CN103514743A - Method for recognizing abnormal traffic state characteristics of real-time index data matching memory range - Google Patents

Abstract

The invention relates to a method for recognizing abnormal traffic state characteristics of a real-time index data matching memory range. The method is characterized by including the steps of firstly, obtaining all historical traffic index data of a certain road section in a certain time period, setting up index memory spaces corresponding to the moment according to all historical traffic indexes at the same moment, and calculating mid-values of all the index memory spaces; secondly, at least dividing each index memory space into a positive buffer range and a negative buffer range, defining areas smaller than the minimum index value as negative overflow areas, and defining areas larger than the maximum index value as positive overflow areas; thirdly, obtaining a real-time traffic index Index(Tn) of the road section at the current moment Tn, and judging the traffic state according to partitions. The invention provides a novel traffic index application method, and therefore the potential of traffic indexes is fully exploited, and the rapid recognition of urban road real-time traffic abnormality can be achieved.

Description

A kind of real-time index-matched is remembered interval abnormal traffic state characteristic recognition method

Technical field

The present invention relates to a kind of real-time identification method of urban road abnormal traffic state feature, the method is usingd the historical law data of urban road real-time traffic index monitoring road conditions accumulation and is remembered as experience, and set up " memory space " according to the statistical property of data memory, real-time traffic index-matched current time section " memory space ", and whether the target road object of being monitored with this discrimination index is current abnormal in traffic behavior, belongs to intelligent transportation system field.

Background technology

Traffic index is a kind ofly can express with serial number a class relativity index of urban road traffic state or traffic congestion, according to practical application request, choose special traffic parameter and build according to certain functional rule, along with the development of transport information technology, domestic each main cities meets the traffic index model of city characteristic road traffic running status is monitored in real time as Beijing, Shanghai, Hangzhou, Shenzhen etc. have all built.Monitor in real time three significant mathematical features of traffic index tool of dynamic road operation characteristic, the first, index is present in fixing numerical value interval, as [0,100], [0,5] etc., choosing of data interval determined jointly by exponential model and issue requirement; The second, index result is continuous numerical value, can cover in theory all real number values in data interval; The 3rd, index has monotonicity, and the trend that the dull reaction of exponential quantity road conditions improve or degenerate, does not exist ambiguity, and the present invention take that the larger road conditions of index are worse carries out as example.Just based on these three mathematical features, index can be with the form record object road of numerical point real-time traffic states constantly, and then can draw the curve of a reflection whole day traffic behavior Change and Development trend, by this curve, not only can differentiate target road or road network region in one day, the block up moment and the degree on peak, can also compare to different road objects or road network colony the assessment serious jam road of normality and unimpeded road.

To urban highway traffic in general, being subject to population distribution, road layout, traffic trip rule etc. factor is metastable affects, traffic behavior often has certain time space distribution, the road occurred frequently that blocks up is in the ordinary course of things relatively-stationary, and the time occurred frequently of jam road is also relatively-stationary.Traffic control department can deploy to ensure effective monitoring and control of illegal activities according to the police strength of this specific character formulation normality, optimum management resource, and smooth work is protected in commander's unimpeded.But the road network system that city level is numerous, the scale of construction is huge has the traffic congestion feature that occurred frequently, traffic rule is suddenlyd change at random, especially when in the face of factors such as inclement weather, eve festivals or holidays, large-scale social activitieies, the road that normality is not blocked up can produce and be difficult to seriously blocking up of precognition, or the normality section that blocks up can present significantly unimpeded etc., this situation tends to do not ignored in early days because the road network that blocks up does not belong to the emphasis section of deploying to ensure effective monitoring and control of illegal activities, and causes road conditions to continue to worsen and even causes serious consequence long-time, that large area is blocked up.Traffic index can be realized the digital expression to road conditions as a kind of efficient information tool, if can real-time road the further change of the personalized traffic behavior feature of every road of discovery of intelligence on the basis of monitoring, the abnormal state degree of quantitative analysis assessment road, will bring important intelligence value and economic worth for confirming that fast cause of problem and site traffic are managed.By the retrieval to prior art and system, do not find to obtain the known correlation technique that can meet the abnormal ONLINE RECOGNITION of real-time traffic states and method.

Summary of the invention

The technical problem to be solved in the present invention is to provide a kind of new traffic index application process, and the potentiality of fully excavating traffic index realize the abnormal quick identification of urban road real-time traffic.

In order to solve the problems of the technologies described above, technical scheme of the present invention has been to provide a kind of real-time index-matched and has remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, step is:

Step 1, in all traffic flow parameters, choose arbitrarily a real time data at the traffic flow parameter of the interval dynamic fluctuation of fixed data as traffic status identification parameter, obtain all historical traffic status identification supplemental characteristic of some time period Nei Moutiao sections or Regional Road Network, according to all traffic status identification parameters under synchronization, create the traffic status identification parameters memorizing space corresponding with this moment and calculate the intermediate value in each traffic status identification parameters memorizing space, wherein, traffic status identification parameters memorizing space corresponding to T is IMS (T) constantly, minimal index value MIN (T) and greatest exponential value MAX (T) in all historical traffic status identification supplemental characteristic of T is negative edge and the positive boundary of traffic status identification parameters memorizing space IMS (T) constantly, the intermediate value MEDIAN (T) of traffic status identification parameters memorizing space IMS (T) is the average of all traffic status identification supplemental characteristics under moment T,

Step 2, each traffic status identification parameters memorizing space is at least divided between Jian Ji negative buffer, positive buffer zone, and will be less than the zone definitions Wei Fu overflow area of minimal index value, will be greater than the zone definitions Wei Zheng overflow area of greatest exponential value; Between the positive buffer zone of traffic status identification parameters memorizing space IMS (T), be PB (T), span be (MEDIAN (T), MAX (T)]; Between negative buffer, be NB (T), span be [MIN (T), MEDIAN (T)); Negative overflow area is NO (T), span be [MIN, MIN (T)); Positive overflow area is PO (T), span be (MAX (T), MAX], wherein, MIN is the minimum possible value of traffic status identification parameter, MAX is the maximum possible value of traffic status identification parameter;

Step 3, obtain the real-time traffic states identification parameter Index (Tn) of Gai Tiao section under current time Tn, judge whether real-time traffic states identification parameter Index (Tn) is positioned between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn) NB (Tn) between PB (Tn) or negative buffer, if, the traffic behavior of Ze Gaitiao section under current time Tn is normality, if not, the traffic behavior of Ze Gaitiao section under current time Tn is abnormal, this extremely refers to extremely unimpeded or extremely blocks up, to be that traffic status identification parameter is more high more block up the traffic status identification parameter model that Ruo Gaitiao section adopts, if real-time traffic states identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely blocking up, if real-time traffic states identification parameter Index (Tn) is positioned at negative overflow area NO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, the traffic status identification parameter model that Ruo Gaitiao section adopts is that traffic status identification parameter is more high more unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at the traffic behavior of negative overflow area NO (Tn) ，Gai Tiao section under current time Tn for extremely blocking up.

