A green wave maximum bandwidth phase difference optimization method

Abstract

The present application relates to the field of urban traffic signal optimization control, and particularly relates to a green wave maximum bandwidth phase difference optimization method, which comprises the following steps: obtaining all intersection vehicle data in a target area, and using the vehicle data to count lane level traffic data of the intersection; storing phase timing data, distance data and traffic data of the target line; calculating key intersections of the green wave line to obtain a green wave cycle; using a green light allocation ratio method to obtain signal timing data of each intersection of the green wave line with the same cycle; calculating the passing time of associated traffic flow; using a particle swarm algorithm to initialize particle phase difference, reducing the positive and negative bandwidth gap through a penalty mechanism, obtaining phase differences of all intersections of the green wave, and obtaining phase differences of all intersections of the green wave with the maximum green wave bandwidth through an iterative method. Through the use of the green wave maximum bandwidth phase difference optimization model, the traffic signal is adjusted according to real-time traffic flow, and the traffic flow liquidity and efficiency are optimized.

Description

Technical Field

[0001] This invention relates to the field of urban traffic signal optimization control, specifically to a method for optimizing the phase difference of green wave maximum bandwidth. Background Technology

[0002] Green wave phase difference optimization is a traffic signal control technology designed to optimize traffic flow and vehicle throughput on urban roads. It achieves this optimization by adjusting the phase difference of traffic lights (i.e., the timing difference between the switching of traffic lights at adjacent intersections).

[0003] Commonly used green wave phase difference optimization techniques employ intelligent traffic signal control algorithms, such as optimization control algorithms, genetic algorithms, and tabu search, to calculate the optimal green wave phase difference setting. These algorithms consider factors such as traffic flow, vehicle delays, and intersection priorities. Traffic signal optimization software: Dedicated traffic signal optimization software is developed that automatically calculates the optimal green wave phase difference setting and generates the corresponding signal control scheme by inputting parameters such as road network topology, traffic flow data, and objective functions. Sensors and intelligent transportation systems: Traffic sensors, cameras, and intelligent transportation systems are used to monitor road traffic conditions in real time and feed the data back to the traffic signal control system. This allows for dynamic adjustment of the green wave phase difference based on real-time traffic conditions.

[0004] Traditional bidirectional green wave phase difference optimization suffers from several drawbacks, such as weak correlation between different intersections, inability to meet diverse traffic conditions and special circumstances, and difficulty in adapting to traffic flow changes with traditional fixed phase difference settings, leading to signal light misalignment or incoordination, thus reducing traffic flow and efficiency. Existing methods for coordinating path chain green wave bandwidth optimization rely on maintaining the existing green wave bandwidth, resulting in an incomplete optimization process. Traditional computational optimization models do not consider the actual phase difference or that signal timing is an integer objective in actual use; if floating-point calculations are used during optimization, it is easy to get trapped in local optima, failing to find the optimal result within the integer set. Existing methods for generating green wave control schemes based on time-distance maps simulate manual debugging steps and brute-force solve phase difference and bandwidth. However, due to the large number of objectives to be solved, including but not limited to forward and reverse phase differences and phase sequence, it is difficult to quickly and conveniently obtain the optimal solution. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a method for optimizing the phase difference of the maximum bandwidth of green waves.

[0006] The method includes:

[0007] A method for optimizing the phase difference of green wave maximum bandwidth is characterized by the following steps:

[0008] S1: Obtain vehicle data from all intersections within the target area through electronic police checkpoint equipment, and use the vehicle data to statistically analyze traffic flow data at the intersection lane level.

[0009] S2 stores the phase timing data that is currently allowed at each intersection on the target route, the distance data between two intersections, and the traffic flow data for each direction of lane turning at each intersection;

[0010] S3, based on phase timing data, distance data, and traffic flow data, calculates the key intersections of the green wave route and obtains the green wave cycle;

[0011] S4 uses the green light allocation ratio method, which allocates signal timing according to the difference between the original intersection signal timing and the green wave cycle, as well as the proportion of traffic flow demand in each phase, to obtain the signal timing data of the same cycle for each intersection on the green wave line.

[0012] S5 generates associated traffic flow based on the coordinated direction of the adjacent intersections on the green wave line. The forward and reverse directions of the two-way green wave lines are separated, and the travel time of the associated traffic flow is calculated based on the green wave speed limit.

