CN113393673A - Traffic signal scheduling plan and time interval optimization method and device - Google Patents
Traffic signal scheduling plan and time interval optimization method and device Download PDFInfo
- Publication number
- CN113393673A CN113393673A CN202110944066.6A CN202110944066A CN113393673A CN 113393673 A CN113393673 A CN 113393673A CN 202110944066 A CN202110944066 A CN 202110944066A CN 113393673 A CN113393673 A CN 113393673A
- Authority
- CN
- China
- Prior art keywords
- time interval
- samples
- group
- time
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic signal scheduling plan and time interval optimization method and a device, wherein the method comprises the following steps: acquiring lane-level flow data of an intersection; performing principal component analysis on the lane-level flow data to obtain a principal component coefficient of each sample day, clustering the principal component coefficients of each sample day, and grouping the sample days based on a clustering result, wherein each group of samples corresponds to respective date type; acquiring steering flow sequence data of the intersection; based on the steering flow sequence data, performing time interval division on the date type corresponding to each group of samples by using a fisher optimal segmentation method to obtain a time interval division result of the date type corresponding to each group of samples; and obtaining a recommended period table of all steering of the intersection in the date type corresponding to each group of samples according to the period division result of the date type corresponding to each group of samples. The invention can obtain a more refined scheduling plan, and the finally obtained time interval division result is more suitable for the actual traffic condition.
Description
Technical Field
The invention relates to the technical field of traffic control, in particular to a method and a device for traffic signal scheduling plan and time interval optimization.
Background
The traffic has certain time-space law, and the traffic can be controlled by identifying the time-space law of the traffic so as to ensure smooth traffic and reduce congestion. The traffic signal control refers to intersection signal lamp control, and is generally divided into a working day (monday to friday) and a rest day (saturday and sunday) when the traffic signal control is performed, namely one set of traffic signal control scheme is used in the working day, and the other set of traffic signal control scheme is used in the rest day. The dispatching plan is more based on life experience, and cannot ensure that all intersections are optimal, for example, the traffic travel characteristics of Monday and Friday at some intersections are different from those of other working days, and if the same traffic signal control scheme is used in the same working day, the used traffic signal control scheme has poor effect.
Disclosure of Invention
The invention solves the problem that the existing dispatching plan may cause the poor effect of the traffic signal control scheme of each intersection.
The invention provides a traffic signal scheduling plan and time interval optimization method, which comprises the following steps:
acquiring lane-level flow data of an intersection, wherein the lane-level flow data comprises the flow of the intersection on each sample day;
performing principal component analysis on the lane-level flow data to obtain a principal component coefficient of each sample day, clustering the principal component coefficients of each sample day, and grouping the sample days based on a clustering result, wherein each group of samples corresponds to respective date type;
acquiring turn flow sequence data of the intersection, wherein the turn flow sequence data comprises the flow of all turns of the intersection on each sample day;
based on the steering flow sequence data, performing time interval division on the date type corresponding to each group of samples by using a fisher optimal segmentation method to obtain a time interval division result of the date type corresponding to each group of samples;
and obtaining a recommended period table of all steering of the intersection in the date type corresponding to each group of samples according to the period division result of the date type corresponding to each group of samples.
Optionally, the performing principal component analysis on the lane-level flow data to obtain a principal component coefficient for each sample day includes:
taking each sample day as a variable, and calculating a correlation coefficient matrix between each sample day;
calculating the eigenvalue and the contribution rate of the principal component by the correlation coefficient matrix, determining the number of the principal component by combining a preset threshold value of the cumulative contribution rate of the principal component, and determining an effective principal component based on the number of the principal component;
and calculating the load on the effective principal component on each sample day as the principal component coefficient of each sample day.
Optionally, the obtaining of the recommended time period table of all the turns of the intersection in the date type corresponding to each group of samples according to the time period division result of the date type corresponding to each group of samples includes:
acquiring the existing phase structure of the intersection;
generating a time-interval division constraint matched with the existing phase structure according to the existing phase structure;
the time interval division result of the date type corresponding to each group of samples is divided again through the time interval division constraint to obtain a secondary division result of the date type corresponding to each group of samples;
and obtaining a recommended period table of all steering of each group of samples corresponding to the date type of the intersection according to the secondary division result.
