CN104183134B - The highway short-term traffic flow forecast method of vehicle is divided based on intelligence - Google Patents
The highway short-term traffic flow forecast method of vehicle is divided based on intelligence Download PDFInfo
- Publication number
- CN104183134B CN104183134B CN201410427899.5A CN201410427899A CN104183134B CN 104183134 B CN104183134 B CN 104183134B CN 201410427899 A CN201410427899 A CN 201410427899A CN 104183134 B CN104183134 B CN 104183134B
- Authority
- CN
- China
- Prior art keywords
- data
- vehicle
- traffic flow
- sequence
- car
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a kind of highway short-term traffic flow forecast method based on intelligence point vehicle, comprise the steps: step one, obtain data; Step 2, data prediction; Step 3, stationary test is carried out to the total vehicle flowrate data sequence obtained after pre-service, if total vehicle flowrate data sequence is stable data sequence, then adopt time series method predicting traffic flow amount; If total vehicle flowrate data sequence is non-stationary data series, then two class vehicles are divided to predict: to adopt time series method to carry out traffic flow forecasting to compact car and container type car; Secondary Exponential Smoothing Method is adopted to carry out traffic flow forecasting to in-between car and large car; Step 4, will to predict the outcome according to the Passenger car equivalent of different automobile types and to be converted to standard vehicle, calculate total vehicle flowrate predicted value; Step 5, wait until next time Data Update time, perform step one.The present invention can improve precision of prediction, helps traffic administration person or traveler to hold road conditions better.
Description
Technical field
The invention belongs to the traffic control system technical field of road vehicle, concrete is a kind of highway short-term traffic flow forecast method of dividing vehicle based on intelligence.
Background technology
Along with the increase of highway mileage, to the information system management requirement of traffic and traveler, the Informatization Service demand to traffic is also increasing traffic administration person gradually, the traffic holding following section in real time is accurately the prerequisite of efficient management and countermeasure, and is the key held section future transportation situation to the prediction of Short-Term Traffic Flow.The how rule of accurate assurance traffic flow, improving the precision of volume forecasting, is the important directions of existing research.
Intelligent Forecasting, the method based on combined prediction and the Forecasting Methodology scheduling theory based on the traffic flow mode basis that the Forecasting Methodology of existing highway Short-Term Traffic Flow relates generally to the method based on lineary system theory, the method based on nonlinear theory, knowledge based find.Based on these theoretical foundation, have developed the Forecasting Methodology of multiple highway Short-Term Traffic Flow at present:
(1) Five City University's journal (natural science edition) (the 18th volume the 3rd phase, in September, 2004) disclose a kind of time series modeling and Forecasting Methodology of traffic flow, it adopts time series models to predict Short-Term Traffic Flow, real data the result shows that this model can matching Traffic Flow Time Series preferably, and can obtain higher precision of prediction;
(2) Communication and Transportation Engineering and information journal (the 8th volume the 1st phase, in March, 2010) disclose a kind of road section short-term traffic flow forecasting model based on chaos time sequence, its chaotic characteristic for the Short-Term Traffic Flow sequence of road section is analyzed, and actual measurement traffic flow data the result indicates this forecast model and has validity to a certain extent;
(3) system engineering theory with put into practice (the 30th volume the 2nd phase, in February, 2010) disclose a kind of Short-time Traffic Flow Forecasting Methods based on k nearest neighbor non parametric regression, and the forecast interval utilizing K value to construct is used for, in the prediction of special road conditions, have also been obtained obvious improvement effect;
(4) Jilin University's journal (engineering version) (the 40th volume the 5th phase, in September, 2010) disclose a kind of traffic parameter combination forecasting method based on wavelet analysis, the method predicts the variation tendency of traffic parameter more exactly, there is universality, and simpler than traditional combined prediction process based on wavelet analysis, for the real-time application of macrooperation amount provides possibility;
(5) Beijing Jiaotong University Dong Chun spoils, for the spatio-temporal distribution characteristic of traffic flow parameter under city expressway freestream conditions, crowded stream mode and blocked flow state, propose urban freeway network short-term traffic flow prediction theory and methods under a kind of multimode, it have studied urban expressway traffic stream short-term prediction under admixture, experiment shows, traditional Forecasting Methodology of comparing has higher precision of prediction.
