CN113012449B - Smart city signal lamp timing optimization method based on multi-sample learning particle swarm - Google Patents

Smart city signal lamp timing optimization method based on multi-sample learning particle swarm Download PDF

Info

Publication number
CN113012449B
CN113012449B CN202110271046.7A CN202110271046A CN113012449B CN 113012449 B CN113012449 B CN 113012449B CN 202110271046 A CN202110271046 A CN 202110271046A CN 113012449 B CN113012449 B CN 113012449B
Authority
CN
China
Prior art keywords
particle
particle swarm
timing
dimension
signal lamp
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
Application number
CN202110271046.7A
Other languages
Chinese (zh)
Other versions
CN113012449A (en
Inventor
詹志辉
邓壮杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN202110271046.7A priority Critical patent/CN113012449B/en
Publication of CN113012449A publication Critical patent/CN113012449A/en
Priority to LU503042A priority patent/LU503042B1/en
Priority to PCT/CN2021/119200 priority patent/WO2022188387A1/en
Application granted granted Critical
Publication of CN113012449B publication Critical patent/CN113012449B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • Y02A30/60—Planning or developing urban green infrastructure

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Biophysics (AREA)
  • Development Economics (AREA)
  • Primary Health Care (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Educational Administration (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Game Theory and Decision Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physiology (AREA)
  • Genetics & Genomics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a smart city signal lamp timing optimization method based on multi-sample learning particle swarm, and mainly relates to the field of smart traffic control and intelligent optimization algorithms. Aiming at the problem that the traditional particle swarm algorithm is easy to fall into local optimum when being applied to a signal lamp optimization timing scene, the method adopts a new multi-sample learning strategy, and the strategy enables the particles to learn different dimensionality samples of other particles while learning to the self optimum position and the global optimum position, so that the diversity of the algorithm is enhanced, and the particles are prevented from falling into the local optimum position. In addition, the method adopts a knowledge embedding auxiliary strategy when generating the initial particle species group, and uses the distribution characteristics of the traffic flow input at the intersection as the front knowledge to assist the generation of the initial species group. The optimization results of the vehicle flows with different saturation degrees at the single intersection are displayed, the method is better in diversity compared with other timing optimization methods on the premise of ensuring the convergence speed, and the optimized timing scheme is better in comprehensive performance.

