CN109727455A - A kind of processing method of traffic information - Google Patents
A kind of processing method of traffic information Download PDFInfo
- Publication number
- CN109727455A CN109727455A CN201910163145.6A CN201910163145A CN109727455A CN 109727455 A CN109727455 A CN 109727455A CN 201910163145 A CN201910163145 A CN 201910163145A CN 109727455 A CN109727455 A CN 109727455A
- Authority
- CN
- China
- Prior art keywords
- model
- traffic information
- sequence
- zero
- mean
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Traffic Control Systems (AREA)
Abstract
The present invention relates to a kind of processing methods of traffic information, belong to technical field of information processing.The following steps are included: S1 acquisition of information: acquiring the sample sequence of traffic information in the form of time series；S2 tranquilization processing: sample sequence is converted to zero-mean sequence, stationarity judgement is carried out to newly-generated zero-mean sequence using nonparametric method, if newly-generated zero-mean sequence is unstable, difference is carried out to newly-generated zero-mean sequence, and so on until steady；S3 statistical procedures: it is calculated tranquilization treated the auto-correlation function and deviation―related function of zero-mean sequence；The identification of S4 model: MA (q) model, AR (p) model or ARIMA model are used according to auto-correlation function and deviation―related function judgement；S5 model order: rank is determined using AIC progress model, ideal order is obtained when AIC value reaches minimum；S6 obtains predicted value: predicted value is calculated using corresponding model according to ideal order.
Description
Technical field
The present invention relates to technical field of information processing, more particularly to a kind of processing method of traffic information.
Background technique
There is the urban road traffic network of oneself in each city, and people are by selecting different routes that can quickly reach
The destination of oneself.However, many times due to traffic congestion, if still according to previous traffic path, user will
It is difficult to reach the destination of oneself in regulation duration.It is well known that urban road congestion is a kind of dynamic congestion, Er Feijing
The congestion of state.This also means that the degree of road congestion changes with the variation of duration.
In the prior art by being monitored to Traffic Information, obtain past and present traffic information to future into
Row prediction.But this mode is in the case where vehicle flowrate changes at random, can not comprehensively, calculate to a nicety to obtain road congestion
Situation.
Summary of the invention
This patent provides a kind of processing method of traffic information, and this method modeling is simple, is readily appreciated that, abundant in data
In the case where, there is higher precision of prediction.The scheme is as follows:
The embodiment of the invention provides a kind of processing methods of traffic information, method includes the following steps:
S1 acquisition of information: acquiring the sample sequence of traffic information in the form of time series, and the traffic information includes handing over
Through-flow flow, density or speed；
S2 tranquilization processing: sample sequence is converted to zero-mean sequence, using nonparametric method to newly-generated zero-mean sequence
Stationarity judgement is carried out, if newly-generated zero-mean sequence is unstable, difference is carried out to newly-generated zero-mean sequence, with
This analogizes until steady；
S3 statistical procedures: it is calculated tranquilization treated the auto-correlation function and deviation―related function of zero-mean sequence；
The identification of S4 model: MA (q) model, AR (p) model or ARIMA mould are used according to auto-correlation function and deviation―related function judgement
Type；
S5 model order: rank is determined using minimum information criterion AIC progress model, ideal rank is obtained when AIC value reaches minimum
Number；
S6 obtains predicted value: predicted value is calculated using corresponding model according to ideal order.
Further, method provided by the invention further include:
S7 verification: obtaining actual value, be compared with predicted value, if error does not meet pre-provisioning request, S3 pairs of return step
The parameter of model is fitted.
Wherein, zero-mean sequence YtIt is obtained using following methods:
,
Wherein, xtFor sample sequence, mean value:。
Wherein, auto-correlation function are as follows:
,
Wherein, covariance:,For zero-mean sequence YtMean value.
Wherein, deviation―related function are as follows:
。
Specifically, if only Autocorrelation Detection q not truncation, MA (q) model is used；If only partial correlation detection p is not cut
Tail then uses AR (p) model；If auto-correlation and partial correlation detect not truncation, ARIMA model is used.
Wherein, AIC are as follows:
,
Wherein, N is the length of sample sequence,For the variance of residual error.
Technical solution provided in an embodiment of the present invention has the beneficial effect that the present invention by the historical data to time series
This rule is extended to future by the rule changed over time, so that the future to the phenomenon makes prediction.Its it is specific using from
Seek its statistical law and statistical property in a large amount of measured data sequence；It is predicted especially by model, including model
Type estimates the processes such as the parameter of model and the order of determining model.This method modeling is simple, is readily appreciated that, abundant in data
In the case where, there is higher precision of prediction.
Detailed description of the invention
Fig. 1 is the flow diagram of the processing method of traffic information provided in an embodiment of the present invention.
Specific embodiment
With reference to embodiment, the embodiment of the present invention is furthur described in detail.Following embodiment is used for
Illustrate the present invention, but is not intended to limit the scope of the invention.