[0013] S6, based on urban road intersection traffic flow data, intersection topology, and associated traffic flow travel times, uses the particle swarm optimization algorithm for particle phase difference initialization, generating a total of... An initial number of particles is set according to actual needs, denoted as . ,in The dimension is n-1, and the signal period is ;

[0014] By using a penalty mechanism to reduce the bandwidth difference between the forward and reverse directions, the phase difference of all intersections in the green wave is obtained, and the phase difference of all intersections in the green wave that maximizes the green wave bandwidth is obtained through iteration.

[0015] Furthermore, step S3 is based on the existing optimized scheme where the maximum value of the signal period at all intersections along the entire green wave line is the green wave period; or the green wave period is set according to requirements.

[0016] Furthermore, the iterative part of step S6 includes: calculating the green light time mapping of associated traffic flows, calculating the time overlap of associated traffic flows, calculating the minimum bandwidth and the sum of bandwidths of the green wave, calculating the optimization objective, and updating the phase difference data using the particle swarm algorithm.

[0017] Furthermore, the calculation of the associated traffic flow green light time mapping includes: Using different sets of phase difference data, calculate the start and end times of the coordinated phase green light time at the starting intersection of each associated traffic flow within the same cycle:

[0018] ;

[0019] in Indicates associated traffic flow. The table shows the start and end times of the originating intersections for associated traffic flows. Indicates the start and end times of the intersection where the associated traffic flow ends, with a time range of [time range missing]. Within the range;

[0020] The coordinated time period will be mapped to the start and end times of the downstream flow according to the formula. The calculation formula for a related traffic flow is as follows:

[0021] ;

[0022] in This represents the range of start and end times that the upstream of a related traffic flow maps to the downstream based on travel time and phase difference; if it exceeds the cycle, the remainder is taken after dividing by the cycle, and the range becomes... ; To determine the phase difference between the upstream and downstream intersections of the related traffic road; The travel time is calculated based on the length of the starting and ending road segments and the green wave speed; the starting and ending time periods are obtained after mapping all associated traffic flows.

[0023] Furthermore, the time overlap of related traffic flows satisfies:

[0024] ;

[0025] in To determine the overlap of related traffic flows, To map the upstream of the associated traffic flow to the downstream start time based on travel time and phase difference, The upstream of the associated traffic flow is mapped to the downstream end time based on the travel time and phase difference;

[0026] The steps for solving the time overlap of related traffic flows include:

[0027] S6.1, two lists of closed intervals for comparison, denoted as intervals_A and intervals_B, each interval contains left and right endpoints, and the endpoint values ​​are 0 or positive integers. Sort the intervals in the two lists from left to right according to the X coordinate.

[0028] S6.2, define an intersection length variable with an initial value of 0;

[0029] S6.3, starting from the first item, compare each item of intervals_A with each item of intervals_B;

[0030] S6.4, intervals_A[a] represents using the a-th item in intervals_A, and intervals_B[b] represents using the b-th item in intervals_B. If intervals_A[a] and intervals_B[b] have an intersection, the current intersection length is added to the intersection length variable;

[0031] S6.5, traversal is achieved by moving the comparison interval:

[0032] If the right endpoint of intervals_B[b] is less than the right endpoint of intervals_A[a], and intervals_B[b] is not the last item in intervals_B, then shift to the right to the next interval in intervals_B and enter the next round of iteration for comparison;

[0033] If the right endpoint of intervals_B[b] is greater than or equal to the right endpoint of intervals_A[a], and intervals_A[a] is not the last item in intervals_A, then shift to the right of the next interval in intervals_A and enter the next round of iteration for comparison;

[0034] S6.6 After the iteration is complete, return the sum of the intersection lengths.

[0035] Furthermore, the specific steps for calculating the minimum bandwidth of the green wave and the sum of the bandwidths, calculating the penalty for the difference between the forward and reverse bandwidths, and optimizing the target value are as follows:

[0036] Calculate the sum of forward and reverse bandwidths of associated traffic flows :

[0037] ;

[0038] ;

[0039] ;

[0040] in For positive associated traffic flow bandwidth, For the bandwidth of reverse traffic flow, each has strip;

[0041] Calculate the minimum bandwidth of the green wave band :

[0042] ;

[0043] in This is a related traffic flow, shared in both directions. strip, Indicates the first Related traffic flows;

[0044] Calculate the penalty for the difference between forward and reverse bandwidth :

[0045] ;

[0046] in This represents the difference between the forward and reverse bandwidths.

[0047] Target value calculation:

[0048] ;

[0049] in To optimize the objective; This is the penalty coefficient.

