CN107591001A - Expressway Traffic Flow data filling method and system based on on-line proving - Google Patents
Expressway Traffic Flow data filling method and system based on on-line proving Download PDFInfo
- Publication number
- CN107591001A CN107591001A CN201710802197.4A CN201710802197A CN107591001A CN 107591001 A CN107591001 A CN 107591001A CN 201710802197 A CN201710802197 A CN 201710802197A CN 107591001 A CN107591001 A CN 107591001A
- Authority
- CN
- China
- Prior art keywords
- data
- traffic flow
- flow data
- model
- detector
- 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
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of Expressway Traffic Flow missing data complementing method based on on-line proving and system, methods described, including data acquisition：To detector packet numbering, the traffic flow data collected according to the detector establishes traffic flow data storehouse；Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake；Model is established：According to the time trend and spatial relationship of each track data of each group detector, multivariate linear model is established；On-line proving：Parameter Self-regression Forecast Model is established, the model parameter based on model parameter prediction current time obtained by the Primary Stage Data in traffic flow data storehouse；Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.The present invention has the advantages that calculating is simple, memory space is small, prediction result is accurate, effectively compensate for deficiency of the static models method in terms of data filling.
Description
Technical field
The invention belongs to intelligent traffic control system field, more particularly to a kind of Expressway Traffic Flow based on on-line proving
Data filling method and system.
Background technology
In recent years, there is the traffic congestion of getting worse because transport need is continuously increased in city expressway.In order to slow
Congestion is solved, the intelligent transportation system for through street control has been applied in practice.Intelligent traffic control system utilizes traffic
Current sensor collects data, carries out traffic behavior estimation or prediction, and optimizes and propose control decision.Quick path control system exists
Real time traffic data is largely dependent upon, yet with hardware fault or communication failure, the data of real-time collecting are easy to
Run into loss of data or it is wrong the problems such as.Effective operation of traffic control system needs accurate complete data, and traffic data exists
Being lost in the very short time can cause traffic control system pause to use, and temporary transient stopping will cause systematic function unreliable, and
It may endanger safe traffic.Therefore, before application traffic control strategy, it is necessary to be designed to solve shortage of data or mistake is asked
The complementing method of topic, to improve the quality of data of traffic control system.Pin is on the other hand, The present invention gives a kind of traffic flow missing number
According to complementing method.
Traffic detector in city expressway can all collect substantial amounts of data daily, due to memory space and processing time
Limited, real-time road control needs practicality, accurately and efficiently data filling method.Existing research can application on site
Data filling method.However, these online complementing methods only account for one kind of time and spatial relationship：Missing data is changed to
Historical data average value is to reflect the diurnal variation trend of flow in the same time；Or filled up using the spatial coherence with periphery detector
Missing data is to reflect flow real-time fluctuations.However, only consider that time relationship can not reflect the real-time feature of daily traffic flow, and
Only consider that spatial coherence be able to not can still provide in the case of whole periphery detector failures and fill up result.It there is no and fill out online
Compensating method repairs missing data in combination with time trend and spatial coherence.In addition, conventional data filling method should
With static models parameter, it is possible to the mistakenly change of predicting traffic flow over time and space.How the time is being taken into account
With accurately and reliably data filling is realized on the basis of spatial relationship and in the case of a variety of shortage of data, be current this area skill
The technical problem that art personnel urgently solve.
The content of the invention
To overcome above-mentioned the deficiencies in the prior art, the invention provides a kind of Expressway Traffic Flow based on on-line proving to lack
Data filling method and system, the data that this method is detected in real time based on data detector, diagnostic data missing or mistake are lost, is built
Vertical multivariate linear model, by Primary Stage Data on-line proving and the model parameter of prediction current point in time, and pass through multiple linear
Model fills up missing data.The present invention has the advantages that calculating is simple, memory space is small, prediction result is accurate.