Make a general survey of the Forecasting Methodology of the above various magnitude of traffic flow, in order to improve the effect of prediction, need the variation tendency better excavating traffic behavior, hold the regularity of traffic flow inside, and above-mentioned traffic flow forecasting method is all for research object carrys out the change of analyses and prediction traffic flow with total magnitude of traffic flow.
Summary of the invention
Found by point vehicle data on flows analysis gathered microwave vehicle checker: the compact car on highway is all lower with the flow of container type car every day, and flow rate fluctuation in one day is more steady; The fluctuations in discharge of in-between car every day presents two obvious crests, but it is all more steady in fluctuation sooner or later, and except morning, evening peak exist except one section of rise and fall trend respectively in the large car flow curve of a day, remaining time period fluctuations in discharge is relatively steady.As can be seen here, the fluctuation Changing Pattern of different automobile types flow is different, and the fluctuations in discharge of vehicle in one day is also along with the change of time changes separately, and total vehicle flowrate of comparing, the regularity of the magnitude of traffic flow change of each vehicle is more obvious.
In view of this, the object of the present invention is to provide a kind of highway short-term traffic flow forecast method based on intelligence point vehicle, can according to the feature of Real-Time Traffic Volume data, adjustment forecast model, thus reach intelligence point vehicle prediction, improve precision of prediction.
For achieving the above object, the invention provides following technical scheme:
Divide a highway short-term traffic flow forecast method for vehicle based on intelligence, it is characterized in that: comprise the steps:
Step one, acquisition data: obtain total vehicle flowrate data sequence of highway, point vehicle data on flows sequence, average speed and average occupancy;
Step 2, data prediction: reject the data not meeting traffic actual conditions;
Step 3, stationary test is carried out to the total vehicle flowrate data sequence obtained after pre-service, if total vehicle flowrate data sequence is stable data sequence, then adopt time series method predicting traffic flow amount; If total vehicle flowrate data sequence is non-stationary data series, then vehicle is divided into two classes, first kind vehicle comprises compact car and container type car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts time series method to predict; Equations of The Second Kind vehicle comprises in-between car and large car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts Secondary Exponential Smoothing Method to predict;
Wherein, time series method predicting traffic flow amount is adopted to comprise the following steps:
(1) stationarity judgement is carried out to point vehicle data on flows sequence obtained, if point vehicle data on flows sequence stationary of correspondence, then directly perform next step; If point vehicle data on flows sequence of correspondence is not steady, then difference processing is carried out to this point of vehicle data on flows sequence, after obtaining new stable point vehicle data on flows sequence, then perform next step;
(2) adopt ARIMA model to predict as forecast model, and for the new stable point vehicle data on flows sequence obtained after difference processing, also needing predicts the outcome to it carry out inverse transformation, is converted to the volume forecasting value of corresponding vehicle;
Secondary Exponential Smoothing Method predicting traffic flow amount is adopted to comprise the following steps:
(1) one-accumulate is carried out to point vehicle data on flows sequence obtained, and curve after cumulative is chosen by the calculating of the coefficient of determination form K best neighbour's sequence of the linearity;
(2) method determines that the smoothing factor α minimum with prediction standard error is postfitted orbit coefficient by experiment;
(3) predict according to Secondary Exponential Smoothing Method forecast model, and by being converted to the volume forecasting value of corresponding vehicle;
Step 4, will to predict the outcome according to the Passenger car equivalent of different automobile types and to be converted to standard vehicle, calculate total vehicle flowrate predicted value;
Step 5, wait until next time Data Update time, perform step one.
Further, in described step 2, threshold theory and traffic flow theory is adopted to reject the data not meeting traffic actual conditions respectively;
Described threshold theory is: within a data update cycle, sets the threshold range of total vehicle flowrate data as [0, Q
max], the threshold range of average speed is [0, V
max]; If when the data of the total vehicle flowrate data collected or average speed are not in the threshold range of correspondence, then show that these group data are unreliable, and rejected; If when the data of the total vehicle flowrate data collected and average number of vehicles all drop in corresponding threshold range, then show that these group data are reliable, retain this group data; Wherein, Q
max, V
maxbe illustrated respectively in the flow maximum in the data update cycle and speed maximal value;
Described traffic flow theory is: first, sets up misdata judgment rule, namely reject rule according to traffic flow theory; Then, judge whether the data sequence gathered meets and reject rule; When satisfied rejecting rule, the data of correspondence are needed reject; When not meeting rejecting rule, retain corresponding data.
