CN113344239B - Traffic congestion condition prediction method and system based on two-stage spectral clustering - Google Patents
Traffic congestion condition prediction method and system based on two-stage spectral clustering Download PDFInfo
- Publication number
- CN113344239B CN113344239B CN202110447756.0A CN202110447756A CN113344239B CN 113344239 B CN113344239 B CN 113344239B CN 202110447756 A CN202110447756 A CN 202110447756A CN 113344239 B CN113344239 B CN 113344239B
- Authority
- CN
- China
- Prior art keywords
- street
- traffic flow
- clustering
- period
- time period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 230000003595 spectral effect Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000013508 migration Methods 0.000 claims abstract description 44
- 230000005012 migration Effects 0.000 claims abstract description 44
- 239000011159 matrix material Substances 0.000 claims abstract description 33
- 238000010586 diagram Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002092 carbon dioxide Inorganic materials 0.000 description 1
- 239000001569 carbon dioxide Substances 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Operations Research (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a traffic jam condition prediction method and system based on two-stage spectral clustering, which comprises the steps of firstly, acquiring time information, street information and meteorological information of urban streets; then, clustering the streets based on two-level spectral clustering to obtain the class cluster of the streets in each time interval; predicting the street traffic flow of the next period, the street traffic flow of each time period in the next period and a street vehicle migration matrix by using a gradient enhanced regression tree model and a multi-similarity reasoning model; and finally, comprehensively analyzing the street congestion condition according to the predicted street traffic flow of the next period, the street traffic flow of each time period in the next period and the street vehicle migration matrix, and dredging vehicles.
Description
Technical Field
The invention belongs to the field of traffic congestion condition prediction, and particularly relates to a traffic congestion condition prediction method and system based on two-stage spectral clustering.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
With the development of economy, in the aspect of traveling, modes are more and more, the modes of motor vehicles such as private cars, trains and public transport means for people to choose for traveling are gradually changed, and the appearance of the motor vehicle traveling modes is convenient for the life of people to a certain extent, so that the time is saved for the people to go out, and the cost is saved. However, as more and more vehicles are used on the street, the traffic jam occurs, and the jam is slow to dredge for a long time. The problem of traffic jam is solved, the emission of carbon dioxide can be reduced, and the environment is protected.
The traffic jam situation can be predicted to adjust the traffic flow of different streets in advance, the problem of uneven traffic flow of different streets is solved, and the traffic is guaranteed to keep smooth at any time and any place, so that the traffic jam problem is solved, therefore, how to predict the traffic flow of the urban traffic intersection in a period of time in advance is a technical problem to be solved urgently.
Disclosure of Invention
The invention provides a method and a system for predicting traffic jam conditions based on two-stage spectral clustering to solve the problems,
according to some embodiments, the invention adopts the following technical scheme:
a traffic jam condition prediction method based on two-stage spectral clustering comprises the following steps:
acquiring street information and meteorological information of each time period of a city;
clustering is carried out according to street information of each time interval to obtain a cluster to which the street of each time interval belongs;
predicting the street traffic flow of the next period by using a gradient enhanced regression tree model according to the cluster to which the street belongs and meteorological information at each time interval;
according to the cluster to which the street in each time period belongs and meteorological information, a multi-similarity reasoning model is used for predicting the traffic flow of each street in different time periods in the next period;
and determining the street congestion condition according to the predicted street traffic flow of the next period and the traffic flow of each street in different time periods in the next period.
Further, the specific steps of clustering according to street information of each time interval include:
dividing streets in each time period into a plurality of levels according to the street traffic flow and the street traffic flow threshold value in each time period;
and clustering the streets of each level by using a two-level spectral clustering algorithm to obtain the clusters to which the streets belong in each time interval.