To achieve the above object, the present invention adopts the following technical scheme that：
A kind of Expressway Traffic Flow data filling method based on on-line proving, including：
Data acquisition：To detector packet numbering, traffic flow is established according to the traffic flow data that the detector collects
Database；
Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake
By mistake；
Model is established：According to the time trend and spatial relationship of each track data of each group detector, multiple linear mould is established
Type；
On-line proving：Parameter Self-regression Forecast Model is established, based on model obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at parameter prediction current time；
Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
The traffic flow data storehouse includes the Primary Stage Data of traffic flow real time data, historical data average value and each moment,
The data gathered before the Primary Stage Data at each moment refers to the moment in specified time step-length.
The diagnosis is based on traffic flow data threshold value set in advance, acceptable area and the corresponding upstream of diagnostic data
The diagnostic module that track data are carried out.
If certain moment traffic flow data of collection exceeds threshold value or acceptable area, diagnosing the moment traffic flow data is
Wrong data；If certain moment traffic flow data of collection is zero, and its corresponding upstream track data non-zero, then the moment is diagnosed
For missing data.
It is each detector acquisition data integrity index by real-time diagnosis：
M is detector packet numbering, and n is that every lane detector is numbered.
The model is established and is based on time trend and spatial coherence；The time trend passes through historical data in the same time
Average value reflects, the spatial coherence embodies the estimate that the representative obtained by correlation between the detectors is fluctuated immediately.
The modelling phase establishes multivariate linear model：
Wherein, xestTraffic flow data vector is represented, α, β and γ are model parameter, and α is used to embody traffic flow data space
Correlation；β is used to embody traffic flow data and the correlation of time average tendency；
β=diag (β1,β2,...,βN)
Δ=diag (△1,△2,...,△N)
X=(x1,...,xN)T
γ=(γ0,...,γN)TFor constant term；αijRepresent the car obtained when filling up track i according to data space correlation
Coefficients of the road j in multivariate linear model；βiTrack i time average tendency is in multiple linear mould during to fill up track i data
Coefficient in type；Δ is data integrity degree matrix；Ω=diag (1 ..., 1)；X is the flow or density that detector detects；For the average value of historical data in the same time；ε=(ε0,...,εN)TFor error term.
The on-line proving specifically includes：According to the multivariate linear model of foundation, before in preceding several time steps
Issue evidence, model parameter is optimized with least square method, makes error term close to zero；By repeated optimization process, obtain
To model parameter corresponding with Primary Stage Data；Model parameter based on model parameter prediction current time corresponding to Primary Stage Data.
According to the second object of the present invention, present invention also offers a kind of Expressway Traffic flow data based on on-line proving
System, including processor and computer-readable recording medium are filled up, processor is used to realize each instruction；Computer-readable storage medium
Matter is used to store a plurality of instruction, and the instruction is suitable to be loaded by processor and perform following processing：
Data acquisition：To detector packet numbering, traffic flow is established according to the traffic flow data that the detector collects
Database；
Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake
By mistake；
Model is established：According to the time trend and spatial relationship of each track data of each group detector, multiple linear mould is established
Type；
On-line proving：Parameter Self-regression Forecast Model is established, based on model obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at parameter prediction current time；
Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
According to the third object of the present invention, present invention also offers a kind of computer-readable recording medium, it is stored thereon with
Computer program, for Expressway Traffic Flow data filling, the program performs following steps when being executed by processor：
Data acquisition：To detector packet numbering, traffic flow is established according to the traffic flow data that the detector collects
Database；
Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake
By mistake；
Model is established：According to the time trend and spatial relationship of each track data of each group detector, multiple linear mould is established
Type；
On-line proving：Parameter Self-regression Forecast Model is established, based on model obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at parameter prediction current time；
Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
Compared with prior art, the beneficial effects of the invention are as follows：
The present invention is set time based on measurement threshold value, acceptable area and corresponding upstream data and diagnosed in fact, is distinguished
Wrong data and missing data；Model is joined according to preceding several data collection cycle data application least square methods of closing on online again
Number is demarcated；And based on the model parameter at calibration result prediction current time；Traffic flow data time trend and sky are taken into account
Between correlation data are filled up, data filling of the detector data temporarily under whole deletion conditions, present invention tool can be achieved
Have the advantages that method is accurate, calculating is simple, memory space requirement is small.