[0050] Furthermore, the particle swarm optimization algorithm updates the phase difference data, including:

[0051] Particle swarm optimization algorithm updates phase difference data: initial The initial calculation of each particle yields... Target value results The local and global optimal solutions for all particles are calculated and updated iteratively.

[0052] Calculate the local optimum :

[0053] ;

[0054] in, This refers to the number of iterations. For local optimal solutions 3D phase difference;

[0055] Calculate the global optimal solution :

[0056] ;

[0057] ;

[0058] in, The target value of the local optimum in the current iteration. The target value is the globally optimal solution from all previous iterations. For the global optimal solution 3D phase difference;

[0059] Based on the particle swarm optimization algorithm, the velocity value is updated using the global optimum and local optima. :

[0060] ; ;

[0061] in, This is the inertia factor for the particle swarm optimization algorithm; The discrete threshold is a random number between 0 and 1; This refers to the update speed in this iteration; The speed value updated in the previous iteration. =0;

[0062] Discretize the values ​​that need to be updated into integers, and use a formula to change the magnitude of each update to ±1 or 0:

[0063] ;

[0064] ;

[0065] Update the timing for each particle:

[0066] ;

[0067] in , , , All are dimensional vector, corresponding The timing is determined and updated based on the above calculations. For normalized variables, To update the step size discretely.

[0068] Furthermore, M takes the value 50.

[0069] Furthermore, .

[0070] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0071] 1. The calculation method of the phase difference optimization model combines the minimum bandwidth of the green wave and the total bandwidth of the green wave as the target value, so that the minimum bandwidth of the green wave reaches the maximum and the total bandwidth also reaches a large value. The traffic imbalance caused by excessive bandwidth difference between the forward and reverse green waves is limited by the form of penalty value. The maximum value of the target value is used to obtain the optimal phase difference in the final result, which can more accurately complete the green wave line phase difference configuration task and meet the signal control needs of real road travel.

[0072] 2. The calculation of model overlap, i.e. green wave bandwidth, is not applicable to the discontinuous coordination stage when using formula calculation. The algorithm proposed in this paper, which uses the intersection of multiple closed intervals, can calculate the green wave bandwidth more conveniently and quickly. This is the first application of this algorithm in the field of signal control for green wave phase difference calculation. It can adapt to the calculation of various overlapping phases and uncertain number of interval intersections, and is the best application of this algorithm in green wave signal optimization phase difference calculation.

[0073] 3. Traditional computational optimization models do not consider that the actual phase difference or signal timing is an integer target in practical use. In actual optimization, if calculations are performed using floating-point numbers, it is easy to get trapped in local optima and fail to find the optimal result within the set of integers. The discretization method of this model ensures that the result of each round of calculation is an integer value, and the iteration update is also performed according to integers. Combined with the randomness of multiple particles in the particle swarm, the optimal solution of the model is easier to find, and the optimal phase difference can converge to a stable value more quickly.

[0074] 4. The entire process is an improvement on the complex engineering implementation process. It simplifies the steps of single-intersection timing optimization, applies new algorithms and optimization methods, and, together with the overall process, can quickly calculate the green wave phase difference. Especially for green waves currently running on existing roads, following the steps in this article and inputting data can quickly calculate and output a set of optimal phase differences, achieving best practices for online applications. Attached Figure Description

[0075] Figure 1 This is a schematic diagram illustrating the overall implementation steps of an embodiment of the present invention;

[0076] Figure 2 A schematic diagram of associated traffic flow provided in an embodiment of the present invention;

[0077] Figure 3 The flowchart illustrates the particle swarm optimization algorithm solution provided in this embodiment of the invention. Detailed Implementation

[0078] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Before describing the technical solutions of each embodiment of the present invention in detail, the terms and terms involved will be explained. In this specification, components with the same name or the same reference numerals represent similar or the same structures and are limited to illustrative purposes.

[0079] Terminology Explanation:

[0080] (1) Green Wave: Green wave is a traffic signal control method that configures control signal equipment on urban trunk roads, main roads, and other roads with high vehicle traffic demand. By adjusting the timing of traffic lights, vehicles can maintain a continuous and smooth driving state at multiple intersections along the route. By maximizing the bandwidth phase difference, that is, setting different phase differences between adjacent intersections, traffic flow can be better coordinated. According to one-way and two-way roads, it is further divided into different control methods such as one-way green wave and two-way green wave.