Further, the two-stage spectral clustering algorithm comprises the following specific steps:
(1) Counting a street vehicle migration matrix in each time period according to street information in each time period;
(2) Clustering streets in each time interval into a plurality of clusters by using spectral clustering according to the positions of the streets;
(3) Calculating a vehicle migration matrix among the similar clusters according to the street vehicle migration matrix at each time interval;
(4) Clustering by reusing a spectral clustering algorithm according to the street position, the street traffic flow of each time period and the vehicle migration matrix among the clusters to obtain the latest clustering result;
(5) And (4) repeatedly executing the steps (3) and (4) until the clustering result is not changed any more.
Further, the gradient enhanced regression tree model predicts residuals of previous trees by constructing each successive tree.
Further, the determining of the street congestion condition specifically includes:
judging whether the traffic flow of the street in a certain time period exceeds the traffic flow of the street in the next period, if so, judging that the street is very congested in the time period; if not, the street is generally congested during the time period.
Further, the method for predicting traffic congestion conditions based on two-stage spectral clustering further comprises the following steps: and predicting the immigration values and the immigration values of the streets in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which the streets belong and the meteorological information in each time period.
A traffic jam prediction system based on two-stage spectral clustering comprises:
the data acquisition module is used for acquiring street information and weather information of each time period of a city;
the clustering module is used for clustering according to street information of each time interval to obtain a cluster to which the street of each time interval belongs;
the first prediction module is used for predicting the street traffic flow of the next period by using a gradient enhanced regression tree model according to the cluster to which the street belongs and the meteorological information in each time interval;
the second prediction module is used for predicting the traffic flow of each street in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which the street in each time period belongs and the meteorological information;
and the street congestion condition determining module is used for determining the street congestion condition according to the predicted street traffic flow of the next period and the traffic flow of each street in different time periods in the next period.
A computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to execute a method for predicting traffic congestion conditions based on two-stage spectral clustering.
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the traffic jam condition prediction method based on two-stage spectral clustering.
Compared with the prior art, the invention has the beneficial effects that:
the invention firstly uses a two-level spectral clustering algorithm to perform clustering, and divides individuals or objects into categories according to the similarity degree, so that the similarity between elements in the same category is stronger than that of elements in other categories, and the follow-up learning and prediction are convenient to perform.
The two-stage spectral clustering algorithm has high speed and strong generalization capability, and can well predict and solve the problem of traffic jam.
According to the invention, the migration matrix is added for clustering, the flow trend among streets is considered, and the obtained clustering result is more accurate.
The invention uses the gradient enhanced regression tree (GBRT) and the multi-similarity reasoning Model (MSI), has high model precision and strong error correction capability, and can well meet the requirements of people on transportation.
According to the street traffic flow and the street traffic flow threshold value of each time interval, the street of each time interval is divided into a plurality of levels; and clustering the streets at each level by using a two-level spectral clustering algorithm to obtain the clusters to which the streets belong at each time interval, so that the clustering precision is improved, and the clustering result is optimized.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a block diagram of a traffic congestion prediction model;
FIG. 3 is a diagram of a gradient enhanced regression tree (GBRT) structure;
FIG. 4 is a diagram of a multiple similarity inference Model (MSI) architecture.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example 1
As shown in fig. 1 and fig. 2, the method for predicting traffic congestion conditions based on two-stage spectral clustering according to the present embodiment includes the following steps:
step 1, street information and meteorological information of each time period in a plurality of historical periods of a city are obtained;
the street information comprises street positions, street traffic flow of each time interval and traffic flow directions of each road junction of each time interval;
in this embodiment, one cycle T is divided into a plurality of time periods T, and the street traffic volume, the street vehicle migration matrix, and the weather information of each time period within the past 7 days, that is, 7 cycles (days), are acquired, and each day is divided into 4 time periods: a peak period between weekdays and weekends (07-00, 00), a daytime period (11;
the weather information includes: weather W, temperature K and wind speed Y, and extracting historical weather information in each time period; in this embodiment, the weather can be classified into five types of common weather, namely sunny weather, cloudy weather, rainy weather, snowy weather and heavy fog.