Brief description of the drawings
The Figure of description for forming the part of the application is used for providing further understanding of the present application, and the application's shows
Meaning property embodiment and its illustrate be used for explain the application, do not form the improper restriction to the application.
Fig. 1 is a kind of Expressway Traffic Flow data filling flow chart based on on-line proving of the present invention.
Fig. 2 is the present invention effect that each track density data is filled up when lacking 20% data.
Embodiment
It is noted that described further below is all exemplary, it is intended to provides further instruction to the application.It is unless another
Indicate, all technologies used herein and scientific terminology are with usual with the application person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singulative
It is also intended to include plural form, additionally, it should be understood that, when in this manual using term "comprising" and/or " bag
Include " when, it indicates existing characteristics, step, operation, device, component and/or combinations thereof.
In the case where not conflicting, the feature in embodiment and embodiment in the application can be mutually combined.
Embodiment one
Fig. 1 is a kind of Expressway Traffic Flow data filling flow chart based on on-line proving of the present invention.
Present embodiments provide a kind of Expressway Traffic Flow data filling method based on on-line proving, including following step
Suddenly：
Step 1：Establish traffic flow data storehouse
To detector packet numbering, the traffic flow data collected according to the detector establishes traffic flow data storehouse.
By carrying out packet numbering m=1 ..., M by upstream and downstream order to detector, to every lane detector numbering n
=1 ..., N, traffic flow data storehouse is established, the database includes real time data, historical data average value and Primary Stage Data, its
Described in Primary Stage Data refer to current point in time before the data that are gathered in specified time step-length.
Step 2：Real-time diagnosis
The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake.
The diagnosis is based on measurement threshold value, acceptable area and corresponding upstream data, wherein the upstream data is
Refer to and now gather the data that the traffic flow upstream of position correspondence is gathered.Predefine the threshold value of traffic flow data and acceptable model
Enclose, once the traffic flow data of real-time collecting exceeds threshold value or acceptable area, then it is mistake to diagnose the moment traffic flow data.
Also, if detector output flow value is zero, but its corresponding upstream track has non-zero delivery, then the zero delivery is diagnosed
For missing data.Diagnosis algorithm obtains data integrity index for each track of each group detector
At the track of Δ=0, on-line proving and data filling module are excited.
Step 3：Model is established
According to the time trend and spatial relationship of each track data of each group detector, multivariate linear model is established.
The present invention proposes a kind of data filling method based on multivariate linear model, while when considering traffic flow data
Between trend and spatial coherence.The average value of historical data reflects traffic flow change general trend in time in the same time, and
The estimate obtained by correlation between detector then represents instant fluctuation.The present invention is by time trend and the instant ripple in space
It is dynamic to be combined, so as to obtain more accurate data filling result.Traffic flow data vector xestWith other lane detection data x and
History average in the same timeRelation be shown below：
Wherein, α, β and γ are model parameter, will determine (δ=(α, beta, gamma)) in the on-line proving stage；α can embody traffic
Flow data spatial coherence, as shown in formula (2), such as αijRepresent the car obtained when filling up track i according to data space correlation
Coefficients of the road j in multivariate linear model；β can embody traffic flow data and the correlation of time average tendency, such as formula (3) institute
Show, such as βiCoefficient of the track i time average tendency in multivariate linear model during to fill up track i data；γ=
(γ0,...,γN)TFor constant term；Δ is the data integrity degree matrix of every group of detector, as shown in formula (4)；Ω=diag
(1,...,1)；X is the flow or density that detector detects, as shown in formula (5)；For the average value of historical data in the same time
(as shown in formula (6))；ε=(ε0,...,εN)TFor error term.
β=diag (β1,β2,...,βN) (3)
Δ=diag (△1,△2,...,△N) (4)
X=(x1,...,xN)T (5)
Wherein, x1,...,xNWithFor the traffic measured value and history number in the same time in detector track 1 to track N
According to average value.
Step 4：On-line proving
Parameter Self-regression Forecast Model is established, is worked as based on model parameter prediction obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at preceding moment.