Optionally, the obtaining of the recommended period table of all turns of the intersection in the date type corresponding to each group of samples according to the secondary division result includes:
acquiring an existing time period table of the intersection, and comparing the secondary division result with the existing time period table by taking a preset minimum time period length as a step length;
and removing the time intervals of which the adjustment quantity relative to the existing time interval table is smaller than a preset value in the secondary division result, and reserving the time intervals corresponding to the existing time interval table to obtain the recommended time interval table of all the steering of each group of samples at the intersection in the corresponding date type.
Optionally, the fisher optimal segmentation method includes a total variation objective function, and the obtaining a time interval division result of the date type corresponding to each group of samples by using the fisher optimal segmentation method based on the steering flow sequence data includes:
obtaining the value range of the time segment number;
calculating an objective function value of the total variation objective function under different segment number values to obtain a curve of the objective function value changing along with the segment number, and determining the optimal time segment number based on a curve inflection point;
and performing time interval division on the date type corresponding to each group of samples based on the optimal time interval number to obtain a time interval division result of the date type corresponding to each group of samples.
Optionally, the total variation objective function includes:
wherein the content of the first and second substances,
for the number of time periods,
、
、
in order to be the flow data,
、
in order for the data within the segment to be degraded,
、
in order to be the objective function of the total variation,
is the length of time series data.
Optionally, the fisher optimal segmentation method uses an optimization objective with minimum intra-segment variance and maximum inter-segment variance, and the objective function is as follows:
wherein the content of the first and second substances,
wherein the number of time periods
With a period index of
(ii) a Length of time series data
Data subscript
；
To be composed of
Division of data into
As a result of the division of the segments,
to be composed of
Division of data into
In a period of time, the
An intra-segment data set of segments;
to be composed of
Division of data into
In a period of time, the
The intra-segment data of the segment is deteriorated,
is the mean value of the data within the segment,
is the number of samples in a segment;
to be composed of
Division of data into
In a period of time, the
The inter-segment data of the segment becomes poor,
for the average of all the data,
the mean value of the data variation between the segments,
mean of data variation within a segment.
Optionally, the obtaining of the recommended time period table of all the turns of the intersection in the date type corresponding to each group of samples according to the time period division result of the date type corresponding to each group of samples includes:
acquiring an existing time period table of the intersection, and comparing the time period division result with the existing time period table by taking a preset minimum time period length as a step length;
and removing the time intervals of which the adjustment quantity relative to the existing time interval table is smaller than a preset value in the time interval division result, and reserving the time intervals corresponding to the existing time interval table to obtain the recommended time interval table of all the steering of each group of samples at the intersection in the corresponding date type.
Optionally, the obtaining of the recommended time period table of all the turns of the intersection in the date type corresponding to each group of samples according to the time period division result of the date type corresponding to each group of samples includes:
and directly taking the time interval division result of the date type corresponding to each group of samples as a recommended time interval table of all the steering of the intersection in the date type corresponding to each group of samples.
The invention also provides a traffic signal scheduling planning and time interval optimizing device, which comprises a computer readable storage medium and a processor, wherein the computer readable storage medium is used for storing a computer program, and the computer program is read by the processor and runs to realize the traffic signal scheduling planning and time interval optimizing method.
According to the method, historical flow data, namely lane-level flow data of an intersection, is obtained, principal component analysis is carried out on the historical flow data, the principal component coefficients of each sample day are clustered, the sample days are grouped based on a clustering result, each group of samples are divided into time periods, a set of daily plan is adopted for the date type corresponding to each group of samples, and therefore division and grouping of different dates are carried out according to trip characteristics based on actual flow data of the intersection, and optimization of a scheduling plan is achieved. And then, time interval division is carried out on each group of samples by adopting a fisher optimal segmentation method to obtain time interval division results of date types corresponding to each group of samples, and further time interval division is carried out on the basis of the optimized scheduling plan, so that the finally obtained time interval division results are more consistent with actual traffic conditions, and the time interval division scheme is more accurate.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a traffic signal scheduling plan and time interval optimization method according to the present invention;
fig. 2 is a schematic diagram of another embodiment of the traffic signal scheduling plan and time interval optimization method of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
As shown in fig. 1, a method for traffic signal scheduling planning and time interval optimization in an embodiment of the present invention includes:
step S100, obtaining lane-level flow data of the intersection, wherein the lane-level flow data comprise the flow of the intersection on each sample day.
The intersection lane-level flow data is time sequence data of the flow of the intersection to be optimized, the time granularity can be set manually, and the time granularity is 5-15 min. The intersection lane-level flow data is historical flow data, comprises flow data of multiple days, and can be obtained through induction of a road induction coil or through an image recognition method or a vehicle detection algorithm.
And S200, performing principal component analysis on the lane-level flow data to obtain a principal component coefficient of each sample day, clustering the principal component coefficients of each sample day, and grouping the sample days based on a clustering result, wherein each group of samples corresponds to respective date type.
The lane-level flow data comprises data samples of multiple days, the data samples of the multiple days are processed by adopting principal component analysis, and the sample days are clustered, so that the optimization of a scheduling plan is realized. Generally, signal control is divided into weekdays (monday to friday) and weekdays (saturday, sunday), but different intersections are affected by the type of land used in the periphery, and the traffic travel characteristics of monday and friday may be different from those of other weekdays. In order to realize more refined signal control, the lane-level flow data is subjected to principal component analysis, the principal component coefficients are clustered, and the sample days are grouped based on the clustering result so as to divide and classify the dates, wherein the sample days in the same group have time sequence characteristic similarity, so that the dates in the same type adopt the same set of daily plan, therefore, samples with inconsistent characteristics in the sample days can be found, and the samples are divided separately in time intervals so as to realize scheduling plan optimization. For example, dates are divided into four categories: monday; tuesday to tuesday; friday; saturday to sunday. After the sample days are grouped, the time interval division is carried out on the date type corresponding to each group of samples, namely, the following steps S300-S500 are respectively carried out on each group of samples.
Optionally, the performing principal component analysis on the lane-level flow data to obtain a principal component coefficient for each sample day includes:
taking each sample day as a variable, and calculating a correlation coefficient matrix between each sample day;
calculating the eigenvalue and the contribution rate of the principal component by the correlation coefficient matrix, determining the number of the principal component by combining a preset threshold value of the cumulative contribution rate of the principal component, and determining an effective principal component based on the number of the principal component;
and calculating the load on the effective principal component on each sample day as the principal component coefficient of each sample day.
And taking each sample day as a variable, taking the flow data of each sample day as variable data, and performing principal component analysis. The preset threshold value of the accumulated contribution rate of the principal components is 80% -90%. The main component coefficients are clustered to realize grouping of sample days, the sample days in the same group have time sequence characteristic similarity, the samples with dissimilar time sequence characteristics are separated, time interval division is respectively carried out, and then optimization of a scheduling plan is completed.
Step S300, obtaining turning flow rate sequence data of the intersection, wherein the turning flow rate sequence data comprises all turning flow rates of the intersection on each sample day.
The turn traffic sequence data herein refers to time-series data of intersection traffic, and the time granularity thereof may be 5-15 min. Optionally, after the steering flow sequence data of the intersection is acquired, filtering and smoothing are performed on the steering flow sequence data to avoid that the steering flow sequence data is affected by cycles and fluctuates greatly.
The steering flow sequence data contains the flow of all the steering, so that the flow of all the steering based on the intersection is divided into time intervals at the same time in the following, the proportional relation difference between the steering is reflected, and the obtained time interval scheme is more accurate. And based on the steering flow sequence data, performing time interval division on the date type corresponding to each group of samples by using a fisher optimal segmentation method to obtain a time interval division result of the date type corresponding to each group of samples.
The fisher optimal segmentation method is a segmentation method suitable for ordered data, each group of samples are respectively used as ordered data sets to be given to the fisher optimal segmentation method, time interval division is carried out on the basis of each group of samples by adopting the fisher optimal segmentation method, the time interval number can be a pre-specified optimal time interval number, the time interval number can also be automatically calculated, and the optimal time interval number can be determined according to a relation curve of a target function and the time interval number.
The time interval division is divided by taking a time ring as a unit, namely 0 point and 24 points in 24 hours of a day are head-to-tail connection points, for example, a time interval division result can exist: 23 pm to 1 pm. The division is more in line with the actual traffic condition, so that the time interval division result is more in line with the actual traffic condition and more accurate.
The time interval division result in the embodiment of the invention is not limited by data granularity and supports division to minutes.
And S500, obtaining a recommended period table of all steering of the intersection in the date type corresponding to each group of samples according to the period division result of the date type corresponding to each group of samples.
In one embodiment, after the time interval division result of the date type corresponding to each group of samples is obtained, the time interval division result is directly used as a recommended time interval table, that is, the optimization scheme is directly used as a recommended result.
In another embodiment, after time interval division results of date types corresponding to each group of samples are obtained, a current time interval table of the intersection is obtained, and the time interval division results are compared with the current time interval table by taking a preset minimum time interval length as a step length; and removing the time interval of which the adjustment quantity relative to the existing time interval table is smaller than a preset value in the time interval division result, reserving the time interval corresponding to the existing time interval table, obtaining the recommended time interval table of all the turning directions of the date type corresponding to each group of samples at the intersection, if the adjustment quantity of the time interval division result relative to the existing time interval table is smaller than a preset threshold value for a certain time interval, reserving the original time interval of the existing time interval table as the recommended time interval, and if the adjustment quantity of the time interval division result relative to the existing time interval table is larger than or equal to the preset threshold value for a certain time interval, reserving the new time interval in the time interval division result as the recommended time interval. Wherein the length of the preset minimum time interval can be selected to be 15-20 min.
In the embodiment, historical flow data, namely intersection lane-level flow data, is obtained and subjected to principal component analysis, principal component coefficients of each sample day are clustered, the sample days are grouped based on a clustering result, each group of samples are divided into time periods, a set of daily plans is adopted for date types corresponding to each group of samples, and therefore, the fact that different dates are divided and grouped based on intersection actual flow data according to trip characteristics is achieved, and optimization of a scheduling plan is achieved; and then, time interval division is carried out on each group of samples by adopting a fisher optimal segmentation method to obtain time interval division results of date types corresponding to each group of samples, and further time interval division is carried out on the basis of the optimized scheduling plan, so that the finally obtained time interval division results are more consistent with actual traffic conditions, and the time interval division scheme is more accurate.
Optionally, step S500 includes:
and acquiring the existing phase structure of the intersection.
In this embodiment, because an applicable actual intersection may have a time interval division scheme, in order to be compatible with the existing time interval division scheme, the present embodiment obtains the existing phase structure of the intersection, and implements constraint on the optimized time interval through the existing phase structure, so as to avoid conflict between the optimized time interval and the existing phase structure, and implement compatibility between the two. The existing phase structure of the intersection comprises a currently adopted phase structure of the intersection and a corresponding time period, for example, a phase structure I corresponding to 0:00-6:00 and a phase structure II corresponding to 6:01-12:00, wherein one set of phase structure scheme comprises phases and combinations thereof and a phase release sequence, for example, a 4-phase intersection, 4 phases are respectively composed of orientation signal lamps, and the sequence of the 4 phases in the same row.
And generating time interval division constraints matched with the existing phase structure according to the existing phase structure, and dividing time interval division results of the date types corresponding to each group of samples again through the time interval division constraints to obtain secondary division results of the date types corresponding to each group of samples.
Specifically, the time interval division constraint matched with the existing phase structure means that one optimized time interval can only correspond to one phase structure, for example, the existing phase structure is 0:00-6:00 corresponding to a first phase structure, 6:01-12:00 corresponding to a second phase structure, and the optimized time interval cannot appear in the range of 5:00-7:00 because 5:00-7:00 corresponds to two phase structures, and the optimized time interval corresponds to either the first phase structure or the second phase structure, and only one of the two phase structures can be selected.
And the time interval division result of each group of samples is divided again through time interval division constraint, for example, the time interval division result comprises 5:00-7:00, the existing phase structure is 0:00-6:00 corresponding to the first phase structure, 6:01-12:00 corresponding to the second phase structure, and then 5:00-7:00 are divided again according to the existing phase structure, so that 5:00-6:00, 6:01-7:00 can be obtained.
And for a certain period in the period division results, searching a phase structure corresponding to the period with the closest start-stop time in the existing period table, and dividing the period division results of each group of samples again based on the period division constraint corresponding to the phase structure.
And obtaining a recommended period table of all steering of each group of samples corresponding to the date type of the intersection according to the secondary division result. Specifically, the secondary division result may be directly used as the recommended period table. In another optional embodiment, after the secondary division result of the date type corresponding to each group of samples is obtained, the existing time interval table of the intersection is obtained, the secondary division result is compared with the existing time interval table by taking the preset minimum time interval length as a step length, if the adjustment amount of the secondary division result relative to the existing time interval table is smaller than the preset threshold value for a certain time interval, the original time interval of the existing time interval table is reserved as the recommended time interval, and if the adjustment amount of the secondary division result relative to the existing time interval table is larger than or equal to the preset threshold value for a certain time interval, the new time interval in the secondary division result is used as the recommended time interval. Wherein the length of the preset minimum time interval can be selected to be 15-20 min.
The time interval division constraint is generated through the existing phase structure of the intersection, so that the optimized time interval is compatible with the existing phase structure scheme of the intersection, and the applicability of the embodiment of the invention is improved.
Optionally, the obtaining of the recommended period table of all turns of the intersection in the date type corresponding to each group of samples according to the secondary division result includes:
and obtaining the current time interval table of the intersection, comparing the secondary division result with the current time interval table by taking the preset minimum time interval length as a step length, removing the time interval of which the adjustment quantity relative to the current time interval table is smaller than the preset value in the secondary division result, and reserving the time interval corresponding to the current time interval table to obtain the recommended time interval table of all the steering directions of the intersection in the date type corresponding to each group of samples.
Wherein the length of the preset minimum time interval can be selected to be 15-20 min.
And comparing the secondary division result with the existing time period table, if the adjustment quantity of the secondary division result relative to the existing time period table is smaller than a preset value for a certain time period, keeping the original time period of the existing time period table as a recommended time period, and if the adjustment quantity of the secondary division result relative to the existing time period table is larger than or equal to the preset value for a certain time period, taking a new time period in the secondary division result as the recommended time period. For example, the preset value is 10 minutes, the time interval in the secondary division result is 8:00-10:00, the time interval of the current time interval table is 8:05-10:00, the time interval is adjusted for only 5 minutes, if the time interval is smaller than the preset value, the original time interval of the current time interval table is reserved, and the time interval is not adjusted.
Because the existing time interval of the intersection is adjusted, other time intervals need to be adjusted in a linkage mode, and when the adjustment amount is small, the optimization effect is not worth of overlarge adjustment workload, so that adjustment is not performed, and balance is achieved between the optimization effect and the adjustment workload.
Optionally, the fisher optimal segmentation method includes a total variation objective function, and the obtaining a time interval division result of the date type corresponding to each group of samples by using the fisher optimal segmentation method based on the steering flow sequence data includes:
and obtaining the value range of the time segment number. And calculating the objective function value of the total variation objective function under different segment number values to obtain a curve of the objective function value changing along with the segment number, and determining the optimal time segment number based on the curve inflection point.
Specifically, the maximum time period number T can be set, the time period number k is 2, 3, …, and T (the maximum time period number), and objective functions under different time period number values are calculated. The total variation is reduced as an optimization target, the total variation is reduced when the number of time periods is larger, but the number of time periods is not expected to be too large in the actual operation and timing process, so that the curve of the change of the objective function value along with the number of the time periods can be found out at the inflection point of the curve, the number of the time periods is continuously increased, the change of the objective function value is small, and therefore the inflection point of the curve can be selected as the optimal number of the time periods.
The total variation objective function is as follows:
wherein the content of the first and second substances,
for the number of time periods,
、
、
in order to be the flow data,
、
in order for the data within the segment to be degraded,
、
in order to be the objective function of the total variation,
is the length of time series data.
And performing time interval division on the date type corresponding to each group of samples based on the optimal time interval number to obtain a time interval division result of the date type corresponding to each group of samples.
And after the optimal time period number is determined, time period division is carried out on the date type corresponding to each group of samples based on a fisher optimal segmentation method. The time interval division result is not limited by data granularity and supports division into minutes.
And determining the optimal time period number through the curve inflection point of the target function along with the time period number, so as to obtain the optimal time period number and further obtain the optimal time period division result.
Optionally, the fisher optimal segmentation method optimizes the objective with the minimum intra-segment variance and the maximum inter-segment variance. The objective function is:
wherein the content of the first and second substances,
wherein the number of time periods
With a period index of
(ii) a Length of time series data
Data subscript
；
To be composed of
Division of data into
As a result of the division of the segments,
to be composed of
Division of data into
In a period of time, the
An intra-segment data set of segments;
to be composed of
Division of data into
In a period of time, the
The intra-segment data of the segment is deteriorated,
is the mean value of the data within the segment,
is the number of samples in a segment;
to be composed of
Division of data into
In a period of time, the
The inter-segment data of the segment becomes poor,
for the average of all the data,
the mean value of the data variation between the segments,
mean of data variation within a segment.