[0081] (2) Green wave bandwidth: Green wave bandwidth refers to the period of continuous passage created between adjacent intersections in a traffic signal control system. It represents the time window during which vehicles can travel continuously between a set of adjacent intersections without stopping under green wave control. The length of the green wave bandwidth is designed and adjusted based on factors such as traffic flow, road length, vehicle speed, and traffic demand. A longer green wave bandwidth means that vehicles can travel continuously over a longer distance, reducing waiting time and traffic congestion. A shorter green wave bandwidth is suitable for areas with low traffic flow or situations where traffic lights need to be adjusted more frequently.

[0082] (3) Green wave cycle: In traffic signal control, the green wave cycle refers to the changing cycle of a group of traffic lights at adjacent intersections, so as to form a green wave passage time interval between these intersections. The green wave cycle is usually used to coordinate traffic flow, so that vehicles can pass through multiple adjacent intersections continuously at a certain speed, reducing waiting time and traffic congestion.

[0083] (4) Key intersections: Among the intersections that constitute the green wave line, the intersections whose phase timing and passage time have the greatest impact on determining the green wave cycle and passage demand.

[0084] This invention discloses a method for optimizing the phase difference based on the maximum bandwidth of green waves. It combines the original signal configuration scheme of the actual intersection with the green wave coordination ratio setting, provides a method specifically for the phase difference, and gives a complete green wave optimization flow optimization process.

[0085] This invention is achieved through the following technical solution:

[0086] like Figure 1 As shown, the overall implementation steps include three parts:

[0087] 1. Equipment basic data collection and acquisition

[0088] 1.1 Equipment transit data collection

[0089] Traffic data is collected through electronic traffic enforcement checkpoints. This data is then used to compile traffic flow data at the lane level at intersections, which can be used for automated or manual signal timing optimization at single intersections.

[0090] 1.2 Acquisition of Basic Information for Green Wave Lines

[0091] Green wave is a traffic signal control scheme that aims to enable vehicles to smoothly pass through multiple adjacent intersections within a certain area by rationally adjusting the timing of traffic lights at intersections, thereby forming a continuous and smooth green light passage route.

[0092] Phase timing: The phase timing of traffic lights at intersections in a green wave circuit is coordinated to ensure that vehicles encounter a green light when passing through an adjacent intersection. Phase timing includes green light duration, yellow light duration, and red light duration.

[0093] Distance settings: The design of green wave routes also takes into account the distance between adjacent intersections. By calculating the average vehicle speed and segment length, the start and end points of the green wave route are determined to achieve coordination of green light duration throughout the route.

[0094] Traffic flow and demand: The design of green wave routes also needs to consider traffic flow and demand. The timing parameters of the green wave routes are adjusted according to changes in traffic flow during different time periods to meet traffic demand in different directions.

[0095] This section of the embodiments stores and records the phase timing data that is allowed at each intersection on the route, the distance data between two intersections, and the traffic flow data of each direction of lane turning at each intersection, which is used for the generation of green wave schemes and the calculation of phase differences.

[0096] 2. Intersection signal timing optimization and green wave cycle calculation

[0097] 2.1 Identify key intersections and green wave cycles

[0098] This embodiment is based on existing optimization schemes. It selects the intersection with the longest cycle as the key intersection and uses the signal cycle of this intersection as the cycle of the green wave line; or for lines with special needs, it designates a key intersection (such as an intersection in a densely populated area, a major congestion intersection, etc.) as the key intersection and manually sets the green wave cycle duration according to the needs.

[0099] 2.2 Fixed-period intersection timing adjustment

[0100] Fixed-cycle signal timing optimization targets signal control methods with fixed time cycles, aiming to improve traffic efficiency and reduce congestion. Common methods include the green light allocation ratio method, the maximum average vehicle time method, the saturation method, and simulation optimization. The green light allocation ratio method, in particular, allocates the cycle time proportionally to the green light time for each direction based on traffic flow and demand. Important traffic directions can be allocated more green light time to improve traffic efficiency.

[0101] This embodiment uses the green light allocation ratio method, allocating green light according to the difference between the original intersection signal timing and the green wave cycle, based on the proportion of traffic flow demand in each phase. This yields signal timing data for the same cycle at each intersection along the green wave route, primarily the green light duration at each intersection in each phase.

[0102] If there is a need for priority clearance in the coordinated direction during the green wave, the green light duration can be configured according to the ratio of coordinated to non-coordinated directions. For example, if the minimum coordination ratio is set to 0.4 and the cycle is 100 seconds, then the total green light duration for coordinated directions must be greater than 40 seconds, and the total phase duration for non-coordinated directions must be less than 60 seconds.

[0103] 3. Phase difference optimization model based on maximum bandwidth of green wave

[0104] 3.1 Green wave generates associated traffic flows at adjacent intersections and calculates standard travel time.

[0105] like Figure 2 As shown, based on the coordinated turning directions at adjacent intersections along the green wave route, associated traffic flows are generated. Two-way green waves separate the forward and reverse directions. The travel time of the associated traffic flows is calculated based on the green wave speed limit. A straight-ahead green wave is relatively common, but it still applies if there are left-turn or right-turn coordinated phases.

[0106] The coordinated directions that constitute the associated traffic flows are shown in Table 1.

[0107] Table 1. Coordinated directions constituting associated traffic flows

[0108]

[0109] Wherein, … indicates that all associated traffic flows are derived according to the associated traffic flows in the table. f1, f2, … represent associated traffic flows, from the starting intersection to the ending intersection, which are the coordinated directions for the entry and exit of the current associated traffic flows. The coordinated direction can be left turn, straight, or right turn; These represent the forward and reverse green wave velocities, which can be the same or different. The travel time for each traffic flow is calculated based on the length of the starting and ending road segments and the green wave speed.

[0110] 3.2 Phase Difference Optimization Model Based on Maximum Bandwidth of Green Wave

[0111] The phase difference optimization model based on the maximum bandwidth of green waves is a method for traffic signal optimization. This model aims to maximize traffic flow and green wave bandwidth by adjusting the phase difference of traffic signals (i.e., the time difference between the turn-on of traffic lights). In traditional traffic signal control, the phase difference of traffic lights at each intersection is usually fixed, which can lead to traffic congestion and inefficiency. The goal of the phase difference optimization model is to find the optimal phase difference configuration to maximize traffic flow and the capacity of the green wave path.

[0112] This model, based on traffic flow and intersection topology, uses optimization algorithms to determine the optimal phase difference configuration. The model flow is as follows: Figure 3 As shown, the optimization method maximizes the bandwidth of the green wave band by maximizing the objective and prevents excessive differences in bandwidth between forward and reverse directions through penalties, thus obtaining the optimization method for the phase difference of the green wave at all intersections. The optimization algorithm for solving the model can employ traditional mathematical programming methods, such as linear programming, integer programming, or nonlinear programming, to minimize or maximize the defined objective function. In addition, metaheuristic algorithms such as evolutionary algorithms, genetic algorithms, particle swarm optimization, and simulated annealing are also frequently used to solve this type of problem.

[0113] By using a phase difference optimization model based on the maximum bandwidth of the green wave, traffic signals can be adjusted according to real-time traffic flow to achieve optimal traffic flow and efficiency. This can reduce traffic congestion, improve transportation efficiency, and provide a better travel experience.

[0114] 3.2.1 Phase Difference Initialization and Range Limitation

[0115] Based on the previous diagram, the number of intersections for the bidirectional green wave route is set to 4, and the green wave cycle is set to... Define the range of minimum and maximum phase differences, the period range, and set the phase difference of any intersection to 0. Let's assume the phase difference of intersection 1 is 0. All other intersections will calculate their phase differences relative to intersection 1.

[0116] Signal period:

[0117] Intersection 1: Phase difference is: 0

[0118] Intersection 2: Phase Difference Range integer

[0119] Intersection 3: Phase Difference Range integer

[0120] Intersection 4: Phase Difference Range integer

[0121] Particle phase difference initialization is performed using the particle swarm optimization algorithm.

[0122] According to the particle swarm optimization algorithm, particles are the variables to be optimized, the phase difference at intersection 1 is 0, and the number of intersections on the line is... If so, then the number of phase differences that need to be optimized is The number of intersection phase differences that need to be optimized for the particle is 3. Following the above range constraints, initial particles are generated cyclically. Each particle generates integer data within the phase constraint range. If the range is not met or the calculation period does not meet the range, the particle is regenerated. For example, if the generated particle is... Then the phase difference of the four intersections is Co-generated An initial particle, in this embodiment Take 50. Set the number according to actual needs, and denote it as... ,in The dimension is dimension;

[0123] 3.2.2 Calculate the green light time mapping of associated traffic flows

[0124] Received Each particle can be considered as Using different sets of phase difference data, calculate the start and end times of the coordinated phase green light time at the starting intersection of each associated traffic flow within the same cycle:

[0125] ;

