CN106652443B - Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions - Google Patents
Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions Download PDFInfo
- Publication number
- CN106652443B CN106652443B CN201610916541.8A CN201610916541A CN106652443B CN 106652443 B CN106652443 B CN 106652443B CN 201610916541 A CN201610916541 A CN 201610916541A CN 106652443 B CN106652443 B CN 106652443B
- Authority
- CN
- China
- Prior art keywords
- data
- server
- vector
- year
- sequence
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 239000013598 vector Substances 0.000 claims description 66
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000012417 linear regression Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 abstract description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method for predicting short-time traffic volume of a highway with similar longitudinal and transverse dimensions, which is characterized in that adopted equipment comprises a server A, a server B, a server C and a computer; wherein, the server A is the existing system server, the server B is the data ETL server, and the server C is the data warehouse server; the original data in the server A is cleaned and converted according to corresponding rules through the server B and finally loaded into the server C, and the computer acquires the data through the report display system to display and analyze the data; the method improves seasonal anti-interference capability in traffic volume prediction, and enlarges the application range of the short-time traffic volume prediction method.
Description
Technical Field
The invention relates to a method for predicting short-time traffic volume of highways with similar longitudinal and transverse dimensions, and belongs to the technical field of computer data mining.
Background
At present, traffic prediction has been considered as a problem from different angles: such as time series problems, regression and function approximation problems, clustering or pattern recognition problems, and even all of the above. Cheng et al uses the spatio-temporal autocorrelation structure of the road network to build a spatio-temporal prediction model. Thomas et al performed short-term and long-term predictions using a single time series after extensive learning of 20 intersections in almiro, the netherlands. Qi et al propose an adaptive single exponential smoothing method to predict short-term traffic flow by optimizing an exponential smoothing coefficient through approximate dynamic programming. Jiang et al introduced a multiple linear regression minimum absolute shrinkage and selection operator method (Lasso) in combination with the nonlinear characteristics of Neural Networks (NN), and proposed a Lasso-NN combinatorial model. Xu et al propose a short-time traffic prediction Cluster-NN model of spatial clustering. Wang et al propose prediction methods based on k-nearest neighbor non-parametric regression. Xu et al propose an adaptive weight particle swarm neural network traffic flow prediction (PSOA-NN) model. Li et al propose a mode to predict short term traffic flow using a Support Vector Machine (SVM) with a time dependent structure. Castillo et al propose a traffic model adapted to random dynamic demands using a generalized Beta-Gaussian Bayesian network. Vlahogianni et al tested short term traffic using an autoregressive time series model in conjunction with a neural network.
In summary, in the conventional short-time traffic volume prediction, an operator is generally required to have a deeper understanding and mastering on the related knowledge, which is not beneficial to large-scale popularization and application.
Disclosure of Invention
The invention aims to provide a method for predicting short-time traffic volume of a highway with longitudinal and transverse similarity, which is used for predicting according to the similarity of traffic volume sequences and the recent similarity combined with a historical distribution rule. The traffic data is first calibrated for the number of weeks and days per week according to the rules of the week and day of the year, and the data is classified according to the calibrated days. And constructing traffic data of the same week number and the same day in different years as a longitudinal data sequence matrix, constructing traffic data of a week before a predicted day in the same year as a transverse data sequence matrix, respectively calculating average aggregate vectors of the longitudinal sequence and the transverse sequence, and obtaining a longitudinal sequence aggregate vector and a transverse sequence aggregate vector through weighted summation of the longitudinal sequence average aggregate vector sequence and the transverse sequence average aggregate vector sequence. Then, the total traffic volume of a single day is predicted, a distribution rule data sequence vector is obtained by combining the traffic volume distribution rule of the single day, and a final traffic volume prediction data sequence vector is determined according to the sum and average of the vertical sequence aggregate vector and the horizontal sequence aggregate vector and the distribution rule data sequence vector, so that the short-time traffic volume prediction process is completed; the method improves seasonal anti-interference capability in traffic volume prediction, and enlarges the application range of the short-time traffic volume prediction method.
