CN108665708A - A kind of urban traffic flow imbalance mode excavation method and system - Google Patents
A kind of urban traffic flow imbalance mode excavation method and system Download PDFInfo
- Publication number
- CN108665708A CN108665708A CN201810510479.1A CN201810510479A CN108665708A CN 108665708 A CN108665708 A CN 108665708A CN 201810510479 A CN201810510479 A CN 201810510479A CN 108665708 A CN108665708 A CN 108665708A
- Authority
- CN
- China
- Prior art keywords
- pattern
- window
- indicate
- traffic flow
- isochronous surface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
Abstract
The invention discloses a kind of urban traffic flow imbalance mode excavation method and system, the method includes：First, the correlation model between road net data and track data is established；In turn, a kind of statistical significance of linear scan statistical method evaluation traffic flow imbalance pattern is proposed；Finally, the dynamic evolution for finding imbalance of flow pattern, that is, the pattern that takes place frequently and accidental pattern are calculated based on pattern similarity.The present invention finds network neighborhood using line segment extended method, avoids calculating the network distance between point, to improve algorithm performs efficiency.Linear scan statistical method proposed by the present invention is influenced the differentiation result of uneven pattern by scanning window size influence that is smaller, therefore reducing the subjectivity of conventional method artificial settings threshold value.Apply the present invention to actual traffic data, the producing cause of uneven pattern is analyzed with urban facilities in conjunction with urban function region, certain reference can be provided for Urban Traffic and traffic programme.
Description
Technical field
The present invention relates to intelligent transportation fields, more particularly, it is related to a kind of urban traffic flow imbalance mode excavation
Method and system.
Background technology
The scientifically and rationally detection evaluation unbalanced degree of urban road traffic state, contributes to traffic administration person to fully understand road
Road traffic noise prediction deeply excavates potential path resource, is equilibrium assignment path resource, and induction traffic flow rationally runs offer
Data supporting and decision-making foundation.
The assessment of traffic behavior is the characterization traffic flow operation degree of crowding, implements the basis of traffic administration and control measure.
So far from the 1980s, urban road traffic state detection discrimination technology achieves many progress, and domestic and foreign scholars propose
A large amount of characteristic index, descriptive model and computational methods, as Assessment of Serviceability of Roads and road crowding/regional traffic state are sentenced
Other model and regional traffic state space-time hierarchical mode etc..But these indexs and model are directed to single section, intersection mostly
Or the detection and differentiation of the road grid traffic degree of crowding, and ignore the inspection of entire road net traffic state disequilibrium, unbalanced degree
It surveys and evaluates.The lack of uniformity of road network traffic flow causes different zones traffic congestion totally different, although overall traffic circulation shape
State is identical, but the traffic behavior otherness in each region is larger.Therefore, it in order to preferably reflect road grid traffic operating status, excavates
The path resource of non-congestion regions quickly dredges congestion regions traffic flow, preferably guides urban highway traffic resources balance
Development, there is an urgent need to obtain the unbalanced degree information of urban road traffic state, establishes the unbalanced degree of urban road traffic state
Method is detected and determined.
Unbalanced traffic stream refers in interval at a fixed time, and the magnitude of traffic flow for flowing into a region is significantly higher than or significantly
Less than the magnitude of traffic flow flowed out from the region.The methods for excavating unbalanced traffic stream mode most of at present are based on theorem in Euclid space
It is assumed that determining the conspicuousness of unbalanced traffic stream using self-defined threshold value.However, traffic flow is strictly by the pact of road network
Beam, and it is difficult to a suitable threshold value is determined to assess the conspicuousness of uneven pattern.
Invention content
In order to solve the defect present in the prior art, embodiment of the present invention provides a kind of urban traffic flow imbalance mould
Formula method for digging and system can accurately and reliably assess the conspicuousness of traffic flow imbalance pattern.
On the one hand, embodiment of the present invention provides a kind of urban traffic flow imbalance mode excavation method, including：
S1, road net data and track data are obtained；
S2, correlation model between the road net data and the track data is established；
S3, the statistical significance that traffic flow imbalance pattern is evaluated using linear scan statistical method；
S4, the dynamic evolution for obtaining the traffic flow imbalance pattern is calculated based on pattern similarity.