Street location, i.e., geographic information L of the street (including longitude and latitude coordinates of the street start and end points);
the street traffic flow N is the traffic flow of each street in each historical time period;
data of the flowing direction of the vehicles among the intersections in each time period;
counting street vehicle migration matrixes in each time interval according to street information in each time interval, and specifically counting street vehicle migration matrixes in each time interval according to the traffic flow direction at an intersection;
and (3) obtaining a vehicle migration matrix of each street in each time period according to the statistics of the flowing direction data of the vehicles at each intersection in each time period: counting an inter-street migration trend matrix in a time period t according to the flowing direction data of vehicles at each intersection in each time period, wherein Ci and Cj in the migration trend matrix represent migration records of streets, namely the number of the vehicles migrating from the street Ci to the street Cj in the time period t; the vehicle migration matrix of the street in each time interval is historical migration data of each street in each time interval, specifically, for each street number, vehicles can flow among the streets, the number of vehicles in each street in each time interval transferred to other streets and the number of vehicles in other streets transferred to the street are calculated, and the vehicle migration matrix of the streets, namely the vehicle migration matrix of each street is formed.
For example, a migration matrix of a street numbered 1 during a day may be a first column numbered with other streets, a second column numbered with vehicles for the street numbered 1 to migrate to other streets during the time period, and a second column numbered with vehicles for the street numbered 1 to migrate to other streets during the time period.
And 2, clustering urban streets by adopting a two-stage spectral clustering algorithm according to the street positions, the street traffic flow of each time period and the street vehicle migration matrix, dividing the streets into a plurality of clusters, and obtaining a cluster C to which each street belongs and the vehicle migration matrix among the clusters.
Before performing two-stage spectral clustering, streets in different time periods are divided into two levels of congestion and general level according to the traffic flow of the streets in each time period through a manually set threshold.
For streets with general grades, clustering is carried out by using a two-grade spectral clustering algorithm according to the geographic information L of the streets, the traffic flow N of the streets and the belonged time period t, the streets with the general grades are clustered into a plurality of classes, and a clustering result C is obtained 1 。
For the streets with the grades of congestion, clustering is carried out by using a two-grade spectral clustering algorithm according to the geographic information L of the streets, the traffic flow N of the streets and the belonged time period t, the streets with the grades of congestion are clustered into a plurality of classes, and a clustering result C is obtained 2 。
Clustering the result C 1 And C 2 And combining the clusters to obtain the cluster to which the street belongs in each time interval.
The two-stage spectral clustering algorithm comprises the following specific steps:
(1) Counting street vehicle migration matrixes in each time interval according to street information in each time interval;
(2) Clustering by using spectral clustering according to the street geographic information L, and clustering traffic flows of different streets in each time period in a city into a plurality of classes;
(3) Calculating vehicle migration matrixes among the clusters according to the vehicle migration matrixes of the streets;
(4) Clustering by using a spectral clustering algorithm again, and clustering according to the street geographic information L, the street traffic flow N, the belonged time period t and the migration matrix to obtain the latest clustering result;
(5) And (4) repeatedly executing the steps (3) and (4) until the number of iterations reaches the maximum value or the clustering result is not changed any more.
And 3, predicting the traffic flow of each street in a next period T (namely one day) by using a gradient enhanced regression tree (GBRT) prediction model according to the cluster to which each street in the past 7 days belongs, the time T to which each street belongs and the meteorological information obtained in the step 2, wherein in the embodiment, the meteorological information only comprises weather when the traffic flow of each street in the next period is predicted by using the gradient enhanced regression tree (GBRT) prediction model.
And training by using a training data set to obtain a gradient enhancement regression tree prediction model, wherein each piece of data in the training set comprises a time interval t of streets in a plurality of historical periods, a cluster C which belongs to the time interval t, weather W and a label (the traffic flow of the street in the next period).