An implementation is given as shown in fig. 2:
acquiring lane-level flow data;
judging whether the traffic data of a plurality of days exist or not;
if the flow data of multiple days exist, performing principal component analysis on the flow data, clustering the principal component coefficients of each sample, grouping the sample days based on the clustering result, and then respectively performing time interval division on the date type corresponding to each group of samples, namely respectively performing subsequent time interval division steps on the date type corresponding to each group of samples;
if the traffic data of a plurality of days does not exist, the following time interval division steps are directly executed:
acquiring steering flow sequence data of the intersection;
acquiring an existing phase structure of the intersection, and generating a time interval division constraint matched with the existing phase structure based on the existing phase structure; meanwhile, judging whether the designated time segment number exists or not, if so, directly using a fisher optimal segmentation method to perform time segment division on the date type corresponding to each group of samples, and if not, calculating the optimal time segment number and then using the fisher optimal segmentation method to perform time segment division on the date type corresponding to each group of samples to obtain a time segment division result of the date type corresponding to each group of samples;
the time interval division result of each group of samples is divided again through time interval division constraint to obtain a secondary division result of the date type corresponding to each group of samples;
and judging whether the existing time period table exists or not, if so, comparing the existing time period table, filtering the time period with small adjustment amount by taking the preset minimum time period length as a step length, and outputting the recommended time period table, otherwise, directly outputting a secondary division result as the recommended time period table.
In one embodiment, the traffic signal scheduling planning and time interval optimizing device of the invention comprises a computer readable storage medium storing a computer program and a processor, wherein the computer program is read by the processor and executed to realize the traffic signal scheduling planning and time interval optimizing method. Compared with the prior art, the traffic signal scheduling plan and time interval optimization device has the advantages that the traffic signal scheduling plan and time interval optimization method is consistent with the traffic signal scheduling plan and time interval optimization method, and the details are omitted.
The reader should understand that in the description of this specification, reference to the description of the terms "one embodiment," "some embodiments," "an example," "a specific example" or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A traffic signal scheduling plan and time period optimization method, comprising:
acquiring lane-level flow data of an intersection, wherein the lane-level flow data comprises the flow of the intersection on each sample day;
performing principal component analysis on the lane-level flow data to obtain a principal component coefficient of each sample day, clustering the principal component coefficients of each sample day, and grouping the sample days based on a clustering result, wherein each group of samples corresponds to respective date type;
acquiring turn flow sequence data of the intersection, wherein the turn flow sequence data comprises the flow of all turns of the intersection on each sample day;
based on the steering flow sequence data, performing time interval division on the date type corresponding to each group of samples by using a fisher optimal segmentation method to obtain a time interval division result of the date type corresponding to each group of samples;
and obtaining a recommended period table of all steering of the intersection in the date type corresponding to each group of samples according to the period division result of the date type corresponding to each group of samples.
2. The traffic signal scheduling plan and time interval optimization method of claim 1, wherein the performing principal component analysis on the lane-level traffic data to obtain a principal component coefficient for each sample day comprises:
taking each sample day as a variable, and calculating a correlation coefficient matrix between each sample day;
calculating the eigenvalue and the contribution rate of the principal component by the correlation coefficient matrix, determining the number of the principal component by combining a preset threshold value of the cumulative contribution rate of the principal component, and determining an effective principal component based on the number of the principal component;
and calculating the load on the effective principal component on each sample day as the principal component coefficient of each sample day.
3. The traffic signal scheduling plan and time interval optimization method according to claim 1 or 2, wherein the obtaining of the recommended time interval table of all the turns of the intersection at the date type corresponding to each group of samples according to the time interval division result of the date type corresponding to each group of samples comprises:
acquiring the existing phase structure of the intersection;
generating a time-interval division constraint matched with the existing phase structure according to the existing phase structure;
the time interval division result of the date type corresponding to each group of samples is divided again through the time interval division constraint to obtain a secondary division result of the date type corresponding to each group of samples;
and obtaining a recommended period table of all steering of each group of samples corresponding to the date type of the intersection according to the secondary division result.
4. The traffic signal scheduling plan and time interval optimization method according to claim 3, wherein the obtaining of the recommended time interval table of all turns of the intersection at each group of the sample corresponding to the date type according to the secondary division result comprises:
acquiring an existing time period table of the intersection, and comparing the secondary division result with the existing time period table by taking a preset minimum time period length as a step length;
and removing the time intervals of which the adjustment quantity relative to the existing time interval table is smaller than a preset value in the secondary division result, and reserving the time intervals corresponding to the existing time interval table to obtain the recommended time interval table of all the steering of each group of samples at the intersection in the corresponding date type.