Description

Smart city signal lamp timing optimization method based on multi-sample learning particle swarm
Technical Field
The invention relates to the field of intelligent traffic control and intelligent optimization algorithms, in particular to a multi-sample learning particle swarm-based intelligent city signal lamp timing optimization method.
Background
With the development of urbanization, the automobile holding capacity of urban residents is improved year by year, and the traffic pressure of urban roads, particularly intersections, is gradually increased. The traffic signal lamp can strengthen road traffic management and effectively solve the problem of traffic conflict, but the road traffic situation becomes increasingly complex along with the increase of the traffic flow saturation of the road. The traditional signal lamp timing scheme, such as a Webster timing scheme, is to establish a traffic mathematical model based on a target of minimizing intersection delay, calculate an optimal signal lamp period according to the model, and divide corresponding green lamp time. However, the webster model is only suitable for traffic models under unsaturated traffic flow conditions, so that when the traffic flow of a road is supersaturated, the webster timing scheme is not suitable any more, and the traditional signal lamp timing scheme cannot meet the requirements of smart city traffic more and more.
As an important evolutionary optimization algorithm, the particle swarm optimization shows excellent performance in various fields such as an electric power system, medical image registration, multi-objective optimization, machine learning and the like. The particle swarm algorithm has strong global search capability and convergence capability, and is suitable for search optimization of the optimal signal timing scheme under different saturation conditions, so that many researchers also apply the particle swarm algorithm to the field of signal timing optimization of intelligent traffic.
Most of the existing particle swarm algorithms for traffic signal optimization have the defect of easy falling into local optimal solutions. The reason is that the particle swarm searches for an optimal value in a continuous space, and a signal lamp scheme finally used for timing is a discrete integer value, wherein the former and the latter have an integer relation, so that when a certain dimension of the particles converges to a certain degree at a local optimal solution, a local optimal position is difficult to jump out at the dimension.
In addition, most of the existing researches do not use the flow distribution characteristics input at the intersection for assisting the particle swarm optimization to optimize the signal lamp. The traditional particle swarm optimization algorithm is completely randomly initialized when a seed group is initialized, but in the field of traffic signal timing optimization, if the distribution characteristics of traffic flow input are acquired, an initial seed group which is more likely to cover the area where the optimal solution is located can be purposefully generated by using the information, so that the searching capability of the algorithm is enhanced. Therefore, based on the above analysis, the conventional particle swarm optimization algorithm needs to be improved and incorporate the inherent characteristic knowledge in the traffic signal timing domain to enhance the optimization capability of the algorithm for the traffic signal timing domain.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a smart city signal lamp timing optimization method based on multi-sample learning particle swarm. In addition, the traffic flow distribution characteristics input by the intersection are used as the front knowledge when the particle population is initialized, so that the initial population distribution is more likely to cover the area where the optimal solution is located.
The purpose of the invention can be achieved by adopting the following technical scheme:
a smart city signal lamp timing optimization method based on multi-sample learning particle swarm comprises the following steps:
s1, the traffic flow and proportion of each entrance in the intersection are the biggest factors influencing the traffic signal lamp setting, in the existing method for optimizing the timing problem of the traffic signal lamp by using an intelligent algorithm, the timing optimization of the traffic signal lamp is taken as a random optimization problem, and the important factor is not taken into consideration, in the invention, an initial seed solution is generated according to the traffic flow distribution characteristics of each entrance of the intersection, and then an initial particle swarm is generated according to the initial seed solution, and the process is as follows:
s101, since the number of lanes at each entrance is not necessarily the same, that is, for two entrances with different number of lanes but the same input traffic flow, the entrance with a small number of lanes will be more "congested", and theoretically, a relatively more green light passing time needs to be allocated to this entrance, so an average lane flow ratio index needs to be defined to measure the distribution ratio of the traffic flow, and the average lane flow ratio is defined as the following formula:
Figure BDA0002971570810000031
Figure BDA0002971570810000032
wherein, ALFjIndicates the average lane flow magnitude, VI, of the jth inletjIndicating the magnitude of the incoming traffic at the jth access opening, LNjIndicating the number of lanes at the jth access opening, RjRepresents the average lane flow rate of the jth access port;
s102, calculating an initial total green light time length according to the proportion relation between the total input traffic flow of the whole intersection and the maximum traffic flow which can be accommodated by the intersection design, distributing the green light time length for each entrance based on the average lane flow ratio to generate an initial seed solution, wherein the green light time of each dimensionality of the initial seed solution is calculated by the following formula:
Figure BDA0002971570810000033
Figure BDA0002971570810000034
wherein g'jIndicating the green time of the seed solution at the jth entrance, G' indicating the total green time of the seed solution, G indicating the total green time of the maximum allowable signal lamp cycle period, VImaxRepresenting the maximum amount of incoming traffic allowed at the intersection,
Figure BDA0002971570810000035
indicating a downward rounding operator;
s103, the initial seed solution represents a possible area of the potential optimal timing scheme, appropriate Gaussian disturbance is added based on the dimension values of the initial seed solution, an initial particle population covering the vicinity of the seed solution is generated, and the initial particle population is generated according to the following formula:
Figure BDA0002971570810000041
wherein
Figure BDA0002971570810000042
A gaussian distribution with a mean value of 0 and a standard deviation of 5 is represented by N (0,5) representing the j dimension of the ith particle in the population;
s2, performing constraint correction on the generated particle swarm individuals to enable the particle swarm individuals to meet constraint requirements of feasible solutions, wherein the constraint requirements comprise maximum green light time constraint, minimum green light time constraint and maximum cycle time constraint;
s3, after an initial population is generated, the speed and the position of the population can be updated according to the iteration steps of the particle swarm optimization algorithm, however, the traditional particle swarm optimization algorithm has the defect that the particles are easy to fall into the local optimal solution, a multi-sample learning strategy is added in the particle speed updating link for overcoming the defect, the particles of the population are subjected to speed updating according to the multi-sample learning strategy, and the updating formula is as follows:
Figure BDA0002971570810000043
wherein
Figure BDA0002971570810000044
Represents the j-th dimension velocity of the ith particle in the t +1 th generation,
Figure BDA0002971570810000045
represents the j-th dimension velocity of the ith particle in the t generation,
Figure BDA0002971570810000046
represents the j-th dimension position value of the ith particle in the t generation, omega represents the velocity weight,
Figure BDA00029715708100000410
representing the historical optimum position, gBest, of the ith particle in the jth dimensionjRepresents the global optimal position in the j-th dimension,
Figure BDA0002971570810000047
the sample individual value of the jth dimension of the ith particle is represented, more specifically, for the sample value of the jth dimension of the ith particle, another two particles are randomly selected, the jth dimension value of the historical best position of the particle with the better adaptive value is selected as the sample value of the ith particle in the dimension, and c1，c2And c3There are three update coefficients that are,
Figure BDA0002971570810000048
and
Figure BDA0002971570810000049
is the j dimension [0,1]]Three differences therebetweenRandom number, the particle position value is updated by adding the original position and the updated new speed in the formula (6) as the new position of the particle;
s4, performing simulation evaluation on the traffic signal lamp timing scheme represented by the downward rounding of the position value of the particle through microscopic simulation software VISSIM, and taking the evaluation result as the adaptive value of the particle;
and S5, when the algorithm is iterated to the specified maximum iteration number, taking the global optimal scheme as a final timing scheme, otherwise, continuing to execute the steps S3-S4 until the maximum iteration number preset by the algorithm is terminated.
Further, in the step S1, in generating an initial seed solution according to the traffic flow distribution characteristics of each entrance at the intersection, a knowledge ratio threshold KT is set, when generating population individuals, a random number between [0,1] is first taken, if the random number is less than KT, the individuals are generated by using a knowledge-assisted strategy, otherwise, the individuals are generated according to a random method.
Further, in step S3, a board update interval threshold RT is set, and if the global optimal position of the algorithm is not improved after the RT generation, the board update is performed, otherwise, the board of the previous generation is continuously used.
Further, the update coefficient c1、c2And c3Set to 0.75, 0.75 and 1.50, respectively.
Further, the knowledge-to-ratio threshold KT is set to 0.8.
Further, the chart update interval threshold RT is set to 7.
Further, the maximum number of iterations is set to 40 generations.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the method and the device help the initial population distribution to cover the area where the optimal solution is more likely to be located by adopting the intersection traffic flow distribution characteristics as the pre-knowledge to assist the initialization of the particle population.
2. The multi-sample learning strategy provided by the invention is beneficial to learning the different dimensions of the particles to different samples respectively, so that the algorithm diversity is enhanced, and the particles are prevented from falling into the local optimal position in the searching process.
3. In the initial particle population generation process, a knowledge ratio threshold is used for controlling the proportion of knowledge auxiliary generation solutions and random solutions, and the mode is beneficial to integrating the respective advantages of a knowledge embedding auxiliary strategy and a random search strategy.
4. In the multi-sample learning strategy provided by the invention, the update frequency of the samples is controlled by using one sample update threshold, so that the method is beneficial to avoiding the problem that the particle search is difficult to converge due to frequent updating of the learning samples of the particles and is beneficial to timely updating of old sample individuals incapable of continuously guiding the particle evolution.
Drawings
FIG. 1 is a flow chart of a multi-sample learning particle swarm optimization algorithm in an embodiment of the invention;
FIG. 2 is a geometric plan view of the New lake road in the Jeans region of Shenzhen City-Yuan road intersection in China in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
Referring to fig. 1, the intelligent city signal lamp timing optimization method based on the multi-sample learning particle swarm comprises the following operation steps:
s1, selecting the Shenzhen New lake road-Yuan road intersection signal lamp as shown in FIG. 2 for timing optimization. The intersection is a four-way intersection, four-phase signal lamps are adopted for control, the duration of a green light of each phase is used as a dimension value of a particle, the traffic flow is input according to four entrance ports of the intersection, and the average lane flow ratio is defined as the following formula:
Figure BDA0002971570810000071
Figure BDA0002971570810000072
wherein, ALFjIndicates the average lane flow magnitude, VI, of the jth inletjIndicating the magnitude of the incoming traffic at the jth access opening, LNjIndicating the number of lanes at the jth access opening, RjRepresents the average lane flow rate of the jth access port;
an initial seed solution is then generated based on the average lane flow ratio according to the following equation, with the green time for each dimension of the initial seed solution calculated by the following equation:
Figure BDA0002971570810000073
Figure BDA0002971570810000074
wherein, g'jIndicating the green time of the seed solution at the jth entrance, G' indicating the total green time of the seed solution, G indicating the total green time of the maximum allowable signal lamp cycle period, VImaxRepresenting the maximum amount of incoming traffic allowed at the intersection,
Figure BDA0002971570810000075
indicating a downward rounding operator;
and finally, generating an initial population according to the following formula based on the dimensional values of the initial seed solution:
Figure BDA0002971570810000076
wherein
Figure BDA0002971570810000077
A gaussian distribution with a mean value of 0 and a standard deviation of 5 is represented by N (0,5) representing the j dimension of the ith particle in the population;
a knowledge ratio threshold KT is set when a seed group is initialized, the knowledge ratio threshold KT is set to be 0.8 in the embodiment, each time the initial particle is generated by adopting a formula (11), a random number in the range of [0,1] is generated, if the random number is less than KT, the initial particle is generated by using a formula (11), otherwise, the particle is initialized randomly;
s2, most particles of the initial population are generated by Gaussian sampling based on the seed solution in the step S1, so that the situation that each dimension value does not meet the feasible solution requirement may occur, and therefore the generated particle swarm individuals need to be restricted and corrected to meet the constraint requirement of the feasible solution;
and S3, before the speed of the particles is updated, judging that the sample individuals of the particles do not need to be updated. A list update threshold RT is set here, set to 7 in this embodiment. If the global optimal position of the algorithm is not improved after the RT generation, the sample is updated first, and then the speed and the position of the particles are updated. In the process of updating the sample, for the jth dimension of the sample individual, different particles in the other two groups are selected, and the jth dimension of the historical optimal position of the particle with good adaptive value is taken as the jth dimension of the sample individual. If the global optimal position of the algorithm is improved in the RT generation, which shows that the sample individual can still continue to guide the particle evolution, the speed and position of the particle can be directly updated without updating the sample,
the particle adopts a multi-sample learning strategy to update the speed, and the updating formula is as follows:
Figure BDA0002971570810000081
wherein
Figure BDA0002971570810000082
Represents the t +1 th generationThe velocity of the ith particle in the jth dimension,
Figure BDA0002971570810000083
represents the j-th dimension velocity of the ith particle in the t generation,
Figure BDA0002971570810000084
represents the j-th dimension position value of the ith particle in the t generation, omega represents the velocity weight,
Figure BDA0002971570810000085
representing the historical optimum position, gBest, of the ith particle in the jth dimensionjRepresents the global optimal position in the j-th dimension,
Figure BDA0002971570810000086
a list of individual values representing the jth dimension of the ith particle, c1，c2And c3There are three update coefficients that are,
Figure BDA0002971570810000087
and
Figure BDA0002971570810000088
is the j dimension [0,1]]The particle position value is updated by adding the original position and the updated new speed in the formula (12) as the new position of the particle;
s4, after the speed and the position of the particle are updated, performing simulation evaluation on the traffic signal lamp timing scheme represented by the downward rounding of the position value of the particle through microscopic simulation software VISSIM, and taking the evaluation result as the adaptive value of the particle;
s5, when the algorithm iterates to the specified maximum iteration number, the iteration is set to 40 generations in the embodiment, the global optimal solution at the moment is rounded down to be used as a final timing scheme, otherwise, the steps S3-S4 are continuously executed until the algorithm termination condition is met, namely, the maximum iteration number is 40 generations;
in summary, the method adopts a new multi-chart-sample learning strategy aiming at the problem that the traditional particle swarm optimization is easy to fall into the local optimum when being applied to the signal lamp optimization timing scene, and the strategy enables the particles to learn the self optimum position and the global optimum position and simultaneously learn the charts of different dimensions of other particles, thereby being beneficial to enhancing the diversity of the algorithm and avoiding falling into the local optimum position. In addition, the method adopts a knowledge embedding auxiliary strategy when generating the initial particle species group, and uses the distribution characteristics of the traffic flow input at the intersection as the front knowledge to assist the generation of the initial species group. The optimization results of the vehicle flows with different saturation degrees at the single intersection are displayed, the method is better in diversity compared with other timing optimization methods on the premise of ensuring the convergence speed, and the optimized timing scheme is better in comprehensive performance.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.

Claims (7)

1. A smart city signal lamp timing optimization method based on multi-sample learning particle swarm is characterized by comprising the following steps:
s1, generating an initial seed solution according to the traffic flow distribution characteristics of each entrance of the intersection, and generating an initial particle swarm according to the initial seed solution, wherein the process is as follows:
s101, defining the average lane flow rate as follows:
Figure FDA0002971570800000011
Figure FDA0002971570800000012
wherein, ALFjIndicates the average lane flow magnitude, VI, of the jth inletjIndicating the magnitude of the incoming traffic at the jth ingress,LNjindicating the number of lanes at the jth access opening, RjRepresents the average lane flow rate of the jth access port;
s102, generating an initial seed solution according to the following formula based on the average traffic lane flow ratio, wherein the green light time of each dimensionality of the initial seed solution is calculated by the following formula:
Figure FDA0002971570800000016
Figure FDA0002971570800000013
wherein g'jIndicating the green time of the seed solution at the jth entrance, G' indicating the total green time of the seed solution, G indicating the total green time of the maximum allowable signal lamp cycle period, VImaxRepresenting the maximum amount of incoming traffic allowed at the intersection,
Figure FDA0002971570800000014
indicating a downward rounding operator;
s103, generating an initial population according to the following formula based on the dimensional values of the initial seed solution:
Figure FDA0002971570800000015
wherein
Figure FDA0002971570800000021
A gaussian distribution with a mean value of 0 and a standard deviation of 5 is represented by N (0,5) representing the j dimension of the ith particle in the population;
s2, carrying out constraint correction on the generated particle swarm individuals to enable the particle swarm individuals to meet constraint requirements of feasible solutions;
s3, the speed of the particles of the population is updated according to the multi-sample learning strategy, and the updating formula is as follows:
Figure FDA0002971570800000022
wherein
Figure FDA0002971570800000023
Represents the j-th dimension velocity of the ith particle in the t +1 th generation,
Figure FDA0002971570800000024
represents the j-th dimension velocity of the ith particle in the t generation,
Figure FDA0002971570800000025
represents the j-th dimension position value of the ith particle in the t generation, omega represents the velocity weight,
Figure FDA0002971570800000026
representing the historical optimum position, gBest, of the ith particle in the jth dimensionjRepresents the global optimal position in the j-th dimension,
Figure FDA0002971570800000027
a list of individual values representing the jth dimension of the ith particle, c1，c2And c3There are three update coefficients that are,
Figure FDA0002971570800000028
and
Figure FDA0002971570800000029
is the j dimension [0,1]]Three different random numbers in between;
s4, performing simulation evaluation on the traffic signal lamp timing scheme represented by the downward rounding of the position value of the particle through microscopic simulation software VISSIM, and taking the evaluation result as the adaptive value of the particle;
and S5, when the algorithm is iterated to the specified maximum iteration number, taking the global optimal scheme as a final timing scheme, otherwise, continuing to execute the steps S3-S4 until the maximum iteration number preset by the algorithm is terminated.
2. The smart city signal lamp timing optimization method based on multi-sample learning particle swarm according to claim 1, wherein in the step S1, in generating an initial seed solution according to traffic flow distribution characteristics of each entrance of an intersection, a knowledge ratio threshold KT is set, when generating population individuals, a random number between [0 and 1] is first taken, if the random number is smaller than KT, the individuals are generated by using a knowledge-assisted strategy, otherwise, the individuals are generated according to a random method.
3. The method for optimizing the timing of signal lights in a smart city based on multi-board learning particle swarm as claimed in claim 1, wherein in step S3, a board update interval threshold RT is set, and if the global optimal position of the algorithm is not improved after the RT generation, the board update is performed, otherwise, the board of the previous generation is continued.
4. The method for optimizing the timing of signal lights in smart city based on multi-sample learning particle swarm as claimed in claim 1, wherein the update coefficient c is1、c2And c3Set to 0.75, 0.75 and 1.50, respectively.
5. The method for optimizing the timing of signal lamps in a smart city based on multi-sample learning particle swarm as claimed in claim 2, wherein the knowledge-to-ratio threshold KT is set to 0.8.
6. The method for optimizing the timing of signal lights in smart city based on multi-board learning particle swarm as claimed in claim 3, wherein the board update interval threshold RT is set to 7.
7. The method for optimizing the timing of signal lights in a smart city based on multi-sample learning particle swarm of claim 1, wherein the maximum number of iterations is set to 40 generations.
CN202110271046.7A 2021-03-11 2021-03-11 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm Expired - Fee Related CN113012449B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN202110271046.7A CN113012449B (en) 2021-03-11 2021-03-11 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm
LU503042A LU503042B1 (en) 2021-03-11 2021-09-06 Multi-model learning particle swarm-based intelligent city signal light timing optimization method
PCT/CN2021/119200 WO2022188387A1 (en) 2021-03-11 2021-09-18 Multi-model learning particle swarm-based intelligent city signal light timing optimization method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110271046.7A CN113012449B (en) 2021-03-11 2021-03-11 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm

Publications (2)

Publication Number Publication Date
CN113012449A CN113012449A (en) 2021-06-22
CN113012449B true CN113012449B (en) 2022-03-29

Family

ID=76406286

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110271046.7A Expired - Fee Related CN113012449B (en) 2021-03-11 2021-03-11 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm

Country Status (3)

Country Link
CN (1) CN113012449B (en)
LU (1) LU503042B1 (en)
WO (1) WO2022188387A1 (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113012449B (en) * 2021-03-11 2022-03-29 华南理工大学 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm
CN116434576B (en) * 2022-12-12 2024-03-12 中电信数字城市科技有限公司 Traffic light timing scheme determining method, device, system and equipment
CN117523823A (en) * 2023-10-11 2024-02-06 吉林师范大学 Regional traffic signal control optimization method based on quantum genetic algorithm
CN117077041B (en) * 2023-10-16 2023-12-26 社区魔方(湖南)数字科技有限公司 Intelligent community management method and system based on Internet of things
CN117787497B (en) * 2023-12-29 2024-06-25 西安电子科技大学广州研究院 Multi-objective optimization method and terminal applied to automobile insurance pricing
CN118172941B (en) * 2024-05-16 2024-07-19 山东领军智能交通科技有限公司 Traffic light control method and device based on wireless communication