Preferably, in described step 1, utilize influence factor mark MARK, all historical traffic identification parameter data to some time period Nei Moutiao section are carried out mark, influence factor mark MARK has on traffic behavior feature one group of cartesian product configuration item that the non-traffic factor of extensive impact forms in influence factor config set, all historical traffic identification parameter under synchronization is classified according to influence factor mark MARK, for each class creates respectively traffic status identification parameters memorizing space and calculates the intermediate value in each traffic status identification parameters memorizing space,

In described step 3, obtain after the real-time traffic identification parameter Index (Tn) under current time Tn, first to it, utilize influence factor mark MARK to carry out mark, obtain the traffic identification parameter Index (Tn) with mark result _mARK, utilize traffic identification parameter Index (Tn) _mARKfind constantly the traffic status identification parameters memorizing space corresponding with its classification under Tn, then judge that according to the relation of between itself and Jian Ji negative buffer, Zhong Zheng buffer zone, traffic status identification parameters memorizing space and Ji Zheng overflow area, negative overflow area whether the traffic behavior of Gai Tiao section under current time Tn be abnormal.

Preferably, the described non-traffic factor that traffic behavior feature is had to an extensive impact at least comprises Weather information Weather, calendar information Calendar and action message Event, wherein, calendar information Calendar is main factor, each has different influence factor mark MARK constantly, and the influence factor of T is labeled as MARK constantly _t(1,2,3)=MARK _t(Calendar, Weather, Event).

Preferably, in described step 2, by each traffic status identification parameters memorizing space be at least divided between positive buffer zone, normal state is interval, between negative normality interval and negative buffer; Between the positive buffer zone of traffic status identification parameters memorizing space IMS (T), the span of PB (T) is (MPSD (T), MAX (T)], MPSD (T)=MEDIAN (T)+stdev (T), stdev (T) is the standard deviation of all historical traffic identification parameters under moment T; Normal state interval is PN (T), and span is [MEDIAN (T), MPSD (T)]; Negative normality interval is NN (T), span be [MNSD (T), MEDIAN (T)), MNSD (T)=MEDIAN (T)-stdev (T); The span of NB between negative buffer (T) be [MIN (T), MNSD (T));

In described step 3, obtain the real-time traffic identification parameter Index (Tn) of Gai Tiao section under current time Tn, judge whether real-time traffic identification parameter Index (Tn) is positioned at the interval PN (Tn) of normal state or the negative normality interval (Tn) of traffic status identification parameters memorizing space IMS (Tn), if, the traffic behavior of Ze Gaitiao section under current time Tn is normal, if not, the traffic behavior of Ze Gaitiao section under current time Tn is abnormal, this extremely refers to extremely unimpeded or extremely blocks up, to be that traffic identification parameter is more high more block up the traffic identification parameter model that Ruo Gaitiao section adopts, if real-time traffic identification parameter Index (Tn) is positioned at PB (Tn) Huo Zheng overflow area PO (Tn) between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely blocking up, if real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) Huo Fu overflow area NO (Tn) between the negative buffer of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, the traffic identification parameter model that Ruo Gaitiao section adopts is that traffic identification parameter is more high more unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at PB (Tn) Huo Zheng overflow area PO (Tn) between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, if real-time traffic identification parameter Index (Tn) is positioned between the negative buffer of traffic status identification parameters memorizing space IMS (Tn) traffic behavior of NB (Tn) Huo Fu overflow area NO (Tn) ，Gai Tiao section under current time Tn for extremely blocking up.

Preferably, in described step 3, if real-time traffic identification parameter Index (Tn) is positioned between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn) NB (Tn) between PB (Tn) or negative buffer, adopt any one in following two methods to judge the traffic behavior of current section under current time Tn:

First method: if real-time traffic identification parameter Index (Tn) is positioned at PB between positive buffer zone (Tn), threshold value [MAX (Tn)+MPSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, be greater than this value and sentence " extremely ", be less than this value and be greater than MPSD (Tn)/MEDIAN (Tn) and judge " doubtful abnormal "; If real-time traffic identification parameter Index (Tn) is positioned at NB between negative buffer (Tn), threshold value [MIN (Tn)+MNSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, be less than this value and sentence " extremely ", be greater than this value and be less than MNSD (Tn)/MEDIAN (Tn) and judge " doubtful abnormal ";

Second method: judge whether to exist reference point Ref _t, reference point Ref _twhile entering first Huo Fu overflow area, Jian,Zheng overflow area, Jian, negative buffer, positive buffer zone by real-time traffic identification parameter Index (Tn) by the interval PN (Tn) of normal state or the interval NN of negative normality (Tn), intersecting interpolation with MPSD (Tn) or MNSD (Tn) creates, interpolation is constantly rounded to and intersects previous moment constantly, if not, sentence " normality ", if, judge that real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, if be positioned at PB between positive buffer zone (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MPSD (Tn)-Ref _t] * T ₀/ (Tn-t), and wherein, T ₀for the update cycle of traffic identification parameter, t be interpolation constantly, if its result is greater than A, sentence " extremely ", between (0, A] between sentence " doubtful extremely ", A is empirical value; If be positioned at NB between negative buffer (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MNSD (Tn)-Ref _t] * T ₀/ (Tn-t), if be not more than-A of its result sentences " extremely ", between (A, 0] between sentence " doubtful abnormal ".

Preferably, utilize three-dimensional extremely to combine the traffic behavior that judges Mou Tiao section in described step 3, described step 3 comprises:

Step 3.1, obtain Gai Tiao section or the real-time traffic identification parameter Index (Tn) of Regional Road Network under current time Tn;

Step 3.2, judge that whether real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, if not, judges whether to exist reference point Ref _t, reference point Ref _twhile entering Huo Fu overflow area, Jian,Zheng overflow area, Jian, negative buffer, positive buffer zone by real-time traffic identification parameter Index (Tn), intersect interpolation with MPSD (Tn) or MNSD (Tn) and create, if exist, directly enter next step, if do not exist, created reference point Ref _tafter enter next step; If real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, judge whether to exist reference point Ref _tif, exist, sentence " normality ", completing steps 3, if do not exist, has created reference point Ref _tafter enter step 3.4;

Step 3.3, the first dimension abnormality juding:

If real-time traffic identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), calculate Index (Tn) MAX (Tn), if this difference is less than A, sentence " doubtful abnormal ", enter next step, otherwise, sentence " extremely ", enter next step; If real-time traffic identification parameter Index (Tn) is positioned at negative overflow area NO (Tn), calculate MIN (Tn)-Index (Tn), if this difference is less than A, sentence " doubtful abnormal ", enter next step, otherwise, sentence " extremely ", enter next step, A is empirical value;

Step 3.4, the second dimension abnormality juding:

If real-time traffic identification parameter Index (Tn) is positioned at PB between positive buffer zone (Tn) Huo Zheng overflow area PO (Tn), threshold value [MAX (Tn)+MPSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, Index (Tn)/MEDIAN (Tn) is greater than this value and sentences " extremely ", enter next step, Index (Tn)/MEDIAN (Tn) is less than this value and is greater than MPSD (Tn)/MEDIAN (Tn) and judges " doubtful abnormal ", enters next step; If real-time traffic identification parameter Index (Tn) is positioned at NB between negative buffer (Tn) Huo Fu overflow area NO (Tn), threshold value [MIN (Tn)+MNSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, Index (Tn)/MEDIAN (Tn) is less than this value and sentences " extremely ", enter next step, Index (Tn)/MEDIAN (Tn) is greater than this value and is less than MNSD (Tn)/MEDIAN (Tn) and judges " doubtful abnormal ", enters next step;