Inputting the time period, the cluster and the weather information of the streets in a plurality of historical periods into a trained gradient enhanced regression tree prediction model, and predicting to obtain the traffic flow S of each street in the next period 1 。
In the present embodiment, one cycle can be divided into 4 periods, for example, one day can be divided into a peak period (07
As shown in FIG. 3, the gradient enhanced regression Tree (GBRT) prediction algorithm calculates a series of simple regression trees { g }collectively 1 (x),g 2 (x),...，g r (x) And b, constructing each continuous tree, and predicting the residual error of the previous tree according to the method, wherein the formula is as follows:
wherein L in the formula represents a loss function,
represents a training set, G (x) = G 1 (x)+g 2 (x)+…+g r (x)，y t Represents the sum of the actual traffic flow of all streets of the t-th cycle, x t And G (x) represents the traffic flow sum of all streets predicted in the t-th period.
The total quantity of all vehicles predicted in the next intra-city is represented by G (x), the total quantity of all vehicles going out of each street in the actual whole city is represented by y, and for the selection of characteristic variables, the embodiment selects time (one day of the week and one hour of the day) and meteorological vectors, and trains a GBRT prediction model meeting the total quantity of vehicles going out of each street in the actual whole city according to the obtained historical data sets (including the time data set and the meteorological characteristic data set and the data set y of all vehicles going out of each street in the actual whole city).
And 4, predicting the traffic flow and the traffic flow percentage of each street in the next period (one day) in different time periods by using a multi-similarity reasoning Model (MSI) according to the cluster, the time information and the meteorological information of the streets in each time period in the past 7 days obtained in the step 2.
For the multi-similarity reasoning model, the input characteristics are a clustering result C, the belonged time period t, the weather W, the temperature K and the wind speed Y, and the output results are the traffic flow and the traffic flow percentage of each street in the next period in different time periods.
In this embodiment, the percentage of the vehicle flow refers to the ratio of the vehicle flow to the total vehicle flow on the day for a certain period of time.
The number of motor vehicles in each street of a city is predetermined using a multiple similarity inference Model (MSI). As shown in fig. 4, the multiphase analogy inference model assumes that the proportion of motor vehicle traveling (i.e., the percentage of traffic flow) in each time period in the next cycle in the future is represented by P1, P2,., PH, and the proportion Pt, t =1, …, H needs to be predicted again; for P1, P2.,. PH, the variables (including the clustering result, the time vector and the meteorological feature vector) corresponding to it are labeled as f1, f2,..,. FH, and based on this, the similarity W (f 1, ft), W (f 2, ft),..,. W (fH, ft) is calculated, where ft is the feature of the t-th time period in the next cycle corresponding to Pt, and W is a function for evaluating the similarity of the two variables.
The multiple similarity inference Model (MSI) is as follows:
the multiple similarity function can be expressed by the following formula:
wherein W represents a multiple similarity function, T represents a sample size of the historical data, E t ×P t And
respectively representing real and predicted traffic flow in different time periods, L representing a loss function of a measurement prediction error, and H representing time.
In order to better complete the prediction of the vehicle flow trend on the street (i.e. the street vehicle migration matrix), another important factor exists, namely, a migration trend matrix is set, m × m is the inter-street (class) migration trend matrix corresponding to the time period T, in the migration trend matrix, ci and Cj represent the migration records of the street (class), i.e. the migration proportion from the street (class) Ci to the street (class) Cj in the time period T, the MSI model is used for learning during the cluster conversion learning of the street (class), and the inter-street (class) migration trend matrix is used for obtaining the vehicle migration matrix of each street in each time period.
According to the cluster to which the street belongs and the meteorological information in each time period, a multi-similarity reasoning model is used for predicting the immigration and emigration values (street vehicle migration matrix) of each street in different time periods in the next period, and the method specifically comprises the following steps: and (3) using a multi-similarity reasoning Model (MSI), inputting a clustering result C, time T, weather W, temperature K and wind speed Y, and predicting the vehicle flowing direction (namely immigration and emigration values, namely a street vehicle migration matrix) of each street at different time periods.
Step 5, obtaining the traffic flow S of each street in the next period 1 The traffic flow S of each street in the next period in different time periods 2 The traffic flow and the congestion condition of the whole city can be known, the traffic congestion condition is analyzed by combining the street vehicle migration matrix, and a scheme for solving the traffic congestion problem is formulated.