The technical scheme of the invention is realized as follows: the method for predicting the short-time traffic volume of the expressway with the similar vertical and horizontal directions is characterized in that adopted equipment comprises a server 1, a server 2, a server 3 and a computer 1; wherein, the server 1 is the existing system server, the server 2 is the data ETL server, and the server 3 is the data warehouse server; the original data in the server 1 is cleaned and converted according to corresponding rules through the server 2 and finally loaded into the server 3, and the computer 1 acquires the data through a report display system for displaying and analyzing; the method comprises the following specific steps:
step 5, calculating the average resultant vector of the transverse sequenceObtained by weighted summation of the data of 7 days in the transverse data sequence matrix, wherein omega1+…+ω7Since the data on day 1 is similar to the data on the predicted day, a large weight ω is given1Equal to 0.7, the rest are equally distributed;
step 6, determining a longitudinal average resultant vectorSum of transverse average resultant vectorThe weight distribution of (2) determining the weighting coefficient by calculating the vector similarity coefficient, and setting the weighting coefficient of the longitudinal average resultant vector as c (0)<c<1) If the sum of the weighting coefficients is 1, the weighting coefficient of the horizontal average resultant vector is 1-c, using the following formulaSolving a value c;
step 7, calculating a longitudinal average resultant vector according to the weight calculated in the step 6Sum of transverse average resultant vectorThe vertical and horizontal sequence resultant vector obtained by weighted summation
Step 8, constructing a single-day total traffic volume sequence vector [ V ] by using the single-day traffic volume data summarized in the step 11,b,c,V2,b,c,...,Va-1,b,c]In which V isa,b,cRepresents an overall traffic volume on day c in week b of year a, where a is 1, 2, …, n, representing the first year from the earliest year, and b is 1, 2, …, n, representing the week a; c-1, 2, 3, 4, 5, 6, 7 denotes the day after week b of a year;
and 9, planning and solving the vector in the step 8 by using a least square method in linear regression to obtain a next solving result Va,b,c;
Step 10, analyzing a plurality of groups of half-hour traffic data to obtain that the traffic distribution of each day is basically consistent, and obtaining a traffic distribution law according to the proportion of the traffic of each time period in the whole;
step 11, solving result V obtained in step 9a,b,cCalculating a distribution rule data sequence vector according to the distribution law obtained in the step 10
Step 12, integrating the vertical and horizontal sequences obtained in the step 7And the distribution rule data sequence vector obtained in step 11Superposing according to the sum and average to obtain the final traffic prediction data sequence vector fa,b,c,k;
By utilizing the steps, the prediction data of the short-time traffic volume of the expressway can be obtained.
The invention has the advantages that the prediction is carried out according to the synchronous similarity and the recent similarity of the traffic volume sequence and the historical distribution rule, so that the seasonal interference in the traffic volume prediction is greatly reduced; the advantages are that:
1. the method for predicting the short-time traffic volume of the expressway utilizes historical synchronization data to predict, and can greatly reduce the influence of seasonal factors of the traffic volume.
2. The prediction method of the short-time traffic volume of the expressway can better enable the prediction result to accord with the actual situation by combining the half-hour segmented distribution rule.
3. The prediction method for the short-time traffic volume of the expressway analyzes and researches the synchronous similarity and the recent similarity of the traffic volume sequence, and has application continuity.
Drawings
FIG. 1 is a diagram of the structure of equipment required for a method for predicting short-term traffic volume on a highway with similar vertical and horizontal directions
Detailed Description
The invention is further described with reference to the accompanying drawings in which: as shown in fig. 1, the method for predicting short-term traffic volume of highways with similar vertical and horizontal directions adopts equipment comprising a server 1, a server 2, a server 3 and a computer 1; the server 1 is an existing system server, the server 2 is a data ETL server, and the server 3 is a data warehouse server; the original data in the server 1 is cleaned and converted according to corresponding rules through the server 2 and finally loaded into the server 3, and the computer 1 acquires the data through a report display system for displaying and analyzing; the method comprises the following specific steps:
step 5, calculating the average resultant vector of the transverse sequenceObtained by weighted summation of the data of 7 days in the transverse data sequence matrix, wherein omega1+…+ω7Since the data on day 1 is similar to the data on the predicted day, a large weight ω is given1Equal to 0.7, the rest are equally distributed;
step 6, determining a longitudinal average resultant vectorSum of transverse average resultant vectorThe weight distribution of (2) determining the weighting coefficient by calculating the vector similarity coefficient, and setting the weighting coefficient of the longitudinal average resultant vector as c (0)<c<1) If the sum of the weighting coefficients is 1, the weighting coefficient of the horizontal average resultant vector is 1-c, using the following formulaSolving a value c;
step 7, calculating a longitudinal average resultant vector according to the weight calculated in the step 6Sum of transverse average resultant vectorThe vertical and horizontal sequence resultant vector obtained by weighted summation
Step 8, constructing a single-day total traffic volume sequence vector [ V ] by using the single-day traffic volume data summarized in the step 11,b,c,V2,b,c,...,Va-1,b,c]In which V isa,b,cRepresents an overall traffic volume on day c in week b of year a, where a is 1, 2, …, n, representing the first year from the earliest year, and b is 1, 2, …, n, representing the week a; c-1, 2, 3, 4, 5, 6, 7 denotes the day after week b of a year;
and 9, planning and solving the vector in the step 8 by using a least square method in linear regression to obtain a next solving result Va,b,c;
Step 10, analyzing a plurality of groups of half-hour traffic data to obtain that the traffic distribution of each day is basically consistent, and obtaining a traffic distribution law according to the proportion of the traffic of each time period in the whole;
step 11, solving result V obtained in step 9a,b,cCalculating a distribution rule data sequence vector according to the distribution law obtained in the step 10
Step 12, integrating the vertical and horizontal sequences obtained in the step 7And the distribution rule data sequence vector obtained in step 11Superposing according to the sum and average to obtain the final traffic prediction data sequence vector fa,b,c,k;
By utilizing the steps, the prediction data of the short-time traffic volume of the expressway can be obtained.