Correspondingly, the embodiment of the present invention also provides a kind of urban traffic flow imbalance mode excavation method and system, including：
Data acquisition module, for obtaining road net data and track data；
Correlation model establishes module, for establishing the correlation model between the road net data and the track data；
Statistical module, for the statistical significance using linear scan statistical method evaluation traffic flow imbalance pattern；
Computing module, for calculating the dynamic evolution for obtaining the traffic flow imbalance pattern based on pattern similarity.
Beneficial effects of the present invention are as follows：
The present invention proposes urban traffic flow imbalance mode excavation method, compensates for the deficiencies in the prior art, contributes to comprehensively
Detection and evaluation urban highway traffic operation conditions, in conjunction with urban function region with urban facilities to the producing cause of uneven pattern
It is analyzed, certain reference can be provided for Urban Traffic and traffic programme；
The present invention is added to road network the characteristics of considering track data in excavating urban traffic flow imbalance mode process
Constraint, with cyberspace replace theorem in Euclid space, the mould that can be omitted in theorem in Euclid space imbalance mode excavation method can be excavated
Formula so that Result is more accurate and reliable；
The present invention finds network neighborhood using line segment extended method, avoids calculating the network distance between point, be calculated to improve
Method execution efficiency；
Linear scan statistical method proposed by the present invention to the differentiation result of uneven pattern by scanning window size influenced compared with
It is small, therefore reduce the influence that the subjectivity of threshold value is manually set in conventional method；
The present invention analyzes traffic flow imbalance pattern after excavating traffic flow imbalance pattern, and based on pattern similarity
Dynamic evolution pattern, more existing imbalance mode excavation method still further, can provide for the space-time abnormality detection in city
Scientific guidance.
Description of the drawings
Fig. 1 is the flow diagram of urban traffic flow imbalance mode excavation method provided in an embodiment of the present invention；
Fig. 2 is the schematic diagram of urban traffic flow imbalance mode excavation system provided in an embodiment of the present invention；
Fig. 3 is provided in an embodiment of the present invention using P as window center point, using 2Z as the linear scan window of window size；
Fig. 4 is the flow chart provided in an embodiment of the present invention clustered based on density expansion algorithm；
Fig. 5（a）It is the schematic diagram of the pattern provided in an embodiment of the present invention that takes place frequently,（b）Be it is provided in an embodiment of the present invention the first
The schematic diagram of accidental pattern,（c）It is the schematic diagram of second provided in an embodiment of the present invention accidental pattern.
Specific implementation mode
It is described in detail to various aspects of the present invention below in conjunction with the drawings and specific embodiments.Wherein, many institute's weeks
Module, unit and its mutual connection, link, communication or the operation known are not shown or do not elaborate.Also, institute
Feature, framework or the function of description can in any way combine in one or more embodiments.People in the art
Member is it should be appreciated that following various embodiments are served only for the protection domain for example, and is not intended to limit the present invention.May be used also
To be readily appreciated that, module or unit or step in each embodiment described herein and shown in the drawings can be matched by various differences
It sets and is combined and designs.
Fig. 1 is a kind of flow diagram of urban traffic flow imbalance mode excavation method according to the ... of the embodiment of the present invention.
Referring to Fig.1, the method includes：
S1：Obtain road net data and track data.
S2：Establish the correlation model between the road net data and the track data.
S3：The statistical significance of traffic flow imbalance pattern is evaluated using linear scan statistical method.
S4：Calculate the dynamic evolution for obtaining the traffic flow imbalance pattern based on pattern similarity, that is, the pattern that takes place frequently with
Accidental pattern.
Road net data in above-mentioned steps S1 is the multistage road vectors data acquired from OpenStreetMap；Rail
Mark data be vehicle GPS at regular intervals（90 seconds）Acquire the vehicle information data once obtained.The present invention needs first from original
The beginning and end data of all vehicles are extracted in beginning GPS data, specific extracting method is as follows：
GPS point under passenger carrying status with same vehicles identifications is identified as same group of track data, if in entire data set
Shared S different vehicle ID, then can obtain S groups track Ri(i=1,2,…,S)；
Every group of track data that upper step is identified is arranged according to the acquisition time ascending order of GPS point, from each group of track Ri(i=1,
2 ..., S) in extraction beginning and end method be：From RiFirst sampled point P1Start, judges RiMiddle P1Latter point P2's
Sampling time and P1Sampling time interval whether be more than initial data temporal resolution（90 seconds）If being not more than, P is indicated1With
P2On same carrying track；If more than expression P1With P2Not on same carrying track, P1End as a track
Point, P2Starting point as another track；
To RiEach point in (i=1,2 ..., S) executes the operation in above-mentioned steps, finally obtains starting point all in track
And endpoint data.