Judging whether the traffic flow of the street in a certain time period exceeds the traffic flow of the street in the next period, if so, judging that the street is very congested in the time period; if not, the street is generally congested during the time period.
On the basis of determining the street congestion condition, vehicles can be dredged according to the street congestion condition and a street vehicle migration matrix (the migration values of each street in different time periods in the next period), and the problem of congestion is solved.
Example 2
The embodiment provides a traffic jam prediction system based on two-stage spectral clustering, which comprises:
the data acquisition module is used for acquiring street information and weather information of each time period of a city;
the clustering module is used for clustering according to street information of each time interval to obtain a cluster to which the street of each time interval belongs;
the first prediction module is used for predicting the street traffic flow of the next period by using a gradient enhanced regression tree model according to the cluster to which the street belongs and the meteorological information in each time interval;
the second prediction module is used for predicting the traffic flow of each street in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which the street in each time period belongs and meteorological information;
and the street congestion condition determining module is used for determining the street congestion condition according to the predicted street traffic flow of the next period and the traffic flow of each street in different time periods in the next period.
And the third prediction module is used for predicting the immigration values and the emigration values of the streets in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which the streets belong and the meteorological information in each time period.
And the vehicle dredging module is used for dredging vehicles according to the street congestion condition and the street vehicle migration matrix (the migration values of each street in different time periods in the next period), so that the problem of congestion is solved.
Example 3
The embodiment provides a computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the traffic congestion condition prediction method based on two-stage spectral clustering.
Example 4
A terminal device comprising a processor and a computer readable storage medium, the processor for implementing instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the traffic jam condition prediction method based on two-stage spectral clustering.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive changes in the technical solutions of the present invention.
Claims (7)
1. A traffic jam condition prediction method based on two-stage spectral clustering is characterized by comprising the following steps: the method comprises the following steps:
acquiring street information and meteorological information of each time period of a city; one cycle is divided into a plurality of time periods;
clustering according to street information of each time interval to obtain a cluster to which the street of each time interval belongs;
predicting the street traffic flow of the next period by using a gradient enhanced regression tree model according to the cluster to which the street belongs and meteorological information in each time interval; specifically, inputting the time periods, the clusters and the weather information of the streets in a plurality of historical periods into a trained gradient enhanced regression tree prediction model, and predicting to obtain the traffic flow of each street in the next period;
according to the cluster to which the street in each time period belongs and meteorological information, a multi-similarity reasoning model is used for predicting the traffic flow of each street in different time periods in the next period; the method specifically comprises the following steps: for the multi-similarity reasoning model, inputting the characteristics of a clustering result, the belonged time period, weather, temperature and wind speed, and outputting the result as the traffic flow of each street in different time periods in the next cycle;
according to the cluster to which the street in each time period belongs and meteorological information, predicting the immigration value and the immigration value of each street in different time periods in the next period by using a multi-similarity reasoning model; the method specifically comprises the following steps: inputting clustering results, time, weather, temperature and wind speed by using a multi-similarity reasoning model, and predicting immigration values and emigration values of each street in different time periods;
determining the street congestion condition according to the predicted street traffic flow of the next period and the traffic flow of each street in different time periods in the next period; the method specifically comprises the following steps: judging whether the traffic flow of the street in a certain time period exceeds the traffic flow of the street in the next period, if so, judging that the street is very congested in the time period; if not, the street is generally congested in the time period;
and dredging vehicles according to the street congestion condition and the immigration and emigration values of each street in different time periods in the next period on the basis of determining the street congestion condition.
2. The method for predicting traffic congestion conditions based on two-stage spectral clustering as claimed in claim 1, wherein: the specific steps of clustering according to street information of each time interval comprise:
dividing streets in each time period into a plurality of levels according to the street traffic flow and the street traffic flow threshold value in each time period;
and clustering the streets in each level by using a two-level spectral clustering algorithm to obtain the clusters to which the streets belong in each time period.