Further, in described step 3, the stationary test method of total vehicle flowrate data sequence is:
(1) obtain through pretreated total vehicle flowrate data sequence X
t, and postpone k and obtain X
t+k, calculate its average μ separately
t, μ
t+k;
(2) according to its autocorrelation function R (k) of autocorrelation function formulae discovery:
Wherein, σ
2for variance, k is the lag period;
(3) when autocorrelation function R (k) can not level off to 0 or fluctuate near 0 by rapid decay, then total vehicle flowrate data sequence belongs to non-stationary series; When autocorrelation function R (k) can rapid decay to 0, then total vehicle flowrate data sequence belongs to stationary sequence.
Further, in described step 3, the formula of described ARIMA model is as follows:
Wherein, x
tfor a point vehicle data on flows sequence, ε
tfor white noise, B is delay operator (B
jx
t=x
t-j), d is difference number of times; P is the order of model;
Utilize the Correlation Moment estimation technique and BIC criterion to carry out the estimation of model parameter and the determination of model order p, concrete formula is as follows:
Wherein,
with
(i=1,2 ..., p) represent autocorrelation function and the model parameter of point vehicle data on flows sequence respectively; P represents the order of model, and when calculating autocorrelation function, p is taken as
wherein N represents sample size.
Further, in described step 3, the formula of Secondary Exponential Smoothing Method forecast model is as follows:
Wherein:
it is t+T phase predicted value;
A
tand b
tbe respectively model parameter, and
be the single exponential smoothing value of t phase, and
be the double smoothing value of t phase, and
α is smoothing factor, X
t-1for initial value;
Volume forecasting value
conversion formula be:
Wherein,
with
for utilizing Secondary Exponential Smoothing Method to the volume forecasting value in optimum k nearest neighbor sequence t+T-1 moment and t+T moment.
Further, in described step 4, the total vehicle flowrate predicted value after the conversion of t
for:
Wherein, q
1(t), q
2(t), q
3(t), q
4t () is respectively the volume forecasting value of t compact car, in-between car, large car, container type car.
Beneficial effect of the present invention is:
The present invention is based on the highway short-term traffic flow forecast method of intelligence point vehicle, from Real-Time Traffic Volume Variation Features and the angle of the fluctuations in discharge feature of different automobile types in a day, by the stationarity of real-time judgement magnitude of traffic flow sequence, determine whether that needs carry out a point vehicle prediction; Namely the method has held real-time flow feature, have also contemplated that overall fluctuations in discharge trend and the Variation Features of different automobile types vehicle flowrate, compare and do not consider the Forecasting Methodology of vehicle, it more can show the regularity of traffic flow inside, can obtain higher precision of prediction.
Accompanying drawing explanation
In order to make object of the present invention, technical scheme and beneficial effect clearly, the invention provides following accompanying drawing and being described:
Fig. 1 is the process flow diagram of the highway short-term traffic flow forecast embodiment of the method that the present invention is based on intelligence point vehicle;
Fig. 2 is that the present embodiment carries out the process flow diagram of stationary test to total vehicle flowrate data sequence;
Fig. 3 is the process flow diagram that the present embodiment adopts time series method predicting traffic flow amount;
Fig. 4 is the process flow diagram that the present embodiment adopts Secondary Exponential Smoothing Method predicting traffic flow amount.
Embodiment
Below in conjunction with the drawings and specific embodiments, the invention will be further described, can better understand the present invention and can be implemented, but illustrated embodiment is not as a limitation of the invention to make those skilled in the art.