Claims (1)
1. The method for predicting the short-time traffic volume of the expressway with the similar vertical and horizontal directions is characterized in that adopted equipment comprises a server 1, a server 2, a server 3 and a computer 1; wherein, the server 1 is the existing system server, the server 2 is the data ETL server, and the server 3 is the data warehouse server; the original data in the server 1 is cleaned and converted according to corresponding rules through the server 2 and finally loaded into the server 3, and the computer 1 acquires the data through a report display system for displaying and analyzing; the method comprises the following specific steps:
step 1, collecting and counting half-hour traffic volume and single-day traffic volume data of highway charging data according to half-hour segmentation and date;
step 2, according to half-smallThe time traffic data constructs a longitudinal data sequence matrix as f1,b,c,k,f2,b,c,k, …,fa-1,b,c,k]T Wherein f isa,b,c,kRepresenting a traffic sequence vector, a =1, 2, …, n representing the first year from the first year, b =1, 2, …, m representing the week of the year a, c =1, 2, 3, 4, 5, 6, 7 representing the day of the week b of the year a, k =1, 2, …, 48 representing the dimension, i.e. the data, of the sequence vector at day c of the week b of the year a;
step 3, dividing the longitudinal data sequence matrix in the step 2 into k sequences according to different k values, planning and solving each sequence by using a least square method in linear regression respectively to obtain a next solving result, and combining the k results into a longitudinal sequence average resultant vector from small to large according to the k valuesWherein 1_ a-1 denotes the 1 st to a-1 st years from the first year;
step 4, constructing a transverse data sequence matrix [ f ]a,b,c-7,k,fa,b,c-6,k, … ,fa,b,c-1,k]TSelecting data of a week before a predicted day when a transverse data sequence matrix is constructed;
step 5, calculating the average resultant vector of the transverse sequenceAnd c-7_ c-1 represents the sum of the data weighted by the number of 7 days in the transverse data sequence matrix from the c-7 th week to the c-1 th week, wherein ω is1+ … + ω7Since the data on day 1 is similar to the data on the predicted day, a large weight ω is given1Equal to 0.7, the rest are equally distributed;
step 6, determining a longitudinal average resultant vectorSum of transverse average resultant vectorThe weight distribution of the vector is determined by calculating the vector similarity coefficient, the total weight of the influence of the vertical and horizontal sequence synthetic vectors on the prediction is 1, the influence weight of the vertical average synthetic vector is c, the influence weight of the horizontal average synthetic vector is 1-c, and the following formula is utilizedSolving a value c;
step 7, calculating a longitudinal average resultant vector according to the weight calculated in the step 6Sum of transverse average resultant vectorThe vertical and horizontal sequence resultant vector obtained by weighted summationThe sum vector is obtained by fusing a longitudinal traffic flow average sum vector calculated based on the c week of the b month from the first 1 st year to the a-1 st year with a transverse traffic flow average sum vector calculated based on the c-7 week to the c-1 week of the b month of the a year;
step 8, constructing a single-day total traffic volume sequence vector using the single-day traffic volume data summarized in step 1V 1,b,c ,V 2,b,c , ... ,V a-1,b,c ]WhereinV a,b,c Represents a total traffic volume representing day c in week b of a year, where a =1, 2, …, n, represents the year from the first year, b =1, 2, …, m, represents the week of a year; c =1, 2, 3, 4, 5, 6, 7 denotes the day after b weeks of a year;
and 9, planning and solving the vector in the step 8 by using a least square method in linear regression to obtain the next solving resultV a,b,c ;
Step 10, analyzing a plurality of groups of half-hour traffic data to obtain that the traffic distribution of each day is basically consistent, and obtaining a traffic distribution law according to the proportion of the traffic of each time period in the whole;
step 11, solving results obtained in step 9V a,b,c Calculating a distribution rule data sequence vector according to the distribution law obtained in the step 10;
Step 12, integrating the vertical and horizontal sequences obtained in the step 7And the distribution rule data sequence vector obtained in step 11Superposing according to the addition and the average to obtain the final traffic prediction data sequence vectorf a,b,c,k ;