3. The method for predicting traffic congestion conditions based on two-stage spectral clustering as claimed in claim 2, wherein: the two-stage spectral clustering algorithm comprises the following specific steps:
(1) Counting a street vehicle migration matrix in each time period according to street information in each time period;
(2) Clustering streets in each time interval into a plurality of clusters by using spectral clustering according to the positions of the streets;
(3) Calculating a vehicle migration matrix among the similar clusters according to the street vehicle migration matrix at each time interval;
(4) Clustering by reusing a spectral clustering algorithm according to the street position, the street traffic flow of each time period and the vehicle migration matrix among the clusters to obtain the latest clustering result;
(5) And (4) repeatedly executing the steps (3) and (4) until the clustering result is not changed any more.
4. The method for predicting traffic congestion conditions based on two-stage spectral clustering as claimed in claim 1, wherein: the gradient enhanced regression tree model predicts the residuals of previous trees by constructing each successive tree.
5. A traffic jam prediction system based on two-stage spectral clustering is characterized in that: the method comprises the following steps:
the data acquisition module is used for acquiring street information and weather information of each time period of a city; one cycle is divided into a plurality of time periods;
the clustering module is used for clustering according to street information of each time interval to obtain a cluster to which the street of each time interval belongs;
the first prediction module is used for predicting the street traffic flow of the next period by using a gradient enhanced regression tree model according to the cluster to which the street belongs and the meteorological information in each time interval; specifically, inputting the time periods, the clusters and the weather information of the streets in a plurality of historical periods into a trained gradient enhanced regression tree prediction model, and predicting to obtain the traffic flow of each street in the next period;
the second prediction module is used for predicting the traffic flow of each street in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which the street in each time period belongs and the meteorological information; the method specifically comprises the following steps: for the multi-similarity reasoning model, inputting the characteristics of a clustering result, the belonged time period, weather, temperature and wind speed, and outputting the result as the traffic flow of each street in different time periods in the next cycle;
the third prediction module predicts the immigration and emigration values of each street in different time periods in the next period by using a multi-similarity reasoning model according to the cluster to which each street belongs and the meteorological information; the method specifically comprises the following steps: inputting clustering results, time, weather, temperature and wind speed by using a multi-similarity reasoning model, and predicting immigration values and emigration values of each street in different time periods;
the street congestion condition determining module is used for determining the street congestion condition according to the predicted street traffic flow of the next period and the traffic flow of each street in different time periods in the next period; the method specifically comprises the following steps: judging whether the traffic flow of the street in a certain time period exceeds the traffic flow of the street in the next period, if so, judging that the street is very congested in the time period; if not, the street is generally congested in the time period;
and dredging vehicles according to the street congestion condition and the immigration and emigration values of each street in different time periods in the next period on the basis of determining the street congestion condition.
6. A computer-readable storage medium, comprising: a plurality of instructions stored therein, the instructions being adapted to be loaded by a processor of a terminal device and to perform a method for two-stage spectral clustering based traffic congestion condition prediction according to any one of claims 1 to 4.