As shown in Figure 1, for the present invention is based on the process flow diagram of the highway short-term traffic flow forecast embodiment of the method for intelligence point vehicle.The present embodiment divides the highway short-term traffic flow forecast method of vehicle based on intelligence, it is characterized in that: comprise the steps:
Step one, obtain data: read collect through highway microwave vehicle checker total vehicle flowrate data sequence, point vehicle data on flows sequence, average speed and average occupancy, vehicle is divided into four kinds by the present embodiment, be respectively compact car, in-between car, large car and container type car, concrete, the present embodiment freeway traffic data field definition is as shown in table 1:
Table 1: freeway traffic data field definition table
The stationarity of sequence is not only subject to the impact of data itself, is also subject to the impact of sample size size.Along with the increase of data sequence sample size, the stationarity of sequence often declines thereupon, and sample size crosses the conference increase coefficient of autocorrelation of sequence and the calculated amount of partial correlation coefficient; Again due to uncertainty and the randomness of magnitude of traffic flow change, the flow sequence that sample size is too little is not only difficult to by stationary test, and cannot reflect the variation tendency of traffic flow, will produce larger error when predicting.By many experiments, the sample size of the data on flows sequence of the present embodiment chooses 25.
Step 2, data prediction: reject the data not meeting traffic actual conditions; The present embodiment adopts threshold theory and traffic flow theory to reject the data not meeting traffic actual conditions respectively;
Threshold theory is: read apart from current time a × m minute, length is the data sequence of m, and m value be 25, a the same as the sample size in step 1 is the time interval of a data update cycle here; Be set in the data update cycle, the threshold range of total vehicle flowrate data is [0, Q
max], the threshold range of average speed is [0, V
max]; If when the data of the total vehicle flowrate data collected or average speed are not in the threshold range of correspondence, then show that these group data are unreliable, and rejected; If when the data of the total vehicle flowrate data collected and average number of vehicles all drop in corresponding threshold range, then show that these group data are reliable, retain this group data; Wherein, Q
max, V
maxbe illustrated respectively in the flow maximum in the data update cycle and speed maximal value, the present embodiment Q
maxvalue is 300, V
maxvalue is 150km/h.
Described traffic flow theory is: first, sets up misdata judgment rule, namely reject rule according to traffic flow theory; Then, judge whether the data sequence gathered meets and reject rule; When satisfied rejecting rule, the data of correspondence are needed reject; When not meeting rejecting rule, retain corresponding data.The present embodiment traffic flow theory misdata judgment rule is as shown in table 2.
Table 2: based on the misdata decision rule of traffic flow theory
Step 3, carry out stationary test to the total vehicle flowrate data sequence obtained after pre-service, stationary time series refers to the time-independent time series of average, variance and Autoregressive Functions.As time series { x
tfor stationary stochastic process time, for any one period 1≤t
1< t
2... < t
mwith h>=1,
joint distribution be equal to
joint distribution; From definition, stationarity is equivalent to: all x
tall there is identical distribution; Within whole period, the degree of correlation between any two adjacencies is all identical; With mathematic(al) representation be:
(1) to any t, average perseverance is constant: Ex
t=μ (with the constant that t is irrelevant).
(2) variance Var (x
t)=σ
2(with the limited constant that t is irrelevant).
(3) to arbitrary integer t and k, autocorrelation function r
t, t+konly relevant with k, r
t, t+k=r
k.
The stationary test method of total vehicle flowrate data sequence of the present embodiment is:
(1) obtain through pretreated total vehicle flowrate data sequence X
t, and postpone k and obtain X
t+k, calculate its average μ separately
t, μ
t+k;
(2) according to its autocorrelation function R (k) of autocorrelation function formulae discovery:
Wherein, σ
2for variance, k is the lag period;
(3) when autocorrelation function R (k) can not level off to 0 or fluctuate near 0 by rapid decay, then total vehicle flowrate data sequence belongs to non-stationary series; When autocorrelation function R (k) can rapid decay to 0, then total vehicle flowrate data sequence belongs to stationary sequence.
If total vehicle flowrate data sequence is stable data sequence, then adopt time series method predicting traffic flow amount; If total vehicle flowrate data sequence is non-stationary data series, then vehicle is divided into two classes, first kind vehicle comprises compact car and container type car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts time series method to predict; Equations of The Second Kind vehicle comprises in-between car and large car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts Secondary Exponential Smoothing Method to predict;
Wherein, time series method predicting traffic flow amount is adopted to comprise the following steps:
(1) stationarity judgement is carried out to point vehicle data on flows sequence obtained, if point vehicle data on flows sequence stationary of correspondence, then directly perform next step; If point vehicle data on flows sequence of correspondence is not steady, then difference processing is carried out to this point of vehicle data on flows sequence, after obtaining new stable point vehicle data on flows sequence and difference number of times d, then perform next step;
(2) adopt ARIMA model to predict as forecast model, and for the new stable point vehicle data on flows sequence obtained after difference processing, also needing predicts the outcome to it carry out inverse transformation, is converted to the volume forecasting value of corresponding vehicle.