By utilizing the steps, the prediction data of the short-time traffic volume of the expressway can be obtained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610916541.8A CN106652443B (en) | 2016-10-21 | 2016-10-21 | Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610916541.8A CN106652443B (en) | 2016-10-21 | 2016-10-21 | Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106652443A CN106652443A (en) | 2017-05-10 |
CN106652443B true CN106652443B (en) | 2020-07-07 |
Family
ID=58855474
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610916541.8A Expired - Fee Related CN106652443B (en) | 2016-10-21 | 2016-10-21 | Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106652443B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110517479B (en) * | 2018-05-22 | 2020-11-03 | 杭州海康威视系统技术有限公司 | Urban road traffic prediction method and device and electronic equipment |
CN111335876A (en) * | 2020-03-02 | 2020-06-26 | 北京四利通控制技术股份有限公司 | Self-adaptive tracking prediction control method for petroleum drilling well track |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
CN105046953A (en) * | 2015-06-18 | 2015-11-11 | 南京信息工程大学 | Short-time traffic-flow combination prediction method |
CN105551251A (en) * | 2016-01-19 | 2016-05-04 | 华南理工大学 | No-signalized-intersection motor vehicle conflict probability determining method |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9159228B2 (en) * | 2012-11-26 | 2015-10-13 | Xerox Corporation | System and method for estimation of available parking space through intersection traffic counting |
US9183743B2 (en) * | 2013-10-31 | 2015-11-10 | Bayerische Motoren Werke Aktiengesellschaft | Systems and methods for estimating traffic signal information |
-
2016
- 2016-10-21 CN CN201610916541.8A patent/CN106652443B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
CN105046953A (en) * | 2015-06-18 | 2015-11-11 | 南京信息工程大学 | Short-time traffic-flow combination prediction method |
CN105551251A (en) * | 2016-01-19 | 2016-05-04 | 华南理工大学 | No-signalized-intersection motor vehicle conflict probability determining method |
Also Published As
Publication number | Publication date |
---|---|
CN106652443A (en) | 2017-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107180530B (en) | A kind of road network trend prediction method based on depth space-time convolution loop network | |
Li et al. | Prediction for tourism flow based on LSTM neural network | |
CN110570651B (en) | Road network traffic situation prediction method and system based on deep learning | |
CN106981198B (en) | Deep learning network model for travel time prediction and establishing method thereof | |
CN109285346B (en) | Urban road network traffic state prediction method based on key road sections | |
Fu et al. | Using LSTM and GRU neural network methods for traffic flow prediction | |
CN113313303A (en) | Urban area road network traffic flow prediction method and system based on hybrid deep learning model | |
CN108205889B (en) | Method for predicting highway traffic flow based on convolutional neural network | |
Panahi et al. | Forecasting plug-in electric vehicles load profile using artificial neural networks | |
CN105702029A (en) | Express way traffic state prediction method taking spatial-temporal correlation into account at different times | |
CN111242395B (en) | Method and device for constructing prediction model for OD (origin-destination) data | |
CN110164129B (en) | Single-intersection multi-lane traffic flow prediction method based on GERNN | |
CN109840587A (en) | Reservoir reservoir inflow prediction technique based on deep learning | |
CN108256724B (en) | Power distribution network open capacity planning method based on dynamic industry coefficient | |
Cordova et al. | Combined electricity and traffic short-term load forecasting using bundled causality engine | |
Xuemei et al. | Particle swarm optimization-based LS-SVM for building cooling load prediction | |
CN115204478A (en) | Public traffic flow prediction method combining urban interest points and space-time causal relationship | |
CN107665376A (en) | A kind of Wetland Space changes in distribution framework analogue and Forecasting Methodology | |
CN111178585A (en) | Fault reporting amount prediction method based on multi-algorithm model fusion | |
Zhuang et al. | Long-lead prediction of extreme precipitation cluster via a spatiotemporal convolutional neural network | |
CN106652443B (en) | Method for predicting short-time traffic volume of expressway with similar longitudinal and transverse dimensions | |
CN109377761A (en) | Traffic factor network establishing method based on Markov-chain model | |
CN105913654B (en) | A kind of Intelligent traffic management systems | |
CN104299114A (en) | Estate information analysis system based on databank | |
Naheliya et al. | A hybrid deep learning method for short-term traffic flow forecasting: GSA-LSTM |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200707 Termination date: 20201021 |
|
CF01 | Termination of patent right due to non-payment of annual fee |