On-line proving is divided into two parts.First, the parameter calibration stage uses Primary Stage Data to model (formula (1)) parameter (δ
=(α, beta, gamma)) with least square method demarcation is optimized to model parameter, make error term ε close to 0.Then, weight is passed through
Multiple optimization process, obtain to model parameter corresponding to Primary Stage Data difference in preceding several time steps, and apply autoregressive prediction
The model parameter at current timeIt is shown below：
Wherein, w-auto-regressive parameter；ε-stochastic error.
Step 5：Data filling
Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
The model parameter drawn according to step 4Data are carried out using the multivariate linear model of step 3 to fill out
Mend；Storehouse will be updated the data in case subsequent fills up use by filling up result.The traffic flow data vector x that need to be filled upestCount as the following formula
Calculate, the combination of time trend and spatial coherence ensure that the algorithm can be estimated in the case where all track data lack
Missing values.
Embodiment two
The purpose of the present embodiment is to provide a kind of Expressway Traffic Flow data filling system based on on-line proving.
To achieve these goals, the present invention is using a kind of following technical scheme：
A kind of Expressway Traffic Flow data filling system based on on-line proving, including processor and computer-readable storage
Medium, processor are used to realize each instruction；Computer-readable recording medium is used to store a plurality of instruction, the instruction be suitable to by
Reason device loads and performs following processing：
Data acquisition：To detector packet numbering, traffic flow is established according to the traffic flow data that the detector collects
Database；
Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake
By mistake；
Model is established：According to the time trend and spatial relationship of each track data of each group detector, multiple linear mould is established
Type；
On-line proving：Parameter Self-regression Forecast Model is established, based on model obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at parameter prediction current time；
Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
Embodiment three
The purpose of the present embodiment is to provide a kind of computer-readable recording medium.
To achieve these goals, the present invention is using a kind of following technical scheme：
A kind of computer-readable recording medium, is stored thereon with computer program, for Expressway Traffic Flow data filling,
The program performs following steps when being executed by processor：
Data acquisition：To detector packet numbering, traffic flow is established according to the traffic flow data that the detector collects
Database；
Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake
By mistake；
Model is established：According to the time trend and spatial relationship of each track data of each group detector, multiple linear mould is established
Type；
On-line proving：Parameter Self-regression Forecast Model is established, based on model obtained by the Primary Stage Data in traffic flow data storehouse
The model parameter at parameter prediction current time；
Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
In above example two and embodiment three, each step is corresponding with embodiment of the method one, and embodiment can join
See the related description part of embodiment one.Term " computer-readable recording medium " is construed as including one or more instructions
The single medium or multiple media of collection；Should also be understood as including any medium, any medium can store, encode or
Carry for the instruction set by computing device and make the either method in the computing device present invention.
Experimental result：
Now by taking certain urban expressway traffic current sensor -2016 annual datas in 2013 as an example, case verification is carried out.
Table 1 lists flow and the Pearson correlation coefficients of density between wherein two groups of detector tracks, therefrom finds out track
Between flow and density specific rate it is more relevant.Find out from the data of detector 1, track 1 and 2 is than the flow between other tracks and close
Degree is more related.Equally, it is more related than the flow and density in other tracks to track 1 and 4 for detector 2, track 2 and 3.
Therefrom find out, the data dependence between track is very different.
The Pearson correlation coefficients of table 1
Two indices are chosen to quantify to fill up precision：Average absolute percent error (MAPE) and normalization root-mean-square error
(NRMSE).The two indexs are calculated as follows：
Data filling method is encoded and realized in MATLAB softwares.The test program can create data integrity matrix,
And missing data therein is deleted.Missing data can completely random, can also be determined in advance.Data volume is lost in total amount of data
Percentage, i.e. missing data ratio is adjustable.A test program database data of inquiry per minute, passes through diagnosis
Algorithm diagnoses integrality, generates a data integrity matrices, and be entered into and fill up in module.Module is filled up to be filled up
Calculate and calculated value is filled up into missing data position.In subsequent calculating, actual value will be considered as in case filling up by filling up value
Other data.