Concrete, the formula of ARIMA model is as follows:
Wherein, x
tfor a point vehicle data on flows sequence, ε
tfor white noise, B is delay operator (B
jx
t=x
t-j), d is difference number of times; P is the order of model,
for model parameter;
Utilize the Correlation Moment estimation technique and BIC criterion to carry out the estimation of model parameter and the determination of model order p, concrete formula is as follows:
Wherein,
with
(i=1,2 ..., p) represent autocorrelation function and the model parameter of point vehicle data on flows sequence respectively; P represents the order of model, and when calculating autocorrelation function, p is taken as
wherein N represents sample size, i.e. the model order p=5 of the present embodiment.
Secondary Exponential Smoothing Method predicting traffic flow amount is adopted to comprise the following steps:
(1) one-accumulate is carried out to point vehicle data on flows sequence obtained, and curve after cumulative is chosen by the calculating of the coefficient of determination form K best neighbour's sequence of the linearity, in the present embodiment, K=25;
(2) method determines that the smoothing factor α minimum with prediction standard error is postfitted orbit coefficient by experiment, wherein, and α ∈ [0.6,1];
(3) mean value getting optimum K front 3 values of neighbour's sequence, as initial value, in conjunction with postfitted orbit factor alpha, is predicted according to Secondary Exponential Smoothing Method forecast model, and by being converted to the volume forecasting value of corresponding vehicle.
Concrete, the formula of Secondary Exponential Smoothing Method forecast model is as follows:
Wherein:
it is t+T phase predicted value;
A
tand b
tbe respectively model parameter, and
be the single exponential smoothing value of t phase, and
be the double smoothing value of t phase, and
α is smoothing factor, X
t-1for initial value;
Volume forecasting value
conversion formula be:
Wherein,
with
for utilizing Secondary Exponential Smoothing Method to the volume forecasting value in optimum k nearest neighbor sequence t+T-1 moment and t+T moment.
Step 4, will to predict the outcome according to the Passenger car equivalent of different automobile types and to be converted to standard vehicle, calculate total vehicle flowrate predicted value.Concrete, the total vehicle flowrate predicted value after the conversion of t
for:
Wherein, q
1(t), q
2(t), q
3(t), q
4t () is respectively the volume forecasting value of t compact car, in-between car, large car, container type car.
Step 5, wait until next time Data Update time, perform step one.
The present embodiment divides the highway short-term traffic flow forecast method of vehicle based on intelligence, from Real-Time Traffic Volume Variation Features and the angle of the fluctuations in discharge feature of different automobile types in a day, by the stationarity of real-time judgement magnitude of traffic flow sequence, determine whether that needs carry out a point vehicle prediction; Namely the method has held real-time flow feature, have also contemplated that overall fluctuations in discharge trend and the Variation Features of different automobile types vehicle flowrate, compare and do not consider the Forecasting Methodology of vehicle, it more can show the regularity of traffic flow inside, can obtain higher precision of prediction.
The above embodiment is only that protection scope of the present invention is not limited thereto in order to absolutely prove the preferred embodiment that the present invention lifts.The equivalent alternative or conversion that those skilled in the art do on basis of the present invention, all within protection scope of the present invention.Protection scope of the present invention is as the criterion with claims.