Above-mentioned steps S2 is specifically included：
By in road network at wired online node, intersection interrupt as straight line, every straight line indicates the side of network, directly
The endpoint of line is network node, to the network structure of one specification of structure.
Each edge in network has both ends point coordinates A1(X1,Y1), A2(X2,Y2), use mathematical method by two point coordinates
Find out the linear equation of every straight line：y=ki*X+bi(i=1,2 ..., N (N is straight line number)).
Each starting point O that step S1 is extractedj(j=1,2 ..., M) and terminal Dk(k=1,2,…,M)（M is beginning or end
Number）There is its corresponding coordinate, judges that the method for the straight line belonging to a tracing point P (x, y) is：By the x coordinate minute of P points
Every linear equation is not substituted into obtains yi(i=1,2 ..., N), finds out yiStraight line with the y-coordinate difference minimum of P points is P points institute
The straight line of category.It can determine whether the straight line belonging to all beginning and ends in this approach.
Sentence the storage relationship for needing to establish after the straight line belonging to section beginning and end between road network and tracing point.This hair
The storage organization table of the bright storage organization table and track data for establishing road net data respectively.The wherein storage organization of road net data
Table（Table 1）Every record indicate a network edge, include side ID mark, the length on side, rise node ID, rise a node connected
Other all sides ID, caudal knot point ID, the ID on other all sides that caudal knot point is connected, the ID of all tracing points on side；
The storage organization table of track data（Table 2）Every record indicate a point, include the ID marks of point, put affiliated classification（It rises
Point or terminal）, the ID on the side where putting, the distance that plays node of the point apart from side where it.
The storage form of 1. road network of table
| ID | Length | Start-node | Edges linked to start-node | End-node | Edges linked to end-node | Points |
| 1 | L 1 | n 1 | E 1 | n2 | E 2 | S 1 |
| 2 | L 2 | n 3 | E 3 | n4 | E 4 | S 2 |
| …… | …… | …… | …… | …… | …… | …… |
| N | L N | n (2N-1) | E 2N-1 | n(2N) | E 2N | S N |
The storage form that 2. points of table
| ID | Category | Edge ID | Distance to Start-node |
| 1 | O | e 1 | l 1 |
| 2 | O | e2 | l 2 |
| …… | …… | …… | …… |
| M | D | E M | L M |
After establishing the correlation model between road net data and track data, point in a certain range is searched in next step S3
Road network neighborhood when can the first a certain range of network line segment of fast search, according to incidence relation correspond to starting point on line segment and
Terminal, the network distance between without cumbersome calculating a little.
After establishing the correlation model between road net data and traffic track data, linear scan system can be carried out
Meter, the first step of scan statistics should determine the statistic of the null hypothesis and alternative hypothesis and scan statistics of scan statistics first.
There are different definition, the present invention to use scan statistics according to the different null hypothesis of statistics purpose and alternative hypothesis in scan statistics
Purpose is to obtain the number of two kinds of point in window there are significant differences, therefore null hypothesis is defined as：
（1）
Alternative hypothesis is defined as：
（2）
Wherein, in formula（1）With（2）In, pOIndicate that the number of starting point O in scanning window accounts for the ratio always counted in window；qOTable
Show that the number of the outer starting point O of scanning window accounts for the ratio always counted outside window；pDIndicate that the number of terminal D in window accounts for total point in window
Several ratios；qDIt indicates that the number of the outer terminal D of window accounts for the ratio always counted outside window, then has。
The purpose of statistic defined in scan statistics is in order to determine the distribution situation of data inside and outside scanning window, for sweeping
Each window in statistics is retouched, all there are one statistical values.The calculation formula of test statistics is defined as follows by the present invention：
（3）
Wherein, Z indicates that a scanning window, C indicate the sum of beginning and end in data set,Indicate starting point institute in data set
The ratio accounted for,Indicate the ratio shared by terminal in data set,Indicate the number put in window,It indicates in window Z
Ratio shared by starting point,Indicate the ratio shared by terminal in window Z.
The second step of S3 is the detailed process of linear space scan statistics, should be with each of point data concentration point first
Window center point builds the linear scan window in network, the points inside and outside statistical window, and the test statistics of calculation window.
The detailed process of all the points in linear scan window is found in description by taking Fig. 3 as an example, and P is window center point in Fig. 3, and 2Z is window
Size, side where P points are E1, and two leafs are N1 and N7, and E2 two leafs in side are N1 and N2, two leafs of side E3 be N1 and
Two leaf of N3, side E4 is N1 and N4, and E5 two leafs in side are N4 and N5, and E6 two leafs in side are N4 and N6, are found in window
The detailed process of point be：
1. since P points, P is judged at a distance from N7, and the traversal in this direction is then terminated more than Z；
2. judging P at a distance from N1, then continue the traversal of the direction less than Z, finds all sides not being traversed of N1 connections
E2, E3, E4, judge whether side E2, the most short network distance of another the leaf N2, N3, N4 and P of E3, E4 are more than Z respectively；
3. the most short network distance of N2, N3 and P are more than Z, then N2, the traversal in the directions N3 are no longer carried out；
4. the most short network distance of N4 and P is less than Z, then all sides of N4 connections are continually looked for, repeat 2-4, until traversal is whole
Only.
Finally obtain the network edge of red line part in Fig. 3, the distance of the terminal P1, P2, P3, P4, P5 to P of network edge in figure
It is Z.It, can be fast after obtaining the side in window by the correlation model between the S2 network datas established and track point data
Speed obtains the point on the side in window, and then the statistic of window is calculated.Since the number on road net data side is far smaller than
The number of track data point, and in network side length it is known that in the present invention it is this find linear scan window in point side
Method has only traversed the side in network, and avoids the network distance calculated between point, substantially increases the execution efficiency of algorithm.
In practice, beginning and end data are divided into M different isochronous surfaces according to data acquisition time
On（T1,T2,…,TM）, with time interval per hour be an isochronous surface, below on an isochronous surface starting point and end
Cluster process introduction is carried out for point, and all N number of points on an isochronous surface are established into space correlation model with road network, it is defeated
Enter sizes of the parameter 2Z as linear scan window, the linear scan window for corresponding to a 2Z size is each put in data set
Mouthful（Z1,Z2,…,ZN）, calculate the statistical value of each window。
The statistical value of obtained window is subjected to Monte Carlo and assumes test, judges the conspicuousness of statistical value.By network
It is taken a little with interval d on side, as candidate point set, is randomly selected from candidate point concentration with the same number of point of starting point as generation mould
Quasi- starting point is randomly selected with the same number of point of terminal as the simulation terminal generated, together with simulation terminal by simulation starting point
As simulated data sets.It carries out Monte Carlo and assumes test R times, it is right in given window to calculate separately each of data set point
In the logarithm of the likelihood estimator of each random data set for assuming test generation（j=1,2,...,R）, with truthful data
The statistical value of the corresponding window of collection is compared, and then obtains the p value each put in data set：
（4）
Wherein, i indicates that each point in beginning and end data set, R indicate that the number of test is assumed in Monte Carlo, and I (*) is one
A discriminant function, takes 1 if meeting condition, and 0 is taken if being unsatisfactory for condition.
The problem of is all calculated by p value, is considered as multiple testing at this time for window corresponding to N number of point in data set,
The present invention is used corrects multiple testing problem based on linear adaptive-adaptive program, the N number of point concentrated to data according to p value
The arrangement of progress ascending order (, then fromStart to find out first and meets following formula：
（5）
Here α indicates significance, if the p value ratio of windowIt is small, then refuse null hypothesis, receives alternative vacation
If.
After correcting for the error generation rate of multiple testing, the level of significance α after a correction can be obtainedadj, i.e.,
FromStart find out first satisfaction（5）Formula.In all scanning windows, p value is less than αadjWindow be
For comprising the significantly unbalanced scanning window of two kinds of point, p value be less than αadjThe central point of scanning window can be considered as
Mix core point.Two kinds of mixing core point can be defined as：
CoreO>D：In a uneven scanning window, rises and count out more than terminal number；
CoreO<D：In a uneven scanning window, rises and count out less than terminal number.
It, can be by density expansion algorithm by two kinds of mixing after identifying two kinds of mixing core point
Core point clusters respectively obtains two kinds of cluster HOLDAnd LOHD：
HOLD：By mixing core point CoreO>DCluster obtains, and the cluster of influx is more than for traffic discharge；
LOHD：By mixing core point CoreO<DCluster obtains, and the cluster of influx is less than for traffic discharge.
Fig. 4 is to form HOLDThe specific steps of density expansion algorithm are introduced for the cluster of type, form LOHDType
The method of cluster be identical with this：
1. randomly choosing a non-classified Core firstO>DPoint m, and distribute ID number 1 for it；
2. from CoreO>DThe middle road network distance for finding range points m is less than or equal to all non-classified points of Z, and is distributed for it
ID number identical with point m 1；
3. the road network distance for continually looking for all the points found apart from previous step is less than or equal to all non-classified points of Z, and
ID number identical with point m 1 is distributed for it；
4. previous step is repeated, until in CoreO>DIn can not find non-classified mixing core point；
5. ID number adds 1；
6. repeating the above steps, until in CoreO>DIn there is no non-classified mixing core point.
After the present invention gets more significant traffic flow imbalance pattern by S3, and it is uneven to analyze this traffic flow
Evolutionary pattern is defined as the pattern of taking place frequently and accidental pattern by the dynamic evolution of pattern.The pattern definition that takes place frequently is in continuing study
Between in section, the pattern that occurs on each isochronous surface（Fig. 5 (a)）, indicate as follows：
（6）
Wherein, N indicates that the sum of isochronous surface, n indicate that n-th of isochronous surface, k indicate k-th of frequency on each isochronous surface
Hair pattern cluster, m indicate different time slice on cluster between it is believed that same position distance threshold.
Accidental pattern is divided into the first accidental pattern and second of accidental pattern by the present invention, the first accidental pattern is determined
Justice is not occur on isochronous surface there are one, the pattern all occurred on remaining isochronous surface within the continuing study period
（Fig. 5 (b)）, indicate as follows：
（7）
Second of accidental pattern definition is to occur on isochronous surface there are one, remaining time cuts within the continuing study period
The pattern that on piece does not all occur（Fig. 5 (c)）, indicate as follows：
（8）
Wherein, in formula（7）With（8）In, N indicates that the sum of isochronous surface, n indicate that n-th of isochronous surface, d indicate d-th of time
Slice, k indicate that k-th of pattern cluster that takes place frequently on each isochronous surface, m indicate to be believed that same between the cluster on different time slice
The distance threshold of one position.
Using distance, the dynamic evolution of traffic flow imbalance pattern between the average cascade synthesis measurement cluster in space cluster analysis
Analytical procedure it is as follows：
1. finding out all H on each isochronous surface respectivelyOLD(LOHD) distance on cluster to remaining all isochronous surface is most short
HOLD(LOHD) cluster and corresponding distance；
2. for the H on any one isochronous surface TiOLD(LOHD) cluster, to remaining all isochronous surface on distance it is shortest
HOLD(LOHD) the both less than given distance threshold of distance of cluster is HOLD(LOHD) take place frequently pattern, cluster position be traffic flow not
The position that takes place frequently of balanced mode；
3. for any one isochronous surface TiOn HOLD(LOHD) cluster, to remaining all isochronous surface on distance it is shortest
HOLD(LOHD) only there are one isochronous surface, to be more than given distance threshold be H for the distance of clusterOLD(LOHD) the first accidental pattern,
Cluster position is the first accidental position of traffic flow imbalance pattern；
4. for any one isochronous surface TiOn HOLD(LOHD) cluster, to remaining all isochronous surface on distance it is shortest
HOLD(LOHD) the both greater than given distance threshold of distance of cluster is HOLD(LOHD) second of accidental pattern cluster, cluster position is
Second of accidental position of traffic flow imbalance pattern.
Fig. 2 is a kind of schematic diagram of urban traffic flow imbalance mode excavation system according to the ... of the embodiment of the present invention, described
Urban traffic flow imbalance mode excavation system is used for urban traffic flow imbalance mode excavation, and with reference to Fig. 2, which includes number
Module 202, statistical module 203, computing module 204 are established according to acquisition module 201, correlation model.It illustrates separately below.
Data acquisition module 201, for obtaining road net data and track data.
Correlation model establishes module 202, for establishing the correlation model between the road net data and the track data.
Specifically, it is additionally operable to find network neighborhood using line segment extended method, avoids calculating the network distance between point, so as to carry
High algorithm performs efficiency.
Statistical module 203, for the statistical significance using linear scan statistical method evaluation traffic flow imbalance pattern.