Referring to Fig. 1, the embodiment of the invention provides a kind of processing methods of traffic information, method includes the following steps:
S1 acquisition of information: the sample sequence of traffic information is acquired in the form of time series from local or server
(preferably with the sequencing natural ordering of time)；Wherein, traffic information be save historical data, including traffic flow flow,
Density or speed etc..
S2 tranquilization processing: sample sequence is converted to zero-mean sequence, using nonparametric method to newly-generated zero-mean
Sequence carries out stationarity judgement, if newly-generated zero-mean sequence is unstable, it is poor to carry out to newly-generated zero-mean sequence
Point, and so on until steady.
S3 statistical procedures: it is calculated tranquilization treated the auto-correlation function and partial correlation letter of zero-mean sequence
Number, the process are handled using classical statistical.
S4 model identification: according to auto-correlation function and deviation―related function judgement using MA (q) model, AR (p) model or
ARIMA model, MA (q) model, AR (p) model or ARIMA model are existing model, be can be used directly.
S5 model order: using minimum information criterion AIC carry out model determine rank (while use auto-correlation function and inclined phase
Close function), ideal order is obtained when AIC value reaches minimum.
Above step can obtain the order of corresponding model, model parameter and model, carry out standard for the calculating of predicted value
It is standby.
S6 obtains predicted value: predicted value is calculated using corresponding model according to ideal order, is such as put into time value
Predicted value is calculated in corresponding model.If step S4 judgement use MA (q) model, then the step using MA (q) model come
Predicted value is calculated.
Further, method provided by the invention further include:
S7 verification: obtain actual value (a small amount of), such as obtain (be manually entered, the data of invoking server or acquisition real value) some
The actual value at time point or some period；It is compared with predicted value (corresponding time point or period), if error is not
Meeting pre-provisioning request, (error is designed according to actual requirement, is such as set difference and is no more than 20%), then return step S3 is to model
Parameter be fitted；If error meets pre-provisioning request, corresponding model can be used, traffic information is predicted.
Wherein, zero-mean sequence YtIt is obtained using following methods:
,
Wherein, xtFor sample sequence, mean value:。
Wherein, auto-correlation function are as follows:
,
Wherein, covariance:,For zero-mean sequence YtIt is equal
Value.
Wherein, deviation―related function are as follows:
。
Specifically, if only Autocorrelation Detection q not truncation, MA (q) model is used；If only partial correlation detection p is not cut
Tail then uses AR (p) model；If auto-correlation and partial correlation detect not truncation, ARIMA model is used.
Wherein, AIC are as follows:
,
Wherein, N is the length of sample sequence,For the variance of residual error.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (7)
1. a kind of processing method of traffic information, which comprises the following steps:
S1 acquisition of information: acquiring the sample sequence of traffic information in the form of time series, and the traffic information includes handing over
Through-flow flow, density or speed；
S2 tranquilization processing: sample sequence is converted to zero-mean sequence, using nonparametric method to newly-generated zero-mean sequence
Stationarity judgement is carried out, if newly-generated zero-mean sequence is unstable, difference is carried out to newly-generated zero-mean sequence, with
This analogizes until steady；
S3 statistical procedures: it is calculated tranquilization treated the auto-correlation function and deviation―related function of zero-mean sequence；
The identification of S4 model: MA (q) model, AR (p) model or ARIMA model are used according to auto-correlation function and deviation―related function judgement；
S5 model order: rank is determined using minimum information criterion AIC progress model, ideal order is obtained when AIC value reaches minimum；
S6 obtains predicted value: predicted value is calculated using corresponding model according to ideal order.
2. the processing method of traffic information according to claim 1, which is characterized in that the method also includes:
S7 verification: obtaining actual value, be compared with predicted value, if error does not meet pre-provisioning request, S3 pairs of return step
The parameter of model is fitted.
3. the processing method of traffic information according to claim 1, which is characterized in that zero-mean sequence YtIt uses with lower section
Method is obtained:
,
Wherein, xtFor sample sequence, mean value:。
4. the processing method of traffic information according to claim 3, which is characterized in that auto-correlation function are as follows:
,
Wherein, covariance:,For zero-mean sequence YtMean value.
5. the processing method of traffic information according to claim 4, which is characterized in that deviation―related function are as follows:
。
6. the processing method of traffic information according to claim 5, which is characterized in that if only Autocorrelation Detection q is not cut
Tail then uses MA (q) model；If only partial correlation detects p not truncation, AR (p) model is used；If auto-correlation and partial correlation
Not truncation is detected, then uses ARIMA model.