In order to assess the effect of time trend in a model, the present invention is first by the model with only considering spatial coherence
Model compares.Model 1 is proposed model (i.e. formula (1)), and model 2 is that a utilization space correlation is filled up
The model (i.e. formula (10)) of missing data.Off-line calibration is used in this part of assessment, it is assumed that missing data is in predetermined car
Road, rather than completely random, the data on predetermined track are assumed that whole day all lacks.
What using in May, 2015 in detector 1, the workaday data of the last fortnight were demarcated, then using of that month 3rd week
Workaday data are assessed model.Table 2 lists obtained model coefficient and statistical value.As a result find, all
In the case of consider that the model of time trend and spatial coherence improves simultaneously and fill up precision.The R of the gained of model 12Value is all higher than
0.8.In addition, all MAPE and NRMSE values are smaller than model 1, the estimation of particularly the 3rd track has been worth to obvious improvement.Cause
This, time trend is included to fill up to significantly improve in model and fills up precision.
The coefficient and statistical value of the model of table 2
(a) data on flows missing is filled up
(b) density data missing is filled up
, next will contrast on-line proving and off-line calibration after the validity of time trend is introduced during model is determined
Effect in data filling, on-line proving is illustrated to filling up the lifting of precision and its applicability in traffic control system.
On-line proving model dynamic generation model parameter, and off-line calibration model demarcates fixed model parameter from historical data.According to
Certain missing data ratio, the random erasure data from database.Missing data ratio is from 5% to 30%.For every
A kind of missing data ratio, generate different missing data situation and run 20 times and fill up.Fig. 2 shows the missing of detector 1 20%
Data carry out the effect that density data is filled up.Show that online and offline nominal data fills up the ratio of result and partial data in figure
Compared with.Compared with off-line calibration data filling, on-line proving data filling is more bonded initial data.In the evening peak period, online mark
Surely the situation that density increases or decreases suddenly that can capture is filled up.On the contrary, off-line calibration, which is filled up, causes irrational fill up
As a result fluctuate.MAPE values drop to 0.075, NRSME values from 0.101 and drop to 0.1490 from 0.1978 under on-line proving, than from
Line demarcation, which is filled up, provides more preferable data filling effect.
It will be understood by those skilled in the art that each module or each step of the invention described above can be filled with general computer
Put to realize, alternatively, they can be realized with the program code that computing device can perform, it is thus possible to which they are stored
Performed in the storage device by computing device, either they are fabricated to respectively each integrated circuit modules or by they
In multiple modules or step be fabricated to single integrated circuit module to realize.The present invention be not restricted to any specific hardware and
The combination of software.
Although above-mentioned the embodiment of the present invention is described with reference to accompanying drawing, model not is protected to the present invention
The limitation enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
Need to pay various modifications or deformation that creative work can make still within protection scope of the present invention.
Claims (10)
- A kind of 1. Expressway Traffic Flow data filling method based on on-line proving, it is characterised in that including：Data acquisition：To detector packet numbering, traffic flow data is established according to the traffic flow data that the detector collects Storehouse；Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake；Model is established：According to the time trend and spatial relationship of each track data of each group detector, multivariate linear model is established；On-line proving：Parameter Self-regression Forecast Model is established, based on model parameter obtained by the Primary Stage Data in traffic flow data storehouse Predict the model parameter at current time；Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
- A kind of 2. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 1, it is characterised in that The traffic flow data storehouse includes the Primary Stage Data of traffic flow real time data, historical data average value and each moment, when described each The data gathered before the Primary Stage Data at quarter refers to the moment in specified time step-length.
- A kind of 3. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 1, it is characterised in that The diagnosis is based on traffic flow data threshold value set in advance, acceptable area and the corresponding upstream track data of diagnostic data The diagnostic module of progress.
- A kind of 4. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 3, it is characterised in that If certain moment traffic flow data of collection exceeds threshold value or acceptable area, it is error number to diagnose the moment traffic flow data According to；If certain moment traffic flow data of collection is zero, and its corresponding upstream track data non-zero, then the moment is diagnosed as missing Data.