Claims (6)
1. divide a highway short-term traffic flow forecast method for vehicle based on intelligence, it is characterized in that: comprise the steps:
Step one, acquisition data: obtain total vehicle flowrate data sequence of highway, point vehicle data on flows sequence, average speed and average occupancy;
Step 2, data prediction: reject the data not meeting traffic actual conditions;
Step 3, stationary test is carried out to the total vehicle flowrate data sequence obtained after pre-service, if total vehicle flowrate data sequence is stable data sequence, then adopt time series method predicting traffic flow amount; If total vehicle flowrate data sequence is non-stationary data series, then vehicle is divided into two classes, first kind vehicle comprises compact car and container type car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts time series method to predict; Equations of The Second Kind vehicle comprises in-between car and large car two kinds of vehicles, and the traffic flow forecasting method of such vehicle adopts Secondary Exponential Smoothing Method to predict;
Wherein, time series method predicting traffic flow amount is adopted to comprise the following steps:
(1) stationarity judgement is carried out to point vehicle data on flows sequence obtained, if point vehicle data on flows sequence stationary of correspondence, then directly perform next step; If point vehicle data on flows sequence of correspondence is not steady, then difference processing is carried out to this point of vehicle data on flows sequence, after obtaining new stable point vehicle data on flows sequence, then perform next step;
(2) adopt ARIMA model to predict as forecast model, and for the new stable point vehicle data on flows sequence obtained after difference processing, also needing predicts the outcome to it carry out inverse transformation, is converted to the volume forecasting value of corresponding vehicle;
Secondary Exponential Smoothing Method predicting traffic flow amount is adopted to comprise the following steps:
(1) one-accumulate is carried out to point vehicle data on flows sequence obtained, and curve after cumulative is chosen by the calculating of the coefficient of determination form K best neighbour's sequence of the linearity;
(2) method determines that the smoothing factor α minimum with prediction standard error is postfitted orbit coefficient by experiment;
(3) predict according to Secondary Exponential Smoothing Method forecast model, and by being converted to the volume forecasting value of corresponding vehicle;
Step 4, will to predict the outcome according to the Passenger car equivalent of different automobile types and to be converted to standard vehicle, calculate total vehicle flowrate predicted value;
Step 5, wait until next time Data Update time, perform step one.
2. divide the highway short-term traffic flow forecast method of vehicle according to claim 1 based on intelligence, it is characterized in that: in described step 2, adopt threshold theory and traffic flow theory to reject the data not meeting traffic actual conditions respectively;
Described threshold theory is: within a data update cycle, sets the threshold range of total vehicle flowrate data as [0, Q
max], the threshold range of average speed is [0, V
max]; If when the data of the total vehicle flowrate data collected or average speed are not in the threshold range of correspondence, then show that these group data are unreliable, and rejected; If when the data of the total vehicle flowrate data collected and average number of vehicles all drop in corresponding threshold range, then show that these group data are reliable, retain this group data; Wherein, Q
max, V
maxbe illustrated respectively in the flow maximum in the data update cycle and speed maximal value;
Described traffic flow theory is: first, sets up misdata judgment rule, namely reject rule according to traffic flow theory; Then, judge whether the data sequence gathered meets and reject rule; When satisfied rejecting rule, the data of correspondence are needed reject; When not meeting rejecting rule, retain corresponding data.
3. divide the highway short-term traffic flow forecast method of vehicle according to claim 1 based on intelligence, it is characterized in that: in described step 3, the stationary test method of total vehicle flowrate data sequence is:
(1) obtain through pretreated total vehicle flowrate data sequence X
t, and postpone k and obtain X
t+k, calculate its average μ separately
t, μ
t+k;
(2) according to its autocorrelation function R (k) of autocorrelation function formulae discovery:
Wherein, σ
2for variance, k is the lag period;
(3) when autocorrelation function R (k) can not level off to 0 or fluctuate near 0 by rapid decay, then total vehicle flowrate data sequence belongs to non-stationary series; When autocorrelation function R (k) can rapid decay to 0, then total vehicle flowrate data sequence belongs to stationary sequence.
4. divide the highway short-term traffic flow forecast method of vehicle according to claim 1 based on intelligence, it is characterized in that: in described step 3, the formula of described ARIMA model is as follows:
Wherein, x
tfor a point vehicle data on flows sequence, ε
tfor white noise, B is delay operator (B
jx
t=x
t-j), d is difference number of times; P is the order of model;
Utilize the Correlation Moment estimation technique and BIC criterion to carry out the estimation of model parameter and the determination of model order p, concrete formula is as follows:
Wherein,
with
represent autocorrelation function and the model parameter of point vehicle data on flows sequence respectively; P represents the order of model, and when calculating autocorrelation function, p is taken as
wherein N represents sample size.