Step 3.5, third dimension degree abnormality juding:

If be positioned at PB between positive buffer zone (Tn) Huo Zheng overflow area PO (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MPSD (Tn)-Ref _t] * T ₀/ (Tn-t), and wherein, T ₀for the update cycle of traffic identification parameter, t be interpolation constantly, if its result is greater than A, sentence " extremely ", enter next step, between (0, A] between sentence " doubtful extremely ", enter next step; If be positioned at NB between negative buffer (Tn) Huo Fu overflow area NO (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MNSD (Tn)-Ref _t] * T ₀/ (Tn-t), if be not more than-A of its result sentences " extremely ", enter next step, between (A, 0] between sentence " doubtful abnormal ", enter next step;

Step 3.6, adopt " an abnormal voting adopted is fixed " or " the minority is subordinate to the majority " extremely to combine differentiation, wherein, " an abnormal voting adopted fixed " refers to: at traffic behavior from " normality " to " extremely " cognitive phase, the first dimension abnormality juding result described in step 3.3 has a ticket power to make decision; And from " extremely ", returning to " normality " stage, the result at the first dimension abnormality juding is on " normality " basis, by the second dimension abnormality juding described in step 3.4 and the third dimension degree abnormality juding described in step 3.5, jointly confirms abnormal restoring;

" the minority is subordinate to the majority " refers to: the second dimension abnormality juding result described in the first dimension abnormality juding result, 3.4 described in step 3.3 and the third dimension degree abnormality juding result described in step 3.5 have identical weight, according to the majority of output result of determination, judges final recognition result; If the output of the result of determination of three dimensionality is all not identical, or according to " criterion of pessimism " towards " extremely " direction identifies, or identify according to " criterion of optimism " towards " normality " direction.

The abnormal traffic state feature real-time identification method that the present invention proposes match index memory space is a kind ofly to utilize real-time index to combine the abnormal intelligent method of identification traffic behavior with history index, difference is, history index is on the basis through influence factor classification such as date, weather, for creating out " memory space " in data cover region on each time slice, and carry out off-note identification as total environment and algorithm triggers condition." memory space " marks off six domain logics by minimal value, negative bias difference, intermediate value, positively biased difference, five controlling values of maximum value: just overflow, just buffering, normal state, negative normality, negative buffering, negative overflowing, form the “Liu territory of division index span " wherein just overflow, negative overflowing for " memory space " exterior domain.Whether real-time index-matched " memory space " is also differentiated and is needed trigger condition to start the analysis of multidimensional abnormity diagnosis according to place domain logic, multidimensional exception diagnosis algorithm comprises that differential analysis, proportion grading, trend analysis etc. can extensive diagnostic methods, and each dimension independence Output rusults is all fuzzy logic judgement.Gather multidimensional abnormity diagnosis analysis result, input is combined differentiation algorithm and is carried out aggregative weighted differentiation, and its final Output rusults is the traffic behavior feature output of time slice.(note: whether all use above constantly, unify, time slice is to be constantly used on real-time index with difference slightly constantly, time slice is used in memory space on corresponding all statistics dates constantly.This my sensation still keeps present literary style, because be easier to understand)

The match index that the present invention proposes is remembered interval abnormal traffic state feature real-time identification method beneficial effect and can be embodied in the following aspects:

The first, from traffic information system aspect, realize the comprehensive abnormal monitoring to urban road the whole network, section, path, section, eliminate road conditions abnormal monitoring space-time blind area;

The priority scheduling of resource of the second ,Wei city relevant departments provides information instrument, and additional related department, to the genetic analysis of off-note and decision-making, especially has important value to the early warning of some pernicious traffic congestion;

Three, can greatly improve the recognition efficiency of the whole traffic abnormity of urban traffic network, improve traffic administration note abnormalities rapidity and the accuracy of traffic problems, for multidisciplinary combine to dispose save time;

Four, abnormal recognition result output is after confirming, for statistics the road traffic frequency and degree annual, monthly generation abnormality feature provide quantification analysis index helpful.

Accompanying drawing explanation

Fig. 1 real-time traffic abnormality recognition method logical architecture;

Fig. 2 real-time traffic abnormality recognition method main-process stream;

Fig. 3 index memory space builds sub-process;

The real-time index-matched memory space of Fig. 4 sub-process;

Fig. 5 is that multidimensional abnormity diagnosis is analyzed sub-process;

Fig. 6 abnormal state connection is appraised and is analysed sub-process;

North and south, Fig. 7 Shanghai City elevated bridge section normal case of status flag on June 19 (Wednesday) in 2013;

Middle Ring Line, Fig. 8 Shanghai City section abnormal case of morning peak status flag on April 19 (Thursday) in 2012.

Embodiment

For the present invention is become apparent, hereby with preferred embodiment, and coordinate accompanying drawing to be described in detail below.

Method provided by the invention all has effect, such as traffic index or the road travel speed of a motor vehicle etc. for having the traffic flow parameter of real time data in the interval dynamic fluctuation of fixed data.In the present embodiment, the traffic index of take is described further the method that provides of invention as example, and traffic index model adopts traffic index more more to block up, if traffic index is less of more blocking up, its principle is identical with the present embodiment, as long as by the contrary setting in relevant position.With regard to the basic terms of using in the present invention, provide definition below:

Time slice (Time Slice): each moment value of whole day that the real-time traffic index update cycle T O of take obtains as step-length, Tn=n * TO (n=1,2 ...).In existing traffic index model, update cycle TO is the longest is half an hour, for the present invention, and update cycle TO more short better (suggestion is 2min, 5min).By align index historical data base create memory space of time slice, if current time is T, measurement period is 100 days, in corresponding 100 days each constantly the index history value on T form a time slice.

To be Weather information Weather, calendar information Calendar in influence factor config set, action message Event etc. have on traffic behavior feature one group of cartesian product configuration item that the non-traffic factor of extensive impact forms to influence factor mark MARK. influence factor mark MARK.In influence factor mark MARK, calendar information Calendar column number is 1, comprise Monday, Tu. ..., Sun., lunar calendar red-letter day etc.; Weather information Weather column number is 2, comprises fine day, light rain, moderate rain, heavy rain, heavy rain, slight snow, heavy snow, dense fog, hail etc.; Action message Event column number is 3, containing normality, World Expo, the Olympic Games, car exhibition, college entrance examination, winter and summer vacation etc.; As the MARK that is labeled as in the T moment _t(1,2,3)=MARK _t(Calendar, Weather, Event)=MARKT (Monday, fine, normality), wherein, calendar information Calendar is main factor, also bears the basic function to historical data classification.

Index memory space IMS (Index Memory Space): in the complete or collected works of the index historical data measurement period (as 2 years) after MARK mark, the index extreme difference that the maximum value occurring on a time slice in office and minimal value are sealed is interval.If index history value is 2 years, wherein there are 112 days Monday, the memory space in the 8:00 moment on Monday is exactly these 112 8:00 data intervals that index covers constantly.

The minimal index value occurring in index memory space IMS on minimal index value MIN. time slice, lower limit 0.

Greatest exponential value MAX: the minimal index value occurring in index memory space IMS on time slice, higher limit 100.

Intermediate value MEDIAN: the index memory space IMS on time slice, according to ascending sequence, the Mean value of index obtaining in the concentrated statistics of general classification is as intermediate value.