[0126] in Indicates associated traffic flow. The table shows the start and end times of the originating intersections for associated traffic flows. Indicates the start and end times of the intersection where the associated traffic flow ends, with a time range of [time range missing]. Within this range, if the green light duration for a certain intersection is in the second phase, ranging from 40 seconds to 65 seconds, then the green light start and end times are: Each related traffic flow has a coordination time period at its starting intersection, covering both forward and reverse directions. strip;

[0127] The coordinated time period will be mapped to the start and end times of the downstream flow according to the formula. The calculation formula for a related traffic flow is as follows:

[0128] ;

[0129] in This represents the range of start and end times that the upstream of a related traffic flow maps to the downstream based on travel time and phase difference; if it exceeds the cycle, the remainder is taken after dividing by the cycle, and the range becomes... ; To determine the phase difference between the upstream and downstream intersections of the related traffic road; The travel time is calculated based on the length of the starting and ending road segments and the green wave speed; the calculation of other related traffic flows is similar, resulting in the mapped start and end time periods for all related traffic flows.

[0130] 3.2.3 Calculation of Time Overlap of Associated Traffic Flows

[0131] Based on the downstream intersection time period and the upstream-to-downstream time period of the associated traffic flow obtained in the previous step, calculate the overlap between these two time periods, which yields the bandwidth value for each associated traffic flow. The formula for calculating the overlap is:

[0132] ;

[0133] in To determine the overlap of related traffic flows, To map the upstream of the associated traffic flow to the downstream start time based on travel time and phase difference, The upstream of the associated traffic flow is mapped to the downstream end time based on the travel time and phase difference. This step is essentially calculating the length of the intersection between the mapped time period and the original downstream time period. In the actual code implementation, an algorithm that calculates the intersection of multiple closed intervals can be used to calculate the length. This length is the overlap length, which is also the bandwidth of the associated traffic flow.

[0134] Briefly describe the algorithm for finding the intersection of multiple closed intervals:

[0135] (1) Two comparison closed interval lists, denoted as intervals_A and intervals_B, each interval contains left and right endpoints, and the endpoint values ​​are 0 or positive integers. Sort the intervals in the two interval lists from left to right according to the X coordinate.

[0136] (2) Define an intersection length variable with an initial value of 0;

[0137] (3) Starting from the first item, compare each item of intervals_A with each item of intervals_B;

[0138] (4) Let intervals_A[i] represent the i-th item in intervals_A, and intervals_B[j] represent the j-th item in intervals_B. If intervals_A[i] and intervals_B[j] have an intersection, add the current intersection length to the intersection length variable;

[0139] (5) Traversal is achieved by moving the comparison interval:

[0140] a. If the right endpoint of intervals_B[j] is less than the right endpoint of intervals_A[i], and intervals_B[j] is not the last item in intervals_B, then shift the interval to the right in intervals_B and enter the next round of iteration for comparison;

[0141] b. If the right endpoint of intervals_B[j] is greater than or equal to the right endpoint of intervals_A[i], and intervals_A[i] is not the last item in intervals_A, then shift the interval to the right in intervals_A and enter the next round of iteration for comparison;

[0142] (6) After the iteration is completed, return the sum of the intersection lengths.

[0143] 3.2.4 Calculate the minimum bandwidth and sum of bandwidths for the green wave.

[0144] (1) Calculate the sum of the forward and reverse bandwidths of the associated traffic flows:

[0145] ;

[0146] ;

[0147] ;

[0148] in For positive associated traffic flow bandwidth, For the bandwidth of the reverse associated traffic flows, there are n-1 flows.

[0149] (2) Calculate the minimum bandwidth of the green wave band:

[0150] ;

[0151] (3) Among them This is a related traffic flow, shared in both directions. The term "traffic flow" is represented by the number k, where k represents the kth associated traffic flow.

[0152] 3.2.5 Calculate the optimization objective: the maximum value of the penalized green wave bandwidth.

[0153] (1) Calculate the penalty for the difference between forward and reverse bandwidth

[0154] ;

[0155] in This represents the difference between the forward and reverse bandwidths.

[0156] (2) Target value calculation:

[0157] ;

[0158] Where OBJ is the optimization objective; The penalty coefficient is taken here. It can be adjusted according to actual needs and effects;

[0159] The optimization model design aims to maximize the green wave bandwidth with penalty and minimize the difference between the penalized forward and reverse bandwidths to meet the requirements of practical applications. The variable is the phase difference of each intersection of the green wave. Due to the uncertainty of the number of green wave intersections and the uncertainty of the set result, the model becomes nonlinear and needs to be solved. Here, the particle swarm optimization algorithm is used to calculate the maximum value of the model and obtain the optimal phase difference result.