- A kind of 5. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 4, it is characterised in that It is each detector acquisition data integrity index Δ by real-time diagnosis：M is detector packet numbering, and n is that every lane detector is numbered.
- A kind of 6. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 1, it is characterised in that The model is established and is based on time trend and spatial coherence；The average value institute that the time trend passes through historical data in the same time Reflection, the spatial coherence embody the estimate that the representative obtained by correlation between the detectors is fluctuated immediately.
- A kind of 7. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 5, it is characterised in that The modelling phase establishes multivariate linear model：<mrow> <msubsup> <mi>x</mi> <mrow> <mi>e</mi> <mi>s</mi> <mi>t</mi> <mo>,</mo> <mi>t</mi> </mrow> <mi>m</mi> </msubsup> <mo>=</mo> <msubsup> <mi>&Delta;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>&CenterDot;</mo> <msubsup> <mi>x</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>+</mo> <mrow> <mo>(</mo> <mi>&Omega;</mi> <mo>-</mo> <msubsup> <mi>&Delta;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>)</mo> </mrow> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msubsup> <mi>&alpha;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <msubsup> <mi>&Delta;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>&CenterDot;</mo> <msubsup> <mi>x</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>+</mo> <msubsup> <mi>&beta;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>&CenterDot;</mo> <msubsup> <mover> <mi>x</mi> <mo>&OverBar;</mo> </mover> <mi>t</mi> <mi>m</mi> </msubsup> <mo>+</mo> <msubsup> <mi>&gamma;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>+</mo> <msubsup> <mi>&epsiv;</mi> <mi>t</mi> <mi>m</mi> </msubsup> <mo>)</mo> </mrow> </mrow>Wherein, xestTraffic flow data vector is represented, α, β and γ are model parameter, and α is used to embody traffic flow data space correlation Property；β is used to embody traffic flow data and the correlation of time average tendency；β=diag (β1,β2,...,βN)Δ=diag (△1,△2,...,△N)X=(x1,...,xN)T<mrow> <mover> <mi>x</mi> <mo>&OverBar;</mo> </mover> <mo>=</mo> <msup> <mrow> <mo>(</mo> <msub> <mover> <mi>x</mi> <mo>&OverBar;</mo> </mover> <mn>1</mn> </msub> <mo>,</mo> <mo>...</mo> <mo>,</mo> <msub> <mover> <mi>x</mi> <mo>&OverBar;</mo> </mover> <mi>N</mi> </msub> <mo>)</mo> </mrow> <mi>T</mi> </msup> </mrow>γ=(γ0,...,γN)TFor constant term；αijRepresent the track j obtained when filling up track i according to data space correlation Coefficient in multivariate linear model；βiTrack i time average tendency is in multivariate linear model during to fill up track i data Coefficient；Δ is data integrity degree matrix；Ω=diag (1 ..., 1)；X is the flow or density that detector detects；For The average value of historical data in the same time；ε=(ε0,...,εN)TFor error term.
- A kind of 8. Expressway Traffic Flow data filling method based on on-line proving as claimed in claim 1, it is characterised in that The on-line proving specifically includes：According to the multivariate linear model of foundation, using the Primary Stage Data in preceding several time steps, fortune Model parameter is optimized with least square method, makes error term close to zero；By repeated optimization process, obtain and preceding issue According to corresponding model parameter；Model parameter based on model parameter prediction current time corresponding to Primary Stage Data.