Negative bias difference MNSD (Median Negative Standard Deviation): the index memory space IMS on time slice is used to standard deviation function, obtain stdev () value, MNSD=MEDIAN-stdev (), if MNSD≤MIN gets MIN value.

Positively biased difference MPSD (Median Positive Standard Deviation): the index memory space IMS on time slice is used to standard deviation function, obtain stdev () value, MPSD=MEDIAN+stdev (), if MPSD >=MAX, gets MAX value.

Positive overflow area PO (Positive Overflow): the index codomain being greater than on time slice more than maximum value MAX is interval, and span (MAX, 100], if MAX=100 under extreme case is just overflowing and is equaling maximum value MAX.Positive overflow area PO is the exterior domain of index memory space IMS, belong to the just half side of overflow area, in real time index drops on this interval, and explanation monitoring road is current has occurred the poorest traffic behavior, belong to suspicious exceptional value, need to start the analysis of multidimensional abnormity diagnosis and combine differentiation algorithm.

PB (Positive Buffer) between positive buffer zone: on time slice, be greater than that positively biased difference MPSD is above and to be less than the index codomain of maximum value MAX interval, span (MPSD, MAX].Between positive buffer zone, PB is index memory space IMS inner region, belongs to the just half side of buffer zone, though in real time index to drop on this interval abnormal, be also as extremely directly output, need and combine in conjunction with the analysis of multidimensional abnormity diagnosis and differentiate algorithm output and monitor.

The interval PN (Positive Normal) of normal state: on time slice, be less than or equal to positively biased difference MPSD and be more than or equal to the index codomain of intermediate value MEDIAN interval, span [MEDIAN, MPSD].The interval PN of normal state is index memory space IMS main areas, belongs to the just half side of normality district, and it is normality that real-time index drops on this interval, only need maintain monitoring without launching other measures.

The interval NN (Negative Normal) of negative normality: on time slice, be more than or equal to negative bias difference MNSD and be less than the index codomain of intermediate value MEDIAN interval, span [MNSD, MEDIAN).Negative normality is index memory space IMS main areas, belongs to the negative half side of normality district, and it is normality that real-time index drops on this interval, only need maintain monitoring without launching other measures.

NB between negative buffer (Negative Buffer): on time slice, be more than or equal to that minimal value MIN is above and to be less than the index codomain of negative bias difference MNSD interval, span [MIN, MNSD).Between negative buffer, NB is index memory space IMS inner region, belongs to the negative half side of buffer zone, though in real time index to drop on this interval abnormal, be also as extremely directly output, need to start multidimensional abnormity diagnosis analysis and combine and differentiate algorithm output and monitor.

Negative overflow area NO (Negative Overflow): the index codomain being less than on time slice below minimal value MIN is interval, [0, MIN), if MIN=0 under extreme case, negative overflowing equals MIN to span.Bear and overflow the exterior domain for index memory space IMS, belong to the negative half side of overflow area, in real time index drops on this interval, and explanation monitoring road is current has occurred best traffic behavior, belong to suspicious exceptional value, need to start the analysis of multidimensional abnormity diagnosis and combine differentiation algorithm.

Reference point Ref _t(Reference): the interim reference point that is used for carrying out trend judgement, position is on positively biased difference MPSD or negative bias difference MNSD, when real-time traffic index Index (Tn) enters buffer zone or overflow area by normality region, intersect interpolation with positively biased difference MPSD or negative bias difference MNSD and create, interpolation constantly t is rounded to the previous moment that intersects the moment.Reference point Ref _tafter generation, have and only have one, its numerical value is with constantly t is all constant, and while only having traffic index Index (Tn) on certain time slice on the same day again to enter normality district by buffer zone or overflow area, reference point is automatically deleted and emptied.

As shown in Figures 1 and 2, a kind of real-time index-matched provided by the invention is remembered interval abnormal traffic state characteristic recognition method, the steps include:

Step 1, obtain all historical traffic index in certain time period Nei Moutiao section, according to the influence factor mark MARK of traffic impact set of factors output _t(1, 2, 3) the historical traffic index of moment Tn is divided into groups (not containing the up-to-date moment), Main classification influence factor is calendar information Calendar, the factors such as Weather information Weather and action message Event are as auxiliary packet factor, take sorted historical traffic index as Exponential Sample overall building index memory space IMS (Tn), in conjunction with Fig. 3, comprise intermediate value MEDIAN (Tn), positive extreme value MAX (Tn), negative pole value MIN (Tn), positively biased difference MPSD (Tn)=MEDIAN (Tn)+stdev (Tn), if MPSD (Tn)>=MAX (Tn), get MAX (Tn) value, negative bias difference MNSD (Tn)=MEDIAN (Tn)-stdev (Tn) is if MNSD (Tn)≤MIN (Tn), get MIN (Tn) value.After " five values " line creates, NB (Tn) ,Fu overflow area NO (Tn) between PB (Tn), the interval PN (Tn) of normal state, the interval NN (Tn) of negative normality, negative buffer between Nature creating “Liu territory ”，Ji Zheng overflow area PO (Tn) ,Zheng buffer zone.If MPSD (Tn)=MAX (Tn), does not have PB between positive buffer zone (Tn).In like manner, if MNSD (Tn)=MIN (Tn) does not have NB between negative buffer (Tn).Creating index memory space IMS (Tn) totally carries out based on sub-index historical data, therefore do not need to carry out in real time, in system idle periods unification in morning every day, be each road, each cycle to carry out index memory space IMS (Tn) to create, result data is that the real-time index-matched of second day is prepared.

Step 2, the real-time index result of mark: the influence factor mark MARK exporting by traffic impact set of factors _t(1,2,3), carry out mark to the real-time traffic index Index (Tn) of moment Tn, generate the data I ndex (T) with mark result _mARKand deposit historical data base in, mate the index memory space IMS (Tn) of Tn constantly simultaneously.

Step 3, real-time index-matched index memory space IMS (Tn), in conjunction with Fig. 4: with MARK _tthe real-time traffic index Index (Tn) of (1,2,3) mark _mARKmatch index memory space IMS (Tn), differentiates Index (T) according to NB (Tn) ,Fu overflow area NO (Tn) between PB (Tn) between positive overflow area PO (Tn) ,Zheng buffer zone, the interval PN (Tn) of normal state, the interval NN (Tn) of negative normality, negative buffer _mARKregion.

If real-time traffic index Index (Tn) _mARKbe positioned at the interval NN (Tn) of the interval PN (Tn) of normal state or negative normality, directly output " normality " empties reference point Ref after differentiating result _t, and jump out whole step and enter next cycle.If real-time traffic index Index (Tn) _mARKbe positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, differentiate and whether had reference point Ref _t, if there is no, generate reference point Ref _t, and directly output " normality " is differentiated and is jumped out whole step after result and enter next cycle.If there is reference point Ref _t, the second dimension and the third dimension in the multidimensional abnormity diagnosis analysis of setting up procedure 4 and extremely combine differentiation, obtains status flag and differentiates after result, jumps out whole step and enters next cycle.If real-time traffic index Index (Tn) _mARKbe positioned at positive overflow area PO (Tn) Huo Fu overflow area NO (Tn), differentiate and whether had reference point Ref _t, if there is no, generate reference point Ref _t, and first in the multidimensional abnormity diagnosis analysis of setting up procedure 4 dimension, the second dimension and the third dimension and extremely combine differentiation, obtain status flag and differentiate after result, jump out whole step and enter next cycle.If reference point Ref _texist, first in the multidimensional abnormity diagnosis analysis of setting up procedure 4 dimension, the second dimension and the third dimension and extremely combine differentiation, output " doubtful abnormal " or " extremely " result, do not empty reference point Ref _t, jump out whole step and enter next cycle.