7. A terminal device is characterized in that: the system comprises a processor and a computer readable storage medium, wherein the processor is used for realizing instructions; a computer readable storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform a method for two-stage spectral clustering based traffic congestion condition prediction according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110447756.0A CN113344239B (en) | 2021-04-25 | 2021-04-25 | Traffic congestion condition prediction method and system based on two-stage spectral clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110447756.0A CN113344239B (en) | 2021-04-25 | 2021-04-25 | Traffic congestion condition prediction method and system based on two-stage spectral clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344239A CN113344239A (en) | 2021-09-03 |
| CN113344239B true CN113344239B (en) | 2022-11-29 |
Family
ID=77468522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110447756.0A Expired - Fee Related CN113344239B (en) | 2021-04-25 | 2021-04-25 | Traffic congestion condition prediction method and system based on two-stage spectral clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344239B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217593A (en) * | 2014-08-27 | 2014-12-17 | 北京航空航天大学 | Real-time road condition information acquisition method orienting to cellphone traveling speed |
| CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
| CN107045788A (en) * | 2017-06-28 | 2017-08-15 | 北京数行健科技有限公司 | Traffic Forecasting Methodology and device |
| CN108960476A (en) * | 2018-03-30 | 2018-12-07 | 山东师范大学 | Shared bicycle method for predicting and device based on AP-TI cluster |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100570664C (en) * | 2008-01-11 | 2009-12-16 | 孟小峰 | A kind of system and method thereof of coming the monitoring and controlling traffic congestion based on cluster |
| CN103456192B (en) * | 2013-09-01 | 2015-07-15 | 中国民航大学 | Terminal area prevailing traffic flow recognizing method based on track spectral clusters |
| CN110738856B (en) * | 2019-11-12 | 2020-09-22 | 中南大学 | Mobile clustering-based urban traffic jam fine identification method |
-
2021
- 2021-04-25 CN CN202110447756.0A patent/CN113344239B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217593A (en) * | 2014-08-27 | 2014-12-17 | 北京航空航天大学 | Real-time road condition information acquisition method orienting to cellphone traveling speed |
| CN104778837A (en) * | 2015-04-14 | 2015-07-15 | 吉林大学 | Multi-time scale forecasting method for road traffic running situation |
| CN107045788A (en) * | 2017-06-28 | 2017-08-15 | 北京数行健科技有限公司 | Traffic Forecasting Methodology and device |
| CN108960476A (en) * | 2018-03-30 | 2018-12-07 | 山东师范大学 | Shared bicycle method for predicting and device based on AP-TI cluster |
Non-Patent Citations (1)
| Title |
|---|
| 共享单车系统站点失衡调整及流量预测方法研究;高萌;《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》;20200215;第C034-567页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344239A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109670277B (en) | Travel time prediction method based on multi-mode data fusion and multi-model integration | |
| CN109658695B (en) | Multi-factor short-term traffic flow prediction method | |
| CN107610464A (en) | A kind of trajectory predictions method based on Gaussian Mixture time series models | |
| CN112132264A (en) | Regional exhaust emission prediction method and system based on space-time residual perception network | |
| CN110942211A (en) | Prediction arrival time prediction method and device based on deep neural network | |
| US20240054321A1 (en) | Traffic prediction | |
| CN114450557B (en) | Route deviation quantification and vehicle route learning based thereon | |
| CN107195177A (en) | Based on Forecasting Methodology of the distributed memory Computational frame to city traffic road condition | |
| Lee et al. | Long short-term memory recurrent neural network for urban traffic prediction: A case study of seoul | |
| CN111368879A (en) | Track data mining method based on deep semi-supervised neural network | |
| CN112738209A (en) | Data analysis method based on big data and artificial intelligence and cloud computing server | |
| CN112633602A (en) | Traffic congestion index prediction method and device based on GIS map information | |
| KR20230057558A (en) | Short-term traffic flow prediction method and device using deep learning | |
| Yang et al. | Feature selection in conditional random fields for map matching of GPS trajectories | |
| CN113159403A (en) | Method and device for predicting pedestrian track at intersection | |
| Xing et al. | A short-term traffic flow prediction method based on kernel extreme learning machine | |
| CN113888867B (en) | Parking space recommendation method and system based on LSTM (least squares) position prediction | |
| CN114662376A (en) | Distributed learning for learning context-specific driving patterns | |
| CN113344239B (en) | Traffic congestion condition prediction method and system based on two-stage spectral clustering | |
| CN112710316A (en) | Dynamic map generation focused on the field of construction and localization technology | |
| CN114724414B (en) | Method and device for determining urban air traffic sharing rate, electronic equipment and medium | |
| CN113962460B (en) | Urban fine granularity flow prediction method and system based on space-time comparison self-supervision | |
| CN114023074A (en) | Traffic jam prediction method, device and medium based on multiple signal sources | |
| Kang et al. | Bayesian path inference using sparse GPS samples with spatio-temporal constraints | |
| CN111524354B (en) | Method, system, medium and device for predicting urban traffic network path selection behavior based on language model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20221129 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |