103646534A (en) * | 2013-11-22 | 2014-03-19 | 江苏大学 | A road real time traffic accident risk control method |
| CN105788249A (en) * | 2014-12-16 | 2016-07-20 | 高德软件有限公司 | Traffic flow prediction method, prediction model generation method and device |
| CN106033425A (en) * | 2015-03-11 | 2016-10-19 | 富士通株式会社 | A data processing device and a data processing method |
| CN105701571A (en) * | 2016-01-13 | 2016-06-22 | 南京邮电大学 | Short-term traffic flow prediction method based on nerve network combination model |
| CN108229724A (en) * | 2017-12-06 | 2018-06-29 | 华南理工大学 | A kind of transport data stream Forecasting Methodology in short-term based on Spatial-temporal Information Fusion |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109859469A (en) * | 2019-02-15 | 2019-06-07 | 重庆邮电大学 | A kind of vehicle flowrate prediction technique based on integrated LSTM neural network |
| CN110009429A (en) * | 2019-04-10 | 2019-07-12 | 金瓜子科技发展（北京）有限公司 | A kind of method, apparatus and computer equipment of predicted flow rate data |
| CN111833594A (en) * | 2019-04-23 | 2020-10-27 | 杭州海康威视系统技术有限公司 | Traffic flow prediction method, traffic flow prediction device, electronic device, and storage medium |
| CN110222739A (en) * | 2019-05-23 | 2019-09-10 | 中国银行股份有限公司 | Model construction, site cash Demand Forecast method, apparatus and storage medium |
| WO2021012342A1 (en) * | 2019-07-23 | 2021-01-28 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for traffic prediction |
| CN112884190A (en) * | 2019-11-29 | 2021-06-01 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
| CN112884190B (en) * | 2019-11-29 | 2023-11-03 | 杭州海康威视数字技术股份有限公司 | Flow prediction method and device |
| CN111091708B (en) * | 2019-12-13 | 2020-11-03 | 中国科学院深圳先进技术研究院 | Vehicle track prediction method and device |
| CN111091708A (en) * | 2019-12-13 | 2020-05-01 | 中国科学院深圳先进技术研究院 | Vehicle track prediction method and device |
| WO2021169174A1 (en) * | 2020-02-29 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Road congestion degree prediction method, apparatus, computer device, and readable storage medium |
| CN111401609A (en) * | 2020-03-04 | 2020-07-10 | 平安科技（深圳）有限公司 | Prediction method and prediction device for traffic flow time series |
| WO2021174811A1 (en) * | 2020-03-04 | 2021-09-10 | 平安科技（深圳）有限公司 | Prediction method and prediction apparatus for traffic flow time series |
| CN111679952A (en) * | 2020-06-08 | 2020-09-18 | 中国银行股份有限公司 | Alarm threshold generation method and device |
| CN111679952B (en) * | 2020-06-08 | 2023-09-19 | 中国银行股份有限公司 | Alarm threshold generation method and device |
| CN112149901A (en) * | 2020-09-23 | 2020-12-29 | 华北电力科学研究院有限责任公司 | Method and device for predicting low-pressure final-stage pressure field of steam turbine |
| CN112149901B (en) * | 2020-09-23 | 2024-04-16 | 华北电力科学研究院有限责任公司 | Steam turbine low-pressure final-stage pressure field prediction method and device |
| CN112652172B (en) * | 2021-01-19 | 2022-01-25 | 东南大学 | Road section traffic volume analysis method based on vehicle GPS track |
| CN112652172A (en) * | 2021-01-19 | 2021-04-13 | 东南大学 | Road section traffic volume analysis method based on vehicle GPS track |
| CN115311842A (en) * | 2021-05-07 | 2022-11-08 | 杭州海康威视数字技术股份有限公司 | Traffic flow prediction model training and traffic flow prediction method, device and electronic equipment |
| CN113222266A (en) * | 2021-05-22 | 2021-08-06 | 深圳市创仁科技有限公司 | Prediction model matching method based on large daily water meter data |
| CN114596702A (en) * | 2021-11-12 | 2022-06-07 | 南方科技大学 | Traffic state prediction model construction method and traffic state prediction method |
| CN114999182A (en) * | 2022-05-25 | 2022-09-02 | 中国人民解放军国防科技大学 | Vehicle flow prediction method, device and equipment based on LSTM feedback mechanism |
| CN117475633A (en) * | 2023-11-07 | 2024-01-30 | 西南交通大学 | Event-oriented traffic flow prediction method and device |
| CN117475633B (en) * | 2023-11-07 | 2024-05-28 | 西南交通大学 | Event-oriented traffic flow prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109300310B (en) | 2021-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109300310A (en) | A kind of vehicle flowrate prediction technique and device | |
| CN102110365B (en) | Road condition prediction method and road condition prediction system based on space-time relationship | |
| CN111292525A (en) | Traffic flow prediction method based on neural network | |
| CN109034505A (en) | Metering table demand on-line prediction method and terminal device | |
| CN108052387B (en) | Resource allocation prediction method and system in mobile cloud computing | |
| CN106781489A (en) | A kind of road network trend prediction method based on recurrent neural network | |
| CN104239687B (en) | Reliability modeling and evaluation method based on aerospace product signal transmission path | |
| CN103514566A (en) | Risk control system and method | |
| CN102693633A (en) | Short-term traffic flow weighted combination prediction method | |
| CN106022614A (en) | Data mining method of neural network based on nearest neighbor clustering | |
| CN101739819A (en) | Method and device for predicting traffic flow | |
| CN110232584A (en) | Parking lot site selecting method, device, computer readable storage medium and terminal device | |
| CN104217258A (en) | Method for power load condition density prediction | |
| CN106991534A (en) | Engineering Project Construction process appraisal procedure | |
| CN103440377A (en) | Aircraft aerodynamic configuration optimum design method based on improved parallel DE algorithm | |
| CN110020739A (en) | For the method, apparatus of data processing, electronic equipment and computer-readable medium | |
| CN107945534A (en) | A kind of special bus method for predicting based on GMDH neutral nets | |
| CN108111353A (en) | Prepaid card residual flow Forecasting Methodology, the network terminal and storage medium | |
| CN110705080A (en) | Precise assembling method and device for digital twin model | |
| CN112613542A (en) | Bidirectional LSTM-based enterprise decontamination equipment load identification method | |
| CN103020433A (en) | Evaluation model engine of electric equipment condition | |
| CN105913654B (en) | A kind of Intelligent traffic management systems | |
| CN103208036B (en) | A kind of short-term load forecasting method of electrically-based user data | |
| Couckuyt et al. | Towards efficient multiobjective optimization: multiobjective statistical criterions | |
| CN110231593A (en) | Indoor orientation method, device, computer readable storage medium and terminal device |
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 |