Computing module 204, for calculating the dynamic evolution for obtaining the traffic flow imbalance pattern based on pattern similarity,
The pattern that takes place frequently and accidental pattern.
Based on the present invention, the excavation of urban traffic flow imbalance pattern may be implemented, contribute to complete detection and evaluation city
City's road traffic operation conditions analyzes the producing cause of uneven pattern with urban facilities in conjunction with urban function region, can
Certain reference is provided for Urban Traffic and traffic programme.
The term and wording used in description of the invention is just to for example, be not intended to constitute restriction.Ability
Field technique personnel should be appreciated that under the premise of not departing from the basic principle of disclosed embodiment, to the above embodiment
In each details can carry out various change.Therefore, the scope of the present invention is only determined by claim, in the claims, unless
It is otherwise noted, all terms should be understood by the broadest rational meaning.
Claims (8)
1. a kind of urban traffic flow imbalance mode excavation method, which is characterized in that the method includes：
S1, road net data and track data are obtained；
S2, correlation model between the road net data and the track data is established；
S3, the statistical significance that traffic flow imbalance pattern is evaluated using linear scan statistical method；
S4, the dynamic evolution for obtaining the traffic flow imbalance pattern is calculated based on pattern similarity.
2. the method as described in claim 1, which is characterized in that the step S2 further includes：It is found using line segment extended method
Network neighborhood.
3. method as claimed in claim 1 or 2, which is characterized in that the step S3 further includes：
The statistic of S31, the null hypothesis for determining scan statistics and alternative hypothesis and scan statistics, the null hypothesis are defined as：, the alternative hypothesis is defined as：, pOTable
Show that the number of starting point O in scanning window accounts for the ratio always counted in window；qOIndicate that the number of the outer starting point O of scanning window accounts for window
The outer ratio always counted；pDIndicate that the number of terminal D in window accounts for the ratio always counted in window；qDIndicate the outer terminal D's of window
Number accounts for the ratio always counted outside window, then has；
S32, it is put as the linear scan window in window center point structure network with each of point data concentration, inside and outside statistical window
Points, and the test statistics of calculation window, the calculation formula of the test statistics are as follows：
Wherein, Z indicates that a scanning window, C indicate the sum of beginning and end in data set,Indicate starting point institute in data set
The ratio accounted for,Indicate the ratio shared by terminal in data set,Indicate the number put in window,It indicates in window Z
Ratio shared by starting point,Indicate the ratio shared by terminal in window Z；
S33, the test statistics being calculated are carried out to Monte Carlo hypothesis test, judges the aobvious of the test statistics
Work property.
4. method as claimed in claim 3, which is characterized in that the dynamic of the traffic flow imbalance pattern in the step S4 is drilled
Turn to the pattern of taking place frequently and accidental pattern, wherein the pattern definition that takes place frequently is each isochronous surface within the continuing study period
On the pattern that all occurs, indicate as follows：
,
Wherein, N indicates that the sum of isochronous surface, n indicate that n-th of isochronous surface, k indicate k-th of frequency on each isochronous surface
Hair pattern cluster, m indicate different time slice on cluster between it is believed that same position distance threshold；
The accidental pattern is divided into the first accidental pattern and second of accidental pattern, the first described accidental pattern definition be
In the continuing study period, do not occur on isochronous surface there are one, the pattern all occurred on remaining isochronous surface, indicates such as
Under：
Wherein, d indicates d-th of isochronous surface；
Second of accidental pattern definition be within the continuing study period, only there are one occurring on isochronous surface, remaining when
Between the pattern that does not all occur on slice, indicate as follows：
。
5. a kind of urban traffic flow imbalance mode excavation system, which is characterized in that the system comprises：
Data acquisition module, for obtaining road net data and track data；
Correlation model establishes module, for establishing the correlation model between the road net data and the track data；
Statistical module, for the statistical significance using linear scan statistical method evaluation traffic flow imbalance pattern；
Computing module, for calculating the dynamic evolution for obtaining the traffic flow imbalance pattern based on pattern similarity.
6. system as claimed in claim 5, which is characterized in that the correlation model is established module and is additionally operable to be extended using line segment
Method finds network neighborhood.