5. The traffic signal scheduling planning and time-interval optimizing method according to claim 1 or 2, wherein the fisher optimal segmentation method includes a total variation objective function, and the time-interval segmentation of the date type corresponding to each group of samples by using the fisher optimal segmentation method based on the steering flow sequence data includes:
obtaining the value range of the time segment number;
calculating an objective function value of the total variation objective function under different segment number values to obtain a curve of the objective function value changing along with the segment number, and determining the optimal time segment number based on a curve inflection point;
and performing time interval division on the date type corresponding to each group of samples based on the optimal time interval number to obtain a time interval division result of the date type corresponding to each group of samples.
6. The traffic signal dispatch plan and time period optimization method of claim 5, wherein the total variation objective function comprises:
wherein the content of the first and second substances,
for the number of time periods,
、
、
in order to be the flow data,
、
in order for the data within the segment to be degraded,
、
in order to be the objective function of the total variation,
is the length of time series data.
7. The traffic signal scheduling planning and time interval optimizing method according to claim 1 or 2, wherein the fisher optimal segmentation method is an optimization objective with minimum intra-segment variance and maximum inter-segment variance, and the objective function is:
wherein the content of the first and second substances,
wherein the number of time periods
With a period index of
(ii) a Length of time series data
Data subscript
；
To be composed of
Division of data into
As a result of the division of the segments,
to be composed of
Division of data into
In a period of time, the
An intra-segment data set of segments;
to be composed of
Division of data into
In a period of time, the
The intra-segment data of the segment is deteriorated,
is the mean value of the data within the segment,
is the number of samples in a segment;
to be composed of
Division of data into
In a period of time, the
The inter-segment data of the segment becomes poor,
for the average of all the data,
the mean value of the data variation between the segments,
mean of data variation within a segment.
8. The traffic signal scheduling plan and time interval optimization method according to claim 1 or 2, wherein the obtaining of the recommended time interval table of all the turns of the intersection at the date type corresponding to each group of samples according to the time interval division result of the date type corresponding to each group of samples comprises:
acquiring an existing time period table of the intersection, and comparing the time period division result with the existing time period table by taking a preset minimum time period length as a step length;
and removing the time intervals of which the adjustment quantity relative to the existing time interval table is smaller than a preset value in the time interval division result, and reserving the time intervals corresponding to the existing time interval table to obtain the recommended time interval table of all the steering of each group of samples at the intersection in the corresponding date type.
9. The traffic signal scheduling plan and time interval optimization method according to claim 1 or 2, wherein the obtaining of the recommended time interval table of all the turns of the intersection at the date type corresponding to each group of samples according to the time interval division result of the date type corresponding to each group of samples comprises:
and directly taking the time interval division result of the date type corresponding to each group of samples as a recommended time interval table of all the steering of the intersection in the date type corresponding to each group of samples.
10. A traffic signal scheduling planning and time slot optimizing apparatus comprising a computer readable storage medium storing a computer program and a processor, the computer program being read and executed by the processor to implement the traffic signal scheduling planning and time slot optimizing method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944066.6A CN113393673A (en) | 2021-08-17 | 2021-08-17 | Traffic signal scheduling plan and time interval optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110944066.6A CN113393673A (en) | 2021-08-17 | 2021-08-17 | Traffic signal scheduling plan and time interval optimization method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113393673A true CN113393673A (en) | 2021-09-14 |
Family
ID=77622681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110944066.6A Pending CN113393673A (en) | 2021-08-17 | 2021-08-17 | Traffic signal scheduling plan and time interval optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113393673A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117609814A (en) * | 2024-01-24 | 2024-02-27 | 广东奥飞数据科技股份有限公司 | SD-WAN intelligent flow scheduling optimization method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050070A (en) * | 2014-03-28 | 2014-09-17 | 国家计算机网络与信息安全管理中心 | High-dimensional flow data changing point detection method in distributed system |