- 9. a kind of Expressway Traffic Flow data filling system based on on-line proving, including processor and computer-readable storage medium Matter, processor are used to realize each instruction；Computer-readable recording medium is used to store a plurality of instruction, it is characterised in that the finger Order is suitable to be loaded by processor and perform following processing：Data acquisition：To detector packet numbering, traffic flow data is established according to the traffic flow data that the detector collects Storehouse；Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake；Model is established：According to the time trend and spatial relationship of each track data of each group detector, multivariate linear model is established；On-line proving：Parameter Self-regression Forecast Model is established, based on model parameter obtained by the Primary Stage Data in traffic flow data storehouse Predict the model parameter at current time；Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
- 10. a kind of computer-readable recording medium, is stored thereon with computer program, for Expressway Traffic Flow data filling, Characterized in that, the program performs following steps when being executed by processor：Data acquisition：To detector packet numbering, traffic flow data is established according to the traffic flow data that the detector collects Storehouse；Real-time diagnosis：The traffic flow data collected in real time to the detector diagnoses, and whether diagnosis lacks or mistake；Model is established：According to the time trend and spatial relationship of each track data of each group detector, multivariate linear model is established；On-line proving：Parameter Self-regression Forecast Model is established, based on model parameter obtained by the Primary Stage Data in traffic flow data storehouse Predict the model parameter at current time；Data filling：Traffic flow data is carried out according to the model parameter at the multivariate linear model and current time to fill up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710802197.4A CN107591001B (en) | 2017-09-07 | 2017-09-07 | Expressway traffic flow data filling method and system based on online calibration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710802197.4A CN107591001B (en) | 2017-09-07 | 2017-09-07 | Expressway traffic flow data filling method and system based on online calibration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107591001A true CN107591001A (en) | 2018-01-16 |
| CN107591001B CN107591001B (en) | 2019-12-27 |
Family
ID=61050640
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710802197.4A Active CN107591001B (en) | 2017-09-07 | 2017-09-07 | Expressway traffic flow data filling method and system based on online calibration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107591001B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109191846A (en) * | 2018-10-12 | 2019-01-11 | 国网浙江省电力有限公司温州供电公司 | A kind of traffic trip method for predicting |
| CN109272760A (en) * | 2018-10-18 | 2019-01-25 | 银江股份有限公司 | A kind of online test method of SCATS system detector data outliers |
| CN109710659A (en) * | 2018-12-16 | 2019-05-03 | 苏州城方信息技术有限公司 | The complementing method of detector missing data based on temporal correlation |
| CN109726503A (en) * | 2019-01-12 | 2019-05-07 | 国电联合动力技术有限公司 | Missing data complementing method and device |
| CN114639233A (en) * | 2020-12-15 | 2022-06-17 | 腾讯科技（深圳）有限公司 | Congestion state prediction method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103050005A (en) * | 2012-11-16 | 2013-04-17 | 北京交通大学 | Method and system for space and time analysis of urban road traffic states |
| CN103247177A (en) * | 2013-05-21 | 2013-08-14 | 清华大学 | Large-scale road network traffic flow real-time dynamic prediction system |
| CN103578274A (en) * | 2013-11-15 | 2014-02-12 | 北京四通智能交通系统集成有限公司 | Method and device for forecasting traffic flows |
| CN105632193A (en) * | 2015-12-25 | 2016-06-01 | 银江股份有限公司 | Data missing road speed calculation method based on time-space relevance |
| CN105702029A (en) * | 2016-02-22 | 2016-06-22 | 北京航空航天大学 | Express way traffic state prediction method taking spatial-temporal correlation into account at different times |
| KR20170073220A (en) * | 2015-12-18 | 2017-06-28 | 한국과학기술원 | Apparatus and method for correcting transportation data |
-
2017
- 2017-09-07 CN CN201710802197.4A patent/CN107591001B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103050005A (en) * | 2012-11-16 | 2013-04-17 | 北京交通大学 | Method and system for space and time analysis of urban road traffic states |
| CN103247177A (en) * | 2013-05-21 | 2013-08-14 | 清华大学 | Large-scale road network traffic flow real-time dynamic prediction system |