7. the processing method of traffic information according to claim 5, which is characterized in that AIC are as follows:
,
Wherein, N is the length of sample sequence,For the variance of residual error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910163145.6A CN109727455A (en) | 2019-03-05 | 2019-03-05 | A kind of processing method of traffic information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910163145.6A CN109727455A (en) | 2019-03-05 | 2019-03-05 | A kind of processing method of traffic information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109727455A true CN109727455A (en) | 2019-05-07 |
Family
ID=66300143
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910163145.6A Pending CN109727455A (en) | 2019-03-05 | 2019-03-05 | A kind of processing method of traffic information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109727455A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5884037A (en) * | 1996-10-21 | 1999-03-16 | International Business Machines Corporation | System for allocation of network resources using an autoregressive integrated moving average method |
| CN103685014A (en) * | 2013-11-29 | 2014-03-26 | 武汉日电光通信工业有限公司 | System and method for strengthening reliability of routing computation by using time series prediction model |
| CN103903452A (en) * | 2014-03-11 | 2014-07-02 | 东南大学 | Traffic flow short time predicting method |
| CN105703954A (en) * | 2016-03-17 | 2016-06-22 | 福州大学 | Network data flow prediction method based on ARIMA model |
| CN107293118A (en) * | 2017-07-18 | 2017-10-24 | 扬州大学 | A kind of traffic speed motion interval Forecasting Approach for Short-term |
| CN107967803A (en) * | 2017-11-17 | 2018-04-27 | 东南大学 | Traffic congestion Forecasting Methodology based on multi-source data and variable-weight combined forecasting model |
| CN108877223A (en) * | 2018-07-13 | 2018-11-23 | 南京理工大学 | A kind of Short-time Traffic Flow Forecasting Methods based on temporal correlation |
-
2019
- 2019-03-05 CN CN201910163145.6A patent/CN109727455A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5884037A (en) * | 1996-10-21 | 1999-03-16 | International Business Machines Corporation | System for allocation of network resources using an autoregressive integrated moving average method |
| CN103685014A (en) * | 2013-11-29 | 2014-03-26 | 武汉日电光通信工业有限公司 | System and method for strengthening reliability of routing computation by using time series prediction model |
| CN103903452A (en) * | 2014-03-11 | 2014-07-02 | 东南大学 | Traffic flow short time predicting method |
| CN105703954A (en) * | 2016-03-17 | 2016-06-22 | 福州大学 | Network data flow prediction method based on ARIMA model |
| CN107293118A (en) * | 2017-07-18 | 2017-10-24 | 扬州大学 | A kind of traffic speed motion interval Forecasting Approach for Short-term |
| CN107967803A (en) * | 2017-11-17 | 2018-04-27 | 东南大学 | Traffic congestion Forecasting Methodology based on multi-source data and variable-weight combined forecasting model |
| CN108877223A (en) * | 2018-07-13 | 2018-11-23 | 南京理工大学 | A kind of Short-time Traffic Flow Forecasting Methods based on temporal correlation |
Non-Patent Citations (2)
| Title |
|---|
| 戎亚萍: "《基于路况数据的交通流预测模型及其对比分析》", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 林剑辉，苏洋，施振浩: "《基于ARIMA模型的道路交通流量预测方法及其应用》", 《PROCEEDINGS OF 2012 2ND INTERNATIONAL CONFERENCE ON AEROSPACE ENGINEERING AND INFORMATION TECHNOLOGY(AEIT 2012 V2)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106503840B (en) | Available parking space prediction method and system for parking lot | |
| CN111104981B (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
| CN104994539B (en) | A kind of wireless sensor network Traffic anomaly detection method based on ARIMA models | |
| Feng et al. | Vehicle trajectory reconstruction using automatic vehicle identification and traffic count data | |
| CN108108844A (en) | A kind of urban human method for predicting and system | |
| CN104504099B (en) | Traffic trip state cutting method based on location track | |
| CN109377752A (en) | Short-term traffic flow variation prediction method, apparatus, computer equipment and storage medium | |
| Zhang et al. | Bus dwell time estimation and prediction: a study case in shanghai-china | |
| Van Hinsbergen et al. | Bayesian neural networks for the prediction of stochastic travel times in urban networks | |
| EP2350819A2 (en) | Determining user similarities based on location histories | |
| CN109379240A (en) | Car networking flux prediction model construction method, device and electronic equipment | |
| CN110047291A (en) | A kind of Short-time Traffic Flow Forecasting Methods considering diffusion process | |
| CN110598775B (en) | Prediction method, system and storage medium based on fuzzy clustering and BP neural network | |
| Deb Nath et al. | Modified K-means clustering for travel time prediction based on historical traffic data | |
| Świąder et al. | Spatial-temporal diversification of poverty in Wroclaw | |
| CN108415884A (en) | A kind of modal parameters real-time tracing method | |
| Sevlian et al. | Travel time estimation using floating car data | |
| WO2016123867A1 (en) | Method and device for position search cognition | |
| CN111380530A (en) | Navigation method and related product | |
| Khetarpaul et al. | Bus arrival time prediction using a modified amalgamation of fuzzy clustering and neural network on spatio-temporal data | |
| Hosseini et al. | Traffic flow prediction using MI algorithm and considering noisy and data loss conditions: An application to Minnesota traffic flow prediction | |
| CN107193824B (en) | Abnormal data detection method and device | |
| CN109727455A (en) | A kind of processing method of traffic information | |
| CN107133636B (en) | Method and system for obtaining similar typhoons | |
| Pinelli et al. | Robust bus-stop identification and denoising methodology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190507 |
|
| RJ01 | Rejection of invention patent application after publication |