Step 4, in conjunction with Fig. 5, multidimensional abnormity diagnosis is analyzed: as real-time traffic index Index (Tn) _mARK, when the interval PN (Tn) of normal state or the interval NN of negative normality (Tn), do not start the analysis of multidimensional abnormity diagnosis.It is all " normality " that multidimensional abnormity diagnosis is analyzed each dimension default value, different according to the trigger condition of real-time matching IMS process, starts respectively the abnormity diagnosis analytical algorithm of different dimensions, finally enters and extremely combines all conduct inputs of all dimension results while differentiating.

The first dimension is that difference is differentiated: calculate real-time traffic index Index (Tn)<sub TranNum="179">mARK</sub>with the deviation of greatest exponential value MAX (Tn) or minimal index value MIN (Tn), if Index (Tn)<sub TranNum="180">mARK</sub>>MAX (Tn), deviation is positivity bias, real-time traffic index Index (Tn)<sub TranNum="181">mARK</sub>be positioned at positive overflow area PO (Tn), calculate Index (Tn)<sub TranNum="182">mARK</sub>-MAX (Tn), and difference is compared with empirical value (in this implementation column, empirical value is taken as 5).If these difference<5, are output as " doubtful blocking up extremely ", if this difference>=5, be output as " blocking up abnormal ".If Index (Tn)<sub TranNum="183">mAR</sub>k<MIN (Tn), deviation is negative sense deviation, real-time traffic index Index (Tn)<sub TranNum="184">mARK</sub>be positioned at negative overflow area NO (Tn), calculate MIN (Tn)-Index (Tn)<sub TranNum="185">mARK</sub>, and difference is compared with empirical value (in this implementation column, empirical value is taken as 5).If these difference<5, are output as " doubtful extremely unimpeded ", if this difference>=5, be output as " extremely unimpeded ".Empirical value can be demarcated according to concrete road object.

The second dimension ratio is differentiated: calculate real-time traffic index Index (Tn) _mARKwith the scale-up factor of intermediate value MEDIAN (Tn), i.e. Index (Tn) _mARK/ MEDIAN (Tn), if this scale-up factor is greater than MPSD (Tn)/MEDIAN (Tn), real-time traffic index Index (Tn) _mARKbe positioned at PB (Tn) between positive overflow area PO (Tn) Huo Zheng buffer zone.According to threshold value [MAX (Tn)+MPSD (Tn)]/[2 * MEDIAN (Tn)], it is boundary line, be greater than this value and be output as " blocking up abnormal ", be less than this value and be greater than MPSD (Tn)/MEDIAN (Tn), be output as " doubtful blocking up extremely ".If scale-up factor is less than MNSD (Tn)/MEDIAN (Tn), real-time traffic index Index (Tn) _mARKbe positioned at NB (Tn) between negative overflow area NO (Tn) or negative buffer, according to threshold value [MIN (Tn)+MNSD (Tn)]/[2 * MEDIAN (Tn)], it is boundary line, be less than this value and export " extremely unimpeded ", be greater than this value and be less than MNSD (Tn)/MEDIAN (Tn) and export " doubtful extremely unimpeded ".

Third dimension degree trend discrimination: calculate real-time traffic index Index (Tn)<sub TranNum="192">mARK</sub>with MPSD (Tn) or MNSD (Tn) with respect to reference point Ref<sub TranNum="193">t</sub>slope differences, if Index (Tn)<sub TranNum="194">mARK</sub>with respect to reference point Ref<sub TranNum="195">t</sub>slope>MPSD (Tn) is with respect to reference point Ref<sub TranNum="196">t</sub>slope, deviation is positivity bias, real-time traffic index Index (Tn)<sub TranNum="197">mARK</sub>be positioned at PB (Tn) between positive overflow area PO (Tn) Huo Zheng buffer zone, according to current time index and positively biased difference, calculate [Index (Tn)-Ref<sub TranNum="198">t</sub>] * T<sub TranNum="199">0</sub>/ (Tn-t)-[MPSD (Tn)-Ref<sub TranNum="200">t</sub>] * T<sub TranNum="201">0</sub>/ (Tn-t), and wherein, T<sub TranNum="202">0</sub>for the update cycle of traffic index, t be interpolation constantly, if its result>5, output " is blocked up abnormal ", if its result between (0,5] between, output " doubtful block up abnormal ".If Index (Tn)<sub TranNum="203">mARK</sub>with respect to reference point Ref<sub TranNum="204">t</sub>slope<MNSD (Tn) with respect to reference point Ref<sub TranNum="205">t</sub>slope, deviation is negative sense deviation, real-time traffic index Index (Tn)<sub TranNum="206">mARK</sub>be positioned at NB (Tn) between negative overflow area NO (Tn) or negative buffer, according to current time index and negative bias difference, calculate [Index (Tn)-Ref<sub TranNum="207">t</sub>] * T<sub TranNum="208">0</sub>/ (Tn-t)-[MNSD (Tn)-Ref<sub TranNum="209">t</sub>] * T<sub TranNum="210">0</sub>/ (Tn-t), if its result≤-5, output " extremely unimpeded ", if its result between (5,0] between, output " doubtful block up abnormal ".Threshold value can be demarcated according to concrete road object.

Step 6, extremely combine differentiation: according to the judgement demand of blocking up abnormal, the abnormality diagnostic decision rule of customizable expansion multidimensional.If Index (Tn)<sub TranNum="212">mARK</sub>>=MEDIAN (Tn), the decision rule of all dimensions is all exported: { " normality ", " doubtful blocking up extremely ", " blocking up abnormal " }, if Index (Tn)<sub TranNum="213">mARK</sub><MEDIAN (Tn), output: { " normality ", " doubtful extremely unimpeded ", " extremely unimpeded " }.Being input as of associating differentiation: { the first dimension output, the second dimension output, the output of third dimension degree ..., according to the difference to IMS degree of dependence, can be divided into two kinds of logic identification abnormal traffic state features: " an abnormal voting adopted is fixed " and " the minority is subordinate to the majority ".

As long as " an abnormal voting adopted is fixed ": at from " normality " to " extremely " cognitive phase, the first dimension result of determination has a ticket power to make decision, thinks Index (Tn) _mARKin index memory space IMS (Tn), be not just identified as doubtful abnormal or abnormal; And from " extremely ", returning to " normality " stage, and in the first dimension result of determination, be on " normality " basis, by other dimensions, jointly confirm abnormal restoring.

" the minority is subordinate to the majority ": all dimension diagnostic results have identical weight, the majority of judging according to output is judged final recognition result.Three dimensions of take are example, and two " normalities " are identified as normality, if three result outputs are all not identical, can according to " criterion of pessimism " towards " extremely " direction, identify respectively, or identify according to " criterion of optimism " towards " normality " direction.