| CN103578274A (en) * | 2013-11-15 | 2014-02-12 | 北京四通智能交通系统集成有限公司 | Method and device for forecasting traffic flows |
| KR20170073220A (en) * | 2015-12-18 | 2017-06-28 | 한국과학기술원 | Apparatus and method for correcting transportation data |
| CN105632193A (en) * | 2015-12-25 | 2016-06-01 | 银江股份有限公司 | Data missing road speed calculation method based on time-space relevance |
| CN105702029A (en) * | 2016-02-22 | 2016-06-22 | 北京航空航天大学 | Express way traffic state prediction method taking spatial-temporal correlation into account at different times |
Non-Patent Citations (1)
| Title |
|---|
| 昝艳 等: "基于混合模型的城市历史交通流数据分析方法", 《第五届中国智能交通年会暨第六届国际节能与新能源汽车创新发展论坛优秀论文集（上册）》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109191846A (en) * | 2018-10-12 | 2019-01-11 | 国网浙江省电力有限公司温州供电公司 | A kind of traffic trip method for predicting |
| CN109272760A (en) * | 2018-10-18 | 2019-01-25 | 银江股份有限公司 | A kind of online test method of SCATS system detector data outliers |
| CN109710659A (en) * | 2018-12-16 | 2019-05-03 | 苏州城方信息技术有限公司 | The complementing method of detector missing data based on temporal correlation |
| CN109710659B (en) * | 2018-12-16 | 2022-11-25 | 苏州城方信息技术有限公司 | Method for filling detector missing data based on space-time correlation |
| CN109726503A (en) * | 2019-01-12 | 2019-05-07 | 国电联合动力技术有限公司 | Missing data complementing method and device |
| CN109726503B (en) * | 2019-01-12 | 2020-12-18 | 国电联合动力技术有限公司 | Missing data filling method and device |
| CN114639233A (en) * | 2020-12-15 | 2022-06-17 | 腾讯科技（深圳）有限公司 | Congestion state prediction method and device, electronic equipment and storage medium |
| CN114639233B (en) * | 2020-12-15 | 2024-02-02 | 腾讯科技（深圳）有限公司 | Congestion state prediction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107591001B (en) | 2019-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107591001A (en) | Expressway Traffic Flow data filling method and system based on on-line proving | |
| CN104134999B (en) | Distribution network based on multi-data source measures the practical method of calculation of efficiency analysis | |
| CN102289731B (en) | Method for maintaining state of power transmission equipment based on system risk | |
| CN103793752B (en) | A kind of equipment failure number Forecasting Methodology based on modeling of degenerating | |
| CN104573881B (en) | A kind of military service equipment residual life adaptive forecasting method based on degraded data modeling | |
| CN103399281B (en) | Based on the ND-AR model of cycle life deterioration stage parameter and the cycle life of lithium ion battery Forecasting Methodology of EKF method | |
| CN103971171A (en) | State evaluation method for power transmission equipment | |
| CN103699668B (en) | Electric distribution network electrical equipment assembled state appraisal procedure based on data section uniformity | |
| CN105427005A (en) | Operation risk assessment method of wind power station | |
| CN106251040A (en) | A kind of method that cable run is carried out health state evaluation | |
| CN104218570A (en) | Method and system for online evaluating overall measuring errors of electric energy measuring device | |
| CN104881003A (en) | Effectiveness evaluation method for metering production facilities | |
| CN111476471B (en) | Comprehensive energy fault diagnosis system and method based on comprehensive energy model | |
| CN103544542A (en) | Power system transient stability margin predicting method | |
| CN106251042A (en) | A kind of method that distribution transformer is carried out health state evaluation | |
| CN107121639A (en) | A kind of multi-Dimensional parameters straight-flow system accumulator cell management method and device | |
| CN104021304A (en) | Installation priority level evaluation method for on-line monitoring devices of transformers | |
| CN107293118A (en) | A kind of traffic speed motion interval Forecasting Approach for Short-term | |
| CN111178725A (en) | Protective equipment state early warning method based on analytic hierarchy process | |
| CN111191832A (en) | Typhoon disaster power distribution network tower fault prediction method and system | |
| CN106595788A (en) | Large pump station flow monitoring method based on multi-parameter coupling correction | |
| CN104331849B (en) | Electrical power system risk ranking method and application | |
| CN104915192A (en) | Software reliability modeling method based on moving points and imperfect error removing | |
| CN107607820A (en) | A kind of inside transformer Hidden fault rate Forecasting Methodology based on birth and death process | |
| CN102751725A (en) | Overload risk state identifying method for power distribution network |
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 |