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1889114A (en) * 2006-07-17 2007-01-03 中国科学院地理科学与资源研究所 Traffic signal off-line time distribution optimizing method basedon particle group operation method
CN104134356A (en) * 2014-06-30 2014-11-05 南京航空航天大学 Control method of city intersection model reference self-adaptive signals
CN106022463A (en) * 2016-05-13 2016-10-12 安徽教育网络出版有限公司 Personalized learning path optimization method based on improved particle swarm optimization algorithm
CN106384521A (en) * 2016-09-18 2017-02-08 广东工业大学 Single-intersection traffic signal optimization control method based on public transport priority
CN109035811A (en) * 2018-08-28 2018-12-18 大连理工大学 A kind of intelligent traffic lamp real-time monitoring method based on digital information element

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113012449B (en) * 2021-03-11 2022-03-29 华南理工大学 Smart city signal lamp timing optimization method based on multi-sample learning particle swarm

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1889114A (en) * 2006-07-17 2007-01-03 中国科学院地理科学与资源研究所 Traffic signal off-line time distribution optimizing method basedon particle group operation method
CN104134356A (en) * 2014-06-30 2014-11-05 南京航空航天大学 Control method of city intersection model reference self-adaptive signals
CN106022463A (en) * 2016-05-13 2016-10-12 安徽教育网络出版有限公司 Personalized learning path optimization method based on improved particle swarm optimization algorithm
CN106384521A (en) * 2016-09-18 2017-02-08 广东工业大学 Single-intersection traffic signal optimization control method based on public transport priority
CN109035811A (en) * 2018-08-28 2018-12-18 大连理工大学 A kind of intelligent traffic lamp real-time monitoring method based on digital information element

Also Published As

Publication number Publication date
LU503042B1 (en) 2023-03-13
WO2022188387A1 (en) 2022-09-15
CN113012449A (en) 2021-06-22

Similar Documents

Publication Publication Date Title
CN113012449B (en) Smart city signal lamp timing optimization method based on multi-sample learning particle swarm
CN111696370B (en) Traffic light control method based on heuristic deep Q network
CN112669629B (en) Real-time traffic signal control method and device based on deep reinforcement learning
CN110136455B (en) Traffic signal lamp timing method
CN109785619A (en) Regional traffic signal coordination and optimization control system and its control method
CN112700664A (en) Traffic signal timing optimization method based on deep reinforcement learning
CN109887297B (en) Method for dividing urban traffic control subareas based on rapid global K-means spectral clustering
CN113223305B (en) Multi-intersection traffic light control method and system based on reinforcement learning and storage medium
CN106251649A (en) Based on alleviating the control strategy of intersection congestion under hypersaturated state
CN109269516B (en) Dynamic path induction method based on multi-target Sarsa learning
CN113963553A (en) Road intersection signal lamp green signal ratio control method, device and equipment
CN108629970B (en) Intersection signal parameter optimization method based on Monte Carlo tree search
CN113538910A (en) Self-adaptive full-chain urban area network signal control optimization method
CN112365713B (en) Main branch intersection signal timing optimization method
CN115578870B (en) Traffic signal control method based on near-end policy optimization
CN116992928A (en) Multi-agent reinforcement learning method for fair self-adaptive traffic signal control
CN113421439A (en) Monte Carlo algorithm-based single intersection traffic signal timing optimization method
CN113392577B (en) Regional boundary main intersection signal control method based on deep reinforcement learning
CN117711173A (en) Vehicle path planning method and system based on reinforcement learning
CN113077642A (en) Traffic signal lamp control method and device and computer readable storage medium
CN117351729A (en) Urban intersection traffic signal real-time control method with black matrix assisting gradient optimization
CN115472023A (en) Intelligent traffic light control method and device based on deep reinforcement learning
CN113421444B (en) Urban road network signal control method and device based on vehicle path information
CN114723156A (en) Global traffic signal lamp regulation and control method based on improved genetic algorithm
Wei et al. Intersection signal control approach based on pso and simulation

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: 20220329

CF01 Termination of patent right due to non-payment of annual fee