In actual applications, the target road network scale of identification is different as required, can use respectively Different Logic, Ru Dui through street whole network, surface road the whole network or administrative section are identified extremely because whole index fluctuation range is less, recommendation " an abnormal voting adopted is fixed " identification, Ru Duimoutiao section or path, because index fluctuation range is larger, recommendation " the minority is subordinate to the majority " identification.

7) history index upgrades: real-time traffic index Index (Tn) _mARKcomplete after index memory space IMS (Tn) coupling, be stored in original historical data base, matching stage is used index memory space IMS (Tn) not comprise current time latest index value.When index memory space IMS (Tn) whole updating, the measurement period that time slice covers is according to removing one day the earliest, add and slide for up-to-date one day, according to forgetting one day farthest, remembeing up-to-date one day, guarantee that the classification historical statistics cycle is constant value.

In addition, the exponential quantity that the inventive method is used is all through data check screening and repairing, and quality of data work for the treatment of completed before each link involved in the present invention.Small part date or time shortage of data does not affect the unitary construction of index memory space IMS (Tn).

With a specific embodiment, illustrate the present invention below.

Step 1) the real-time index result of mark: the influence factor mark MARK that export by traffic impact set of factors (1) _t(1,2,3), carry out mark to T up-to-date real-time index Index of the moment (T), and (2) generate the data I ndex (T) with mark result _mARKand deposit historical data base (step 2) in, (3) mate IMS simultaneously _t(step 4).

Step 2) create index classification history library: (1) is in the original historical data base of index, according to MARK _t(1,2,3) mark result is divided into groups (containing the up-to-date moment) to historical data in the same time, and Main classification influence factor is calendar, the factors such as weather and activity are as auxiliary packet factor, and (2) take sorted historical data as Exponential Sample overall building IMS (step 3).

Step 3) build IMS:1) each sorted T moment history index sample overall (step 2) is carried out to " memory space " establishment, (1) comprise that intermediate value line MEDIAN (T), positive extreme value MAX (T), negative pole value MIN (T), positively biased difference MPSD (T)=MEDIAN (T)+stdev (T) are if MPSD (T) >=MAX (T), get MAX (T) value, negative bias difference MNSD (T)=MEDIAN (T)-stdev (T) if MNSD (T)≤MIN (T) gets MIN (T) value; (2) after " five values " line creates, Nature creating “Liu territory " PO (T), PB (T), PN (T), NN (T), NB (T), NO (T), if MPSD (T)=MAX (T), there is no PB (T) interval, if in like manner MNSD (T)=MIX (T), does not have NB (T) interval; (3) creating IMS memory space totally carries out based on sub-index historical data, therefore do not need to carry out in real time, in system idle periods unification in morning every day, be each road, each cycle to carry out IMS establishment (step 7), result data is that the real-time index-matched of second day is prepared.

Step 4) real-time index-matched IMS: with MARK _tthe real-time index Index (T) of (1,2,3) mark _mARK(step 1) coupling IMS _t(step 3), according to “Liu territory " differentiation Index (T) _mARKbetween location; (1) if [Index (T) _mARK∈ PN (T)] ∩ [Index (T) _mARK∈ NN (T)], directly result is differentiated in output " normality ", empties reference point Ref _tand jump out and differentiate to wait for and to enter the T+I cycle; (2) if [Index (T) _mARK∈ PB (T)] ∩ [Index (T) _mARK∈ NB (T)], differentiate and whether had reference point Ref _t, if there is no, generate reference point Ref _t, and output " normality " result waits for and to enter the T+1 cycle, if there is Ref _t, start multidimensional abnormity diagnosis analysis (the second dimension, the third dimension) (step 5) and extremely combine differentiation (step 6), obtain status flag and differentiate result, wait for and enter the T+I cycle; (3) if [Index (T) _mARX∈ PO (T)] ∩ [Index (T) _mARK∈ NO (T)], differentiate and whether had reference point Ref _t, if there is no, generate reference point Ref _t, and start multidimensional abnormity diagnosis and analyze (all dimensions) (step 5) and extremely combine differentiations (step 6), obtain status flag differentiation result, and wait enters the T+I cycle, if reference point Ref _texist, start multidimensional abnormity diagnosis and analyze (all dimensions) (step 5) and extremely combine differentiation (step 6), output " doubtful abnormal " or " extremely " result, do not empty reference point Ref _twait enters the T+1 cycle.

Step 5) multidimensional abnormity diagnosis is analyzed: as Index (T) _mARKnot when PN (T) or NN (T) (step 4), start the analysis of multidimensional abnormity diagnosis.It is all " normality " that multidimensional abnormity diagnosis is analyzed each dimension default value, different according to the trigger condition of real-time matching IMS process, start respectively the abnormity diagnosis analytical algorithm of different dimensions, finally enter all conduct inputs of all dimension results while extremely combining differentiation (step 6).

(1) first dimension is that difference is differentiated: calculate in real time Index (T)<sub TranNum="251">mARK</sub>with the positive error of MAX (T) or MIN (T), if Index (T)<sub TranNum="252">mARK</sub>∈ PO (T), Index (T)<sub TranNum="253">mARK</sub>-MAX (T) also judges it is in " doubtful blocking up extremely " (<5) or " blocking up abnormal " (>=5 according to threshold value (as poor in 5 vertex degrees)), if Index (T)<sub TranNum="254">mARK</sub>∈ NO (T), MN (T)-Index (T)<sub TranNum="255">mARK</sub>and be in " doubtful extremely unimpeded " (<5) or " extremely unimpeded " (>=5 according to threshold decision); Threshold value can be demarcated according to concrete road object.

(2) second dimension ratios are differentiated: calculate in real time Index (T) _mARKwith the scale-up factor of MEDIAN (T), i.e. Index (T) _mARK/ MEDIAN (T), if [Index (T) _mARK∈ PO (T)] ∩ [Index (T) _mARK∈ PB (T)], according to threshold value [MAX (T)+MPSD (T)]/(2MEDIAN (T)), it is boundary line, be greater than and sentence this value and sentence " block up abnormal ", be less than this value and be greater than MPSD (T)/MEDIAN (T) and judge " doubtful blocking up extremely ", if [Index (T) _mARK∈ NO (T)] ∩ [Index (T) _mARK∈ NB (T)], according to threshold value [MN (T)+MNSD (T)]/(2MEDIAN (T)), it is boundary line, be less than and sentence this value and sentence " extremely unimpeded ", be greater than this value and be less than MNSD (T)/MEDIAN (T) and judge " doubtful extremely unimpeded ".

(3) third dimension degree trend discrimination: calculate in real time Index (T) _mARKwith MPSD (T) or MNSD (T) with respect to reference point Ref _tslope differences, if [Indcx (T) _mARK∈ _pO(T)] ∩ [INdex (T) _mARK∈ PB (T)], according to current time index and positively biased difference, calculate [Index (T) _mARK-REF _t] T ₀/ (T-t)-[MPSD (T)-REF _t] T ₀/ (T-t), according to threshold value 5 vertex degrees, judge, if>5, sentence " blocking up abnormal ", between (0,5] between sentence " doubtful block up abnormal ", if [Index (T) _mARK∈ NO (T)] ∩ [Index (T) _mARK∈ NB (T)], according to current time index and negative bias difference, calculate [INdex (T) _mARK-REF _t] T ₀/ (T-t-[MNSD (T)-REF _t] T ₀/ (T-t), according to threshold value-5 vertex degree, judge, if≤-5, sentence " extremely unimpeded ", between (5,0] between sentence " doubtful block up abnormal "; Threshold value can be demarcated according to concrete road object.