7. such as system described in claim 5 or 6, which is characterized in that the statistical module is additionally operable to：
Determine that the statistic of the null hypothesis and alternative hypothesis and scan statistics of scan statistics, the null hypothesis are defined as：, the alternative hypothesis is defined as：, pOTable
Show that the number of starting point O in scanning window accounts for the ratio always counted in window；qOIndicate that the number of the outer starting point O of scanning window accounts for window
The outer ratio always counted；pDIndicate that the number of terminal D in window accounts for the ratio always counted in window；qDIndicate the outer terminal D's of window
Number accounts for the ratio always counted outside window, then has；
It is the linear scan window in window center point structure network, the point inside and outside statistical window with each of point data concentration point
Number, and the test statistics of calculation window, the calculation formula of the test statistics are as follows：
Wherein, Z indicates that a scanning window, C indicate the sum of beginning and end in data set,Indicate starting point institute in data set
The ratio accounted for,Indicate the ratio shared by terminal in data set,It indicates in window
The number of point,Indicate the ratio shared by starting point in window Z,Indicate the ratio shared by terminal in window Z；
The test statistics being calculated are subjected to Monte Carlo and assume test, judge the notable of the test statistics
Property.
8. system as claimed in claim 7, which is characterized in that the dynamic evolution of the traffic flow imbalance pattern is the mould that takes place frequently
Formula and accidental pattern, wherein the pattern definition that takes place frequently is to occur on each isochronous surface within the continuing study period
Pattern indicates as follows：
,
Wherein, N indicates that the sum of isochronous surface, n indicate that n-th of isochronous surface, k indicate k-th of frequency on each isochronous surface
Hair pattern cluster, m indicate different time slice on cluster between it is believed that same position distance threshold；
The accidental pattern is divided into the first accidental pattern and second of accidental pattern, the first described accidental pattern definition be
In the continuing study period, do not occur on isochronous surface there are one, the pattern all occurred on remaining isochronous surface, indicates such as
Under：
Wherein, d indicates d-th of isochronous surface；
Second of accidental pattern definition be within the continuing study period, only there are one occurring on isochronous surface, remaining when
Between the pattern that does not all occur on slice, indicate as follows：
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810510479.1A CN108665708B (en) | 2018-05-24 | 2018-05-24 | Urban traffic flow imbalance mode mining method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810510479.1A CN108665708B (en) | 2018-05-24 | 2018-05-24 | Urban traffic flow imbalance mode mining method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108665708A true CN108665708A (en) | 2018-10-16 |
| CN108665708B CN108665708B (en) | 2020-06-05 |
Family
ID=63776568
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810510479.1A Active CN108665708B (en) | 2018-05-24 | 2018-05-24 | Urban traffic flow imbalance mode mining method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108665708B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111574A (en) * | 2019-05-16 | 2019-08-09 | 北京航空航天大学 | A kind of urban transportation imbalance evaluation method based on the analysis of flow tree |
| CN110930281A (en) * | 2019-12-04 | 2020-03-27 | 中南大学 | Method and system for statistical detection of urban traffic flow community structure |
| CN112419131A (en) * | 2020-11-20 | 2021-02-26 | 中南大学 | Method for estimating traffic origin-destination demand |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127159A (en) * | 2007-09-18 | 2008-02-20 | 中国科学院软件研究所 | Traffic flow data sampling and analysis based on network limited moving object database |
| CN101783074A (en) * | 2010-02-10 | 2010-07-21 | 北方工业大学 | Method and system for real-time distinguishing traffic flow state of urban road |
| US20120226434A1 (en) * | 2011-03-04 | 2012-09-06 | Board of Regents, University of Arizona | Active traffic and demand management system |
| CN103974311A (en) * | 2014-05-21 | 2014-08-06 | 哈尔滨工业大学 | Condition monitoring data stream anomaly detection method based on improved gaussian process regression model |
| CN104573116A (en) * | 2015-02-05 | 2015-04-29 | 哈尔滨工业大学 | Taxi GPS data mining based traffic abnormality recognition method |
| CN105489006A (en) * | 2015-12-15 | 2016-04-13 | 浙江工业大学 | Multi-scale road flow visual analysis method based on taxi GPS data |
| JP2016197423A (en) * | 2016-06-16 | 2016-11-24 | 株式会社Ｚｍｐ | Server system |
-
2018
- 2018-05-24 CN CN201810510479.1A patent/CN108665708B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101127159A (en) * | 2007-09-18 | 2008-02-20 | 中国科学院软件研究所 | Traffic flow data sampling and analysis based on network limited moving object database |
| CN101783074A (en) * | 2010-02-10 | 2010-07-21 | 北方工业大学 | Method and system for real-time distinguishing traffic flow state of urban road |
| US20120226434A1 (en) * | 2011-03-04 | 2012-09-06 | Board of Regents, University of Arizona | Active traffic and demand management system |
| CN103974311A (en) * | 2014-05-21 | 2014-08-06 | 哈尔滨工业大学 | Condition monitoring data stream anomaly detection method based on improved gaussian process regression model |
| CN104573116A (en) * | 2015-02-05 | 2015-04-29 | 哈尔滨工业大学 | Taxi GPS data mining based traffic abnormality recognition method |
| CN105489006A (en) * | 2015-12-15 | 2016-04-13 | 浙江工业大学 | Multi-scale road flow visual analysis method based on taxi GPS data |
| JP2016197423A (en) * | 2016-06-16 | 2016-11-24 | 株式会社Ｚｍｐ | Server system |
Non-Patent Citations (2)
| Title |
|---|
| 盛子豪: "基于数据挖掘技术的交通拥堵判别与预测算法研究及应用", 《中国优秀硕士学位论文全文数据库》 * |
| 许明: "基于车辆轨迹挖掘的城市路网分析关键问题研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111574A (en) * | 2019-05-16 | 2019-08-09 | 北京航空航天大学 | A kind of urban transportation imbalance evaluation method based on the analysis of flow tree |
| CN110930281A (en) * | 2019-12-04 | 2020-03-27 | 中南大学 | Method and system for statistical detection of urban traffic flow community structure |
| CN110930281B (en) * | 2019-12-04 | 2023-10-03 | 中南大学 | Method and system for statistical detection of urban traffic flow community structure |
| CN112419131A (en) * | 2020-11-20 | 2021-02-26 | 中南大学 | Method for estimating traffic origin-destination demand |
| CN112419131B (en) * | 2020-11-20 | 2022-07-08 | 中南大学 | Method for estimating traffic origin-destination demand |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108665708B (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| An et al. | A network partitioning algorithmic approach for macroscopic fundamental diagram-based hierarchical traffic network management | |
| CN110111574B (en) | Urban traffic imbalance evaluation method based on flow tree analysis | |
| CN108665708A (en) | A kind of urban traffic flow imbalance mode excavation method and system | |
| Khezerlou et al. | A traffic flow approach to early detection of gathering events: Comprehensive results | |
| CN108645342B (en) | Road width extraction method based on road track and high-resolution image | |
| CN108320511A (en) | Urban highway traffic sub-area division method based on spectral clustering | |
| CN103345566B (en) | Based on the geochemical anomaly discrimination and evaluation method of Geological Connotation | |
| CN106572493A (en) | Abnormal value detection method and abnormal value detection system in LTE network | |
| CN110677859A (en) | Method and device for determining weak coverage area and computer readable storage medium | |
| Liu et al. | Relationships between street centrality and land use intensity in Wuhan, China | |
| CN106355881A (en) | Space-autocorrelation-based traffic accident blackspot identification method and device | |
| CN108009972A (en) | A kind of multimode trip O-D needs estimate methods checked based on multi-source data | |
| CN107832910B (en) | Method for evaluating influence of road traffic characteristics on carbon monoxide emission concentration | |
| CN110738853B (en) | Key node identification method based on complex network correlation | |
| Yu | A mathematical morphology based method for hierarchical clustering analysis of spatial points on street networks | |
| Janus | A new approach to calculating distances to parcels: A way to increase the accuracy of farm efficiency analyses and the assessment of land consolidation projects | |
| CN107368480A (en) | A kind of interest point data type of error positioning, repeat recognition methods and device | |
| Michau et al. | Combining traffic counts and Bluetooth data for link-origin-destination matrix estimation in large urban networks: The Brisbane case study | |
| CN109710720A (en) | A kind of crime risk such environmental effects analysis method for taking background effect into account | |
| Liu et al. | A new multiple-factor clustering method considering both box fractal dimension and orientation of joints | |
| Hasan et al. | Cluster analysis for automated operational modal analysis: A review | |
| CN113379334B (en) | Road section bicycle riding quality identification method based on noisy track data | |
| Kłos et al. | Performance evaluation of roundabouts using a microscopic simulation model | |
| Kao et al. | Optimization models for siting water quality monitoring stations in a catchment | |
| Badii et al. | Real-time automatic air pollution services from IOT data 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 |