| CN105225539A (en) * | 2015-09-30 | 2016-01-06 | 中国民用航空总局第二研究所 | Based on the method and system of the sector runnability aggregative index of principal component analysis (PCA) |
| CN105703954A (en) * | 2016-03-17 | 2016-06-22 | 福州大学 | Network data flow prediction method based on ARIMA model |
| CN106920402A (en) * | 2016-11-21 | 2017-07-04 | 中兴软创科技股份有限公司 | A kind of time series division methods and system based on the magnitude of traffic flow |
| CN111341110A (en) * | 2020-05-22 | 2020-06-26 | 深圳市城市交通规划设计研究中心股份有限公司 | Signal coordination control subarea division method and device, storage medium and terminal equipment |
| CN113034940A (en) * | 2019-12-25 | 2021-06-25 | 中国航天系统工程有限公司 | Fisher ordered clustering-based single-point signalized intersection optimization timing method |
-
2021
- 2021-08-17 CN CN202110944066.6A patent/CN113393673A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104050070A (en) * | 2014-03-28 | 2014-09-17 | 国家计算机网络与信息安全管理中心 | High-dimensional flow data changing point detection method in distributed system |
| CN105225539A (en) * | 2015-09-30 | 2016-01-06 | 中国民用航空总局第二研究所 | Based on the method and system of the sector runnability aggregative index of principal component analysis (PCA) |
| CN105703954A (en) * | 2016-03-17 | 2016-06-22 | 福州大学 | Network data flow prediction method based on ARIMA model |
| CN106920402A (en) * | 2016-11-21 | 2017-07-04 | 中兴软创科技股份有限公司 | A kind of time series division methods and system based on the magnitude of traffic flow |
| CN113034940A (en) * | 2019-12-25 | 2021-06-25 | 中国航天系统工程有限公司 | Fisher ordered clustering-based single-point signalized intersection optimization timing method |
| CN111341110A (en) * | 2020-05-22 | 2020-06-26 | 深圳市城市交通规划设计研究中心股份有限公司 | Signal coordination control subarea division method and device, storage medium and terminal equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117609814A (en) * | 2024-01-24 | 2024-02-27 | 广东奥飞数据科技股份有限公司 | SD-WAN intelligent flow scheduling optimization method and system |
| CN117609814B (en) * | 2024-01-24 | 2024-05-07 | 广东奥飞数据科技股份有限公司 | SD-WAN intelligent flow scheduling optimization method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7577513B2 (en) | Traffic information prediction system | |
| Li et al. | Studying the benefits of carpooling in an urban area using automatic vehicle identification data | |
| CN109711591B (en) | Road section speed prediction method, device, server and storage medium | |
| CN108389406B (en) | Automatic division method for signal control time interval | |
| CN106767872A (en) | A kind of intelligent travel reminding method and its client | |
| JP2013148574A (en) | User's route selection taste extraction system and route selection taste extraction method | |
| CN113393673A (en) | Traffic signal scheduling plan and time interval optimization method and device | |
| CN108615361A (en) | Crossing control time division methods and system based on multidimensional time-series segmentation | |
| CN108281033A (en) | A kind of parking guidance system and method | |
| CN108305459B (en) | Road network operation index prediction method based on data driving | |
| JP2004157814A (en) | Decision tree generating method and model structure generating device | |
| CN106203717A (en) | Tax hall intelligent navigation method based on data analysis | |
| CN111815941B (en) | Frequent congestion bottleneck identification method and device based on historical road conditions | |
| CN112243025A (en) | Node cost scheduling method, electronic device and storage medium | |
| CN116842060B (en) | Reasoning query optimization method and device based on agent model rearrangement technology | |
| CN109922212B (en) | Method and device for predicting time-interval telephone traffic ratio | |
| CN113256472B (en) | Intelligent traffic control method and system and brain-like computer readable storage medium | |
| JP2000270473A (en) | Power demand estimating method | |
| CN112652164B (en) | Traffic time interval dividing method, device and equipment | |
| CN116433245B (en) | Customer visit plan generation method and system | |
| CN116431923B (en) | Traffic travel prediction method, equipment and medium for urban road | |
| Waury et al. | Assessing the accuracy benefits of on-the-fly trajectory selection in fine-grained travel-time estimation | |
| CN113358130A (en) | Method, device and equipment for acquiring planned path and readable storage medium | |
| JP4495746B2 (en) | Traffic jam travel time prediction database creation device, traffic jam travel time prediction database creation method, traffic jam travel time prediction database creation program implementing the method and recording medium recording the program, traffic jam travel time prediction device, traffic jam travel time prediction method, Program for predicting traffic jam travel time and a recording medium recording the program | |
| JP5561643B2 (en) | Demand forecasting device and water operation monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210914 |