Step 6) extremely combine differentiation: according to the judgement demand of blocking up abnormal, the decision rule of customizable expansion multidimensional abnormity diagnosis (step 5).If Index (T)<sub TranNum="282">mARK</sub>>=MEDIAN (T), the decision rule of all dimensions is all exported: { " normality ", " doubtful blocking up extremely ", " blocking up abnormal " }, if Index (T)<sub TranNum="283">mARK</sub><MEDIAN (T), output: { " normality ", " doubtful extremely unimpeded ", " extremely unimpeded " }.Being input as of associating differentiation: the first dimension output, the second dimension output, the output of third dimension degree ..., according to the difference to IMS degree of dependence, can be divided into two kinds of logic identification abnormal traffic state features: " a ticket power to make decision " and " the minority is subordinate to the majority ".

In actual applications, the target road network scale of identification is different as required, can use respectively Different Logic, Ru Dui through street whole network, surface road the whole network or administrative section are identified extremely because whole index fluctuation range is less, use " an abnormal voting adopted is fixed " identification, Ru Duimoutiao section or path, because index fluctuation range is larger, used " the minority is subordinate to the majority " identification.

Step 7) history index upgrades: real-time Index (T) _mARKcomplete after IMS coupling, be stored in original historical data base, matching stage is used IMS memory space not comprise current time latest index value (step 4).The measurement period that time slice T covers, according to removing one day the earliest, added and slided for up-to-date one day when IMS memory space unitary construction every day, guaranteed that the classification historical statistics cycle is constant value.

Claims (6)

1. real-time index-matched is remembered an interval abnormal traffic state characteristic recognition method, it is characterized in that, step is:

Step 1, in all traffic flow parameters, choose arbitrarily a real time data at the traffic flow parameter of the interval dynamic fluctuation of fixed data as traffic status identification parameter, obtain all historical traffic status identification supplemental characteristic of some time period Nei Moutiao sections or Regional Road Network, according to all traffic status identification parameters under synchronization, create the traffic status identification parameters memorizing space corresponding with this moment and calculate the intermediate value in each traffic status identification parameters memorizing space, wherein, traffic status identification parameters memorizing space corresponding to T is IMS (T) constantly, minimal index value MIN (T) and greatest exponential value MAX (T) in all historical traffic status identification supplemental characteristic of T is negative edge and the positive boundary of traffic status identification parameters memorizing space IMS (T) constantly, the intermediate value MEDIAN (T) of traffic status identification parameters memorizing space IMS (T) is the average of all traffic status identification supplemental characteristics under moment T,

Step 2, each traffic status identification parameters memorizing space is at least divided between Jian Ji negative buffer, positive buffer zone, and will be less than the zone definitions Wei Fu overflow area of minimal index value, will be greater than the zone definitions Wei Zheng overflow area of greatest exponential value; Between the positive buffer zone of traffic status identification parameters memorizing space IMS (T), be PB (T), span be (MEDIAN (T), MAX (T)]; Between negative buffer, be NB (T), span be [MIN (T), MEDIAN (T)); Negative overflow area is NO (T), span be [MIN, MIN (T)); Positive overflow area is PO (T), span be (MAX (T), MAX], wherein, MIN is the minimum possible value of traffic status identification parameter, MAX is the maximum possible value of traffic status identification parameter;

Step 3 is obtained the real-time traffic states identification parameter Index (Tn) of Gai Tiao section under current time Tn, judge whether real-time traffic states identification parameter Index (Tn) is positioned between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn) NB (Tn) between PB (Tn) or negative buffer, if, the traffic behavior of Ze Gaitiao section under current time Tn is normality, if not, the traffic behavior of Ze Gaitiao section under current time Tn is abnormal, this extremely refers to extremely unimpeded or extremely blocks up, to be that traffic status identification parameter is more high more block up the traffic status identification parameter model that Ruo Gaitiao section adopts, if real-time traffic states identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely blocking up, if real-time traffic states identification parameter Index (Tn) is positioned at negative overflow area NO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, the traffic status identification parameter model that Ruo Gaitiao section adopts is that traffic status identification parameter is more high more unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at the traffic behavior of negative overflow area NO (Tn) ，Gai Tiao section under current time Tn for extremely blocking up.

2. a kind of real-time index-matched as claimed in claim 1 is remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, in described step 1, utilize influence factor mark MARK, all historical traffic identification parameter data to some time period Nei Moutiao section are carried out mark, influence factor mark MARK has on traffic behavior feature one group of cartesian product configuration item that the non-traffic factor of extensive impact forms in influence factor config set, all historical traffic identification parameter under synchronization is classified according to influence factor mark MARK, for each class creates respectively traffic status identification parameters memorizing space and calculates the intermediate value in each traffic status identification parameters memorizing space,

In described step 3, obtain after the real-time traffic identification parameter Index (Tn) under current time Tn, first to it, utilize influence factor mark MARK to carry out mark, obtain the traffic identification parameter Index (Tn) with mark result _mARK, utilize traffic identification parameter Index (Tn) _mARKfind constantly the traffic status identification parameters memorizing space corresponding with its classification under Tn, then judge that according to the relation of between itself and Jian Ji negative buffer, Zhong Zheng buffer zone, traffic status identification parameters memorizing space and Ji Zheng overflow area, negative overflow area whether the traffic behavior of Gai Tiao section under current time Tn be abnormal.

3. a kind of real-time index-matched as claimed in claim 2 is remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, the described non-traffic factor that traffic behavior feature is had to an extensive impact at least comprises Weather information Weather, calendar information Calendar and action message Event, wherein, calendar information Calendar is main factor, each has different influence factor mark MARK constantly, and the influence factor of T is labeled as MARK constantly _t(1,2,3)=MARK _t(Calendar, Weather, Event).

4. a kind of real-time index-matched as claimed in claim 1 is remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, in described step 2, by each traffic status identification parameters memorizing space be at least divided between positive buffer zone, normal state is interval, between negative normality interval and negative buffer; Between the positive buffer zone of traffic status identification parameters memorizing space IMS (T), the span of PB (T) is (MPSD (T), MAX (T)], MPSD (T)=MEDIAN (T)+stdev (T), stdev (T) is the standard deviation of all historical traffic identification parameters under moment T; Normal state interval is PN (T), and span is [MEDIAN (T), MPSD (T)]; Negative normality interval is NN (T), span be [MNSD (T), MEDIAN (T)), MNSD (T)=MEDIAN (T)-stdev (T); The span of NB between negative buffer (T) be [MIN (T), MNSD (T));

In described step 3, obtain the real-time traffic identification parameter Index (Tn) of Gai Tiao section under current time Tn, judge whether real-time traffic identification parameter Index (Tn) is positioned at the interval PN (Tn) of normal state or the negative normality interval (Tn) of traffic status identification parameters memorizing space IMS (Tn), if, the traffic behavior of Ze Gaitiao section under current time Tn is normal, if not, the traffic behavior of Ze Gaitiao section under current time Tn is abnormal, this extremely refers to extremely unimpeded or extremely blocks up, to be that traffic identification parameter is more high more block up the traffic identification parameter model that Ruo Gaitiao section adopts, if real-time traffic identification parameter Index (Tn) is positioned at PB (Tn) Huo Zheng overflow area PO (Tn) between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely blocking up, if real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) Huo Fu overflow area NO (Tn) between the negative buffer of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, the traffic identification parameter model that Ruo Gaitiao section adopts is that traffic identification parameter is more high more unimpeded, if real-time traffic identification parameter Index (Tn) is positioned at PB (Tn) Huo Zheng overflow area PO (Tn) between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn), the traffic behavior of Gai Tiao section under current time Tn is for extremely unimpeded, if real-time traffic identification parameter Index (Tn) is positioned between the negative buffer of traffic status identification parameters memorizing space IMS (Tn) traffic behavior of NB (Tn) Huo Fu overflow area NO (Tn) ，Gai Tiao section under current time Tn for extremely blocking up.