5. divide the highway short-term traffic flow forecast method of vehicle according to claim 1 based on intelligence, it is characterized in that: in described step 3, the formula of Secondary Exponential Smoothing Method forecast model is as follows:
Wherein:
it is t+T phase predicted value;
A
tand b
tbe respectively model parameter, and
be the single exponential smoothing value of t phase, and
be the double smoothing value of t phase, and
α is smoothing factor, X
t-1for initial value;
Volume forecasting value
conversion formula be:
Wherein,
with
for utilizing Secondary Exponential Smoothing Method to the volume forecasting value in optimum k nearest neighbor sequence t+T-1 moment and t+T moment.
6. divide the highway short-term traffic flow forecast method of vehicle according to claim 1 based on intelligence, it is characterized in that: in described step 4, the total vehicle flowrate predicted value after the conversion of t
for:
Wherein, q
1(t), q
2(t), q
3(t), q
4t () is respectively the volume forecasting value of t compact car, in-between car, large car, container type car.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410427899.5A CN104183134B (en) | 2014-08-27 | 2014-08-27 | The highway short-term traffic flow forecast method of vehicle is divided based on intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410427899.5A CN104183134B (en) | 2014-08-27 | 2014-08-27 | The highway short-term traffic flow forecast method of vehicle is divided based on intelligence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104183134A CN104183134A (en) | 2014-12-03 |
CN104183134B true CN104183134B (en) | 2016-02-10 |
Family
ID=51964136
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410427899.5A Active CN104183134B (en) | 2014-08-27 | 2014-08-27 | The highway short-term traffic flow forecast method of vehicle is divided based on intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104183134B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105957329A (en) * | 2016-06-29 | 2016-09-21 | 肖锐 | Intelligentized information system for highway |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899663B (en) * | 2015-06-17 | 2019-02-26 | 北京奇虎科技有限公司 | A kind of data predication method and device |
CN105761489A (en) * | 2016-04-13 | 2016-07-13 | 南京邮电大学 | Cubic exponential smoothing optimal method of traffic flow prediction |
CN105913654B (en) * | 2016-06-29 | 2018-06-01 | 深圳市前海绿色交通有限公司 | A kind of Intelligent traffic management systems |
CN106128101B (en) * | 2016-06-29 | 2018-09-04 | 南京城建隧桥经营管理有限责任公司 | The visualization device of magnitude of traffic flow short-term prediction result |
CN108428338B (en) * | 2017-02-15 | 2021-11-12 | 阿里巴巴集团控股有限公司 | Traffic road condition analysis method and device and electronic equipment |
CN107256636A (en) * | 2017-06-29 | 2017-10-17 | 段晓辉 | A kind of traffic flow acquisition methods for merging laser scanning and video technique |
CN107798871B (en) * | 2017-10-27 | 2020-07-03 | 云南大学 | Traffic flow prediction method and system for highway toll station |
CN107798877B (en) * | 2017-11-22 | 2020-10-02 | 北京云星宇交通科技股份有限公司 | Method and system for predicting traffic volume based on highway charging data |
CN108898851B (en) * | 2018-06-20 | 2020-11-27 | 东南大学 | Combined prediction method for traffic volume of urban road section |
CN109887290B (en) * | 2019-03-30 | 2021-03-23 | 西安电子科技大学 | Traffic flow prediction method based on balance index smoothing method and stack type self-encoder |
CN111696369B (en) * | 2020-04-10 | 2023-04-28 | 北京数城未来科技有限公司 | All-market road time-sharing and vehicle-division type traffic flow prediction method based on multi-source geographic space big data |
CN111951556B (en) * | 2020-08-20 | 2022-04-29 | 重庆大学 | Bus travel time prediction method considering truck mixing rate based on RFID data |
CN111951557B (en) * | 2020-08-21 | 2022-05-24 | 成都工业学院 | Regional short-term traffic flow prediction method and system based on Internet of vehicles big data |
CN111932898B (en) * | 2020-09-23 | 2020-12-22 | 南京感动科技有限公司 | ETC portal system-based short-time flow prediction method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20090124197A (en) * | 2008-05-29 | 2009-12-03 | 주식회사 세인시스템 | Apparatus for forecasting traffic information with multi-detection and method for operating the same |
CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
CN102722982A (en) * | 2012-03-30 | 2012-10-10 | 上海市金山区青少年活动中心 | Background and inter-frame difference algorithm-based traffic flow and motion state detection method |
CN103606292A (en) * | 2013-11-13 | 2014-02-26 | 山西大学 | Intelligent navigator and realization method for path navigation thereof |
CN103903452A (en) * | 2014-03-11 | 2014-07-02 | 东南大学 | Traffic flow short time predicting method |
-
2014
- 2014-08-27 CN CN201410427899.5A patent/CN104183134B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20090124197A (en) * | 2008-05-29 | 2009-12-03 | 주식회사 세인시스템 | Apparatus for forecasting traffic information with multi-detection and method for operating the same |
CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
CN102722982A (en) * | 2012-03-30 | 2012-10-10 | 上海市金山区青少年活动中心 | Background and inter-frame difference algorithm-based traffic flow and motion state detection method |
CN103606292A (en) * | 2013-11-13 | 2014-02-26 | 山西大学 | Intelligent navigator and realization method for path navigation thereof |
CN103903452A (en) * | 2014-03-11 | 2014-07-02 | 东南大学 | Traffic flow short time predicting method |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105957329A (en) * | 2016-06-29 | 2016-09-21 | 肖锐 | Intelligentized information system for highway |
CN105957329B (en) * | 2016-06-29 | 2019-04-19 | 芜湖达成储运有限公司 | A kind of highway information intelligence system |
Also Published As
Publication number | Publication date |
---|---|
CN104183134A (en) | 2014-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104183134B (en) | The highway short-term traffic flow forecast method of vehicle is divided based on intelligence | |
CN104157139B (en) | A kind of traffic congestion Forecasting Methodology and method for visualizing | |
Kim et al. | Urban traffic flow prediction system using a multifactor pattern recognition model | |
CN104183119B (en) | Based on the anti-arithmetic for real-time traffic flow distribution forecasting method pushed away of section OD | |
CN102693633B (en) | Short-term traffic flow weighted combination prediction method | |
CN101794507B (en) | Method for evaluating macroscopic road network traffic state based on floating car data | |
CN102568205B (en) | Traffic parameter short-time prediction method based on empirical mode decomposition and classification combination prediction in abnormal state | |
CN109215350B (en) | Short-term traffic state prediction method based on RFID electronic license plate data | |
CN102610092A (en) | Urban road speed predication method based on RBF (radial basis function) neural network | |
CN103646542A (en) | Forecasting method and device for traffic impact ranges | |
CN103413443A (en) | Short-term traffic flow forecasting method based on hidden Markov model | |
CN104778835B (en) | High-grade road multi-bottleneck-point congestion evolution space-time range identification method | |
Theja et al. | Short term prediction of traffic parameters using support vector machines technique | |
CN102346964A (en) | Real-time jam prediction and intelligent management system for road traffic network area | |
CN103903433A (en) | Real-time dynamic judgment method and device for road traffic state | |
CN104484994A (en) | Urban road network traffic operation index evaluation method and device based on array radar | |
CN103745110A (en) | Method of estimating operational driving range of all-electric buses | |
CN106295888A (en) | A kind of public building based on measured data is joined and is built parking position and share time window and determine method | |
CN106887141A (en) | Continuous Traffic node congestion degree forecast model based on queueing theory, system and method | |
CN106898142A (en) | A kind of path forms time reliability degree calculation method of consideration section correlation | |
CN104050809A (en) | Method for automatically analyzing and predicting traffic condition | |
CN106548625B (en) | A kind of urban highway traffic situation combination forecasting method | |
Wang et al. | Using k-means clustering to identify time-of-day break points for traffic signal timing plans | |
Guo | Application of entropy weight method in the evaluation of the road capacity of open area | |
CN108053646A (en) | Traffic characteristic acquisition methods, Forecasting Methodology and system based on time-sensitive feature |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20201230 Address after: 402460 station No.14, no.6, 10th floor, innovation and development center, No.19 Lingfang Avenue, Changzhou street, Rongchang District, Chongqing Patentee after: Chongqing kezhiyuan Technology Co.,Ltd. Address before: 400030 No. 174 Sha Jie street, Shapingba District, Chongqing Patentee before: Chongqing University |
|
TR01 | Transfer of patent right |