[0160] 3.2.6 Particle Swarm Optimization Algorithm for Updating Phase Difference Data

[0161] The initial M particles were used to calculate the initial M target values. The local and global optimal solutions for all particles are calculated and updated iteratively.

[0162] (1) Calculate the local optimal solution :

[0163] ;

[0164] Where t is the number of iterations;

[0165] (2) Calculate the global optimal solution

[0166] ;

[0167] ;

[0168] in, The target value of the local optimum in the current iteration. This is the target value of the globally optimal solution in all previous iterations;

[0169] (3) Based on the particle swarm optimization algorithm, update the velocity value using the global optimum and local optima. :

[0170] ; ;

[0171] in, The inertia factor for the particle swarm optimization algorithm is set to 1. The random number t between 0 and 1 represents the update rate in this iteration;

[0172] (4) Discretize the values ​​that need to be updated into integers. Since the phase difference of the green wave is an integer, use the formula to change the amplitude of each update to ±1 or 0:

[0173] ;

[0174] ;

[0175] Update the timing of each particle

[0176]

[0177] in , , , All are n-1 dimensional vectors, corresponding to n-1 time allocations, and are updated through the above calculations.

[0178] 3.2.7 Repeat steps 3.2.2-3.2.6 N times to obtain the optimal phase difference result. In this embodiment, N is 2000.

[0179] Additional notes:

[0180] (1) Method for fixed-cycle signal timing at intersection S04. Commonly used methods include green light allocation ratio method, maximum average vehicle speed method, saturation method, and simulation optimization method. These methods can be used individually or in combination for signal timing optimization. This method adopts the simplest green light allocation ratio method and combines it with a pre-set coordination direction ratio to allocate the difference between the fixed cycle and the original cycle. Under the premise of ensuring that the coordination ratio setting is met, the cycle time difference is then allocated to the original scheme according to the ratio.

[0181] (2) The particle swarm optimization algorithm used to solve the optimization objective in the S06 model: The optimization objective of the model is essentially a nonlinear optimization model based on the minimum value, and it is solved by a heuristic algorithm. The particle swarm optimization algorithm is one of them. Other heuristic algorithms that can solve nonlinear optimization can be replaced, such as simulated annealing algorithm, ant colony algorithm, etc. The particle swarm optimization algorithm can find the optimal solution of the nonlinear model faster by using local optimal solution and global optimal solution, which is more suitable for the actual application scenario requirements.

[0182] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for optimizing the phase difference of green wave maximum bandwidth, characterized in that, Includes the following steps: S1: Obtain vehicle data from all intersections within the target area through electronic police checkpoint equipment, and use the vehicle data to statistically analyze traffic flow data at the intersection lane level. S2 stores the phase timing data that is currently allowed at each intersection on the target route, the distance data between two intersections, and the traffic flow data for each direction of lane turning at each intersection; S3, based on phase timing data, distance data, and traffic flow data, calculates the key intersections of the green wave route and obtains the green wave cycle; S4 uses the green light allocation ratio method, which allocates signal timing according to the difference between the original intersection signal timing and the green wave cycle, as well as the proportion of traffic flow demand in each phase, to obtain the signal timing data of the same cycle for each intersection on the green wave line. S5 generates associated traffic flow based on the coordinated direction of the adjacent intersections on the green wave line. The forward and reverse directions of the two-way green wave lines are separated, and the travel time of the associated traffic flow is calculated based on the green wave speed limit. S6, based on urban road intersection traffic flow data, intersection topology, and associated traffic flow travel times, uses the particle swarm optimization algorithm for particle phase difference initialization, generating a total of... An initial number of particles is set according to actual needs, denoted as . ,in The dimension is n-1, and the signal period is ; By using a penalty mechanism to reduce the bandwidth difference between the forward and reverse directions, the phase difference of all intersections in the green wave is obtained, and the phase difference of all intersections in the green wave that maximizes the green wave bandwidth is obtained through iteration.

2. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 1, characterized in that, Step S3 is based on the existing optimized scheme where the maximum signal period at all intersections along the entire green wave line is the green wave period; or the green wave period is set according to requirements.

3. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 1, characterized in that, The iterative part of step S6 includes: calculating the green light time mapping of associated traffic flows, calculating the time overlap of associated traffic flows, calculating the minimum bandwidth and the sum of bandwidths of the green wave, calculating the optimization objective, and updating the phase difference data using the particle swarm algorithm.

4. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 3, characterized in that, The calculation of the associated traffic flow green light time mapping includes: Using different sets of phase difference data, calculate the start and end times of the coordinated phase green light time at the starting intersection of each associated traffic flow within the same cycle: ； in Indicates associated traffic flow. The table shows the start and end times of the originating intersections for associated traffic flows. Indicates the start and end times of the intersection where the associated traffic flow ends, with a time range of [time range missing]. Within the range; The coordinated time period will be mapped to the start and end times of the downstream flow according to the formula. The calculation formula for a related traffic flow is as follows: ； in This represents the range of start and end times that the upstream of a related traffic flow maps to the downstream based on travel time and phase difference; if it exceeds the cycle, the remainder is taken after dividing by the cycle, and the range becomes... ; To determine the phase difference between the upstream and downstream intersections of the related traffic road; The travel time is calculated based on the length of the starting and ending road segments and the green wave speed; the starting and ending time periods are obtained after mapping all associated traffic flows.

5. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 3, characterized in that, The time overlap of related traffic flows satisfies: ； in To determine the overlap of related traffic flows, To map the upstream of the associated traffic flow to the downstream start time based on travel time and phase difference, The upstream of the associated traffic flow is mapped to the downstream end time based on the travel time and phase difference; The steps for solving the time overlap of related traffic flows include: S6.1, two lists of closed intervals for comparison, denoted as intervals_A and intervals_B, each interval contains left and right endpoints, and the endpoint values ​​are 0 or positive integers. Sort the intervals in the two lists from left to right according to the X coordinate. S6.2, define an intersection length variable with an initial value of 0; S6.3, starting from the first item, compare each item of intervals_A with each item of intervals_B; S6.4, intervals_A[a] represents using the a-th item in intervals_A, and intervals_B[b] represents using the b-th item in intervals_B. If intervals_A[a] and intervals_B[b] have an intersection, the current intersection length is added to the intersection length variable; S6.5, traversal is achieved by moving the comparison interval: If the right endpoint of intervals_B[b] is less than the right endpoint of intervals_A[a], and intervals_B[b] is not the last item in intervals_B, then shift to the right to the next interval in intervals_B and enter the next round of iteration for comparison; If the right endpoint of intervals_B[b] is greater than or equal to the right endpoint of intervals_A[a], and intervals_A[a] is not the last item in intervals_A, then shift to the right of the next interval in intervals_A and enter the next round of iteration for comparison; S6.6 After the iteration is complete, return the sum of the intersection lengths.

6. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 3, characterized in that, The specific steps for calculating the minimum bandwidth of the green wave and the sum of the bandwidths, calculating the penalty for the difference between the forward and reverse bandwidths, and optimizing the target value are as follows: Calculate the sum of forward and reverse bandwidths of associated traffic flows : ； ； ； in For positive associated traffic flow bandwidth, For the bandwidth of reverse traffic flow, each has strip; Calculate the minimum bandwidth of the green wave band : ； in This is a related traffic flow, shared in both directions. strip, Indicates the first Related traffic flows; Calculate the penalty for the difference between forward and reverse bandwidth : ； in This represents the difference between the forward and reverse bandwidths. Target value calculation: ； in To optimize the objective; This is the penalty coefficient.

7. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 3, characterized in that, The particle swarm optimization algorithm updates the phase difference data, including: Particle swarm optimization algorithm updates phase difference data: initial The initial calculation of each particle yields... Target value results The local and global optimal solutions for all particles are calculated and updated iteratively. Calculate the local optimum : ； in, This refers to the number of iterations. For local optimal solutions 3D phase difference; Calculate the global optimal solution : ； ； in, The target value of the local optimum in the current iteration. The target value is the globally optimal solution from all previous iterations. For the global optimal solution 3D phase difference; Based on the particle swarm optimization algorithm, the velocity value is updated using the global optimum and local optima. : ； ； in, This is the inertia factor for the particle swarm optimization algorithm; The discrete threshold is a random number between 0 and 1; This refers to the update speed in this iteration; The speed value updated in the previous iteration. =0; Discretize the values ​​that need to be updated into integers, and use a formula to change the magnitude of each update to ±1 or 0: ； ； Update the timing for each particle: ； in , , , All are dimensional vector, corresponding The timing is determined and updated based on the above calculations. For normalized variables, To update the step size discretely.

8. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 1, characterized in that, M takes the value 50.

9. The method for optimizing phase difference based on maximum bandwidth of green wave according to claim 6, characterized in that, 。

Images