5. a kind of real-time index-matched as claimed in claim 1 is remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, in described step 3, if real-time traffic identification parameter Index (Tn) is positioned between the positive buffer zone of traffic status identification parameters memorizing space IMS (Tn) NB (Tn) between PB (Tn) or negative buffer, adopt any one in following two methods to judge the traffic behavior of current section under current time Tn:

First method: if real-time traffic identification parameter Index (Tn) is positioned at PB between positive buffer zone (Tn), threshold value [MAX (Tn)+MPSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, be greater than this value and sentence " extremely ", be less than this value and be greater than MPSD (Tn)/MEDIAN (Tn) and judge " doubtful abnormal "; If real-time traffic identification parameter Index (Tn) is positioned at NB between negative buffer (Tn), threshold value [MIN (Tn)+MNSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, be less than this value and sentence " extremely ", be greater than this value and be less than MNSD (Tn)/MEDIAN (Tn) and judge " doubtful abnormal ";

Second method: judge whether to exist reference point Ref _t, reference point Ref _twhile entering first Huo Fu overflow area, Jian,Zheng overflow area, Jian, negative buffer, positive buffer zone by real-time traffic identification parameter Index (Tn) by the interval PN (Tn) of normal state or the interval NN of negative normality (Tn), intersecting interpolation with MPSD (Tn) or MNSD (Tn) creates, interpolation is constantly rounded to and intersects previous moment constantly, if not, sentence " normality ", if, judge that real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, if be positioned at PB between positive buffer zone (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MPSD (Tn)-Ref _t] * T ₀/ (Tn-t), and wherein, T ₀for the update cycle of traffic identification parameter, t be interpolation constantly, if its result is greater than A, sentence " extremely ", between (0, A] between sentence " doubtful extremely ", A is empirical value; If be positioned at NB between negative buffer (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MNSD (Tn)-Ref _t] * T ₀/ (Tn-t), if be not more than-A of its result sentences " extremely ", between (A, 0] between sentence " doubtful abnormal ".

6. a kind of real-time index-matched as claimed in claim 1 is remembered interval abnormal traffic state characteristic recognition method, it is characterized in that, utilizes three-dimensional extremely to combine the traffic behavior that judges Mou Tiao section in described step 3, and described step 3 comprises:

Step 3.1, obtain Gai Tiao section or the real-time traffic identification parameter Index (Tn) of Regional Road Network under current time Tn;

Step 3.2, judge that whether real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, if not, judges whether to exist reference point Ref _t, reference point Ref _twhile entering Huo Fu overflow area, Jian,Zheng overflow area, Jian, negative buffer, positive buffer zone by real-time traffic identification parameter Index (Tn), intersect interpolation with MPSD (Tn) or MNSD (Tn) and create, if exist, directly enter next step, if do not exist, created reference point Ref _tafter enter next step; If real-time traffic identification parameter Index (Tn) is positioned at NB (Tn) between PB between positive buffer zone (Tn) or negative buffer, judge whether to exist reference point Ref _tif, exist, sentence " normality ", completing steps 3, if do not exist, has created reference point Ref _tafter enter step 3.4;

Step 3.3, the first dimension abnormality juding:

If real-time traffic identification parameter Index (Tn) is positioned at positive overflow area PO (Tn), calculate Index (Tn)-MAX (Tn), if this difference is less than A, sentence " doubtful abnormal ", enter next step, otherwise, sentence " extremely ", enter next step; If real-time traffic identification parameter Index (Tn) is positioned at negative overflow area NO (Tn), calculate MIN (Tn)-Index (Tn), if this difference is less than A, sentence " doubtful abnormal ", enter next step, otherwise, sentence " extremely ", enter next step, A is empirical value;

Step 3.4, the second dimension abnormality juding:

If real-time traffic identification parameter Index (Tn) is positioned at PB between positive buffer zone (Tn) Huo Zheng overflow area PO (Tn), threshold value [MAX (Tn)+MPSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, Index (Tn)/MEDIAN (Tn) is greater than this value and sentences " extremely ", enter next step, Index (Tn)/MEDIAN (Tn) is less than this value and is greater than MPSD (Tn)/MEDIAN (Tn) and judges " doubtful abnormal ", enters next step; If real-time traffic identification parameter Index (Tn) is positioned at NB between negative buffer (Tn) Huo Fu overflow area NO (Tn), threshold value [MIN (Tn)+MNSD (Tn)]/[2 * MEDIAN (Tn)] of take is boundary line, Index (Tn)/MEDIAN (Tn) is less than this value and sentences " extremely ", enter next step, Index (Tn)/MEDIAN (Tn) is greater than this value and is less than MNSD (Tn)/MEDIAN (Tn) and judges " doubtful abnormal ", enters next step;

Step 3.5, third dimension degree abnormality juding:

If be positioned at PB between positive buffer zone (Tn) Huo Zheng overflow area PO (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MPSD (Tn)-Ref _t] * T ₀/ (Tn-t), and wherein, T ₀for the update cycle of traffic identification parameter, t be interpolation constantly, if its result is greater than A, sentence " extremely ", enter next step, between (0, A] between sentence " doubtful extremely ", enter next step; If be positioned at NB between negative buffer (Tn) Huo Fu overflow area NO (Tn), calculate [Index (Tn)-Ref _t] * T ₀/ (Tn-t)-[MNSD (Tn)-Ref _t] * T ₀/ (Tn-t), if be not more than-A of its result sentences " extremely ", enter next step, between (A, 0] between sentence " doubtful abnormal ", enter next step;

Step 3.6, adopt " an abnormal voting adopted is fixed " or " the minority is subordinate to the majority " extremely to combine differentiation, wherein, " an abnormal voting adopted fixed " refers to: at traffic behavior from " normality " to " extremely " cognitive phase, the first dimension abnormality juding result described in step 3.3 has a ticket power to make decision; And from " extremely ", returning to " normality " stage, the result at the first dimension abnormality juding is on " normality " basis, by the second dimension abnormality juding described in step 3.4 and the third dimension degree abnormality juding described in step 3.5, jointly confirms abnormal restoring;

" the minority is subordinate to the majority " refers to: the second dimension abnormality juding result described in the first dimension abnormality juding result, 3.4 described in step 3.3 and the third dimension degree abnormality juding result described in step 3.5 have identical weight, according to the majority of output result of determination, judges final recognition result; If the output of the result of determination of three dimensionality is all not identical, or according to " criterion of pessimism " towards " extremely " direction identifies, or identify according to " criterion of optimism " towards " normality " direction.

Images